Manu Sporny: Alright hello everyone welcome to the VC API call this is January 18th 2022. ✪

Manu Sporny: On the agenda today is just an agenda review which is what we're doing right now introductions and reintroductions if we need any we will then move on to relevant Community updates spend five minutes on that look at to pull requests around workflow apis and hope to try to at least have a tiny bit of discussion there put 30 minutes to it but we can move on from it if. ✪

Manu Sporny: I don't want to dwell on that and then finally we'll be doing some issue processing as time allows and then any other business folks one are there any updates or changes to the agenda anybody want to talk about anything else today? ✪

Manu Sporny: Okay if not we'll go ahead and get started. ✪

Manu Sporny: See first up on the agenda is our introductions and reintroductions intros and reintros, is there anyone new on the call or anyone that has a chance had a change that wants to give us an update. ✪

Manu Sporny: Okay no introductions are reintroductions relevant Community updates does anybody have anything that the group should know about with respect to VC API or protocols or anything in general happening in the in the API space around verifiable credentials. ✪

Manu Sporny: All right no updates there just forgot to mention that the verifiable credential 11 spec the final date to vote on that is today if you have not voted in favor of publishing it as an updated wreck please please do so it doesn't have a lot of boats on it and for those of you that can see the votes you'll see some very interesting commentary by a very big company. ✪

Manu Sporny: It's on a good track it's just, anyway take a look -- https://www.w3.org/2002/09/wbs/33280/vc-11-2021-corrections/results ✪

Manu Sporny: Any other community updates before we move on? ✪

Manu Sporny: I guess it's worth mentioning that interrupts going to be a big part of this year with respect to kind of be Capi in some of the other stuff that we're doing in other spaces in I expect us to probably have an interop discussion not next week but the following week just to cover cover some interrupt things. ✪

Manu Sporny: Dimitri I shared a link in the ccg minute sorry I don't have it in front of me right now but you should be able to find it in the Raw logs public logs there. ✪

Manu Sporny: Right on to pull request so the first pull request is about the workflow apis so just to remind everyone on the last call we talked in the fair amount of detail about the non-normative workflow API things it just you know the messages. ✪

Manu Sporny: You know going back and forth. ✪

Manu Sporny: Around credential refresh and how that could be generalized this PR pull 255 covers actually let me go ahead and share my screen here. ✪

Manu Sporny: This PR covers a realization of that API so it actually defines it so PR 255 basically takes the non-normative text that we had last time moves it into a new section on workflows it adds two endpoints and defines them in open API specification. ✪

Manu Sporny: It identifies that we're talking about generalizations to the exacting to the existing VC API and then it provides two diagrams one of them is the generic VCA Pi flow this kind of back and forth showing how the workflow calls are made so you basically request to start a workflow a verifiable. ✪

Manu Sporny: Issuer/Verifier can request information for from you in a way of interacting and then you respond back to that with a verifiable presentation with the things that for example the issuer asked for or the verifier asked for and then once the receiving app that you sure the verifier processes the information they let you know. ✪

Manu Sporny: How it went so they'll either give you a new verifiable credential if you're refreshing or they'll say hey thanks I got that data looks good to me or they'll kick back an error or they'll say hey I need more information with you within additional interaction request there so there's a generalized flow in the spec now in that generalized flow is the exact same thing that we have proposed in the refresh spec. ✪

Manu Sporny: Refresh is a specialization of the generalized flow in the in the VC API PR that's here let's see let me see if I can get a preview so the new section is here under workflows. ✪

Manu Sporny: There are two endpoints workflows the name and start and then workflows in an active workflow ID and presentations so you either start a workflow and then you're asked to give it some data and you post the data at the presentations and point in these two are just kind of like works-in-progress the naming whether or not this is a design pattern or an actual endpoint URL still to be determined the interaction thing, here is the same one that you saw in the previous screen. ✪

Manu Sporny: And then it's just got a whole bunch of you know it has the examples of the thing that we talked about last time it also talks about the start workflow here and it's got you know Json schema on validating it and Response Center codes and things of that nature so that's that PR 255 let me pause. ✪

