1 of 26

Prioritization Process for FOLIO development

Revised proposal by Prioritization Process Working Group

May 2022

| www.folio.org

1

2 of 26

Initial proposal - highlights

  • Additional SIG prioritization
  • New tools for SIG and institutional ranking
  • Higher level ranking (epics instead of features)

  • task after initial presentation: gather feedback

| www.folio.org

2

3 of 26

Considerations

  • Use surveys for institutional prioritization?
  • Combine SIG and institutional ranking?
  • Frequencies

| www.folio.org

3

4 of 26

Charge

The prioritization working group is to propose a new process for identifying priorities for new development in FOLIO that is equitable, transparent and clear. The group will consider how to get input from the community on priorities, taking into consideration institutional as well as SIG requirements. 

The group will as part of this process investigate possible tools that could be used for rankings. For this purpose it may research ways of prioritizing in other open source or commercial projects.

| www.folio.org

4

5 of 26

Background

There is an interest to no longer use JIRA for institutional rankings, as it is not seen as the appropriate tool. Existing JIRA rankings should be exported and made available in a spreadsheet for historical purposes.

We may need a way to import rankings back into JIRA.

There was a pointing process introduced for Kiwi that was based on JIRA rankings, as only R1 and R2 features made it into a spreadsheet for institutions to point on. With JIRA being no longer used for rankings, there is either a new process or a new tool needed.

| www.folio.org

5

6 of 26

Working group members

  • Thomas Trutt
  • Marie Widigson
  • Michelle Futornick
  • Jacquie Samples
  • Debra Howell
  • Martina Tumulla
  • Jana Freytag (convener)
  • Martina Schildt (convener)

| www.folio.org

6

7 of 26

Links and documentation

| www.folio.org

7

8 of 26

Flowchart

| www.folio.org

8

9 of 26

Considerations

  • Submission of new requests
    • not directly in JIRA (quality assurance)
    • preferred: submit via SIG or PO
  • App Ideas page: https://wiki.folio.org/display/PLATFORM/FOLIO+App+Ideas
  • add new requirements to JIRA like new apps; clear differentiation to existing tickets
    • Julie in chat: Or they get added in Jira, but not as a "feature" (which has a clear Definition and structure for dev work), but as [Name tbd] "new request"
    • POs no time to review these; That work can be done by the SIG; From the SIG's to PO's after first review
  • first via SIG or PO; maybe unclear size
  • no “permission” needed
  • document requests
  • Charlotte in chat: Often new requirements is driven by libraries, who have this requirement as a road blocker for them in order to adopt FOLIO

| www.folio.org

9

10 of 26

Considerations

  • Add SIG rank
    • additional field for calculated rank in JIRA
    • prioritization as SIG via separate ranking tool
    • regularly; frequency up to SIG/PO
    • several weeks to rank
    • extensive communication
    • cross-app features: all related SIGs rank
  • what are SIGs ranking on? only new requests? development? analysis?
  • prioritization can feed into roadmap
  • Kirstin in chat: That prioritization is the base to decide on development and open up the opportunities for hopefully future developer Teams to get orientation, what is needed most.

| www.folio.org

10

11 of 26

Feedback

  • Charlotte in chat: And also we need to plan work on features, which span over several apps; so the work being rolled out is useful for the implementing libraries, and not just half-baked functionality
  • be careful how prio is presented
  • separate streams (prio, roadmap, resources)
  • Jana in chat: We have discussed, that NFR and basic industry standards should not be part of the ranking but rather something that will go into development automatically as well.
  • Kirstin in chat: +1 Owen. Prioritizing gaps / needs/ requirements is just one half of a process to get things done. But it is the necessary start.
  • Harry: Owen, I think you have that exactly right. While the priorities help inform interest, it in no way decides what gets worked on next.
  • Kirstin in chat: But development Teams / Stakeholder can base their decisions what they will develop on that Information. That is an indicator what customers want. And other Teams can pick up, what is left over …. it Would be an opportunity to have a transparent process who picks up what - on each Stakeholders own decision. That is not questioned.

| www.folio.org

11

12 of 26

Feedback

Maura Byrne in Chat: Should features fit into epics?

