ABCDEFGHIJKLMNOPQRSTUVWXYZAAABACADAEAFAGAHAIAJAKALAMANAOAPAQARASATAUAVAWAXAY
1
More ressources: Questions for Wednesday 07.12.22:
- How granular is our FAIR evaluation ?
- Should we use conventions defined by us (see convention table below) to have non-ambivalent evaluation criteria for FAIR in our evaluation, i.e. like in F-UJI automated evaluation ?
-
-
2
https://en.wikipedia.org/wiki/List_of_academic_databases_and_search_enginesSee mini-evaluation belowTO DO:
3
https://en.wikipedia.org/wiki/List_of_digital_library_projectsFill out detailed FAIR sub-table
4
https://en.wikipedia.org/wiki/List_of_preprint_repositoriesRemember: If you can not find information about some details, then send an email to ask (answers from emails will be marked in the table with an asteriks)
5
6
RESPONSIBLEEloiEloiDanielDanielJohannesJohannesDanielLarissaTimDanielLarissaEloiTimJohannesLarissaTimDanielLarissaTim
7
PORTAL NAME4TU Research DataarXivArchive of Formal ProofsAustralian Research Data CommonsFigshareHALHarvard DataverseMathrepoModelicaRepoNetwork RepositoryOEISOpenScience FrameworkOpenScience Library (CodeOcean)Science Data Bank (ScienceDB)SuiteSparse Matrix CollectionThe House of GraphsWikidataZenodo Directory,Biomodels
8
LINKdata.4tu.nlarxiv.orghttps://www.isa-afp.org/researchdata.edu.au/figshare.com/hal.archives-ouvertes.frdataverse.harvard.edumathrepo.mis.mpg.de/https://modelica.org/libraries.htmlnetworkrepository.com/oeis.orgosf.iocodeocean.com/explore?query=Mathematicswww.scidb.cn/ensparse.tamu.edu/https://houseofgraphs.org/wikidata.orgzenodo.orghttps://www.ebi.ac.uk/biomodels
9
CommentsaggregatorFigshare free of charge, but users can buy more storageno reply to email
10
InfrastructurePlatformFigsharearXIv codebase partially opensource (https://github.com/arxiv)Hugoproprietary (not open source)MySQL, SOLR, CINES (Github in future, asked with Mail)DataverseGithubProprietary, partly based on GitHubproprietaryMediaWikiOSF (Open Source https://github.com/CenterForOpenScience/osf.io)ProprietarySpring/Lucene/TripleDB(GraphDB)ProprietaryProprietaryMediaWiki/WikibaseCERN Data Centre/Invenio
11
CostFreeYYYYYYYYYYYYYYYYYY
12
Free to access, but contribution needed for depositNNNNNNNNNNNNNNNN
13
Size> 8400 datasets> 2M700208188 from 105 research organizations6.131.816~3.016.908150.247> 5630 libraries with 100s of models~7000 networks> 350,0007M public files, >75k preprints>1005.777.129> 28931.4B statements about 100M items using 10k propertiesabout 2.5 billion
14
CommentsAll data is stored in object storage at two locations in Delft and in one location in Leiden (see: https://data.4tu.nl/info/en/about-4turesearchdata/frequently-asked-questions)yesText files in pdf format or image files are sent to CINES for long-term archiving.
