- From: CCG Minutes Bot <minutes@w3c-ccg.org>
- Date: Thu, 04 Jul 2024 19:14:58 +0000
- To: public-credentials@w3.org
Thanks to Our Robot Overlords for scribing this week! The transcript for the call is now available here: https://w3c-ccg.github.io/meetings/2024-06-24-vc-education/ Full text of the discussion follows for W3C archival purposes. Audio of the meeting is available at the following location: https://w3c-ccg.github.io/meetings/2024-06-24-vc-education/audio.ogg A video recording is also available at: https://meet.w3c-ccg.org/archives/w3c-ccg-weekly-2024-06-24.mp4 ---------------------------------------------------------------- VC for Education Task Force Transcript for 2024-06-24 Agenda: https://lists.w3.org/Archives/Public/public-vc-edu/2024Jun/0003.html Organizer: Kerri Lemoie, Simone Ravaioli, Dmitri Zagidulin Scribe: Our Robot Overlords Present: Ildiko Mazar, Dmitri Zagidulin, Naomi, Stuart Freeman, Alex H, Andy Griebel, TallTed // Ted Thibodeau (he/him) (OpenLinkSw.com), Geun-Hyung, Michal Jarmolkowicz, Eugene, Phil Barker, Juan Caballero, Rajesh, Keith Hackett, Mahesh Balan - pocketcred.com, Eric Shepherd, PL/T3, Nate Otto Our Robot Overlords are scribing. Juan Caballero: Can you hear me. Juan Caballero: Just in case anyone is talking I cannot hear anyone. Juan Caballero: Thank you so much yes that's reassuring to know that I have no issue with the audio thank you. Dmitri Zagidulin: Welcome everyone uh I think I was I was muted. Dmitri Zagidulin: All right. Dmitri Zagidulin: Welcome everyone to the weekly uh VC Edge call. Dmitri Zagidulin: These calls are recorded anyone can participate in the calls you do not need a w3c account however to make any substantive contributions. Dmitri Zagidulin: Uh to any of the specifications you do have to be a member of the ccg and sign uh IPR agreement so if you haven't done that and you want to contribute please let us know reach out to myself or to ildo or to Simone and we'll set you up. Dmitri Zagidulin: Uh The Columns are recorded and transcribed uh Auto transcribed and the. Dmitri Zagidulin: Available at the ccg uh GitHub repo. Dmitri Zagidulin: We use a queuing system so uh click on the raise raise your hand icon in uh the jitsui or type Q Plus in the chat and it'll uh it should queue you up. Dmitri Zagidulin: Let's uh let's do introductions and reintroductions does anyone new here and would like to introduce themselves uh go ahead and type Q Plus or raise your hand. Dmitri Zagidulin: Or if anybody hasn't uh introduced themselves in a while uh go ahead Juan. Juan Caballero: Yes good afternoon Juan carvio here from C and being part of also the neon team that is working on on exploring the utilization of wallets within the educational sector. Juan Caballero: I work as an aspiring designer architecture manager within the couch University in Saudi Arabia. Dmitri Zagidulin: A fantastic welcome we're glad to have you here do you mind um. <ildiko_mazar> Welcome, Juan. Dmitri Zagidulin: Pasting in chat uh a URL to to the university or to your projects uh. <juan> www.kaust.edu.sa Dmitri Zagidulin: Anyone else uh want to reintroduce introduce or reintroduce themselves if you haven't uh spoken up. Dmitri Zagidulin: Thank you thank you Juan I see the the URL. Dmitri Zagidulin: All right uh let's do announcements and reminders uh. Dmitri Zagidulin: Does anybody have uh community events that they want to remind people of. Dmitri Zagidulin: I know we have iaw coming up in the fall closer to Halloween. Dmitri Zagidulin: Uh I know we've got badge Summit and uh Boulder Colorado coming up uh later in July. Dmitri Zagidulin: Any other uh relevant conferences uh. Dmitri Zagidulin: That I'm. <ildiko_mazar> Here in Europe we have a big international conference called Online Educa Berlin: https://oeb.global/ Dmitri Zagidulin: Okay uh so in that case let's uh let's move on to our we go mentions that there's the European conference in chat online educa Berlin. Dmitri Zagidulin: Uh sounds really interesting I wish I was there. Dmitri Zagidulin: Any any other announcements uh while we. Dmitri Zagidulin: What people are connecting. Dmitri Zagidulin: All right uh so the main topic of this week's call is just free form it's uh office hours uh and uh open Agenda which means. Dmitri Zagidulin: Free to ask questions share how your projects have been going what you've been encountering if there's any roadblocks you've been running into. <ildiko_mazar> It's only in November, but registration is now open with early bird rates available. Digital credentials are one of the sub-themes. Dmitri Zagidulin: Or what's on your mind in general in the area of. Dmitri Zagidulin: Uh educational credentials technology and so on. Dmitri Zagidulin: Or report outs from any interesting conferences uh you've been to. Dmitri Zagidulin: I can start on on our end uh on so I wear several several different hats aside from uh co-chair of uh this excellent task force here. Dmitri Zagidulin: Uh so. Dmitri Zagidulin: I work with mostly with mit's digital credentials Consortium. Dmitri Zagidulin: Uh where uh I'm I'm technical architect for an open-source Mobile Wallet. Dmitri Zagidulin: A uh issuer Suite. Dmitri Zagidulin: Uh so UI and back-end services and a verifier Suite in addition uh I also help out the Arizona State University team on very similar uh kind of things so I work on uh their corner pocket. Dmitri Zagidulin: Mobile Wallet mobile credential wallet uh the issuer services they're verifiers and so on and I also help out. Dmitri Zagidulin: Um Phil Long with various projects uh for Chamber of Commerce for T3. Dmitri Zagidulin: And and so on uh so we're doing. Dmitri Zagidulin: So with MIT and and ASU continuing work on uh just adding features to the wallet uh and so we've got a open source Mobile Wallet written in um JavaScript in react native. Dmitri Zagidulin: And we're. Dmitri Zagidulin: We're preparing it to. Dmitri Zagidulin: Handoff stewardship to the open Wallet Foundation uh which if you're not familiar with uh is uh part of the Linux Foundation uh it's a Linux Foundation project and its goal is not to write specifications it leaves that to the other species like ITF and w3c and ISO and so on. Dmitri Zagidulin: But it's it's mandate is to. Dmitri Zagidulin: And uh and curate a set of open-source libraries libraries and wallet implementations. Dmitri Zagidulin: Uh also it it work it's working on a sort of best practices and architecture