Fandom Platform of the Future✨

Requirements Gathering and Features Discussion

Table of contents:

INTRODUCTION

BACKGROUND

CONTINUE THE DISCUSSION/KEEP TABS

A TEMPERING OF EXPECTATIONS

SPECIFIC POPULATION FEEDBACK

HOUSEKEEPING NEEDS

NEXT STEPS

VOTING AND SUGGESTIONS

Account Management

User Interface & Accessibility

Content Presentation

Metadata & Discovery

Privacy & Content Controls

Integrations & Export

Revenue & Income (survivability)

TOS, Moderation, Abuse Team(s)

Groups and Communities

OTHER / AWAITING CATEGORIZATION

COMPLEX ISSUES / BRAINSTORMING

Flagging & Tagging of Adult, Sexual, and NSFW Content

Pros and cons of “reblogging” and nested conversations

Single Central Identity; Tiered protections

Should image descriptions (“alt-text”) be required or strongly encouraged?

Orphaning Posts & Entire Blogs

Tags as Metadata vs Whisperspace

Mobile App vs. Responsive Design

Age Requirements or Restrictions

Allowing alts while preventing bots

Legalities

What’s in the ToS?

Logging and backups

On the Question of Organizational/Platform Resilience

Off-site notifications

SOCIAL CORNER


INTRODUCTION

Hi! I made a tweet, and then Maciej Ceglowski DMed me and said if fandom (I realize this is not going to be all parts of fandom) could come to a consensus about what such a platform should look like, he said he’d see what he can do.

pearwaldorf (owner of the document) here -- my comments are in blue. Planning to stay out of this for the most part, just keeping an eye on duplicates and organization

BACKGROUND

As you may have figured out by his handle, Maciej is Pinboard guy. He stepped in after the whole Delicious debacle and made Pinboard more fandom-friendly. Fandom put together a list of requirements and nice-to-haves in this Google Doc. I do not know if this one will reach the size of the Pinboard doc, but I guess we’ll see.

CONTINUE THE DISCUSSION/KEEP TABS

There are a number of places where discussion and planning are taking place. It’s not specifically about FPOTF, but definitely related and will probably be of interest.

post-tumblr-fandom Dreamwidth community (broad-based discussion, moderated)

distributed_fandom Discord (focuses on modifying existing federated open-source software for fandom use; lurkers welcome)

fandom-after Tumblr (commentary and snark about The Purge)

I’ve also created a Trello board (read-only) to keep track of things that I know are going on.

A TEMPERING OF EXPECTATIONS

I cannot stress how incredibly theoretical this is at the moment. Maciej has seen the doc, but he has other obligations (Pinboard, among other things) that he has to tend to. If/when he could get to it, it does not seem like it’s going to be at the top of his list, and that’s fine! +5

There does appear to be interest though, and that’s really great. I know just enough to know how much I do not know, and this is a thing that would require many, many people and a lot of money. At the very least, we need a product manager and a project manager, programmers, DBAs, QA testers, UI/UX designers, a documentation person, customer support, moderators. Who are not going to work for free, nor would I expect them to. If you have experience scoping these sorts of things, I’d love to hear from you (email me at pearwaldorf at gmail).

SPECIFIC POPULATION FEEDBACK

I’d like to get input from sex workers who are/were on Tumblr or considered Tumblr as a platform. This is a place for you too, and I want to make sure it meets your needs. If you know anybody who fits this description, please send them here. (Note: you do not need to identify yourself as a sex worker in the doc to give feedback.)

Not to shuffle off responsibility or anything, but there are platforms that exist now that are aware of the needs of sex workers and designed specifically for them: Assembly Four has some apps, there is also Sharesome. I don’t know anything about them, but if you know people who have need of services like this, let them know. +1

If you think there are certain segments of the population we should be soliciting feedback from, please let me know.

HOUSEKEEPING NEEDS

  1. Lots of stray +1s added. Somebody needs to consolidate them. [I think I went through all the stray +1s up until 15:55 UTC, December 6] Thank you!
  2. Announce a cut-off time for up/downvoting. It’s becoming clear tracking these numbers in a doc is unsustainable.
  3. After that’s done, I need someone to put features into an app/survey that can support up/downvotes. Ideally in categories (such as below) and with a place to discuss said feature. Are you interested? Email me at pearwaldorf at gmail or DM @pearwaldorf

NEXT STEPS

I’m putting a stop to up/downvoting at 6pm Pacific time (time converter). I will switch the doc into view-only mode. Fear not, you will still get a chance to put in your thumbs up/down when we get all this migrated into a format that is more suited to this than a Gdoc.

DONE. Thank you so much for your contributions everybody. It’s been a delight and amazement watching you come together and hash and design and discuss.

Things are going to go quiet for a little bit, because I’m going to start separating out must-have at launch features from things we can add later. Future suggested features will be migrated to a more appropriate forum for up/downvotes and discussion.

VOTING AND SUGGESTIONS

Please +1/+N things Only add +1 at a time, regardless of how strongly you feel! Do not add arbitrary numbers that aren't +1 or -1!

  • The votes can ultimately be used to identify what features are in highest demand.
  • Negative votes can be added at the end of the count (+5/-2) please do not subtract from the number.
  • H’okay. Due to somebody changing the numbers in ways they weren’t supposed to, I had to revert back to a version from about an hour ago (as of 12/5 2:50pm Pacific). Apologies if you lost work, but I wasn’t going to manually compare all those likes. Thank you for your patience.
    Also, don’t be a jerk. It’s not nice. +
    15
  • DO NOT USE COMMENTS. They are hard to see. If you have something to say, put it in-line in the doc.

Please do a Ctrl+F at the very least to see if somebody’s already mentioned the feature you want to add. If they have, go ahead and +1 it. If not, please add your suggestion to the Other/Awaiting Categorization section.

Account Management

  • Accounts not tied to user’s RL identity - no “real name” policies. +2112
  • That said, some kind of verification sufficient to stop people endlessly creating accounts for spamming/harassment/etc seems necessary. +46 /-2 for the reasons listed out below
  • I would potentially be okay with verified identities if they were secret and not openly tied to your account like on Facebook. But I’d also be okay with not doing this. I could see it for age verification/anti-spam, but for absolutely nothing else. -20 (-5 because Facebook, can we keep Facebook out of our lives please?) (hey sorry for unclear wording, I meant “do NOT do what Facebook does,” not “use Facebook to implement this.” Please don’t use Facebook to implement anything.)
  • How would you even verify? +3
  • Disagree with implementations of this that would make it difficult to create multiple accounts for separate pseudonymous identities, e.g. one fandom identity and one RL identity that should never be linked. +17
  • (the RPers will want multiple accounts probably) +8
  • We seem to be leaning towards one master account that can’t be viewed by others that can control many “sideblog”-type accounts (that can all do everything you’d expect an account to be able to do) which would be ideal for RPers. +8
  • Also a concern with data security. If it's linked, it can be mined. +21
  • It was my own suggestion and I +1’d this. I agree with you, I’m just anxious in advance about potential for abuse if it gets huge. Twitter is such a cautionary tale.
  • I’m not an RP’er but they ideally need the ability to create ‘in character’ accounts etc. One of the big things that social media platforms offer for role playing is the ability to create a space where a community of shared in character accounts interact. As long as those activities don’t spill into general tags etc. I think it should be something people can do and depends on the interface the site provides. +4
  • Typically on other fandom platforms, RPs create an ad post for their RP, then when interested RPers contact them from it, take their RP off to the side, to a dedicated RP community or interacting mainly with each other from RP blogs. You might have to scroll past ad posts, but not past all the RPing and replies to each other.
  • Also not an RPer, but I sometimes write smut and would like the ability to keep it separate from my other writing so it doesn't get in front of eyes it shouldn't. +2
  • An ability to flag individual posts as NSFW should already keep it away from anyone who’s a minor/hasn’t opted into smut. However, having multiple accounts seems like a popular feature anyway, and lots of writers maintain different profiles for different types of work just for organization. +2

  • A “display name” separate from an “@ name” so that we can have a joke display name like on twitter but keep the “@ name” the same. +78 (-9 on joke names.)
  • Ability to view feed without display names, only usernames. Long, changing display names (and name emoji) can be an accessibility issue. +70
  • Ability to customize how we view names, so we can select to see usernames, display names, or both. +2039
  • Ability to nickname other users in ways only we can see? (Like on Skype??) +194/-10
  • I feel like this is logistically unsound considering every user could possibly nickname every other user. Limit number of nicknames and/or only be able to nickname mutuals? (nicknaming limited to mutuals +2)
  • Reddit has flair, or the ability to “tag” usernames, essentially displaying a nickname beside the username. You also have the option to toggle them on/off. Would this be something people want? +5
  • Reddit also has a userscript that’s commonly employed to leave tags on usernames you want to remember you got in an argument with, or whatever. I’d be personally fine leaving private nicknames for other users to the optional world of userscript land +6
  • Discord allows you to leave notes (that only show up for you) on people’s profiles +8
  • LiveJournal/Dreamwidth permits leaving yourself a comment on/about a user; it’s a useful feature. This might be a more general alternative to “create a tag or private nickname”. +10
  • Limit the number of emoji in a display name for accessibility reasons.  +18/-4
  • Could we accomplish this with a gentle reminder in the interface? +5/-6
  • This could also be useful for indicating what people should refer to you as so you have something to be called other than “donutlover456” if that winds up being your account name (“taylor” as a display etc.) +2

  • A “pronouns” field
  • Viewable if you hover over a name/icon. +5
  • Hovering doesn’t really work on mobile sites or touch screen computers if you’re using your finger. +10
  • Hovering doesn’t work, but you can do a long press and get a popup
  • Possibly something similar to flair on Reddit? Where it’s a separate thing that shows up after your name. +4 /-1
  • I support it being more prominent than buried in a freeform bio for users who would benefit from it, but a constantly-visible flair edges into social pressure to disclose. +12
  • Have hovering as an option, otherwise visible (toggleable? +2) +7
  • If there’s going to be a pronoun field, it has to be opt-in. Many people strongly prefer to leave their gender unspecified, and no, gender-neutral pronouns don’t quite count. +51 +1
  • Strongly agree. I’m nonbinary and my pronouns are it’s/complicated or leave/me/alone. But I actually suggested the feature because I know others would benefit from this.
  • Ability to specify/leave blank per pseud +6
  • Does this need to be a specific feature? With other platforms, people just put their pronouns in their bio/description/profile information if they want to or leave it out if they don't. +26/-2
  • Specific pronouns field encourages use and self-identification +4 / -3
  • See above re: “opt-in” and “many people prefer to leave their gender unspecified”
  • Please never “encourage” me to disclose personal information I’m uncomfortable with disclosing. Disclose what you want and don’t judge. +5
  • Have options for he / her / it / they as well as ‘input your own’ for neopronouns?
  • This is culturally specific and also language-specific. Billions of people use no-pronoun languages to communicate on the internet.
  • You’re right, but lots of this is culturally specific and language-specific. We’re designing a site for English-language fandom needs at the moment. Anyone who finds it irrelevant for any reason can simply leave it blank. -2
  • Fair I just get hot under the collar about American Social Justice Is Culturally Universal. “Blank/Optional” is good enough
  • Imo software ought to be designed with localization in mind from the beginning. It’s one of those things that’s hard to retrofit later.
  • It’s easier to check at a glance than go wading through someone’s entire life story, all their ships, and their allergy to mangoes. But this area could also contain other very brief need-to-know information, like previous usernames, or what they prefer to be called (e.g. “Allison, she/her.”) Perhaps nickname, pronouns, and previous username(s) could all be optional fields in this area. +75
  • (sort of a Mad Libs feature might be good - “My name is X; my pronouns are Y; hobbies are Z” etc and anything you don’t fill in is just left out of the end display) +8 /-3
  • I don’t want hobbies there, that’s regular profile/bio material. It’s not for getting people’s life story or deciding if you want to be friends, just “who is this person and how do I refer to them?” +2
  • What about just an open field with a character limit and hint text +3


  • Bio/Description/Profile per account/pseud +6 and single central identity (if there is one) +12
  • “Username history” if renaming is allowed +22/-2
  • Either optional, or optional beyond a certain time frame. (eBay requires it for one month.) +7
  • I dunno, I see no downside to a username history and plenty of upsides. Keeps the “everyone on my dash changed their name and I have no idea who anyone is” thing from happening. +2
  • Stalkers, mostly. But an option to hide your username history when you make a change might help. Just warn them that it might confuse people etc etc. +4
  • This should definitely be a feature but it should be 100% optional with no mandatory time frame. In case of issues with harassment. +7

  • Option for multiple accounts and easy switching. +156 +1
  • Or at least something like tumblr’s sideblog feature +26
  • Would prefer sideblogs over multiple accounts so people couldn’t make bots as easily +14 +1
  • But only if there are no features a new account allows you that a sideblog doesn’t. On tumblr, sideblogs can’t follow, send asks, or “like” posts. +7
  • If free accounts have data caps, people will make multiple accounts as a workaround. This isn’t really ideal for anyone. +5
  • All sides share the same cap then maybe?
  • I mean that they will create a new “root” or “master” account, like with a new email or whatever. All sides sharing the same cap would just encourage people to make a new root account for every side, instead of actually using the side feature.
  • Not to be a “criminals break laws” person, but if people want more room bad enough to make two full accounts they’re gonna do it regardless of caps.
  • Re easy switching, needs to be as simple as “post as:” and a dropdown; no log out/log in shenanigans. +10

  • Reply, like, send messages as any of your blog identities. +92

  • Consider the option of having a single central identity (which can’t post or read, but only exists to manage one or more aliases / aspects / views - which are the equivalent of “accounts” on other platforms) +35

  • OpenID or similar for cross-platform commenting? +11/-14
  • Can we also have optional fields in the profile that encourage us to cross-link our usernames and identify who we are on other social media platforms, e.g. AO3/dreamwidth/twitter etc? +14
  • This would be a good candidate for a sort of “quick bio” or “identity” field, that also includes previous usernames, and optionally, what you wish to be addressed as, as in nickname/pronouns. Like something you could pull up quickly to answer, “Who is this person? Do I know them from somewhere? How do I address them?”
  • I would love if we could work with Dreamwidth to achieve a certain level of interoperability esp since they’re the only other ‘fan owned’ online community. OpenID, and two way RSS support would be a nice start +5
  • Can I get an explanation of why OpenID is contentious from anyone who has -1’d? Just for curiosity. +2
  • I was replying to this before when some edits got reverted due to trolling yesterday. Anyway, I didn’t +1 or -1 it, because I’m ambivalent, but I can see potential problems. One is that commenters from offsite might not, as we used to say, “lurk moar,” but simply charge into conversations without observing local cultural norms. Fandom can be insular and not entirely eager to deal with a flood of comments on a post that suddenly went viral from people with Twitter IDs who are all kinda rude. It might lower the investment required to engage even lower than the simple bar of “make an account,” encouraging flyby interactions from people who don’t add to the community overall. People may be wary of crossposts, like a lot of people find crossposting from Instagram and Twitter obnoxious, because it’s content that wasn’t “addressed” to them, and the OP may not even be logging in or monitoring notes? +10
  • Also, OpenID has the chance of nullifying a few other suggestions that I like elsewhere in the thread. Basically - openness is a great ideal there may be tradeoffs between that and insulating/protecting ourselves from the rest of the internet. +5

  • "Gated" communities requiring 18+ but accounts can be made by anyone (similar to reddit's new nsfw subreddit policies) +36
  • At least paying lip service to NSFW not being for people under 18 is probably necessary, even if the site doesn’t ask your age. (AO3 does not verify or require age disclosure.)
  • It needs to be possible to change it later on. There are all sorts of issues with it being set with what you signed up with.
  • At least warn people that minors can’t use the site in certain ways, so RPers don’t end up with broken accounts. Don’t show anyone’s birthday anyway so RPers have no reason to do that in character. +5
  • Birthday display (M/D or M/D/Y) should be user choice. Gift giving’s a pretty big thing in some sections, and some may be shy about announcing their birthday. (also, letting you know about upcoming birthdays is nice :) ) +10
  • If we’re doing an invisible root account with multiple sides that can each have a bio, maybe birthday *display* would be a separate thing per-side, while “how old are you REALLY” is a hidden part of your root account? That way, RPers can have fake birthdays, or people can put their birthday on their NSFW account so everyone knows they’re over 18, but not on their other accounts where it’s irrelevant. But I don’t want adults pretending to be minors for realsies, so maybe there needs to be a kind of RP flag that basically says “this account is fictional” or something? (Also useful for say, political and celebrity roleplay--you can roleplay Barack Obama but not actually impersonate him.) Also, while I agree with month/day for gift-giving, year-only should be possible for people who want to use it as a declaration of age *without* disclosing their actual birthday and getting bothered if they find gifts and messages bothersome. (+8 for flagging fictional accounts)

  • Stable links (if you change your username, you have the choice to redirect to new handle or break links.) +84/-1  
  • Since constant redirects can be expensive (some people really like to change their usernames all the time), maybe put a slight hurdle like in the old LJ days. (payment? time limitation? one free change and then payment thereafter?) +63/-1
  • It needs to be a choice and not automatic in case someone needs to disappear from the net. +45
  • Ability to orphan specific posts or entire blogs - change author to "orphan" and decouple links to original author? +35/-2

User Interface & Accessibility

  • Built-in accessibility from the ground up
  • Image descriptions, ability to mark images as decorative (AKA part of design template) +59
  • Ability to disable gifs and video autoplay. +897 (?!)
  • As two separate options if possible +51
  • Also ability to disable audio-only autoplay, if people can add that to their posts or themes. +24
  • As a possible alternative: no video or audio autoplay at all, instead of an opt-out. +23/-4
  • Neutral on this for audio/video, but very against this for gifs. Gif autoplay is essential at least as an opt-in. +9
  • Or only as an opt in +2 (this seems the same as the above suggestion?)
  • Distinction is the default setting being on or off. +1
  • Muting video autoplay is also something people might want to choose. +8 Autoplaying video with sound is much worse than autoplaying video. Autoplaying video can be nice sometimes if it’s quiet, like a huge gif. Though so expensive.
  • Ability to toggle some kind of “plain” skin when you visit someone else’s blog page to make it more readable/accessible? I miss LJ’s thing where you could add “=?format=light” on the end of the blog url and it would clear the user’s theme to plain text on white background, which was great for themes with like gray text on beige backgrounds or sparkle mousepointers or sites with sidebar frames that wouldn’t load right.  → yeah: i've got a similar suggestion under responsive design that suggests saving profiles/skins/settings so that you can switch when you head into work, or are out having only mobile data +20
  • “Plain” skin needs to at least be WCAG compliant (Web accessibility guidelines: https://www.w3.org/WAI/standards-guidelines/wcag/), preferably WebAIM guidelines. +3
  • Would prefer this to be something I can toggle for an always thing, and not have to do it for every single blog I visit if I have issues for accessibility reasons, but would also like to be able to do it case by case like LJ because some people think Dark Blue on a black background is a good idea. +5
  • Yeah: my idea is that it could be associated to device, but in addition to that I'd like a "switch views" button where if someone likes yellow text on a grey background and uses a zillion images you could go "... I'm just gonna view this on my mobile view" and not load the images and have your default light screen or whatever but still see NSFW stuff. +2
  • Ability to toggle a theme to invert all colours, for those of us who enjoy having bright white themes (looking at you, theme-hunter on tumblr) but also for those who need automatic dark modes, or vice versa
  • I wouldn’t necessarily invert the colours, but just have a standardised ‘dark mode’ skin/theme
  • I have a visual impairment (cataracts) and anything-on-black  is unreadable for me.  Please build in a standard way to recolor a page to black-on-white, with no background pattern; the "Zap Colors" bookmarklet is working less and less well.
  • I believe - please delete this if I’m wrong - that there are browser extensions that do this automatically, such as the High Contrast extension for Chrome.
  • There are browsers that have a “reader view” (eg Vivaldi) that strips stuff from the page and leaves it black text on white, but it needs to be supported by the webpage itself. So I would say barebones support for this is a minimum.

  • Ability to switch between infinite scroll/pagination. +129
  • Paginated feed URL always leads to same content (tumblr does this well, as opposed to DW’s ?skip=20) (Tumblr uses a Unix time stamp in the URL to do this.) +29
  • Side note: Very important for low/limited network connectivity users +15
  • Global setting/ability to override a blog’s infinite scroll setting would be nice, too +21
  • I don’t really feel like having infinite scroll is necessary, personally. Are there people who would hate having only pagination? (+1 to pure pagination)
  • I dislike it only because my hardware is outdated and can’t handle it, but on good hardware it is “frictionless.” Whether that’s a good or bad thing I’m not entirely sure; it makes you spend more time on the content without realizing, which is good if the site wants your attention, bad if you hadn’t meant to spend three hours just then. It seems to be more of a way to trick or entice users into deeper engagement than something users actually ask for or think they want. (+3 infinite scroll is a way to trick users, not desirable for a fandom-friendly site)
  • I like infinite scroll on a dashboard but on individual blogs it always ends up killing my browser, so ability to swap there would be nice. +3errrrrrJ(pHYJp²
  • If infinite scroll is easy to implement, we should have it as an option for the people who like it. If it has significant costs associated, then eh, chuck it - it’s not that important

  • Some way to ping/@ other users and have them notified. +105
  • A way to block-in-advance pings (limit pings to people you follow, people who follow you, mutuals, or none at all.) +66/-1
  • Something similar to Twitter where you can block notifications from people you don't follow, but it will still show up in your mentions would be helpful. +21
  • A way to be pinged under multiple different names or accounts. Maybe the ability to make a namelist and use of any of the things on the list to try and contact you  will ping you. +1
  • If the secret “root” account has a master activity page, and if you can sort your activity by type (all reblogs, all comments, all follows, etc) then you could simply check your root activity and sort by pings to see if anyone had pinged any of your accounts.

  • Native mobile/responsive interface from launch. +39
  • Fully functional on both iOS and Android. +14
  • Strong, specific notifications controls for mobile. +18
  • Relatively low/”old” spec requirement for phone hardware. +25
  • Tested and functional in all major web browsers +43
  • Ability for users to switch between skins/views/settings easily: both have them saved (and easy to switch to) or associated with a device. This would include night mode, "at work", "using mobile data so no pictures" etc. +7 Easy access to night mode (aka, dark theme)(covered in the above point, I think) [also mentioned under accessibility/toggle “plain” skin]

  • Alternatively, possibly paid ability to skin the site, a la AO3. +28/-4
  • See Revenue & Income (survivability)
  • Skinning the site is a way for people to change the site to fit their accessibility needs [as mentioned further up]. After all, a site should be accessibility focused but it can't perfectly fit everyone's needs, but allowing people to code how it displays things or what things not to display gives them more recourse than just asking the site admins/coders to implement something. I don't think it should be put behind a paywall.  +58
  • Putting skins behind a paywall wouldn't work, anyway, since people would just use client-side tools like Stylish to get around it. +1
  • Could just be some extra-fancy skins that are “premium content”? +12
  • This could be a good way to make some money (pretty layouts without having to do the work yourself) without disadvantaging anyone who needs skins for accessibility.
  • I think we should distinguish between skins, which alter the interface of the site when you browse your dash/feed, tags, communities, etc, and layouts/themes, which personalize your blog and let you sparkle it up in ways other users will see when they visit your blog. Skins can sometimes be an accessibility issue, layouts are fluff. I see no problem with charging for layouts, at least. And a couple of choices of accessible skins to meet different needs on a free tier could make skins that go beyond that fluff. Though as others have said, Stylish is an option, which someone could use if they really needed a change for accessibility and didn’t have money to support the site. I don’t have a problem with people doing that. +1 (for free site skins, possibly paid layouts)

  • Ability to change view to hide all images, hide all video, resize all images above a certain size.
  • This could be merged with the “replace all images with their image descriptions” accessibility option, and might be more coherent than simply removing images. +21/-1

  • Activity feed showing likes/reblogs/etc., with toggle to include all activity or just activity with additions (see the xkit feature for an idea of how this might work) +90
  • Multiple activity feeds for each sideblog/pseud, and one big account-wide one that shows all your activity in one spot. +30
  • Ability to mute notifications about a specific post of yours from your activity +55

  • Ability to view archives/posts by date - and by tag +43
  • Tumblr's archive view is more navigable than LJ's was (small previews of posts sorted by month rather than a calendar with dates highlighted) +10
  • Ability to view archive by original post and/or reblogs (similar to xkit find original posts, yes please!) +6

  • Flags for things like jump scares and strobes, and, if there’s advertising on the platform, a hard line against using that sort of thing in ads. +57
  • Mastodon has an interesting approach to CWs that might be useful here +6
  • Mentioning a site that does something well is good, but please elaborate when you do for people who don’t use that site!
  • Mastodon has a strongly-encouraged option to add content warnings to a post, which is visible as an eye symbol on the top info bar of the post. Clicking it reveals the content warnings. This enables people to opt in/out of CWs per post. There is also an option to always see the CWs on a post.
  • This is incorrect. Using a content warning on Mastodon hides the entire post (“toot”, aka tweet) behind the warning (you decide what the warning says). It’s very easy - one click/tap to see the toot, one click/tap to hide it. It’s very easy to use, but if you are a hardcore Choose Not To Warn type, this option would need to be more flexible than the version provided by Madtodon.
  • Couldn’t this be handled with blacklist? Or disabling gifs/videos? (is there a way this could be done automatically?)
  • Not really? Disabling all gifs/videos means that you never get to see anything because clicking on anything is a crapshoot, and re: blacklist, at least on some parts of the internet, this kind of thing shows up as a reaction gif, so blacklisting for strobe means you miss the whole post.
  • For strobing lights in particular, that flag could have the separate option to remove autoplay. So in your settings: “Autoplay gifs? yes/no, autoplay gifs with flashing/strobing lights? yes/no, autoplay gifs with jump scares? yes/no.” A single frame can’t strobe at you or jumpscare.
  • Actually, this might be more useful than “warning” for such things. Strobing lights and jumpscares aren’t something you want to hide entire posts for, you just want it to not autoplay, and know that’s why it’s frozen before you unfreeze it. A banner along the top of the frozen video/gif that says “Strobing lights” is plenty. +1
  • Maybe a bit crazy, but it might be cool to be able to apply some universal tags (Flashing lights, violence, injury, etc) to other people’s posts. A content creator might not realize that their post is ‘strobey,’ and if one of their users can tag that for them, that saves everyone some trouble. However, there is potential for harassment, too. +1 / -4
  • Ooh. I’m just seeing so many posts labelled as “q-slur” by well meaning viewers. I’d go with ‘ask creator to tag’ here. +13
  • Might it be possible to have a small collection of tags (thinking like the major warnings on AO3) which can be universally added, with anything else limited to ‘ask creator to tag’? Still a potential harassment point around what tags are universal. -1
  • I think that might still be too open to abuse? I mean, most people tag anyway (and if they don’t, consider unfollowing), but if they deliberately didn’t tag for those few mandatory tags they’re not going to be happy about people adding it on after (in fact, I’d suspect they’d be downright pissy)
  • There are a lot of suggestions here that amount to “hardcoding having good manners”. I think a central switch on an account that disables moving gifs or other autoplay of any kind is hard-codeable: making someone else tag accessibility is not. +3
  • OP here, I didn’t mean “users tag posts for creator” so much as one universal tag for a SELECT few things. I don’t have any hard triggers, but everyone on my list has a different way of tagging triggers: ‘tw gore’ ‘tw - blood’ ‘trigger - gory’ ‘gore mention’’ etc. I’m proposing a universal “one gore tag to rule them all” so that people who don’t wanna see that can hide it, and so that a user doesn’t have to play “guess the trigger tag’ when they find a new person. There is some potential for harassment though, that is true-- but I was thinking of ‘broad’ tags, not specific ones. :) +2
  • Like AO3’s archive warnings, or like the concept of canonical tags. I was working on the assumption of this being a good way to handle it already.
  • Maybe an ability to take it to the abuse team to allow for things like strobing lights that could cause harm in case of creator refusal? Allowing other users to directly tag a post seems way too abusable
  • But that said there was some discussion on AO3 of using an optional ability to view the tag cloud applied by readers for this as well as its many other uses… however, that makes more sense with fanworks, where the presumption is the ability to bookmark and organize those bookmarks for search, than with people’s blogs. (I would strikethrough this whole bullet point if I could find it in the interface now, sry)
  • I’d rather have an AO3-like system with enforceable platform warnings + CNTW. If you see something that has such content and wasn’t warned/CNTW’d, you can report it and it’ll be changed to CNTW. I’m not completely sure what types of platform warnings we’d need--the fact that it’s a blogging platform rather than a fiction platform changes things. I think there needs to be a distinction between, say, Hannibal fanart and graphic war crime footage. Whether the sensitive content is real or fictional matters. This could apply to NSFW as well--fictional NSFW is stuff like art and fic, real NSFW is stuff like IRL porn. (Spoilers count as a “fictional” warning.) Visual triggers like flashing/strobing lights and jumpscares are a separate category. I don’t want to get too granular here, but I’d like if CNTW was divided into “real content,” “fictional content,” and “visual triggers.” I can click and take my chances if I know a fic might contain graphic fictional content, but still want to know if there will be embedded jumpscare gifs or a graphic personal recounting of sexual assault in the A/N. +4
  • +4 for ability to mark post as fictional/non-fictional. This would help distinguish between fictional violence for fun and drama (i.e., whump) and real-world traumatic events (i.e., news/history).

