-----------------------------
Application message id: 11350250202991/12473 X400
Grade of Delivery: Normal
-----------------------------
VMSmail To information: MUVAXA::MRGATE::"mci
_mta::*emsinternet::*mbx1ietf-822(a)dimacs.rutgers.edu::su=822-List"
Sender's personal name: John C Klensin
After thinking overnight about SGML and richtext, and rereading Erik
Naggum's comments, I would like to express very strong votes for (i)
pulling richtext out of the MIME document, (ii) letting Erik prepare
some notes on the minimum changes possible to make it fully SGML
compliant and what a DTD would look like, and (iii) letting the real
SGML community have a look at what comes out.
I find the arguments about existing tools, compatibility with a
widely-used and increasingly supported International Standard, a
careful
review of generic vs formatting tags (an issue I've tried to raise
before) in the context of not assuming a particular display model,
etc.,
to be very persuasive.
We are risking here doing what many of us often accuse the ISO OSI
community of doing: adopting a not-invented-here attitude toward an
established and proven body of practice. If we are going to have
something that is nearly-SGML, or that addresses issues that SGML
claims
to solve, then I think it should either *be* SGML or we should be
prepared to write a rationale for the departures that is a little
stronger than "well, some people have complained about X".
--john klensin(_at_)infoods(_dot_)mit(_dot_)edu