Re: Status of WebID proposal

Happy New Year all,

> On 10. Jan 2023, at 22:55, Melvin Carvalho <melvincarvalho@gmail.com> wrote:
> 
> Hi All and Happy new year!
> 
> Just wondering if there might be some movement on the WebID spec
> 
> Could 2023 be the year that we get the spec over the line?
> 

Don’t hold your Breath Melvin.

Standards take a huge time to come together. Just look at Http Message Signatures which
started in 2014 or so and is now progressing through the IETF HTTP Bis WG.

https://datatracker.ietf.org/doc/draft-ietf-httpbis-message-signatures/

They have implementations from Amazon, Mastodon and many others at each stage
of its development. Implementations drive standardisation. And even here the
background is quite splintered, with say Mastodon using a 6 year old version.

WebID is being adopted most seriously by Solid, which is progressing on a
number of fronts, where WebID is a key part. The WebID specs as they are are
not getting in the way it seems.

Last year we made a whole bunch of editorial fixes to WebID. Nothing major that
would require publishing a new version though.

WebID is currently used by the following projects (I am aware of)

Webid-Profile group
----------—————————

They are working on profile vocabularies
https://github.com/solid/webid-profile

I am not following that, due to lack of time, but should pop
perhaps to see where it is going.

Solid OIDC
---———————

 https://solidproject.org/TR/oidc-primer

There is some idea of using it also to identify App instances,
thinking of  them as agents. That could require json-ld support
in the spec, which is not unreasonable. So I think that is on
the cards at some point, but there is no consensus yet, and the
authors have found it easier to use another word.

Http Signature
-------———————

I am working on Http Signatures, which I see as the successor of
WebID TLS. After writing an implementation in Scala of

https://github.com/bblfish/httpSig

and after implementing in a Solid Server [1] I am now writing client
libraries to test the idea fully. So I will be updating the HttpSig
spec soon with what I have learnt from all these implementations.

https://github.com/bblfish/authentication-panel/blob/main/proposals/HttpSignature.md

Here the KeyId takes a central role and could be all that is needed in a
minimal setup.  But linking to a KeyID is fragile, and one would prefer
a more stable identifier like a WebId. And indeed they can link up as I show
in that spec.

Those are some of the project I know of that are using WebID. There are
certainly others. Eg. WAC uses it too for access control.

Henry


[1] https://github.com/co-operating-systems/Reactive-SoLiD/pulls

> 
> Best regards,
> Jacopo.
> 
> ---
> 
> Jacopo Scazzosi
> https://jacoscaz.com
> https://treesandrobots.com
> 
> 
> 
> > On 21 Jul 2022, at 22:51, Jeremy Lewi <jeremy@lewi.us> wrote:
> >
> > Hi Folks,
> >
> > What's the current status of the WebID proposal? It seems like the proposal has been around for 10+ years but hasn't become a W3C recommendation.
> >
> > This thread suggests that progress has stalled in part because the spec has gained very little adoption. Can anyone provide more context/opinions on why progress/adoption has stalled? Were their technical objections to the spec or was the spec solving a problem most folks didn't have?
> >
> > How does the future of WebID relate to DID's which I believe recently became a W3C recommendation (blog)? Would WebID be compatible or superseded by DID web?
> >
> > Thanks
> > J
> 
>