Brian E Carpenter <brian.e.carpen...@gmail.com> wrote:
> Actually I agree with Martin. I think there's a gap in the style guide,
> because the principle that diagrams are illustrative rather then
> normative should be clearly stated, and that is orthogonal to SVG vs
> ASCII art. (It should have been stated in RFC 2360/BCP 22, but that's
> an IETF document.)I agree that we should really say this somewhere. Probably needs an IETF-wide Consensus Call. I think that this needs to be accompanied by concrete suggestions on how to properly document a packet structure. We shouldn't all re-invent the wheel here. I would also be in favour of a machine-readable format (such as my 25-year old PAX PDL). That work probably belongs elsewhere, and deserves a GENDISPATCH round. That probably needs to go to providing normative text (machine-readable formats) for much of our existing core protocols. rfc791, rfc2460, etc. (RFC2460 section 3 text is actually pretty good, since it lists the size of each field. Many descriptions do not). Not a trivial job. The draft that was mentioned about making the diagrams parseable should be included. In principal all the diagrams could then be generated.
-- ] Never tell me the odds! | ipv6 mesh networks [ ] Michael Richardson, Sandelman Software Works | network architect [ ] m...@sandelman.ca http://www.sandelman.ca/ | ruby on rails [
signature.asc
Description: PGP signature
-- rswg mailing list -- rswg@rfc-editor.org To unsubscribe send an email to rswg-le...@rfc-editor.org
