- From: Daniel Thompson-Yvetot <drthompsonsmagickindustries@gmail.com>
- Date: Tue, 30 Jul 2019 18:49:42 +0200
- To: Dave Longley <dlongley@digitalbazaar.com>
- Cc: Henry Story <henry.story@gmail.com>, Credentials Community Group <public-credentials@w3.org>, Manu Sporny <msporny@digitalbazaar.com>
- Message-ID: <CAF_jFpu+z4JnKrS1GXe8twnpgGP2F1nV=AhjNaSBAr=cEe4qzw@mail.gmail.com>
I guess my point is that any additional operations on a canonical DID ID increase the overhead. And once we're thinking about resolving potentially quadrillions or more of rows of id's - even with stored DB methods - such ops force infrastructural overhead and presuppose backpressure. If I can append a did directly to a base url resolver without needing to parse for colons I can leverage existing and well known CRUD patterns and still parse if I need to... On Tue, 30 Jul 2019, 18:39 Daniel Thompson-Yvetot, < drthompsonsmagickindustries@gmail.com> wrote: > Dereferencing and parsing from the DID are two additional processes. Is > there a real reason why it has to be : delimited instead of / delimited? > > On Tue, 30 Jul 2019, 18:05 Dave Longley, <dlongley@digitalbazaar.com> > wrote: > >> >> On 7/30/19 8:23 AM, Henry Story wrote: >> > >> > >> >> On 30 Jul 2019, at 13:16, Henry Story <henry.story@gmail.com >> >> <mailto:henry.story@gmail.com>> wrote: >> >> >> >> Would it be possible to make a minor extension to HttpSignature >> >> so that one can use https WebIDs [0] just as a way to get a simple >> >> thing going? That could be completely compatible with DIDs, but >> >> would allow one to get going in cases where one does not need >> >> all of the extra goodies that DIDs give one. >> >> >> >> I did some work on that to implement a server side HTTP-Signatures >> >> and before that I worked on how one could use JS Crypto to create >> >> keys in the browser. >> > >> > >> > Actually I think one could get the two to get quite close by adding a >> > notion of a KeyId ie perhaps an URI for a key that could be linked to >> > a WebID. One could then have >> > >> > - https:// KeyIds. These could be located in the WebID Profile, >> > in a different document on the same server, or perhaps even on >> > a different server. >> > - or dids. >> > >> > A WebID could point to the KeyID or DID (or perhaps these are >> > the same?), and authentication using Https Signature could then >> > pass both in the header. The DID for authentication, the WebID for >> > social network type identifier, and the link from the WebID to the >> > Did/KeyId in the Profile Document would be the proof. >> >> My understanding is that HTTP signatures already supports something like >> WebID-style authentication. Just use a URI for the `keyId` parameter. >> >> We already use HTTP signatures in conjunction with LD proofs and DIDs. >> To my knowledge, it wouldn't work any differently if the `keyId` pointed >> at a key that resolved to a document that referenced a WebID as its >> `controller`. You'd then dereference that WebID to ensure its profile >> document had an appropriate back link to the key. (This is generally how >> LD proofs work, so it is agnostic to DIDs or WebIDs). >> >> >> > >> > Henry >> > >> >> >> >> Henry >> >> >> >> [0] https://www.w3.org/2005/Incubator/webid/spec/identity/ >> >> [1] https://github.com/read-write-web/akka-http-signature >> >> [2] https://github.com/read-write-web/solid-client >> >> >> >>> On 30 Jul 2019, at 04:22, Manu Sporny <msporny@digitalbazaar.com >> >>> <mailto:msporny@digitalbazaar.com>> wrote: >> >>> >> >>> On 7/29/19 1:35 PM, Joe Andrieu wrote: >> >>>> TL;DR What’s after DIDs? >> >>> >> >>> From a technical standards perspective, these are currently pain >> points >> >>> for Digital Bazaar and our customers that seem like we might be able >> to >> >>> collaborate on with other developers / companies in this group: >> >>> >> >>> * True multi-DID interop >> >>> * True multi-wallet/issuer/verifier interop >> >>> * Collaboration on the Credential Handler API >> >>> * Linked Data Proofs/Signatures (W3C WGs for these) >> >>> * Secure Data Hubs (or, how do we make storage privacy-aware >> >>> and self-sovereign) >> >>> * Verifiable Credentials 1.1 (we're not done yet) >> >>> * Verifiable Credentials Extensions >> >>> * VC/DID/LDP Registries >> >>> >> >>> There are non-technical things we should do as well: >> >>> >> >>> * Non-Violent Communication (and other approaches) to mend some of the >> >>> damage across the identity community >> >>> * Field Work - More Use Cases from real people/customers >> >>> * Bite sized material to communicate our work to the general public >> >>> >> >>> -- manu >> >>> >> >>> -- >> >>> Manu Sporny (skype: msporny, twitter: manusporny) >> >>> Founder/CEO - Digital Bazaar, Inc. >> >>> blog: Veres One Decentralized Identifier Blockchain Launches >> >>> https://tinyurl.com/veres-one-launches >> >>> >> >> >> > >> >> >> -- >> Dave Longley >> CTO >> Digital Bazaar, Inc. >> http://digitalbazaar.com >> >>
Received on Tuesday, 30 July 2019 21:38:08 UTC