[MINUTES] CCG Incubation 2026-02-05

Meeting Summary: CCG Incubation

*Date:* 2026-02-05

*Attendees:* Benjamin Young, Dave Lehn, Dave Longley, David C, Manu Sporny,
Parth Bhatt, Phillip Long
Topics Covered:

   - *Review of PR 40: Add verifiable recognition credentials section to
   data model.*
   - *Discussion on the recognizedBy property at the recognizedEntity
   level.*
   - *Education use cases for representing educational institutions and
   diplomas.*
   - *Consideration of adding a present action to recognizedAction.*
   - *Discussion on self-issued credentials and their implications.*

Key Points:

   - *PR 40 Updates:* Manu Sporny integrated updates to the data model,
   including definitions for general properties, recognizedEntity,
   recognizedIssuer, and recognizedAction. Examples were also updated.
   - *Removal of recognizedBy at recognizedEntity level:* The consensus was
   to remove the top-level recognizedBy property for recognizedEntity as it
   was seen as redundant with the issuer's role and could lead to confusion.
   This aligns with the idea that the issuer is inherently recognizing the
   entity.
   - *Education Use Cases:*
      - *K-12 Diploma Issuance:* A significant discussion centered on how
      to represent the scenario where a school issues a diploma but a
district is
      the official issuer. The proposed solution is to have the district as the
      formal issuer in the data model, with the high school's details embedded
      within the credential. This avoids complex multi-issuer scenarios.
      - *Program Specificity:* The data model can accommodate detailed
      lists of programs offered by institutions, potentially leading to long
      lists for highly specific offerings. The flexibility of using URLs to
      external lists or more generic terms was discussed.
   - *present Action:* The utility and necessity of adding a present action
   to recognizedAction were debated. While initially proposed to cover use
   cases like the directory example and self-issued claims, the group
   concluded that it might be implicitly covered by existing mechanisms or can
   be added later if a clear need arises. The focus shifted to ensuring the
   core use cases are met without adding unnecessary complexity.
   - *Self-Issued Credentials:* The group affirmed that self-issued
   credentials (where both issuer and subject are the same entity) are a core
   part of verifiable credential and SSI principles and are generally
   supported. The concern was raised that formalizing certain aspects might
   inadvertently restrict this capability.
   - *Governance and Validation:* A point was made about the importance of
   governance in the validation process. The specification should clarify that
   recognizers need to validate all fields for which they are claiming
   issuance.
   - *Meeting Time Adjustment:* The team acknowledged the need to find a
   more suitable meeting time that works for all attendees and will be
   discussed via email.

Text: https://meet.w3c-ccg.org/archives/w3c-ccg-ccg-incubation-2026-02-05.md

Video:
https://meet.w3c-ccg.org/archives/w3c-ccg-ccg-incubation-2026-02-05.mp4
*CCG Incubation - 2026/02/05 09:58 EST - Transcript* *Attendees*

Benjamin Young, Dave Lehn, Dave Longley, David C, Manu Sporny, Parth Bhatt,
Phillip Long
*Transcript*

Phillip Long: Yes.
00:05:00

Benjamin Young: Hey y'all. Sorry to be late. I notice we don't have David C
here though.

Benjamin Young: Hey, Mona. Go ahead.

Manu Sporny: Yeah, that is a problem…

Manu Sporny: because he's, one of the main people that's pushing for part
of the spec. I'd say we've got enough people here to let's just meet and
try to get some stuff done on the spec. I was able to get the updates from
last time integrated in. So, I'd like to review those. and hopefully Phil
can kind of represent the education use cases. I think we've got those
covered at this point. but we do need to try and just finalize the data
model so that we can get this PR in so that we've got some kind of base
shape to the specification that at least achieves two out of the three
major use cases we're trying to achieve. and then maybe we can talk a
little bit about David Chadwick's use case. I think we have enough of a
basis on what he's trying to accomplish.

Manu Sporny: There is a lot to talk about there. So maybe we can get to
some kind of cohesive design that we think is reasonable and…

Manu Sporny: then whenever he shows up on a future call we can pass it by
him. That's at least my suggestion on at least using the time we have today
productively.

Benjamin Young: Yeah. No,…

Benjamin Young: that sounds fine to me. did you say you made a PR or that
you were Yeah.

Manu Sporny: Yes. Yes.

Benjamin Young: Do you have a link to that handy? I do not.

Manu Sporny: Yeah. I can go ahead and screen share.

Benjamin Young: Yeah, that'd be great.

Manu Sporny: Let's see. Let me go ahead and bring this up. And then it is a
pull request.

Benjamin Young: No, not at Let's dive in.

Manu Sporny: It's this one. PR 40 add verifiable recognition credentials
section to data model.

Manu Sporny: I was also able to integrate Ted's changes as well. So, real
quick to dive into this, unless Benjamin, you wanted to cover something
else before we jump in. All right. the PR's, getting to be fairly big but
focused. Let me bring up the preview. so last time we covered this general
properties section this concept called a recognized entity. We also talked
about an recognized issuer and then recognized action and then stop short
of doing the EU use cases which have to do with list operators and
publishing information about trust lists and things of that nature.

