Note: Jitsi went down at this point. The below content was recorded with Otter.ai and the audio file only reflects what is below (Jitsi recorded over the first part of the meeting).

Simone Ravaioli 24:22 Apparently not, maybe just as a comment, we know that there are the work on use cases is important. That's why other groups are working on it. Namely, you know, the CLR work group and IMS is defining their set of use cases. We know that there are some being specked out by d3. So we index an opportunity for us to check whether we are on the right track as well. We are loosely covering the same ones. All right. Thanks, Carrie.

Unknown Speaker 25:00 Any other question?

Simone Ravaioli 25:05 All right, let me try to send Juliet. Julian, would you like to try to come back on stage and share? I'm going to type in my email anyways.

Unknown Speaker 25:28 Hello,

Unknown Speaker 25:35 can you can you hear me I have an issue. Here.

Simone Ravaioli 27:34 All right. Something happened for sure. Sorry about that, everyone.

Unknown Speaker 27:41 Yeah, that was really changed. Like just do it down for a little bit.

Simone Ravaioli 27:46 Awesome. Yeah. So I'm going to start the recording again

while we get everyone back recording is on.

Unknown Speaker 28:04 Let's see.

Simone Ravaioli 28:06 Julian Are you back with us yet? No, I don't see him yet.

Unknown Speaker 28:11 Then yeah.

Simone Ravaioli 28:12 All right.

Unknown Speaker 28:16 Yeah. So let's set it to a desert and at that moment, you know, it is what it is go wrong. So I'm going to share my screen now and it should be okay.

Unknown Speaker 28:39 So, think Nope. Nothing. You're not seeing what I want to share. It's just got so many options now. Okay, so let's see. Was Firefox as the education this one yes. And now you see the presentation. We do. Yes, perfect. Okay. Nope. This link Okay, so blog so the free so yeah, the main change is coming to Brooks refreeze to the alignment is to be to see verifiable credentials, spec. And we also brought in a couple of things did support mostly a couple of new properties and I want to expose a bit of the roadmap what we want to do in the coming year because looks refreeze one term because a lot of things to do, obviously to keep on

Simone Ravaioli 29:52 Julian, I'm sorry to interrupt. Could Yes, it looks like people are having issues seeing your screen. So there's a black screen right now so we're not seeing your slides.

Unknown Speaker 30:06 Okay. Maybe you should just say if you don't go full screen Yeah, yeah. Okay. Yeah, like this. Is this working now?

Simone Ravaioli 30:19 I'm not for me. I wonder whether others are see. Okay, nope. Nope. Not working. Okay.

Unknown Speaker 30:28 Okay.

Unknown Speaker 30:31 Not easy. today. Okay. Again, this one, alo and then. Yeah, now you're seeing it? Yes. Okay. Okay, so I'm not going to go full screen and stagger. So yeah, like WCC verifiable credentials. Did support new properties in the roadmap. So yeah, well, I mean, obviously, before the version two was the language open badges, as you may know, and now we wanted to go with the standards. gear for PC and this group to the, you know, part of the greater ecosystem. So yeah, the idea is to facilitate interoperability, but intelligibility. We would like to, you know, find credentials, verify in our system, and vice versa. And yeah, to align with the rest of the companies and industry on that. We also wish to make blockcerts more versatile. So not just education, but you know, have more use cases, trying to liberate physically the signature and the display property that was two things I think that make books stand out, in so we remain with a strong credential proofing we're using NACA proof 2019 which was developed in partnership with volted and the its it remains a blockchain anchoring solution. So that tively it's Bitcoin and Aetherium. But we also got some proof of concept that you can work with us chains and Muslim Muslim. Namely, we are the Matic example working, especially on the verification side, but then you know, you'd need some customization. So it's a little bit technical, to make it work, but it's, it's doable, and it remains tamper proof anything that's the great value proposition of cocktails that you cannot just go about and modify because we do the hash comparison. Yeah, for so for those who are not familiar with fluxes, what we do is is as a maca root of issuance that we anchor on the blockchain, as well as a hash of the document and the Merkel root is is just a macro free of all those caches of all the documents that were out of decisions big together into one maca root, and then that we save on the blockchain and on the verification side, then we're going to go fetch the transaction associated with the issuance, find the maca root, find the hash and make sure that we can we compute those things locally. And if everything matches, then you'll most likely have a valid, verifiable credential. There are more steps to the verification to the term saying most likely, but that's the recall of the proposal. But we also added decentralised identifiers to Bloxels be free. So that's at this moment only for the issue of we want to do the holders a little bit later. But the idea is to augment the trusts in the issuing entity. So yeah, make sure that the issuer is particularly there and we are using by default, the API from the D Universal resolver although it's advised to not use that in production and you know, you should spin up your own resolver and provide that to the verification library. It's fairly easy. It's just a parameter in the API. And then yeah, what we do is that we take the deed and the public key that's in the deed and we recreate the issuing address and with that, we prove that the deed and issuing address of the

