[rfc-i] Arguments about markup

Nico Williams nico at cryptonector.com
Fri Feb 1 12:29:12 PST 2013

In section 2.2.2 I see this:

   Arguments against a markup language as the Revisable format:

      *  Having the Revisable format be in a markup language instead of
         in a simple text-formatting structure ties us in to specific
         tools and/or tool support going forward.

This isn't obviously true, and we have existence proofs of it being false.

Currently we have several markup languages for Internet-Draft
typesetting that produce XML in the xml2rfc schema, while xml2rfc can
output roff.  At least one tool can also output LyX/TeX/LaTeX.  So we
have: a) several markup languages (including two mark-down types, one
XML, and LyX), b) several typesetting languages (two, really: roff and
TeX), c), several different tools, d) the ability to convert all of
the markup inputs to roff, and soon enough we may be able to convert
in all directions between the markup languages anyways (since this is
easy to do with XML and XSLT).

I would also note that there are several tools for typesetting
directly from XML in the xml2rfc schema -- at least three: xml2rfc
1.x, xml2rfc 2.x, and Greenbytes' XSLs.  I intend to eventually
produce an XSL that outputs LyX with proper formatting, which will
then allow typesetting to PostScript, DVI, PDF, ... and all in a text
format editable with $EDITOR but also with a wonderful WYSIWYM GUI.
This is the existence proof.

I'd go further: XML is extremely malleable because of XSLT.  It's
perfectly reasonable to have XML on the one hand as a format used by
authors and roff/TeX/whatever as an interim step in the RFC-Editor
editing process.  (Indeed, this is what we do today in many, if not
most cases.)


More information about the rfc-interest mailing list