Manu Sporny: So changes made since last time is in the general properties
section. We now have definitions for all the general properties like each
one of those definitions points to where the value space is defined the ID
property is defined in the verifiable credentials data model 20
specification in section 4.4. So we just say go there if you want to know
what the appropriate value for this is. and then we start off with just a
really quick update intro on what this value is. so there's that stuff.
last time I think Dave Longley you asked that we create a new type called
recognized issuer.

Manu Sporny: kind of put it in snuck it in here. The other alternative is
to create a new section on the left here where we talk about recognized
entity and we also talk about recognized issuer and this is the only
normative statement that pops in. Everything else is just a regular link
data object. so that and that's it for the general property section. So I
think we're done with the general property section at least applied all the
changes people asked for. And then in recognized entity, the only real
update was saying that hey, you can supply these properties, but you can
also add in mixin properties from general properties.
00:10:00

Manu Sporny: And you're not limited to those. You're just like it's linked
data. You can add anything you want to. But, we specifically call out that
you can add things from the general properties to, give more, information
for the recognized entity. go ahead Dave.

Dave Longley: When do you want feedback as you're doing this?

Manu Sporny: Yeah, let me summarize kind of just to bring everyone up to
what the changes are and then at the end we can go back. if that's All
right. missing things up, thumbs ups. so recognized entities, in here with
recognized by and recognized too. I know there's a recognizer conversation
we need to have, So, that's recognized entity. So, I think that's up to
date now. Recognized action. There was a request last time to add
recognized by recognizer. We need to have that bike shedding discussion.
But a recognized action now has a recognized by a recognizer.

Manu Sporny: there and then didn't change output validation. I know we
don't really like it but that didn't get changed. and then the examples
were updated and largely it was So the new thing this used to just be an
issuer in a URL. it's a issuer with a type of recognized issuer. And then
we reference that object in these recognized entities here. That's not new.
We were doing that before. But in the recognized actions example, now we
have a recognized action and we added to the recognized action.

Manu Sporny: The reason we did this last time was to make sure that if
recognized action just exists on its own, who recognized the action. and I
guess this is somewhat disconnected from the recognized entity at this
point now. So we might want to talk about those are the updates that I saw
from the minutes and notes. sorry if I missed anything. There was a lot of
conversation last time. Okay, so let's jump to feedback. go ahead, Dave.

Dave Longley: So, I was thinking that I have a couple thoughts about that
top level ra recognized by property. I don't know that it accomplishes
much. I was also thinking about use cases where you might have a list where
you might want to state multiple parties recognize some tity. and really I
think when you recognize some other entity, what you're doing is you're
recognizing them to do certain things. I think in the absence of
recognizing them to do certain things, then you're not really recognizing
them.

Dave Longley: I think in the directory case where you're just describing
information about the entity really the only recognition there is sort of
identifying I guess and I think that can all be done without saying someone
else declares someone else recognizes these properties because that's
really just r at that point the issuer of a credential that makes claims
about the legal name image URL description about a particular entity that's
going to be the issuer because that's where that trust comes from. So I
don't know that makes sense there. And if you imagine trying to make a list
where you would have multiple parties that recognize actions which might
even be different actions about the same entity. I think you're going to
get a lot of confusion in the model and that's even if you don't go and
merge it with some other list.
00:15:00

Dave Longley: So in the same list you might say let's take this web
university example as the first credential subject here. If you had two
actions and the first action was for issuing it was recognized by A and the
second action was recognized by B. that top level recognized by property if
you didn't say anything in the actions would be a problem. I understand
that you added it to the action. but I think we could probably remove it
from the top level. I don't know that it really serves any function.

Dave Longley: And if we want to have an action that would be put in here to
talk about who recognizes another party, I think there are two good action
names that make I'll offer one that I think makes the most sense because it
fits in with the other actions that we're trying to capture here. I think
we have issue verify and we have present. And if I think that you recognize
another party to present ties very tightly with these are the properties I
would expect of this entity and I recognize them to present something. I'm
not going to say what that something is but if they are presenting using
this identifier then I expect these other properties that the issuer is
claiming to be true. I think that's what is being said.

Dave Longley: So, minimally, if you didn't want to put any other actions in
there, I think you could put in a present action, and that gives you a
place to put the recognized by property there.

Manu Sporny: I didn't follow the last bit.

Manu Sporny: So we might need to come back to that.

Dave Longley: So just to summarize what's on the screen,…

Dave Longley: if you deleted what's highlighted where it says recognized by
and that was no longer a thing and action said present, I don't think you
would need any output validation for that. That would be a minimal example
of saying that the learning commission recognizes the example polytenic
university and they recognize them to present whatever credentials that
thing wants to present. We recognize that if you are presenting using the
ID did webun university.ample

Dave Longley: We recognize that that will be the example polytenic
university.

Manu Sporny: I'm trying to match it to a use case.