and security sort of paper for wallets but but in general it's much like many other Linux Foundation projects. Dmitri Zagidulin: Like like the Apache foundation and so on. Dmitri Zagidulin: It's it's goal is to uh sort of gather and Steward uh open source either. Dmitri Zagidulin: Full-on wallet implementations or component libraries so we've uh. Dmitri Zagidulin: We've been talking to the open world foundation and are in the process of migrating the. Dmitri Zagidulin: Uh the learner credential wallet which is what the DC wallet project is called over to open Wallet Foundation. Dmitri Zagidulin: Uh so we'll we'll MIT will continue sort of uh uh. Dmitri Zagidulin: Leading the software development uh as we can we're fairly small uh development team. Dmitri Zagidulin: But we're by moving it to the Linux Foundation we're inviting broader participation hopefully the project will be interesting to uh. Dmitri Zagidulin: Other teams as well. Dmitri Zagidulin: So if anybody has uh questions about the wall Foundation uh please feel free to ask in chat or queue up I'm probably not the best person to uh answer them but we can we can get your we can forward them to the open world Foundation folks and and. Dmitri Zagidulin: Uh get your questions answered. Dmitri Zagidulin: The stuff that we've been working on is uh. Dmitri Zagidulin: Yeah so basically cleaning house for uh. Dmitri Zagidulin: B batching up uh various like warnings and errors and uh Tech debt uh in order to invite guests over for learning credential wallet that's handing off the. Dmitri Zagidulin: Uh open source wallet. Dmitri Zagidulin: To the broader Community uh and aside from that have been working on um. Dmitri Zagidulin: Issuer Suites uh to um Consortium member universities. Dmitri Zagidulin: We we've kicked off the. Dmitri Zagidulin: Uh quote unquote trust Registries project I think quote unquote because trust Registries is. Dmitri Zagidulin: Kind of a a controversial and um not the best name. Dmitri Zagidulin: In in our credential Community for the thing that we're studying and what do we mean by trust registries. Dmitri Zagidulin: So anybody dealing with credentials in the. Dmitri Zagidulin: Space but really in any space this this absolutely. Dmitri Zagidulin: Covers any sort of credentials or anytime you sign an object digitally. Dmitri Zagidulin: Because what's a signature right you're you're signing it essentially with a cryptographic key or with another a baked identifier like it did at decentralized identifier. Dmitri Zagidulin: And which which brings all sorts of useful properties. Dmitri Zagidulin: But the major challenge with digital signature is that they identifiers opaque when you're receiving a credential you don't know who the key belong to that you signed it with. Dmitri Zagidulin: And so you must necessarily. Dmitri Zagidulin: Have access to directories to lists of known issuers so you can say okay this key belongs to this University or this key belongs to uh this particular company or even an individual person right we're in the area of Education we're interested in. Dmitri Zagidulin: Self issued credentials which presents its own. Dmitri Zagidulin: It's kind of interesting challenges in terms of. Dmitri Zagidulin: If we can get issuers. Dmitri Zagidulin: Companies and organizations on a list what about uh natural persons so we can we can talk about that what some of the techniques there. Dmitri Zagidulin: But anyway so. Dmitri Zagidulin: Regardless of what you do with credentials regardless of what technology stack. Dmitri Zagidulin: What sort of uh. Dmitri Zagidulin: Specifications you're using. Dmitri Zagidulin: You're going to need to have access to these directories to these lists of known issuers and that's what we mean by trust registry. Dmitri Zagidulin: Uh at at Baseline it is a list mapping opaque identifiers to known entities of course that that's not all that is that it can be. Dmitri Zagidulin: Industries can have other Advanced um features such as they can carry. Dmitri Zagidulin: About what kind of verification steps were taken. Dmitri Zagidulin: Get the entity on that list uh meaning we performed full uh personal kyc the know your customer. Dmitri Zagidulin: Terminology from from Financial world. Dmitri Zagidulin: We can say that okay this is our governance framework and in order to get it on this list you have to meet these criteria and this is these are the verification steps we're going to take. Dmitri Zagidulin: All of these things to raise the confidence level of the verifier. Dmitri Zagidulin: When they're hopefully will be accepting your credentials in the future. Dmitri Zagidulin: Aside from a base level this is a list of identifiers. Dmitri Zagidulin: These Registries have uh lots of other useful features. Dmitri Zagidulin: To the ecosystem and so uh credential engine and uh digital credentials Consortium and many other many other groups uh have have. Dmitri Zagidulin: Joined into this uh 1 year uh research projects uh or or practically research project where essentially we're comparing. Dmitri Zagidulin: Um half a dozen to a dozen. Dmitri Zagidulin: Testing or in progress specifications. Dmitri Zagidulin: Because all these groups working with uh credentials Registries have have run into this problem. Dmitri Zagidulin: Registries that means we need a data model and sometimes we need a protocol or an API. Dmitri Zagidulin: Switch to interface them with. Dmitri Zagidulin: And that leaves implementers like us. Dmitri Zagidulin: With lots of questions okay I'm entering into this world I have a wallet or I have a verifier or I have an issuer which 1 of these trust registry specifications do we use. Dmitri Zagidulin: And there's there's already uh a handful of choices so 1 of the 1 of the goals of This research project is we'll be uh simply comparing the specifications and trying to make sense of them these are the trade-offs that they're making uh this is a data model specification this is uh just an API these are the affordances they have for uh how to specify governance. Dmitri Zagidulin: How terms of service how to get on that list and how to get off that list and how to. Dmitri Zagidulin: All the information is outdated how do I simply change. Dmitri Zagidulin: The the information right all of these things need to be. Dmitri Zagidulin: Thought through so the trustee project will be comparing the specifications. Dmitri Zagidulin: Selecting 