Re: When Technical Standards Meet Geopolitical Reality

On Thu, Jul 17, 2025 at 5:41 AM Daniel Hardman <daniel.hardman@gmail.com> wrote:
> I respond that, although the primary objective of a tech standard is unlikely to be human rights, socially consequential tech comes with an ethical responsibility to first (try to) do no harm and then (try to) make things better. Obviously we can have different opinions about what "harm" and "better" mean, but I think we mostly agree that ethical considerations are not out of scope.)

Yes, I do agree that most of the people that are involved in this work
agree with your statement above.

> > So maybe the questions to ask ourselves, in relation to any new standard, are:
>
> * To what extent does this standard force implementers to make positive progress or remove an old and obnoxious compromise on human rights or dignity?
> * To what extent does this standard add new forced or optional compromises on human rights or dignity?

Yes, I agree that those are good questions to ask of any work we do.
Where's the bar, though? Is it good enough that it removes one thing
while adding another? How many of these things need to be
removed/added... and to what degree, for us to deem the work worth
doing? I expect the answer is an "it depends", and the only way of
going about it is to have a discussion on each work item.

> Sometimes they got less than they demanded (new compromises), but the net effect trended in the right direction. Compromises in many dimensions (conciseness of the legal code, ease of implementation, political outcomes...) might have increased, but compromises in the human rights dimension decreased. Real progress occurred.

Yes, thank you for making my point far more eloquently than I did.
It's not that we've wanted to make some of the compromises we have,
but that was the reality of the situation when we did -- the choice in
front of us (at the time) was compromise or watch them kill the work.

For example, the Controlled Identifiers specification is a distasteful
specification. I say this as the lead editor of that specification...
but there was a small group of people that were adamant that the
specification exist, that controlled identifiers that were built on
HTTPS should exist, and that if that didn't happen, a fork or more
centralized mechanisms would be put forward as alternatives. So, a
compromise was made and both sides of the debate got some of what they
wanted. The pro-centralization folks got to keep building their
centralized systems, and the pro-decentralization people got some
decentralized stuff that wasn't standardized yet... standardized...
and by doing that, they were able to push some of that decentralized
stuff into real production systems.

There are over 1.8 million digital driver's licenses in production in
the state of California now that are Verifiable Credentials and that
use Decentralized Identifiers. There are many more Verifiable
Credentials issued in the retail sector that use Decentralized
Identifiers as well. Both of those industries needed the assurance
that there was a global standard to do so and we wouldn't have gotten
there if we did make some compromises along the way. So, while we
compromised and put forth the Controlled Identifiers specification...
they ended up adopting the Decentralized Identifiers specification.
While California DMV has adopted did:web, and while that's not as
decentralized as we'd like them to go, we haven't yet put something
better in front of them that achieves their goals. I would expect
they'd jump to did:webvh *if* there was a standard there... but that
is work that we still need to complete. I'd expect them to go even
farther, but again, we don't have a more decentralized DID Method for
them to adopt yet.

Similarly, I can almost guarantee that the retail sector would adopt a
more decentralized DID Method... again, IF we can provide a global
standard to them for that purpose. Their industry is massively
decentralized and they get the benefit of both DIDs and VCs over the
alternatives out there.

So, we need to build toward that future... the work we've done over
the past decade or more is out there in production in significant ways
and we have achieved a good chunk of the initial vision and mission.
That said, there is more work to do -- decentralization (and human
rights) isn't a destination -- it's a constant struggle. There will be
setbacks along the way, but that is inevitable. There will always be
people building centralized systems, or systems that are not as
respectful to human rights as we'd want -- our mission is to provide
better options and put better systems out there.

-- manu

-- 
Manu Sporny - https://www.linkedin.com/in/manusporny/
Founder/CEO - Digital Bazaar, Inc.
https://www.digitalbazaar.com/

Received on Thursday, 17 July 2025 14:43:15 UTC