Manu Sporny: Any questions or concerns I see Mike's provided some feedback Marcus you provided some feedback and enough you'll want to elaborate on that here. ✪

Manu Sporny: Go ahead Justin. ✪

Justin Richer: There we go so my main question with this is that if that hey I need to get the user involved thing and a Marxist comment I think it was starts to touch on this can be abstracted in tough to be able to act as a jumping off point for a variety of different systems because this it it very much feels like WWW-Authenticate header style thing. ✪

Justin Richer: You're basically saying I can't fulfill this request but here's the information that you need in order to go do some to do a different protocol to fulfill the request that you want right because it is it is about API access and that is an API response and this I think from my perspective is where Adrian's questions of meshing with things like GNAP really come into play where things like you know did come messaging and stuff really come into play is is in that part of the step. ✪

Manu Sporny: Yes that is absolutely correct in what you just said is where the next PR picks up so the next PR talks about interaction and bootstrapping and other protocols and you know I can't do anything with you here you need to go over there style interaction so that is not. ✪

Manu Sporny: The pr here is just kind of a general kind of structure but I think your comments Justin apply more to like interact if interact comes back and says I need you to go somewhere else and do something else I think that's where the majority of your comments apply if you just you know if all if all the response then you can just respond. ✪

Manu Sporny: Clean that you know that could be the end of it. ✪

Justin Richer: So I would actually like to push back on that and ask why that is a different thing because in both cases you are telling the client here's something that I need from you. ✪

Manu Sporny: Yeah and so maybe it's not a different thing. ✪

Justin Richer: And that's as so that's basically that's how we're treating it in cap and which is where so which is where I think that there are some similarities in the pattern here whether or not there's overlap and then abstraction in the protocol I think you know remains we figured out but this similarity is that I'm seeing in the pattern are are around that type of thing where sometimes. ✪

Justin Richer: You talk to somebody else and go do a whole other protocol and then get back to you with something sometimes you just want them to answer a question that you're asking directly give me a verifiable presentation that answers. ✪

Justin Richer: This query or what have you. ✪

Justin Richer: But they're materially very similar there's semantically identical things where even though the process for fulfillment is very different. ✪

Manu Sporny: Yep yep so I think we need to kind of as a group drill into that and see if there's stuff here we can simplify or if there's overlap where we can remove you know unnecessary differentiation and things of that nature Adrian you're on the queue. ✪

Adrian Gropper: That was quick so having spent about a decade in the UMA context and then watching the healthcare people go and invent something called you'd App instead of adopting Boomer or looking forward to good nap I'm very sensitive to what Justin just said so I'll just leave it at that. ✪

Adrian Gropper: There with what good app is now fixing that was not fixed before and the the healthcare specific approach I don't think it's going to catch on even in healthcare but they're doing it. ✪

Manu Sporny: All right thanks for that input Adrian okay if there are no other comments on this PR please take a look at it and tell us what you think of it that sort of thing the next PR up is in a different repo but this goes to what Justin was. ✪

Manu Sporny: Was speaking to. ✪

Manu Sporny: Is you might want to bootstrap into another protocol or they may be another protocol that comes into effect so in the verifiable presentation request spec which a number of vendors are using in this space like the traceability folks digit bizarre uses it there is a section on interaction so last I think it was a week or so ago. ✪

Manu Sporny: For mediated presentation, you need a human being to step in and do something or you don't need a human being to step in and you can just respond through HTTP that takes basically verifiable presentation right so those are the two things that we had before but interaction as we talked about last week has you can provide a whole bunch of different ways of interacting in these are usually known as. ✪

Manu Sporny: Has different protocols. ✪

Manu Sporny: And so when you say I want to start a workflow in the response comes back to you in it there can be an interaction interact mechanism and it could for example be an open IDC credential provider interact mechanism where it expects you to then bootstrap into an open IDC credential provider approach so this is the work that matter has been working on. ✪

Manu Sporny: To respond using open ID Connect credential provider there's a it says I understand that and you can go here to do that or if you wanted to interact in communicate over did calm so you wanted to switch from VC API into did Cam you have that capability through this Edition and you can list multiple interaction mechanisms so you can respond. ✪