Manu Sporny: I kind of get what you're saying.

Dave Longley: The use case is an identifier comes in. I don't know who
that's for,…

Manu Sporny: Mhm.

Dave Longley: but this list tells me gives me these other properties about
who that is. And those other properties might be useful. there's a URL in
there. There's an image. That's the directory use case. Yeah, I don't think
you need recognize by and…

Manu Sporny: And you're saying for that use case you don't need recognized
by.

Dave Longley: I think it's a potential for significant confusion in any
other use case.

Manu Sporny: But you're saying in recognized action there's a use for
recognized by.

Dave Longley: And I'm saying you have to put it there because that's the
actual differentiator.

Manu Sporny: So the argument you're making is remove recognize by at this
level…

Manu Sporny: because who's recognizing them. It's the issuer. So it's
redundant, right? Mhm.

Dave Longley: Yeah. And…

Dave Longley: if you think about it as reading as an English sentence,
example polytenic university is recognized to issue by the learning
commission. And if you don't do it like that, you lose those pieces. And if
you put other data in this graph, I imag put a second get rid of recognized
in the action and…

Dave Longley: now put a second action in there that now you're recognized
by at the top level has two values, right? And which one goes with which
action? You don't know. Yes.

Manu Sporny: …

Manu Sporny: so the ar general argument is get rid of at the recognized
entity level. All right. Go ahead, Phil. Yep.

Phillip Long: Yeah. can you hear me?

Phillip Long: Okay. I'm trying to map this into two education examples. One
is the circumstance that's found in K12 actually where and the individual
school does the instruction and what have you and creates a diploma for an
individual, but it is the district in which that school is located that is
actually the official entity that the diploma from that school gains its
validity from.
00:20:00

Phillip Long: And I'm not sure how that gets worked into this. And the
second question I have is you'll noticed in the output value there is a
recognition of their ability to offer a bachelor's. institutions offer all
kinds of different flavors of formal programs. Bachelors being one of them.
And of course this I guess would handle any bachelor degree that they give
whe regardless of the discipline. But I'm wondering therefore whether there
is any other categories here or whether this ID list might end up getting
quite long for specific programs within them. So those are the two things
that I'm trying to map this representation to.

Phillip Long: I'm open to feedback.

Manu Sporny: Yeah, maybe we take the second second one's easier I think to
answer yes this list could get very very long if you have a thousand
variations of things that you offer the list I would expect and you want to
be very specific about it the list would become a thousand long,…

Manu Sporny: but that's unavoidable…

Phillip Long: you could in fact represent that thousand things rather than
bachelors saying programs or…

Manu Sporny: because you want to specify a thousand different things that
you offer and therefore you have to specify what those things are
somewhere, right?

Manu Sporny: Yeah,…

Phillip Long:

Phillip Long: something more generic. Okay. …

Manu Sporny: in the JSON schema for it could have a whole bunch of reg x's
to reduce it from a th00and down to 200, right? Or it could have a whole
bunch of literal values. you can always compress it down and make the JSON
schema really more complicated, so that's something that they could do. and
it could be something like degrees and if it's, bachelor's, masters,
whatever. And then very specific, yeah.

Phillip Long: certificates and all these other sort of variations that are
available. I'm just wondering if you could point it to something at the
institution and say whatever's there.

Manu Sporny: Yeah. Yeah. Yeah. I mean, that's what this URL is for, right?

Phillip Long: Yeah. Right.

Manu Sporny: So you could use the URL and…

Phillip Long: Okay. I just want to Yeah.

Manu Sporny: you could point it to the institution and the institution
could update those things from day to day if they don't do the digest
multibase thing right so there are ways to do that and…

Phillip Long: Yeah. Yeah.

Manu Sporny: make it dynamic and let them have full control over what those
allowable values are. so I think we support that use case today and we give
full power to the institution to figure out the best way for them to do
that. Okay.

Phillip Long: Yeah. I mean there are things Yeah.

Manu Sporny: So, that was just Go ahead.

Phillip Long: One last thing just to put it in the people's minds the
commission or that we're talking about here that is recognizing that
themselves needs to be clear that this institution can't put something in
there that they haven't vetted and that might imply…

Manu Sporny: Yep.

Phillip Long: therefore that the school can make all kinds of claims about
things and point to the commission and saying I'm recognizing by them for
this when in fact that that's not what they actually did their verification
on.

Manu Sporny: Yeah.

Phillip Long: Yes,…

Manu Sporny: I mean that's a governance thing, right? That's between the
institution being Yep.

Phillip Long: just we need to put that in comment in the explanatory text
somewhere.

Manu Sporny: Yep. Okay. All right. let's see. is that a new issue?

Manu Sporny: that's probably a new issue. explain that output validation
requires governance. specification should explain that recognizers need to
validate all fields that they are claiming that they are issuing claims for
such as
00:25:00

Manu Sporny: The is for example if constitution geez is providing a JSON
schema for validation. The recognizer needs to vet that schema to make sure
what it allows is the same as is

Manu Sporny: showing something like that. Okay, I'm tracking that.

