Re: State of the art in Docbook-XSL
Hi Michael!
>>>>> "Michael" == Michael Smith <smith@xml-doc.org> writes:
[...]
Michael> Currently, you'll probably get better PDF output from
Michael> Jade/Openjade and the DSSSL stylesheets (which you can
Michael> use with XML, not just SGML).
I was using the DSSSL stylesheet but I got some annoying "bugs" (like
no support for "bottom-of-page" footnotes), and prefer to put my time
on a technology with more long term support like XSL.
>> What about long term development?
Michael> The open source XSL-FO engines (FOP and Passivetex) ain't
Michael> quite there yet, but still, IMHO, the world would start
Michael> to be a better place if more people tried to make the
Michael> move away from DSSSL to the XSL toolchain.
Well, I hope I help to make the world a better place than ;).
Michael> Norm and other DocBook Open Repository developers working
Michael> on the DocBook stylesheets are focused on the XSL
Michael> stylesheets, not DSSSL.
That's the main reason of my switch from DSSSL to XSL.
Michael> The more people who use the XSL-FO toolchain, the more
Michael> more incentive there will be for improvement of the
Michael> existing open-source XSL-FO engines and/or development of
Michael> alternative ones.
Michael> There are a couple proprietary engines -- RenderX XEP and
Michael> the XSL-FO implementation in Arbortext Epic -- that
Michael> pretty much work flawlessly. So a good complete
Michael> open-source implementation of the spec is do-able.
I don't think passivetex will be able to implement everything... it's
really a different technology. But I very like the ideas behind it
and TeX is a so much great typesetting system. There is also the
Db2Latex stylesheets which looks very great too, if I can't make them
work... Too exclusive to docbook however. But my first version of my
RPG manual was in latex and I still miss a lot the quality of those
version. Current stylesheet aren't just quite to the typographical
quality of LaTeX (no offense Norm!) although yet very good, especially
with passivetex.
[...]
Michael> xsltproc is a great for working with DocBook, Xalan
Michael> isn't. There's been a known bug for a while now that
Michael> makes some features of the DocBook XSL stylesheets
Michael> unusable with Xalan. There have been some bugs with
Michael> xsltproc and the stylesheets, but once Daniel Veillard is
Michael> able to reproduce them, he fixes them very quickly. Saxon
Michael> also works with DocBook.
Encoding support (especially re-encoding) seems to be better in Saxon,
but the current support of xsltproc seems just adequate for me.
Michael> There hasn't been significant new development on
Michael> Passivetex in a long time, despite known reproducable
Michael> bugs and lots of enhancement requests. Its sole
Michael> developer, Sebastian Rahtz, doesn't have time to work on
Michael> it, I guess, and nobody else has jumped in to help out.
The version I upload has a CVS date of 2002/08/10... Sebastian is
currently in vacancy 'til 9 september but I don't think we can said
passivetex is dead. However, I think that some inherent limitation of
the implementation will never make it a completely compliant XSL-FO
processor, lack which are, IMHO, far compensated by the use of TeX
has the final rendering engine.
[...]
Michael> If/when PSGML and whatever other open-source XML tools
Michael> begin to support schema syntax other than DTDs, it would
Michael> be great if instead of W3C XML Schema, they started by
Michael> supporting RELAX NG, which is every bit as powerful (or
Michael> more) than the W3C XML Schema language, but massively
Michael> easier to learn and use and to implement support for in
Michael> tools.
Have to look at them, then... The problem between DTD vs. XML Schema
is they have a very different philosophy for the representation of
Data and so the translation is not trivial and I still prefer the XML
Schema design. But has long as PSGML doesn't support them, I will not
switch my DTD to them.
Ciao!
Fabien
--
Fabien Niñoles Debian Maintainer
fabien@debian.org http://www.debian.org
GPG KeyID: C15D FE9E BB35 F596 127F BF7D 8F1F DFC9 BCE0 9436
Reply to: