Transition from CG to WG

[acka47]

For some time now the transition to a W3C Working Group has been discussed, last time in the September 2024 meeting. In this ticket, we will gather relevant information and begin planning the transition process.

Some CG history and facts

For some time now the transition to a W3C Working Group has been discussed, last time in the September 2024 meeting. In this ticket, we will gather all relevant information and plan the transition process.

Some CG history and facts

• 2019-06-11: The Entity Reconcilation CG is launched
• 95 issues have been closed since then over all five repos of the reconciliation-api GitHub organization
• 2019-09-16: The first commit in the specs repo, 247 commits have been made to the master branch since
• There are 38 members in the reconciliation-api GitHub organization
• Seven people have contributed to the specs repo, six to the testbench repo and six to the census repo with some but not complete overlap between contributors
• Two versions of the specification have been published as CG reports
• Work on a draft WG charter has already been started in this repo (still early stage)
• Several implementations exist and are collected in the census repo and – for the testbench – on Wikidiata:
  • hosted services
  • clients
  • libraries
  • local services

The listed facts show that the specification is very mature and has a lot of traction, especially because the popular data cleaning tool OpenRefine can act as a reconciliation client.

CG participants and W3C membership

Currently, six CG members are employees at a W3C member organisation: Hochschulbibliothekszentrum des Landes NRW (2 participants); Ontotext; imec; J. Paul Getty Trust; Institut Mines-Télécom (1 participant each).

Some – including some of the most active – CG members are not working at a W3C member organization and would need to participate in the WG as invited experts.

Next steps

There are some informational materials about the transition process, e.g. https://www.w3.org/Guide/process/cg-transition.html. Furthermore, the whole transition process is currently under revision in the Community Council (see also the CG Council slides from TPAC 2024).

In the Entity Reconciliation CG, there is not much experience with this step in W3C processes so that we would be happy to get some support from W3C. There is the W3C Chair Buddy system, however it currently rather looks dormant (last mail to the list is from 2018). As the CG council is currently working on the transition process, we hope to get some support from there, thus, I am pinging @ianbjacobs and @dontcallmedom.

[ianbjacobs]

Hi @acka47, I'll schedule a chat with the CG chairs to sync up (which is a piece evaluating the transition landscape). Thanks for the ping!

[acka47]

@ianbjacobs met with @wetneb, @fsteeg and me on 2024-12-17. (Thanks for the quick help, Ian!) It was a very informative exchange about bringing the Entity Reconciliation Service API to the official standardization process within a (to be created) working group. Generally, Ian noted that the preliminary work seems substantial and fitting for standardization in a WG, if there is support in W3C membership.

After the meeting, @ianbjacobs has set us in contact with @pchampin who is a member of the W3C Strategy Team, with a focus on data interoperability. Pierre attested "if we can get enough traction among members, this would be a nice WG to create" and hinted at Wikimedia as a W3C member that is implementing the reconciliation API and might be supportive of the project.

In the process, it became clear to me that I am missing basic knowledge about the requirements for creating a working group. So I looked it up and found https://www.w3.org/Guide/process/charter.html#baseline-support where it reads:

Generally, the Team [this refers to W3C staff, A.P.] will expect to receive reviews for Charter proposals from at least 5% of the Membership. If the 5% threshold is not met, the Charter may still be approved, but additional scrutiny is warranted, and resource allocation may be limited. Additionally, the Team will continue to consider the number of declarations of intent to participate or implement the output of the work group.

Currently, there are 357 W3C members, which means, we will need at least 18 reviews when proposing a WG charter (357*0,05=17,85). So, we will have to look for W3C members that have an interest in standardizing the API and 1) raise their awareness of the API and the planned standardization and 2) ideally receive a review by their Advisory Committee representatives (short, AC reps).

@fsteeg and me went through the W3C member list to identify those members who are or might be interested in the standardization (we included all academic institutions plus orgs with community group members and known users of the API):

• ETH Zürich
• Fraunhofer Gesellschaft
• DFKI
• hbz
• Hitachi (@pbnjay could provide contact)
• imec
• Inria (@pchampin could provide contact)
• Inrupt
• Institut Mines-Télécom
• Keio University
• King's College London
• Library of Congress
• Nanjing University
• National Library of Sweden
• Ontotext
• Open Geospatial Consortium
• Shenzhen Institute of Information Technology
• Stanford University
• SURF bv
• The National Archives
• The New York Times
• The Open University
• Universidad Politécnica de Madrid
• Université Côte d'Azur
• University of Edinburgh
• University of Illinois
• University of Milano Bicocca
• University of Oxford
• University of Southampton
• Vrije Universiteit
• Wikimedia Foundation
• Zhejiang University
• Institut Mines-Télécom
• J. Paul Getty Trust
• Keio University
• King's College London
• Library of Congress
• Nanjing University
• National Library of Sweden
• Ontotext
• Open Geospatial Consortium
• Shenzhen Institute of Information Technology
• Stanford University
• SURF bv
• The National Archives
• The New York Times
• The Open University
• Universidad Politécnica de Madrid
• Université Côte d'Azur
• University of Edinburgh
• University of Illinois
• University of Milano Bicocca
• University of Oxford
• University of Southampton
• Vrije Universiteit
• Wikimedia Foundation
• Zhejiang University

[pbnjay]

It's been a while since I contributed much of anything, but I'm happy to contribute more if it's helpful to establishing the WG. I've changed jobs so I'm now at a subsidiary of Hitachi which is a member.

[VladimirAlexiev]

Ontotext will support the establishment of this WG.
One major piece of work will be a test suite. The Testbench is relevant, but the WG will also have to define conformance tests and implement them as Test Manifest rdf (Turtle) files.