Gang Zhou (SHL)in Chat: +1 Kristin

Ian Walls in chat: I think all this assumes that we have a single pool or resources that are working against a single queue of features/fixes. I don't think that's realistic in an open source setting

different groups are going to have different priorities, and they're free to apply their own resources, and those of like-minded institutions, towards those priorities

Harry in Chat :+1 Ian

Von Gang Zhou (SHL) an alle 04:19 PM +1 Ian

Von Thomas Trutt an alle 04:19 PM That is defiantly true.

Von Marc Johnson an alle 04:19 PM Ian, I think that is a really important point. Which becomes even more important when a feature crosses app boundaries and potentially involves multiple teams

Von Ian Walls an alle 04:20 PM as a vendor, should I get to the point of being able to contribute code, I'm going to do what my partner libraries need done. I want to be sure that's given back, but that's where my directive is going to come from

Von Marc Johnson an alle 04:20 PM I would also note that development teams don’t tend to make autonomous decisions about what they build. Those decisions tend to come from outside

Von Kristin Martin (University of Chicago; she/her) an alle 04:20 PM I hear from POs that they want these ranking from both the SIG and the institutions. I'm not hearing that they feel bound, but they want to get input into direction.

Von Kirstin Kemner-Heek an alle 04:22 PM Yes, from the Stakeholder who hires then, for example. That happens in our FOLIO Environment. Other Options are possible. As well as that team decide to do something completely different. All this doesn't exclude any other decisions. But helps to keep an overview.

| www.folio.org

12

13 of 26

Considerations

  • Institutional ranking
    • keep one field for calculated institutional rank in JIRA
    • same feature/epic list as SIG
    • same ranking tool as SIG
    • frequency: twice a year
    • several weeks (> month) to rank
    • add calculated total rank to JIRA
    • make variance of ranking available through tool

| www.folio.org

13

14 of 26

Considerations

  • Implement weighting for ranking institutions
    • for implementation status
    • for consortia vs. single institution
    • not: OLF members

| www.folio.org

14

15 of 26

Considerations

  • What do we rank on?
    • not preferred: change to ranking epics instead of features
    • feedback: “Something in between”
    • find a way to have rankable bundles (discuss with POs?)

    • NFRs do not need to be ranked
    • spend X amount of time in development on those (TBD)

| www.folio.org

15

16 of 26

Considerations

  • Ranking values
    • PO rank
      • keep PO rank as is, because the numbers allow to put features in numerical order
    • SIG and institutional ranking
      • will be using 5 ranking levels (similar to R1 to R5)

| www.folio.org

16

17 of 26

Considerations

  • Surveys
    • frequency: once a year
    • institutions can list most important gaps as free text
      • e.g. 10 most important functions
    • will be related to JIRA tickets by PO or SIG convener

| www.folio.org

17

18 of 26

Considerations

  • Track features | dashboards
    • Add label or tag in JIRA
    • for individual institutions to keep track of individual important features
    • for broad functional areas
      • e.g. MM, RA, ACQ, ERM, Cross-app
      • for easy overview over all SIG JIRAs

| www.folio.org

18

19 of 26

Considerations

  • Communication
    • Extensive
    • For request and results
    • In all Committees,
      • PC SIG convener updates
      • SIGs
      • Slack channels
      • Mailing lists

| www.folio.org

19

20 of 26

Possible ranking tools

  • https://airtable.com/ 
    • Could be used for broader voting processes
  • https://seatable.io/en/
    • Cheaper than airtable
  • https://uservoice.com/
    • Could be used for broader voting processes
  • Miro
    • Suitable for prioritization discussions within a group
    • Miro offers an Education account (with full features) that is free to educational institutions. There can be unlimited users interacting with the boards.
  • Evaluation of tools

| www.folio.org

20

21 of 26

Next steps

  • Gather feedback ✅
    • please comment on these slides
  • Joint PO and SIG convener meeting
    • recording will be made available
    • offer to visit any other interested group
    • gather feedback via a survey
  • Presentation of possible tools
    • separate meeting;pros and cons (incl. costs/budget)
    • Evaluation of tools
  • Wolfcon session

| www.folio.org

21

22 of 26

What is your feedback?

| www.folio.org

22

23 of 26

Chat

15:44:04 Von Peter Murray an Alle:

App Ideas page: https://wiki.folio.org/display/PLATFORM/FOLIO+App+Ideas

15:50:07 Von Julie Bickle an Alle:

Or they get added in Jira, but not as a "feature" (which has a clear Definition and structure for dev work), but as [Name tbd] "new request" ?