Unknown Speaker 35:13 issuer are matching. So there's a little bit of a graphic here that's going to be a little bit maybe easier to understand. So we start with a CD please. Following the beep 49 recomendation. And with that, we generate the Fisher's private key with 60 256 k one, B, two proposition in between. And then with that we can have the issue as public key from which we derive the issuing address either on Bitcoin or Aetherium. And we also create a deed documents from night tests I've used iron mostly that was one with which I could use 60 to 561 keys and then you into documents. The final document two would have the issuer ID needs to be the deed and you also need to provide the verification method in the proof section. And then that's how you get you know, everything gets baked into the document. Like she's signed and then anchored on the blockchain and later on, we take the we do the the way backwards. Like I said, to match the issuing address and the document. As for new properties, what they're not so new if you're familiar with book certs, we just made them into the stand up that we were providing with learning machine, a couple of them which were metadata, JSON and display HTML. So we change that a bit, put them in a standard article base since they were the value proposition. And so now we've got metadata. Which is basically a way to provide more keys that are you know, that you didn't define in the context in the JSON LD context and that are specific to issuer the issuance, the older you know, whatever is needs to be flexible around the issuance in the document, the verifiable credential. And we also have the display property, which is still the visual layer but support supports HTML as it did before and v2, but now supports also images. So these formats PNG, jpg, BMP, and GIF, and as text, so like, you know, it could be anything but basically would be base 64. And we also do support PDFs as well, in this base 64 So you need to put that into the documents that may end up with a big document, but if you need to, you know, VF increases and put into a verifiable credential. That's the way we solve that. And on the broadness of maybe a little bit longer, but that's the first point is to align the revocation list with the WT CCG. Works. That's the status list. 2021 So we started looking at that, we still want to improve the trust with the issuer. So the main point that I cannot resolve on my own and maybe if you could ideas, I like to hear them but I still think you know, the song is not enough to prove it's the the the identity of the holder of what I mean by that is that you know, it's easy to create a document and then create a verifiable credential. And, you know, claim something. So, I'm trying to think of ways to provide maybe a challenging point or, you know, other solutions like this registry where you could go interrogate to check that it is indeed, does indeed belong to any and social media we've got on our plate this year. And we want also to focus on the olders experience. For now we've got a word which was created like I think 2018 or 2017, and needs, you know, a little bit more love, you know, from the point of view of the UX and also to align again with the standards and the ideas people have brought in the space so we want to make it easy to hold and share credentials, we want to

Unknown Speaker 40:15 provide the ability to create verifiable presentation and potentially also LP selective disclosure. proofs. Select to limit what knew what the holder can share with any verifier and we also want to improve the old does identity proofing to make sure you know that obviously, that somebody show presents a credential that they are they are you know, again, and we also want to see if the things around the decentralised hosting. So like maybe using IPFS is something that's more like proof of concept and maybe a little bit longer before releasing anything there. But it's something we want to explore yeah, I've got a couple of things if you want to go further so electrical distributors website it doesn't change the communities you wherever it is still there. Then if you want to see what the context is, like, on GitHub and as a public open source repository, and the issuing library, also in the verifying components were put during to the web components so something that you can use directly on your website, as we click on looks us at all expense. It's this piece here. And the verification happens on different labour, that we also thinking maybe putting an API on an API somewhere. So that's it. So I don't know if you've had any questions to answer them.

Simone Ravaioli 42:12 Thank you, Julia. Let's see if there are any questions in the queue here. There were a couple here as you were speaking. Hey, go carry carry as a question. Thank you.

Unknown Speaker 42:28 I was wondering if you could put the link to the context in the chat for us we could.

Unknown Speaker 42:34 Yeah, so it's such as the context it's a bit more than just a little bit of modifications. I'm just going to show you but basically, it's got all these files. So you can just go through them and read them, but that's where JSON LD contexts will be defined. So I'm just gonna show you the whole

Simone Ravaioli 42:57 thing. Julian, while you're there. Nate was asking if we could see an example of Jason of initial credential.