Multple Datacenters for redundancy(See Mail Answer)
Files uploaded to OSF Storage are stored in various storage locations, Configurable per User (https://help.osf.io/article/203-faqs#Backup-Preservation-Policy). Twice daily backups.yes (publication chengzhan: storage has backup)
redundancy: Y, answered in Mail
15
PreservationRedundancyNoneNNNNNNNNNYYNNNYYNN
16
Multiple redundant copiesYYYYYYYNYNNYY (through CLOCKSS)YNNYY
17
Geographically distributed redundant copiesYYYYYYYNY (through Git)NNYY (through CLOCKSS)YNNYY
18
CommentsDOI, handles and others (see https://ardc.edu.au/services/ardc-identifier-services/ for details); No backups: https://tutorials.rc.nectar.org.au/object-storage/01-overviewyesDOI, internal id (IdHal)Nointernal IDDOI, CSTRNoneNo
19
Persistent Ids.NoneNNYNNNNYYYNNNNNYYN
20
DOIYYNYYYYNNNNY (optional)YYNNYY
21
other persistent ID (Which?)NY (arXiv ID)NNNIdHAL (internal)NNNNY (internal id)Y (GUID: Globally Unique Identifier)CSTRY (internal id)NYN
22
Comments-yes public metadata not deleted admins can unpublishyes? see: could i remove ....Yes, archival storage (Google cloud)yes (see 8)Wikidata IDs and thousands of external ones
23
Persistent data depositLong term data preservationYYNYYYYY (Github)Y (through GIT)NNYY (through CLOCKS)YNNYY
24
CommentsRegistration + Review process + Standard measuresAuthentication, Checksums,AuthenticationThe COS uses commercially reasonable, industry-standard measures to protect the security of the COS's Websites and its Services (see https://help.osf.io/article/226-faqs-security, https://github.com/CenterForOpenScience/cos.io/blob/master/TERMS_OF_USE.md, https://help.osf.io/article/391-security-and-privacy)Authentication for web-version, API has no auth (only retrieval possible)not systematic
25
Security / PrivacySecurityAuthentication mechanism availableYYYYYYYYNYYYYYNYYY
26
CommentsData collection policy (https://data.4tu.nl/info/fileadmin/user_upload/Documenten/Data_collection_policy_2020.pdf)Privacy measures (https://arxiv.org/help/email-protection)https://ardc.edu.au/resource/sensitive-data/Private dataNo private data (has ftp server though)Privacy policy (https://github.com/CenterForOpenScience/cos.io/blob/master/PRIVACY_POLICY.md)Distinction between public and private dataNo Private data, Embargo possible
27
PrivacyDistinction between public and private dataYNNNYNYNNNNYYNNNNY
28
CommentsNot for every author, but authors must register and can add ORCID and arXiv author identifier (https://arxiv.org/help/author_identifiers) email address or ORCIDemail, ORCID, affiliations listingno, but unique user namesyes, ORCID and Global Unique Idenifier (GUID) for each object in OSF.Name, Affiliation, Email, ORCIDno
29
ArchivingAuthor IDNoneNYYYNNYYYNNNNNYYYN
30
ORCIDYY (optional)Y (optional)Y (optional)Y (optional)Y (optional)Y (optional)NNNNY (optional)NY (optional)NNYY
31
SCOPUSNNNNNNNNNNNNNNNNYN
32
Other (Which?)NY (optional arXiv author identifier)NNemail as idemail as idNNY (unique user name)Y (unique user name)NY /Name/Affiliation)email id NNYN
33
Commentsinternal identifieryes, DOI and internal, BIBCODEyes, DOI and Global Unique Idenifier (GUID) for each object in OSF.DOI, CSTR, internal in urlhundreds
34
Publication IDNoneNNYNNNYNYNYNNNYYYN
35
DOIYYNYYYYYNNNY (optional)YYNNYY
36
Other (Which?)NY (arXiv ID)NNNgenerates IdHAL (internal), can be added: PMID, arXivID YY (MIS-Preprint, ARXIV)NNNY (GUID: Globally Unique Identifier)NCSTRNNYN
37
Commentsyesinternal identifieryesyesinternal identifierthere is timestamping but unclear whether it is saved every time and if it is for sequence or only for e.g. commentsyesyes, by version uploadmany
38
Time stampingNoneNNNNNNYNNYNNNYY (maybe in meta-data)YN
39
Timestamp upon uploadYYNYYYYYYNYYYNNYY
40
Timestamp for every versionYYYYYYYNYNYNYNNYY
41
Comments4TU.ResearchData strongly encourages the use of standard, exchangeable or open file formats to prevent loss of access to files due to file format obsolescence (see https://data.4tu.nl/info/fileadmin/user_upload/Documenten/Preferred_File_Formats_2019.pdf for preferred file formats)(La)TeX, PDF, HTML for text / PostScript, JPEG, GIF, PNG or PDF for figures (https://arxiv.org/help/submit)Timestamp upon AFP releaseAll can be uploaded, categoriesAll can be uploaded, Publications, Academic work, Research data, DocumentsAll can be uploaded (focus on data, not documents), preferred file formats