15:51:07 Von Peter Murray an Alle:

It was the "APPIDEAS" project in Jira.

15:51:14 Von Kirstin Kemner-Heek an Alle:

That work can be done by the SIG's

15:51:39 Von Kirstin Kemner-Heek an Alle:

From the SIG's to PO's after first review

15:55:11 Von Charlotte Whitt an Alle:

Often new requirements is driven by libraries, who have this requirement as a road blocker for them in order to adopt FOLIO

15:56:06 Von Owen Stephens an Alle:

I like that Sharon

16:04:44 Von Owen Stephens an Alle:

I agree with that Ian - and there's definitely potential for frustration from the SIGs to see things they have prioritised not getting developed (and I think some SIGs are already in this situation)

16:06:47 Von Kirstin Kemner-Heek an Alle:

+1 Martina

| www.folio.org

23

24 of 26

Chat

16:08:02 Von Kirstin Kemner-Heek an Alle:

That prioritization is the base to decide on development and open up the opportunities for hopefully future developer Teams to get orientation, what is needed most.

16:09:41 Von Charlotte Whitt an Alle:

And also we need to plan work on features, which span over several apps; so the work being rolled out is useful for the implementing libraries, and not just half-baked functionality

16:12:12 Von Jana Freytag | VZG an Alle:

We have discussed, that NFR and basic industry standards should not be part of the ranking but rather something that will go into development automatically as well.

16:12:27 Von Kirstin Kemner-Heek an Alle:

+1 Owen. Prioritizing gaps / needs/ requirements is just one half of a process to get things done. But it is the necessary start.

16:12:51 Von Harry an Alle:

Owen, I think you have that exactly right. While the priorities help inform interest, it in no way decides what gets worked on next.

16:13:33 Von Gang Zhou (SHL) an Alle:

+1 Owen

16:13:52 Von Harry an Alle:

It has influence. But not a decision maker.

16:15:05 Von Thomas Trutt an Alle:

Sorry, i got tied up.

16:15:46 Von Kirstin Kemner-Heek an Alle:

But development Teams / Stakeholder can base their decisions what they will develop on that Information. That is an indicator what customers want. And other Teams can pick up, what is left over …. it Would be an opportunity to have a transparent process who picks up what - on each Stakeholders own decision. That is not questioned.

| www.folio.org

24

25 of 26

Chat

16:17:35 Von Maura Byrne an Alle:

Should features fit into epics?

16:17:47 Von Gang Zhou (SHL) an Alle:

+1 Kristin

16:19:00 Von Ian Walls an Alle:

I think all this assumes that we have a single pool or resources that are working against a single queue of features/fixes. I don't think that's realistic in an open source setting

16:19:15 Von Ian Walls an Alle:

different groups are going to have different priorities, and they're free to apply their own resources, and those of like-minded institutions, towards those priorities

16:19:16 Von Harry an Alle:

+1 Ian

16:19:35 Von Gang Zhou (SHL) an Alle:

+1 Ian

16:19:42 Von Thomas Trutt an Alle:

That is defiantly true.

16:19:58 Von Marc Johnson an Alle:

Ian, I think that is a really important point. Which becomes even more important when a feature crosses app boundaries and potentially involves multiple teams

16:20:05 Von Ian Walls an Alle:

as a vendor, should I get to the point of being able to contribute code, I'm going to do what my partner libraries need done. I want to be sure that's given back, but that's where my directive is going to come from

| www.folio.org

25

26 of 26

Chat

16:20:38 Von Marc Johnson an Alle:

I would also note that development teams don’t tend to make autonomous decisions about what they build. Those decisions tend to come from outside

16:20:48 Von Kristin Martin (University of Chicago; she/her) an Alle:

I hear from POs that they want these ranking from both the SIG and the institutions. I'm not hearing that they feel bound, but they want to get input into direction.

16:22:18 Von Kirstin Kemner-Heek an Alle:

Yes, from the Stakeholder who hires then, for example. That happens in our FOLIO Environment. Other Options are possible. As well as that team decide to do something completely different. All this doesn't exclude any other decisions. But helps to keep an overview.

16:23:35 Von Marie Widigson, Chalmers an Alle:

No additions, Martina, it was a very good summary!

| www.folio.org

26