  • Ability to disable emoji for people who use screen readers. +30
  • Emojis are harder on screen readers when there are long strings of them. Perhaps an accessibility option to limit emojis to N in a row, or N per post? One screen-read emoji is tolerable, 30 repetitions of the same emoji is awful. +4
  • If this isn't implemented, maybe at least condense repeat emojis, like it'd say "heart emoji 13 times" instead of actually saying the emoji thirteen times +8
  • Idk if it’s a screenreader thing or a setting thing, but instead of going ‘heart emoji smile emoji double heart emoji’ (you know the ones, with 30+ different emojis at once), go ‘emojis: heart, smile, double heart three times, etc etc’ +3

Content Presentation

  • Chronological timeline: content displayed in reverse chronological order +427

This is very popular - please +1 to above and not add separately :)

  • Can we bookmark our place in the timeline, to come back to it later? URLs with Unix timestamps would basically do that, but it being part of the site would allow it across devices without having those devices synced, and also help you if you had a crash and lost your tabs. +28
  • I think that there’s an interesting challenge because fandom needs are both for a scrolling chronological timeline that is essentially ephemeral but we also have needs to be able to create archival hubs, some small ones but with the ability to scale up too. Recovering and exploring old content on either Reddit or Tumblr, for example, is a massive pain that almost fights the interface given. So it needs both the ability to have a chronological timeline and to self curate a non-chronological space, with tools to aid that if possible. +3
  • Robust search tools are essential to this end re: finding older stuff. Tumblr’s tag search becomes a black hole if you try to look farther back than a couple months +14
  • I don’t know how it is post-purge, and tumblr’s search is very broken about the past (possibly because the search feature itself was a later addition, and it didn’t scan back that far past its implementation?) but tags never die, I can use a Unix timestamp to see what was going on in a tag in 2012. The only difference is tumblr blogs delete so often a lot of content that was there then is gone now, but that isn’t the platform design’s fault. It isn’t terrible useful though, because hardly anyone goes, “Show me everything that was posted in the tag on March 1st five years ago.” Search, and the ability to browse multiple tags at once (e.g. not just #finnrey, but #finnrey + #fanart + #au) would be more archival. As would communities in general. +2
  • Enable Boolean search terms, e.g.: (stormpilot OR finn/poe OR finn&poe) AND fanart +13
  • Me, chanting, a la the TEETH TEETH TEETH tweet: BOOLEAN SEARCH! BOOLEAN SEARCH!

  • Ability to host images, audio, video +512
  • This seems like the most necessary element -- it’s what’s missing from Dreamwidth (nb DW does have 500MB of image hosting, with talk of providing more for purchase.) +3
  • Prioritize image hosting over audio/video to start, but with options to embed from other sources like Youtube/Spotify/etc. +44
  • I agree that as a practical matter images might need to be prioritized but the hosting of new audio is needed. Things like podfic, discussion almost like podcasts and so on.
  • I’m not a podficcer or podcaster, but are there really no platforms that host that? Discussion podcasts in general are some of the easiest fanworks to host from a copyright perspective, and podcasts as a medium are so popular now I’m genuinely surprised that demographic is being failed. I agree if you don’t have anything you do need something, though!
  • There is an audiofic archive but in 2016 an enormous amount of the archived podfic was lost during a server move
  • Ability to post images with an individual Sensitivity Screen (“WARNING: Click to View”) for each NSFW image (nudity, animal abuse, sexual content, violent news photos etc) so that they can be included alongside SFW images in a set +63
  • Can use warning categories from sites like doesthedogdie.com (includes specific categories like suicide, eye gore, clowns, etc.)
  • We’re already super granular, but some of that stuff might need to be freeform. Like, “graphic violence” is a broad enough category, dividing that into fictional vs. real is reasonable, but things like “eye gore” or “hand trauma” or “cannibalism” you can work out with tags. I actually do want more granular than AO3, but I can’t deal with a page with 50 warning checkboxes. One thing that might be helpful though is that when Xkit’s blacklist blocks a post for me, I can still read the tags on the post, and use those tags to guess the content. When tumblr’s native blacklist blocks a post, I see nothing but the term it was blocked for, which makes it much less useful. If I block the “graphic violence” flag, it makes a difference if I can see that OP’s tags (or marginalia, or something) say something like, “warning just to be sure, a little blood,” or “clowns, cannibalism, eye gore.” I can make a much more informed decision about where I click than if they both just say, “blocked for containing graphic violence” with no other clues. I also want to be able to see who posted it, because that’s another clue. Something like that, where people have the chance to elaborate on broader categories, but only broad categories are official platform warnings, might work. +2
  • Also make it possible to warn for content (both images and text) for spoilers with the same “click to view” you would for triggers/sensitive content. +25
  • Maybe we could use different colours/formats to flag separate warnings? I.e. one style for spoiler content, one style for violent imagery, one style for nudity, etc -- to make it easy to identify and allow users to decide what they want to click on? +8/-3
  • Having differentiating colours/formats raises accessibility issues - it needs to be something that isn't just visual. +2
  • If we’re gonna colorcode things we need shapes or symbols with them too, don’t forget the colorblind people, they won’t like opening what they think is NSFW and is actually a spoiler. +6
  • Could use symbols like AO3?
  • Simply allow uploader to say why they’re putting an image behind a sensitivity screen (Hidden for NSFW/Hidden for Gore/Hidden for female presenting nipples) +4 / -1
  • I might be okay with adding freeform and optional areas for extra warnings if people really want, but the idea is that certain warnings, like on AO3, should be mod-enforceable as either warn or flag CNTW. Plus the presence of those warnings avoids people just “forgetting” what kinds of content you’re supposed to warn/CNTW for. And it would also allow a permanent opt-in in settings, like “never warn me for violence.” That makes warnings you don’t need not constantly interfere with your browsing and annoy people, while giving those who do need them access to them. Really, I’d rather just have extra warnings put in freeform tags on a per-post level. If you really need to warn for penguins, just tag the whole post with penguins. Also please let us disable freeform warnings appearing as we browse if they’re implemented so I can look at fanart that contains representations of eyes without having to click through a scopophobia warning. +2
  • Reddit distinguishes between “NSFW” and “NSFL” for saucy vs violent content iirc, and as far as global filtering goes I think it’s reasonably granular especially if it’s coupled with a tag system. Might add Spoiler to the list but more than 3 could get clunky. Even AO3 maxes out at 4 in the form of Archive Warnings
  • This would be easier done in settings? Just tick a list of things you want flagged +1
  • I agree with opting into what you do and don’t want flagged, but I want to know which warning I’m waiving each time I waive it, of the ones I asked to be warned for. +5
  • Show colors / shape styles / etc, allow to pair certain styles / colours with certain parts (etc, I can set ‘blue rectangle’ to nsfw and ‘green oval’ to spoilers, and someone else can use red or hearts or smth.) +2
  • Just use text. +5/-1
  • Click-through spoiler warnings that automatically expire after a set amount of time. You check a Spoiler Warning tickybox, input the name/installment/whatever of the piece of media you’re spoiling in the designated field, and select up to 6 days or 4 weeks or 12 months from a drop down menu. That, or you select a specific date. (Any more than a year feels like overkill though, and would lead to disproportionately inconvienced users if the poster doesn’t manually remove the warning at some point.) +1
  • Better still, can you set the date the content being spoiled for came out? So you click spoilers, select that it’s for today’s content, 12/6/18. Someone else sees the post and they know they’re not caught up yet since the warning shows the date, so they skip it. Then someone sees it a week later--they haven’t seen the newest content, but they go “oh 12/6, I’ve seen that one” and click. Useful too with people just getting into older canons. Maybe they’re not ready for that thing that came out in 1971 yet! +3
  • This would also help with the different worldwide release dates for major films

  • Ease of posting for images comparable to tumblr, capping of size is ok but please no uncontrollable compression and resizing. +71
    (Please if you love fanart and gifsets vote for this.)
  • Decide what post sizing we’re using and stick to it. Remember when tumblr went from 500px posts to 543px, resized everyone’s perfect gifsets, and nobody stopped being angry for over a month? Don’t do that. +5
  • I don’t mind the full view allowing for more potential, it’s the smaller “dash/TL view” that shouldn’t be resized. That’s what actually affects browsing the most as you scroll. No one complained when tumblr allowed gifs to be 2 GB rather than 1 GB, as the old 1 GB gifs weren’t affected in any way. Allowing bigger full view sizes doesn’t resize old stuff and is helpful. Messing with the scale of dash view damages older works. +3
  • Oh god yeah, pick one and don’t change it. If you change something, up the file size cap, not the pixel size / margins.
  • Please be generous with file size caps--tumblr was the ‘best’ source for full resolution pictures for a long time. +2
  • Note: allowing full resolution also means allowing the highest use of storage space and thus server/hardware needs on the backend, aka going to cost a lot more money +1
  • Tumblr was generous, but not unlimited. It had limits both in image dimensions and file size. Several times I had to compress my own art to meet tumblr’s upload requirements. I was okay with that, though. I had control over how the art was compressed and got to tweak it before uploading. Tumblr did also sometimes damage art with compression/resizing a bit, it wasn’t perfect, but I think as long as most pictures look basically okay we won’t complain. It doesn’t need to be ultra high-res unlimited size, just good enough to be usable. But good enough to be usable is extremely important. (This is similar to how writers might tolerate a character limit on posts that requires them to rework chapter sizes a bit before uploading, as long as most chapters fit most of the time, but not a system that truncates or removes chunks of text silently to make it fit without giving them any control, or an excessively tight character limit that constrains most fics.)
  • As there are other sites that allow image hosting, this might not be an immediate priority - if there were, for example, an easy way to post photos that are stored in Google Drive, that would relieve some of the pressure on the site +1 / -2
  • No, actually, there aren’t any good services that meet fandom’s needs for image hosting, especially wrt having our back on explicit content. If it was so easy to just use other services for hosting, fans would have made a platform for it ages ago, or AO3 would be sufficient. We need something that tumblr did and AO3 doesn’t.
  • Stacked/slide-show image option (as in Instagram) to show progress/WIP shots for artists without clogging the newsfeed +25/-5
  • Not only would we artists use this for its intended purpose, I just know we’d do creative things with it! +6
  • I voted for it and wrote the above, but also FYI if this coexists with reblogging and slideshows can be updated after a reblog and the reblog will show the update, that could be abused. However, I think it almost never will be. No one goes to all the trouble of working hard on art, building a blog and a following and getting people to trust them, just to one day sneak a shock image in and destroy their own reputation and get reported/blocked by everyone. One could just as easily make an entire blog just to one day shock your followers. It’s far too much effort for trolls, I think it will be used positively.
  • I can see this happening with "20 pictures of cute cats!" or stuff like that, though. I'm glad you brought this up, I would never have considered that abuse potential +1
  • Slide-shows aren’t great for alt-text and therefore for accessibility. If we do slide-shows we have to figure out if the alt-text is for the slideshow as a whole or if it’s for individual images. Neither is perfect. +2
  • If image descriptions are handled better than simple alt text and are a full feature, we should be fine including slideshows.
  • Is it possible to add something like “photo 1: [alt text]. end photo 1.” between the alt text photos to demarcate the different photos of a slideshow?
  • I like this idea, but it should definitely be optional. +2
  • Easy, beautiful, and intuitive, no coding knowledge required +9
  • But HTML editing as an option +104
  • Maybe Markdown, too5/-1
  • CSS as well, for those of us who like to code layouts! +10
  • Ability to turn off custom layouts because they can be havoc for accessibility +12
  • Make custom layouts for thememakers an actual part of the settings, kind of like tumblr’s themegarden but way more submission-friendly, so you don’t have to dive through one thememaker’s tags in hopes of finding a different thememaker. +1 DW’s organization of custom themes is nice. I often wanted a way to modify the Tumblr dash- ie MY experience of the site, not necessarily (or in addition to) my blog. LJ-style your feed/flist/dash is skinned automatically, but in Tumblr the Stylish hacks to do it kept breaking and I gave up
  • Keep history of saved themes for a while, for those of us who accidentally delete something important and need to go back. Maybe, as a stretch goal, allow users to save specific themes and switch between them like site skins, for favourite theme settings?
  • This sounds like a decent paid feature? +7
  • Some amount of Javascript would be nice, too, if possible, for more complex/interactive themes
  • And not being blocked from embedding them on other platforms.+9 (-7 contributes to loss of control by artists)
  • However if this is a sticking point to getting something up and running, I think fandom is willing to forgo it for safety and a stable link at this point. (Other sites have a tendency to change the links at random so you can’t even direct people to your work.) +6
  • Hotlinking as a paid option? +14/-3
  • Let artists control if their work can be hotlinked. +25
  • Offer an embedding option (similar to how tweets are embedded) so artists can be linked offsite without losing credit/control. +25
  • Maybe auto-watermark pictures embedded on other sites?
  • Let artists opt into this, and maybe even have a separate watermark file they can make themselves as a transparent PNG or something to apply to all the works they check “watermark” on. I know some artists would love this, but personally I do not want a platform modifying or defacing my art without permission. Deviantart allows optional watermarking (though not your design of watermark, and for the main site rather than only when embedded to other sites). A lot of people use it even though I think it looks awful! +1 (for opt-in watermarking)
  • Provide a reasonable amount of hosting space with an account, but for those who want to host more/larger files, give them an option to pay to upgrade their storage space. +29/-5
  • The ability to view uploaded media in a sane and human-readable way, as well as downloading it, either item by item or en-masse (e.g. as a Zip)
  • Not sure default to downloadability is a great idea (e.g. for IP reasons), but perhaps as an option? +1
  • Should be uploader defined. E.G. if I upload a vid and don't want you to be able to download it, I should be able to control that option. +12/-2
  • I +1’d this, but jsyk nothing will stop a determined person from ripping anything. I download lots of video sites didn’t want me to. I don’t mind making it a little harder as a good-faith gesture to vidders, though, and to possibly appease copyright holders. (Vids have particular copyright issues fic and fanart doesn’t. Even though gifs are similar, gifs haven’t been enforced often while vids have.) +3
  • If pictures can be downloaded, have the filename and/or meta automatically include the username and site of the artist +10
    (in the meta at least if not in the file name, not everyone deletes the meta and that stuff stays so long as you don’t copy / paste the pixels themselves and make a new file - this is the best way to make sure that I can find the source of stolen artwork.)
  • Someone should start a section specifically for dealing with art theft and art reposting, but okay. Just as vids being downloaded is optional, some sites (I think flickr was one) offer the option of disabling right click + save, which some artists and photographers value. They even disable the screenshot key! Personally I don’t think that’s that great a feature, as where there’s a will there’s always a way, but as long as it’s optional I know it would be very attractive to a subset of artists. It could cut down at least on the constant casual reposting by people who aren’t even very savvy. One time my art was reposted while the original was still on the first page of the tag! You literally just had to scroll down a few posts to see my OP. +2
  • Meta tags can help with, “I got this art from somewhere, I don’t know where, I want to be nice and credit the artist but I don’t know who drew it,” but it won’t help with, “My art was passed around from Pinterest to WeHeartIt to Reddit to 9gag and 20 other sites and ten other reposts stolen from those sources all have more notes than my OP” problem.
  • As another suggestion for this problem, can the platform “remember” images that have already been uploaded to it? Like not similar, but the exact same image to the pixel. That way, when you try to repost art that’s already on the site, the site says that’s already here--would you like to reblog it instead? And if you catch someone reposting your art, either not caught by the system or from another site, if you can prove it’s your art you can “claim” the post--basically, you become the OP, and you can keep it up, edit it, or delete it. +1
  • Adding to this, because as I wrote it I wondered, “What happens if the artist deletes their OP?” I don’t want to make the image unpostable, especially since a very basic “is this literally the exact same image to the pixel” matching would be very easy to fool if you were trying to, by altering the art, which the original artist would likely hate even more. But perhaps the site remembers that it’s seen this image before in a deleted post, so it knows not to give you credit like you made it, but just sort of does nothing. That way, artists would be encouraged not to delete their art, because having their art maintained on the site actively prevents reposting and gives them more control. Currently, artists often delete as an attempt to staunch the flow of art theft, which is bad for everyone.
  • I understand the financial reasons for this, but as an artist, this is a showstopper. What happens to all your “extra” hosted art when you can’t afford your yearly fee, or even die? If you do pay for extra space, it needs to be a one-time expense. Fandom values permanence and archiving. I don’t mind paying, but I don’t want my stuff inevitably deleted when I stop paying for any reason. +48
  • I definitely respect the financial realities involved with multimedia hosting. If hosting space is going to be limited, it should be on a monthly basis, rather than a hard cap for the life of the account - say, 250MB per month for a free user, 4GB per month for a paid user (specific numbers aren’t important). A hard cap makes life hard for invested, long-term users, and that’s bad for the site culture and finances. +15
  • Vimeo does this, seems to work. There’s a weekly quota unless you pay. Needs to be large enough to handle short videos/gifs for animators if this is the route chosen.
  • But what is the more serious expense, hard drive space or bandwidth? It seems to me fandom’s traffic demands are what will hug the site to death, not storage space. And you absolutely can't cap bandwidth. +1 Fandom will turn on you if they become unviewable the moment they finally start to get popular. And BNFs will drive site traffic, popularity, and community. Making it so no one can view the most popular content on the site is platform suicide. Limiting server space won't stop the real gushing money hole, it'll only make content creators delete their old content so they can post more. This will just destroy archiving and make everything ephemeral, while not cutting bandwidth/expense much because the new stuff gets the most hits anyway.  +1
  • Make it so that you pay for uploading a certain amount of content per month. That way, if one month you can’t pay, nothing gets deleted, but you can’t upload over the free limit +6
  • And deleting old content won’t raise your cap! You don’t want to incentivize that. +3
  • Full compatibility with the embed code from major streaming sites, e.g. YouTube (no code-stripping/oEmbed.) +24
  • Ability to gift storage space to another user (if you’re not going to upload any media, you can add your cap to someone else’s) +18 +1
  • If this isn't supported natively and there are data caps, users will do this anyway by posting things for their friends. That will make attribution, control over content, plagiarism, and so on an absolute nightmare to moderate. +8
  • If video hosting is going to exist, in order to be useful for fanvidders, it needs to be fairly lax on copyright detection -- cf YouTube copyright strikes over music that make it difficult to host vids there, and have led to the loss of older vids that get taken down after years
  • Either it needs to not have any detection algorithms at all and only remove user-reported full piracy, or, and this is logistically a lot more complicated but one of the only ways I can think of to calm down the MPAA/RIAA, someone needs to approach those agencies and convince them to allow transformative works like vids with some compromise like letting them put ads on vids that use their IP, and letting them make money on those ads. So for example, you mark that your vid uses footage from Captain America and music from Evanescence, ads appear on it, and Marvel and Evanescence make the profit. It seems like the only way they’ll ever be convinced vidders aren’t stealing from them is when it actually becomes profitable for them. And offering copyright holders the opportunity to sign up as owning that IP and make some money with no extra work has to be a lot more attractive than filing thousands of DMCAs at their own expense. Maybe have a disclaimer with all the vids that they’re fanmade and the content owner does not create, oversee, or individually approve content as a cover-your-ass if some vid clashes with their brand or something. Part of the contract that allows them to make money has to be that they don’t get a say in what kinds of vids get made--you can call it piracy or you can worry it will clash with your brand, but it can hardly be BOTH, right?
  • I’d be careful with how we do or do not identify, regulate, and report copyright videos and photos. There are enough recent court cases that not making any attempt will not go over well. Flagging automatically through a bot with the OP able to contest and self-reporting by the original owner might be the best options. It’s a little strict but will save you from legal issues in the long run.
  • I know human moderation is rarely the cheaper option, but I have to wonder if that’s feasible? Like, we’re not YouTube. How many vids even get made a day in fandom? If it’s like five, someone could just check them before approving them. If it’s like 5,000, you might need a bot. +2
  • Note that hosting and distribution of vid content is liable to be extremely expensive, particularly wrt ‘edge’ (close-to-user for latency) caching. No one but a big cloud provider can do that; it means putting machines in physical cities. But big providers will NEVER be free of copyright detection or friendly to explicit material. Vid hosting might be the feature req to sink this ship. See also lim’s post for more context …. Whiiich you can’t because lim deleted it, okay. Tl;dr we might be better off w/ good integrated support for offsite vid-hosting. +1
  • I understand that we probably can’t do the most expensive features out of the gate, but I don’t want to give up on ever doing them. Embedding would probably have to be enough at first. Would a SFW vid requirement help make that more feasible as a middle step on that road? Most vids are SFW anyhow, so it would still be a huge help, even if it’s not the full dream. +1
  • This might also help cut down on porn bots….as long as the method of determining what is/is not SFW is more accurate/human-driven than tumblr’s current fiasco.
  • So I’m thinking about this, and the issue of vids and the issue of videos are two separate matters. Fanvids are something there’s really no safe archive for anywhere on the web, and there aren’t all that many of them, compared to other types of video content that exist. Nor are they typically more than a few minutes long, some as short as 30 seconds. Other types of content might include: viral things like cute cat videos, personal vlogs, and amateur porn, especially if we’re courting tumblr’s sex workers. Viral cat videos can be embedded from almost anywhere. Vlogs are generally just fine on YouTube. Porn may or may not be embeddable from existing porn sites (e.g. pornhub or xhamster). The thing is, hosting sex workers’ amateur porn is going to require age verification, or we’ll have the exact same C-generated CP problem tumblr did. Amateur porn sites age verify. It’s also a lot quicker to make a porn video and there’s a lot higher volume than for vids, as well as it being harder to find moderation for if there’s a moderated queue. If we’re going to make both embeds and do video hosting, it makes sense to prioritize our space and bandwidth for actual fanvids, and let people embed the cat videos from Instagram. +1

  • The ability to create more customized pages that don’t automatically have your blog theme applied (a la Tumblr) +1

  • Threaded comments on a post +509
  • And the comments go on the original even when it’s reblogged, so they’re all findable +429 /-2 +1017 (1013 magical x-files number!) 21 was the last “real” number I could find so we’re reverting back to that (if you lowered this number, add a -1 instead, thanks.)
  • But a reblog with a comment has the ability to view the comments starting with that reblogger’s subthread. +5 /-3 (Be careful with this - it has the ability to strip something of responses and context.  In many massive threads on tumblr, corrections and counter-arguments are invisible because people can only see the particular thread created by the reblogger they specifically follow, rather than other responses as well.)
  • I think you may be confused about how tumblr works? Reblogs always link to and credit the OP, no matter how many reblogs deep or how many additional comments. Both the OP and the reblog chain that brought OP to that point are clearly visible and attributed. It's obvious what the OP was and obvious what the most recent comments on that branch of the reblog tree were. One usually assumes a person reblogging agrees with the most recent comments unless they add a comment of their own to disagree. All of that is already transparent on tumblr, with the possible exception of the older view of the site that made posts less legible the more deeply they were nested. This was fixed by the redesign on dash view and mobile, though individual blogs on desktop remained broken, because tumblr. ¯\_(ツ)_/¯
  • What I think the person you responded to meant is those posts with false information that get a gazillion notes, and even if someone reblogs with a correction, it’s not easily spread, because most people won’t see it unless they follow the person who posted it. Not sure there’s really any way to prevent this, though…
  • Oh, I see, that’s different. Well, on tumblr, I can click on the number of notes, and if it’s a huge post with tons of reblogs and comments, I can scroll through those--tumblr prioritizes these over fluff like “likes.” And if we’re also having threaded comments in a centralized place, it’s easy enough to skip over to that and see what people are saying. OP changing their own mind and editing their post live is another good feature, but if you want to somehow add “corrections” to other people’s posts that that person disagrees with and can’t divorce from their post, rather than simply adding a comment, that just won’t be functional, everyone who disagrees with everyone else will want their say.
  • Also, perhaps relatedly, while you can’t hard-code manners or wits, we could disable embeds from “fake news” conspiracy theory sites to at least make the most obvious misinformation a little harder to spread. I think Facebook has a list of such dubious sites or something? +1
  • A way to hide reblogs that are just ‘x reblogged’ with no additions from the ‘notes’ view, so you’re not going through 12853 additionless reblogs in the notes trying to find the person who continued that collaborative story +15
  • This could be merged with the ability to view only activity with added content, a la Xkit.
  • Different options for “hide reblogs without new comments” and “hide reblogs that don’t have tags” for those who comment in the tags but don’t technically add new comments
  • Depends on how we’re handling “whisperspace”/marginalia. That type of ephemeral comment might be its own thing, which you should definitely be able to view! But you should be able to view tags too, just because some will put marginalia there from force of habit, and if OP can’t see tags people will use them to avoid being read by the OP, which is annoying. +1
  • SBNation-style comment features that highlight new comments +8 and let users skip to them with hotkeys etc (low-priority feature) +10 (I use a greasemonkey script to do this on DW and it makes the site like ten times more usable.)
  • LJ/DW style threads, please, they are still my favourite for conversations on the internet. +23/-1
  • No reddit-style upvote/downvote system. +106
  • What about just upvotes? That way, you can say "yes, I like/agree with this comment". +84/-3
  • Like say…..a like button. +35
  • Separate “like” and “bookmark” buttons +27
  • No dislike button, too easy to abuse +152
  • maybe a dislike button that all it does is remove 1 like/upvote -98
  • Do you mean a "hide" button?
  • That’s reddit-style downvoting, which I think kind of sucks. You deserve to keep all the likes you got! +7
  • Ravelry uses "educational" "interesting" "funny" "agree" "disagree" and "love", which is probably too many, but I think "agree" and "disagree" work better than "like" and "dislike" or "up" and "down" because they are more specific.  +65/-11
  • There’s a forum I’ve seen that has agree, winner, informative, witnessed, useful, and like as options for rating posts
  • If there’s a positive and a negative option, people will use the negative one to dump on +11
  • Informative helps separate well-thought-out and researched reblogs from virulent rants +1
  • I’m not convinced this can’t be gamed. (Plus every anti having their “reminder that shipping is bad!” marked super informative. 9__9)
  • Or maybe a “dislike” button that hides the thing from your personal view, but doesn’t affect anyone else +1
  • Isn’t this essentially muting a post? (see “mute a conversation” below) +5
  • I like the idea of having a few “emotive” reactions--that way you can “aww” when someone posts about being ill, or “like” something happy, etc. Always hate when I have to “like” someone’s post about their sick kitten or whatever :C +7
  • I’m torn between agreeing and finding those automated responses a bit alienating after a point. I’m in favor of some degree of passive engagement through “likes,” but at some point I want people to use their words. (The other day Cortana tried to suggest sentences to type to my mom on Skype. Human interactions shouldn’t be too automated.) Perhaps if the “like” feature didn’t have the happy-positive implications, but was something like, “noteworthy.” “You and 5 others have marked this as noteworthy.” Or something that communicates, “I appreciate you sharing this,” though I’m not sure what catchy way to condense that. That’s what likes really boil down to. Making it seem unambiguously happy was more of a commercial social media mood management thing. +2
  • People already use the heart/like button as a generic “I see this” or “I like this”--before facebook added their other non-like reactions, it was full of people saying “Not liking this!” or similar. Youtube also has conflict -- do you dislike a video that’s sad where you want to support the creator? Sometimes you don’t want to come up with words, but want to quietly support people.
  • Call it a ‘support’ or something instead of a ‘like’? Kinda conveys both meanings.
  • I would like to “like” both posts and comments (looking at you, AO3) +16 +1
  • If a post I liked gets deleted and my likes are viewable a la tumblr, and the deleted post is inaccessible to me now, take it out of my likes. I have 153 ghost likes on tumblr that I can’t unlike and bring the number down. +8
  • No limit on how deep comments can thread, even hundreds of comments in you can still tell what comment is replying to what. +16
  • This might be a technical limitation
  • How is this a technical limitation, when Livejournal was doing it in the 90s? I don't mean unlimited, stupendous numbers like 80 billion, but if Livejournal had a cap on thread nesting it almost never got in my way, while Imzy's got in my way constantly. +2
  • Numbered comments would help here - “replying to [name] at [number]”.  Each comment keeps its original number even if deleted/moderated out. ⇐ A la 4chan. +6
  • Definitely need a hard limit on how much of the thread is visible before it collapses? And you need to expand it. For visuals sake. Maybe like first 4 comments in a thread and then a … newest comment? +4
  • Needing to hit "expand" is completely fine, even ideal. Threads quickly turning to soup where all new replies are at the same nested level renders the site near-useless for vigorous discussions. +8
  • Being able to contract the expanded bit again after reading it might help with how much is loaded at once +10
  • Setting for how much is viewable before it collapses, for those of us with really small / big monitors +3
  • “Parent up” feature that lets you see the thread starting from the comment that comment was replying to. +32
  • And thread from start ala DW +15
  • "Context view" a la Reddit would be helpful too; show the parent comment but not any of the parent comment's other children comments, to prevent clutter in humongous threads. (I'm not proposing this instead of the above, but in tandem with it.)  +3
  • Images in comments the same as in posts -- with the same opt out/accessibility features. +11