Phillip Long: Anybody that right?

Manu Sporny: So that's 41. Okay. And then going back. That was your second
comment. Phil, did we do we So,…

Phillip Long: Right. Yeah.

Manu Sporny: I think we're good and tracking that now.

Phillip Long: We just have to get the language right so that one says,…

Manu Sporny: Yep.

Phillip Long: wait a minute, they can do something that I didn't bet.

Manu Sporny: Okay. So, let's go back to your first comment.

Manu Sporny: Unfortunately, it has gone from my mind right now because I
was thinking about your first comment.

Manu Sporny: Can you please re rehash that? Mhm.

Phillip Long: Yeah. The first comment is that there in the K12 environment
at least,…

Phillip Long: the high school, let's say, issues a diploma for the
graduates in that program. in the way in which that is treated by most
states the issuer of record this has always been a problem the issuer of
record is not the high school it's the district within which the high
school is organized in a sense the state passes the recognition legal
authority to the district and…

Phillip Long: then that is operationalized by the high school issues the
diploma in their name, but the district has to sign it before it's legal.
And so I'm just trying to make sure that you represent that in a way that
is clear here.

Manu Sporny: is yeah I guess let's dig into that.

Manu Sporny: In that case who ends up on the issuer it's not a dual
signature situation is it Phil or…

Phillip Long: That's a good question.

Manu Sporny: is it I mean,…

Phillip Long: That might be the simplest,…

Manu Sporny: so here's …

Phillip Long: frankly. Yeah.

Manu Sporny: I was thinking that would be the most complicated. the
simplest thing would be for the district to actually be the issuer. right
and…

Phillip Long: Or not. Yeah. Right.

Manu Sporny: in that case then the credential can say and so high school
and the credential itself can have that metadata in it but the actual
issuer of it is the district.

Manu Sporny: And so then you have a single issuer that district is
recognized to issue high school diplomas and…

Manu Sporny: then the recognizer is who would that be like the state
education commission or something like that.

Phillip Long: Why don't we try that now and…

Phillip Long: I can verify that with a couple of districts to see whether
or…

Phillip Long: not that will fly for them.

Manu Sporny: Yeah. If…

Manu Sporny: if that doesn't fly for them, the alternate is that the high
school is the one that issues. The high school is recognized by the
district and then the district has to be recognized by some higher
authority and that higher authority needs to be recognized by some higher
authority above there.

Manu Sporny: Yeah.

Phillip Long: Yep. That gets to be complicated.

Phillip Long: That last one.

Manu Sporny: I mean, it would be nice to avoid that. We can certainly do it
and support it. It's just getting that rolled out across the US would be
challenging, I think.

Phillip Long: Yep. Yeah. No. Yeah.

Manu Sporny: And it adds complexity where I don't think you really need
that flexibility and complexity. we can certainly support it, but I think
from an operational standpoint, then you have the job of educating every
single high school across the US on how to, use the system. And usually
they would probably just punt it to district it or state it to do it
anyway, right?

Phillip Long:

Phillip Long: Right. Yeah.

Manu Sporny: Okay,…

Phillip Long: Let me run this by a couple of folks. I can get a quick
answer to this because this has always been a point of contention. And it
is resolved somewhat similar to what you're describing now using the
comprehensive learner record version two that one has for this where
there's an outer credential that's issued by the district if you will and
the inner credentials are the details that the school is responsible for.
They did it as a compound credential with two different with an embedding
so to speak. right, right?
00:30:00

Manu Sporny: got it.

Dave Longley: I was just wondering…

Dave Longley: if there's a significant review process after the school, you
said the school issues or the school puts out a document that has to get
the signature from the district. Is there a significant review process or
is it more or less the school just delivers a packet of these things?

Phillip Long: No, there's a review process about the qualifica the high
schools procedures by which they make the final decision that this is a
valid diploma that the person has in fact passed the courses that are
essentially required for them to pass because the state imposes various
kinds of rules associated with what those courses

Phillip Long: need to cover and be. And so it's the district's
responsibility to make sure the school's curriculum at the ground level
touches them and that they've been assessed properly.

Phillip Long: So that's right.

Dave Longley: And is that assessment ahead of time and…

Dave Longley: then at the diploma, more or less they do that assessment at
some point,…

Phillip Long: They do it annually.

Dave Longley: And so I guess what I'm in the back of my mind I'm wondering
sort of mechanically if all of that's been done if these schools can be
given the capability to hit some service that's tailored specifically to
their school that would push out the diploma with the issuing signature
with the district having given delegated that to them.

Dave Longley: I'm wondering if that's a viable model.

Phillip Long: potentially. I don't know.

Phillip Long: All of this obviously is highly political in the sense of who
is actually responsible from the perspective of the parents view of the
legal situation there. So that's why it's been complicated to date. but
you're saying that you could have the issuer essentially being the district.

Dave Longley: Yeah, the district had set up a bunch of rules and…

Phillip Long: And how would you right…

Dave Longley: and guard rails.