Unknown Speaker 43:05 Sure. So let me see I've got one here. Yeah. Okay, sorry. I'm just going to put it in the in Firefox so fine because it's on my machine.

Unknown Speaker 43:32 We're just gonna see if it's correct. And so I'm just gonna show you like first of all, verification

Unknown Speaker 43:54 it's working. I don't know it's okay. So it's not starting for some reason. So I'm just gonna trip this year. Yes. Okay. Very nice. Maybe it is. So yeah, that's a huge context here. May or may change but it's, it's like one situation of peace and stillness to show you know, extent of how we're going to use it. So the things that are flexible, but here we can see that three issue is that logo issuer so the basics for quite some space, and each state credential subject at this point is very simple. In the last version with these repeat more information, we decided to use claims as the main property for describing what CERT is about, is the metadata as I was describing, and then we display is the text that would be type. Then we have the HTML that gets displayed. So I don't know why I cannot see these two Yeah, I'm just gonna try get dropped maybe.

Unknown Speaker 45:44 No, I don't know.

Unknown Speaker 45:49 I don't nothing else. I don't know what's happening, but I can demo for some reason. That

Simone Ravaioli 45:56 no worries. Maybe you can isolate an example of it in in the main news, too.

Unknown Speaker 46:04 Sure. Yeah. Yeah.

Simone Ravaioli 46:08 There was another question from T long as you were presenting. This is a question Are there any blockchain is the blockcerts support that are not proof of work consensus algorithm based?

Unknown Speaker 46:23 natively? I'd say no, because we are using Bitcoin in a theorem what obviously theorem is trying to move towards full steak as you know. Like I said, you could customise your own chain, I think, as long as you you, there is like an RPC client on diversification, but then even probably on the Asian side, you should be able to to use any blockchain you you'd want. So yeah, I haven't tried any other like, I'd be happy to try maybe I'll grant or sort of chain will be permitted. To to see but at this moment, no.

Simone Ravaioli 47:10 I carry you're in the queue again. Thank you. Sure.

Unknown Speaker 47:14 Thank you. I'm not sure everybody here has a lot of experience with blockcerts. If you don't mind, I'm just going to sort of like recap my understanding of how it works. Maybe confirm that just so like from an overall height picture show effective, which is that the issuer like the the there's a badge this an Open Badge fibration to have an open badge that an issuer issues to a student, let's say has saved that's a scenario to their public key. And then it does, it gets hashed and the hash of this badge is put on the blockchain. And the verification actually compares the data that's being presented to the hash that's a blockchain that the hashes match. And that's how the verification works. It's correct. Correct. Okay, good, right. It's making sure I had it right. And so as you get closer to like aligning with verifiable credentials, do you think any of that will change because, you know, verifiable credentials, have their own proof method, and cryptography and ways to, you know, ensure tamper evidence and things like that?

Unknown Speaker 48:28 No, I think blockcerts is gonna, I don't know I can there may be some business decisions that are not aware of, and I'm on the engineering side. So, you know, I'm not the notes there. But I think it's I think it's it's, it's not gonna be changing that. It's still a value proposition that we have something that son killed on the blockchain and then the proof is going to be there. as well. I don't know. I think actually the maybe some ways to do like a lighter version. Maybe Jason actually. Can can wait on that one. There. Case, any skill resist?

Unknown Speaker 49:17 Yeah, I'm still here. So could you rephrase that question? For me?

Unknown Speaker 49:20 The question is a symptom maybe like, Will everyone be using blockcerts issuing on the blockchain or will there be possibilities to use blockcerts with different cryptographic suites?

Unknown Speaker 49:43 Gotcha. So as best I know, for the immediate roadmap, those are still intended to be anchored to blockchain. I don't I don't think I've heard anything different from that.

Unknown Speaker 49:56 Okay, nice. So just wanted to

Unknown Speaker 50:01 thank you

Simone Ravaioli 50:04 I do have a question. For Julian, and this is more around what kind of support do you have around blockcerts At this stage mean, in terms of community, I mean, who else is contributing to the development? We're finding help or what's what's the status of the community?

Unknown Speaker 50:24 It's it's mostly me and the team so my colleague mature was not under call to and yeah, we're, we're still we're very limited team at this moment. Working on books, so that's why the development is slow. But with good alignment with the company to you know, pursue the fault and bring more value to two blocks out. It's another tech project, but it's, it's still very small. capacity as it was with running machine.