  • Reposting content with or without comment aka reblogging, retweeting, quote tweeting +64
  • If we do this, can we have it per user? As in, I follow x for their original content, and y for their reblogs, and never the twain shall meet. +19
  • Should adding comments/extra info/reactions make a difference? Maybe reposting content without comment being treated differently from reposting content with comment? +4
  • Control of the entire reblog chain (think Pillowfort, which links to the original post; not Tumblr, which replicates the content on your site.) +26 / -5
  • Counterargument: Linking the original post means that if the original post changes, users could end up with surprise goats or whatever on their dash (agreed, look how much people enjoy messing with embedded buzzfeed posts. +5  Compromise that you can’t edit, but if there’s a way to delete the original post all reblogs are deleted?) +3/ -71
  • re: pushing IMAGE revisions, iirc this *was* a ye olde tumblr feature and it *was* abused to make people reblog kittens that turned into swastikas. See also: the glorious tradition of making people look like jerks on Facebook by editing your half of a convo with them. Aka the likely reason why (edited) is universal now
  • I would rather have transparent edits and no delete than delete and no edits. +6
  • No delete + transparent edits (marking where the post was edited, possibly showing the original?) seems like the way to go to me. Deleting an original post eating the entire chain has potential for abuse and can lead to content loss. Think of an OP saying something wrong, a whole bunch of people adding in extremely thoughtful corrections / discussion leading to an entire chain which… just goes poof if OP deletes. Same with, say, a collaborative storytelling thread disappearing if OP deletes. That’s no fun for anyone. +3 (I’m so relieved that someone pointed this out)
  • Guys Tumblr’s way is like quotes on a forum, except instead of an actual quote function it just copies the entire text, xenforo-style quotes are way better, you can’t delete other people’s quotes of you (THAT OPENS UP SO MUCH ABUSE POTENTIAL) but people can both snip quotes and easily see if a quote’s been edited, as well as the URLs updated automatically if you change URL rather than it just breaking +5
  • I love the idea that if someone quotes you, they can’t misquote you, and you can edit your quote live (even if it’s just to fix a typo that’s going to embarrass you to no end) but you can’t hide your edits, so you can’t change it to a radically different statement to make the other person look bad without everyone seeing what you did. +2
  • (tl;dr tumblr reblogs are a Lot like forum quoting except drastically worse)
  • Counter-counter argument: when posting something that then gets reblogged/goes viral/etc, control of the original content is lost.  This makes it impossible for the OP to retract, edit, correct, hide, delete, etc the original content.  
  • Give OP the ability to make edits, with transparency (all users can see former versions of the post) and to optionally push edits as "important updates" that notify everyone who reblogged it +3 (useful for things like retracting misinformation, apologizing for a blunder, clarifying something that was misinterpreted due to unclear wording, turning a WIP into a finished product.) That way, you aren't watching with your hands tied as you're repeatedly reamed for the same misunderstanding or a hoax you accidentally started spreads, nor do you have the problem of 50,000 notes on a WIP and 10 notes on the finished product, but you also don't have the ability to permanently remove content that was public and became part of someone else's blog. If you have a problem with that, don't enable reblogging of your posts, it's part of what reblogging fundamentally is.+4
  • OPTION: art cannot be edited/removed (so kittens don’t change to swastikas) BUT text can be edited (so typos can disappear and information can be updated) -1
  • Devil’s advocate: you can do the same kittens-to-swastikas thing with editing text. Make a positive post that garners positive reblogs of the “yasss”, “you tell ‘em OP”, etc type, and once it’s got enough notes, edit it to “nazis are great, they’re just misunderstood”... +3
  • As an artist, this is unfair! We make mistakes too in our pictures (forgetting to color an area, disasters like backwards hands, leaving out a costume detail, whatever!) and it would also be awesome to update WIPs to finished works so we could share early, without the effect of “losing” notes because people think, “eh I already reblogged a version of that, I don’t need two.” The original picture would still be available to view as what the post originally was, so if your mutual is posting swastikas you can see that they used to be kittens, and your mutual just got trolled, they aren’t a Nazi. Since as above text could also be edited to include all kinds of slurs, hate speech, graphic content, even serious things like doxxing, I don’t see why images should be treated as so much higher risk. +3
  • Also, how about we just make being a really obvious Nazi (trolling people into posting swastikas seems sufficient there) a bannable offense. I’m less interested in squinting at every single political opinion and trying to guess who might be a secret Nazi, but someone editing a post to just say “heil hitler” 50 times is a kind of content no one will miss. +4
  • Being able to send push notifications to someone who doesn’t follow you seems...bad +4
  • No, because they reblogged from you. It’s a bit hypocritical to say, “I’m literally posting your content on my blog, but don’t ever speak to me.” It would just be an automated notification that the post was edited, anyway, which knowing about helps you catch much faster in cases of bait & switch abuse. +2
  • Ability to orphan your portion of the post so that it no longer leads back to you but the content (and any replies) remains +4 +1
  • Maybe a similar function to Twitter, which deletes all reblogs if the original post gets deleted? (Also edits all reblogs if the original post is edited, to stop bad or old information from spreading, which is a problem on Tumblr) +5/-6
  • Ability to update with a correction that notifies people who reblogged the old version of your change is more useful for this than quietly making the post disappear after the damage is done. +8
  • Reblogs of deleted posts continue to exist, thus preserving everybody's comments, but the post being reblogged is replaced by "no longer available".  That's what Twitter does, and it works well for me.  On the other hand, we all know how to screenshot.  On the other other hand, most of us know how to Photoshop, so....? (No, not all of us know how to Photoshop)  (OP: I don't myself; my point is that screenshots are no longer probative) -6 (don’t leave other people’s blogs full of holes and lost content.)
  • Explicit decision with documented reasoning on whether or not to allow edits to existing content. +3
  • Ex. we know this wouldn’t work on Twitter as it currently exists, but that doesn’t have to determine usage here; think through ramifications, particularly re: reposting/reproduction on other feeds/in other contexts.
  • Mark post as edited and potentially show original in mouseover or something, like various other platforms do. +38
  • Replace old version with edited content, with a little symbol or highlight underline or something indicating the content was edited after the reblog happened. Ability to click that and see what they changed. So you can fix your typos, but you can’t cover your tracks. +13 / -1
  • So basically you’re asking for a diff view
  • As someone who works in IT, preserving views in the audit trail is gonna use space like HELLA. +3
  • Worth it. Text edits can’t be that much worse than images, gifs, audio, and video, which we already want.
  • Have option to notify people who have reblogged content when an edit happens. +19 +1
  • That seems like a sensible option, as long as it’s opt-out-able so people who don’t care don’t get bothered +4
  •  Types of changes to a post I reblogged I want to be notified of: a new image was uploaded, an image was deleted and replaced, OP marked it as an important update to correct misinformation, clarify, apologize, etc. Types of changes I don't want to be notified of: OP realized they meant "they're" instead of "their." +4
  • Image revision capabilities are ripe for trolling, though it would depend on platform. On tumblr absolutely, but I don’t remember it being an issue on Deviantart. Might have to do with how much of the OP is reproduced with each ‘share’/reblog action— on DA nothing ever superceded actually viewing the piece in its native page, while on tumblr reblogging implies a sort of ownership of endorsement (which is then exploitable).
  • They are kinda ripe for trolling, in a huge general-interest setting where you’re reblogging kittens from everyone without even looking at the username. But you could pull that kind of stunt once before everyone hated your guts. Maybe people should think about who they reblog from for half a second. Glancing at OP’s blog to see if they look like they’re going to troll you like that might mean fewer people reblogging from Nazis or paid Russian state operative bloggers  or bots or whatever the heck went on in tumblr’s big glory hole reblog system. The vast majority of people wouldn’t light their social cred on fire by trolling in that way, and if people have to pause and think about whether they trust someone before blindly slamming reblog, that might do the culture some good. Image revisions would also be subject to the same platform-wide rules about warning/CNTW for image content, of course, so if they’re abusing it that far and not using the mandatory flags, they’re cruising for a ban. Sites like Twitter don’t have AO3-like content warnings. +1
  • Maybe a couple of tickbox options when you make an edit so you can distinguish eg ‘correction’, ‘addition or clarification’, vs ‘typo’ +4
  • This could be useful for a history of editing on the original post (e.g., at the bottom of the OP edits are recorded like "edited on 03 January, 2019, at 16:15 for typo" "edited on 21 January, 2019, at 09:00 for addition or clarification" "edited on 01 February, 2019, at 20:08 for addition or clarification" "edited on 01February, 2019, at 20:32 for typo"). Maybe not as something immediately visible but that you can click on to expand the history?+2
  • This + notifs about posts you reblogged/commented on being edited would be a good alternative to having to store all original and edited versions of a post, esp for posts that get a lot of edits. You can just read thru the edit history instead, and people who really want receipts will probably screenshot things anyway.
  • I think minor revision/major revision would be sufficient. (Replacing an image or video would always be a major revision, sorry artists who just realized they forgot to color in a spot, it’s an anti-trolling thing.) Since areas that had been edited would be marked in some way, figuring out the history of a post would be easy and require no explanation. You can see exactly what they did. And that way it’s not on the honor system, either. +1
  • Comparable to AO3? Where if you look at your history, if a fic has been changed they say “major” or “minor” changes. Too detailed info could be a hassle considering the volume of posts & reblogs to take into account? +2
  • Also make opt out/in also possible by number of notes. If it’s a five note post, it’s probably something I’ll remember reblogging, but 50k notes? Ehhhh.        
  • No, I don’t think you should be able to opt out of major revisions. Say you reblogged a news story that OP realized was a hoax? No one can stop you from dismissing the notification, of course. Deleting the reblog should prevent additional notifications, so if someone is being an ass you can just get rid of their post. +1

  • Ability to make specific posts rebloggable/not rebloggable (at creation? At user whim, similar to freezing comments on a LJ/DW post?) +292/-2 +1
  • Ability to turn off comments on a post but still make it rebloggable, or make a post rebloggable but not allow comments. +663/-2 +1
  • Enabling reblogging on your post should be a one-way (per-post) street. No use closing the barn door after the horse has left. Once it's part of other people's blogs, you're overstepping by micromanaging how they interact with it, beyond blocking them or editing your own words transparently. +10
  • Ability to limit reblogs to a certain subset of users? -1
  • If you mean criteria like “over 18,” cosigned. (Though marking your post as explicit should make your content invisible to minors anyway, if we even allow minors on the site.) If you mean stuff like, “Do not interact if you are a man, eat meat, or think shipping is okay” then no. +14
  • Yeah marking something as 18+ shouldn’t hide your post from search automatically. It should only hide it from those users who either have explicit turned off or are under 18. +5 [we’re not talking about search features in this thread/section, should probably be moved elsewhere or removed]
  • I read this as more like "only my friendslist can reblog," which seems reasonable, but does that mean other people cannot reblog the same post from them? How feasible is that kind of limit? +10
  • IMO, not feasible at all. I want reblogging because I want exposure and "signal boosting." I want to spread like wildfire. For friendslist-only interactions, it makes much more sense to show appreciation with likes, comments, or preserve the post by privately bookmarking it. Reblogging should only be possible for posts that are already public. You can have fame or you can have privacy, but it's a contradiction to have both. +6
  • Seems fair - option to make a post ‘rebloggable’ and anyone can put it on their blog or an option to keep it within a friendslist and they can comment as much as they like, while you keep total control. +1
  • Tbh if you only want a friendslist to reblog it, then I’d say make it so the post itself is only viewable to said friendslist. -1
  • But then who can view the reblogs? Now friends of friends are viewing it--or are only people in your circle still allowed to see it on someone else’s blog? Can the friends of friends reblog it if they can view it? “Underground” reblogging makes no sense. Reblogging is a publicity feature for going viral, for people who want that. Only public posts should even have the option to be rebloggable. +1 +1
  • A way of prioritising a post you want to monitor, thus staying on top of what reblogs are added to it. +22
  • A post tracker vs a like? <essentially the idea is that if you want to see the sequence of reblogs and additions from other users on a post in your TL you click a little button on it and it stays in a dedicated section of your feed that you can visit to keep an eye on developing reblog chains.
  • Like the Subscription function on AO3? < yes applied to any post you choose with one click. And a feed for said posts separate from your main feed or your likes. +29
  • Like one "new" feed and one "watched" feed
  • But a way to add limitations on who can subscribe to your posts, like only people you follow, or only people who follow you. Or let them request to track the post if they don’t meet those criteria and you can allow/not allow. This is to avoid making stalking/harassment super easy and low-effort.
  • I'm not sure this applies entirely to this suggestion? The suggestion was for you to be able to individually mark posts you want to subscribe to so you keep track of further replies, not subscribing to a person to see all their posts.
  • True, I guess, but I’d be weirded out by people tracking my notes that closely and easily and not being able to deny them access. How would you feel about an anti tracking ship posts and keeping track of all the interactions? It might make people shy about interacting at all. +1
  • So are you creating new specific security filters there? This could potentially just merge with the different security filters a la lj/dw’s custom friendgroups?

  • Vote for NO reblogging/reposting. +4/-26
  • Can I ask why? (Out of curiosity, not jerkiness) I envision this as a place for all kinds of media, not just text. If you didn’t want reblogging you could just stay on LJ/DW
  • Of course :) I wrote a whole paragraph down in the “Pros and cons of “reblogging”” section. The short answer is that I think if likes and comments could show up on your feed, they could take the place of reblogs. While keeping the original content in one spot, and all the commentary around it attached to the original post. I say this as someone whose tumblr and twitter feeds are like 95% reposts. Addendum: It seems that maybe what I’m actually thinking of is Pillowfort-style reblogging where all reblogs link back to the original post? As opposed to just creating infinite copies of the original post.
  • Vote against Pillowfort style reblogging; allowing users to interact with a post on its own merit lets them develop relationships gradually instead of feeling like they’re intruding on OP’s space. Pillowfort reblogging also doesn’t take into account the “appending” function of reblogging that isn’t always appropriate to threaded conversation. This is expanded on further in the “pros and cons of reblogging” section

  • Likes/kudos on posts. +85
  • Ability to show/hide likes. +25/-2
  • Do you mean your own likes, or to hide how many likes a post receives, period?
  • If regarding the privacy of your own likes: A vote for private being default. This is one of Tumblr's main advantages over platforms like Twitter. +18 +1
  • Vital! I don’t want an easily accessible page of everything I’ve ever hit “like” on. If I wanted it all in one place like that, I'd reblog it.
  • Counterpoint: On Tumblr I use likes to bookmark (this might be addressed somewhere else) and wish there were an easier way to find/navigate/search through things I've liked. That said, I don't want things I've liked to be public- my liked posts and reblogs are usually mutually exclusive. +5
  • Most people I know use likes as a way of keeping track of posts that they’ll reblog later. I suppose you could accomplish the same thing with a drafts feature, but I would hate not being able to browse my likes.
  • I don’t think I was clear--I meant an easily accessible page to other people, not something that was inaccessible to me personally. Private likes on tumblr are still browsable by me, it just makes it harder for people to stalk everything I’ve been looking at on the site. If I go into some weird porn tag and hit “like,” people looking at the notes of the weird porn will know, and I’ll be able to find it again, but no one will be able to pull up my likes page and go, “Yikes, nonny’s into THAT, huh?” It’s still usable as a bookmark or “to reblog later” feature, though I hope the site would have better options (actual bookmarks, drafts) for both. +1
  • I also use likes to bookmark, definitely let me navigate that stuff.
  • I also use likes to bookmark on tumblr, but that’s only because tumblr doesn’t have a separate save post/bookmark feature. I would prefer being able to give something a like, that I like but don’t necessarily want to reblog. I’d also like having a separate feature to bookmark a post like a recipe or a how-to post. +2
  • This is currently available as an x-kit extension on Tumblr (post archive) - could this be replicated on a new site?
  • Ability to show likes to followers/mutuals, hide from others. +10/-3
  • Ability to specify exactly which notifications you receive (replies, mentions, likes, reblogs etc.) +44
  • Ability to like/kudos a comment rather than the original post. +369 +1
  • Ability to remove the option to like a post (such as an art post. Something you want shared rather than just “hearted”) +5/-13 -1
  • I really don’t think this will increase interactivity with posts/content in the desired way. The fact is that not everyone likes all art/etc that they see enough to reblog it every time, and artists turning off “likes” for a piece are going to end up discouraged when they don’t see more reblogs as a result. It’ll create drama +8
  • Forcing users to reblog isn’t something you can realistically dictate. +3 +1
  • There could possibly be other reasons to not want likes on a post - maybe posts that also don’t allow reblogs only? +2
  • I’m fine with posts that don’t allow you to interact with them at all, but trying to “force” users into boosting you by disabling likes is just obnoxious. +1
  • Okay, I actually argued against this because it’s a peeve of mine when artists whine for reblogs (I’m an artist myself and I do understand their feelings--perhaps that’s why I cringe so much) but I joked to someone that if you really wanted to armtwist people into reblogging, you’d make it so they can only ‘like’ the post after they’ve reblogged it, instead of taking away likes entirely. This is a joke, not a feature request--I think it would be a terrible idea. But then I realized that perhaps giving rebloggers some kind of “bonus” isn’t terrible? Not withholding stuff they’re used to having access to, like ‘likes,’ of course, and I don’t want to veer into anything too major being held hostage for reblogs, but content creators being able to show a bit of appreciation for their supporters bothers me a lot less than being manipulative and demanding about it. On Patreon, you get certain extras for donating at different tiers. If we can find some way for creators to give rebloggers some kind of nice little reward or bonus for their support, that might give frustrated artists some tools to drive interaction without coming off as whiny or entitled. +3
  • honestly i think the reward is that they have content on their blog that they didn't have to make? that's the function of a reblog

  • Pin a post to the top of your page. +80
  • Ability to pin a post that's not yours. +6/-6  (Pardon me, but could you clarify this please?  Like, pin a post on someone else’s X or pin someone else’s post to your own X?) (I think pin someone else’s post to your page)
  • Pinned posts are good, but should be limited to ones you’ve created, to reduce the risk of harassment  +2/-2
  • Yes, but, what if I want to pin the Yuletide Sign Ups post on my writing comm or such? <valid, my + is retracted
  • On twitter, people can quote retweet and pin that post instead of the original one and original poster will still be notified (through the quote retweet). People can do the same within this platform. Allowing people to pin other’s post in their own profile would just open a new permission quandary. +3 +1

  • No auto-updating feed/toggle. +39 +1
  • Does this mean updating the feed while you are looking at it? If so I’m in favor.
  • Yes, PLEASE. Having text move while I’m reading it is a massive migraine trigger for me

  • Ability to queue posts so that they post automatically at a specified time. +27
  • Separate abilities to schedule vs. queue posts? Schedule being like “post this at this time on this day” and queue being tumblr-style, “post from this pile at a steady rate” +25
  • Ability to put queue on pause, for hiatuses and the like, so you don’t have to go add 200+ posts to your queue when you return, you can just pause it and the posts will stay. +3 +1
  • Ability to shuffle queue -- or move posts to other specific spots +3
  • Ability to add queue tag for all posts that are queued, so you don’t have to individually tag them with a queue tag, they can just appear there automatically.+8
  • Not trying to be a jerk, but what is the function of a queue tag? Do you use it to navigate your own blog/categorize or is it intended for the reader? I see them all the time ofc but I’m still baffled
  • Not OP, but I think it’s a way of saying “I’m not actually here, don’t expect a reply”? Or “I did this ages ago, so if this picture of a gorilla coincides with unfortunate gorilla news, know it’s a coincidence.” +12

  • Ability to save and return to draft posts. +75 +1
  • With auto-save. +37
  • Ideally multiple auto-saves by time (low-priority) +7
  • Don’t hide the drafts page when you don’t have a draft like tumblr does +5 +1

  • “Read more” option to limit size of long posts. +35 +1
  • Auto-read-more after X characters? (toggleable?) +10/ -6
  • Auto-cut length should be set by the reader, not the OP, and apply globally to content that user browses. Default would be off. +15
  • Needs to also be a manual “read more” which OP can add. Thinking of fic posts, where teaser/opening is above the cut, full post beneath where timing is part of the marketing, or even just ‘triggery content below cut’ with warning text beforehand. No use if your auto-cut length kicks in halfway through the triggery content. +4
  • Auto-cut length that is very long (3,000 words, 5000 pixels, etc.) so most users aren't impacted but deliberately trolly posts like the infamous sky image on Tumblr are hampered? +7
  • I don’t think the sky one was actually trolling, but antis do spam tags with long chains of images (usually just a meme reading "I'm calling the police" x 80) and that gets annoying. As long as expanding long posts you actually do want to view inline, like webcomic dumps, is an easy one-click affair.
  • Like ye olde LJ-Cut, can be used to hide images too. +16 merged in with general “read more option”
  • Spoiler text cuts or some way to expand a cut on your reading list without going to that post. +32 (I use xkit [“read more now”] to do this on tumblr and forgot this wasn’t just how it worked.)
  • Ability to ‘cut’ posts at a certain height that you can then expand. +1 merged with configurable length
  • Ability to end a cut/readmore, like LJ/DW and unlike tumblr, so everything under the cut/readmore isn’t indiscriminately hidden, but you can say, hide the text of a fic, end the cut/readmore, and say something like “I’ll be updating every week” that people would see without expanding the cut/readmore. +28/-1 +1
  • Why couldn’t you just put the text at the top? Although the ability to have multiple cuts is useful.
  • Because of flow. Sometimes there's text that just doesn't make sense at the beginning of the post, but it still shouldn't go under the cut. +3
  • Multiple cuts/readmores could also be used as an index of sorts -- think headers -- for longer posts if they’re able to designate a beginning and end per cut/readmore. +9 +1
  • They’re also useful for aside clarifications - like having a footnote or a tiny bit of spoiler content that is otherwise avoidable +1
  • I think this would be most useful if readmores can be labeled for what they contain, something like "read more: [input text here]" so you can have "read more: spoilery image" or "read more: sources" or "read more: digression/technical details." I'd appreciate any feature that could be used for something like footnotes. +3
  • Ability to add comments under a readmore, so you don’t just have ‘comment: readmore || comment2: readmore’ and just have ‘comment: readmore (comment 2 is under the first readmore), but has to be enabled by the OP. It’s useful for NSFW RPs where you want all of the RP under one big cut. Nested readmores, you mean? If so +2
  • It seems like some of these (hide spoilers, hide NSFW) are similar in concept to the filtering/warning things being discussed, but people want it to be granular down to the word rather than per-post. Readmores are a workaround for that, but a possible complication is that if content flags (including CNTW as an option) are mandatory, you might have to flag the whole post--at least as CNTW--anyway just to observe the rules, which defeats the purpose. An innovative yet probably difficult to implement alternative would be to be able to apply those warnings not to an entire post, but to areas of highlighted text. (Others have already suggested doing it on a per-image basis for reaction gifs, or just one “sensitive” gif in a gifset.) It’s similar in concept to the idea to be able to highlight areas of text and “like” that text. I mean, it’s not that different from how a lot of forums and I think Reddit allow you to mark a single word or area of text as a spoiler, it just lets you mark it as more things.
  • Add readmores that open a new tab to the blog itself, and some that are just ‘click to view’ within timeline / dashboard.
  • Make the different types obvious at the very least, sometimes I really only want one tab open at a time. But I can see that first option becoming very annoying very quickly.
  • Can’t I just, say, click to expand, right click + “open in new tab” to do that? Instead of the OP having to guess how people will want to use their cuts. +1
  • This should apply to number of images as well as number of characters.  I use xkit to keep one post with multiple images from dominating my dashboard, and I'd love to see this available as part of the system.

  • Public follower count a la twitter rather than tumblr (mostly wanting to raise it as an issue since I don’t see it mentioned yet.) +3/-35
  • Or maybe option to show/not show follower count +17/-3 conditional on it being set by blog owner like “Option to make your follower count public” +1/-1

  • Mutual flag to see if a user you follow is also following you (xkit has this as an extension and twitter has the “following you” flag.) +47/ -2 Visibility in context (hover) info like DW eg “this person follows you. You give them access.”
  • Maybe have this as a toggle-able function, like being able to turn read receipts on or off on iMessage?+3

  • Receiving asks with option for anonymity; being able to respond to/post ask on blog, tag, and reblog. +25 +1
  • Ability to turn entire feature off. +47 +1
  • Ability to turn only anonymity off. +52 +1
  • Ability to block anon IPs. +38/-2 +1

  • Submissions, which is to say, sending a post/content for moderator review on someone else’s blog (and the ability to turn it off) +24
  • Maybe only for particular blog types, like a community zine style blog with moderators posting their own as well as others’ contributed works versus a personal blog where only you make posts? +2/-2
  • That would require designating the blog as a certain type. Easier might simply be the ability to turn submissions off +10
  • I think what you’re thinking of is something more like Communities on LJ/DW/et al It’s like a closed comm where only the people the mod admits as writers can post but everyone can follow/potentially comment (I mean, sounds like it can be done with existing DW features without a special name/category) +1
  • Communities already do all this stuff better, and it was used much more vigorously on LJ/DW than the submit feature on tumblr was. I feel like regardless of what the site design enabled/intended, user practice was rarely to submit content to moderated blogs in lieu of communities, but rather to treat public tags as a big unmoderated community. Finding distinct social roles for communities vs. tags would be a fascinating venture, something tumblr never accomplished because the multiple-modded sideblog + submit features just never achieved traction.
  • If communities are also a thing, they should be followable like individual users. +1
  • Completely agree! They are on LJ/DW. I never imagined them otherwise.
  • This would already be the case if you have the ability to turn it on and off like OP suggested

  • Ability to turn off any kind of inbox feature, including any messaging system, to reduce anxiety and provide control for people with anxiety disorders. +16
  • You don’t even need an anxiety disorder to want this, bullies and abusers are encouraged if you can’t close off access to them. (Deleting your entire account and losing all your work, friends, etc, doesn’t count--that’s a win scenario for some bullies.) While you can block individual users, in some kind of dogpile situation you need to be able to board up the windows, at least until things quiet down.
  • Added to this, if one does turn off any kind of messaging/communications, can we have a whitelist? Like, maybe there’s one single person you’re okay with messaging you. Just add usernames to your whitelist on a case-by-case basis. +3

  • Controversial, but ability to hide number of notes on your posts +2 /-11
  • Those against: why not? IME posts with high note numbers continue to snowball while low note ones fade. This looks like a soft way to discourage sharing, to me.
  • People don’t like having information and features they’re used to snatched away from them? It’s different if you want to hide note counts from your view for yourself, but if you want to take something away from other people, some will mind that.

  • Ability to hide number of notes on posts you’re viewing +1/-1

  • Even more controversial, but no number of notes on any posts ever. Going back to the forum life baybee +1 / -22
  • Even forums had hit numbers and comment numbers +1
  • Only if there's some other easy way to tell that there's more  
  • Not great for creators who want to gauge their reach e.g artists who are trying to build audience base +2 +1
  • As a content creator I very much want to see what’s popular and what’s not, so I know what content to make more of… +1
  • Reduces the pressure to gain fame and increases the emphasis on having fun as fans together. Makes social media a saner world. Would be great to have this as a toggleable option.
  • A setting so that you don’t see any note counts is acceptable. I don’t mind people using this as a filter to step back from whatever competition was triggering feelings of inadequacy or harming their mental health. But taking it away from everyone, on the whole platform? No.
  • All the forums I’ve been on lately actually have a “thanks” button, which is effectively a “like.” Forums have, as mentioned, hit counts as well as comment counts. Worse (for what you seem to want), all the forums I’ve been on put the most recently-commented-on thread on top. So a really hot thread will always be on top, while an unpopular one will get pushed further and further back, where fewer people will see it. And which is even worse: 500 hits and 0 comments, or 2 hits and 0 comments? Newer forums I’ve been on add a “hot!” badge to threads that are getting a lot of comments on them at the moment. If you want to disable any kind of metric at all for people to know what’s popular and what isn’t, I don’t think that’s a good idea. Even in real life, you can tell the difference between a crowded street and an empty one.

Metadata & Discovery

  • Ability to tag/categorize content (ie: #Supernatural or #nsfw) +930
  • Be able to use tags to see just your own content with that tag, or just someone else’s, or all. +41 (Couldn’t viewing just your own content be achieved by searching the tag on your own blog rather than on the entire site?) +3
  • Fandom tends to use tags in two ways: as metadata, and as supplementary text (for example, long and poetic stream of consciousness commentary). The second is controversial. Some solutions include “tag wrangling” a la AO3 but this only works when something is canonical, like media.
  • Vote for metadata tags only. +10/-3
  • Vote for metadata and supplementary tags. +90/-3
  • Vote for hiding number of tags. +3
  • With ability to expand to view full list of tags. +12
  • With the ability to opt in or out. +2
  • Vote for metadata tags and supplementary tags as two different categories of tags, with the ability to show/hide supplementary tags as desired. +65 (-11 This seems like it would be very time consuming for users, perhaps to a burdensome degree.)
  • I like the two-tier tag system, with globally-searchable tags (like tumblr) and organizational tags that are specific to the user, so for example I can tag something with “hockey” and choose to have it appear either in the global hockey tag and in my own hockey tag OR only in my own hockey tag. Is that part of what we mean with the supplemental tags? +7
  • No, because reblogs wouldn’t appear in global search anyway (or they don’t on tumblr) and if you wanted to tag a post with #hockey on your blog but not appear on the global #hockey, you could just not make that post public if the privacy settings we’re talking about turn out anything like I’m hoping! These are both effectively metadata tags. The second type of tag wouldn’t be like #hockey, but like “#kdsjhdhks hockey players have nice butts i can’t even” and no one would ever use it for organization. However, tumblr only let the first five tags on a post be global, so people added filler tags if they wanted to stay off search, ergo “sports, /, //, ///, ////, hockey.” We could do something like that (I would like a higher limit than five public tags, maybe 10? It does cut down on spam after a certain point) but formalizing it into three tiers when there are other privacy settings might be getting a bit byzantine. People do use personal tags like “#nonny likes hockey” for organization though, and if we have public canonicals, you could just be sure to stay off those. +7
  • There’s gotta be something here that draws people to using tags as supplementary text, and I think it’s the appeal of adding “noncommital” text. On tumblr this means text that the OP isn’t necessarily alerted to, that isn’t hard-preserved in the reblog record. I think this carrying over into ao3 is a remnant of tumblr habits. With the other privacy controls discussed, I think tags should be metadata/organization focused, with supplementary text as either a low priority or not at all. +2
  • Supplementary tags are an essential component for me. They allow conversation without sidetracking from the posts direction, or quieter commentary, which, while public, needs to be sought out and won’t simply be carried forwards by future reblogs by default. +4
  • I’m in favor of these needs being met, and met well, but not in tags specifically. We do need a space for quiet, ephemeral commentary.
  • I think that some in fandom, and some users in general, will use ALL capacities of the tools they are given to express their personality or creativity. You can’t really attempt to designate a part of it, like the tag system, as purely functional. If you give people a bit of UI that can express creativity at all then some will use it to make jokes, some to be obnoxious, some to tell stories etc.+5
  • If you really want to hamstring tags as marginalia, not only do you need to provide an alternate space for marginalia to thrive without pushing it into the spotlight, but you could make tags just a little bit more annoying to use, like not allowing spaces. #deanwinchester is as useful for organization as “#Dean Winchester,” but #thisisadeanwinchesterssadeyesappreciationblog just doesn’t work as well. If you format the new marginalia area at least a little like tags in some way and put the input/display near the tags, people might get the idea. +1

I’m coming at this from an LJ/DW mindset, but “just not make that post public” really seems like overkill to me. There needs to be a way to make content visible to anyone who comes into your space to see it without turning on the fire hose. Maybe have a toggle to designate all tags on a post as Shareable/Global or not?