Manu Sporny: Just through VC API you can respond with open IDC credential provider you can do respond with did calm in the receiving server the issuer for example supports all three protocols or two or whatever you know it is so this PR is meant to demonstrate that we could bootstrap into other protocols. ✪

Manu Sporny: Through you know verifiable presentation request it tells you how it can interact now clearly the market it's best for the market if we could agree on one thing you know simpler if it solves all the use cases you know simpler security characteristics less code to go wrong you know all that kind of stuff but what we're trying to do here is kind of show. ✪

Manu Sporny: All the different wallet. ✪

Manu Sporny: Two calls that there is a way to go about this where not everyone has to choose to die on their own Hill right it's a we can at least agree to the mechanism in which we then bootstrap into a separate protocol which you know is is one way that we could go about this so This PR is pretty simple it just adds the oid see credential. ✪

Mike Varley: Hi there can you hear me oh super first time using Safari here okay my question on the flow this is the the holder reaches out to the issuer in the very first call in order to get this message back that says how the issuer can interact correct for all cases. ✪

Mike Varley: Yep yep no I can see it. ✪

Manu Sporny: Yes yeah that's correct can you can you see my screen I'll try and point so on the on the diagram step one is so this is the generic flow but it lets assume holder on the left issue or on the right if holder reaches out to an issuer and says I want to start this workflow the issuer can respond back with that's great here's the information I need from you and here's how you can interact with me in that at that point. ✪

Manu Sporny: And say I want you to interact with me using kind of the VC API or I want you to interact with me using open ID c-- credential provider or did convey to or wacky packs or this is a fully automated exchange just you know give me a presentation back in return all of those are options could the options. ✪

Mike Varley: Okay so the slow step so this first API endpoint is like a almost like a dynamic Discovery API call. ✪

Mike Varley: That would be kind of standardized through this API saying any wallet implementing this Dynamic Discovery can make this API call and then look for an interact response which then might switch them over to a did Chamorro IDC flow where that stack is fully implemented that's it I'm getting that sense I'm getting that flavor I don't know if that's the intention. ✪

Manu Sporny: That is the intent I'd stop by you know I don't know if it's a discount I see how you could see it as a discovery mechanism it's really meant to be like I want to do this thing with you and then you say great here are the ways that we can interact accomplish that. ✪

Mike Varley: Right and so I guess my other comment there is it's not like a static dot well-known how could we possibly talk together type meta document it's a context specific Discovery API. ✪

Manu Sporny: Yes so the place that that's kicked off let me bring up this other PR here there's this start workflow thing where it's like workflows the name of the workflow and start there's a presumption that this is shared out-of-band somewhere right and it could be that you get this URL from a well-known thing that is yet to be invented. ✪

Manu Sporny: At you you have a specifically like a specific end point you go to to just kick the process off and from there you can discover how they interact for that specific process for that specific workflow you're talking about that make sense. ✪

Mike Varley: It does I'm wondering than would a valid implementation this may be a huge question so I apologize if it explodes but could a an implementer of the VC API just implement this workflow start and response message and then for example the open ID connect or a did. ✪

Mike Varley: Our core but so the only piece they really Implement from VC API is his this component. ✪

Manu Sporny: I think the short answer is yes I think at least that's kind of been the proposal but that's up for the group to kind of discuss and see if that's a you know a good idea or not. ✪

Mike Varley: Absolutely and I just wanted to hear your thoughts on that and you said you know yes that would be compliance okay cool great thank you. ✪

Manu Sporny: All right thanks Mike Mike Pro rocker on the queue. ✪

Mike Prorock: Yeah thanks man who and and mr. Marley I think you hit on something key which is it's highly context-specific right this is not like a capabilities Discovery mechanism ride this is not saying oh what crypto do you support or anything like that right it's very specific around you know oh you've got this thing let me go ahead and deal with this specific item right like a credential type item or something else the. ✪

Mike Prorock: I think the other key thing you hit on is that you may we may well see micro service type deployments that are like yep I'm just handling workflows or I'm just who you know handling verification etcetera as subsets of this so. ✪

