- From: Moses Ma <moses.ma@futurelabconsulting.com>
- Date: Tue, 27 Nov 2018 00:08:41 -0800
- To: public-credentials@w3.org
- Message-ID: <195427ff-bc81-77bb-a7a8-96772300b7dd@futurelabconsulting.com>
Samantha Chase's excellent paper can be found here: https://github.com/WebOfTrustInfo/rwot7/blob/master/topics-and-advance-readings/CanCurationMarketsEstablishSustainableTechnologyCommons.pdf On 11/26/18 9:57 PM, Carlos Bruguera wrote: > Thanks for the update. > > On this regard, can anybody share the link paper mentioned during the > call: /Furthering sustainable commons/? Appreciated. > > On Tue, Nov 27, 2018 at 10:58 AM <kim@learningmachine.com > <mailto:kim@learningmachine.com>> wrote: > > Thanks to for scribing this week! The minutes > for this week's Credentials CG telecon are now available: > > https://w3c-ccg.github.io/meetings/2018-11-20/ > > Full text of the discussion follows for W3C archival purposes. > Audio from the meeting is available as well (link provided below). > > ---------------------------------------------------------------- > Credentials CG Telecon Minutes for 2018-11-20 > > Agenda: > https://lists.w3.org/Archives/Public/public-credentials/2018Nov/0129.html > Topics: > 1. Introductions and Reintroductions > 2. Announcements, reminders > 3. Action items > 4. Work Items > 5. Pain points > Organizer: > Joe Andrieu and Kim Hamilton Duffy and Christopher Allen > Scribe: > > Present: > Christopher Allen, Bohdan Andriyiv, Andrew Hughes, Manu Sporny, > Dmitri Zagidulin, Ryan Grant, Brent Zundel, Moses Ma, Joe > Andrieu, Lucas Parker, Ted Thibodeau, Lionel Wolberger, Markus > Sabadello, Drummond Reed, Joe Kaplan, Sam Smith, Nate Otto, > Michaela Casaldi, Jarlath O'Carroll, Jeff Orgel, Chris Webber, > Andrew Rosen, Adrian Hope-Bailie > Audio: > https://w3c-ccg.github.io/meetings/2018-11-20/audio.ogg > > Joe Andrieu: Connections > Ryan Grant: Does voip-ccg association still work if you do it? > > Topic: Introductions and Reintroductions > > Lionel Wolberger: ... Main topic, the pain points that DIs are > solving. > Drummond Reed: Note: I can only stay for the first 30 mins today. > Moses Ma: Spoke with his partners about our work, and we have a > volunteer. Dr. Wu [scribe assist by Lionel Wolberger] > Lionel Wolberger: ... VC who ran a $billion fund > Lionel Wolberger: ... Templates for DID monetization > Lionel Wolberger: .... List different ways we can monetize the > DID market > Manu Sporny: +1 To that effort, would be very helpful to the CCG. > Lionel Wolberger: ... Dr. Wu was a lead investor on Tivo, is good > at revenue models. > Joe Kaplan: Will this be a work item? How can the community > support? [scribe assist by Lionel Wolberger] > Moses Ma: Paper for next RWoT [scribe assist by Lionel > Wolberger] > Sam Smith: Furthering sustainable commons, [scribe assist by > Lionel Wolberger] > Lionel Wolberger: ... If looking to monetize, this paper is > related. Will share it. > Moses Ma: Let's have the community participate. Should stipulate > how a standard can create a fair method to enable monetization > models. [scribe assist by Lionel Wolberger] > S/Furhtering/Furthering > Lionel Wolberger: .... A mockup of the UX would be helpful, > perhaps in Adobe XD > Joe Kaplan: Send email and we will follow up. [scribe assist by > Lionel Wolberger] > Lionel Wolberger: Jarlath to the mic! > Jarlath O'Carroll: CEO and founder of Jobs___ [scribe assist by > Lionel Wolberger] > Lionel Wolberger: ... Connects students to jobs > Lionel Wolberger: ... Interested in CCG/VCs for credentials > regarding skills, etc > > Topic: Announcements, reminders > > Joe Kaplan: Dec 10 workshop, Microsoft [scribe assist by Lionel > Wolberger] > Manu Sporny: 55 People are signed up, room for 15 more. [scribe > assist by Lionel Wolberger] > Manu Sporny: > https://www.w3.org/Security/strong-authentication-and-identity-workshop/cfp.html > Lionel Wolberger: ... Seeking more lawyers, regulatory and > compliance types > Lionel Wolberger: ... Seeking more European (GDPR) and China > focus > Lionel Wolberger: ... Still time to register! > Lionel Wolberger: ... Note that new proposals will compete with > some critical proposals that we must present at the workshop > Lionel Wolberger: ... Agenda is being formulated and will be > shared soon. > Lionel Wolberger: RWoT #8 