See Tags as Metadata vs Whisperspace

  • Fans use tags as “whisperspace” on tumblr in part because comments are perma-added to the reblog chain. Even if we didn’t implement that (though I’d like it, actually) people would still use whisperspace to avoid appearing the OP’s threaded comment section. An ephemeral comment area that can’t be replied to and doesn’t follow reblogs would fill the function fans use whisperspace tags for, though how quickly habits would change I don’t know. +19
  • Also people use tags as whisperspace because they started writing tags and it turned into a tag novel, or because of tags-as-footnotes +2
  • Separate into ‘tags’ and idk ‘notes’ or ‘myfeels’ or something? So make one explicitly a ‘here is how we organise content’ and have another space to enter stream-of-conscious/squee? +7
  • Just because tumblr has a (totally unintentional) affordance for whisperspace doesn’t mean a tumblr replacement has to implement the exact same thing. Tumblr whisperspace grew out of tag searching being totally broken; splitting it into metadata vs whisperspace is going to have a very high learning curve for new users and people aren’t going to use it the exact same way they use it on tumblr. Consider letting it go. +2/-4<people use nonfunctional tags on other sites too, eg ao3, twitter, though
  • Some people opt to give comments on contents by using the so-called whisperspace because they don’t want to officially add anything to the original post for their blog; these comments/feedbacks are the stuff that fuel content creators. It’s a hard thing to do in Tumblr precisely because Tumblr only has metadata that is not immediately accessible to content creators.
  • Content creators adapted to that first with Xkit making the retags visible, then tumblr itself finally offering some of that functionality natively, in recognition for how it was used. I think if a feature catches on that powerfully, even overflowing onto sites like AO3 where it doesn’t even offer the advantages it did on tumblr, it fills some kind of previously unmet community need. I agree that OP should have access to all additions to their post, however, and I don’t think that conflicts with whisperspace needs, since most people on tumblr knew that content creators were monitoring retags. Avoiding OP entirely was never the idea, if they wanted to do that they’d make their own post.
  • Tags as whisperspace also serve a function in making reblog chains more manageable. Though this might be eliminated in a pillowfort type model. It keeps posts from getting too long and convoluted because if you just want to say something like “nice post”  that doesn’t really add to the conversation. You would put that in the tags and not on the original post and you would add to the original post if you had something to add to the conversation. I remember in like 2012 on tumblr using tags as whisperspace was not a convention and it definitely impacted the user friendliness of the site because every major post had a ton of random comments on it that you had to scroll through. I do like the xkit feature where you can see the tags in the notes though.
  • Ability to search on multiple tags. For example, you want to see posts (yours or all) tagged with both "Marvel" and "fanart." +54
  • Ability to search on multiple tags including exclusions, ie “+Marvel +fanart -Thanos” +54
  • Ability to search your own page for the tag as well as the global site for the tag, independent of each other. +32(I literally just found out today Twitter doesn’t have this and I was flabbergasted.)
  • View and click through all tags a user has used. +12/-2
  • If all tags used, ever, appear on people’s blogs as a sort of table of contents, this could definitely nudge whisperspace into some new, provided field and out of tags, since putting it in tags would fill your blog with useless, single-post tags.
  • I think this feature would get unmanageable really quickly. +1
  • Depends on how you tag.
  • Ability to bulk edit tags. +38
  • Clarification on that: if I tagged 100 posts “mystery pearl” (a ship name) but then we learn the other half of the ship’s name so the ship name changes, I can change all 100 posts from the old tag to the new tag at once, instead of manually editing each one. +28
  • Also be able to select multiple posts and add / delete / change tags +2
  • Toggleable ability to “read more”/view full list of tags over a certain length +2
  • Ability to view all tags that have been added to a post without going to each individual page, similar to what xkit allows you to do on tumblr when looking at notes on a post +8
  • Common content warning tags being able to be selected from a list and added to your tags, or some other way of prompting their use and suggesting some common ones +4
  • Ability to ‘save’ a handful of tags you use most often for easy reference/addition to posts (e.g. might be to have ‘queue/art/personal/such good doggos!’ show up either near the top, or in a separate ‘common tags’ list).+4
  • Tumblr used to store your previously used tags and suggest them when you tag a post eg. you could type “ha” + enter, and it would tag as “harry potter” (assuming you don’t have any other previous tags starting with “ha”)
  • On tumblr there’s an issue where tags become unbrowseable if they contain hyphens, which is incredibly inconvenient for obvious reasons (e.g. if I tag a post “obi-wan kenobi” I can’t look for it again), we should make sure tagging allows for a variety of different text characters (hyphens, periods, quote marks) for practicality reasons +3
  • Don’t forget “/”

  • Tag search is viewable by ‘most recent’, ‘popular in the last day’, ‘popular in the last week’, ‘popular in the last month’, ‘popular of all time’ (like reddit or lofter) +16
  • I do also like Tumblr’s dedicated tag system (unlike their search) where you can just scroll infinitely through a particular tag, though, so maybe it can be organized with an option to just scroll through a tag like normal before filtering through popularity? +10
  • The main difference, on tumblr, between /tagged/ and /search/, is that /tagged/ displays posts where the exact string was tagged in the first five tags (if we did something like this, please make it more like 10) while /search/ finds any of those words anywhere in the body of the posts or any of the tags. So the /tagged/ view of “art school” shows only posts tagged with “art school,” but the /search/ view of it shows any post where the words “art” and “school” both appear anywhere in the body, and will get you drawings of schools. Regardless of what kind of search features also exist, and searching within tags sounds absolutely lovely, I do want dedicated tags that just show stuff tagged prominently with that and no more.
  • But unlike tumblr’s infinite tag viewing page, make it optionally paginated +10
  • Chronologically, reverse and normal would be awesome for tags.+3
  • Agreed, but can we make “chronologically” the “normal,” with “popular” as something you have to select? It can remember your choice if you want.+1
  • Browsing top posts in a sub like that is one of the handiest things about reddit. +1

  • Tagviewer ability like xkit. Good for if you want to read the tags people tagged a post with, without having to open each and every reblog. Especially nice for artists and writers who want to see how people are reacting to and tagging their works. +26
  • I think this is dependent on how exactly the tag system will be implemented. Like, there’s currently a discussion above on tagging vs metadata vs whisperspace that is pertinent.
  • It’s a good point to bring up; to answer, sometimes I check tagviewer because the version of the post is not tagged at all and I’d like to know from what fandom the fanart or screenshot is from.+6
  • This is a really good point, and shows the functionality beyond whisperspace!
  • Sorry, I did a skim-read and have a general idea of the discussion, but tagviewer should still be possible no matter how the tags work, right? It’s just like showing all the tags people have put on a post when they reblog a post. Ahhh unless you mean that by only using formal tags, people won’t be making comments in the tags anymore, so there’s no point in having a tagviewer?
  • Yes, exactly :) I think a tagviewer ability is going to be dependent on what kind of tagging system we get and how the users end up using it, so we may have to wait and see what happens rather than having it be implemented from the beginning.
  • Okay, thanks! Put me down as supporting having a tagviewer if whisperspace is still allowed. But if whisperspace is taken away, I’d like a working comment section instead. <3 +7
  • I don’t think whisperspace will ever die, or can be killed without actively limiting how users can tag in a way many users will find limiting, even those who don’t use whisperspace, but I think we might be able to choose whether whisperspace/marginalia exists as a type of tag or as a type of comment/reblog addition that isn’t a tag. Whatever type of whisperspace/marginalia we create a place for, I agree that the ability to view all of them on a post should exist -- though like tumblr tags, one shouldn’t be able to reblog with their content included or reply to them directly, though no one can really stop anyone from copy/pastes or screenshots. +6

  • “Discovery” type page to see people outside of your feed if you want -- or a way to see friends-of-friends ala LJ/DW. +26
  • Wattpad has a networking system of seeing what your friends are commenting on that has allowed for discovery of those posts and new authors, maybe something like that?
  • Only if it’s opt-in. +20
  • I assume it would be a page you would choose to visit, separate from your timeline/feed/friendslist/dash. +7

  • Blacklist +95
  • Of both tags and post content. +29
  • Make it an option to Blacklist only on tags/flags, or on both tags/flags and post content +9
  • Make it an option to automatically blacklist content that has been reported (with the ability to select which type of report you want blacklisted i.e. blacklist content reported for spam but not for nsfw). +2/-4
  • Content often gets reported while not being in violation of the thing it’s reported for; either on purpose or by people misunderstanding the system/TOS/accidentally clicking a thing +1
  • Not of tags and random words appearing in the post, but of tags and flags that function like archive warnings on AO3. So for example you could blacklist the flag for “noncon” and also blacklist tags like “daddy kink” since that’s unlikely to have a global flag. But your blacklist would catch noncon posts flagged as noncon but not containing the word or tag noncon. +7/-1
  • With the option to view individual posts (see: ye olden Tumblr Savior, in which the post doesn’t disappear but hides, and you click on it to see it.) +13
  • But you can see the tags on the hidden post so you get an idea of whether you want to click on it or not
  • Absolutely--what it does with the posts should be configurable. Some people want a heads up and clickthrough, some want total hiding, both should be possible.
  • Wouldn’t that require ‘archive warnings’ like on AO3 then? Would be a massive undertaking for any kind of support/abuse team to keep on top of users actually doing so..
  • I mean, AO3 does it and that’s all volunteer hours…
  • For this to be useable you need to be very specific and very limited as to what those required ‘warnings’ then are, as AO3 is with only 4 required warnings. There’s also a large scale difference between a social platform and AO3, and thus the amount of volunteer hours (even paid hours!) to enforce such a thing would be massive. +6
  • Most things posted to ao3 take time and effort to make, so I don’t mind having to go through all the prerequisite tagings and warnings, but something more for social media use shouldn’t require you to choose the “n/a” option for 5 different things if you’re just saying, “I had a good day at work today! Look at this photo of my cat on my coworker’s laptop!”
  • Nah, they’d just be empty ticky boxes then. You might not even need to expand that section.
  • Why would you want the noncon flag/warning to be treated differently from the noncon tag? They both specify the post contains noncon, there’s no reason to treat them differently in terms of blacklisting. +1
  • Maybe some are discussion of noncon rather than actual noncon content? +2 (typical for oppression terms  eg ‘racism’, ‘misogyny’, often used in discussions of it)
  • Yes but then you just discuss it/use it in the post but choose not to put it as a tag. (Also, surely no racist person is going to be nice and helpfully tag their views as ‘racism’)
  • One way to deal with this kind of distinction would be to have two kinds of blacklisting, one for tags and one for just the word in text. That way the tags/flags are for what is being depicted and if you don’t want to see discussions of something you can block that, too.
  • Could be discussion, could be wanting a public post but to stay off the public tag for that to reduce exposure/scrutiny without making it so no one at all can find it, or even because you think your post, while it needs a warning, isn’t going to be interesting or seen as a good contribution over at #noncon and you don’t want to spam an actual tracked tag. Also helps to distinguish “platform warning that’s enforceable” from “things I warn for to be courteous, but no one is required by the platform to warn/CNTW these.”

  • Timestamps +52
  • (Seriously, just bring in the Xkit people and make those options standard opt-ins.) +25
  • Options for either absolute (monday december 3rd etc) +1 or relative (two days ago) +5
  • And then how to deal with timezones? Timestamps relative to the reader +2, or the poster?
  • I’m sure that a really easy bit of code converts all time stamps to viewer!local or else why does everything ask what timezone I’m in? If this isn’t a thing, obvious answer is all dates/times in NZ timezone as only correct timezone ;p +2
  • Timestamps on OP showing when it was posted (to prevent out of date info being spread as if it’s new) (instead of how tumblr does it, where it just shows the time it was reblogged) +5

Privacy & Content Controls

  • Blog visibility (mutuals only, public, accessible only with a password.) +47
  • On tumblr, accessible only with a password made posts from that blog not show up on your dash at all, you had to visit the blog itself each time. This made that feature useless for just about everything except “soft-deleting” a blog you were done with so that only you had access to the content. It’s closer to a deletion feature than a privacy feature.
  • As a more useful privacy feature, you make it so you have to individually approve or even invite new follows. However, DW-style per-post privacy and circle/access settings would make this redundant, as you could simply only allow access to people you chose and make your posts friends only. Even LJ-style filters would make it redundant, as I could make my posts only visible to a group of users I hand-selected who followed me. So it’s pointless if you create such robust privacy settings.
  • That said, I would actually prefer an entire-blog “invite-only” setting to LJ/DW’s filter settings, which in my experience just made wank worse. +2/ -5
  • To clarify: I’m fine with simple filters like “people I follow only,” “people who follow me only,” or “mutuals only.” It’s the making hand-picked lists of specific people while leaving some of your mutuals out of the loop that in my experience caused a lot of talking behind people’s backs and wanky gossip stuff I could do without. The only good thing about it was leaving people out of content they don’t want to see, but blacklist does that better. -1
  • I found that to be one of the most useful features on LJ/DW and while I’m certainly not discounting your experience I never dealt with any drama relating to filters in the 13 years I was active on LJ. I found it incredibly helpful to be able to filter both my reading and my posting by fandom or just the people I most urgently needed to be in touch with. +1
  • Could you have invite-only posts as well? Or posts anyone can look at but only certain people can comment on/reblog? +3 / -1 (like for rp)
  • Staunchly against granular controls for who can reblog anything, beyond blocking people. It’s a permissions disaster because a reblog essentially becomes part of someone else’s blog: it’s their post now. So either it’s going to get out of your control in a more convoluted way than before (but still get out of your control) or it’s going to be pretty useless since you broke the main functionality of reblogging, and would have been better off just using comments for a select-people-only discussion. Posts only certain people can look at, e.g. for RP, could be handled by either the RP having its own community that the RP is done in, or having RP sideblogs all friend each other and go friends-only. Though one possible useful exception to the “only public posts should be rebloggable” rule is self-reblogs--because the permissions don’t change. Since communities have one unified set of permissions for view access anyway, you should also be able to reblog private posts within a community--but only to that same community. The main benefit here is that you can add content while putting it on the top of people’s dashes/timelines, instead of a comment which you’d need a notification on that thread to know about.
  • I agree with this but also would not want restricted visibility to ever be the default because then there’s a lot of inaccessibility and restriction of content for browsers that happens more out or inertia than intent. It means that if someone goes back years later or a blog gets orphaned that, even if the runner wouldn’t care really, it might be hard to access. I’m thinking partially of comms on livejournal  that when they were ‘live’ were barely restricted at all in intent but then years later become impossible to browse because it relies on someone okaying you. +2
  • Levels of "mutuals" so you can grant access to people without actually having to follow them. +6
  • Split "followers" from "access to locked posts" like Dreamwidth does. +73
  • Access lists would be great, ie ability to lock one post to “personal” list, another to “politics” list, another to “fandom” list, etc +18 /-2
  • If this is implemented, the ability to choose a type/s of access list when you follow might be nice? >When you follow? How would that work? +1 / -1
  • Please no. Splitting following and access is good. Sometimes your friend is in a fandom you don’t care about, but you don’t want to give them the complete cold shoulder. But I don’t want “some mutuals are more mutual than others.”
  • Fair enough - I was just thinking it might be nice to follow only someone’s fandom content without also reading all their politics posts (or vice versa). That said, I’ll take a functional blacklist from the get-go.
  • Back in LJ days, this could helpfully spare people content they weren’t interested in. Blacklist does that better today. If you really want to have separate sets of mutuals for completely different blogging topics, running multiple sideblogs seems an easier way to manage that, if we do get full-featured sides. -1
  • Counterpoint - some of us want to be able to spare our mutuals from whatever our latest fannish passion is but don’t want to have totally separate blogs for totally separate groups of people. I find that the sideblogs lessens the feeling of community/connection with an actual person :/
  • Can’t they just blacklist your fandom tag if they find it boring? That’s even more frictionless, you don’t have to assume for them or talk about it with them, and that way they can un-blacklist it if they end up getting into that fandom later. I don’t share fandoms anymore with a lot of my older friends and this is what we all do. The effect is the same anyway, they’re still not seeing those posts, except the person locked out of those posts is the one in control of their own access.
  • You may also want to consider that there may be mutual following situations where you’re fine with X mutuals seeing your fannish content but not your personal stuff and Y mutual

  • Privacy settings at all levels of posting. +68
  • Choose privacy levels for each post (this is private; this is mutuals; this is public.) +40
  • Privacy settings at a minimum including "public; friends/approved followers only; private." +42
  • Per-post settings for a post to be visible to only a certain list of users. +16
  • (If you mean public vs. friends only, definite +1. If you mean freeform lists of users, -1.)
  • Ability to make your entire blog have a default privacy setting other than public. +18
  • I would divide it as: 1) visible to everyone and on public tags/search, 2) visible to anyone with a link or who follows me but not visible on public tags/search, 3) visible only to my followers or mutuals or something, 4) nobody but me knows it exists. +5
  • That said, limit ability to fine-tune among your friends who can see your posts. Only people who follow me is fine, only people I follow is fine, only mutuals is fine, but “only these six mutuals who hate that bitch we’re fake mutuals with” nah. +3/-9
  • This has been used in the past to filter out people having a birthday surprise organized for them, or to make sure your IRL friend who isn’t fannish but found you can’t see all your dirty porn +4 /-1
  • People either gotta either hide their blogs better or be better at saying no, even if it’s awkward. I’ve told IRL friends I don’t want them on my fandom blogs, it went fine. I don’t want a feature that enables lying to your friends just because some of your friends might not be fannish so that’s okay. It’s a feature that’s mainly for lying to your friends, that gets used for fandom friends too and it’s a bad feature. The birthday surprise thing I’m sure there are other ways of doing? -2
  • The filtered lists on FB as a means of curating who sees your content can be really important for people who only have that platform to stay in touch with people and need to customize who does and doesn’t see their posts. I realize this conversation is mostly focused around fandom but you can’t build a platform and assume it will only be used for fannish things, and discounting other potentially useful features because it can be used to be mean isn’t a useful rubric for designing a system. +1
  • Good point, Facebook has the least drama and wank of any social media. /s But more seriously, I’m not that interested in something that does exactly what Facebook does. People who have non-fannish needs that Facebook meets have Facebook. And there are obviously still ways to lie and whisper and gossip and evade in various ways. People can talk in PMs/DMs. They can whisper about you on other platforms, like Discord. They can make a sideblog/community just to be Mean Girls (or Mean Gender-Neutral Youths) in. Most of these could be used if you really did desperately need a feature like that. But I don’t want it to be that easy, that built-in as a feature. I don’t want the platform to ask you, “Do you want to lie to your friends now?” Make people work for their drama at least a little.
  • Look, if my years on LJ & Fandom Wank taught me anything it’s that even something as simple as “friends-locked/mutuals-locked” inevitably winds up bringing all the drama llamas to the yard. Just because a very useful feature could potentially lead to interpersonal drama doesn’t mean that it shouldn’t be included.
  • This isn’t a feature that just could be used for drama, this is a feature that basically only has drama left as a reason for including it now that the non-drama benefits of it are better met by other, more modern features. I know that any privacy setting at all can be used for drama, but friendslocking has enough non-drama benefits to be worth it. I know that DMs/PMs can be used for drama, and sideblogs can be used for drama, but those have enough legitimate uses to be worth it. I know that reblogging, for that matter, can be gasoline on a trash fire of drama, but that also has enough legitimate uses to be worth it. This feature is just a drama enabler, and I guess, a birthday party planner, for how often that’ll get used. (If you guys want to brainstorm a less wanky feature for planning birthdays, I’m all for it.) Trust me that this also comes from a lot of experience with LJ wank. The wank potential/actual usefulness ratio on this is just bad. I feel that tumblr anti abuse is fresher in our memories and we’re doing a better job engineering against that, but I remember F_W too, I remember FFR (and FFRR), I remember sf_drama, I remember a lot of other wank and garbage, including stuff that never got written up on wank comms and most fans don’t know about. We should be engineering against the worst of our nostalgic past, too.

  • Mute conversation feature like Twitter for popular posts. +46(one that actually works, unlike Twitter's :P) (Xkit does this too.)
  • To clarify the contrast between this and the mute two points down, this is to hide a post so that if all your friends are reblogging it you can make it stop showing up on your dash, while that one is to hide all posts by a user.
  • Instead of doing this on a per-post basis, can we make it so that reblogs of posts you’ve already seen don’t automatically show up in your feed? (Opt-in or out of reblogs you’ve already seen showing up?) +5 / -1 I’m not keen on how Twitter does this automatically, especially when the person goes on to respond to the hidden retweet. Opt-in/out might work.
  • Can this only automatically trigger if there’s no change in content? I might want to see discussion additions, but I’ve already seen that kitten video, it was cute, I get it. Also an expiration date on it (like a year?) since gandalfihavenomemoryofthisplace.png. Never hide the version it let you see the first time (i.e. that exact reblog on your dash) and never hide the OP when viewing OP’s blog.
  • What about muting certain branches of replies on a post, like if there’s a really cool discussion going on in one corner and in another it’s devolved into bickering you don’t want to hear
  • This is achieved with ‘everything that follows this comment is bad, but everything that follows the comment above it is not bad’ sort of deal. +2

  • Ability to show/hide who you follow. +30/-1
  • But with the caveat that users can opt out of being shown on other people’s (globally, not individually) “people I follow” lists. For example, some people have blogs such as say, a trans health issues vent blog, that they don’t want everyone browsing their BNF friend’s profile to learn about the existence of, but they want their BNF friend to be able to follow them, and their BNF friend wants to show who they follow. +13

  • Block/mute features that actually work (i.e. if a person reblogs a blocked/muted person it won’t show up on your feed.) +114
  • Ability to block users/posts from all your blog identities at once. +16
  • A big problem with Tumblr’s block function is that you still see reblogged posts originally authored by them. This should be fixed. +5 [already part of suggestion description above]
  • Separate “mute” (hide all posts by this person so I don’t have to look at them) from “block” (don’t allow this person to interact with me). So for example I can mute the person who ships my OTP in a way that annoys me, but let them keep following me and reblogging my posts if they like MY content. (Good for friends whose blog styles you hate but you love them.) +12
  • Muted blogs will still be muted even if your bestest friend mutual reblogs from them. You won’t see that reblog on your dash/timeline/friendslist. +14 [already part of suggestion description above]
  • Have this as a toggleable option +4
  • Make it impossible to tell if you’ve been muted. It needn’t hurt feelings, it’s just curating your experience. +34/-1
  • “Limit people” option like what xkit allows on tumblr - if someone makes more than 2 (or other number) consecutive posts it hides them unless you click to show it +5
  • Have it both by post number or by ‘X amount of posts in X amount of time’, for those who have 20+ posts in a row in ten minutes and you see half of them because someone else reblogged something right in the middle of their excitement
  • Can this be by tag? For example, your friend is really excited about the new Star Wars and just made approximately 90 posts about it, you only see the first 5 because you think Star Wars is okay but calm down, but then when they post a completely unrelated thing that comes through fine. Wouldn’t help much for people who don’t tag anything, though.

  • Public blogs should be visible to non-signed in users (don't have to have an X to see stuff on X.) +54
  • Multiple levels of “public,” e.g. the ability to visit the page as a non-member and see only posts set to “public” +16
  • Pinterest is a very annoying example of not having this in place -- even if you come from an embedded post, trying to click through will FORCE you to make an account/sign in before you can view the content [unless you have the removable overlay extension on chrome...works like a charm]. +10
  • Keep in mind that this will mean web (what do you mean by this?) (maybe they mean… the non-fandom internet populace can see?) +1
  • Yes, I see what you mean. I was going to say that privacy is what friendslock is for, but if those posts aren’t rebloggable or accessible within the site itself, so that’s much more hidden than something everyone with an account can see and interact with. This also affects (or could affect, depending on how it’s done) search engines, scrapers, roving analytics, and other robots, which presumably won’t have accounts if we can help it. Tumblr does have an option to only be visible to logged-in users, and that hides you from Google as well. Might be especially useful for when you want the fourth wall to be sturdy, such as keeping celebrities from self-googling RPF, or generally trying to avoid being right in the face of brands/celebrities who aren’t giving fandom space. Content creators keep interacting with fanmade content on Twitter and Instagram, and not everyone likes that. It’s similar to the option to archive-lock, on AO3.
  • Related: Toggleable search-engine indexing, something I really appreciate about Tumblr. +1

  • Good faith effort invested in user data privacy and security testing/GDPR etc. compliance from the beginning. +72
  • The “how to stay out of jail and keep people from getting doxxed” feature. +15
  • NO TRACKING or anything like it. Only use cookies to keep us logged in. +10
  • Exceptions are probably okay for things like which browser / monitor size, for statistic data viewable only by the team to improve experiences. +2
  • Would cookies help with any of these granular “edge” controls we want? I know the standard way to do it is to cache it nearer to the user or something, which is expensive, and cookies won’t transfer between devices, but I don’t know, is this a thing, could it save money?
  • Disallowing the posting of information that identifies where you live. I don’t know if this is something that people would hate, but especially if we’re considering allowing minors onto the site, I don’t think being able to post anything more than regional information should be allowed. -6
  • Not sure if this is in any way practical. People do like to talk about their actual lives in addition to fandom, and meet-ups are a time honored fannish tradition. +1
  • I used to participate in sales comms on Livejournal and have done so again on Tumblr. We paid actual money and shipped actual fannish items to actual addresses, seems like this would ban transaction-critical info relevant to several hobbies (like, how much it costs to ship to where you live). Platformwide ban is way overkill even for minors, at least as much as banning anything above zip code (necessary to calculate shipping cost before making an invoice) +1
  • At the very least disallow “outing” info like posting another person’s info. Basically don’t allow doxxing. +4
  • AO3 has a good policy on this that I think is worth taking over, especially because in fandom we also have the added issue of (potentially) wanting to keep our fannish and RL identities strictly separate +1
  • Don’t post your own zip codes etc etc. Not the same as saying “I’m from Southern California” or what have you. Mainly I’m worrying about kids posting stuff about themselves. They don’t experience fear when it comes to revealing their information. Don’t post what high school you go to etc. To prevent targeting.
  • How on earth would you enforce this?
  • As a kid (16) we do have a sensible level of caution when it comes to internet safety, with the exception of really young kids like preschoolers and elementary schoolers. (But I’m sure they all don’t. We adults get alarming cautionary tales from time to time) +1
  • Is this supposed to be a “strongly discouraged” or a “warnable/bannable offense” kind of thing? Because I wouldn’t feel comfortable or safe on a website where saying, “I live in Miami” might get me banned. Plus, what about RPers, who want to be able to specify where their characters live? If location specification is bannable/warnable, that makes RP a lot more difficult. +9
  • There should be a “fictional profile” setting for RP blogs! I mean I agree with the rest of it anyway, but that’d just be neat, for example if location were more supported and I wanted to, say, look for users in New York City, I wouldn’t want to get more RP blogs than people. +1
  • If worried about kids specifically, could make the feature limited to blogs of a certain age. +1
  • Make it clear that region is okay, no closer than province / state / prefecture / etc for account owners. Character locations are fine, but not for real people. If you want to get closer (say two different cities if you found out your friend lives close), take it to Discord or smth. +1 / -6
  • This just seems really messy and complicated and hard to enforce and adjudicate??? +7
  • Once you get into this level of content micromanagement, you open yourself up to a whole host of legal liabilities and costs of monitoring and enforcement. May as well just make the site 18+ only at that rate. +3
  • Actually, I kind of want the exact opposite. I mean, don’t post your own full home address in your bio, probably, but that isn’t a use case I really foresee anyway. But when you mentioned location, I actually thought how great it would be to filter for users close to me. I shouldn’t be able to see their addresses, but maybe being able to search like, “within N miles/kilometers from me”? (And obviously not being exact enough to allow multiple users to triangulate someone’s house or anything.) This would all be 100% optional and opt-in, but you know, I’d love that? I meet all these awesome people and they’re all around the world, which is lovely but the times a fandom friendship has blossomed into something IRL have been so wonderful for me. It’d be so damn great to be able to search for users near me and see if any of them seem like people I’d like to be friends with! If you want zero information on where you live, good for you, protect your privacy. But if adults want to meet other fannish adults nearby? Let them be friends! +1
  • Also, this is really restrictive when it comes to talking about your life, like a lot of college/uni students might find it relevant to say what school they go to because it’s part of their life. If you got into Cal Arts and are training to be a professional illustrator, you’d talk about that. If you just moved to a new city and want to post photos of the historic landmarks, that’s normal stuff. +1

  • Like Tumblr, a way to denote your blog might have explicit content and a way to opt out of seeing it +50
  • The ability to do this at individual post level. +41
  • If minors are allowed on the site, anyone who has disclosed their age as being under 18 should not be able to view or post explicit content. +8
  • Wouldn’t different countries have different age limits though? Perhaps it’s perfectly legal for someone in country X to view or post said explicit content? +2 → what matters is how the site’s TOS defines this. We don’t need this level of bikeshedding here, y’all, I’m sure there will be someone handling content policy at a later date if this does go forward.+1 → I didn’t mean to get into a TOS discussion here, I mostly meant to point out that this is a TOS/content policy discussion and not a feature discussion (but I might have gone about that the wrong way). But agreed this thread isn’t really useful discussion to have currently/here
  • This will probably be by where the admins are located - if it’s US, they have to follow US law, if somewhere else, then that law. So the admins don’t get taken to court, y’know.
  • Users over 18 should be able to choose between “hide all the explicit content and don’t tell me about it, “tell me it’s explicit and let me decide,” and “just show all of it, never warn me.” (And be able to select ‘everyone’ or different settings for different blogs) +24
  • Would it be possible to make this setting device-specific? Like, I don’t want to see nsfw anything on my phone because I often browse on my phone in public, but my computer rarely leaves my apartment so nsfw content is fine there. +1

  • Ability to opt-out of being tagged in a photo +3 (is there necessarily photo tagging? Description in an image for accessibility purposes ≠ tagging.)
  • Vote for no face-tagging. +70 +2032 I couldn’t find a previous version with an accurate number for this so I’m going to say 32? If I was trying to goose numbers I wouldn’t be telling you this. +2
  • I don’t mind my friend tagging me “hey pal look at this fanart of a thing you love!” but I never want to be tagged as “this photo contains you, Friend1, Friend2”, by bot +1 or by humans. +35
  • This was my suggestion but I realized it’s restrictive for cosplayers who want to credit each other in group photos. Maybe instead of being tagged by the OP, you can tag yourself as being in someone else’s photo (which OP would have to approve to be visible). That way you don’t cross boundaries tagging other people--but you also don’t get weird fictionkin tags on screencaps. Or instead of being tagged as “in” the photo (which is a very Facebooky feature anyway) it could simply be an attribution thing, where collaborations can have multiple “sources”? Like two cosplayers and a photographer are all content creators of a cosplay photo post. Two artists and a writer are all content creators of a webcomic post. That feels like a more fannish way to give credit, and removes the temptation for people to tag themselves as “in” anime screencaps, lol.
  • Would we still be able to tag ourselves as being in photos (like a selfie tag)? What about tagging cosplayers in group photos? (I might be confused here. I’m assuming we mean tags like metadata tags. If we mean those little “identify a face in the picture and having it lead directly to the person’s profile” thing that facebook does, hard pass on that).+3
  • Semi-related, can we make it so i can ping someone to a post, without having to do that publicly?  So we don’t get 300 @blank comments on every post like facebook? +8/-1
  • You can just send the post through private messaging +2

Integrations & Export

  • Crossposting to/from other platforms: Twitter, WordPress, AO3, DW/LJ, Tumblr, etc. +31
  • API Hooks for apps such as IFTTT. +9

  • Ability to export personally authored or updated posts (ex., not reblogs) +34
  • With metadata. +20
  • And comments +14
  • Where are we drawing the line between personal authorship and comments… What about communal fic posts, with multiple contributors - allowing only the OP (potentially a prompt blog) to export the post, including other people’s fills is a problem. +1
  • Why can’t we be able to export our reblogs as well? When I back up my tumblr, I want everything, because I use it as a way to not just produce my own content but also save things I want to see later. +1

  • RSS feeds (whole individual timeline or by tag.) +15

  • Google Analytics compatibility. +6/-4
  • Giphy compatibility (might save some space if users can insert something already hosted on giffy instead of everyone re-uploading all the reaction gifs)  +7
  • API with programmatic access to everything a user can see. +4
  • This is absolutely vital for accessibility. People seriously underestimate how badly Tumblr's lack of a robust, consistent API has hurt its adoption rate - too many people simply cannot use the website or app.
  • Full Federation -- any person with some technical skill can stand up an instance of the service on their own hardware, and any user on any instance can fully communicate/interact with anyone else on any instance. +2/-6
  • Makes moderation IMPOSSIBLE, has implications on blocking, has legal implications with not all content being legal in all countries, if someone else is running it they wouldn't necessarily be under the same TOS.
  • It adds additional difficulties with regards to moderation, and you’re absolutely right that blocking needs to take into account the safety of vulnerable groups. It offers one advantage that a single centralized platform doesn’t, however — it allows more resilience towards events like Tumblr’s imminent collapse and LiveJournal’s Russian takeover and eventual homophobic TOS, when the entity that owns a particular platform, say, gets bought out or acquired. I’ve seen this same song and dance too many times in the past decades and I’m sick of it. There has got to be a better way to organize. +3/-2
  • The issue with tumblr and livejournal wasn’t that they were centralized, it was that they were owned by either private individuals or shareholders.  Stick this project in an organization led by people who are elected by users, and you get all of the benefits of centralization without having to worry about getting sold out. +1 (This is a "we own the damn servers" issue)
  • Make the code open source, but remove the "communication between instances" options. That way, you want it badly enough, you can run your own and the rest of us don't get harassed because whoever's running them just goes "Lol, I don't need a TOS or enforcement" and has nazi shitlords in their federation. It's essentially like giving 4chan a megaphone into your living room and then expecting each individual user to block them rather than STOPPING THE PROBLEM IN THE FIRST PLACE. +1
  • In addition to that, we're very attached to our pseuds in fandom. If you've got a Phoenix@Twitch and someone comes along and is Phoenix@twitcll, tell me there aren't going to be problems with impersonation. With a single instance you can control unique usernames. This is so ripe for abuse.
  • I agree with all the reasons against, but at the same time, wouldn’t a distributed network like this lower expenses of any one host significantly? Wouldn’t it share the burden of keeping the whole thing running? Are there other ways to somehow share some of those burdens, perhaps even on fan user-hardware, to make a more distributed, less top-heavy network, purely for expense/logistics reasons? I realize this sounds like a security nightmare, but the other way is a financial nightmare, I’m juggling here.

Revenue & Income (survivability)

  • A sustainable (not ad-based) revenue model. +85
  • Basic functionality must be provided for no charge -- extra services (more than one name change? more than x sideblogs? certain premium layouts? Gallery of avatars? Reaction GIFs?  Almost anything can be monetized - look at Twitch and their specialty emoticons) can be paid services (like the dreamwidth model when it first got off the air.) +161
  • Seed accounts (ie paid accounts that early adopters chose to pay for) may also help with start-up cash.  That’s what DW did [and what Pillowfort does now]. +45
  • Are seed accounts permanent accounts? (they are on DW) +28 for being able to buy a permanent account at startup.
  • At the very least selling seed accounts can be a quick way to gather revenue at the start of this site, even if they aren't sold anymore later on.
  • DW only offered permanent accounts at seed specifically for the seed money. Denise actually wrote a post explaining why they don’t regularly sell them, namely that selling permanent accounts all the time isn’t a sustainable business model because it relies on infinite growth. I’d actually go old school LJ on this and have an occasional permanent account sale once in a blue moon. +19
  • Sorry, but what are permanent accounts? → On LJ, paid accounts offered a lot of extra features like voice posts, more icons, premium layouts, etc. You could buy a paid account for 1, 3, 6, or 12 months. A permanent account was a one-time payment to have a paid account forever.
  • Friendly reminder based on Pillowfort's current troubles: you need to have a slow drip of incoming users to start out, or be prepared to ramp up really fast if fandom latches on.
  • An invite queue that lets you know upon checking where you are in the queue. +27

  • Offset cost of individual users by charging a one-time fee for brand/marketing accounts. +19/-6 (marketing accounts should be able to afford a monthly fee, surely?)
  • My problem with this isn’t that they can’t afford it, it’s the tacit approval of their presence on the site at all, and the implication that, as they are the customers, they need to be getting their money’s worth off fandom in some way. When brands want marketing accounts, they aren’t paying for a blogging platform, they’re paying for the users. It’s dirty money. Whether or not we can stop them from being there, let’s not get beholden to them or make them feel welcome.
  • What constitutes a brand/marketing account? Like, if say you’re an artist and you want to go freelance and start selling commissions or similar. The line can be blurry there & a lot of indie artists might not be able to afford a fee. +9
  • I’m thinking things like how Sony and Buzzfeed have official YouTube accounts. Basically that if you’re part of a marketing team for a product, you can approach about buying a specific package that will get your account’s content on the reading pages of people who are interested in it. E.g, If Sony bought an account and then advertised a Venom sequel, people who are subscribed to the Venom tag would see it. (And be able to block it, because some people don’t want to see ads even if it’s for stuff they like.) Though this does feel very cart before the horse. +10/ -1
  • Alternately, just let brands in on exactly the same terms as everyone else. Any sustainable site needs to not be beholden to advertisers--and our data needs to not be for sale--but I also wouldn’t want there to be any sort of ‘are you a real person’ test that would keep out accounts like the JSTOR tumblr or the Moon Pie twitter. +4
  • Arguably this whole document is a cart. (+9003 this is an amazing comment)
  • If this service attracts a lot of fans, then it is likely to appeal to social media marketers who make fan engagement part of their content strategy. And I think the goal is to attract a lot of fans!
  • But won’t this cause the same kind of problem as with YouTube and now Tumblr? Once advertisers and revenue get involved, they start wanting to control the content that appears on the site (ie, YouTubers can’t cuss in their videos because it’s not “ad friendly” and now on Tumblr you can’t post nsfw stuff)
  • If we allow advertisers—which tbh I would prefer we didn't (+19) —the platform needs to understand that the corporations are choosing to enter fandom's space and not the other way around, TPTB aren't allowed to control what we post about their media +22
  • There would also be the inevitable back and forth tug over marketing teams wanting information on those who engage, and users wanting their privacy.  For example, they might want information on how long users spend on certain tag pages and what other tags they use. +3
  • No ads. That's what eventually brought down our past platforms. +21

  • Pay for individual Feature Packages such as: +11
  • Ability to hotlink (although keep embedding free) media files (images, audio, video) to other social media.
  • If I pay to upload high-res art/gifs, my art/gifs can’t be downgraded/deleted later due to non-payment, even if I lose the ability to upload new ones. +21
  • +1 only if it's your own media files. No paying to embed someone else's art.
  • Upload GIFsets above N megabytes.+5
  • Change username more than once.  +7
  • With Twitter and Tumblr allowing name changes, though, this has become a standard feature. Shouldn’t we keep to this instead of going backwards to LJ standards? +2/-1
  • if we go with having both a display name and a username, it can work without seeming too outdated. +34
  • This would make it prohibitive to do a name change to avoid a stalker, depending on how other systems (redirects etc) work.
  • Would also make it difficult for trans/NB people who are trying different names on for size
  • Hi I’m trans/nb and have significant name dysphoria, and legally changed that name. This isn’t really helpful to me because my meatspace/wallet name was a whole other can of worms from my fandom pseud. I chose my fandom pseud, unlike my deadname. Most people’s fandom pseuds are not their wallet names--and probably 0 of them are trans with name dysphoria, because why would you do that to yourself? A name others call you, which may be a name you try on for size as a potential wallet name, is much less formalized than a username. Changing that is just a matter of updating your bio or informing your friends. A much bigger problem for trans people with name dysphoria is whenever financial stuff gets linked, e.g. to pay for premium features, and the site somehow attaches their deadname to their account at the hip--forces it into their account info, associates it in a database with ties that cannot be broken, uses it to address them in emails, puts “Hi, [Deadname]!” on a banner at the top of the page. I’ve deleted accounts because they just would not stop deadnaming me. In some cases, even after showing them the court order name change and having it updated in some areas, they still had the old name stuck in various database areas and nothing could root it out, the only way to get rid of it was to nuke the entire account and start over. (Looking at you, Skype! Serves me right for giving you money!) So my request here is this: when you take wallet names as part of payment information, treat them as part of the payment information, as you would their credit card number, and do absolutely NOTHING else with it. Removing it should be as simple as deleting that credit card from your account and adding a new one in the right name. A wallet name derived from payment information should NEVER appear in any non-financial area, and in financial areas only where absolutely required by law to process the payment. If age verification takes place, for any reason, including to verify sex workers, all of this also applies. A scan of their ID should be used for NOTHING more than to confirm age. No deadnaming/misgendering trans sex workers, thank you. And I really don’t care if redoing your username is a paid feature or not. It’s not even close to the same trauma as trans name issues.  +1
  • Customised domain name. +31
  • Multiple / more than N user icons. +9 / -3
  • Switching avatars seems clunky outside the LJ model where urls never changed and so keeping track of people was easy. The only benefit to multiple avatars imo is for RP blogs, in which case maybe having different blog types might be a consideration? Ie personal blogs, shared blogs, community blogs w/ contributions, RP blogs, art portfolio blogs etc. +4/-1
  • But switching icons allows for “reaction shots” in comments, and created its own mini gift economy, where people might create/draw icons for friends, etc.  Allowing different icons at least on comments allows greater freedom of expression and gives people the incentive to pay for larger icon packages (something that was gifted from fan to fan regularly on LJ/DW)+9 <Many people remember this system and would be delighted to resume it I’m sure, even if others aren’t interested
  • Switching icons is also used as a marketing tool for some artists - creating icon sets that spread as more people use them, linking back to the original artist and getting greater visibility +1
  • Custom themes? -2
  • Theme customization is how I make sure I can physically read my own blog (migraines, eye strain, screen reader friendly formatting, etc). I wouldn’t want to see that paywalled.+6
  • See also the theme/skin related threads under User Interface & Accessibility. Certain pre-made/premium themes however could be sold/monetized, as long as the main accessibility ones are free, and users can create their own as well. +2

  • Ability to gift added services or credit for future use of services to other users. +56
  •  By name or anonymously, and even to random folks? (Possibly opt-in for the last.) +24

  • If ads are inevitable, a type of premium account that does not have ads. +84
  • If on free account using an adblocker, don’t use fake ads that beg me to buy a subscription. I get why, but let’s not guilt trip people here.
  • I tend to forget I have my ad blocker on, honestly. I appreciate when sites give me a nudge sometimes. +3
  • I’ll take a nag over completely breaking the site. And I’ll even take a “you can’t view this site with adblocker on” honest screen over just breaking the site silently with a never-ending loading gif. But I do hate all of these.

.

  • Payment processor needs to be one that’s okay with adult content on the service. Paypal, for example, is not. +47
  • Look into who Dreamwidth is using, because they got in trouble w/Paypal for the same reason +5
  • More than one processor if possible, so it’s accessible +4

  • A process for flagging ads for content that, itself, violates the TOS, or is otherwise problematic (funds bigoted causes, etc.) +48
  • If there are ads, the ability to opt out of or into certain topics (no horror movies, yes to video games, no to anything food-related, etc.) +31
  • This is a standard type of warning in indie video game fandom. It could function just like other kinds of content warnings.
  • Similar to Ravelry, if ads are inevitable, perhaps they could be smaller/indie/fandom-specific (contrasted to a random ad server somewhere). To expand on this, Ravelry ads must all be related in some way to the site, and most are from small yarn shops, pattern sellers, etc. Perhaps a fandom-specific version could be applicable to the fandom social media site if ads are necessary. +13
  • Also if ads are REQUIRED to run the site and keep it semi-profitable, ability to sell ads for the site specifically wouldn’t be a bad idea (see, facebook selling ads to anyone who wants to buy one, but not facebook’s inability to deny the ones that are problems) +2
  • Fandom revival of classified ads? Toss the platform a few bucks, get the word out that you need a DragonCon roommate or that you're selling off some of your merch collection, spare everyone the need to read a dozen cross-posts and promo reblogs. +32
  • Holy crap, this is the only ad-related idea I’ve actually liked. I don’t want huge corporate interests involved in fandom. I don’t mind some Etsy seller buying ads for their custom plushes or something. If fans are buying the ads themselves, it removes all my discomfort with ads and even makes me excited about it!
  • I really like the idea of ads being something primarily created & paid for by the userbase of the site! Though I’m not sure if the ability to pay for everyone to see your dumb meme is a good thing or not. Maybe an ad review process.+4
  • Ad reviews sure, but if people want to blow their hard-earned cash on a dumb meme and it supports the site, it’s their money? But I wouldn’t want, say, antis buying space in the Reylo tag to piss in the cornflakes.
  • Maybe an opt-in opportunity to view ads (and answer promo surveys?) instead of paying for a premium account? (+1 for optional ads for perks, -1 for promo surveys.)
  • Okay, here’s a little bit of a wild idea riffing on the above. Say some users pay in exchange to give their posts a little boost, possibly get recommended on people’s dashes similar to how posts were recommended on tumblr. But wait--I know, plenty of people will hate that! So we don’t just force it on all the users. We offer some kind of premium perks (like extra icons/themes/etc) for users who allow themselves to be recommended to. So in essence, the users who want their posts promoted and pay, and the users who don’t pay but allow themselves to be promoted to, are collaborating with each other to contribute money to the site. This is a way for people without a lot of money to support the site without getting disruptive, corporate ads. They’re just seeing normal users’ posts. And it offers yet another way for people who are trying to build a following to give themselves a boost. Blacklist would still be in effect, so you wouldn’t be exposed to things from your blacklist just because they were recommended. +1

  • Some sort of monthly donation opportunity with special membership rewards? (stickers, etc.)(we need a cute mascot like DA’s Llama :D +10) +2
  • This works very well for AO3 and, in my experience, increases positive perceptions of the site - I am happy to donate to AO3 because fandom changed my life for the better. I’d be happy to donate to a fan-owned fan-run site that offers me lovely fannish content.
  • Kickstarter for this whole hullabaloo? +8
  • Kickstarter takes 9% of your money right off the top. What you’re paying for in that is exposure to a lot of people (I’m going to be arrogant and say we probably can do that ourselves), and a centralized place that handles payment. Also, they are an all or nothing crowdfunding site.
  • Kickstarter will get you money (if you’re lucky). It won’t get you the vast amount of people you’re going to need to actually create and sustain a platform -- project managers, coders, QA, lawyers, etc +2
  • Also is not a long-term solution +1
  • Site-themed merch and swag for sale? +3
  • Not likely to become a main source of revenue, but could be a side stream.
  • Could offer redbubble-type option for artists to sell their art and give a cut to the fansite. +3
  • Redbubble artist here! So in order to do that, you’d need a lot of printing equipment and raw materials. Redbubble is print-on-demand that puts your art on products, such as mugs, phone cases, shirts, blankets, and of course, plain old paper prints. This would be expensive to set up to scale, though I’d be completely down for it if you did it! Starting with just selling prints would be easier to set up. Deviantart already does this, though the finances don’t favor artists much. There’s also copyright issues with fanart to consider, which are legally more complicated than the kneejerk “selling fanwork BAD.” Fanworks with really recognizable symbols, such as superhero costumes and logos, are the most fraught to sell on merch. This happens a lot regardless, but it is potential trouble. But not all fanart is necessarily of a copyrighted thing. For example, if I draw something that you and I know is Finn from Star Wars, I could argue that actually, it’s John Boyega. John Boyega’s face isn’t copyrighted. It’s legal to sell illustrations of celebrities. My case is better if he isn’t drawn in a costume from Star Wars, but a costume I designed. I also couldn’t advertise it by calling it Finn, or mentioning Star Wars. However, Disney is one of the worst when it comes to being a legal bully about this stuff, so even though this might be winnable, you might have to fight it anyway, which costs money. Original work obviously avoids this whole problem. And you could also have commissions rather than print on demand merch, or both. It being fandom, though, people will absolutely want to, and try to, sell fanworks. That’ll be a constant thing.
  • Doesn’t Dreamwidth have a Patreon or something?

TOS, Moderation, Abuse Team(s)

  • TOS that clearly outlines what is and is not allowed (e.g. Absolutely No Nazis Allowed +1073)
  • Example: Dreamwidth. +24
  • Clear policy of what will happen if someone breaks the rules, including different levels of infraction and escalations if deemed appropriate. +36
  • Comments/additions to posts should be flaggable separate from the parent post +31
  • No automated bans based on quantity of reports, this is highly abusable. +40
  • Once something gets a certain number of reports, either in number or in frequency, it must be reviewed by a human moderator.
  • Separate rules for user-generated content from the main TOS document, so people can find and study them without wading through 5000 words of legalese.
  • Or, just have a human readable TOS like Dreamwidth does. +22
  • With up-front messaging signalling that it actually is human-readable.+8
  • Proposal for a sustainable way of moderating and enforcing these consequences, segueing neatly into

  • An abuse team
  • This is work and should be paid as such. +2095
  • The Metafilter model may be suitable: paid moderators working in shifts for worldwide coverage. +22
  • Preferably people familiar with fandom should be hired for this. +8

  • If we want to have audio and video, a solid Abuse team is extremely important for any platform to be viable. Otherwise we could have actually terrible content by the truckload uploaded to it: real violence, real child porn, all kinds of horrible stuff. Not to mention copyrighted content, which would get studios all up in arms about it. It’s not easy work :/ +29
  • Develop/aggregate resources/training on Fair Use. (assuming these are US servers?) +10
  • Can we have servers in multiple countries? I quite like the rights protecting my intellectual/artistic creations in mine → A lot will depend on the legislation in your country as well, nonnie, but it’s important to remember that if this site has servers in a country that subjects it to its law just by having a server there, it’ll likely also have to have lawyers there. That will cost money in multiple countries, then.→ Makes sense.+2
  • Also note that specific countries may have local laws that restrict the types of content users from that country see (e.g. country-specific hate speech laws). Other social products have implemented country-specific visibility restrictions on content deemed violating in this respect, to avoid litigation. +18
  • I would love if you made it accessible to people in countries with homophobic/repressive regimes (you can’t control government blocking, but not blocking them on your end) without censoring content to comply with homophobic governments. +14
  • Detailed reporting system: fields for flagging specific ToS violations such as spam, bots, *ism, *phobia, harassment, extremely violent content, animal abuse, doxxing, scripting hacks, plagiarism. +40
  • For anti-style harassment campaigns, where a group coordinates to abuse one person anonymously, when the target realises it's happening, an "I'm being harassed/bombarded" button which flags/logs further messages for review by a human so staff can potentially ban on IP or something? I mean, the solution is for the target to turn off anon, but I'd like to see people who do this get some blowback. +9
  • Definitely an ‘I need help please fix where this is coming from’ button, so the staff can review interactions with that blog / account to see what’s going on and how to help someone who’s under a smear campaign. +7
  • And potentially if they pick a target/move on/pick a different target/move on review all interactions from that IP? Kind of a "this is a pattern of behaviour, you're out?" deal? +3
  • On tumblr, when you get a DM from someone you’ve never talked to before, there’s an easily visible/accessible button to mark it as spam. Having something like this for harassment as well, on any DM/PM/message from someone you’ve never interacted with before in private, would make it easiest to report. Every time I’ve gotten those harassment/bombardment campaigns, it’s a whole bunch of accounts I don’t know. This is a really different thing from say, having a dramatic falling-out with a friend. +1
  • One thing I considered, because of a smear campaign I saw on tumblr, is like...if a specific user is consistently sending in false reports, them getting a warning about the volume of false reports that mentions that they’re wasting the moderation team’s time, and then having to do a captcha to send in reports, and then perhaps if they’re REALLY persistent in malicious reports perhaps making them unable to send reports for x amount of time? Or some other punishment? +13
  • If a user has sent in more than a certain number (5?) of reports that a human moderator has determined were groundless, there should be a much more severe punishment for that behavior, rather than it being tacitly tolerated or merely slowed down. On the other hand, not so if they’ve reported 500 things that a mod thought were legitimately infringing. Perhaps your score of “frivolous attack reports” vs. “justified reports” can’t go over a certain percentage.
  • Having the ability to both tag multiple posts by an offending user for a specific ToS violation as well as tag multiple ToS violations in one or more separate posts. +10
  • Extreme paranoia exercised on the subject of automated moderation. (Using AI to recognize specific known abusive images {beheadings, animal torture etc} as flagged by moderators to prevent duplicates from being re-uploaded ad nauseum would be an example of appropriate use. So that once a hideous image is verified as a ToS violation, any duplicates or mirror images can be automatically blocked forever.) +27
  • Is there AI tech capable of regularly and reliably identifying indescribable content? I was under the impression there wasn't really, yet.
  • As I understand, in the above comment, AI would only be used to match images to ones previously deemed abusive by human mods.
  • Yeah AI can reliably identify SPECIFIC images there’s apparently a whole database, but try to make it identify images it’s never seen before and it just cannot. +1
  • I agree with this, but I also see no problem with using existing databases for identifying known and verified child porn, beheadings, etc.

  • IF automated moderation is used, and there’s an appeal, AUTOMATICALLY reviewed by a human. No questions asked. +41
  • This is sustainable to a point, but once/if the user base evolves beyond a certain number of regular users the volume of submissions will be excessive for a human team. +6
  • Bot-generated flags should always be subject to human review by request. If there's too many review requests to keep up with, the bots are over-sensitive. Bots should mostly be identifying bots +1, or content that, after a human review, gets someone banned. (Or potentially, content that requires a form letter response. But it still shouldn't be so excessive that real mods can't keep up.) +8
  • Could be made out of volunteers that are recruited regularly from within the userbase, like AO3's policy & abuse team.
  • This cannot be a volunteer position specifically because of the potential child pornography content. This will require a trained and dedicated team to either answer issues as reported by users -- a team which has established connections to the appropriate law enforcement and NCMEC -- or it will require a pre-screening of every single piece of content that's published to the site, which IMO is not sustainable. It's possible there's another way, but I'm not an engineer by any means. If there are other ways, it's out of my wheelhouse. +16
  • Could be a mix ala Reddit, where moderators for communities are allvolunteer, but there are still site admins for the really bad parts/etc +4
  • Could potential child pornography be flagged for paid+trained mods to deal with, while volunteer settle smaller disputes? +1
  • Is that sustainable for a full social media platform, especially when this is a for-profit model, unlike AO3?  IMO no. +18
  • We’re still in the batting ideas around stage -- I have no idea what the structure for the organization that takes care of this will be
  • Why does it have to be for-profit?
  • Non-profits typically need to have a cause and also have greater restrictions placed on them.
  • Also, what is the training/credentials needed to be a volunteer? [seems outside the scope of a document about specs and features?]
  • A training program after the AO3’s? +4
  • It’s worth looking up with DW’s Denise has written/tweeted on the subject since she has extensive experience on abuse teams. They need the right kind of screening and the right kind of psychological support as well.  +23 (but I think they need that and pay, I don’t think it’s acceptable to ask volunteers to do things they’re going to need therapy to get through.) (Definitely) +11
  • I know we’re very angry at Tumblr around here at the moment, but Oath’s model for taking care of their abuse employees is quite good. There's a dedicated therapist plus time restrictions on how long an employee can look at certain kinds of content, among other things. +19
  • LJ/DW have a volunteer forum area; almost anyone can answer questions, and those who do so and get marked as good, get training for more complex/sensitive questions. Let the userbase volunteer to help each other sort out things like "how do I delete a post" and "what are filters?" +10
  • Can anyone do it and then flag any content they don’t like (including pairings)
  • Pillowfort has a peer review moderation system that has more than one (possibly two?) mod look at a flagged thing to reduce the likelihood of something like this happening +2 Scientific crowd labor uses this method too eg ‘identify which animals are in this photo’ but there are always multiple people responding
  • If all antis need is critical mass of people who agree with them, let’s not count on this. Lots of people may hate a ship, without it being actually infringing of any rules.
  • People flagging for things they don't like would happen no matter who's looking at the flagged posts. +5
  • I think that there should be a difference between flagging something as a violation of terms of service vs flagging something as content you don’t want to see. For instance, thumbs downing a show or movie on Netflix means you won’t be shown that thing in your searches or recced similar things, not that the content will be removed from the site. +15 [this is the abuse team/abuse volunteers section -- this comment/thread seems.. unrelated/in the wrong place]
  • Wouldn’t a blacklist accomplish the Netflix thumbs down effect?
  • A blacklist would blacklist exact matches. If a “thumbs down/less like this” were implemented, it would need to be algorithm-based. That’s less reliable at the moment, but potentially better at anticipating things you won’t like rather than you having to specifically say everything you don’t like. For example, if you don’t like piss and scat kinks, you probably won’t like vomit either -- though if you do, you’d tell the algorithm so and it’d learn. It could also guess based on, “users similar to you didn’t like these posts over here.” I don’t like this feature myself, I’m just guessing how it could work.
  • Preference to NOT have an algorithm-based system of showing you things… chronological posts ftw +10 [this thread is about blacklisting, not order of appearance in a feed; chronological post order is already dealt with elsewhere -- recommend removing this comment]
  • Yes, but an algorithm removing things from your feed is different than adding things. Maybe have it give a warning and show the tags on the post and have an expand button instead of removing the content entirely
  • Build in a flagging protocol (series of popup windows) where users can block/hide non-ToS violating content with a Tag Filter before going through farther to report something as a violation.
  • By which I think you mean, when you click “report/flag,” it asks you, “Why are you flagging this? Is it against the TOS? Do you just not want to be seeing it?” and if you say you just don’t want to look at it, it suggests blacklisting. Facebook has a feature like this. +16

Groups and Communities

  • Communities +75
  • Communities have personal moderators ala reddit mods +1/-1
  • Ability for communities to have chronological timelines in addition to a timeline of a tag or of an individual.
  • What would the hierarchy be if there is a problem with a community’s mods? (e.g., they join an anti campaign against a user)
  • If what they are doing breaks the TOS and it counts as abuse, the mod/s potentially get booted and a new mod steps up, or the community does what it ALWAYS has done and just leave to join another community. It's fandom. People get reputations: you get hard done-by, you post in your personal journal so people know the facts, it gets around. +2
  • I think what a lot of people forget about LJ-style fandom is that mods were never actual authority figures. Sometimes they had to beg people to come participate in their communities. It was a lot of hard work, and they had to build a place users would want to be in. The moment they fucked that up, someone would make a rival community, and all the users would go flock there and they could mod tumbleweeds and dolls by themselves. Community creation was easy and there was almost always healthy competition. When users in one fandom I was in disliked the mods of one drabble community, two more were made in the same fandom. You aren’t beholden to the mod of any comm. If it’s TOS-violating abuse, report it, if it’s just unpleasant, you can leave and make more.

