[TLS] Re: draft-connolly-tls-mlkem-key-agreement

[Watson Ladd <[e-mail]>]

Having MLKEM without a hybrid as an option in TLS when the interoperable
choice is a hybrid is not going to exclude people.

Furthermore we didn't hybridize x25519 and RSA. It's clear some people
believe ML-KEM is secure enough for their uses.

There also is no protocol police: code points for this will exist
regardless of what the IETF does with the draft.

Sincerely,
Watson

On Sat, Dec 7, 2024, 9:26 AM D. J. Bernstein <[e-mail]> wrote:

> Salz, Rich writes:
> > The IETF has lawyers who are familiar with how the IETF works, and
> > that legal counsel finds our processes acceptable level of risk.
>
> RFC 9680 was issued in October 2024 and says that it's "intended to
> educate IETF participants about how to reduce antitrust risks in
> connection with IETF activities". Sounds like the lawyers are worried.
> More to the point, they _should_ be worried.
>
> > The IETF has long accepted that saying "we have customers who want
> > this" as a metric into its decisions. If Foo.Com says they have
> > customers for RFC X, then the IETF community believes it is more
> > likely that Foo.Com will implement RFC X. Market acceptance is
> > important to what the IETF considers success.
>
> The whole point of antitrust law is to put constraints on _how_ market
> success can be pursued, so praising market success is missing the point.
> Take any corporation (whether for-profit or not) that lost an antitrust
> case, and look at its briefs; you'll see that it argued, unsuccessfully,
> that what it was doing was healthy market activity.
>
> As one of many examples of the constraints that antitrust law places on
> standards-development organizations, look at
>
>
> https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:52011XC0114(04)
>
> asking, inter alia, whether "the procedure for adopting the standard in
> question is transparent". Saying that IETF is pursuing market success
> does nothing to answer this transparency question, or any of the other
> procedural questions asked by antitrust law.
>
> The same link also asks whether there are "objective criteria for
> selecting the technology to be included in the standard". Where are
> IETF's objective criteria for deciding what to select? Where's the
> documentation demonstrating systematic enforcement of those criteria?
>
> As illustrated by this discussion, IETF allows a company to push for a
> standard by simply claiming, without evidence or quantification or a
> technological rationale, that there are customers for the standard.
> History shows that IETF is more likely to act on such claims from large
> companies than from small companies.
>
> Yes, of course a company pushing for something to be standardized will
> claim that the standard will have customers. This is content-free beyond
> the company name---and IETF doesn't follow procedures that demand more
> information. Evidently the IETF "metric" is not the market prospects of
> what's being proposed for standardization, but rather the size of the
> company pushing for it. That's inherently anti-competitive, rewarding
> the existing big players instead of establishing a level playing field.
>
> A court will want to know how IETF procedures stop a large company or a
> cartel from manipulating IETF's decision-making process to suppress
> competition. What the court will instead see is evidence of IETF often
> allowing its process to be manipulated without even asking questions.
>
> > Please don't quote other lawyers to us.
>
> You're complaining about my providing a link to
>
>
> https://www.google.com/books/edition/Handbook_on_Antitrust_Aspects_of_Standar/zin5tgAACAAJ
>
> from the American Bar Association? I think this is a great starting
> point for WG participants who want to learn more about the topic.
>
> ---D. J. Bernstein
>
> _______________________________________________
> TLS mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]
>