[TLS] Re: Algorithm agility, was Re: Re: WG Last Call : draft-ietf-tls-mlkem-08 (Ends 2026-07-08)

On 7/7/26, 11:58 PM, "Peter Gutmann" <[email protected]> wrote:

  *
Which is what's made it so secure.  It obeys Grigg's Law, "there is only one
  *
mode and that is secure". ... One person, working by themselves, has come up with a secure-
  *
A-to-B mechanism that's better than anything the IETF has ever produced.


Collaboration is hard and security often involves trade-offs.  Especially when national(istic) forces come into play. These days it’s not enough to have an X.509 OID to be a country (as some wag once wrote[1]), you often have to have a a cipher too.

That one person also benefits from not having to work with an installed base that inherits a legacy of crypto patents, little public review and analysis of algorithms, and export and import controls.  Chalking it all up to an argument against crypto agility seems a little one-sided

[1] https://www.cs.auckland.ac.nz/~pgut001/pubs/x509guide.txt

_______________________________________________
TLS mailing list -- [email protected]
To unsubscribe send an email to [email protected]