  • Communities can be set to public or private (members-only) +47
  • Or viewable to the public, but cannot interact with unless you’re a member (protected) +17

  • Moderated posting access and moderated posting options. +18

  • Roles: Owner, moderator, member, visitor (optional) +24
  • To potentially clarify, the owner is the only one who can delete the community, and the only one who can add or remove moderators, as well as the only one who can’t be removed. It could potentially be shared, with owners being unable to remove each other. All would need to leave or remove the community to delete it. If the only owner leaves a community, they must designate another user of that community the owner or delete the community. They can make another user owner and step down without leaving. Moderators/administrators would be able to delete people’s posts, handle the queue of posts waiting for approval (if implemented), screen threads and posts (if implemented) and make other changes to the community such as changing the theme and settings -- they can change things from public to private, but private to public requires the consent of individual people whose posts they change. They too can step down without leaving.
  • Can mods/admins be removable by owners? +1
  • Yes, I was intending this.
  • Members/contributors can make posts, but not delete posts other than their own or make other community-wide changes.
    Visitors/followers, I’m guessing, are users who can follow the community and see even members-locked content, but not make posts. Would they be able to post comments? +7
  • I’m not sure! There are arguments for and against. On the one hand, you might not want to create a barrier to feedback. On the other, you might want to create a barrier to interaction. In LJ days, I’d say leave comments on, you can block trolls if you get any. I honestly don’t know if that’s still viable or not.
  • If a reblogging feature is included, we may also need a "may reblog posts from this community" access level. +9
  • I would have assumed this was the basic member level, so maybe an option to allow regular members reblog, or deny that option to only a select group, customisable per-community.
  • Potentially controversial: anyone who can see a community post can reblog it +1
  • Restricting access in any way and reblogging are not compatible features. If a post is rebloggable at all, it has to be rebloggable by everyone. The relevant setting would be allows reblogs/doesn’t allow reblogs, not who can reblog, with two exceptions: 1) posts can always be reblogged within the account they were posted to (in this case, the community, not the user who posted it in the community--so any user who can post to the community can reblog it to that same community, but the user can’t reblog it to their own blog unless everyone can reblog it--because this kind of self-reblogging doesn’t change permissions) and 2) blocklists of the OP and of the community merge for community posts, so neither someone the OP has banned nor someone the community has banned can reblog it, allowing communities to protect their users from accounts known to troll that comm, while also protecting the user from people who troll them specifically. So basically, the community would have a global setting, “Allow users to make their posts rebloggable.” (Dependent on users being able to make posts public to begin with.) If not checked, no posts in that community are rebloggable. If checked, users have a “Make this post rebloggable” checkbox if they’ve set their privacy to public. Anyone not blocked by that community or that user may reblog.

  • Enable opt-in anonymous comments at the community level (yes - I’m thinking about a return of the kinkmemes). Owner and/or mods can choose to enable anon comments for their comms,obviously default would be off. +12

OTHER / AWAITING CATEGORIZATION

Ability to recommend a post. Aka, the post will appear on your followers’ dashboard but won’t appear as reblog on your blog. This is nice for people who like to only post their own original content on their blogs but still want to be able to share other posts with their followers without clogging up their blog. +3/-3

  • I don’t see how this avoids blog clutter, since it’s still content on the blog and on followers’ feeds/dashes/timelines/friendslists? If viewing someone’s blog already let you easily sort between all posts and original posts only, there would be no reason to avoid reblogging to keep one’s blog “clean.” Additionally, users could do what tumblr users already do: create multiple sideblogs, and use one for original content and one for reblogs, allowing people to decide for themselves which to follow. +8
  • I kind of like the original suggestion when it comes to things like sharing PSAs, boycott information, fundraising, and important news stuff that someone doesn’t want to have on, say, their art portfolio blog but still wants to share to their followers as something they think is important.+3
  • Describe the functional differences between this and a reblog (how the recommend appears/functions, if the permissions are different?) other than it having a different name. Do you mean it would appear on followers’ dashes, but not on blog view? I don’t like that. I don’t like the idea that visiting a blog doesn’t show you everything they’ve been showing to their followers, and it sorta breaks the transparency/openness concept of reblogs by having what’s essentially a followers-only reblog. However, what a lot of users do now is delete stuff that’s ephemeral (news, announcements) while keeping an archive of their creative content. Could we just give reblogs a “self-destruct” button, that they’ll autodelete in N hours/days? That would reduce clutter, without creating confusing permissions/openness levels on reblogs. A blog that has ephemeral/news content is a poor place to host an art portfolio anyhow. Not only are there already sites catering to that, but if you want a more static archive to showcase your art, use a sideblog. Or I’d even be happy to suggest implementing some kind of “gallery” feature for artists to archive their work in! It isn’t fair to users who think they’re following an artblog that only posts art to get 500 petitions a day on their dash that the artist hid they were “recommending” to their followers. Since access and follow might be different anyway, it’s kind of using follow as a sort of backdoor access.

Defined access filters similar to LJ/DW (blacklist might solve this issue.)

  • Custom access filters. +4
  • Separation of who you follow and who you give access to a la DW.
  • [pretty sure this topic is already being discussed in one of the threads]

I don’t know if this has been mentioned but I actually like the ability to reblog a post from just the op or only up to a certain comment on tumblr.

Private messaging/IM? +43

  • Messaging system that keeps a record of the conversation. +164
  • And doesn’t delete it after a certain amount of time +3
  • Counterpoint: warns you that it's going to delete and gives you the OPTION to archive off system (I'm just worried about space) +8
  • I don’t think chat logs will be our biggest space concern, all my rather prolific logs from 2009-present currently fit on my tablet.
  • That you can download +1
  • My kingdom for an API that will let you use it with third-party chat clients. I still use Pidgin (yes, in 2018) and it merges Discord, Skype, Hangouts, Steam, and other IM clients into one handy bundle. It lets me log every conversation locally, merge contacts and their logs, and search my logs on a per-contact basis, so I can search my local log for something someone said, regardless of whether it was on AIM a decade ago or on Discord last week. It lets me be available on chat without having a browser open. Server-side logs are nice too (searchable by text + viewable by date would be great) but as long as I can pipe it into my own client and log it myself I’m good.
  • Ability to answer messages publicly/privately like Tumblr. +30
  • Ability to report? DNW [acronym = “do not want”] abuse via PM. +16
  • That gets into the abuse team details but yes. & ability to block. By IP address? username?
  • I think it should be the account's database ID (to avoid having to reblock the same user under a different name) +3
  • No, blocking isn’t sufficient (though of course we should have that too). When one anti decides to harass me, they put a post on their blog and tell all their friends to do it too. I almost never get just ONE harassing DM. I’ll get a flood of them, and harassing asks too. Similarly, if I’m DMed by a spambot, it’s not like that bot is only spamming me. It should be reported. And types of abusive IMs (sexual harassment, physical threats) are serious enough to be dealt with by the abuse team and a block is a band-aid solution that only lets the problem get worse. If DMs allow images, you also have the potential of people sending shock images (gore, dead bodies, animal mutilation) and CP via DM as an intimidation technique.
  • Maybe to limit this, a user could choose who could send them GIFs/images over DMs. +5
  • A call to dogpile on someone should maybe be a ToS violation? +3
  • Be very clear about what a call to dogpile consists of, though; if it’s “this person is an anti callout post” that’s one thing, but if it’s “This is a nazi/supremacist/cp blog” that’s a whole other thing
  • Most of the time, it wasn’t worded in a way where they actually told anyone to do it (in public) they’d just complain that an icky X shipper touched their precious pure fanart of a different ship with a reblog. Sometimes they wouldn’t even post my username, but people must have been DMing them to ask who they have to fight, or something, because the hordes came nonetheless. Usually they just play the victim and when their friends go, “Oh no, how can I help?” I’m sure they have links--but it’s all in private. Other times, they’d just reblog my reblog with a “how dare you touch my post, you filthy X shipper,” but then all their friends would see that and know what to do.
  • Ability to control who can PM you (Everyone? Only people you follow? etc.) +23
  • Ability to block anonymous messages (if those are allowed). +13 (On tumblr, it’s the only way to limit the amount I’m harassed.)
  • Ability to create channels (like IRC.) +1/ -6 (eh, we've got discord +1)
  • Counterpoint: ability of community moderators to make a group chatroom for their community. I think Imzy was playing with something like this shortly before they went down. Reddit also has this feature +6
  • There are third party chat services. Why reinvent the wheel? +8
  • Also you can always put a link/invite to the gc in the com’s about/description
  • Essentially, I do agree that we should focus on features we don’t already have. On the other hand, when tumblr implemented DMs, people who’d been shy about IMing chat handles I posted suddenly started talking to me? I don’t know why, it’s a friction thing or something. But Discord is bigger than anything we’re spitballing here at the moment, so if anything, the Discord might be bigger than the comm and help the comm get users, lol.
  • Ability to export DMs to a text file, for those of us with rps in dms who are looking Sad about having to copy paste each individual message. +1
  • (can rp via google doc instead?)
  • If it allows third-party clients, you can just enable logging in those.

Direct chatting in realtime?  +3 -1

  • Not the same thing as DM - tumblr has this, DW/LJ did not.  Allows fen to talk directly and privately
  • On tumblr, the direct realtime chat WAS called DMing by users, and the other types of messages were called asks or fanmail. Asks could be replied to in a public post, fanmail was more like an email. What LJ/DW has is PMs (private messages) not DMs (direct messages). PMs are very important to me, DMs could be nice but aren’t a priority to me personally and seem harder to implement. +6
  • If I want DMs, there's discord. I don't have an objection to inviting someone to take it to discord. +3
  • This is going to be a big-ass project.  When we can delegate a big feature to another system, like chat or email, let's do it. +4

  • Speaking of chatting, group chat which involves a handful of users?? +1
  • Could this be another paid perk? (host of chat needs to have paid, inviting others who don’t necessarily have to, maybe? Don’t want people to be excluded just because they can’t afford the perk +1) +2

Tip jar integration (Kpal, ko-fi, etc) for both other users and for site administrators/costs/etc. +18/-2

NOTE: Ko-: tip

  • I can see this being potentially problematic for the same reasons it’s not allowed on the AO3? +8
  • It’s not allowed on AO3 because they are legally a non-profit.  If this platform is for-profit like PF, DW, and Tumblr then there should be no problem. +12
  • That's not the reason -- AO3 doesn't allow them because the legalities of profiting off fan-created content are iffy at best. +2
  • but the purpose of Ao3 is fan content, and this is presumably a site that also allows/promotes original art. It could be up to the user if they believe their content is profitable. +7
  • A toggle for original content? Maybe on post level rather than (or in addition to) blog level?
  • The thing is, not all fan content is non-monetizable. Sure, you wouldn’t want to do it with fanvids, and most fic/art is dodgy, but public domain stuff is legally fair game. And then there’s stuff like how tumblr user sleemo collects tips for news and reporting on Star Wars current events. Sleemo isn’t making transformative works, she’s checking dozens of sources daily, seeking out things that only exist in print and scanning/transcribing them, sometimes translating nonfiction content like interviews and news. It’s fannish, but it isn’t the sort of activity people worry about when it comes to copyright infringement with transformative works.
  • The thing about tip jars is that they technically aren’t for content at all. If I post reviews of lawnmowers, fanfic, and original art of mermaids and someone tips me on Ko-Fi, it’s really impossible to say which they tipped me for. Being fannish should not be penalized by making it impossible to collect tips for other types of content. +5
  • Unfortunately it does, unless we want the big wigs of Hollywood breathing down our asses for taking money away from them.
  • [citation needed]
  • Oh please, conventional fandom wisdom 20+ years.
  • I’m also a Fandom Old, and times have changed, C&Ds aren’t what they used to be, DMCA is enforced with per-page takedowns where the only penalty is that specific content may be removed if infringing, and it doesn’t endanger the site as a whole, or the other users. The burden is 100% on the copyright holders to report each individual infringing post, and they’re busy reporting actual piracy. Fandom is too big and too nebulous to even bother with at this point, given the amount of human labor it would take. It isn’t Zine Era days anymore. Tip jars aren’t illegal, stop pearl clutching. Block users who have them if you like. +6 / -1
  • Have you never read articles about pirate sites? Or maybe about torrent trackers? I know that’s not what we’re trying to make here, but copyright holders often go after sites that they think are taking money from them. And in the case of torrent trackers, they literally are by definition just directing users. But guess what, because those info hashes point to pirated content, they get lumped in. It’s entirely possible for them to go after both someone with a patreon or whatever and the site for those same reasons. The excuse of “oh its the user not me” isn’t always a safe bet if it’s purported that you’re supporting the violation. DMCA doesn’t just absolve the site of the ability to be sued for said content.
  • We’re definitely not making a torrent tracker, this is catastrophizing. YouTube enables more copyright violations than blogs with third-party tip jars, and of tumblr’s many problems, that never came up.
  • AO3 doesn't allow it because their specific approach is, "if it's not for profit, fanworks are legal." They're not saying that selling fanworks is illegal, just that fanworks involving profit are outside of their mission. A for-profit site that allows fanworks is fine as long as those fanworks fall in the range of fair use. (Or other kinds of permitted use.) +6
  • What about tipping site currency like Dreamwidth and Reddit have? Not money you can cash out, but points to apply toward paid features. +8 / -4
  • This seems like a decent compromise? +1
  • Seems aggressively anti-monetization and anti-self promotion, so the worst of Reddit in this regard. Unless I’m misunderstanding--I read this comment as “don’t buy them ko-fi, buy them reddit gold 2 instead”which does bupkis for you if you’re trying to make money at all +4
  • Linking ko-fi / patreon without keeping that on the site. If we’re going to have site currency, it should be for more than just going to paid features. +2
  • I really like this idea. Like Twitch’s bits.  Though the ability to cash out might be nice too. Some popular people might get more than they can realistically spend--and the site could take another cut there too.
  • These people could gift site currency to other users? Otherwise wouldn’t the cashing out money come from the sites own budget?
  • True. 3am brain is not a good brain it works for twitch though. Maybe it’d be more like DA points, then, I think. Still wary about being able to donate to a ‘black hole’.  ...and about the risk of DA’s seriously underpriced commission culture appearing here…
  • My landlord doesn’t accept rent in site reskins and extra avatars.

A tag/flag for fundraising posts +8

  • And the different kinds - medical is different than school, for example
  • If users can tag their content, I’m sure they’re able to tag their fundraising posts. I don’t see how this should be a separate/specific ‘feature’. +2
  • How many of you want a tag/flag just so you can blacklist ebegging?

Custom views a la LJ/DW reading list. +3

  • I would rather just have individual sideblogs/pseuds be able to have separate follow lists, and robust tag blacklisting. +5
  • Separate dashboards/timelines, too. +1
  • Instagram is actually a good model for this, except ideally you’d be able to do this from one account rather than having all separate ones +1

Pinboard compatibility. +18

  • Would love to auto-feed my public bookmarks as I make new ones into my feed on this site (good for recs blogs, etc.) +2

Ability to link to your profiles on other sites (AO3, pinboard, etc.) +101

Ability to search people’s profile info (e.g. for people with “blanket permission to podfic” listed) +12

  • Other useful things people could opt into being searchable based on: “blanket permission to translate,” “blanket permission to use my art for avatars/graphics,” “offers beta services” “takes fic requests” “takes art requests” “takes vid requests,” “takes gif requests,” “gifs/graphics okay to use in RP,” etc. +5
  • Is this something that could be covered by a list of groups a user is in (maybe with the option for users to make their membership public or private when they join a group?) That way users have a bit more flexibility - someone might want to be part of the ‘stucky beta readers’ group instead of being listed as a non-specific beta reader (for example). Also might give a sense of community / make searching a bit more standard. +2
  • At least, some kind of filtering or search of the results would be more useful than a list of every single user. Communities might do it, but I guess an advantage of having it in the architecture is that it isn’t reliant on having a competent mod for every fandom, and you don’t get people repeating each other’s work. FFN of all places actually has a feature like this. When you add to your account that you’re willing to beta, it lets you select fandoms that apply (of which it has loads of categories, because it’s FFN) and then you can write a sort of beta profile that authors can browse as they look for betas, it’s really neat. So you wouldn’t necessarily be able to narrow it down to Stucky betas (per-ship would get ridiculous with minor ships/small fandoms) but you could narrow it down to MCU betas, see if you recognised any, and browse or search profiles for “Stucky.” If we could somehow hook into a list of AO3’s canonical fandoms for that feature, that’d be really cool.
  • It’d be cool to have a sort of “footnote” option, then, for any posts made by other users that have used that blanket permission you’ve given to link back to the post that you gave it in? (sorta how AO3 has a “works inspired by/based on” section?) +7
  • Also add ability to make these public / private - public “these people are allowed to use X resources of mine” and private “these are my betas for X fic”
  • I’m not sure I understand? This is about discovering new people to collaborate with, e.g. “I need a beta, but I don’t know anyone in this fandom,” or “I want to record a podfic, but I don’t know which authors are okay with that.” Listing a handful of people isn’t useful, because if you already know who you want to work with, you don’t need discovery. If I’m looking for gifs I can use to RP with, finding a list that says, “only these five friends can use my gifs” in the “gifs free to use” list isn’t helpful. That should go under some kind of shared attribution for already-created works/WIPs.

Group blogs?

I think we used to call these Communities? Or did you mean something else? +3

  • Clarification: are we talking a LJ/DW style community or just a blog that a select number of people have access to?
  • Blogs as in multi-mod advice blogs and whatnot, I think? +2 /-1
  • Communities probably make these irrelevant, don’t they? +1

A way to mark posts read so you can keep track of where you are on your reading list. +20

  • In a similar sort of vein, a “read later” section where you could star certain posts that you could come back to later? This would be separate from “Drafts” or “Likes”. +12

Multiple icons (with option for more iconspaces as a paid feature à la DW?) +18/-2

  • Multiple icons would only make sense if you could select a different icon when posting something, which is not a given.
  • Yeah that option should definitely be available, both for posts and comments. +6
  • (this is copied from higher up, but relevant here) Switching avatars seems clunky outside the LJ model where urls never changed and so keeping track of people was easy. The only benefit to multiple avatars imo is for RP blogs, in which case maybe having different blog types might be a consideration? Ie personal blogs, shared blogs, community blogs w/ contributions, RP blogs, art portfolio blogs etc. +1/-1
  • This is a duplicate from above
  • Accessibility note: ability to turn off seeing alternate icons, for people with memory problems who identify people based on their icon +2
  • You could have your main icon and then a smaller icon beside it for reacting to the individual post, that might work <On LJ we called this moodthemes, but reaction images seem to have replaced them…
  • Nah, moodthemes were more like site emojis, not full-size user avatars. Although I remember the community-wide custom moodthemes, a more modern and higher-res version of that could be pretty awesome? Paid feature for communities? Ability for artists to make moodthemes and make them public for anyone to adopt?

Ability to turn off seeing crossposts -- if someone’s automatically posting their delicious recs or their youtube likes, I should be able to disable seeing that. +21

                If it’s not native, that’s the first thing to get made for greasemonkey.

  • (Can this be handled by blacklist? If there’s an autotag or something applied to denote source maybe.)  +7

Ability to add a note to someone else's blog that only you can view (LJ had this functionality, so that accounts you'd added notes to had a star next to their username, and when you hover over them, alt-text would show you the note you'd added). This way, if people change their names, or you want to add a note about how X participated in Y wank, you have an easy place to reference it. +117 -1

  • Ability to add a note that only you can see to the profile of someone you’re following  keep track of where you know people from, who they are if they change names, etc.), LJ-style. +5
  • (This is mentioned in Account Management, above)
  • I see “Ability to nickname other users in ways only we can see?” but that’s not exactly the same thing, although they could be combined. +1
  • I added the suggestion above this about adding notes that only you can view, and I think these two things could be combined into one.
  • Do the thing dA does where if someone changes their username, it automatically adds a comment of ‘Previously known as X’ so you know who just changed their url on you.+2
  • -1, when LJ implemented this feature, people mostly said “Oh, won’t that be great for keeping track of people’s pronouns?” but how people actually used it was snidely commenting, “I’m adding an LJ note about what a horrible person you are” in the stupidest wanks. If you hate so many people you can’t remember them all or why, you should maybe consider letting some of the less important ones go, or just blocking like a normal person. It was bad enough in 2010 or whenever, I don’t want to see antis learn this trick, it was insufferable. The culture of needing to keep “receipts” on everyone instead of just blocking or muting people who aren’t doing anything report-worthy is part of the same culture of harassment and wank. Antis will love it. -3 >Were those notes publicly viewable? I didn’t remember that. If they’re private, I don’t see how it contributes to wank or becomes insufferable. And as someone who likes to use memory aids it would be really handy to be able to add things like ‘from X comm’ or ‘that one x/y fanartist’ or ‘the one who talks about trains’ -1
  • No, they were not public, which was why you had to brag about it, and yes, this was absolutely a staple of wank in the wankiest LJ comms, we’re talking sf_drama, ontd_p, etc. Keeping ~*~receipts~*~ on everyone is a wanker thing. And no one did it quietly, because then you wouldn’t know how virtuous and responsible they are protecting everyone like that! Does this sound like antis yet? Because they haven’t changed their tune all that much.
  • OK, I think I replied to this before but how is this different from any other "I think you're MEAN and I'm gonna tell everyone about it?" comment? People are gonna people. If the notes are private, who cares? At least this way if there's a violent misogynist who isn't in my circle or my communities who I want to keep an eye on but not interact with but only see every year or two, I can make a note and remember them. +5
  • You should block the violent misogynist. If you can tell they’re a violent misogynist, so can everyone else, so warnings aren’t necessary either. (And if something is serious enough that others need to be warned, is the biggest barrier really going to be that you forgot?) If what they did isn’t important enough for you to remember on its own merit, and the folder of wanky screencaps on your computer is too large to possibly go through all of them, they’re not doing anything report-worthy or serious enough for you to block or mute them, maybe letting it go is an option? You don’t need a special feature for receipts so you can sally around the internet “warning” everyone they interact with about this one time they put their foot in their mouth seven years ago, which you saved as a note so you’ll always remember how much you hate what’s-their-name. No one’s stopping you from keeping a file on your computer or a Nonny’s Hate List google doc of all the people who’ve pissed you off in ways too numerous and minor to remember, we just don’t need receipts as part of the platform, it encourages grudges, a permanent note of some minor offense you will now never forget, which you obviously would have otherwise or you wouldn’t need the note. Blocking people who annoy you is vastly superior to this. I’ve blocked literally thousands of antis. I wouldn’t trade my blocklist for a stack of wanky notes on every unblocked anti so I can squint resentfully at them across the internet for all time.
  • -1 me all you like, I don’t really want sf_drama to spring fully formed from the skull of our new platform. I know some of you loved wank--I don’t, and I don’t think LJ-style wank was that different from the Tumblr Discourse crap we have now. But alternately, if you want notes so that people can keep track of username changes or their own pronouns, maybe have a space where people can add notes about THEMSELVES that everyone else can see, where they can note username changes or pronouns as desired.
  • That’s called a bio. Different thing.
  • Bios/profiles are different, this is more like metadata if you see an avatar on your dash/feed and want to quickly check something important about them. The tumblr generation will get the concept of “put your pronouns here, tell us more about yourself there” no problem. Put a short character limit on it, plaintext only. +5
  • Discord has this and it appears to be fine, I mostly see people saying they’ll put pronouns there<- Agreed, wank isn't inherent to this feature +2
  • People on Discord tend to put their pronouns after their usernames if it’s that important to them. There are much better ways of keeping track of pronouns, like for example, anything the person whose pronouns are being recorded has access to/power over?
  • What about when someone changes their username? If you’re supposed to be able to break all links to escape stalking, it’s working against itself if a note from your stalker identifying you follows you invisibly on your profile.

Non compressed images and gifs +8, /-3 (because it would take up so much server space and it’s not a must) THis is a duplicate from above

  • Option to vote for that “within reason.” It should be enough to look good on a screen, but even tumblr had limits. Gifs in particular could get massive with no restraints whatsoever. +5
  • Like, images below ___MB are uncompressed, but anything above is compressed to fit? Perhaps with a paid option to increase the individual image cap? It would have to be a pretty generous size cap to allow for fandom style giffing though. +3
  • Any compressed images you have an option to add a clickthrough to the full pic hosted elsewhere+1
  • One click access to full view… +8
  • With the caveat that the “preview” version should not be a crop or a thumbnail, but just a smaller version of the full image. Tumblr did this right. Imzy did this very wrong. +6
  • Allow thumbnails / crops by OP per post, for those NSFW images. +1 as long as we’re not forcing them on anyone.
  • A sfw/nsfw toggle on images instead of just tagging (think twitter sensitive content opt in toggle.) +1 (also tumblr did in fact have this until very recently -- well, it was per-post, but better than just tags or per-blog.) duplicated above
  • sfw/nsfw/explicit maybe, so there’s a place for milder stuff so it doesn’t get put in sfw just cause someone is hedging
  • This would be a huge server space-eater though.
  • Yes, but it’s more or less the entire point. You can already post all sorts of plaintext on tumblr, AO3, and Dreamwidth. It’s artists and gif makers who just lost a platform. A new platform that doesn’t even bother to support us because it’s expensive and hard defeats the purpose of coming together and trying to do the hard, expensive thing that needs doing. +10
  • Would love to be able to set the site to “I’m at work!” on login, and have it automatically hide/cover/collapse anything marked NSFW -- or have a toggle that tells it to do this. +21
  • But this is one of the major reasons why Twitter isn’t an acceptable replacement, because of degraded image quality. If a platform is meant to be used by artists, full-res images are a must. +8
  • I really don’t want compression either, but I could potentially tolerate a compressed preview with click for a better version as a bandwidth-saving measure. +4
  • Paid feature? Don’t want to make people pay solely for being artists, but is there a possible compromise? A certain (reasonable) resolution free, uncompressed paid? +3
  • If you stop paying, do your existing images get compressed? +1
  • As an artist, a caution if data caps are applied as a money-saving measure: artists will likely get around them by deleting old content in order to post more. This won’t save much bandwidth, because the new content will be the stuff getting the vast majority of views anyway, but it will destroy any sense of archival or fandom history. Artists are already flighty and delete old work at the slightest provocation, don’t give them more incentives. +5 +1
  • If you stop paying, then you should keep what you’ve got, but not be able to add any more - if you’re say, 13/15 and stop paying and go down to 13/10 posts, then you can’t add more content until you’re 9/10 or start paying again, or something like that +1

No cropped images and gifs. +13/-1

  • As an artist, there is nothing worse to have your art’s first impression be made while cropped. Imzy was like that and it was the worst part of the site when it was alive. +1
  • But some illustrations are too big for feasible UI e.g long comics.
  • Tumblr was able to host long illustrations with minimal detractions (the “Colors of the Sky” post being one exception) +1
  • Tumblr had some limits that were workable. Anything that tumblr has as a size/resolution limit is perfectly fine by me--that was never one of the problems with tumblr, it was one of the things artists loved about it. Imzy had much more severe limits that hampered the presentation of art. Artists complained about this at the time, and Imzy devs brushed them off--not because of any technical, financial, or UI restrictions, but because they believed that forcing users to click to see the whole image would drive more clicks/user interaction, and make feedback more likely. They wouldn’t budge on this, like the “you must be logged in to view, like Pinterest,” and like not being willing to disable infinite scroll, this was one of the things Imzy staff did out of belief it would make the site more popular, that fandom hated because it impairs usability. Trust me: artists love getting more interaction/comments, and we know this is bad for our art. It would be like if AO3 replaced summaries with a chunk of text from your fic selected by a robot, because their marketing team believed it would drive user interaction. Just don’t do it, it’s bad design!
  • Or choose where it is cropped, like mastodon.art +5/-1
  • Choosing a cropped area has some value, for example, for making SFW previews of NSFW (and if we do this, I’d like the ability to flag preview and full image as NSFW or not separately) but please, make this opt-in. I don’t want to be forced to crop my art. Resizing down to a reasonable dash/feed scroll size is modern web standard. Cropping makes me feel like I’m on someone’s hand-coded archive in 2001. (Bless them, they tried, but those were truly the Dark Ages for artists.) +11
  • It’d be nice to be able to have the SFW preview show up in the search (with the caveat that there will be a flag on the post itself saying that the full image is NSFW) so that people could see the art style of an artist or decide for themselves if it’s something they want to see.+4
  • Ability to upload a separate preview image, like Deviantart does, would also give users some censorship ability. Native cropping would guarantee it’s exactly the same image, but separate uploads (the preview being optional, it just uses your main image if you don’t add one) would allow for blurs, black boxes, etc. Deviantart is a huge site that uses this all the time without it ever being abused that I can recall.
  • Another good use for this is if you upload a comic page--if you use a much smaller crop as a preview (Deviantart users will put text on it indicating if it’s something that doesn’t show well if you’re not on the page for the work itself, like comics and animations), users can click to view, this is more practical than spoiler warnings or something because you at least get a teaser of the art style! It could be used for good, but I’d rather not have it at all than have it forced on every image.

A personal bookmarking/liking function that does not add any notes to the notes count. +22 (Xkit has this feature.)

  • (Like an option to do private bookmarks like on ao3?) +2
  • Yes, like “saved” posts in FB, but searchable and therefore more useful. +7
  • Something distinct from "likes" so that you can like a post or bookmark it to read later. +4
  • I think if it ends up being Maciej who works on this on any level it is a little rude to ask him to build a Pinboard competitor. :D
  • I agree with that, but my comment (likes vs. read later) is related to how I (and many other people) use twitter likes as bookmarks. I tend to like people's posts less often since having likes and read laters together clutters it all up. (Also, selfishly, I would never add these things to my pinboard and get my excellent organizational system all messed up and would legit just add them to my browser bookmarks or to drafts like I do on tumblr if that was the option.)
  • This would be private-only, while Pinboard is public or private. It would bookmark stuff on this site only, while Pinboard can bookmark any site on the web.
  • Make the bookmarking attach to a Pinboard account. +5

Ability to search entire blog, either by tag or keyword. +106 (IMO this is critical.)

  • Being able to search your own likes. +89
  • Browsing a tag on your blog locally (as well as posts with privacy settings you can see on someone else’s blog) is crucial, I agree. Searching your entire blog by keyword I could pass on.