Simone Ravaioli 51:02 Do you are you getting any help from say, our universities that may be adopting blockcerts? I know there's certainly one in Canada.

Unknown Speaker 51:14 So we get some feedback on the Roxas firm but it's it's not very active. It's not as much as we'd like to. We, at some in the past two years, there was someone in Italy I forgot his name, you know, but what contributed to blockcerts nitride is become a bit slow and reopening with his books of the free we get a little bit more traction again, on the on on this side of development so that it's not just us, driving the thing. But you know, other people are interested in get committed to to the verifiable credential blockcerts.

Simone Ravaioli 52:00 Well, kudos for keeping up the good work. Thank you and just sort of pushing this through. There was a question by Deb we're going back into metadata land. So we're in the credential example that you showed you would include ctds metadata within the alignment field within the badge. Or what if the credential is not an open badge?

Unknown Speaker 52:24 Okay, so could you just clarify for me, what is the CT DT CT CT, the L metadata?

Simone Ravaioli 52:31 Maybe that would you like to open your mouth? Thank you. Sure.

Unknown Speaker 52:34 That's a credential transparency description language. So it's an open schema for describing over 800 different characteristics of credentials.

Unknown Speaker 52:46 Okay, so I've never thought maybe I was, like, don't want to give you know, an answer. I wouldn't be correct.

Unknown Speaker 52:57 You don't need to speculate. That's fine. If you had an answer, it would be useful, but it's also a question that this group is is working on, right? Not all of the metadata is going to be in the VC. So what's what's the best way to reference it? response to that question. Where would you reference additional information about this credential?

Unknown Speaker 53:30 Like would you in traditional like, where I think you probably in the credential subject would be the good place. For any sort of disinformation probably. I guess you would want a polio specific property I guess for for that. That's just my opinion. Now discovering this, and you know, I haven't done any research on that and first time I hear, but it is so again, take this with a grain of salt, but I think well, yeah, the credential subject should be the right place for this sort of data, but you might want to do that at foot level too. Yeah. I guess it depends on on the final purpose. Yeah.

Unknown Speaker 54:21 So what we'll be discussing it in this group. And if you want to get together separately, and we can talk about cpdl and the registry. Thanks.

Simone Ravaioli 54:33 Thank you that thank you, Julian. Carrie, you had the floor.

Unknown Speaker 54:38 I'm sorry. I keep taking up the queue. So I this is in response to a Deb was asking about you. So Julian is part of a mess. 2.0 what some folks have been doing they've been leaking to data about credential that's hosting the cradle edge and finder in the alignment field. Right now, like that's what's been happening so far. I think what you may want to do is join us as they start talking more about Open Badges three point no and how that relates to verifiable credentials, because the structure of how you're doing might change a little bit even if you're not using the proof method. Because one of the relooking at is changing the structure so that the achievement data of the actual like metadata properties except for like issuer and issuance, which is verifiable credential, but like the description of the data, including any information that might be CTO and achievement. Type, which we have to figure out still will do inside the credential subject. Okay, so that was a year issuing this credential about this claim. And the claim is, this one is no match and it has this data inside of it for the Imagine anyway. So, but I think we're looking at that structure overall where the has achieved information would be inside the critical subject. So I will be happy to have you here.

Unknown Speaker 55:56 I'm interested to income.

Unknown Speaker 55:58 Great, thank you. I know the title isn't convenient, but

Unknown Speaker 56:01 it's okay for me.

Simone Ravaioli 56:04 Thank you. Thank you, Carrie. Julian, where are you based? If I may ask.

Unknown Speaker 56:08 I'm in France, as you heard for me, excellent.

Simone Ravaioli 56:13 Well, good. We're closer than most people are in this call, so I may need to do so. Okay, great. All right. Other questions in the queue? But this is kind of Last call for questions on blockcerts update.

Hearing none, I'd say we still have, say 10 minutes left in the call. Worthwhile probably switch into planning for 2022 and I want to share the results of the first round of serving. What I dropped in the chat here isn't linked to that Google sheet where we have started to list out the topics and I've added a temporary column that indicates where they came up in terms of priority after I'd say 14 people have voted. So that's they're free to review. But I think what can make more sense now is to open it up and ask whether we feel there is a need to extend the time to do an another round of voting and how or other ideas on how to then make sense of this indications. You know, how do we consider them mean? How do we tackle them? I mean, we have 20 Right now there so that's lot. We've had an offline file with Deb who I think had a very good way of also connecting them in a way so they're all connected, but how do we organise the work and and even just give it priority? And maybe and also suggest how we could further refine these priorities according to different dimensions, for example, you know, the weather, you know, what's the value for the users and the effort by product team. So maybe this is just a two by two matrix that we could use eventually. Any comments on this?