1 1 or 2 and. Dmitri Zagidulin: Standing up and implementation a a sample registry that we can operate for a while uh use in our wallets and issuers so that we can report uh knowledgeably that okay so this is what it's like to operate 1 of these these These are the downsides these are the the caveats. Dmitri Zagidulin: So as a sort of service to the community. Dmitri Zagidulin: So that's uh that's 1 of the projects but. Dmitri Zagidulin: Uh does anybody else have uh. Dmitri Zagidulin: The discussion uh things that they're working on they want to share. Dmitri Zagidulin: Hop on the list Naomi please go ahead. Naomi: Hi there hi Dmitri um you said some interesting things and uh we'd love to stay in touch with you not only about the trust Registries I think that that was 1 thing um velocity had been inquiring around earlier um in terms of certain lists that we've seen that we were curious about how we could incorporate them into the network so that's definitely and of course we have our own tools so we definitely love to stay in touch with you on that um another um thing that you mentioned of Interest was the wallet the open wallet and um working with the Linux Foundation I'd love to follow up with you on that as well velocity had a couple components of its organization that we were interested in connecting with Linux and maybe there is some dovetailing of interests in terms of your work with um the wallet and uh Linux foundation so um would love to follow up with you on those 2 points but in addition you know you started this um opening up you know the announcements recent events and our CTO on. Naomi: race did. https://www.velocitynetwork.foundation/empowering-self-sovereign-identity-with-trusted-credentials-exploring-velocity-network-checks Naomi: At 1 of the diff um events specifically on uh empowering self Sovereign identity with trust credentials um I did post it on LinkedIn it kicked off on Thursday um but what I'll do is I'll put in the chat at least some of his recent writings and you know everyone on this call may have ideas or thoughts or feedback on that so I'll put um some of his recent writings and blogs in the chat for anyone who kind of wants to read velocities latest opinion um and then the other thing I'll put in the chat again is um. <ildiko_mazar> Thanks, Naomi. https://www.velocitynetwork.foundation/wp-content/uploads/2023/10/Recommended-LER-technical-and-functional-requirements-Sep-2023-.pdf Naomi: Um our Lear uh infrastructure requirements and this was kind of an early Doc and it was retrofitted I think for the Lear term being used in the US I think otherwise um it's almost like a Manifesto of some of velocities core beliefs in terms of why it works the way it works I just think that as a return this corner and we're starting to invest in various products that educate us on the notion of trust it might be a helpful document for everyone to have and and and opine on so um I just stuck those 2 in the chat and um we have a lot going on in velocity Network Foundation as well if anyone's interested in hearing more specific updates um I I believe we'll be publishing like a a quarterly or a 6-month kind of outward-facing newsletter on some of the big changes in velocity in the next week or so that hopefully everyone will take a look at and um be able to keep up on things happening in velocity network with all our members. Dmitri Zagidulin: Thank you so much Naomi I really appreciate it. Dmitri Zagidulin: All right well so while while people are uh searching their minds and um. Dmitri Zagidulin: Hopefully getting on the queue. Dmitri Zagidulin: I want to introduce a topic that's always near and dear to my heart which is self issued credentials. Dmitri Zagidulin: Uh I think we've we've talked a bit about we presented on. Dmitri Zagidulin: On this topic I think with Phil Long on the on the subject on the work that we're doing with link credentials author. Dmitri Zagidulin: In the future on an open source uh resume author demo. Dmitri Zagidulin: Uh but the reason I bring these up is that. Dmitri Zagidulin: We've essentially got these 2 Worlds these 2 aspects of our of our credentials uh we have the in the classical institutionally issued credentials. Dmitri Zagidulin: Which what is. Dmitri Zagidulin: Does that mean it means. Dmitri Zagidulin: But it's an organization our company doing it and they're likely to be on a known issue or list somewhere. Dmitri Zagidulin: Uh whether it's by the government by the local Chamber of Commerce by uh some sort of uh third party. <ildiko_mazar> :+1: Dmitri Zagidulin: Directory something like that right but then we have this whole other world of natural persons who are not likely to be on the list and in some jurisdictions like the European Union are specifically not legally allowed to be kept on list for privacy and and you know very very good legal reasons. Dmitri Zagidulin: Still want people to be able to. Dmitri Zagidulin: Selfish credentials and and for various reasons right so the first time the subject is brought up to people unfamiliar with the verifiable credentials uh the question immediately comes up. Dmitri Zagidulin: Why right what what business value what usefulness do we have in. Dmitri Zagidulin: Issuing a claim of some sort. Dmitri Zagidulin: The answer to that is always well just like in the current non- digital credentials world right the the the business value of it greatly depends on the person greatly depends on their qualifications capabilities reputation in the community and so on. Dmitri Zagidulin: Ah I see we've got mahash on the Queue so I'll pause here Mah go ahead. Dmitri Zagidulin: But if you're talking you might be muted. Dmitri Zagidulin: He's showing us unmuted here though. Dmitri Zagidulin: But we can't hear you otherwise. Dmitri Zagidulin: Or at least I can't hear you it could be a network partition. Dmitri Zagidulin: Uh if somebody else can hear my hash let us know in chat. <mahesh_balan_-_pocketcred.com> I will rejoin <pl/t3> Nope no voice from Mahesh Dmitri Zagidulin: Okay so my my hash is going to reconnect all right. Dmitri Zagidulin: So yeah so so sufficient credentials. Dmitri Zagidulin: The the value of this credential uh greatly depends and 1 of the things that we're we're claiming in in starting this project with Phil Long and others of uh. Dmitri Zagidulin: I'm making more widely available tools for natural persons to. Dmitri Zagidulin: Easily create