Re: A Framework for Human-AI Collaboration in Standards Work

+1 Manu. Apologies for cross-posting with CCG:

The structure of workgroups and SDOs will change with increasingly
capable AI.

For almost a year now, I've been able to ask LLMs how to accomplish what I
want. They read the API documentation, answered my questions and
implemented the API. I never, once looked at a standard or the code.
However, I do ask Claude Code to generate markdown files for documentation,
which helps me feel in control and reduces the cost and risk of future
changes. The documentation includes very valuable analyses of security
vulnerabilities.

Consequently, the role of groups like CCG and standards workgroups is
changing, at least for me. I look forward to learning about business
realities and real-world experience.

I no longer care about new standards. If a vendor or service provider wants
my business, it's up to them to provide and document the APIs. Standardized
APIs can reduce risk and switching costs, of course, but if the tradeoff is
5+ years of discussions on CCG and related forums, the juice is no longer
worth the squeeze.

Adrian

On Sat, Apr 18, 2026 at 5:10 PM Manu Sporny <msporny@digitalbazaar.com>
wrote:

> On Sat, Apr 18, 2026 at 3:29 PM Christoph <christoph@christophdorn.com>
> wrote:
> > I did not want to just leave it hanging like that so I have generated an
> opinion on the perspective that I see and am moving towards in my own work.
>
> Christoph, I know you're an introspective person, and that's the only
> reason I'm responding to this email. You have made Marcus' point so
> thoroughly and convincingly that I am now re-thinking my position on
> "just ignore the noise". This (LLM usage on the mailing list) is
> starting to get out of hand (again). This time, the problem isn't
> Morrow... it's the humans on the list that don't seem to be getting
> the hint that they're overusing LLMs and subjecting the rest of us to
> a tremendous amount of noise.
>
> Even though it is not your intention, you are harming the discourse on
> this mailing list.
>
> Please stop.
>
> Melvin's response was good (as it usually is), he was terse, made his
> point and got on with it.
>
> I can see people trying to be subtle and nice about it, but that
> doesn't seem to be working. So, I am going to attempt to be polite,
> but firm, because I want you to be heard by this community.
>
> There is a limit to what each of us will tolerate, and your email
> (Christoph) shattered that limit for me (not in a good way). I'm
> hoping you, Eduardo, and Michael Herman will take the rest of this
> email to heart.
>
> > I am a fan of maximizing LLM context
> > Prompt:
> > Write an extensive dissertation
>
> I stopped reading when I hit the words "extensive dissertation". I
> know you didn't realize how disrespectful of everyone else's time this
> was, but let me try and spell it out:
>
> You generated something that would take multiple hours to read,
> internalize, and analyze, and then you sent it out to a mailing list
> with over 580 people on it, and you expected a subset of them to read
> something that took you a fraction of that time to generate.
>
> This is a type of asymmetric information overloading attack that a
> number of us on the mailing list have been complaining about for
> weeks.
>
> Eduardo, your DID dashboard thing (your second pass) looks like an
> improvement, but then you hit the mailing list with A LOT of
> LLM-generated projects. Overwhelming the group, or at least me, to the
> point where I lost all motivation to give you feedback on your
> dashboard  revision work (which I thought was a good improvement)...
> but if you're going to keep LLM'ing me, then I don't have the time to
> respond... I give up, I hope you have great success, but I can't
> review every LLM-generated project that you're sending to the list. I
> don't even know which one of them you care about the most. I'm
> overwhelmed. You have overwhelmed me, and I am completely demotivated
> from engaging with your content... and that's terrible, because I
> think ONE of the things I did engage on looked like it was going in a
> good direction.
>
> Same thing for you, Michael:
>
> > Web 7.0: an entire general-purpose Decentralized System Architecture
> platform: code, test cases, Whitepaper, design docs, 15 IETF draft
> specifications, spec compliant, etc. etc. in less that 2 weeks
>
> I have no hope of ever reviewing the vast majority of that, and I
> doubt others on the mailing list do, either. So, I've just shut off.
> Again, I hope you're wildly successful, but I have no way of engaging
> with what your LLM is generating at scale. The human is overwhelmed
> and unable to help you anymore.
>
> The result is that the people that read the mailing list are going to
> increasingly start treating each of your posts as noise. I know that's
> where I am right now. Again, I'm not trying to be rude about this...
> I'm trying to be polite by letting you know how what you are doing is
> affecting my perception of you. The way that each of you are engaging
> with your use of LLMs breaks social norms we've had for a while in
> this community, and I don't think it's having the sort of positive
> effect you think it is. I'd be happy to be in the minority here,
> because that means I can just mark it as noise (for me) knowing that
> others are benefitting from the way you're engaging.
>
> -- manu
>
> --
> Manu Sporny - https://www.linkedin.com/in/manusporny/
> Founder/CEO - Digital Bazaar, Inc.
> https://www.digitalbazaar.com/
>
>