42
SubmissionData typesAny file-type allowedNNNYYYYNNNYNYYNNY
43
File-types restricted (to which?)YYYNNNNY (.rst)YY (wiki-like edits)NY (compute capsules)NNYYN
44
Commentsthy 20GB for free accFrom Mail: They have an FTP Server which be used for storing bigger files no restrictionsJSON, RDF
45
Data sizeNo restrictionsNNYNNNNNYNNNYYNN
46
Restricted to max. size (Which?)Y (5 GB/year free of charge)Y (10 MB each individual file)NY (20GB)Y (Filesize is limited to 200mb)2.5GB per file, 1TB per dataset, but not strictY (for larger than 1GB, contact admins)YNY (50 GB for public projects / 5 GB for private projects)Y (20GB)NNNNY (50GB)
47
Commentsyesyes (https://arxiv.github.io/arxiv-arxitecture/subsystems/announcement.html?highlight=redundancy#id22)yessome metadata entries are necessary for publishingsome metadata entries are necessary for publishingyesyesno metadate required but desiredcurrently about 5MB per item: https://www.wikidata.org/wiki/Special:LongPages
48
MetadataNo metadata necessaryNNNNNNNNNNY (at least didn't find anything about this)NNNYYYN
49
Controlled LanguageNNNY~Yes, metadata description can contain any wording, ANZ FoR codes for categoriescategories are possibly a controlled vocabularyNot requiredYNYNNY~ for categoriesNNYN
50
Readme fileYNNNNot required, description field required thoughNot requiredNot requiredNYNNYNnot necessaryNNNnot necessary
51
Commentsyes (see metadata review process: https://data.4tu.nl/info/fileadmin/user_upload/Metadata_review_guidelines_June_2021.pdf)yes (moderation process: https://arxiv.org/help/moderation)basic metadata, config filesyes, reviewing possible~no peer review option (double check), verification of uploads metadata thoughnoSubmissions are reviewed and approved for metadata and complianceyesMost of Wikidata is metadata; language is controlled but continuously evolving and with no guarantees for consistency
52
Review/Data QualitySubmissions are reviewed and approved for metadata and complianceYYYNNYNYYYYNYYYYYY
53
CommentsOA, restricted access request possible, but metadata publicly availableOAautomated replication and human reviewopen*openopenOAOA (private projects possible)open, most free some embargoedopen
54
Access / SharingOnline AccessData available for free and open downloadYYYYYYYYYYYYYYYYYmixed
55
User registration neededNNNNNNNNNNNNNNNNNmixed
56
CommentsYesyes for uploadingyes for uploadingyesyes (download verifiyed by mail answer)yes for uploadingyes for uploadingyes, but undocumentedYesyesyesYes
57
APINo API availableYYYYNNYYNYNNYNNYNN
58
API for search availableNYNNYYYNYNYYNYYNYY
59
API for submission availableNNNNYYYNN (maybe through GIT?)NNYN~YNNYY
60
API for download availableNNNNYYYNN (maybe through GIT?)NNYN~YYNYY
61
CommentsCC,MIT,Apache,GPL, Copyrightcopyright or CC
62
LicenseCreative Commons (any)YYNY (optional)YYY (optional)YNY (CC BY-SA 3.0)YYYYNNY
63
ODbLNNNY (optional)NNY (optional)YNNNNYNNNY
64
Other license (open)YYYY (optional)YNC Public Domain Mark, ETALAB Y (optional)Y (free choice)Y (Modelica License 2)NNYMIT, and BSD-3-Clause for softwareNNNY
65
Other license (restricted)Y (for personal/confidential info)NNY (optional)Y (GPL)~CopyrightY (optional)Y (free choice)NNNNGNU GPL for softwareNNNY
66
CommentsCC0 as default. Other Creative Commons licenses are available from a predefined list. Diverse (data.4tu.nl/info/en/use/publish-cite/upload-your-data-in-our-data-repository/licencing)Diverse (CC BY, CC BY-SA, CC BY-NC-SA, CC BY-NC-ND, arxiv.org perpetual), https://arxiv.org/help/licenseBSD-style or GNU LGPLDiversePolicy MandatePolicy Mandate , policy texts planned diversediverseCreative Commons Attribution-ShareAlikeCreative Commons Attribution Non-Commercial 3.0 licenseDiversevariousdiverse, mostly CC
mail q succession plan: A:ScienceDB keeps DRAFT datasets for six months and PUBLIC datasets permanently.
