– DRAFT –
Decentralized Identifier WG Special Topic Call

<Wip> rrsaagent, draft minutes

<Wip> Agenda: DID Rubric and Traits Discussion

present`

Wip: today we're talking about the DID rubric and trait stuff
… mainly looking for how to move these forward
… we did pass a resolution which states:

<Wip> https://www.w3.org/2025/02/20-did-minutes.html#0130

Wip: work with traits and rubric editors to bring them into the W3C
… there are questions around the registration process also
… JoeAndrieu maybe you can start?

JoeAndrieu: the rubric itself can handle booleans like the traits work started
… there are good questions in here
… we do need to weigh who's doing the analysis
… and I'd love to get more traits folks involved
… to see how we can combine these things
… Wip and I have also been exploring turning this into JSON
… right now you have to do some rather sophisticated HTML
… and know how the numbering works
… and I think that's slowing submissions
… so I'd like to get that JSON format done
… and then work on incorporating traits

jceb: I'm happy to work on that
… two questions
… do we keep the specification up at the DIF and then import traits into the rubric?
… that seems like how we'd start
… and then maybe when it's all moved over, we shut down the one at the DIF?
… and the other question is how/when we incorporate traits into the extension registration process
… should we do the rubric first?
… or do both in parallel?

Wip: I did want to mention a previous resolution around making the rubric a W3C registry
… so that would mean it exists beyond the lifecycle of this WG
… with some clear policies around how it's maintained
… the rubric as a registry gives us some means to do that long term
… we may have to think through the timing especially around this JSON structure

JoeAndrieu: importing would be the easy lift
… the rubric does let one cite a source
… so we could continue to reference the DIF document
… so when we incorporate them we can point to the traits work
… and then memorialize it when it's ready
… that seems better than "shutting it down"
… once these are in JSON we could also use these for more dynamic uses like filtering, etc.

ottomorac: this pathway makes sense. There are some things for me and jceb to iron out at DIF
… so is the vision then for both of these to be extensions on the registry?
… mostly trying to know how to guide DID method authors

jceb: these are different questions, so maybe replies to ottomorac first

JoeAndrieu: we're going to have a registry for the rubric
… and we're going to incorporate traits into the rubric
… but the rubric itself is about explaining the questions and not about being a storage location for the answers
… I wish we didn't have to put any methods in there
… we'll incorporate some things in the rubric
… the extensions registry is where people could provide their own self-assertions
… I have some doubts about that, but it still may be the right place to start

jceb: regarding the name DID traits. Will we keep it somehow? or will this just become the "big rubric"?
… the other question I had is around how the rubric will become a registry
… could you share more

Wip: I'll try and answer the registry bit first
… the W3C registries are a new thing at the W3C
… essentially a group defines what goes into the lists of things and how they get in there
… then you also appoint a group which maintains those rules long term--beyond the lifecycle of a single WG

<Zakim> Wip, you wanted to mention you could do a rubric evaluation with just the "traits" criteria

Wip: we did want to explore that for both the rubric and DID methods
… we have talked about the naming of the rubric
… you can think of it as a list of criteria
… currently, there are not answers that are true/false
… but when we incorporate the traits work, then we'd have Boolean answers available
… there's no requirement in the rubric that you have to fill the whole thing out
… you can focus on what applies to your usecase

<Zakim> JoeAndrieu, you wanted to about keeping did traits or absorbing them into rubric

<ottomorac> Like minimum criteria?

Wip: but we may still want to have criteria against which each thing is measured against--so a limited set of required questions

JoeAndrieu: that's an interesting idea. It may be controversial because it limits who can get onto the list.
… when we started this none of the W3C registry stuff existed, so we're still very much in a learning curve
… this group can produce a bunch of notes, etc., but their time boxed to the WG
… but if we put these things in an registry, they can live longer than the WG

<Wip> +1

JoeAndrieu: and we can state who will maintain it beyond the WG
… which could include the Credentials Community Group
… One question I still have is how we add the traits. Maybe we still call them traits and have a section just for those Boolean questions
… but it may be better to group them into the categories we already have--security, privacy, governance, etc.

Wip: thanks JoeAndrieu
… I'd love to hear from others on the call

ottomorac: I think the minimum number of traits you'd have to pick is an interesting idea...I'd like to hear more

Wip: I think what I said was "recommend"--I agree with JoeAndrieu that we don't want to block people from joining the list
… that was opposed earlier

<Wip> w3c/did-extensions#619

Wip: so this would be more about recommendation that folks should aim at a specific set at least

JoeAndrieu: I think that's right
… there are challenges to having a registry at all
… I would rather not have one
… it cuts cross grain to a decentralized technology
… So one thought is we may not even need to recommend them explicitly.