Chapter posts. Some kind of simple and easy way to link posts together as a unit/book/series for art and written stuff. *Not* tags. No hunting down the last chapter of something in between a billion other posts of kittens +49

  • Perhaps similar to making something part of a series on AO3? +8
  • Or ability to add to “playlist”-type independently-linkable construction, similar to YouTube? (This is probably the same thing.) +7
  • Doubles as the pseudo-webcomic hosting Tumblr does now +4
  • Could generalize to: Collections of posts arranged in user-specified order. This has so many uses. Series, showcases, starter packs on a topic, scrapbooks, organized "my fic" masterlists, greatest-hit lists that aren't overshadowed by shitposts that went viral. If you could collect other people's posts as well as your own, it'd be an awesome tool for rec lists and transmitting the important parts of a fandom's history to newcomers. The relentless recency bias of chronological feeds and tags was already a problem for fandom on LJ, but on Tumblr it's turned into a gaping memory hole. It'd be really nice to have more native tools to collect what we think is important in the order we think it makes sense in.+7
  • Allow a redirect to a ‘chapter list’ / ‘table of contents’ like youtube’s side playlist shows you everything, so when you stumble upon part 5, you can go back to part 1 or don’t have to click all around to see part 3 again. +3

Still functional when javascript is disabled. +11

  • Query: is this a bandwidth thing? A security thing? I don’t have a lot of familiarity with this issue.
  • This is an appearance and some functionality thing, with the cost of bandwidth (takes longer to load) and potential security issues (site can have j'script doing things you don't see.)
  • Not the requester nor is it important to me, but I have NoScript installed in my browser, so when I visit sites that are completely, 100% unreadably broken with javascript disabled, it’s a frustrating user experience. If I had an account here I’d probably enable javascript anyhow, but it’s a turnoff for new users who are also browser security nerds, I guess. I will say that if you do break the site without javascript, just let it load all broken, don’t give me a blank page and a loading animation that never finishes. Those are the actual worst. (For what it’s worth, AO3 looks great without javascript. A fallback no-JS site looking decent probably bodes well in general for cross-browser compatibility.) +6

Would like a way to view blogs/communities without custom CSS/HTML +1

Moddable communities. Communities with mods with option for private or public membership. Or some kind of communal space where multiple conversations can happen at once and not be on top of each other (not chat or unthreaded, basically). +19 (but I want to +10000 it) (^3) 

  • Ideally it’d allow posting content, commenting on content, and also just chatting. +4
  • And it’d be possible to organize content within an inch of its life. +3
  • 18+ communities. +8
  • What I love about tumblr tags: the anything-goes discoverability. What I love about LJ/DW-style comms: the moderation. Tumblr basically expected users to dedicate themselves to being “curators” and spend their whole lives combing tags for relevant content without anyone helping them--while there were some co-modding features, they were barely more than vestigial. I want the option of the nice friendly moderated living room, or also the firehose. But let’s not try to treat a firehose like a living room. +3
  • Duplicated above

Per-post Creative Commons licensing options. +15

  • Plus option that doesn't exist in CC: Can make derivatives/fanworks, but not copy the original. +9

Business/Corporate transparency: Site admins let people know when there's problems and what they're planning to do to fix them. +16

  • "Known bugs" listed somewhere public. +25
  • No “secret testing features” like forcing algorithms on users and not telling them why their posts aren’t showing up in public tags like they should. +58
  • When suggestions are made and they become known, please let the admins tell us what’s being considered and what’s not, and if not, tell us why so we can improve. Nobody likes screaming into the void about suggestions and never knowing if they’re being heard or not. Obviously not every suggestion, but if it’s big and loud, then I want to know what the admins think. +1

Obviously not a top priority or anything, but fannish next of kin (à la AO3). Someone that in case of your death will be given rights to act in your name. +42

  • Not just in case of death -- long vacation, hospitalization, gone to prison -- for any situation where the account holder doesn't have access, they can designate an agent who's allowed to make changes as needed. +11
  • As an option going beyond actual death (which should be separate, and verify somehow that you are, in fact, dead) control of your account being transferred to the user you specified if you haven’t logged in in X time frame (minimum something long like 6 months to a year). Obviously very opt-in and optional. Would send lots of warnings and notifications before the switch happened. +11
  • I have this set up with a Gmail feature.  If I don't log in for N period, login information is sent to my designated person.  Every few months Google reminds me that I've set this up, and lets me edit the person or drop the feature.  
  • Ability to name a next-of-kin who doesn’t have an account on the site, by naming things like email, phone number, real name and address, or alternative social media account. Might be harder to find the person in the event that it’s needed, but would avoid, “Hey, make an account on this site you don’t care about in case I die.” +15

Another potential income stream for the site (and its users). This might be well beyond the scope of this document, but the sex worker community and the self publishing community both have enormous overlap with fandom. I'm not familiar with the former, but I'd personally kill for a self-publishing site that allowed original fic authors to tag their books like AO3 writers can tag their fanfic. Additionally, I know that a lot of erotica writers in particular would like a way to self-publish certain types of content that are forbidden by Amazon. Would it be possible to have a feature like AO3 that automatically turns designated posts (and even better, designated series or collections of posts) into downloads in several formats, ie .epub, .pdf, and .mobi for text, .zip for images. AO3 does this free, but I was thinking of something like an itch.io payment model, that allows people to offer downloads free, pay-as-you-will, or for a set price. There would have to be some sort of rules about this to stop people from trying to sell their fanfiction, or even worse, something like actual child porn or bomb-making instructions, so again, it's probably beyond the scope of this document, but as long as we're making a wishlist, I thought I'd throw it out there. +10/-1

                Perhaps you would pay a one-time moderation fee for each document for sale you upload, to cover the cost of a human taking a look at it and determining that yep, this is original fiction, carry on. That could make the site a small amount of overhead per-story, and if they sell the site will get a cut of that too! I think original fiction is a great thing to include because it does have that fannish overlap, but can fundraise without the legal hassle.

This isn’t a top priority but to be completely fair, what keeps me on tumblr vs dw/lj/et al is the dashboard layout. Personal blogs aside, the dashboard is simple, clean, and doesn’t overload me sensory wise. +14

  • Also it being the first page; don’t make me navigate to my dash/feed +2

Multiple font sizes without having to zoom in the entire blog/dashboard, particularly when on mobile. Most apps, including tumblr do not allow you to zoom into the text without being unable to see most of the media included. On browser, there is the ability to zoom in the entire site, which while presenting some problems with media, still allows media to be viewable. +9

        Ability to zoom or shrink pictures, text, and emoji separately +2

        

In all absolute sincerity, it seems like Pillowfort already does a lot of the things that we want a site to do.  The people behind it want to build a site for fandom.  Yes, they have no long term monetary plan and aren’t that stable right now, and those are definitely issues, but the structure is already there.  Having used it a little, I was impressed.  If you search “Pillowfort” in this document, there are a lot of cases of people saying, “I think Pillowfort got this bit right.”  Is there any reason we can’t collectively reach out to Pillowfort and say, “We want to help”?  Let us take the ideas laid out in this document, particularly about financial sustainability, and offer to work with them instead of re-inventing the wheel.  Pillowfort is new, it’s spare, it’s not perfect - but it gives us an existing foundation, both in terms of code and in terms of visibility and userbase.  Why not work with them instead of setting up a competitor and further fragmenting fandom? +1 / -1

  • I feel like this document & the initial reach out to Maciej probably arose in part out of a dissatisfaction with what Pillowfort is offering overall, despite liking pieces of it, as well as a sense that they may not be able to make the necessary adjustments in the kind of timeframe we’d like to see. +1 <I doubt that this document is going to lead to a particularly fast implementation of any large # of these asks given the scope involved in some of them. But this doesn’t necessarily mean they’re all compatible with Pillowfort either.

It seems worth it to me that as we discuss desired features, we also need to discuss desired theory unpinning the site, since that will determine what a lot of the features can even be. Maybe this belongs in the Revenue/Sustainability or Organizational Resilience section, but I think unless we start with a very clear and agreed-on understanding of the potential pitfalls and problems with things like corporate, legal and other governmental pressure, we’re just going to have the exact same thing happen again. Establishing an initial framework that underpins every action/feature might help us actually avoid the cycle we’re in. This seems so pivotal to me. AO3 did this - and they dealt with it by being a nonprofit and having a robust legal team. If we can’t do that with the Social Network Of Our Own, [very unrelated, but SNOO is a cute name][thanks! I stole it from someone else somewhere in the doc, but I love it too] how will we protect it from the same problems we’ve always had? I’m thinking of a post by @greywash on tumblr that says “I’m saying: there is a systemic problem within fandom, regarding the fact that we routinely get hit on three fronts: legal rights to the material we transform, sexual content, and governmental disapproval. Protecting fandom means fighting for fandom on all three fronts and putting thought and effort into how to make an archive robust against all three prongs of the attack.” How do we build in those protections in a for-profit endeavor? Can we have a section where we attempt to hash out this theory? +6

  • I was just thinking that this theoretical site seems like more of an archive with social media usability than one or the other
  • I can see that! I don’t know that it has to be that way or not, but it would let us maintain a record of our fannish social media history, almost for the first time.  It just seems so clear the traditional social media for-profit structure isn’t right for fandom, and I’m not sure if that document has dealt with that. Since it’s mostly about desired features, maybe it doesn’t need to, but it seems like... Can a not-for-profit structure work? If it can’t, are there safeguards we can build in from the very beginning of creation - safeguards to the structures, features, and probably most importantly, governance - to ensure that they same-old-thing doesn’t happen again?

When making a new post, have basic things like tags and whatever text/photos/vids you’re posting be easily accessible (i’m not wording this right, but think like how Tumblr sets up it’s new post page), but have things like warnings/privacy options/automatic orphaning/more specific post settings be in a collapsable menu to facilitate easier posting à la Tumblr/Twitter while having more AO3-like options without having them clutter up the entire page.

  • Ability to organize likes (a la memories on lj)?
  • The discussion on bookmarks seems to cover that ground.

Something I haven’t seen brought up yet, but came to mind while reading the Accessibility section:  closed captioning or transcripts on any videos? +1

COMPLEX ISSUES / BRAINSTORMING

Flagging & Tagging of Adult, Sexual, and NSFW Content

  • “I want to see porn” vs “I want to see NSFW content [not porn]” vs “No porn or NSFW content”; “porn blogs can follow me” “NSFW [not porn] blogs can follow me” “All blogs can follow me” +31
  • Suggestion: two types of account with post by post toggles on each.  One type is “primarily adult content” with individual post toggles for “this post is work-safe/child-safe”; the other is “primarily safe-for-work/child-safe content” with individual post toggles for “this post is not safe for work/children”.  +3
  • Unless we are talking about people marking their own blog as nsfw/porn, and we are not expecting this to be checked and regulated by a moderation team, this will run into the problem of defining “what porn is”. What will be the rule? Real people as opposed to drawings/text? The fact you can see sexual organs? It being a sex act or not regardless of other factors? What about non-sexual nudity? And if it is sensual nudity but not a sex act? And in the other side of the problem what about explicit sexual drawings and text? Things like furry art and anime and manga? +1
  • It's probably not "female presenting nipples". +4
  • I added this, and my thought was that users could flag blogs (or posts) as porn or as nsfw (both options should be there) -- I wasn't thinking about a mod-enforced thing, just an easy way to make it clear that this post has female-presenting nipples; this post has two people having sex. I see a lot of both completely untagged on tumblr, and having an option per blog or per post seems like it would encourage people to mark them appropriately.+1
  • I'd really like it if you could mark your posts as explicit/none even if your own blog is marked otherwise. +15
  • I think that to work, this is a setting that would have to be mandatory at both blog and post level.
  • There could be a ‘this blog chooses not to warn and therefore may be explicit’, which would mean not tagging individual posts if someone doesn't want to. But of course this is functionally the same as blanket NSFW. +3
  • This is the same as the default “Not Rated” function on AO3, which does enforce a system like this.
  • If we assume 4 sensitivity types (Porn, NSFW, Safe for All, Unspecified/Choose not to warn), could we have the user setting be which one of these sensitivity groups their posts are marked as by default? Like, if I choose in my user profile to have a blog be “NSFW by default” then the little toggle or dropdown menu where I would specify sensitivity level of a post would go to “NSFW” by default, but I could still change it to porn or safe or no warnings. (I imagine “choose not to warn” would be the global default until a user sets their preferences).+6
  • E621 and other booru-style websites require all posts to be tagged safe, questionable or explicit. It doesn’t seem unreasonable to me to have to hit another button or two during posting. (maybe there could even be text shortcuts like RATING:E in the first line? -- with a default setting of course.) +1

  • While it’s true the edge cases are unendingly murky, AO3 does, in fact, enforce this. It requires that users add a rating, and if the rating is too low and it gets reported, it gets set to [Not rated] by mods.
  • So...ratings? Explicit, non-explicit?
  • Need a functional definition of NSFW, which will include "explicit" but also other content that's not sexual but still not workplace appropriate; this also likely works as a "no small children" filter. +6 +1
  • YES, but possibly like, explicit (sexual), explicit (nudity, no sex acts involved), non-explicit. +7
  • I’m having flashbacks to publishing rules of having to be able to count all 10 fingers on nude pictures, otherwise it would go from nude to sexual
  • We'll need a system that also differentiates between explicit media (audio/visual/text) types
  • I feel like this is such a subjective thing that making too many options could open up a door we don't want to go through. +4
  • The point for me is mostly that I'm happy to see nudity, and I'm not happy to see sex acts, and I feel like there should be a way for me to do that. +5
  • IMO that is tagging, not site functionality -- users will always be better at making that distinction than a system is, and users who do not use best practice tagging can be unfollowed. Creating “official categories” will be dicey at best and a constant source of debate at worst. +21
  • It's literally just an extension of what tumblr already has? There's the option to mark your blog as adult content/nsfw -- the suggestion here is to just mark it as adult content and/or nsfw.
  • NA That's not how I read the original request, sorry! I think allowing someone to list their blog as NSFW or Adult Content or both is fine, but what those terms mean will still have to be user-defined. +2
  • I totally agree with that and didn't mean to suggest otherwise -- it was intended as a form  of self-identification.
  • If we mean "sexual content," that's what the tag should say. "Adult content" can also include violence and/or gore. We might split: Nudity, Sexual Content, Violence/Gore, Other Extreme Content. +5
  • In relation to the above categories, I see merit in further dividing them by 3 sub-categories: written/fanfic, art/fanart and “real people”.
  • This whole thing is getting more and more complicated.. Do you want to have to fill out an entire AO3-style tag settings page before every post you make on a social media platform? In the end, this platform should actually be useable. +1
  • I dunno, requiring users to put a moment’s thought into their posts could be refreshing.
  • I’d just say put a basic filter thing like Deviantart has:

Contains Mature Content?   Yes  / No (Required)
if yes:
Moderate /  Strict
Nudity
Sexual themes
Violence/gore
Strong language
Ideologically sensitive +1

  • How about AO3-style, with its ratings (porn could go under Explicit, nudity under Mature?) and Not Rated always there as the CNTW of ratings. A violence warning is a separate ticky box, just “Contains graphic violence,” bam, done. Let’s not fret over edge cases. AO3’s moderation style is more or less, “if it’s close enough, we’ll give you the benefit of the doubt.” How sexy a sexy image is is endlessly subjective, sure--defer to the OP’s rating. Graphic porn tagged as all-ages is obvious abuse. Only intervene in cases of obvious abuse, let people self-declare otherwise.
  • All blogs can follow other blogs, but NSFW content seeing is a seperate function. +2

  • Disincentives and limitations for people flagging as violations the wrong things with malice (like flagging for pedophilia when it’s about a ship they don’t like and it’s simple age-gap between adults etc.) +10
  • Checks against flagging velocity and flagging accuracy are helpful in ensuring higher flag quality. +11
  • -1, I hate when I visit a site, see bucketloads of actually infringing content out in the open, try to report everything infringing and the site punishes me by assuming I’m a troll. Craigslist and its “community norms” was terrible this way.
  • Maybe you only get a limit on your reports-per-day if you’ve been found to report falsely and maliciously in the past? +5
  • So like, if you start doing a lot of false reporting, you get informed that you’re wasting moderator’s time with false reports, and that you’re getting put on a limited number of reports per day? +4
  • No flag for "pedophilia," which is a mental disorder not identifiable from pictures or text content. Flag for "sexual content involving child", possibly separate from "sexual content involving a teen." ⇐ -3 for disaggregating child CP vs teen CP. It’s all illegal
  • Child porn involving real children should absolutely be a separate category from other content, and when reporting it you should have to verify the post contains images/video of a real person you believe is a minor, with punishments for reporting drawings, fic, etc with this high-priority flag. (<you said better what I tried to say before about disincentives.) +13
  • Simulated CP (e.g. art, fic) should be a separate category
  • “Simulated CP (e.g. art, fic)” should not even be against the rules. It should have a mandatory content flag or CNTW flag, like AO3. +14
  • -2, it absolutely should have a mandatory content flag. CNTW should not apply to CP. I think that's more likely to protect both our asses as the viewer/browser and the site.
  • CNTW applies because it’s not CP :) it’s fanworks you’re wrongfully describing as CP. CP is obviously banned, it’s illegal and reprehensible. AO3 allows CNTW for all fanworks, it’s working fine for them.
  • Whether or not the site allows users such as “MAPs” who self-identify as pedophiles/attracted sexually to children, they should be clear and consistent in their choice of rule. If it’s banned it should be a separate offense from actual child porn. (agree it would make it easier and faster to find and to denounce CP to the authorities if done well.) +4
  • I don’t see any reason why the “eroticization of childhood” should be banned unless there are actual children involved. Otherwise, you’re going to get reports against fictional content, or against fetish content where two grown adults dress up in onesies. Plus, anyone who does any kind of gender or sex education is inevitably going to talk about adolescent sexuality. Basically, I would say, no victim, no offense for something like this. +10
  • There is a difference between content that might squick someone, and someone literally saying, “I am sexually attracted to children, I am a pedophile.” Not something you interpreted as that, but them literally saying that. Banning MAPs, if done, would not be a witch hunt, you just couldn’t blog about being an actual for-real pedophile or build a community around pedophilia as a sexual “orientation.”
  • Sexual harassment or grooming of a minor by an adult should be a separate offense from both of the above (if minors are allowed on the platform) and it should be banned. An example of such an offense would be an adult asking a self-identified minor for nude selfies.  +10
  • Follow up as well with police report of some kind. Maybe IP blacklist or some kind of measure to prevent re-joining under a different account +2
  • If minors are groomed into creating CP (of themselves) for adult predators, the adult should be the one punished for it, and the minor perhaps directed to some type of help or education rather than having their account penalized or reported. +9
  • Said CP would still need to be deleted at the very least, but possibly reported in case other adults downloaded and are keeping the images
  • I'm reasonably certain that this falls into a mandatory reporting category. Hands may be tied as to how to handle this.
  • If minors are allowed on the site, other users sexually harassing/grooming/soliciting CP from them should be separately handled from other types of “harm to minors,” such as peer bullying. +7
  • Things like whether a user may “secretly” be a pedophile, as wildly guessed from their tastes in fiction or other non-evidence are not enforceable in any kind of reasonable way. CP and sexual harassment of minors absolutely need to be banned, self-identifying as a pedophile and discussing attraction to children openly could optionally be banned, but reporting users as “pedophiles” needs to be clear this is not a matter of opinion. Anyone could be a “pedophile” in the opinion of some rando who just doesn’t like them. +8
  • Reblogging a minor’s art and “tagging as ship” (tagging as though the characters depicted in the art were in a relationship for blog organizational purposes, regardless of artist intent) does NOT constitute “sexual harassment of a minor.” +11
  • Providing a specific definition of what “sexual harassment” means would limit the amount of “this user harassed a child by saying they want Rose and Dave to smooch” situations. Basically, if there’s going to be a special “sexual harassment against a minor” that’s separate from the general “sexual harassment” report (which seems redundant, but could be useful for prioritization?), there needs to be a very clear definition of “sexual harassment.”+5

Pros and cons of “reblogging” and nested conversations

Should reblogging be a feature? How to follow conversations and discussions on a post in a coherent way?

  • Reblogging keeps being mentioned as though it is assumed, but I'm not so sure. Is it possible to keep a coherent thread of conversation/discussion and still maintain reblog functionality? A problem that a lot of us have with tumblr is that it makes community building and meaningful discussion difficult. And the reblog feature seems to be at the heart of this. On this front, there are two problems with the tumblr model:
  • If you want to have a discussion about a post, you have to reblog it to all your followers … over and over. It should be possible to comment on a post without broadcasting it to all your followers every time you do. +16
  • Could threaded comments be a thing, and all comments travel with the post? (i.e. you can reblog a post, but comments are attached to it as their own thing, rather than each comment essentially creating a new post, making it impossible to follow all of the conversation about the original post.)  +9 / -1
  • This is how Pillowfort does it and it works fine. +3
  • Benefits of reblogging include boosting art/fic/other creative work -- comments on those things are lovely, but someone reblogging my fic helps get it to new readers who otherwise might never have seen it. I want a reblog/RT feature for this reason, even if it's a "pure reblog with no commentary" like a Twitter RT. +6
  • I do like being able to retweet with commentary and start my own conversation in my own space, without necessarily involving the OP. +2 (Though you don’t need reblogs for that--we did it on LJ just by C&Ping into a locked post with a link to the original.
  • Easy to just skip the ‘link to original’ step in that though… reblogging does offer a stronger guarantee of sourcing, and there’s been a fairly ferocious ‘anti-reposting’ campaign running long enough to have moulded and reformed a lot of people’s habits.
  • How about mini-reblog -- "I have added a comment to this post" + the title + link to the original post, instead of re-showing all eight screens of text and commentary. +6 +1
  • I'd like a non-reblog way to comment, but I'd still like to be able to reblog, even if it's only "without adding commentary" like a RT on Twitter - it's really useful for things like art or fic, where comments are lovely but reblogs help more people see your work +6
  • Have you seen the thing on youtube where people just comment with “+” as a way to like a comment and move it up in the algorithm? There could be something like that, a kind of comment that means “look, isn’t this great”
  • Whisperspace?
  • Losing reblogs would be site-breaking to me, but I agree that they’re not the ideal way to hold many types of conversations. I think we need to recognize that reblogging and comments have different goals, and one is not remotely a replacement for the other. Going back to comments after having reblogs is why I can’t go back to Deviantart, as an artist. People just don’t interact with fanart as much that way, we get so much more benefit from reblogs. Same for stuff like gifsets. However, not every post’s priority is to spread like wildfire. Rebloggability should be an optional feature for public posts, for people who do want the exposure, at the cost of some control of who sees and interacts with their content. That’s a very natural trade-off. Stuff that’s more conversation-like doesn’t need reblogging, it needs threaded comments, and it should get them.
  • When you add commentary on a post, and go back to it later, you can't tell which additional comments were replying to your addition and which to other people or the original post.
  • My strong preference would be for a comment thread to travel with a post each time it’s reblogged -- that there be one central version of a post, and a reblog is a sharing of the central post, not a copy-and paste -- and the original post creator has control over deleting it, or freezing/moderating comments, or changing the privacy level, if they so desire. +8
  • Changing the privacy level is sticky since it’s part of other people’s blogs now. You can’t let people reblog your post and then tell them their own followers can’t see it (unless you have those specific users blocked). Allowing a post to be reblogged is truly releasing it into the wild. It’s the price you pay for all the signal boosting people are doing for you, and it’s a fair one.
  • I honestly feel like Pillowfort got it right in this regard? The content of the post can be reblogged, but you have to click through to the original post to see the comments. If they’re threaded/nested you can clearly see what people are replying to. +17 / -1
  • Could it flag whose reblog the comment was made on?
  • My comment further up got lost, so I’m going to try to recreate it here. My vote is for no reblogging! +3/-4 But this comment seems to be really about changing reblogging/achieving the function with different names, which seems maybe too confusing.
  • To be more specific, I think the ways that people use reblogging could be recreated in different ways. For reblog-as-like, have a like button and an option that your likes show up on your feed for your friends to see (or not see, if they opt out of seeing your or all likes). For reblog-as-commentary, have threaded comments with a ticky box when you’re adding a comment that also shows the comment (addendum: the original post, plus your comment?) on your feed (and again, your friends can opt in or out of seeing your comments). For reblog-as-promotion, have a section of the comments where people just add a “+” or “<3” or whatever in order to make it show up on their feed again (or, and this is just a crazy idea, people could add more than one comment to things they really like! lol). Oh, and I should add that I think the no-reblogging feature might go hand-in-hand with the edit-posts feature. Like one of the main reasons why places like twitter don’t allow you to edit tweets is because retweets exist, you know? +8 / -2
  • +3 being able to follow someone for their commentary rather than original content.
  • The idea of using “+” or “<3” comments to make something show up in other people’s feeds relies on a feed/dashboard/what have you being curated by an algorithm, which I’m personally really against. I dislike how many/most social media platforms are moving towards this model because it narrows what I’m seeing - I’m not choosing to follow someone because I only want to see popular posts of theirs or things similar to what I already like, I’m choosing to see all the things they post. And then blacklisting within that can allow me to curate content I don’t want to see, but imo, a chronological feed has historically been one of the things tumblr got right. +4 >Right, but people DO use reblogs (without commenting) simply to promote the post onto their feed. OP’s suggestion doesn’t rely on algorithms. They’re talking about an ability to subscribe to the posts your friends want to share with you (by reblogging, whatever you choose to call it)(or not subscribe to them).
  • +7 for avoiding algorithms deciding what I see
  • Likes showing up on feeds is one of the worst technical qualities of twitter by far. Likes are a personal curation and minor social interaction, not a boosting method. No one wants to have people’s likes show up in their feeds. +9 (And no one wants their likes showing up in other people’s feeds. +3)
  • This idea works in the way we think about retweeting (promotion), but it doesn't necessarily work for reblogging. People view reblogs on tumblr as building/shaping their blog identities -- they want that material on their blog not to indicate that they like it or to find it again, but to make it part of their blog identity. (Aesthetic blogs are like the extreme version of this. I don't think we need to make this a space for aesthetic blogs but they're an example.) However, I like your idea of thinking creatively around these issues.
  • I posted about this before but it got deleted in the roll-back. I think the major problem with reblogs on Tumblr is that the idea of viral attention is a double-edged sword, because once you make a post, it’s out of your hands. The ability to edit and control all posts from the original (as a responder put, Pillowfort has this feature already), and delete the post entirely (per Twitter), might mitigate some of the mob mentality that has arisen from reblogging. +3/-5
  • I feel like this is not a good approach even while I agree with the danger or viral attention and dogpiling. A user shouldn’t be empowered at an atomic level with the ability to delete or edit the work and content of others. +1
  • In Livejournal (and DW), if someone deleted a comment, all responding comments still existed, even if the original text or the first comment was gone - it would say something like “User deleted this comment”.  If someone deleted a post, the post and all comments made on that post were gone.  I can therefore understand the fear of losing work if one reblogs something with a comment and then the original disappears (say, a long story chain made of reblogs or something - if the OP deleted the first post and all reblogs disappeared, that could be tragic for people working on the story.)  But there’s a simple solution to this: don’t put important content in reblogs.  Use some other method, like linking to it from reblogs.  Content shouldn’t be in reblogs, it should be in comments or in posts. +1 -3
  • ‘Content’ that people care about or put work into will inevitably be everywhere it can be.+2
  • Yes, and it’s so much better for artists that people are actually reblogging our posts rather than reposting! (At least sometimes, hah.) Our art’s gonna get around anyway.
  • It should be that double-edged sword, though. You want viral fame. Other users are contributing their own blog space to give that to you. In return, you don’t get to snatch something off their own blogs.
  • Nb Instagram doesn’t have formal reposting in-app either. They limit it to only being able to re-story posts you’ve been tagged in, and not picture posts at all.
  • Maybe be able to reblog things once? So then the post is on your blog, but you can’t spam it?
  • Mute post should handle that. Or the proposed feature to hide reblogs you’ve already seen with no added content, which helps as well when it’s not one person spamming, but everyone you follow.
  • You’re completely wrong, reblogging is 100% about promotion on tumblr too, that’s why content creators love it so! Yes, people do want it on their blogs for the “aesthetic,” but for the content creator, yay, that’s promotion! Please promote my aesthetic! Follow 4 more.