planned for Feb22/28/Mar 01 > Joe Kaplan: Making decisions about location, to be announced > ASAP. [scribe assist by Lionel Wolberger] > Joe Kaplan: IIW APril3-May 2. Not the same time as RWoT this > time ;-) [scribe assist by Lionel Wolberger] > Manu Sporny: Barcelona proposal for RWoT [scribe assist by > Lionel Wolberger] > Moses Ma: +1 Barcelona > Lionel Wolberger: ... May be just after MWC (mobile world > congress) > Christopher Allen: Take train > > Topic: Action items > > Bohdan Andriyiv: +1 For Barcelona) > Joe Kaplan: Planning to "create Amira as a repo" [scribe assist > by Lionel Wolberger] > Moses Ma: Can someone post URL to Sam's "Furthering sustainable > commons" paper > Joe Andrieu: https://github.com/w3c-ccg/community/issues/18 > Manu Sporny: https://www.w3.org/2018/11/19-vcwg-minutes.html > Manu Sporny: Meeting minutes on how to harmonize with Verifiable > Credentials [scribe assist by Lionel Wolberger] > Lionel Wolberger: ... General pattern for addressing ZKPs > Lionel Wolberger: ... Pattern to host ZKP even as binary BLOBs > Joe Andrieu: > https://github.com/w3c-ccg/community/blob/master/work_items.md > > Topic: Work Items > > Drummond Reed: The Sovrin community intends for ZKPs to NOT be a > "bizarre, out-of-the way format" :-) > Ryan Grant: +1 For Barcelona > Manu Sporny: Drummond -- I expected as much, :) > Manu Sporny: OCAP in JS [scribe assist by Lionel Wolberger] > Lionel Wolberger: ... Library implementation > BLOB = Bizarre Large Object </humor> > Manu Sporny: Regarding, seeking additional funds for people to > implement tools [scribe assist by Lionel Wolberger] > Lionel Wolberger: ... E.g. issue a new type of verfiable > credential, need to define a vocabulary, need a website where you > can go and CLICK to publish such a vocabulary > Lionel Wolberger: ... Cryptographic hash linking specification, > that is more detailed then just "use IPRS" > Lionel Wolberger: ... Will be useful to have a kind of "magnet > link" > Lionel Wolberger: ... This is a problem across the decentralized > blockchain space > Lionel Wolberger: ... Proposing an IETF specification > Nate Otto: +1 To magnet link IRIs for linked data > Lionel Wolberger: ... New problem emerging around vendor lockin > on digital wallets > Lionel Wolberger: ... Ensure that one vendor won't lock out > everyone else, by being specification conforming but not enabing > data portability > Lionel Wolberger: Manu: Exciting stuff +1 > Drummond Reed: BTW, avoiding vendor lock-in is a primary goal of > DKMS, of which the plan is to start a Technical Committee at > OASIS. See http://bit.ly/dkmsv3 > Manu Sporny: Mag links will be important to endurance, the > ability for documents to be addressable over a period of years > [scribe assist by Lionel Wolberger] > > Topic: Pain points > > Manu Sporny: Drummond, What I was talking about goes beyond DKMS, > but yes, that work is important as well. > Chris Webber: We accept the value of decentralization without > much consideration [scribe assist by Lionel Wolberger] > Lionel Wolberger: ... We can now tease out the assumptions and > motivations behind this > Lionel Wolberger: ... These should be made overt in the DID > primer > Lionel Wolberger: ... Let's start with Vendor Lock-in > Lionel Wolberger: ... Many standards and protocols ended up being > locked-in due to some inherent centrality > Lionel Wolberger: ... Example: Twitter had lots of apps in a > broad ecosystem, but by Twitter controlling the API Keys they > constrained that ecosystem > Lionel Wolberger: ... In federated DIDs, some parties took > protocols that were intended to be two way > Lionel Wolberger: ... But then only implemented one side > Lionel Wolberger: \ > Manu Sporny: Every market vertical has its own motivation for > needing DIDs [scribe assist by Lionel Wolberger] > Lionel Wolberger: ... In Healthcare DIDS are useful for X,Y,Z > Lionel Wolberger: ... In banking DIDs are useful for doing n,m,o > Lionel Wolberger: ... Local, provincial and federal governments > do not want to be the system of record for identifiers > Lionel Wolberger: ... It's all knowledge based stuff > Lionel Wolberger: ... These organizations do not want to control > knowledge based identifiers as opposed to cryptographic > identifiers > Lionel Wolberger: ... Since they are almost guaranteed