Mike Varley: +1 Thanks ✪

Manu Sporny: Yep that's that's exactly right Mike you know I think the goal with the VC API is you can choose to only Implement subsets of it so issuers would probably only you know Implement issuing May Implement workflows verification verifiers would maybe just Implement verification without issuing you know present to holder software might just implement this side of it right. ✪

Manu Sporny: Workflow stuff... anyone can Implement so whether you're an issuer you might have workflows for issuing stuff for verifiers you might have workflows for verifying stuff and you could even contact holders and say I want to engage in a certain type of well-known workflow with you but I mean that's that's a bit speculative right I don't know praise ability folks might have a use case it's actually I think traceability does have a use case for that where the issuer contacts you and they're like I've got some credentials for you. ✪

Manu Sporny: And so that could be viewed as like a workflow. ✪

Mike Prorock: Yeah since you went there yeah we have that we absolutely do have that exact use case of saying yep I've got a pile of credentials or a presentation with a bunch of you know credentials inside of it that you know let me know when you're ready to come get this right and and you know it does some exchange of challenge token Etc once again we also from an interop standpoint are starting to look at that capabilities Discovery side right and how does that get done we had been using did config which has some. ✪

Mike Prorock: Likely shifting over to did web on a lot of that so but that but that kind of capabilities this discovery of are you supporting issuing you know are you supporting verifying right basically what are these endpoint groups are you supporting that side of it is getting you know somewhat auto-detected and check box via tests on our own the interop side of things for traceability so. ✪

Manu Sporny: Got it thanks Mike yep yep okay so I mean plenty to dive into here but you know the first step is to try and just Define exactly what the API looks like with the schemas input-output documentation looks like and kind of go from there okay that's PR 15 on the verifiable presentation request spec any other comments or questions on bootstrap into protocol from VP API? ✪

Manu Sporny: Okay if there's nothing else there let's go ahead and jump into issue processing next to processing. ✪

Manu Sporny: https://github.com/w3c-ccg/vc-api/issues?q=is%3Aissue+is%3Aopen+sort%3Aupdated-asc ✪

Manu Sporny: Go ahead Mike. ✪

Mike Prorock: Yeah just a light that posit is one thing that occurred to me and I just wanted some clarification on your thinking on this so far as far as bouncing out into other services obviously things that are very credential you know related such as oid C etcetera makes sense there what about like arbitrary Services right because a case that comes up for us the you know still in the early phases of hashing out as like you know doing things like saying hey. ✪

Mike Prorock: Areas of credentials coming in related to some known thing like a shipment can you go ahead and just start streaming me these credentials you know over to some arbitrary service right that good just think Pub sub type stuff is that something you and vision being able to handle with this kind of thing basically saying yep if you're authorizing X you know yep you could go ahead and do that and then fire off this service here or do you think it should be bound a little more. ✪

Manu Sporny: The former so interaction types are meant to be in extensibility point in so if there's some kind of like let's say you're doing like mqtt streaming of some totally different other type of credential you know I think that's a possibility now you know is it is it a good thing to bootstrap from like a verifiable credentials ecosystem in to some kind of totally different ecosystem I don't know but it's pause it's that it's certainly possible to do that right or maybe it's you know you're sharing verifiable credentials over mqtt instead of this or Bluetooth right or pick anything else pick any other type of you know protocol that's not HTTP you should be able to bootstrap into that from using this interaction type. Did that answer your question Mike? ✪

Mike Prorock: Okay cool it did yeah thank you. ✪

Manu Sporny: Okay alright so let's go ahead and jump into issues. ✪

Manu Sporny: https://github.com/w3c-ccg/vc-api/issues/97 ✪

Manu Sporny: First one up is issue 97 in this is Mike you open this issue confusion around and point naming Mike would you like to remind us what this issue is about it's a year old at this point. ✪

Mike Prorock: Yeah this is yeah what back in the day a little bit and I think bird from discussions going back even into 2020 and before but basically the whole point of this issue is that you know there's some pretty good guidance around how do you name rest endpoints and things like that and it you know we don't really follow that Super Crate here on the PC ABI there have been some improvements to this. ✪

Mike Prorock: Basically kind of getting at you know hey should we go ahead and adopt certain naming you know conventions and if so when is a good time to open a PR on that when it's good time not to etcetera so especially as we think about things like multi-tenant Services Etc gets you know more important properly. ✪