  • In my opinion, reblogging is what makes tumblr tumblr. No other site has reblogs with the ability to create massive quote chains and spread posts to thousands/millions of people. Until there’s an alternative that facilitates the spreading, discussion, and curation of all types of content like Tumblr does, I’m sticking with Dreamwidth. +2
  • That’s mostly an adaptation to lack of threaded comments, which is admittedly a disaster. I want you to get threaded comments with better conversations. Until a site lets people reblog my art, I’m sticking with Twitter.
  • I feel like the ability to reblog has become a big part of fandom behaviour. +8 You see a thing you love and you pass it on to show others. It’s used to curate your own space in fandom the way you would put posters on the walls in your room; it’s a way to show what you like and feel as well as a tool to promote the things you love and a way to lift other creators up and help others reach a wider audience. A lack of reblogs would be a very big feature for a new social fandom hub to not have. We have had access to it for over a decade and have had a pretty good time with it. There are other places without reblogs. DW, Instagram, etc. I wouldn’t join another fandom place without a reblog tool; it’s what I miss in other places! +8
  • Ok but if interacting with a thing caused it to show up on your feed so that your followers could come and interact with it too, would that not serve the same purpose? Except that by keeping the original post and then attaching all the commentary to it, you get to see how lots of other people are interacting with it as well. +1 / -1
  • No, I want reblogs, and I do not want likes forced on your followers at all! They aren’t remotely interchangeable.
  • I’m not completely sure what you mean. Are you thinking like how people can see your twitter likes?
  • Yeah, like your feed could be a list of “username liked” [post] and “username commented on” [post] (to be clear -- where you can actually see the art or the text before the cut or whatever, not JUST a link) and then when your followers click on it, it takes them to the original post (or your comment) so they can also like it or add to the conversation or whatever. -2
  • That’s not something I am looking for in another social media account. I want the ability to reblog the pretty art and have it on my page. My tumblr account is first and foremost for me and what I want on it. When people reblog my work, I enjoy seeing the tags others put on it as they send it further out in the world. When I reblog, I do it both because I want to keep it in my space, and to show others. +1 (So what you are suggesting are two feeds? One where you can see what you liked, and one that is your original posts? So basically a public like page?) (No, just reblogs, why do we need a feed for likes?)+1
  • I feel like this is all getting very complex when it doesn't need to be. The suggestion up-chain about reblogs essentially being the original post, with all commentary attached to that post as threaded comments really addresses most of what is being discussed here. The post that you want to share shows up on your blog, but rather than having infinite iterations of comment chains that are impossible to follow, it's all contained within the comments of the post (ala LJ/DW). As pointed out above, this is the model that Pillowfort currently has, and I think it's an elegant and simple solution to the issue.
  • Yeah, I think I agree (having never seen Pillowfort) -- my whole point with the “no reblogging” opinion was not that I don’t want people to be able to share things they like (obviously that’s a major function of social media) but that I hate the way tumblr reblogs are like, creating infinite copies of the thing rather than just linking to the original. It makes having any sort of coherent discussion almost impossible.
  • I have found having coherent discussion in the... little reply notes perfectly possible. Not the reblogging, but the replies. Just not very long messages, which I think is a feature a reblogged post could have, like on PF. The space for longer replies -- if they could be threaded or not I can’t say, as I lack the technical skills. >I’ve done this too, but it’s infinitely less usable than an actual threaded comment section.
  • I’m very into this model, except that I will very much miss the affordances for terrible jokes that tumblr text post reblog chains provide. No more “randy your snacks”, rip
  • I’m not into this model, and I’d be sad to lose those things you said.
  • I think the PF system can work. This was meant to be a response to the “I vote for no reblogging” argument, because I feel the ability to reblog adds something that set tumblr apart, and what I will miss the most. Then I was just trying to see if I could get a clarification to what the person who replied to my addition meant. (For sure it has negative aspects as well! But for me the positives far outweighs those, and the positives are a big part of why I am even in fandom.) +2
  • Pillowfort’s reblogging would be likely the best option here
  • Oh, I agree. Being able to share content like that is something that has become central. But it definitely has its negative aspects as well, which is why I think the PF system is a good compromise. What I miss about the LJ days is the community building. I haven't made a single new friend through tumblr, whereas I am still friends with people I met on LJ over a decade ago. And I think it's the emphasis that tumblr has on spreading content without having a coherent method for discussion about it that has led to that.  +1 / -1
  • Yes, I agree
  • My experience is the complete opposite, since most of my fandom social circle are people I have met on tumblr. I’ve had people I met on tumblr come meet me from different continents, and traveled to meet friends in turn. That’s why I think it’s important to consider that no experience is absolute, and hear out different viewpoints. Some managed to find coherent discussion and meet people while others didn’t. It might not just be the social media, but a difference between different fandoms, or just random chance.
  • This is my experience too. I was on LJ for years and it felt like a VERY high bar to interact with other fans or individuals if my own personal content wasn’t as compelling. Being able to reblog and curate a feed as an expression of my fandom interests has exploded my involvement with fandom and with others. It turned me from a consumer of fanwork to a creator of fanwork and I’ve built a lot of friendships. More than I ever did in LJ. +3
  • I’m sorry you haven’t made friends, but that isn’t on the platform. I very much made friends, fantastic ones.
  • I realize it’s possible I’m an outlier here, but I feel very strongly that the casual way tumblr’s reblog system allows you to interact with other users and gradually build relationships with them without having to go onto the blog of a complete stranger and directly talk to them is really one of its best features. That tumblr encourages users to treat reblogs as their own, in a way, means that you can feel comfortable adding commentary even when you're not sure of OP's welcome. +27
  • Obviously no one’s experience is universal, but LJ never felt particularly welcoming to me because of exactly this problem. I am opposed to even the PF style system compromise (that still would have a clear owner of a particular post rather than a more ambiguous OP) above for this reason, although if I turn out to be in the minority it seems better than no reblogs at all. +7
  • I also think that original post as hub of all discussion is often not appropriate. If someone posts a picture they found of something, say an artefact from a museum, then my response to it for my followers related to fandom isn’t really meant for the original post in someone else’s space.  +4
  • Right, if I reblog a picture of a bear with a comment that references some fandom specific meme, that’s good fannish content for me & my followers but would be inappropriately derailing in a threaded conversation about habitat on a conservation blog (or w/e). But I also don’t want to just repost the picture because that would lose the credit to OP, & linking would potentially cut off portions of what I’m referencing unless every single follower was willing to click through just to see the picture/its original context. +1
  • I think this is an important point, and indicates reblogging has 3 sub-functions: commenting (LJ/DW/PF style), boosting (related to curating, when you have a complete copy on your own blog), and appending (what’s discussed here, where you might add a comment that significantly alters the original post in a way that may not be appropriate in threaded comments). I think if these functions are separated from each other, appropriate restrictions could be placed on each (for example, a personal blog may want to restrict appending to friends only, a deep discussion blog may want no appending at all, and a blog that posts only pictures of bears may be OK with unrestricted appending. Separating this by post from who can comment/boost would be a huge improvement.). +2 (+1 for the explanation of functions, but -1 the conclusions, I don’t want restrictions on reblogs, reblogs belong to the people, that’s fundamentally what they’re about. You cannot ask people for “boosting” without giving them the opportunity to collaborate in exchange. +1 Rebloggers don’t exist just to fan OP’s ego.)
  • Is there a way to maintain that function of “ownership” and detached commentary without devolving into a million branching tumblr-style reblog chains? Maybe limit the number of reblog “levels” away from the OP?
  • I’m against limiting who can reblog something beyond certain measures that feel reasonable--friends can reblog (meaning friends of friends can see it but not reblog), everyone can reblog, or no one can reblog. Maybe even just everyone or no one, at least to start. I feel like having a comments option would address the concern of making this more manageable than it currently is on tumblr--people would likely keep their on-topic conversations in the threaded comments, so the only branching reblog chains would be ones where an appended comment changes the topic of conversation significantly. It’d be interesting to me if you could then have threaded comments on the appended post, although that may be a more complicated ask. +1 / -1
  • Literally what good is a reblog no one else can reblog? Just give your friends permission to repost, at that point.
  • I’m against Pillowfort-style reblogs because it gives the OP complete control over the reblog chain. Like. Imagine: Russian plant maliciously posts anti-vote stuff or something. Person reblogs it and refutes it. Russian plant deletes the reblog, despite it not being theirs, therefore both deleting content off of someone else’s blog and presenting the illusion that all of the reblogs agree with them. Or, someone starts a smear campaign, is refuted, deletes the refutations off other people’s blogs. Like imagine if you were on a forum and if you quoted someone in a post they could delete your post. And not only was there nothing you could do about it, their ‘right’ to delete your content was baked into the site design. Seebs on Tumblr describes it as “if you make a funny post, and people end up having a long and interesting conversation in the reblogs, you can delete the entire thing.” which would really discourage like...actual conversations. It’s like, if you want to comment on something somebody says, the only safe way to do it is to start a completely new post with an embedded link, or repost. +1 (Whether PF does this or not, I don’t want it.)
  • It … doesn’t do that, I don’t think? I’m on Pillowfort and I don’t think I can delete reblogs without deleting the entire post. Deleting the entire post makes it hard to gain traffic so if I was trying to delete reblogs for some nefarious purpose, surely it’s outweighed by the hassle of then getting my post the same level of exposure again? It’s basically like deleting a forum thread. Sure, your stuff goes but mine goes with it. The ability of a thread creator to delete the thread doesn’t stop people contributing anywhere else, either.
  • Could a PF-like single post that the reblogs link to be a solution if, as in the above comment, the OP can’t delete reblogs? Perhaps once a post has been reblogged, the OP can’t edit it, or can edit it only to add text or links, not to delete things. This would allow correction, clarification, etc, but wouldn’t allow replacing the original content with a jump scare or porn imagery. Perhaps reblogging creates a record/canonical copy on the servers, and the OP can delete their copy or edit its appearance drastically, but the link to the canonical (on the server, identical to the original, but with all social content+possible additions) remains if they don’t delete it?
  • I like the control of pillowfort though. I feel like the solution to people abusing the edit function is simple remove the people abusing the edit function? I like knowing that if I make a typo or a factual error I can correct it later and every version will be the same level of up-to-date.
  • Agreed. The problem with the old Tumblr edit function (bunnies into swastikas) was that the original post could be deleted and the user could ghost with zero consequences. If we're already controlling for bots and throwaway accounts, and deletion is "whole post or nothing", malicious editing would be significantly less attractive from the get-go.
  • I'm not sure "build a system with strong affordances for malicious editing, then ban everybody who takes advantage" is exactly the best way to go about this... maybe some kind of snapshot system, where every time someone reblogs a post, the post appears on the reblogger's blog as it was at the moment they reblogged it, but still links back to the canonical version on the original blog for comments and stuff...? I dunno. <hmm, that sounds promising, but isn’t it the same as asking to save a whole version history again?
  • Idk - my theory is that whatever we do, someone, somewhere, will try use it to be an asshole. Might as well not tie ourselves in knots trying to predict how when we could just hit the assholes with a banhammer. +1
  • If we do end up with a system something like tumblr reblogs, could there be an easy way to lock certain reblogs so that only you can see them? For instance, sometimes I want to reblog 50 star wars fic rec lists in a row so that I can find them later, but I don’t want to subject everyone who follows me to 50 rec lists in a row.
  • I feel like this ties into discussions about bookmarking, etc, further up? A good bookmark feature would essentially fulfill this function.
  • Sounds like the ‘like’ function tumblr has now
  • Not really, because there’s no way to search or tag your likes. Checking tumblr quickly, I have 75k “likes” and no way to sort through them. (Also, trying to view them crashes the page, because tumblr.)

Single Central Identity; Tiered protections

  • This creates the possibility of allowing strong verification of the central identity on a 1-time basis (with verification proofs discarded / destroyed after they have been used to create the strongly verified) - which may be desirable to help exclude bots, etc. +15/ -2 (for anyone who wants nothing to do with tying an identity to a fandom handle) -1
  • Sensitive users (at-risk, SW, trans, suffering DV, etc, etc) can have a separate verification path with different, stronger protections. Establishing at the start that certain classes of user may need additional protections, but still want verification and the benefits that provides would be a powerful signal to marginalized groups of users. +4/ -8
  • I don’t know what you mean here, but if you’re implying only some users should get privacy protections, I strongly disagree. All users should get the strongest privacy protections, or it should just be anonymous. +12
  • Forcing people to admit their identity/struggles for stronger protections is asking for gatekeeping. +65
  • However, "stronger protections + more hassles in verifying ID" could be an option available to people on request. Like 2-factor authentication, or other verification options: "always send me an email before changing major account settings." +5
  • I think there should be an option for 2-factor authentication and/or other strong protection from getting your account hacked, but I don't think your legal identity should be tied to anything (which is what the idea of people who need protection more--sensitive groups--leads me to think of). As long as there are strong options to protect someone from stealing your account and you can remain anonymous and separated from your wallet name if you so desire, there doesn't really need to be tiered protection based on the users themselves. +8
  • If 2-factor authentication is used it should be an option, or at least have the option to just use e-mail or something - sharing your phone number etc should never be required +7
  • Authenticator apps or hardware keys should be the first-choice 2FA options, with SMS only if the user actively wants it that way instead of having it pushed on them. Better for both privacy and security. +1
  • This identity model also allows for formalisation of the “multiple / throwaway account” pattern seen on other sites -- but built-in as an expected (and even encouraged) user mode. This allows for users to have separate aspects that better reflect the separate aspects of their lives. +9
  • Also would be nice to have the ability to immediately see how long an account has existed for (or how active it is). If account is very recent, may be more obvious it’s being used for trolling (or whatever) +35+1 ⇐ I can’t stress how important this is
  • Normal users will be very visibly obviously different from bots / malicious users as the latter will not want to verify, or to create multiple attack personas that are immediately traceable by staff back to a single root identity that can be banned, taking out all identities. +5
  • Access control to allow users to e.g. “Not see anything at all that was not generated by a verified (or sensitive verified) account”. If this was the default mode, then the attack surface for malicious users against casual / new users would be hugely reduced. +2
  • Multiple "aspects" belonging to the same account should be allowed to participate in the same communities, e.g. for roleplaying purposes, throwaway accounts on advice communities, etc -- basically, don't do what Imzy did. (If one aspect gets banned, though, the whole account gets banned.) +4

  • Phone number verification with ability to make multiple (indefinite?) main accounts from one phone number. Still not “your real name” level of identification, but strongly discourages creation of excess burner accounts -9
  • Not necessarily - it’s easy to make a bunch of google voice numbers to get around a # verification barrier, at this point blocking a phone number is probably as effective as blocking an email address
  • I hate phone number verifications; my only phone is paid for by work, and I'm always worried about using it for personal projects. And not everyone has a phone, or a smartphone, or one that they control. +5
  • Telegram’s phone number verification is why I lost track of a lot of my chat-friends when Skype for Linux finally collapsed under its own deprecation. I don’t think phone verification is the way to go.
  • Some countries require registration of phone numbers to avoid the whole burner-phone thing, and it ties it to your national identity. If you’re using the social media account to do something that your country criminalizes, I think it might be best for you to hold nothing that could incriminate your users and expose them to legal prosecution, torture and even death.
  • Also, moving countries/switching sims would lock you out of the account

Should image descriptions (“alt-text”) be required or strongly encouraged?

  • Mandatory image descriptions (and the ability to mark images as decorative) +10/-4
  • What is “marking images as decorative”? +3 to descriptions but -10 to mandatory
  • It’s part of accessibility settings in course management software and other things. So if your image is just a divider or something that doesn’t carry meaning, you can tick “decorative” and not describe it.+8
  • The ability to have image descriptions is great. Not sure mandatory is cool +4
  • How about “strongly encouraged?”
  • Interface itself should encourage doing it (such as having to manually opt-out of describing each image every single

time you upload one), not just policy. +43

  • Similarly, would be nice if the interface could encourage attribution, including a direct link to the original post that contained the image, if available. +30
  • This would be great for recs in the middle of unrelated posts, so that people can reblog from the source. +3
  • I feel at the very least it should be easy to attach the metadata of “I drew this image” to an uploaded image.+1
  • Could have a short standard list of image descriptions ("fanart/person photo/landscape/meme/explicit" etc.) on a dropdown, to encourage some labeling even if in a hurry or on mobile. +17-2
  • I think a “suggested image description” feature would be feasible to do with machine learning.-1
  • lol, no it wouldn’t, where are you getting the AI that can parse images that well? +4
  • This level of image description isn't useful +1
  • Agreed; making it easy to provide bad image descriptions isn’t the answer
  • You should be able to choose in your settings whether to completely hide image descriptions, replace images with their descriptions, have both images and descriptions, or show descriptions first and click to show the image. This covers a broad range of accessibility, from saving bandwidth for people who can’t see the images anyway to effectively acting as a content screener for people who have preferences not usually warned for. +12
  • Make it against the rules to intentionally put something misleading in an image description. I don’t want an accessibility feature hijacked for jokes and commentary rather than just describing the image. This should be a TOS violation. +15 / -4
  • I disagree. If someone uses the alt-text for an image to make a joke (a la xkcd) instead of describe it and the user viewing the content doesn't like that, they should just block that user. Same as if people don't use tags so you can mute their posts, or if people don't put things behind cuts and it makes your feed hard to read. Some things will need to be user decisions, rather than policy. → xkcd just uses title text (their alt text is the comic title, no other description). Image descriptions for accessibility purposes aren’t even visible to most users who don’t look at the source/use a screenreader, so it would be either ignorance or deliberate malice of some kind. Agreed that it’s not at TOS violation level, though. +3
  • Alt text should be separate from image descriptions. <--+8 Image descriptions are an accessibility feature. Don’t use them at all if you want, but don’t make them unusable for blind people who need them. To clarify further, they should be usable by screen readers, but not exclusively for that. People could also use the ability to replace images with text, see images with the text, or preview text before clicking to see the image, to screen content, save bandwidth, cope with sensory processing disorders, to help “explain” images to 0users who struggle to comprehend what they see, especially recognizing faces or emotions, e.g. people with faceblindness, some but not all autistics.+2
  • The TOS violation is not "mislabeling images;" it's "deliberately attempting to interfere with other people's use of the site," which is against the rules in most places. +4
  • This sort of thing seems very difficult to enforce.  What do you define as “intentionally misleading”?  If I have a picture of a border collie playing with some cantaloupes, and I tag it “melancholy” [a bad pun], am I being intentionally misleading?
  • Yes. A blind person doesn’t benefit from your cute pun, because they can’t see what you’re riffing on. A blind person needs to know that it’s a border collie playing with cantaloupes, and that’s why your alt text or comment of “melancholy” (which everyone can see, rather than being hidden by people who don’t need it, which is a bad way to tell a joke anyway!) is funny. Using an accessibility feature to make jokes that disabled people are barred from laughing along with isn’t very funny. You will have so many other text areas to make jokes in. Let disabled people have one thing that lets them keep up with what people are laughing at. +6
  • Put a real image description in, and then put your pun at the end, or in the text area of the post. Crap descriptions are about as frustrating and debilitating as crap captions on video content: they actively make things worse than ‘no description’ would have.
  • This is also “enforcing having good manners”. The platform’s job is to make it possible and easy, so that individual users don’t have to do workarounds. “Don’t be a dick to people who need the alt text” is a reasonable topic for community self-policing, not for a paid team that’s going to be neckdeep in CP takedowns+4
  • This should be enforceable by moderators of given communities but I don’t want rules at a site wide level that require overseeing the content in that way, not just because it’s unfeasible but because it makes every post on the site have to essentially be approved or reported to a central authority.  

  • If we’re going to require or “strongly encourage” alt text, there needs to be some kind of guide somewhere about how to even write good alt text, because a lot of people do not know how to do this. And this sounds like a great way to get a lot of garbage alt text on every image on the site. +58(The dropdown idea would help with this.)+1+1
  • What if we crowdsourced image descriptions? Like, if you see a post go by on your dash that’s got an un-described image in it, you can open up a little popup or something and type in your own image description. That then gets sent to the OP for approval before it gets added to the post for all to see, so you can’t troll. +3
  • There’d need to be a very good guide on creating image descriptions in this case (no good if I (sighted) write a description for OP (sighted) that’s actually useless to the blind comm) <- this is true regardless of how image descriptions are implemented
  • Agree this is tough but it could be a good solution and I’ve seen it in action on Tumblr. There are plenty of people willing and able to contribute descriptions for random undescribed images they come across <-yup, that’s what I was thinking of. This would be a way to do that, but make it so you don’t have to go hunting through the reblogs to see if anyone has described the images yet

Orphaning Posts & Entire Blogs

  • Breaking the link between creator and content or blog. +4
  • Concern I have seen with AO3 is you give up all control of your work when it’s orphaned -- and nothing you do will get it taken down if you change your mind
  • I feel like that’s the point of orphaning, though. It’s saying “I understand this is useful to people but I do not want to be associated with it anymore.” Otherwise you’d just delete it. +13+1
  • Possibly orphan with the ability to delete later? A little flexibility seems sensible to me (though i have no technical knowledge about how this might work.) -1-1
  • Three-status system: Live -> archived or soft-deleted (exists but invisible/inaccessible) -> deleted (gone forever). This is a pretty standard paradigm
  • Invisible/inaccessible is basically the opposite of orphaning, which is “visible and accessible, but not connected to the author.” A soft mid-step between orphaning and deleting would be more like, “author is listed as anonymous and can’t be identified on the front end, but the author still has control over the work,” kind of like the Anon group on Ao3.
  • So maybe 4 post statuses: published, private/invisible, anonymized, and deleted, but the OP-user still has control over anonymized posts?
  • On AO3 I put works like that in Anonymous collections instead, so I can retain control, continue to receive notifications, and can reply as “Anonymous Author.” It obscures my identity without having to make a sock (and even socks have a sort of identity if there are multiple fics/posts on them) and I can undo it later if I want to claim my work. Something like this would probably work better for most people. Imzy had some options for logged-in anonymity (per-post randomly generated “Anonymous[adjective][noun]” accounts that were the same within the post. I’d make that within post or within thread, optionally.) +13
  • But a feature many wanted on tumblr was the opposite -- to remove notifications from a post while not divorcing it from your identity necessarily. +2+1
  • This is somewhat a function of the way Tumblr aggregates activity on your content and how Tumblr disseminates your content, leading to tens of thousands of people not realizing that tens of thousands of people have already made the same useless comment “to your face”. It may be less of an issue on another platform.
  • +8 for orphaning entire blogs; not sure I like the idea of specific posts (mostly because it sounds complicated to do.)
  • Concerned with whether this would allow for circumvention of abuse enforcement (abdicating responsibility from the user to the platform itself.) +8
  • Yeah, the idea that someone could make an extremely controversial post that generates wank (or whatever the kids are calling it these days) and could then decide to orphan that post so that their screw up isn't associated with them anymore is concerning, to say the least.
  • Or just outright making racist or sexist (or etc) posts and immediately orphaning them.
  • In some ways this is a plus, though -- lets people who make a mildly dumb but not offensive post that then gets reblogged 20000 times by people calling them idiots not have to literally move identities to escape. +3+1
  • This could be addressed by giving people the ability to block reblogs.
  • What about having the ability to publicly orphan a post, but the moderators (or whatever sorts the site content) can still trace it back to the OP? This may or may not be still visible as one of “your” posts from the user’s view, but the data would still be recorded somewhere internal. This way only the content mods can identify that person, so they can avoid harassment, but if a user abuses this feature to spread racist etc. content repeatedly, they can still be held responsible by the site for not following the terms.+1
  • I don't see a difference between that and "anon" posting.

Tags as Metadata vs Whisperspace

  • I like the two-tier tag system, with globally-searchable tags (like tumblr) and organizational tags that are specific to the user, so for example I can tag something with “hockey” and choose to have it appear either in the global hockey tag and in my own hockey tag OR only in my own hockey tag. Is that part of what we mean with the supplemental tags? +11
  • That's what I was thinking when I suggested that a user could define what's supplementary or metadata!
  • Users could use their organisational/non-global tags as whisperspace or as organisational or as both
  • so like two tag sections, one that's "global tags" and one that's "personal tags" with the latter being used for whisperspace and/or organization of your own blog without showing up in tracked tags +7
  • I think this is a useful compromise and doesn't require huge energy setting up canonical tags or anything. People know when they tag which space they're tagging in and what that space is intended for, and then can still use it however they like, understanding that if they use the global tag "squishyface" and most people don't know that means Cucumberpatch, it won't be globally found. Etc. And also lets people tag "hockey" in their personal tag space without putting it in the global "hockey" tag +3
  • I mentioned this usage of tags in a paper in college, it’s so fascinating.+1
  • I don’t think the whisperspace even has to be in the form of something tag-like. If it’s just a place in a post where what you say is more ephemeral, whatever “ephemeral” means in relationship to the rest of the post structure (i.e., doesn’t show up in reblogs if there’s reblogging), people will use it to put those meta-thoughts or in-jokes. It could literally just be a box called “whisperspace” that shows up in smaller text. +3
  • I love the idea of having global tags, personal tags (still metadata/organizational), and then this third whisperspace commentary spot that doesn’t require indexing +4
  • If we have a global and personal tags section, why can’t the personal tags section also be the whisperspace? What benefits could a separate space be?
  • But who's going to enforce that? +3
  • Should be a user decision. You're creating your own tags the way you want to, and if you use a "supplementary"-style tag as your own metadata tag on your own blog, you should put that in the metadata section. +5
  • Maybe only metadata tags can be searchable, in order to quantify that distinction? +4/-1
  • AYRT but that only takes care of an individual's person organisation system (a-la dreamwidth); what do we do with tags that are browsable cross-site (a-la tumblr)?
  • AYRT the way I see it, there are two options: 1) people will eventually learn that if they want their posts about X to be searchable when someone searches for X that they have to use the site agreed-upon tag or it will be a separate set of search results; 2) tag wranglers a la AO3 (I'm partial to #1, at least to start, since people who want their stuff to be found will pretty easily figure out the site's consensus for something and use that tag. Or use multiple tags.)
  • My issue with tumblr tags has always been that tags then get used as a community space, so we get etiquette rules like “don’t post hate in the tag”, which makes tagging for personal organizational purposes more difficult imo. +2
  • This might be addressed by layout - if the tags show on the side instead of underneath as part of the post, they won't "feel like" extra post content. +2
  • I definitely do not want tag wrangling. I love AO3, but tag wrangling is an ongoing server maintenance issue that will never go away and that affects the scalability of the site very negatively. Not to mention the fact that tag wranglers have the highest rate of burnout of any volunteer on a fandom site in history. +22
  • On the one hand, I fully support "tag the way you want" (as an over-tagger myself for meta sorting purposes on Pinboard). On the other hand, if your tags show up on my feed/timeline/friends page and takes up two mobile screen sizes without me having the ability to hide it somehow, I will develop murderous tendencies. How to balance the two? >The Xkit extension that auto-crops the tag field when it’s over X rows/lines
  • Nested comments will probably also take care of the supplementary tags anyway. +5
  • Ability to search on multiple tags. For example, you want to see posts (yours or all) tagged with both "Marvel" and "fanart." +20
  • Ability to exclude tags from search results. +16
  • Ability to search your own page for the tag as well as the global site for the tag, independent of each other. +7
  • This would be solved by ability to exclude users from search results, which I think should be implemented as well
  • Suggested tags that you can autocomplete, either from overall site tags, or from your own tags. Could be an option only for metadata (aka searchable) tags, or for both types of tags if we go with a 2-tier system. +4

Mobile App vs. Responsive Design

  • I don't think we need an app straight away (or necessarily at all), as long as there's a good mobile interface with an easily accessible and user friendly posting page and reading view. +51
  • I don’t have space for another app on my phone, but I can always open and close a browser tab. +2
  • Agree, if it doesn’t need to be an app then just focus on having a mobile responsive site
  • No app at all if we want to have NSFW content, that’s just an unnecessary vulnerability. +42 (Hello tumblrgate anyone?)
  • Apps don’t have to be distributed through Google or Apple. Apps can be installed on Android by anyone who enables Developer Mode. (You literally just tick a box.) Apps can be installed on iOS by anyone with a jailbroken device. Cydia is an alternative to Apple’s App Store for people who use non-Apple-approved apps on iOS. Apps could be distributed without worrying about app stores or NSFW.
  • Assuming people are technically capable of doing so. Not having an app on the app store makes it real difficult for lots of people (security restrictions, blah blah). I think it’s better to just optimize for mobile. +10
  •  +7; this doesn’t mean there can’t be apps (which could also be built by a third party provider), it just means developing an app shouldn’t be a priority as long as mobile UI is good.
  • As an addendum to this "Apps" argument: work provides my phone. I literally CAN'T jail-break it without completely messing up some really necessary things. Mobile design is better than an App. +4
  • Some games (Pokemon Go) won’t play on a jailbroken phone anyway.
  • I had to jailbreak my phone after it stopped getting support updates and losing apps like Go and Banking has been hellish. I wouldn’t wish it on anyone. Mobile browser UI should definitely be the initial focus.
  • Email alerts can take the place of push notifications (though that might end up with putting the domain the site's on in some email providers' spam lists.)
  • Digest notifications could also be an option? Like how AO3 gives you one kudos email per day. +15
  • Opt-in SMS notifications in place of push notifications if no app (which I think is a good idea given, well, tumblr).
  • I believe this can be done in-browser as well, e.g. Facebook.
  • Should be built from the ground up with an API that could facilitate app development, because that’s a pain and a half to add later. +4 if the API is good then an official app becomes less important, because third parties could make their own more easily
  • This is why I suggested “API with full programmatic access to everything a user can see.”

Age Requirements or Restrictions

We Likely would have to be 16+ because GDPR, but we could maybe do an AO3 thing and be 13+ depending on the user’s country

  • I think sixteen+ might be an acceptable middle ground there.  But honestly, at this point, I would prefer 18+ just to head off any of this crap before it gets entrenched.  If only because then, when they try to harass/doxx/etc. we can boot them without much fanfare or dogpiling.+6
  • I like the idea of 16+ because it leaves some wiggle room for a user to appeal to Support with “I’m a minor and I’m actually getting creeped on, I Need An Adult” without getting TOS’d themselves. (Even if they’re 13 and are lying about their age.) BUT if we admit there are minors on the site at all there will have to be minor protections, legally. +1

What are people's thoughts on age requirements in order to join, or perhaps separation of blogs/accounts via age? I know that I for one would prefer not to interact with anyone under the age of 18, and I am quite sick of the age discourse that's frequent in fandom these days due to younger teens being quite vocal about it. +3 /-3

  • I think some will be necessary, at least if we want Americans to be able to use the site :) But I like the idea of being able to rate your own blog/posts as appropriate by age group. +4
  • However this is dealt with, I would prefer if it was in a way that didn't play into the assumptions of some people (mainly younger) that the simple fact of being an adult is inherently predatory. +12
  • Yes, agreed 100%. IIRC, LJ was 18+ and there weren't too many hurdles there. TBH, those younger than 18 can still lie and say they're of age (a time honored fandom tradition! ;))  (Yes, exactly. I remember being under 18 and clicking through anyway.)+3
  • When was LJ 18+? I remember it being 13+. What I remember people doing sometimes, was make locked communities that needed admin approval to join, with only adults being allowed, so that was pretty much entirely in the hands of the users.
  • You would set your age as whatever it was. Then certain coms and journals were 18+. If you were under that based on the birthdate… you weren’t getting in. Period. I remember this because I accidentally set a mature RP account for a mature game for a birthdate under 18. I was PO’d when I figured that out.
  • I just lied for like four years until I actually did turn 18. Oh, the olden days.
  • LJ added age limitations on some communities later; I don't remember if that was 6Apart or SUP, but it wasn't always there.
  • Wow, it’s been too long since I was on LJ. I remember that being in the hands of the community admins. But I might well have that mixed up.
  • Oh, gotcha! I knew there was some kind of 18+ moderation, but I was over 18 before LJ became a fandom thing, so I didn't retain any of the specifics, clearly.
  • Yep the only reason I remember it so clearly was that one stupid RP account I’d spent forever making perfect only to not be able to use it. >.<
  • asjdfls how frustrating +2
  • The minors are always going to lie, but still an age requirement could be useful? Or being able to have a blog that the minors possibly can’t see idk?