Re: [Rswg] Making progress on the v3 format

> The RSWG is discussing what would go into a XML v4 grammar. The result of that discussion may become a RFC in the Editorial stream of RFCs. The v4 grammar is unlikely to be backwards compatible with the grammar in [RFC7991]. Similarly, the semantics for the v4 grammar is unlikely to be backwards compatible with the semantics in [RFC7991].

I didn't see this conclusion in Jay's email, though I'm only relying on my memory of the discussion that followed.

What I would like to see:

A document that describes the XML format, 
 * encompassing all of the variations that exist in the RFC series since 7991 
 * AND the variations that this group decides represent the right choice for new documents.  
 * With a clear separation in documentation between what is considered good and what needs to be accepted by a processor in order to process documents that were produced prior to the "good" definition being agreed.

A processor (or processors plural, hah) that can render something comprehensible from each of these variations.  This might need a flag to enable the processing of documents produced prior to the "good" definition, so that it is easy to eradicate certain bad practices.

It might be that we can't achieve this, because we conclude that a clean break is necessary.  At that point, we'll have a v4, but I don't think that anything I've seen so far forces us into that position...yet.

So, while this says "may become v4", it also says "discussing v4".  I think that we should avoid being so definitive about that.  We should iterate on v3 until that becomes untenable.  A hard continuity break is - at least in my opinion - part of why the v3 transition was so rough.

On Thu, Jan 5, 2023, at 11:04, Paul Hoffman wrote:
> Greetings again. Many weeks ago, before the <u> and <bcp14> semantics 
> discussions started, Jay Daley proposed a way forward for the formats 
> at:
>    
> https://mailarchive.ietf.org/arch/msg/rswg/gspp4uvuqkNEnwzAyyrazSkEwOE
>
> The proposal got a reasonable amount of support, so I have updated my 
> personal draft draft-hoffman-rfc-format-framework-as-implemented to 
> reflect it. The new version is here:
>    
> https://datatracker.ietf.org/doc/draft-hoffman-rfc-format-framework-as-implemented/
>
> Note that I'm updating my draft to Jay's proposal because the WG seems 
> to like the proposal, not because I do. I'm trying to keep my document 
> in line with what the WG wants in case the WG wants to adopt a "way 
> forward" document. (My personal preference would be to publish 
> draft-irse-draft-irse-xml2rfcv3-implemented as an RFC and then work 
> only on v4.)
>
> Comments are welcome. If what I have put here doesn't match what people 
> expected from Jay's proposal, or if tweaks are desired, it is easy for 
> me to keep revving this. If there is a desire to adopt this as a WG 
> work item, that's great, but we will have to wait for the xml2rfc tool 
> to be updated to allow Editorial track drafts.
>
> --Paul Hoffman
>
>
>
> -- 
> rswg mailing list
> rswg@rfc-editor.org
> https://mailman.rfc-editor.org/mailman/listinfo/rswg