CC-BY 4.0CC0, which technically is not a license
67
PolicyMandatePolicy text availableYYNYYYYNNNNYYYes, authority not completely clearNNNunclear
68
SecurityPolicy text availableNNNYYNYNNNNYN~Yes (in browser)NNNY
69
Data OwnershipPolicy text availableYYNYyes, claim ownershipNYNNNYNYNoNNYY
70
PreservationPolicy text availableYNNYYNYNNNNYYinfo that preservation is not guaranteed see terminationNNNY
71
Succession planPolicy text availableY (partially included under preservation policy)NNN~Y (this is only info)No, from Mail: but major french infrastructur, in preparatiion NNNNNNNNNNNY
72
FAIR PrinciplesFindabilitySummarized (details in next table)Y (see https://docs.google.com/document/d/1JmLJXMv-1mMHpkQ082lM0p_7iM4z8k2sgoX73VecjLg/pub)YYmostly yesYYYF3 and F4YYAll except F3YYYF1 and F2YYY
73
AccessibilitySummarized (details in next table)YYYmostly yes but counter example https://researchdata.edu.au/two-rocks-moorings-2004-2005/444898/ (linked from https://ardc.edu.au/article/enabling-and-enhancing-the-discovery-and-reuse-of-data-with-metadata/ )Y~yesYYYYUnclear if A2YY~yesA1?YY
74
InteroperabilitySummarized (details in next table)YY*NvariesY~yes, not clear i2Ynot reallyYYyes, unclear if I2Y*Y~yes, not clear i2I3: yes, I1/2: unclearYYY
75
ReusabilitySummarized (details in next table)Y*Y*YvariesY~yesYR 1.2YYUnclear if R1.1Y*Y~yesR 1.1?YY
76
FAIR https://knowledge.figshare.com/publisher/fair-figsharemore FAIRness is planned
77
https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwj3xq2rzt_6AhUoPewKHYqYB94QFnoECAoQAQ&url=https%3A%2F%2Fs3-eu-west-1.amazonaws.com%2Fpfigshare-u-files%2F13848212%2FFigshareandFAIR.pdf&usg=AOvVaw2LkBRT6nGPZZpdNeJrpM7BNext: write mail, check FAIR
78
DETAILS ZU FAIR KRITERIENFigshare say its fair in publication:
Figshare and the
FAIR data principles 2018
79
Maybe helpful for understanding the details:Excel table here may be useful (https://data.4tu.nl/articles/dataset/Evaluation_of_data_repositories_based_on_the_FAIR_Principles_for_IDCC_2017_practice_paper/12694157)
80
http://www.snf.ch/SiteCollectionDocuments/FAIR_principles_translation_SNSF_logo.pdf
81
https://www.go-fair.org/fair-principles/aggregatorno reply to email
82
https://www.f-uji.net/?action=testautomated evaluation of fair criteria, it is important to use DOI url if available otherwise automated checks dont find the persistent id, example
83
https://www.f-uji.net/index.php?action=methodsDownload metadata, evaluate the file manually with this guideline which specifies metadata / Minimal dataset upload to test the forms and also minmal criteria fullfilled
84
this publication has a statement on the measurability of each criterion, it could be cited in case something is not measurable
85
FAIR PrinciplesFindabilitySummary
86
F1. (Meta)data are assigned a globally unique and persistent identifier YYNYYYNYYYYYY
87
F2. Data are described with rich metadata (defined by R1 below) YYYYYYNYYYYYY
88
F3. Metadata clearly and explicitly include the identifier of the data they describe YYNYYYYNYYNYY
89
F4. (Meta)data are registered or indexed in a searchable resource YYYYYYYYYYNYY
90
AccessibilitySummaryA.1.2: there are no uploads possible for API, so its not necessary to auth
91
A1. (Meta)data are retrievable by their identifier using a standardised communications protocol YYYYYYYYYYYYY
92
A1.1 The protocol is open, free, and universally implementable YYYYYYYYYYNYY
93
A1.2 The protocol allows for an authentication and authorisation procedure, where necessary YYYYYYYYYYNYY
94
A2. Metadata are accessible, even when the data are no longer available YYNYY YYNNA.1.2:ScienceDB supports the generation of datasets in draft status through an API that is open to community members and the paper submission and review systems.
N(Y) metadata of deleted entities remain accessible via dumps from before their deletion. In rare cases, the creation and deletion (only of items) might occur so quickly that the item is not part of any of the daily or weekly dumps. such cases are confined to spam/ vandalism entries, so nothing to be preserved in the sense of the FAIR Principles.Y
95
InteroperabilitySummaryI2: categories are a controlled vocabulary
I3: yes, see oai-pmh example
l2: domains are controlled vocabulary (other fields not)
l3: authors seem to be interlinked, not the other fields (ask how to obtain detailed metadata to verify)
l2: categories are a simple controlled vocabulary
l3: rather no, only license contains a ref
96
I1. (Meta)data use a formal, accessible, shared, and broadly applicable language for knowledge representation. YY(Y) "shared" only within an Isabelle contextYYYNYYYNYY
97
I2. (Meta)data use vocabularies that follow FAIR principles YYNYY(Y) The vocabulary for the files in Dataverse is controlled, but there is no control with respect to the vocabularies used within those files.NNNYNYY
98
I3. (Meta)data include qualified references to other (meta)data YNNY~Y (define I3 tbd)YNYNNYYY
99
ReusabilitySummaryR1.3: yes, datacite metadata schema and custom extendability of schemasR1.1: such a license can be selected (not mandatory to select it though)
R.1.3.: from Mail with OAI-PMH metadata may be exported with a Datacite format
R.1.3: https://schema.org/Dataset is the schema
100
R1. (Meta)data are richly described with a plurality of accurate and relevant attributes YYNYY(Y) There are cases where "rich" and "plurality" do not applyNYYYNYY