<ottomorac> sounds good

JoeAndrieu: we could simply use them for filtering, and that promotion of them in the UI would naturally incentivize them being filled out.

Wip: I'll try to summarize
… it feels like we do still want to incorporate traits into the rubric
… but that there may be more work around what that integration looks like
… but that's likely up to the editors

<Zakim> JoeAndrieu, you wanted to mention schema

Wip: so maybe we can focus next on what the next steps are outside of this call

JoeAndrieu: the way we've been thinking about the JSON Schema has been focused on adding to the rubric
… but if some version of this will go into the extensions registry, then we may want to rethink what we have now
… so that it's easier for both uses

<Wip> https://www.w3.org/2025/02/20-did-minutes.html#0eda

Wip: this work on building a JSON-based foundation for the rubric's rendering could use some help also
… we need help around figuring out the JSON and then how we're going to populate the HTML
… so I think we do need to talk through what next steps are, but maybe that happens elsewhere

JoeAndrieu: I think it's a little abstract now

jceb: adding the extensions registry to the considerations is really where I'd like to focus
… and you have way more insights into that work, and it'd be great to learn more
… I'm happy to help where I can
… my main takeaway for today is that we continue the traits work at DIF
… push toward 1.0 and get approval for it there
… and in the meantime, start working on incorporation into the rubric

Wip: that sounds good to me

ottomorac: my thinking is the same as jceb's
… I need to help close out the open issues, so we can accelerate getting to the 1.0
… jceb , how long do you think that may take?

jceb: maybe middle of June?
… then we can align with Kim around the next steering committee meeting

<Zakim> Wip, you wanted to ask about timeline

Wip: great. that timeline sounds good

<Zakim> denkeni, you wanted to comment on DID method standardization, guidebook for DID methods

denkeni: I'm happy to see the rubric and the traits combine
… and it would be a great starting point and more practical as a starting point than DID core
… that's a more ambitious goal for the rubric
… seeing the DID methods scored in the rubric will ultimately lead to people making choices from that data
… so emphasizing some methods during the standardization process may be helpful

Wip: my understanding is that DID method standardization selection has been happening at DIF in part using the traits work for guidance
… and I think it does make sense to use them for guidance
… where an organization is stating which traits they are selecting methods on up-front even before standardization
… I agree with denkeni that this is really about helping people do evaluations

ottomorac: just to expand and agree with you. Exactly what denkeni said.
… opinions around compliance and knowing how organizations are making their decisions
… this set of criteria can help them express their opinions
… the market is telling us there's too many methods, so we need a way to find the ones we want

JoeAndrieu: a question for denkeni , were you considering expanding that rubric to analyze things besides methods?

denkeni: yes, mostly around correlating this work with DID method standardizations

JoeAndrieu: if it's not a DID method that we're evaluating, what else would we analyze?

denkeni: when an organization is working on a standardization, they will set their criteria
… it will be interesting to see when different DID methods by different standardized processes
… but on the other hand it's interesting to see the W3C and DIF work together on rubric and traits

JoeAndrieu: great. thank you. It's primarily been a conversation around making it possible for each person and organization to choose their questions and traits to express what they want
… the US federal government is working on expressing their criteria as is the EU, and having something like the rubric and traits would make expressing and evaluating things easier

markus_sabadello: quickly about the DID method evaluation, we do have a working group in DIF around standardization of a few methods
… but DID methods can be standardized anywhere
… on the DIF side, we've been trying to come up with some criteria to determine what criteria we should use to choose what to standardize
… we've recently been talking about DIF endorsed and DIF recommended
… but that we would not say certain traits or rubric questions were good or bad--just the stability/value of the DID method regardless of whether it's "on the Web" or uses a blockchain or whatever

<Zakim> Wip, you wanted to suggest different orgs can suggest additional criteria

markus_sabadello: but we do care that these methods properly fill out the rubric and traits, so understanding what they support/do would help everyone

Wip: organizations should also feel empowered to build criteria they value, and the rubric could be the place where they do that
… I think many things you expressed markus_sabadello could go into the rubric

KevinDean: from past experience at GS1, there will be a push to have multiple registries
… some governments will not be comfortable with a registry that is under the control of a single country
… they may want it to be hosted only in their environment--or one they trust
… we hit this wrt to DNS standardization

Wip: I think this is a future concern, but thanks for raising it
… we should continue to discuss--which criteria do we want to recommend...or do we even want to do that

Wip: any final thoughts?

ottomorac: maybe next steps?
… maybe in the next month we focus on our own tasks and then come back in June to discuss more steps?

Wip: that sounds good. We'll leave this on the calendar.
… enjoy your days!
… bye all