Phillip Long: how would you designate then what is the recognition of the
school in that context on this data model.

Dave Longley: I'm whatever the you said this data model. I was speaking
more to I was thinking mechanically about how these diplomas would go out.
And so I was I really wasn't even thinking about the data model on the
diplomas.

Dave Longley: I was just thinking about whether or not an issuer the
district can set up a strict guard rails and then delegate permission to
the schools that they have already done their assessments on giving them
the capability to just push those diplomas out mechanically. What comes out
at the end of that process is a diploma that has as the issuer the district…

Dave Longley: if that's what they're looking for.

Phillip Long: Yeah, I've taken notes. I will check that and get back to us
next time.

Manu Sporny: Okay, thanks Phil.

Manu Sporny: The other thing I think to note there is that you asked where
does the high school show they can show up in the actual verifiable
credential in the diploma itself,…

Phillip Long: Right. Yeah.

Manu Sporny: right? So the high school diploma will very clearly state the
student graduated from this high school with but the signer would be the
district. Mhm.

Phillip Long: Right. The OBV3 has credential subjects and types and there
are 33 types within it. One of which is a diploma. And so the credential
subject in that case would be the individual who's getting this particular
diploma,

Phillip Long: but I'm not sure it says anything about the actual school
that's issuing it in that particular framework.

Manu Sporny: interesting. So it may presume it would be the high school
issuing it.

Phillip Long: Right. Right.

Manu Sporny: Mhm.

Phillip Long: Because It's intending to represent the individual has a
valid diploma that was issued.

Phillip Long: And the diploma itself is decorative for your wall.
00:35:00

Manu Sporny: Right. Right.

Manu Sporny: Okay. All right.

Phillip Long: All right.

Manu Sporny: Yeah.

Phillip Long:

Phillip Long: Thank you.

Manu Sporny: I mean, there's still even in that scenario,…

Manu Sporny: as Dave was saying, there are multiple ways to achieve that to
get either the high school as the issuer on there through delegated,
authority.

Phillip Long: We can find out…

Manu Sporny: but have the district run the infrastructure or you just put
the district as the issue on there. So I'm fairly confident we can support
all the variations.

Phillip Long: what Right.

Manu Sporny: It's just very important as to make sure that we've got one
story across for example the US versus a thousand different ones.

Phillip Long: And they have to see it in they have to look at it and…

Phillip Long: see themselves in it pretty clearly. that's more what I'm
concerned about. Very good.

Manu Sporny: Yeah, that's right.

Manu Sporny: Yeah. Yep. Okay. All right.

Phillip Long: Thank you.

Manu Sporny: No, thank you for that feedback. That was all very helpful and
useful. what changes do we need to make, Dave? I'm hearing you lobby for
removing recognized by on the recognized entity. I'm fine with that. I
don't have any objections.

Manu Sporny: I would imagine David Chadwick would be the one with the
strong feelings on that. I don't imagine Dmitri would have an issue with
that either. so I can make that change.

Phillip Long: Yeah, I don't think No,…

Dave Longley: I recommend adding a recognized action of present.

Manu Sporny: Go ahead, Phil. Okay.

Phillip Long:

Phillip Long: I don't think Dimmitri would. I said,

Manu Sporny: And then what else do we need to Okay, so let's see based on
change. wait. add second.

Manu Sporny: And what's the use case there again? You just recognize them
to present anything.

Dave Longley: And it seems like that there was a desire to be able to say
more than just the issuer it would trust or accept the claims made about a
particular entity. So, I'm trying to think of the directory use case.
you've got a bunch of issuing entities by ID and then you attach all these
properties and it's the issuer is making all those claims about those
properties whether they're true or not.

Dave Longley: and in previous discussions we talked about what if the
parties that are making this list we wanted to empower them to essentially
say that these other parties would also accept those claims. they would
accept my directory values is sort of how I was interpreting that. And
that's more or less saying if that's true, what that means is that if a
recognized entity shows up and proves that they are the controller of this
identifier, then these 10 other parties would agree with these other claims
made about that identifier.

Dave Longley: So identifier X says the name of this is Virginia Tech,
University. you could announce in your list any number of other parties
that would say,…

Dave Longley: Yeah, that is the identifier for Virginia Tech." But that's
all you're saying.

Manu Sporny: I don't understand…

Manu Sporny: how present gets us there.

Dave Longley: If Virginia Tech presents anything or anyone presents
anything from Virginia Tech with that identifier that and you have this
list,…

Manu Sporny: That's true. Or you should accept that as well.

Dave Longley: yeah, you would say, this is the identifier for Virginia
Tech." And that's kind of Dimmitri's use case.

Manu Sporny: But what happens when they're like, "But what?" Yeah, but what
happens when they lie?

Dave Longley: The point is you have…

Manu Sporny:

Manu Sporny: I know of this.

