On 19-May-26 08:59, Eric Rescorla wrote:


On Mon, May 18, 2026 at 1:57 PM Brian E Carpenter <[email protected] 
<mailto:[email protected]>> wrote:

    Interested, yes, but not if you do it during Austrian working hours...

    Re draft-rescorla-rfc-jit, I'm all for an RFC12345.1 model but it has to 
come with an improved errata process; I view that as a single package.


Hi Brian,

Do you think you could elaborate a bit on what you would want to see in an 
improved errata process? I have my own ideas, of course, but I'm curious about 
yours.

A good question, and I haven't thought about it enough yet. Mainly, a better 
defined process for each stream to rapidly process each erratum, establish 
whether it's editorial or needs a consensus discussion, and get it done and 
integrated into a new version of the source (presumably .xml), and issue a .1 
version. It really isn't OK that we have unresolved errata going back years.

   Brian


    Regards/Ngā mihi
         Brian

    On 19-May-26 03:31, Eliot Lear wrote:
     > Are others interested? We can do it virtual if that's more convenient.
     >
     > Eliot
     >
     > On 18.05.2026 16:39, Eric Rescorla wrote:
     >>
     >>
     >> On Mon, May 18, 2026 at 7:21 AM Eliot Lear <[email protected] 
<mailto:[email protected]>> wrote:
     >>
     >>
     >>     On 18.05.2026 14:42, Eric Rescorla wrote:
     >>>     As I said, i don't think this is addressing the primary problems 
with
     >>>     mapping specifications onto RFCs, which are much more about
     >>>     the ability to progressively revise than imposing some sort of
     >>>     meta-structure.
     >>
     >>     I'd like to see us focus on that ability to progressively revise 
specifications.  Is that something we might want to grab a room to discuss in Vienna? 
 I'm thinking about the following:
     >>
     >> I won't be in Vienna, but might be able to call in.
     >>
     >>       * What are the process limits?
     >>       * What are the tooling limits?
     >>       * Are the requirements of all levels of the stack the same?
     >>       * Related, what are the necessary interoperability requirements?
     >>
     >> I think there are actually a number of related issues:
     >>
     >> - Non-semantic changes (editorial issues, errata, etc.)?
     >> - Semantic changes that aren't really new versions (e.g., RFC 8446-bis, 
which is largely a clarification of 8446, but does in fact contain new normative 
text)?
     >> - New versions
     >>
     >> My put for non-semantic changes is: 
https://datatracker.ietf.org/doc/draft-rescorla-rfc-jit/ 
<https://datatracker.ietf.org/doc/draft-rescorla-rfc-jit/>. I suspect we'd need 
different processes for the other two.
     >>
     >> -Ekr
     >>
     >>     What I would like NOT to do would be to create the IETF version of 
the NPM/Pypi dependency mess.
     >>
     >>     Even if we keep the process the same, can we improve other aspects, 
like how readers view errata or evolutions of works like TLS.  We've got another one 
coming: TEAPv2.  We don't need to rewrite *all* of TEAP, but rather do some 
incremental changes.
     >>
     >>     Thoughts?
     >>
     >>     Eliot
     >>
     >
-- rswg mailing list -- [email protected] <mailto:[email protected]>
    To unsubscribe send an email to [email protected] 
<mailto:[email protected]>

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

Reply via email to