Manu Sporny: Right so I think yeah excellent point I think we decided to follow I forget the name of the the approach but you know it's where you have a verb as the as the last as the last bit so cart checkout playlist play that kind of thing I think we agreed to using that General pattern in general and then you know the question of can you tack the API on the end of a tenant ID I think I thought we said yeah you can do that meaning that like you know the the endpoints are not intended to be just bear off of the off of a domain I think you can I thought we said we could put them at the end of it. ✪

Manu Sporny: As long as they're the end of the URL we're good but I don't know if I'm misremembering that anyone else remember what decision we came to there. ✪

Joe Andrieu: I also recall the latter one specifically that's requiring it to be at Route was an anti-pattern we seem to recognize. ✪

Manu Sporny: Okay. ✪

Manu Sporny: I'm going to write that down because I think that's what we said anyone have any heartburn over that. ✪

Manu Sporny: And then does anyone have any heartburn over the does anyone remember the name like. ✪

Manu Sporny: Name for what we the approach we took. ✪

Mike Prorock: Yeah I'm looking it up and basically no one uses it which is also part of the confusion anytime I throw this API at a developer is it doesn't match what everyone else in the world does so that's you know which is more noun based followed by you know verb of like I've got credentials and then I wanted you know I have a credential / I want to issue or you know a getter in this context of you know holders for us and stuff like that. ✪

Manu Sporny: So do we want to write that down as a decision we made and it's problematic or do we want to say we never made that decision but for whatever reason that's kind of where the API is these days and we need to talk about it more. ✪

Mike Prorock: I need for my son I'd like to talk about it more or at least revisit maybe with a reorg PR but whenever timing is right so I'm happy to comment on the issue on that. ✪

Manu Sporny: Do you think we need a new specific issue around that or do you want to continue discussing it in this position. ✪

Mike Prorock: I think continuing here is fine I think as I know Ori added some actually some helpful examples up above of different pads that also get followed so. ✪

Mike Prorock: Since then you know you know restful API obviously is gone through and done some very useful things like you know put out an actual design tutorial for restful apis right and stuff like that that we probably should adhere to if we're trying to you know get just in as much as this is pre standards that's part of my concern is let's not get down a certain path too far and then have to go revamp everything right. ✪

Manu Sporny: Yep okay all right that's that item anyone else want to weigh in on that or before we move on to the next one. ✪

Manu Sporny: https://github.com/w3c-ccg/vc-api/issues/104 ✪

Manu Sporny: Next item is issue 104 raised by One enable HTTP request to specify optional verification checks there was a PR that just went in at Marcus I think you were you were involved with that PR and Charles was as well where we took. ✪

Manu Sporny: I believe Marcus do you remember what we did there? ✪

Markus Sabadello: Yeah I remember that the problem was specifically that there was a conflict between what's implemented in the test suite and what's in the open API specification so the test Suite was sending checks option that was not actually Allowed by the by the open API specification and at some point we had a call which I moderated I think it because you ✪

Markus Sabadello: Couldn't that call we had a lot of discussion around verifying and validating and there was pretty strong consensus that some things that we would consider. ✪

Markus Sabadello: Validates validation like checking if check satisfies certain business requirements things like that would be out of scope so a lot of things that would maybe justify a checks option that would make that would add some flexibility in what exactly happens suddenly seemed to become less important than less relevant and the group was supportive of having a more well-defined. ✪

Markus Sabadello: Behavior when specifying something rather than having too much flexibility and there were several open issues related to that and all of that seemed to go into a Direction Where We want it to moved option again first first reason being the the contradiction or the conflict between the specification and the test suite and the second reason being a feeling that we wanted less flexibility. ✪

Mike Prorock: Yeah just recalling some discussion around when this issue was opened I think some of the difficulty was in the way it has been addressed since we've moved it to like actual more open API proper you know definitions so like I know when the trace case were now able to actually test them using normal you know you know Postman Etc tall you bought it you know APA valid API. ✪

Mike Prorock: This is a lot of this in addition to the fact in addition to Marcus's adjustments in that PR you know some of the little while ago so. ✪

