Re: [Tsvwg] WGLC on the SCTP Implementer's Guide (draft-ietf-tsvwg-sctp-impguide-14.txt)

[Joe Touch <[e-mail]>]

Brian F. G. Bidulock wrote:
> Joe,
> 
> I believe the name is ala ITU, where:
> 
>    "Implementation Guide" => "Errata"

Yes, but this being the IETF it would be more useful to indicate it as
"Update to the SCTP Protocol" than to imply that it is either a guide
(rather than a change in the required specification) or focused on
implementation (rather than specification) issues.

> Normally in the SIGTRAN WG we take the imp guide and combine it with the
> protocol spec to form a -bis (more ITU jargon) document, the final
> document being a replacement protocol spec.
> 
> I thinks that this draft goes beyond a BCP, because it make corrections
> and adds some new protocol requirements (such as max burst, cookie
> handling, fixing security holes, adding code points for protocol
> parameters requiring IANA actions, etc.  I does not contain many
> interpretations or operational considerations.
>
> I asked about the status of this document back in Feb 2003 and I was
> told by Michael Tuexen that "the SCTP IG will never become an RFC as far
> as I know.  It is the 'delta' which will be applied to RFC 2960 when
> SCTP is moved from PS to DS or PS again."  A point with which I agree.
> 
> IMHO it has little to no use as an "Informational" RFC if it does not
> actually modify the PS.  It might just as well remain a draft.
> 
> What I would prefer to see is the deltas detailed in the draft applied
> to RFC 2960 to create an rfc2960-bis draft that we could comment on and
> advance as a new, corrected and updated, PS for SCTP.
>
> But that would be a lot of work and I don't know that the authors or the
> WG are up to that challenge with all the other items on the work agenda.

Agreed on all points; IMO, most of the concerns might be addressed by
making the document title more clear, at least, regardless of what track
the document ends up on.

Joe

> 
> --brian
> 
> On Tue, 23 Aug 2005, Joe Touch wrote:
> 
> 
>>2960 is a protocol spec; this is an implementer's guide.
>>
>>It's not clear either the term 'implementer's' or 'guide' should apply
>>to a standards-track doc. Maybe BCP?
>>
>>Joe
>>
>>Brian F. G. Bidulock wrote:
>>
>>>Allison,
>>>
>>>I am confused about the "Informational" designation.  The
>>>Implementor's Guide, as you know, is written in a style that
>>>provides replacement text for RFC 2960, and uses requirements
>>>language.  RFC 2960 is, of course, a "Proposed Standard".  So,
>>>my confusion is how an "Informational" RFC can replace text in a
>>>"Proposed Standard"?
>>>
>>>--brian
>>>
>>>On Tue, 23 Aug 2005, Allison Mankin wrote:
>>>
>>>
>>>
>>>>Hi, TSVWG,
>>>>
>>>>This begins the Working Group Last Call for 
>>>>draft-ietf-tsvwg-sctp-impguide-14.txt for Informational
>>>>publication.  Please send comments and reviews before 
>>>>September 5.  A couple of reviews that state "have read it
>>>>as part of my working group participation and
>>>>view it as ready for publication" would be appreciated.
>>>>
>>>>(We're going to make this request on all the WGLC's,
>>>>not singling out the Implementer's Guide).
>>>>
>>>>Thanks!
>>>>
>>>>P.S. I'll send out an email addressing the four nits the
>>>>idnits tools finds.
>>>>
>>>>_______________________________________________
>>>>tsvwg mailing list
>>>>[e-mail]
>>>>https://www1.ietf.org/mailman/listinfo/tsvwg
>>>
>>>
> 
> 
> 

_______________________________________________
tsvwg mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/tsvwg