verifiable credentials 1 of the things that that we're we're claiming and and are investigating as part of this project is. Dmitri Zagidulin: Our thesis is that. Dmitri Zagidulin: The value of self issue credentials can be greatly increased. <ildiko_mazar> I am one of those individuals who has genuine interest in self-issued credentials as a lifelong non-formal learner couldn't expect an Agent to formally/informally issue credentials proving one's knowledge, skills and competences. Dmitri Zagidulin: With 2 things. Dmitri Zagidulin: 1 Is evidence. Dmitri Zagidulin: Let's say this is a uh. Dmitri Zagidulin: Stuff is should claim of skills or uh. Dmitri Zagidulin: Or a an employment history item right the things that make up potentially make up a resume. Dmitri Zagidulin: The veracity and the value of that can be greatly increased with just adding to the credential. Dmitri Zagidulin: I'll link to uh a video where you demonstrating that skill a link to uh an external non digitally science certification a uh. Dmitri Zagidulin: A code repository right literally evidence of the claim that you're making and then the second thing that that that we think greatly increases the value of the claim is of course. Dmitri Zagidulin: Recommendations endorsements uh comments on the claimed by other people and and in the verify the credential world we have these general purpose mechanisms uh that that we're hoping to see wider use so 1 of the things that we're we're trying to demonstrate is the user interface and the data model. Dmitri Zagidulin: On 1 approach and how this can be done uh by Sima hash has um. Dmitri Zagidulin: Uh rejoined us so go ahead Mash. Mahesh_Balan_-_pocketcred.com: Can you can you hear me let me. Dmitri Zagidulin: Yes we can hear you go ahead yeah. <ildiko_mazar> Using common standards to convert what would normally be on a single-language un-structured CV, can this way become a multi-lingual, machine readable verifiable cradential. Mahesh_Balan_-_pocketcred.com: All right sorry um um I just um I think I I missed the there was probably uh presentation on ASU pocket and the tln network maybe 2 weeks back uh and uh I was wondering if it's possible for you or somebody else uh who was there to give a quick summary of uh where the progress is I'm very interested in that particular project uh it's right uh right in my backyard thank you. Dmitri Zagidulin: Thank you Mahesh. Dmitri Zagidulin: Who who wants to give a summary of uh the ASU presentation. Dmitri Zagidulin: We should have a recording uh up in the minutes. <ildiko_mazar> @Mahes, we are sill to deliver the Pocket presentation. Dmitri Zagidulin: Ah okay so so uh mentioned so the ASU pocket presentation uh was scheduled for previously but had to be moved uh due to speaker. Dmitri Zagidulin: So that's coming up so you haven't missed anything yet. Mahesh_Balan_-_pocketcred.com: All right great thank you so much yeah. Dmitri Zagidulin: Any other questions any questions about the uh. Dmitri Zagidulin: Notion of self-assured credentials on. <ildiko_mazar> The recording of the LER session can be found at: https://w3c-ccg.github.io/meetings/2024-06-03-vc-education/ Dmitri Zagidulin: Uh have any of you either implemented a wallet or have used the wallet that has included tools to create self issued credentials you'll go ahead. <pl/t3> Ildiiko self-issue credentials are simiilarly 'sealed' cryptographically when they are created. Dmitri Zagidulin: Thanks so the account and yeah it's a it's a tough technological problem how do you electronically sign how do you seal these self issued credentials again not that it's difficult to sign. Dmitri Zagidulin: Uh to to create a signature it it's 1 function call right. Dmitri Zagidulin: Signing part is Trivial technologically what's not trivial what what I suspect is referring to and and what 1 of the things that uh we're investigating with Phil long is. Dmitri Zagidulin: Once it's signed how do you get the key or the identifier in a directory somewhere how do you. <ildiko_mazar> EDCs must be signed by an eIDAS compliant electronic seal (can't be sealed by an individual's electronic signature). It's a trust issue. Dmitri Zagidulin: How do you hook into the known issue or mechanism. Dmitri Zagidulin: How do you. Dmitri Zagidulin: To identify the person. Dmitri Zagidulin: And again in the case of uh. Dmitri Zagidulin: Classic universities companies uh other providers. Dmitri Zagidulin: Yeah so as a Zelda mentions in um in chat in the European Union model. Dmitri Zagidulin: Uh credentials must be signed by an EIT compliant electronic seal which means. Dmitri Zagidulin: What does that mean it means that the issuer has to be on a registry on a list this year has had to have gone through. Dmitri Zagidulin: Uh organ. Dmitri Zagidulin: Organizational identity verification. Dmitri Zagidulin: So again and since those uh identity verifications are for organizations and and cannot be uh sort of by legislation uh with natural persons what do we do about people. Dmitri Zagidulin: That there's let's talk about what what the options are and then uh some guesses on what the. Dmitri Zagidulin: Well some of the solutions could be alright so. <ildiko_mazar> And how can third party viewers trust the content of credentials if they contain claims about the issuers themselves. Dmitri Zagidulin: The the trivial option is of course. <nate_otto> My project ORCA enables self-claiming of credentials within the governance rules of a community, but the community (currently using did:web) is the issuer and signer. I haven't yet seen any wallets that are really enabling self-issuing and signing with a key associated with the same ID (e.g. did:key) that they use to receive credentials. I'm sure we'll see some self-issuing features in wallets eventually. Dmitri Zagidulin: Yeah so as I mentioned in chat uh how can third-party viewers trust the contents of the credentials if they are self claimed if their claims about the issues themselves. Dmitri Zagidulin: Yeah that's that's the key question how do we get the signature on a directory somewhere or at least how do we identify who the signer was and then how does the verifier trust the claims. Dmitri Zagidulin: And there's a couple of options uh in terms of uh increasing Trust. Dmitri Zagidulin: 1 Is again evidence. Dmitri Zagidulin: You don't just like in the non-credentialed. Dmitri Zagidulin: You don't. Dmitri Zagidulin: Don't take people's