Manu Sporny: Okay so I so that this decision is the group the group talked about this we wanted less optionality and we have taken this out of the API. ✪

Mike Prorock: Yeah and we can actually validate the API schemas now as well more more fully. ✪

Manu Sporny: Juan do you feel like then the issues addressed like this is not an option anymore we can verify the endpoints we're done with this issue okay. ✪

Manu Sporny: Okay great all right next up what do we want the March 20 21 flood Fest feature free state the be. ✪

Mike Prorock: I would recommend some time in 2021. ✪

Manu Sporny: I mean we can't change the past do we just close this. ✪

Mike Prorock: Close no longer relevant. ✪

Manu Sporny: Overtaken by events it is now 2022. ✪

Manu Sporny: Alright next one up. ✪

Manu Sporny: We're going to have to probably hold off until Orie gets here... this is a learning credential VC edu... Dimitri, Kerri, any thoughts here. ✪

Dmitri Zagidulin: Let me pull that up. ✪

Manu Sporny: https://github.com/w3c-ccg/vc-api/issues/107 ✪

Dmitri Zagidulin: Right. ✪

Dmitri Zagidulin: We should add an example credential like that. ✪

Dmitri Zagidulin: The exact contents of that credential as you can probably imagine has been evolving among very different among various different groups so. ✪

Dmitri Zagidulin: Kerri, would it make sense for us to have the discussion item Paul to get by it. ✪

Kerri Lemoie: Yeah I think so can you hear me okay okay good so there's a couple things going on is when we'll to is that some of the work that describes this data is happening into education standards groups open Badges and CLR but also a VC EDU we've been talking about I'm having a context file that does this either to helps describe this so I think I might be safe too and if we could move it over into VC edu or. ✪

Kerri Lemoie: What do you think, Manu? ✪

Manu Sporny: Yeah we can't we can move it over there's a transfer link somewhere here I forget where exactly. ✪

Kerri Lemoie: Does that make sense to you Dimitri? ✪

Dmitri Zagidulin: It does yeah do we want to leave a comment to that effect and see if Orie is good with that. ✪

Kerri Lemoie: Yeah that'd be great. ✪

Manu Sporny: I was gonna okay all right so where we would. ✪

Kerri Lemoie: Absolutely I am. ✪

Manu Sporny: Okay all. ✪

Kerri Lemoie: Yeah a lot has changed in VC I do since this was made so. ✪

Dmitri Zagidulin: Right. ✪

Kerri Lemoie: Thank you. ✪

Manu Sporny: https://github.com/w3c-ccg/vc-api/issues/105 ✪

Manu Sporny: This one might also be easy this is issue 105. ✪

Mike Prorock: And long since resolved I issued PR is on that last year sometime. ✪

Manu Sporny: Okay. ✪

Manu Sporny: Okay that's that one. ✪

Manu Sporny: Here's a good one. ✪

Manu Sporny: https://github.com/w3c-ccg/vc-api/issues/91 ✪

Manu Sporny: There's a suggestion that we should use a well-known endpoint for an API to advertise optional features supported such as digital signature schemes credential status methods optional apis some discussion between or Daniel Buchner one MikeP you're the last comment on here thoughts. ✪

Mike Prorock: https://github.com/w3c-ccg/traceability-interop/pull/45/files ✪

Mike Prorock: Yeah the obviously this is evolving because did config we have been leveraging over on the trace side for exactly this part of the issue is that did configuration is highly complex for folks that aren't familiar with it to generate and or generate properly I just put in the chat a link to pull 45 which we just merged all... this has to do with the same issue as far as Discovery but leverages a did web for that as far as just saying hey what keys do I need to use for testing it obviously does not cover all capabilities right that the this issue addresses but it does have I think is a good step in the right direction. ✪

Mike Prorock: Since we have gotten away from optional features right on the VC API that in addition to leveraging something like you know did web for at least key Discovery for testing purposes and things like that you know is it probably carries us most of the way there is my personal Sense on this so. ✪

