- From: Ian Jacobs <ij@w3.org>
- Date: Thu, 08 Jan 1998 17:52:50 +0100
- To: mathieu.bois@nomura.co.uk, www-html-editor@w3.org
mathieu.bois@nomura.co.uk wrote:
>
> Hello,
>
> I have got the current proposal for HTML 4.0 (and I used to had an
> older one).
> It is very well done, and I use it a lot (I have taken the .tgz file
> in order to access it locally and quicklier !).
Thanks for the positive feedback!
> Anyway, I have got one question : how was it written ?
> I mean in which original format (SGML / HTML ?) and with which
> software (if not done by hand) [and if you have a little more time to
> answer me, what did you use to convert it in other formats like PS,
> PDF] ?
Without going into the details, we did the following:
(1) We wrote it in legal HTML. Nothing fancy, in fact, but lots
of class and id attributes.
(2) We ran each document through an elaborate series of transformations
and produced legal HTML.
(3) We converted the HTML into various formats as follows:
a) Postscript: We used a patched version of html2ps (see the
Acknowledgments)
b) Text: We used Netscape Navigator to generate plain text.
c) We used Distill (from Adobe, I believe) to generate PDF
The processing of step (2) was done with lots of Perl scripts that we
wrote
by hand. For instance, for the indexes, we generate databases of
elements or
attributes, then sort and format them, etc.
We also ran the spec through a battery of tests (which weren't perfect),
including:
Spell checking: ispell
Link checking: linklint
SGML checking: nsgmls
We managed the whole thing with CVS (the three editors work in three
different
countries).
Each editor used his favorite software to author the source HTML (I use
emacs, for example).
I hope that this answers a few of your questions. Thanks again,
Ian
Received on Thursday, 8 January 1998 11:53:18 UTC