Really, so Did anyone feel like there was other topics that should have been added to the list? Alright, then. Thank you, for the floor is yours.

Unknown Speaker 58:57 Well, I was just gonna suggest that because this is a long list there's probably an order in terms of the things that are most foundational that people need answers to enable us to be able to move forward. And so I think that having an AN EFFORT value matrix can be useful, but then within that to say, the things that are the most valuable and that we can tackle fairly quickly, which are those also the ones that people need sooner rather than later in terms of dependencies. So it's just another way of thinking through how we tackle this list.

Simone Ravaioli 59:46 Thank you. Good. I see. Dimitri. Do you have your hand up? I see. There. I do.

Unknown Speaker 59:51 Yes. I was. I was wondering if the other question we want to ask, in addition to the order is if anybody wants to volunteer to present

Unknown Speaker 1:00:14 so obviously don't have to do an answer right. Now. But something for everybody to think about. If you you've just been dying to discuss a particular topic. Great. Yeah.

Simone Ravaioli 1:00:32 So what I'm hearing is from that, that a possible way to refine the work is to take these results in and add to them in a sense, you know, whether, you know, what is the effort required and the value and maybe continuing in a survey with any scoring high or low so we can give it some additional organisation? So that's one thing the other is maybe you know, out of those many topics, some require a bit more explanation. And so this is something that we can schedule so that it provides more clarity Carrie

Unknown Speaker 1:01:21 you are in the queue, right, and I think

Unknown Speaker 1:01:30 I just started want to find out that went to meet Jason, because we have this fondness of topics. But as far as learning, we can't tackle all of these, we all have full time jobs too. So I really need you all to chip in and help us out with the ones that are most important to you. And we're gonna figure out a way to do that probably. You haven't gotten this far yet. co chairs haven't talked about it. But what they do with the main TCG group is they create issues right and reaches me an example that already created one on the link resources and then discussions happen there. And then you know, that group can really like push that agenda that item forward. We can be here to support that and help with that. And we're interested in these topics too. But we're gonna need folks to jump on and really help drive the work for this too, and do the documentation and presenting and even if you don't know anything about it yet, it's a way to learn about it, and gain support and knowledge from other folks that we work with.

Simone Ravaioli 1:02:27 I think that's a great point. We don't want to lose sight of those issues, even if they are not, they don't make the top 10 or top five. So just spinning, spinning off issues on GitHub, maybe a good way to maintain a conversation on them and we're happy to reprioritize you know, if need be based on that. So, wait, feel you're in the queue.

Unknown Speaker 1:02:53 Yeah, just just to follow up. Can you hear me?

Simone Ravaioli 1:02:56 Yes. Hello. Yes, we can hear you okay.

Unknown Speaker 1:03:00 Just to follow up on Carrie. Just to follow up on Karen's comment, I'll make an entry in the issues list for the Verify credential, endorsement. Credential approach that we're discussing, trying to pull together and perhaps that can be a topic for this group with collaboration between others in the future.

Unknown Speaker 1:03:26 Thanks, thank you.

Simone Ravaioli 1:03:29 Alright, looks like I mean, we as chairs, as you know, may take this up and then get back to you on and say, follow up survey, you know, scoring just priorities as they are in a more articulate way and you know, opening up the issues, the thanks for you for taking, taking that on. We want to make good use of your feedback. And also make sure that we're all on the same page. We understand what these themes are. Yeah, great, thank you feel feel saying it's certainly everyone in this chat is welcome to, you know, to provide some help. There due to adding issues on the GitHub with two minutes to go, sorry for the technical glitches here. Are there any other things I'm also looking at my co chairs here? That we forgot to cover with? We're here

all right. It doesn't look like there's anything outstanding at this stage. You'll hear from us in email. Thanks a lot, Julian. And the Highland team for the work for the update but also more importantly, thank for the work you were doing on supporting blockcerts. And for the presentation here, we look forward to have you engage in this group in the future and thanks everyone who contributed the survey that's important if you have not done so please go in again and squeeze in your your votes that we can review those as well. With that said, All right, thank you. Thank you. Thank you very much. We'll see you next Monday.

Unknown Speaker 1:05:26 Thanks, everybody. Well, thank you thank you like

Unknown Speaker 1:05:32 recording has stopped

Transcribed by https://otter.ai