uh word just at face value you also asked to provide evidence evidence of the skill evidence of the project completion of the output. Dmitri Zagidulin: And in addition. Dmitri Zagidulin: We have these well-known mechanisms of letters of recommendation letters of endorsement uh. Dmitri Zagidulin: We have the. Dmitri Zagidulin: Uh references available upon request at the end of a resume which realistically means that somebody in HR is going to call up your uh previous manager previous teammate and and and just try to. Dmitri Zagidulin: Confirm what you claimed again the the reason that. Dmitri Zagidulin: Self issued claims are valuable uh is really easy to demonstrate a resume is nothing but a collection of self-assured claims. Dmitri Zagidulin: Is valuable absolutely that that is how so much of our employment world. Dmitri Zagidulin: Uh rests on so what we're what we're sort of competing against what we're what we're racing against is can we provide in these digital signatures in these digital credentials can we provide equivalent or hopefully more value than a regular paper resume uh go ahead elico you're on the queue. Dmitri Zagidulin: Yeah excellent point thank you for bringing that up uh thank you for bringing that let's talk about that so. Dmitri Zagidulin: As as a brief side note about is it better than a paper resume uh 1 of my favorite essays uh from uh I think early 2000s Was Won by uh Joel spolsky who is incidentally um the founder of the company that created stack Overflow as well as Trello and and a couple other things uh. Dmitri Zagidulin: Great guy wrote a lot about uh issue tracking and uh managing Developers. Dmitri Zagidulin: 1 Of the. Dmitri Zagidulin: See what we said. Dmitri Zagidulin: Is in developing an issue tracker you're not just competing against jira or Microsoft or other people that are providing the software you're competing against. Dmitri Zagidulin: An open notepad file. Dmitri Zagidulin: You're competing against uh just a simple file in a text editor I always think about that uh in in the area of credentials especially when employment and ress I always think just like when creating a software are you better than. Dmitri Zagidulin: Open text file in terms of jotting down notes Here Again the question is can we be better than a paper resume and as Elder Co points out. Dmitri Zagidulin: The 1 of the low-hanging fruit 1 of the sort of holy Grails of this whole field is not just the digital signature the value of verifiable credentials whether institutionally issued or self issued is not just a signature it's not just the reputation that the issuer brings to the table it's the fact that it is structured data what do we mean by that so. Dmitri Zagidulin: Uh if you look at the world of data uh as a developer there's essentially only 2 kinds of data. Dmitri Zagidulin: Raw text or unstructured text. Dmitri Zagidulin: This is what you. Dmitri Zagidulin: Get when I don't know you write an email or you scan in an ebook or you scan in a resume right you're applying to a job portal and it's like. Dmitri Zagidulin: Upload your resume in word or PDF format there that's on structured data it's uh long paragraph of text meant to to be for human consumption. Dmitri Zagidulin: But what does an analytical system. Dmitri Zagidulin: What does machine readable code. Dmitri Zagidulin: What can go do with that and of course now these days we have large language models we have uh AIS with with all of their benefits and um and drawbacks and and part of the reason why there is such hype about AI is that they let us. Dmitri Zagidulin: Text free form text. Dmitri Zagidulin: Scanned in resume into hopefully. Dmitri Zagidulin: This is still being demonstrated a number of experiments the hope is that we can turn that unstructured text into structured data what do we mean by that. <ildiko_mazar> Very-very true. And while I fear that we won't see many self-claimed degrees floating around :-), in today's world where so much learning is happening non-formally, documenting skills in ANY way, has great advantages (without legally binding signatures). Dmitri Zagidulin: We mean machine readable records we mean uh fields that are defined uh we we're all familiar with these records the these are. Dmitri Zagidulin: Uh anytime you fill out a form or or use a spreadsheet uh although form is slightly better example than a spreadsheet right uh this is where columns and fields are defined the data structure is known and the receiver knows what to do with it it knows that this field is a name this field is um the name of the place that you worked and this field is how long you worked there that sort of thing. Dmitri Zagidulin: In general 1 of the 1 of the main value propositions of verifiable credentials is. Dmitri Zagidulin: Usually the issuer or the implementer of the software that was used to to create this thing. Dmitri Zagidulin: Did the the cognitive work and the user interface the hard ux work of. Dmitri Zagidulin: Turning this record into. <pl/t3> person identity - notarization is the personal identity verifidation. Dmitri Zagidulin: Something that just meant for for human ball into structured data which means that it's actionable hopefully if you have the right system it's actionable uh now let's see there's some good good comments in chat um. Dmitri Zagidulin: Sounds very true. Dmitri Zagidulin: We won't see many self-claimed degrees. Dmitri Zagidulin: Much learning is happening non formerly yeah absolutely absolutely. Dmitri Zagidulin: Uh Phyllis pointing out that uh. Dmitri Zagidulin: In in terms of uh personal identity in terms this is the issue that we're talking about earlier. Dmitri Zagidulin: How do we know who the credential is from 1 of the very powerful Tools in this uh toolkit is. Dmitri Zagidulin: Uh infrastructure right like we're all hopefully familiar with no Republic uh which is if you think about it an interface between the formal world of government and bureaucracy and the messy human. Dmitri Zagidulin: It's no no much like LM is a mechanism to turn on structure text into hopefully structured in a way notaries are a mechanism. Dmitri Zagidulin: Or at least partially. Dmitri Zagidulin: Selfish should claims into witnessed and verified claims and giving them uh a set of structure and formality. Dmitri Zagidulin: Now let's see uh there's also an excellent comment from Nate. Dmitri Zagidulin: That I think I may have missed as the transcriber