Dave Longley: if you have a list

Manu Sporny: You're saying I know of this entity and…

Manu Sporny: they can present anything they want to I recognize they exist
and…

Dave Longley: Yeah, you the …

Dave Longley: you're not accepting…

Manu Sporny: so Yeah,…

Dave Longley: what they present. You're accepting the claims made about
them. They can go and present things and if this party presents things as
this identifier, then you believe what you is in your trust model, in your
acceptance or recognition model is that the party that is presenting is
this party. What they're presenting is a separate question.
00:40:00

Manu Sporny: I'm wondering about the nuance on that being potentially lost.
I'm thinking anyone can present anything at any point. They don't need to
be recognized to do so.

Dave Longley: Yeah, but you don't know who they are. That's where this is
coming from.

Manu Sporny: Yeah. I'm wondering what the pres action present actually gets
us here. Meaning do we even need that?

Dave Longley: This would allow you to do things like self-issued claims
from Virginia Tech. if someone shows up in an online transaction happens
and…

Dave Longley: a presentation comes in with IDX I don't know who IDX is. I
go to my list and this presentation's it's making all these claims about
itself, I go The list says X is Virginia Tech. You could easily have a
model that says, Virginia Tech wants to say whatever it wants to about
itself. I'm fine with that." Yeah.

Manu Sporny: Mhm.

Manu Sporny: And I recognize that they do exist, so they can do that. okay.
Go ahead, Phil.

Phillip Long: Yeah. Yeah.

Phillip Long: This gets into another area that's commonly a problem.
universities often have a very strict process of vetting what the
university's provost signs off as representing academically the
institution. So, a degree program that formally goes through academic
senate and has all kinds of academic review criteria and demonstrated
evidence that it is what the institution believes is a quality program etc.
is one set of things that gets the institution's name on it.

Phillip Long: Individual departments can often issue c credentials that are
representing things that a particular faculty member does in a particular
class as a single assertion verifiable credential and the institution
specifically says we do not recognize that. we recognize that as our
faculty and it is our course but we do not recognize that entity unless it
goes through the faculty senate governant process which that thing did not
go through. So that's where this has challenges.

Dave Longley: I would say that I think this model captures that because
either you're saying this is the university and things that the university
issues would be trusted at the university level…

Dave Longley: but when you talk about the particular set of faculty or
whatever they're going to have different identifiers from And they're not
going to all if the university is set up that way they also should not have
the capability to issue random things using the cryptographic keys and
identifier of the university itself. So if all of that is kosher then you
would expect to be able to to differentiate all those things in this model.

Phillip Long: right? Yeah.

Dave Longley: So, you know that this comes through and there's some faculty
member or some sub college or whatever it is, they're going to have a
different identifier from the university as a whole. and…

Phillip Long: right okay it is very very much the case that the single
assertion OBV3 type credential is something that the faculty member signs
not the institution Nevertheless,…

Dave Longley: I think it would all

Phillip Long: they usually put the institution's name on it and the
institution looks the other way.

Dave Longley: Yeah, just to jump Q and in this model when you see that
identifier it's going to say the faculty member's name. It's not going to
say Virginia Tech. And so you can make that distinction because of how
we've set this

Manu Sporny: I'm wondering if so I'm going back to the use case right the
first one the list of known universities in a particular nation we're
getting rid of this I'm wondering if they're default actions or Not because
present to me feels like it's implied, if you say this, you're implying the
recognized action is present in anything. Go ahead, Dave.

Dave Longley: As an issuer of this list? How do you say that some other
party would trust these claims about that identifier?
00:45:00

Dave Longley: Not about the but about the thing with that identifier.

Manu Sporny: Yeah, I don't Yeah,…

Manu Sporny: again I'm again getting confused by the use case that address.
I don't know if the use case is in scope for the specification. I get the
use case and I get the value.

Dave Longley: So, we had the thing that's on the screen right there is with
recognized by up at that level and…

Manu Sporny: I'm wondering if there's equivalence here this isn't it?

Dave Longley: not attached to an action is where we started from. And the
idea there was a use case presented that the issuer might be different from
the value in recognized by that was the assertion when we first introduced
this model if we don't have that use case. So in my assertion that's better
expressed using the present action.

Dave Longley: But if we don't have that use case at all and we don't need
to cover things like that or we want to cover the self claims with a JSON
schema that says the issuer is yourself and…

Dave Longley: there's I don't know how easily you can cover the self-issued
claims case with a JSON schema. or that bit's not clear to me.

Manu Sporny: Yeah. Yeah.

Manu Sporny: I don't know if that's a use case that is in scope. I get what
you're saying, but I'm trying to think do we have that use case in the
universities in example? Do we have it in the driver's license example? Do
we have it in vital records or retail coupons?

Manu Sporny: I don't

Dave Longley: supporting people's ability to make assertions about
themselves is important.

Dave Longley: So I want us to make sure that if that we support the idea of
having a list from some trusted source so that I could say yeah this
identifier is for this person and then that fits into a trust model…

Dave Longley: where I can then accept for example self-issued claims from
them. Now that might be accomplished in some other way but I think that's a
general use case that should exist for all people.

Manu Sporny: I'm wondering…

Manu Sporny: if it harms that use case to say that you like the way I'm
coming at it is you can always do that. that is 100% what all this
verifiable credential did SSI stuff is about. You can always do that.
nobody needs to give you authority or whatever to do that. And I'm
concerned that if we formalize it,…

Manu Sporny: then all of a sudden we're saying, " no. You can't do that
unless you show up in somebody's list somewhere.

Dave Longley: No, we're miscommunicating.

Dave Longley: I don't know that other identifier is for. And if I have a
directory from a trusted source that tells me who that identifier is for,
then I can start accepting whatever I want to from that party based on
those claims.

Manu Sporny: Right? But…

Manu Sporny: why is this not enough?

Dave Longley: And the question is do we the only reason that it wouldn't be
enough is based on that initial case…

Manu Sporny: Do we need action present for that?

Dave Longley: where I want to if I have this list and this list says who
these parties are that's what the issuer claims but does anyone else that I
care about agree with these claims that that was and…

Dave Longley: that's why we were going to have recognized by be on there in
the be to

Dave Longley: begin with.

Manu Sporny: meaning yeah I don't know…

Manu Sporny: why I'm having such a hard time with this use case me meaning
somebody else is on this list that wouldn't be the learning commission
Right.

Dave Longley: So who's the issue of this credential right now? do we have
an example in the spec where recognized by is used with a different value?

Manu Sporny: No. …

Dave Longley: So that was the first question that came up for the
recognized by field. So we could remove that field and then walk through
the use cases and see if everything's covered.

Manu Sporny: we are removing recognized by, right? Yeah,…

Dave Longley: I'm saying, we could just do that and not add the present
action and see if it covers what needs to be covered.

Manu Sporny: If we say the present action is there so that people can be
explicit about it about something that's implicitly there already then I
don't think that there's a danger I don't know if anybody else is having
issues with the present thing or I'm fine if I'm the only one to back off
on it.
00:50:00

Manu Sporny: Go ahead, Phil.

Phillip Long: Does this imply that…

Phillip Long: if the university is recognized by the commission and there
are other entities that also are associated ated with the thing that the
institution is issuing that they would all be listed in this section. So
for example one obvious case is there's a degree in engineering that a
bachelor's in engineering that the institution issues recognized by the
commission but oops I said I typed I bet by mistake is the professional
society of engineering that offers and many institutions follow their
guidelines for the construction of the curriculum that is specific to
engineering ing and the commission.

Phillip Long: Typic commissions which are more regional recognizers of
institutions will often then they give their vetting of the institution's
quality from its procedures and quality assurance practices and the like
but the AET disation is a stronger one for that particular decision
particular diploma or bachelor award would

Phillip Long: they be listed in here and would that mean that every other
discipline that has something like that would be appropriate to list in
here?

Phillip Long: I'm concerned about this getting out of hand. You see what
I'm saying? Is that Well,…

Manu Sporny: I Yeah,…

Manu Sporny: I see what you're saying. I don't know how it connects to what
Dave's saying.

Phillip Long: I'm not sure. it may be a parallel comment. maybe I'm
confusing things by introducing it now. but I was simply reacting to the
fact that we have a single entity that is recognizing this if we take
recognize by should there be a mechanism for all these other possible
assert list.

Manu Sporny: If they're different recognizers, they would be different
lists is my expectation.

Phillip Long: Okay. Yeah.

Manu Sporny: And so I think that use case is covered by, if you've got, a
thousand different recognizers, then you've got a thousand different lists.

Phillip Long: I'm thinking of it by dis there many disciplines have a
professional body that does Yeah.

Manu Sporny: And each professional body would be a recognizer up here,…

Phillip Long: We'll just do it that way.

Manu Sporny: And they would all have different lists.

Phillip Long: Yep. Very good.

Manu Sporny: So each Yeah.

Dave Longley: So if…

Manu Sporny: Each Go ahead.

Dave Longley: if our assertion is that any party generating this list is
only going to refer to itself as a rec recognized by field in the actions
and we eliminate the recognized by field at the top level of the
description of each of these entities. I think that's probably fine that
the data will merge fine and you can pull things back out if you merged it
all together. And the assertion there is if you accept any one of these
lists, then you're willing to accept any of the claims made about the
entity with that identifier.

Dave Longley: So imagine we consumed 10 of these lists. You might get
slightly different legal names,…

Dave Longley: images, and URLs for that first recognized entity in the
right there. All having the ID did webun university. And what you would be
saying is I accept that all of those claims are true about that ID.

Manu Sporny: Mhm.

Manu Sporny: Mhm. Mhm.

Dave Longley: And if you're just looking at the claims, that's it. So if
you had some wallet interface where you have several of these lists that
you trust and somebody wanted to get the information about did webun
university. It might say this ID is associated with these three names. and
I think that's okay. And it does nothing.

Dave Longley: It really does nothing more than that than the directory case
of allowing you to know…

Manu Sporny: Mhm. Mhm.

