[Ediint] Re: Document Structure Discussion Summary - draft-ietf-ediint-rfc4130bis-01

[Erik Wramner <[e-mail]>]

My comment was specifically for the case we discussed where there would be one normative document not covering the old and only referencing the old ("use 4130 as fallback when..."). In that case, we talked about having a non-normative part with clarifications for 4130. There is a long list of clarifications that are presented during the certification. It would be good for interoperability with non-certified products in the future if those clarifications were available in a non-normative document concerning 4130 if 4130bis does not cover legacy at all.

We can structure things in a number of ways, and I know IETF does not concern itself with the certification. What I'm trying to get at is that it would be valuable to be able to publish clarifications for the old 4130 in some way so that this knowledge becomes public.

What I envisioned originally was a 4130bis that was still backwards compatible with the clarifications and improvements, and then a new standard for AS/2 2.0. That does not seem to be what we are getting, hence a non-normative clarification part would be better than nothing.

-Erik

-----Original Message-----
From: Marc Blanchet <[e-mail]> 
Sent: Friday, 12 June 2026 13:36
To: Erik Wramner <[e-mail]>
Cc: [e-mail]
Subject: Re: [Ediint] Document Structure Discussion Summary - draft-ietf-ediint-rfc4130bis-01


> Le 12 juin 2026 à 02:58, Erik Wramner <[e-mail]> a écrit :
> 
>>> It would also be nice with a non-normative text (if we can't get a normative one through) that covers all the clarifications to 4130 we intended to release as 4130bis. Many things are not really changes, but clarifications. It would be a pity not to get them out somehow.
>> 
>> To me, if these clarifications are to be part of the new version of the protocol (with modern cryptos, ...), then they should be part of the new protocol specification. People implementing the old protocol are free to implement additional features specified in the new protocol specification.
> 
> We are not talking about features here; we are talking about clarifications to the old spec.

Well, there was also new features or enhancements that were discussed and presented in the last meetings.

> Many of them can be part of the new, but it would be nice to also cover the old

One of the alternatives proposed for the structure (2 documents) can cover that. I'm not taking any position on which alternatives should be taken by the working group, but this specific consideration can be handled by 2 documents.

> if the new version is a breaking upgrade, which it seems to become. Kind of "the certified products interpret this sentence in this way; if you do that, things are more likely to work".

Certification is not the remit of IETF.

Marc.

> 
> -Erik