was scrolling the documents ah Nate was talking about project Orca enabling self claiming of credentials within the government's rule of a community. Dmitri Zagidulin: Uh I haven't seen any wallets ah scrolled off uh Nate uh please feel free to hop on the. Dmitri Zagidulin: On the Queue we'd love to hear about uh your project and and it's a a really good point that uh there is. Dmitri Zagidulin: It's not a binary ahnaf between uh traditional institutions and and people we have many other nodes in the middle such as a a given Community whether it's a uh. Dmitri Zagidulin: AA club and Association uh and so on I'm sure you can explain it better. Dmitri Zagidulin: But it's an excellent point. Nate Otto: Um so Orca is an open source piece of software that aims to be a badge platform for communities of practice whatever that community of practice is it could be uh sports team who primarily wants to have people show up to the pickup practices um or it could be a department of a university you know engaging in formal education and they can set up their achievements with whatever governance rule they like in terms of what it takes to earn them um and self claiming is is available so it's interesting to see the presence of many possible communities whether they're these tiny little things that may fly as spin up for a very small purpose or storied institutions that um last for hundreds of years. Nate Otto: Um as intermediaries in the space that exists for the purpose of enabling people to self claim um the skills that they have and then get peer validation and endorsement from the other people in the community um it's 1 possible approach here I think it's also compatible with sort of pure self claiming where maybe you have some key material in a wallet somewhere and you're creating credentials signed by and to the same uh credential subject I haven't seen a whole lot of that happening yet although I'm sure that um we will as there are destinations for. Nate Otto: To actually go and be turned back into real insights and real um. Nate Otto: Uh actionable you know intelligence um like you said. Dmitri Zagidulin: Thank you so much Nate yeah these are really excellent point uh from from Milo and Nate uh it'll go point that uh in addition to the difficulty of uh. <naomi> hi, I struggle with the level of effort required for self-reported or peer-attested credentials given that AI can make us less reliant on structured data. That said, love the idea of community feedback being mobile! We do have to be better than a piece of paper. Dmitri Zagidulin: Identity verification don't forget about the value proposition of structured data uh even even if self issued and then Nate bringing up several excellent points such as. Dmitri Zagidulin: 1 There's communities and and groups and all sorts of other issuers that could be potentially uh. Dmitri Zagidulin: Business value both of reputationally and and organizationally so so that's an excellent point and then uh as Nate mentioned um. Dmitri Zagidulin: We still shouldn't forget about individual people right we've got these wallets and what do wallets do they manage cryptographic Keys which means we have. Dmitri Zagidulin: Much of the infrastructure for people to manage their own keys. Dmitri Zagidulin: But that last mile that the difficulties of course had to identify them how do we get them on a registry that Maps a key to a person uh. <pl/t3> I think those of us interested in self-issued credentials are trying to align existing formal organizational practices that in HE speak are often captured under the practices of accreditation. How do we enable individuals to have forms of third-party assessment of the claims. Dmitri Zagidulin: 1 1 Of the possible solutions was already mentioned by Phil Long and that is notaries right you you can use uh notaries as an active as a trust anchor uh where notaries can issue a notarization credential that says okay I checked this person's. Dmitri Zagidulin: And this key uh allegedly belongs to Demetrius aadan right so that's that's 1 option and then. Dmitri Zagidulin: Uh the other option that I uh. Dmitri Zagidulin: So that is my personal suspicion that may play a role is of course address books. Dmitri Zagidulin: Meaning we all carry in all of our phones. Dmitri Zagidulin: Essentially I trust. Dmitri Zagidulin: Right that we use to filter incoming phone calls that we this this person calling ah okay it's uh it's Grandma it's my car dealership whatever right. <ildiko_mazar> How would this look like in a digitally signed credential? The self-signed cred is re-signed by the notary (envelope in envelope style)? Dmitri Zagidulin: They're identify they're opaque identifier the phone number is in my registry which is the address book right but the nice thing about uh even today's uh phone address books is that they have a bunch of fields emails uh address phone number and they all have we don't really use them it's not really used much but the capabilities there they all have fields to record identifiers and keys and so I definitely want to see more done with that but we've got uh more interesting conversation in chat. <pl/t3> Personal recommendation graphs can become the equivalent of address books Dmitri Zagidulin: Naomi mentioned level of effort. Dmitri Zagidulin: Right uh level of effort required for self-reported or peer tested credentials. Dmitri Zagidulin: Uh an excellent point pill Phil mentions that uh. Dmitri Zagidulin: Those of us interested in South Asia credentials are trying to um align existing formal organizational practices. Dmitri Zagidulin: And then uh elico. Dmitri Zagidulin: How would this look how would this look like in a digitally signed credential the self-signed credential plus the notary uh excellent question and and we hope uh Phil and I hope to present on that in the future and I see. <mahesh_balan_-_pocketcred.com> What about x509 certificate extensions feature - I heard something about it in the last IIW Dmitri Zagidulin: Uh Phil Miku go ahead so. <dmitri_zagidulin> @Mahesh - sure, so, 509x is just another signature / 'sealing' technology, definitely usable for credentials PL/T3: Yeah this is very helpful conversation actually for for those of us um uh like Nate and ourselves and others that are looking at how we try to transpose what institutions and corporate um in nonprofit art work corporate um mechanisms of impose of of uh either professional development