Dave Longley: what entity is being referred to by that ID with these other
properties. And then all of the actions are much more granular. They say
you can issue or you can verify and then have if you're issuing and maybe
if you're verifying, we haven't gotten to that yet. They have out the
validation JSON schemas. And if all that stuff's merged together, since the
actions also say…
00:55:00

Manu Sporny: All right.

Dave Longley: who they're recognized by, you can differentiate. And I think
that model works. So, me trying to put present in here was about trying to
cover something that I think we're now saying you're just never going to
do. And so, I think we're okay.

Manu Sporny: And if that's not correct, we can always put present there in
the future.

Manu Sporny: We can add it without modifying the statement or whatever.

Dave Longley: Yeah. and…

Dave Longley: I think right and in the case of self-issued credentials I
think that case is covered then as well because if some credentials issued
by did webun university.ample and…

Dave Longley: who they are then that's just your own choice as to whether
you accept selfisssued credentials.

Manu Sporny: Mhm. Yep.

Manu Sporny: All So then currently we're not going to add present, but that
was a good conversation to have. and then if we find a need to do it in the
future, it's a fairly simple addition that, doesn't even impact the
context. It would be a single line addition to allowable values and an
action. Okay. I think we're out of time for today.

Dave Longley: Yeah, I just want to clarify for the notes there, self-issued
credentials. to be clear,…

Dave Longley: that doesn't just mean that the issuer is yourself. It also
means that the subject is yourself. So, you're talking about yourself. …

Manu Sporny: Yep.

Manu Sporny: All So, I can wait. No, we're not doing this. So, no changes.
I can wh thank you.

Dave Longley: we're going to remove recognized by at the top level.

Manu Sporny: Remove by what is it? recognized entity and then we're not
going to do that. okay.

Manu Sporny: We go. back over to you, Benjamin. I don't know if we have
anything else we wanted to cover today.

Benjamin Young: No, there hasn't really been a hard agenda on this …

Benjamin Young: because it's been pretty linear working on the spreadsheet
and then this PR. So, hey David, I was going to reach out over email.
you're on mute, too. Yeah.

David C: Sorry, I had it in my mind it was a 4:00 start. So, I just got a
nice time for the start of the meeting and realized it was a 3:00 start.
So, apologize that we're on holiday. We've just been swimming in the sea,
So, not really.

Benjamin Young: Nice. Yeah,…

David C: Yeah. Right.

Benjamin Young: we've been kicking around moving this meeting to a
different time. Sadly, it cannot be this upcoming hour because it conflicts
with the did working group. but maybe that's something we can discuss over
email to find a more successful time that works for you.

Benjamin Young: …

David C: Yeah. Yeah.

Benjamin Young: but we mostly discussed Manu's PR and there should be a
pinned link to that in the chat. if you want to take some time to look
through that and share thoughts asynchronously and then maybe I'll be on a
call next week that'd be great.

David C: Yeah, I've got the URL now.

David C: Just bringing it down now. Yeah, I'd verify our recognition
credential section today. Got it. apologize once again for being an hour
late.

Benjamin Young: And …

Benjamin Young: you're on holiday. So, I'm amazed you're here.

David C: Yeah. Yeah.

Benjamin Young: Yeah. And so, let's just try again next week and then I'll
start an email thread with those of us here to see…

David C: Yeah. Yes.

Benjamin Young: if there's a better time for this regular.

David C: Okay.

Benjamin Young: All right. Thanks all. Yep.

David C: Thank you. Bye-bye. that's the meeting. I missed the meeting.
Darling to the beach. Where is it? Here. Come on.
01:00:00

David C: pay off.

David C: Is that remember to clean up?

David C: this machine. this isn't

David C: we've had it since we came here. Yeah, everybody else is probably

David C: Right now my coffee I really do some whatever's in coffee that's
good for you. That was lovely. Have you got a little bit that I can have?

David C: Yeah, I left you a bit last night and I think you used it today.
Yeah. I'm going to have to make myself with another one. there might be a
tiny little bit there.

David C: could you make me a deep coffee? Where's the coffee? It's here.
The cup's been clean. Nothing is spared from the tidying hand of Maggie
MDonald comes through sweeping through, putting everything in its place
when it didn't need a place. And it was quite happy sitting on the table or
the th
01:05:00

David C: Yes, I shall have a decap and then I shall go and sit on the bed
and have a little maybe a do if you want to. I don't milk. yes there is
some in the fridge.

David C: All right, I think it was up already. Yes, I know. But this
morning when I think it was already open, the green one was finished with
Yeah.

David C: Then after I've had a do we can do cutting two nails and painting
the small glasses. Actually I think there might be one on my dressing table
my bedside table.

David C: Is that around there and is the one on the table there. Yeah. that
this one's for you have to wash That's obviously one for my bedside table.
And if something's full of lipstick in it, it's sort of a cue saying, "Push
me.

David C: push me. No, you're saying this is obviously okay with Chadwick.
01:10:00

David C: I hope this man
Meeting ended after 01:13:44 👋

*This editable transcript was computer generated and might contain errors.
People can also change the text after it was created.*