W3C home > Mailing lists > Public > public-credentials@w3.org > September 2017

Re: Status of DID spec and going to production

From: Manu Sporny <msporny@digitalbazaar.com>
Date: Fri, 22 Sep 2017 12:25:57 -0400
To: Markus Sabadello <markus@danubetech.com>, Daniel Buchner <daniel.buchner@microsoft.com>
Cc: W3C Credentials CG <public-credentials@w3.org>
Message-ID: <e056dbea-9049-4863-0fa5-808bbee48d1e@digitalbazaar.com>
On 09/20/2017 09:54 AM, Markus Sabadello wrote:
> FYI in case you're not aware, at DIF there's been a lot of discussion
> on a "Universal Resolver" that can resolve different types of
> identifiers including DIDs.

+1 on this... as you may be aware, we're working on something like
that... authorization.io being the first roll out of a service like that
several years ago:

https://demo.authorization.io/dids/did:8743453f-e45e-4ac6-b85f-4513ba4c1460

... we've always intended to update it to support different types of
DIDs (and are getting closer to deploying that code base soon-ish...
next couple of months).

There is also this, which is a DID resolution library for the browser
and node.js:

https://github.com/digitalbazaar/did-io

> The idea is that "drivers" can be developed and plugged in for
> different identifier types.

This is one approach that we decided to not pursue due to the deployment
complexity of such a system. I suggest we follow a more RESTful
approach. A DID resolution library really only needs one endpoint to do
resolution (per DID):

GET /dids/{DID}

That's it. If we tell the people writing ledgers that they need to
expose that on their ledger nodes if they want to support DID resolution
directly, then writing resolver libraries becomes a simple matter of
configuring the resolver libraries to use a set of endpoints per DID
they want to support.

If ledger developers don't want to expose that directly on the nodes,
one can still expose an HTTP server that is akin to your "drivers" approach.

This leads to very loose coupling between the DID resolver library and
the systems that do the heavy lifting. No recompilation or
redistribution necessary for the DID resolver libs... you just config
them w/ endpoints that you trust (and you may be running all those
endpoints).

> Not really sure how this implementation work will be 
> organized/coordinated, maybe RWoT and IIW will be good opportunities
> to discuss.

Let's bring it up at RWoT and see where the discussion goes. I won't be
at IIW this year.

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Rebalancing How the Web is Built
http://manu.sporny.org/2016/rebalancing/
Received on Friday, 22 September 2017 16:26:20 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:18:13 UTC