or educational uh acquisition of Knowledge and Skills are have been doing for for centuries frankly um to into a world which is more unstructured and uh and less uh opaque less transparent I should say um and 1 of the things I think that we're continuing to harp on that is trying to bridge this Gap Demetri already mentioned and that is the presidents of third-party evidence and assessments that an individual can offer in in support of the claim that they're in that they're self asserting um so that there is uh there. PL/T3: opportunity for. PL/T3: Entities whether they are um testing center or a in another individual with reputational value or other sorts of um of alignments that are more associated with uh formally structured issued credentials from an entity and organization that has at least in the higher education sense the Notions of accreditation that sort of check on the practices that are being used by which the acquisition is taking place to see if they conform to um things that are well known to um increase the probability that the acquisition claims is actually uh 1 that the person um is capable of to demonstrating in another in another context so I think that's where the the the issues are and any of us that are interested in this could the more sharing we do about possible Connections in this regard will help all of us and the last 1 that Dmitri mentioned was um the the context of the. PL/T3: your address. <nate_otto> How quickly could a verifier consume all of these forms of evidence? How much time will a reviewer take with each of individual's self-issued claims. Or is the relevant user imagined to be in the process of reviewing 300 resumes and unable to spend more than a minute on each one? <dmitri_zagidulin> EXCELLENT question, Nate PL/T3: And in effect we've been talking about the idea of a personal recommendation graphs that 1 might be able to uh develop that are associated with particular dimensions of 1's types of skills that 1 that that 1 is claiming so that there is the this the start of at least of a collection that um facilitates a third another independent third parties opportunity to judge that self-claimed um with a set of things um that is more than just a random individual picked so thanks. Dmitri Zagidulin: Thanks Phil uh Nate brings up another excellent question chat how quickly could a verify consume all of these uh forms of evidence meaning. Dmitri Zagidulin: The the the the promise of digital credentials is a sort of instantaneous uh verification process except it's not right so this is this is 1 of the. Dmitri Zagidulin: 1 Of the fundamental insights uh we're all starting to have as a community in in implementing these credentials is that. Dmitri Zagidulin: We we don't want to oversell it as an instant uh verification human judgment still has to get involved even if it is. Dmitri Zagidulin: Upfront human judgment on which trust Registries do we pick so that if an ident if a signature from that uh entity shows up then we know we can trust it but in so many cases and as nay points out especially in uh self issue credentials and evidence. Dmitri Zagidulin: We're still going to need human judgment which is incidentally uh why it's super important to make sure that your issuing and verify your protocols. Dmitri Zagidulin: Deeply asynchronous that there can be. Dmitri Zagidulin: Minutes and hours and days and weeks between a submission for verification and a response back right we need to make sure to build into our protocols uh this essentially humans being involved. Dmitri Zagidulin: Okay so we've got uh almost nearing time. Dmitri Zagidulin: And and we want to encourage uh everyone at the end of the call to uh re to reach out to us if you have a topic you want to speak on. Dmitri Zagidulin: Right so we're uh as always we're looking for uh presenters and speakers on these weekly calls. Dmitri Zagidulin: Uh so if you. Dmitri Zagidulin: If you have a. Dmitri Zagidulin: Project you want to showcase a topic you want to present on uh please um. <dmitri_zagidulin> dzagidulin@gmail.com Dmitri Zagidulin: Uh um email ildico email myself uh email Simone uh we'll drop some uh emails in the chat here here's mine and chair hers and so on uh please reach out to us. Dmitri Zagidulin: Okay but. <pl/t3> It might be useful to dig a little deeper into what attributes people associate with a well regarded 'brand' which institutionally issued credentials develop and are used as a proxy supporting trust in the assertions claimed. <ildiko_mazar> ildiko.mazar@nttdata.com Dmitri Zagidulin: In the in the 10 minutes or so that's left I I want to make sure uh we highlight the. Dmitri Zagidulin: 2 Things 1 is. Dmitri Zagidulin: Well I I I want to talk about. Dmitri Zagidulin: Again the value and the difficulty of structured versus unstructured data and fill in chat points out might be useful to dig deeper into what attributes people associate with well regarded brand. Dmitri Zagidulin: Uh that are used as proxy supporting trust in the assertions claimed yes absolutely I think that's that investigation is crucial to the whole field of verifiable credentials oh yeah and the other thing that I wanted to briefly address is just to answer all the questions uh what does. Dmitri Zagidulin: A notary or an endorser look like uh does a notary just uh add a signature to the existing uh credential or what uh and yes so adding a signature is always a solution but it's never the best solution uh. Dmitri Zagidulin: There fortunately verify the credentials have other mechanisms uh that are sort of the equivalent of uh the attached. Dmitri Zagidulin: I always find it helpful to think about uh real world processes on um. Dmitri Zagidulin: And then and then mimic them in credential uh scum orphism so to speak right think about what the procedure for a notary is uh I have a document. Dmitri Zagidulin: Know uh self attestation about something and then I attached I include another separate document from the notary that says I reviewed this and and my notary number is this and here's my sign uh I have reviewed the attached right so even in the real world we have this notion of attachments and hyperlinks. Dmitri Zagidulin: We say things like the above the attached document and so on uh and so similar. <ildiko_mazar> I'd also like to add to Phil's observation: at what level of data-richness we start seeing digital credentials as having high quality? We discuss