that the > funding creating the system diminishes over time > Lionel Wolberger: ... As the systems grow, the funding shrinks > and can even be cut > Lionel Wolberger: ... Making the central system suceptible to > failure > Andrew Hughes: Identifiers are useful. The fatal flaw (in our > opinion) is that useful widely-usable identifiers end up with > central authorities or defacto authorities that have ‘kill > switches’. Also all ‘authorities’ must inevitably become > high-value attach target infrastructure while at the same time > facing funding pressures (because it goes into the background as > infrastructure). Decentralization has the promise of a > globally-shared namespace that involved de[CUT] > Andrew Hughes: Governance and operations but universal > resolvability. > Lionel Wolberger: ... Organizations are excited that the DID > enables use without hosting it > Lionel Wolberger: ... Though when you point out the cost, their > enthusiasm cools a bit > Q > Chris Webber: Borders are a pain point [scribe assist by Lionel > Wolberger] > Lionel Wolberger: ... Borders between countries. Borders between > companies. > Lionel Wolberger: ... Different ways we evaluate and think about > trust > Lionel Wolberger: ... Everybody's trust requirements are > different, in sometimes subtle, sometimes kajor ways > Lionel Wolberger: ... A centralized federated system demands tha > tthe trust model propagate throughout the system and mark all > interactions > Lionel Wolberger: ... A decentralized system will support > variation in those trust rules > Lionel Wolberger: ... You may want to rely on something that > other people dont need or dont want to pay for > Drummond Reed: Gotta run now. Bye. > Lionel Wolberger: ... Back in SSL, we defined client certs, and > almost no one ended up adopting that > Joe Kaplan: In solving the double spend problem, we ended up > defining DIDs [scribe assist by Lionel Wolberger] > Lionel Wolberger: ... Interstitial jurisdictionality > Lionel Wolberger: ... There are well defined jurisdictions > Lionel Wolberger: Inbetween these well defined jurisdictions > there are interactions > Lionel Wolberger: ... In these interstices we interact > Lionel Wolberger: ... How can we have an interaction outside a > jurisdiction > Lionel Wolberger: ... E.g. a soviet union master of science, how > will another country e.g. the UK evaluate that > Andrew Rosen: Identifiers are useful. [scribe assist by Lionel > Wolberger] > Lionel Wolberger: ... These have kill switches > Lionel Wolberger: ... DID offers governance but still > resolvability > Lionel Wolberger: ... Identifiers are useful. The fatal flaw (in > our opinion) is that useful widely-usable identifiers end up with > central authorities or defacto authorities that have ‘kill > switches’. Also all ‘authorities’ must inevitably become > high-value attach target infrastructure while at the same time > facing funding pressures (because it goes into the background as > infrastructure). Decentralization has the promise of a > globally-shared namesp[CUT] > Lionel Wolberger: ... Governance and operations but universal > resolvability. > Sam Smith: Offloading personal data liability, avoiding toxic > data [scribe assist by Lionel Wolberger] > Lionel Wolberger: ... Focusing on construction sites, new > construction to create a safety wifi network to mark things on a > job site, track > Lionel Wolberger: ... Generates a safety plan and a 3D model of > the space from floor plans > Lionel Wolberger: ... Sam showed them overlays in the wallet > Lionel Wolberger: ... Proof of data without cost of storage > Lionel Wolberger: ... Given these watches (apple watch) will you > accept this token? > Lionel Wolberger: ... If this succeeds, no one has to store the > data, then through an overlay or an OAuth scope > Lionel Wolberger: ... Hit the threshold > Lionel Wolberger: ... This way create a non-surveillance > ecosystem > Lionel Wolberger: Audio problem > Lionel Wolberger: Go on > Manu Sporny: Centralized ID providers, e.g. legal entity > identifier and large corporations [scribe assist by Lionel > Wolberger] > Lionel Wolberger: ... These are interested in upgrading their > identifiers > Lionel Wolberger: ... E.g. a company whose business model is > issuing identifiers > Lionel Wolberger: ... They seek