Re: [EXT] Re: LS from GSMA EIG to W3C

I'm sorry if this is obvious to everyone else, but why does the current 
BBS DataIntegrityProof spec not enable unlinkability?

Markus

On 9/14/23 03:00, Orie Steele wrote:
> Would the GSMA cryptographers be willing to help address the security 
> issues associated with RDF DataSet Canonicalization, in the context of 
> multimessage zero knowledge proofs built on top of application/n-quad 
> based selective disclosure schemes with unlinkability?
>
> If they are eager to help address those concerns, and they see value 
> in application/n-quads as the base informational unit for unlinkable 
> credential data models, I think it makes sense for us to comment 
> specifically on that in whatever the working group puts forward.
>
> I agree with Manu's comment about signature fingerprints, however I 
> think it's important to acknowledge what is "possible theoretically, 
> and assuming a lot of work that still has not been done and might not 
> be done in the lifetime of the charter" vs "what the spec / draft 
> allows today".
>
> The current BBS DataIntegrityProof spec does not enable unlinkability.
>
> That does not mean that BBS can't support it, but if people are 
> signing up to support the work without understanding that it is still 
> not delivering on its primary value proposition over ecdsa-sd or 
> sd-jwt... That's a problem, and we should ensure the record is set 
> straight before everyone gets disappointed, or the marketing hype of 
> bbs washes away its maturity and current capabilities.
>
> I say this having seen that happen at least once before for BBS and 
> for AnonCreds which offered a similar set of capabilities... It can 
> hurt the mission of digital credentials a lot, when technical 
> capabilities or performance are oversold.
>
> Let's be clear in the liason statement why GSMA is interested in RDF 
> based selective disclosure... let's not phrase it as RDF based 
> unlinkability and selective disclosure, until it can do those things.
>
> If GSMA wants to contribute RDF and crypto experts to help close this 
> gap, let's get to work!... but it will require RDF expertise... not 
> just crypto expertise. I hope this clarifies further my previous comments.
>
> OS
>
> On Thu, Sep 14, 2023 at 2:32 AM Manu Sporny 
> <msporny@digitalbazaar.com> wrote:
>
>     On Wed, Sep 13, 2023 at 6:35 PM Paul Bastian
>     <paul.bastian@posteo.de> wrote:
>
>         To give some perspective, you should follow this thread [1] at
>         the eIDAS ARF. It seems GSMA suddenly woke up and realized
>         they need to be part of it. I've given them seven reasons why
>         BBS+ is currently not favored for the PID, but they seem not
>         to understand.
>
>     Hmm, my read on that thread is a bit different. There are a number
>     of legitimate criticisms of SD-JWT as applied to the PID and a
>     call for better technical solutions. I note that there are
>     individuals from the IETF CFRG, not just GSMA (who have deep
>     expertise in cryptography), that are criticising SD-JWT and
>     calling for BBS+-based solutions.
>
>         So I assume they are privacy advocates at any cost or they
>         might have a hidden agenda.
>
>     Presume good faith; getting an official liaison statement out of
>     GSMA is not a trivial thing to do -- it almost certainly went
>     through multiple approval processes so we can't just cast the
>     request aside based on a presumption of a "hidden agenda". One
>     could say that SD-JWT or ecdsa-sd's "hidden agenda" is tracking
>     people using signature fingerprints -- which is not conducive to a
>     productive discussion.
>
>     -- manu
>     -- 
>     Manu Sporny - https://www.linkedin.com/in/manusporny/
>     Founder/CEO - Digital Bazaar, Inc.
>     https://www.digitalbazaar.com/
>
>
>
> -- 
>
>
> ORIE STEELEChief Technology Officerwww.transmute.industries
>
> <https://transmute.industries>
>