this a lot in the ELM Team. Dmitri Zagidulin: Techniques can be used in the credential World by the notion of um. Dmitri Zagidulin: Links by the cryptographically secured or not uh attachments embedding and so on uh but that sounds like a great topic uh in general. Dmitri Zagidulin: For another presentation uh but super quick. Dmitri Zagidulin: It turns out that. Dmitri Zagidulin: That that turning. Dmitri Zagidulin: Unstructured text even if it is a text in somebody's mind uh into structured machine actionable is actually really hard what do I mean by that. Dmitri Zagidulin: Think about developing a uh free form tool to issue. Dmitri Zagidulin: Self issued verifiable credentials which which is exactly what we're working on with Phil the the link claims author and the resume author. Dmitri Zagidulin: Think about if you're managing a development team or you're a developer and you're sitting down and your task is uh come up with. Dmitri Zagidulin: John purpose tool or somebody can come and produce any kind of credential or but not just any kind of credential let's narrow it down let's say it's a skill assertion. Dmitri Zagidulin: This this tool is going to produce scale assertions. Dmitri Zagidulin: The the trivial sort of solution is okay you have a form uh and it's like your name name of the skill the description of the skill what steps did you did you have to take to to get it do you have any sort of supporting evidence right so that's the. Dmitri Zagidulin: Sort of the. Dmitri Zagidulin: Line you have these uh couple of fields but they're mostly unstructured because most of the value is still in the human readable text inside name of the scale. Dmitri Zagidulin: And and description of the skill. Dmitri Zagidulin: Why is the name of the scale unstructured uh because. Dmitri Zagidulin: People can refer to the same skill in in different languages right uh our our Dream as as developers as bureaucrats as as uh stewards of the ecosystem is always wouldn't it be great to link this skill to an existing skill. <nate_otto> "our dream as bureaucrats" (the name of my new album) Dmitri Zagidulin: Like say that the government has okay this is a directory of skills that we recognize. Dmitri Zagidulin: A dream is bureaucrats as the name of a new album it would be an excellent album um. <ildiko_mazar> 1. reference skills from a published framework Dmitri Zagidulin: That that's 1 of The Avenues to for us to investigate can we tie the skills or any sort of self assertions into published Frameworks uh and the answer is well trivially yes but. <ildiko_mazar> @Nate :-D :-D Dmitri Zagidulin: How to do it with the least friction and the least confusion on part of the user is an interesting topic that is worth a lot of Investigation. Dmitri Zagidulin: All right but we are nearing time uh yeah excellent points in chat uh elico reference skills from published framework associate volume to the learning show proof of skill acquisition including detailed description of the type Mo motive assessment yeah. <naomi> Thank you for the call! -- informative. Dmitri Zagidulin: Anyway so my my whole point was that. Dmitri Zagidulin: The main value proposition is also turns out to be 1 of the hardest things uh to do in this field which is. Dmitri Zagidulin: How do you make the tax machine actionable in structured hopefully you're tying it into Frameworks uh but there's not that many good Frameworks so we're all sort of on the lookout of uh. Dmitri Zagidulin: Frameworks to use and then how to present that workflow to the user do we. Dmitri Zagidulin: Present a giant pulldown drop-down list. Dmitri Zagidulin: That says okay uh here's the 500 skills recognized by the state of Minnesota Select 1 right do do we do that or can we do something better. <pl/t3> In the self-issued credential world, 3rd party testing centers may emerge as of comparable value to institutional issuers.... Dmitri Zagidulin: All right any other uh last minute questions. <ildiko_mazar> Good point, Dmitri. Dmitri Zagidulin: Phil mentioned talks about the role of third-party testing centers May merge. Dmitri Zagidulin: Comparable value to institutional issuers yeah uh great great question. Dmitri Zagidulin: Uh Phil do you have uh anything else uh to say I think. Dmitri Zagidulin: On the Queue from previous. <ildiko_mazar> A good taxonomy should definitely be more tree-like than a giant 'drop-down' list, in my opinion. <dmitri_zagidulin> @Ildiko - definitely PL/T3: No I'm not not specifically I just think I I think what we're all struggling with is the ability to make this connection and do so in a way that doesn't impose on the individual who is making the self-assured claim the expectation of the knowledge and uh and expertise that affect assessment organizations and others that have been doing this for such a long period of time have developed uh in fits and starts over over tens of years we have to figure out a way that there's something that's good enough to convey the brand value recognition that a lot of people currently use who aren't even paying attention to all the things underneath the surface but just assuming them. Dmitri Zagidulin: All right uh we are nearing the top of the hour we should wrap up I think the moral of the story here is that next time somebody wakes you up at 4 a.m. in the morning and says quick. <ildiko_mazar> It was a great call, thank you, Dmitri. Dmitri Zagidulin: What's the value proposition what are the difficult things of self issue credentials then you'll be able to answer them you can say. Dmitri Zagidulin: What are the hard parts. Dmitri Zagidulin: Uh unstructured thoughts into structured ones uh establishing trust in the self issuer uh presenting evidence and presenting endorsements and pulling all of that together in a. <mahesh_balan_-_pocketcred.com> Thank you for the thoughtful discussion team VC-EDU Dmitri Zagidulin: User-friendly way and if you can say all of that at 4 in the morning uh we we salute you. <ildiko_mazar> :) Dmitri Zagidulin: See you all hopefully uh the next week or since uh. <pl/t3> coctail party opening lines ;-) Dmitri Zagidulin: Uh see you all 1 of these next weeks thank you again. Dmitri Zagidulin: Everyone uh as always we really appreciate the good conversation and questions. Dmitri Zagidulin: And uh yeah talk to y'all later. Dmitri Zagidulin: And and reach out to us if you have topics to present on.
Received on Thursday, 4 July 2024 19:15:06 UTC