the addition of a layer of > cryptography to mitigate and prevent theft > Lionel Wolberger: ... They could roll their own crypto, or more > simply adopt DIDs > Lionel Wolberger: ... Centralized authorities want to upgrade > their ecosystem and add cryptography > Lionel Wolberger: *** Can someone scribe temporarily, I will drop > and rejoin **** > Bohdan Andriyiv: One of the issues is longevity in identifiers. > [scribe assist by Manu Sporny] > Bohdan Andriyiv: If I have an identifier, and I want a signature > on something, providers can disappear, there is no certainty that > these centralized identifiers will stay. So I think this is one > of the reasons that digital signatures were not widely adopted. > [scribe assist by Manu Sporny] > Bohdan Andriyiv: DIDs solve this problem. [scribe assist by Manu > Sporny] > Lionel Wolberger: Manu, i'm back > Bohdan Andriyiv: Question to manu - governments do not want to > manage records of centralized identifiers - I do think > governments still want those lists - they still have databases, > data stores, records of who paid how much in taxes, who received > how much and benefits, they need to keep this data, they don't > want to manage passwords for people. [scribe assist by Manu > Sporny] > Lionel Wolberger: ... Still a need to retain the data, just not > manage the task force and make it more secure > Markus Sabadello: Regarding large companies interested in > upgrading their IDs to DIDs [scribe assist by Lionel Wolberger] > Lionel Wolberger: ... I have an IETF draft to discover DIDs based > on the domain name system > Lionel Wolberger: ... Large companies are interested in using > domain names for discovery > Markus Sabadello: > https://datatracker.ietf.org/doc/draft-mayrhofer-did-dns/ > Manu Sporny: +1, That's really neat work that's going on. > Joe Kaplan: In the digital realm things are easily faked [scribe > assist by Lionel Wolberger] > Lionel Wolberger: ... Public key/private key issues > Lionel Wolberger: ... How do you verify that something is not > fake > Lionel Wolberger: ... That is a pain point that DIDs solve > Jarlath O'Carroll: @Lionel - there was a discussion about VC and > Jobs earlier, can you please post the link to the details of this > work in the feed again (I missed it)? > Chris Webber: Keep in mind we had PGP keys for decades and they > were decentralized [scribe assist by Lionel Wolberger] > Lionel Wolberger: ... They did not spread everywhere because > Lionel Wolberger: ... (A) they were not vendor agnostic nor > future proof > Lionel Wolberger: .. .DIDs are rotateble so allow technological > upgrades > Lionel Wolberger: ... The crypto is separated from the actual > identifer > Lionel Wolberger: ... Another reason why PGP fingerprints did not > achieve wide market adoption > Lionel Wolberger: ... Due to the complications of rotating them > Lionel Wolberger: ... Revocation was extremely difficult, you > needed the original key material > Lionel Wolberger: ... You had to notify people > Lionel Wolberger: ... A number of DID methods have fast and > efficient ways to notify about revocation and rotation > Adrian Hope-Bailie: Questions back to Markus, etc [scribe assist > by Lionel Wolberger] > Lionel Wolberger: ... I use corporate centralized user IDs in > general today > Lionel Wolberger: ... If DIDs would be linked to domain names or > email addresses > Lionel Wolberger: ... Would the service provider only persist the > DID and not the email address? > Lionel Wolberger: ... Let's say I use finger > Markus Sabadello: Yes, your understanding is correct. [scribe > assist by Lionel Wolberger] > Adrian Hope-Bailie: That sounds like a powerful value statement. > [scribe assist by Lionel Wolberger] > Lionel Wolberger: ... That ability sounds quite valuable > Lionel Wolberger: Something that wasn't mentioned - DID process > of creating an identifier feels like it's lower friction, more > lightweight. [scribe assist by Manu Sporny] > Lionel Wolberger: So many more digital interactions, so many > more devices, feels like a better way to interact given the > complexity of devices today. [scribe assist by Manu Sporny] > Manu Sporny: Responding to Bohdan [scribe assist by Lionel > Wolberger] > Lionel Wolberger: ... The general assertion is that governments > must continue to manage data > Lionel Wolberger: ... But the identifier is really secondary to > their interest > Markus Sabadello: FYI the August CCG list archive has some > discussion on pros/cons of discovering DIDs from DNS: > https://lists.w3.org/Archives/Public/public-credentials/2018Aug/thread.html > Lionel Wolberger: ... E.g. in the USA the SocSec number is being > used as an identifier but SecSec admin wants to stop this > Lionel Wolberger: ... SSA does not really need the identifier, > they just need to provide their services > Lionel Wolberger: ... This is what we mean by saying geovernments > do not want to be identifier providers > Lionel Wolberger: ... It is not their core value proposition > Lionel Wolberger: ... They still need an identity proofing > process, of course > Lionel Wolberger: ... But then they would not have the > responsibility to maintain and track the identifier > Lionel Wolberger: ... Keep in mind, they still have to store the > ID and that is an attack surface honeypot > Lionel Wolberger: ... They will benefit from the VC architecture, > where they store that they had a verified credential and can tear > down and not store a lot of the artifacts of the proving process > itself > Chris Webber: We are trying to move away from knowledge based > security (e.g. you know my SocSec#, you know my birthdate) > [scribe assist by Lionel Wolberger] > Manu Sporny: Yep, Knowledge Based Authentication is usually a bad > thing... > Lionel Wolberger: ... Human memorizability for DIDs was an > argument that we had > Lionel Wolberger: ... I (Chris) advocated for non-memorizable > IDs, I wanted it to be underlying > Lionel Wolberger: ... But people may want DIDs to last a lifetime > Lionel Wolberger: ... That is not prevented by the standard, > though this would be an inappropriate use > Lionel Wolberger: ... I dont want to give my BTCR identifer, I > want to give a more safe identifer. > Adrian Hope-Bailie: Responding to Manu, that the credentials are > not retained [scribe assist by Lionel Wolberger] > Lionel Wolberger: ... Huge synergy with the upcoming technology > that more and more data stores will be held by individuals > Lionel Wolberger: ... This is a good argument for DIDs in the > broadest sense > Joe Kaplan: Adding pain points from previous notes. [scribe > assist by Lionel Wolberger] > Lionel Wolberger: ... Things change. Email addresses change. > Phone numbers change. Technologies change. Organizations change. > Lionel Wolberger: ... The organization that could have verified > your deed does not exist anymore. > Lionel Wolberger: ... Fakes are a pain point. Signatures prevent > this, but signatures need PKI > Lionel Wolberger: ... Over-identification is a pain point. > Lionel Wolberger: ... Identifier misuse. Successful and useful > IDs tend to get used for more things > Lionel Wolberger: ... Burden of management: DIDs will be easier > for companies and organizations. > Lionel Wolberger: ... Jurisdictional boundaries, where different > groups for different reasons need their own identifiers. > Andrew Hughes: Pain point - vendor lock-in > A world of pain (points) </h> > Chris Webber: One size trust does not fit all [scribe assist by > Lionel Wolberger] > Manu Sporny: Good summary, is really going to help write the W3C > TAG primer > Lionel Wolberger: ... You get to decide what your trust model is > Moses Ma: Bye y'all, have a great thanksgiving! > Lionel Wolberger: HAPPY TURKEY DAY > Joe Kaplan: See you [scribe assist by Lionel Wolberger] > > > > > -- *Moses Ma | Managing Partner* moses.ma@futurelabconsulting.com | moses@ngenven.com v+1.415.568.1068 | skype mosesma | /linktr.ee/moses.tao/ <http://linktr.ee/moses.tao> FutureLab provides strategy, ideation and technology for breakthrough innovation and third generation blockchains. Learn more at /www.futurelabconsulting.com/ <http://futurelabconsulting.com>. For calendar invites, please cc: mosesma@gmail.com Or whet your appetite by reading /Agile Innovation/ <http://www.amazon.com/Agile-Innovation-Revolutionary-Accelerate-Engagement/dp/B00SSRSZ9A> | /Blockchain Design Sprint/ <https://www.amazon.com/Blockchain-Design-Sprint-Workbook-Implement/dp/1548592714> | my blog at /psychologytoday.com/ <http://www.psychologytoday.com/blog/the-tao-innovation>.
Received on Tuesday, 27 November 2018 08:09:23 UTC