Manu Sporny: All right any any other comments on this one? ✪

Manu Sporny: What do you feel are like next steps here Mike. ✪

Mike Prorock: Well we can as we get into kind of implementation of that PR 45 that I linked on the trace side we could always back Port that over 2v Capi once we've got it up going you know with Dynamic testing that might be a path right we've been doing some stuff like that I'm already such as like presentations available Etc so that might be a good way to handle it right. ✪

Mike Prorock: The appetite for other implementers something to like Spruce and stuff that may want to view no weigh in on you know service and key discovering type issues right and everyone shares these you know concerns is just where do you hash it out and test it out first to make sure it even works before you try to bother defining an API around it. ✪

Manu Sporny: Okay so let's just say discussion will continue in this PR with this update does that work for everyone. ✪

Manu Sporny: https://github.com/w3c-ccg/vc-api/issues/101 ✪

Manu Sporny: Next on opposition 101 the readme is not clear on purpose or contribution guidelines let's see my kanuri Mike. ✪

Manu Sporny: What are your thoughts on this one. ✪

Mike Prorock: My thoughts are going to send really nasty messages to Horry on slack the for hanging me out on this one this I believe this is actually a dupe I think because. ✪

Mike Prorock: In some ways right there is that API style guide comment right controller versus controller was the word we were looking for made it earlier that has been more or less kind of moved away to collection style and points this I think we can close as a dupe of others I think the contribution guidelines we got PR din because I wrote those and then the API guidelines are being hashed out in whatever issue that was we looked at earlier. ✪

Mike Prorock: Issue 97. ✪

Mike Prorock: So between those have between prior PR S Plus issue 97 ongoing discussion I think we can close this. ✪

Manu Sporny: K any objections to that from anyone? ✪

Manu Sporny: Okay we will close that then. ✪

Manu Sporny: Next one is from Orie and this is the API does not meet the needs of our customers. ✪

Mike Prorock: Yeah well I can get it some of this right so that we that I know we have addressed right so one of the issues was some of this spun out of contention around including an array of credentials in a verifiable presentation right for being able to bulk exchange you know either do one single compressed etcetera and minimize traffic over the wire so that is one item. ✪

Mike Prorock: This also allow for workflows that could exchange data via other protocols if required and under wow the transcribers amazing the huh that's sorry the you know the the other one is the workflow side right helps with this as well as segmentation around better you know API naming and breaking stuff out. ✪

Mike Prorock: Right but so I think we're headed the right direction on this like as a as a group I don't know about closing the issue my inclination would be say yes assuming Ori has no concerns there with closing it but I do think that it is something that we do need to consider continue to be mindful of right because a lot of these workflows are very different when we're looking at you know tens of thousands of concurrent credential exchanges going on you know sometimes per minute and the supply chain side against you know possibly one server right as a different scenario than we see in other cases it mainly obviously you have flowed cases as well right where you could see definite High load type situations so it's just stuff we need to be mindful of as we're designing this stuff out we need to design it to last and you know follow all best practices whenever possible even in a void. ✪

Mike Prorock: Reinventing the wheel right that's that's I think one of the biggest dangers we could fall into is trying to go oh because it's credentials or web three or whatever the cool new term is you know to forget the lessons that have already been learned by everyone else right over time so that's my soapbox is now set aside and them back. ✪

Manu Sporny: Okay alright so I wrote down you know high level of what you said are we okay I feel like these are General considerations we need to keep in mind and we're headed in the right direction with the group be okay with closing this issue which is pretty big and Broad in then or suggesting that we close the issue to Orie and then asking for more specific issues to be raised if there's anything remaining. ✪

Manu Sporny: We're out of time only two minutes left next one up was going to be 113 the oauth2 bearer token thing for plugfest 2021 we didn't do that this was you know overtaken by events we could probably close it but let's not rush that well I mean this is overtaken by events for the 21 plugfest we could update this for 22. ✪

Manu Sporny: At a time anything else folks would like to comment on before the end of the call anything we should keep in mind for the next week any of that stuff. ✪

Mike Prorock: One single chair note which is looking at certain people on this call to self nominate for the open ccg chair roll. ✪

Manu Sporny: +1, That thanks for the reminder like all right with that thanks everyone for the call today appreciate the engagement we will be back next week. ✪