mrex at sap.com
Fri Jul 27 13:48:58 PDT 2012
Paul Hoffman wrote:
> Julian Reschke wrote:
>> That being said: it doesn't really matter. Conforming HTML5 can
>> easily be transformed into XHTML5. But if we do that, we'll need
>> to change some of the rules currently in the draft, for instance
>> with respect to closing elements (for some elements in HTML,
>> such as <link>, it's not even optional).
> The possibility of these kinds of changes argues for a stable
> (at least 10 years) format for the canonical format and multiple,
> possibly-changing additional formats. This is the basic logic in
> draft-hoffman-rfcformat-canon-others for using a long-lived profile
> of XML as the canonical, non-changing format.
After taking a look at the visual terrifying disaster here:
I'm convinced that using HTML for a submission format would be an
extremely bad idea. It would require huge amount of manpower for
the RFC Editor to rework HTML-based submissions from independent
authors in a fashion that the result will at least have superficial
resemblance in style and coming from the same SDO.
When I compare the style of ITU-T X.509 11/2008 document with the
above HTML document, then the author of the former shows to have
an understanding of typesetting with respect use of type faces,
font size, use of boldface, color and text background style.
Compared to TeX, the XML document structure is significantly more
painful to write and maintain and for HTML it is a nightmare in how
much additinonal typing&maintenance it requires, and how much information,
that is conceptually redundant, must be manually and seperately maintained,
making it prone to inconsistencies.
TeX (manual section numbering):
\beginsection 5.3. Sequence Diagrams
LaTeX (auto section numbering):
rfc2629-XML (auto section numbering, I assume):
<section title="Sequence Diagrams" anchor="websequencediagram-sample">
draft-hildebrand-html-rfc-2012-07-16 (manual section numbering):
<div class='section' id='websequencediagram-sample'>
<h3><a class='self-ref' href='#websequencediagram-sample'>5.3.</a> Sequence Diagrams</h3>
More information about the rfc-interest