forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Marc Portier" <...@outerthought.org>
Subject RE: XHTML2 I have read it and I like it... GO READ IT!
Date Wed, 14 Aug 2002 09:37:03 GMT
some snap thoughts...

[disclaimer] I'm no docuhead, so I really don't see myself as the DTD and
document organizing expert, yet this comes to mind:

- having the pragmatic view on these DTD things I see a big resemblance to
our current one, so the naieve question 'why not reconsider?' comes quite
spontanous
- from the libre/book.xml point of view I like that <nl> :-) (can i be
excused?)
- since we want to have other tools generate content to be included in the
site (javadoc, syntax-highlight...) support for a more standard schema makes
sense (e.g. redoing the alexandria jxr to spit out what we want)
- indeed people with more skills in this area (that have the brain, nerve
and patience to read the full spec on xhtml, which is very unlike me) should
have their say, I'll be happy to learn from this discussion

> > When it does come time for this, is it better to experiment
> > in Forrest's scratchpad, rather than manage branches?
> > We already have some scratchpad build targets for doing
> > transformation and processing of other document types.
>
> But this is about converting all the documentation in XHTML, should I
> copy all of it in scratchapd?
>

mmm, I'm not particulary fond of the branches idea either

1. purely community-wise I'ld like us to come to aggreement before massively
proceeding in all directions, smaller tests and try-outs can be done
off-line and presented to each other, no? (I see how David comes to
proposing scratchpad for those that would have the feeling that these tests
need to be in cvs.  In the same line of thought: if that person would need
ALL documents to be converted to proove a point...  :-))  There is time
enough to convert ALL documentation after we've agreed on doing so.

2. purely feature-wise we should have a system inside forrest that allows us
more easily to handle new upcoming DTDs like this.  It would also allow our
users to take advantage.

So this brings me back to:
as I said in another thread: the dream of the content-aware pipeline isn't
realized yet, so we need to think about how forrest can elegantly learn
about new src-content types (some of them auto-generated by other
tools/cents) and thus new pipelines to publish them.

(I'm more and more thinking of reusing layout.xml and content-type
describing files to be merged into a xslt?-process-generated sitemap... that
in combination to **.dtdx.html like URL-suffixes to get the pipelines going)


> OH well, let's wait for Steven to come back even only to discuss, or the
> sky will fall...
>

Steven is a big guy, he can handle loads of catch-up reading :-)
If momentum is (t)here, his absence should not keep us from discussing and
investigating. (don't think he'ld like us to be paralized)

(oh man, I can't wait to have him around again: moaning on several 1000 msgs
in his box killing his rules etc etc :-))


-marc=


Mime
View raw message