Re: Writing I-Ds and RFCs using XML

>  As a Co-Chair of a joint IETF/W3C WG, I was thinking that it would be
quite
> nice if the W3C and IETF agreed on a common specification schema/DTD.
(This
> could make my life easier when I find a WYSIWYG XML/CSS editor I'm happy
> with for editing specs! <smile>)
>
>  Marshall, the efforts at the W3C [1] to date have been informal within
the
> XML activity and on the spec-prod list [2]. However, the folks involved
have
> come up with quite a nice DTD [3]. What is the status of usage/deployment
of
> your Informational RFC on a schema/DTD for RFCs and have you seen [4]? Any
> interest in harmonizing/subsetting/profiling? And then there's always the
> XSLT option...
>
>  [1] http://www.w3.org/MarkUp/SGML/spec-mgmt.html
>  [2] http://lists.w3.org/Archives/Public/spec-prod/
>  [3] http://www.w3.org/XML/1998/06/xmlspec-report-19980910.htm
>      http://www.w3.org/XML/1998/06/xmlspec-19980910.dtd
>  [4] http://www.ietf.org/rfc/rfc2629.txt
> End Forwarded Text ----

hi. i agree that it would be nice to have a common syntax for authoring w3c
and ietf work products. a few things come to mind:

first, you really ought to be talking to the rfc-editor@isi.edu, since
they're the ones who manage the rfc editorial process.

second, the rfc editorial "rules", which you can find in rfc 2222, have
some, but not a lot of overlap with the goals listed in
http://www.w3.org/XML/1998/06/xmlspec-19980910.dtd

in particular, you don't get to do pdf, or html, and postscript when
publishing an rfc. (well, they'll allow postscript for figures, but you'll
get so much feedback that you'll prefer to chew off your right arm rather
than actually publish anything in postscript...) this means that the
translater is going to have to be very, very smart, or you'll need to allow
dual markup for the same content. messy.

there are about a dozen or so people who write I-Ds/RFCs using the 2629
format. there is one translator for it that produces .txt & .html versions
from it. however, ascii art is the lowest common denominator. the entire RFC
database is available in biblio format for including as references (which is
a major selling point).

it's not really clear to me that a harmonization effort would work, given
the limitations in rfc 2222. however, i could certainly imagine someone
writing some XSLT that would translate from the 2629 format to whatever the
new format becomes, presumably because the new format would be a functional
superset.

/mtr

Received on Tuesday, 13 June 2000 15:31:39 UTC