RE: html to dsssl ?

["Didier PH Martin" <martind@xxxxxxxxxxxxx>]

Hi Sebastian,

Sebastian said:
We are back here again. The W3C did not invent XSL in opposition to
DSSSL. It formed a group to look at a `DSSSL for the Web' and that
group (including many DSSSLers) reluctantly decided to switch
syntax. If they could have persuaded people to use (), they would have.

Didier says:
Don't get me wrong. I do not say that W3C invented XSL to be against DSSSL.
Wahtever the good reasons the editors of XSL got to use a new syntax. They
choosed to start from scratch with a XML based syntax. I also, do not mean,
because some ex-DSSSLers are now part of the W3C working group that XSL is
the best thing on earth. it also, do not mean that because they left DSSSL,
that this latter dead or that no improvements are possible. DSSSL-2 could
learn from CSS, XSL and from the experiance gained from DSSSL-1. If we also
include what we know of languages like Balise or Omnimark DSSSL-2 could get
an improved design. And finally, it does not mean that because W3C make a
recommendation for XSL that the world end there and that we shouldn't learn
and improve with DSSSL-2.

Sebastian said:
we probably agree that if CSS takes over the world, we can all just
leave the room; last one out turn off the light of real typesetting...

Didier said:
So there is room for real typesetting but not for the common mass (anyway
like it always have been) until we improve the model so that Ben schederman
UI principles find their way in the typesetting world.


Sebastian said:
You have a PDF backend, using a very sophisticated page makeup engine
(TeX). Since we can see that even TeX has problems with DSSSL, there
is a way to go. I repeat, the DSSSL model is almost entirely _unproven_

Didier says:
Yes but not an easy process. Here is waht I am playing right now but still
very incomplete and still a work in progress.
a) I create a DSSSL style sheet with common DSSSL constructs
b) Include in the Talva SGML/XML kit the command line for this script but
with a small plus "t- pdf"
c) In the Document Explorer I just then click on the document icon - and the
next thing that I see is the document transformed into PDF and displayed.
Now, I can check if I got the right formatting.

What new in the process I described? Nothing except ease of use. This is
what is lacking now with the Tex backend.

If only the actual JadeTex generated code could be interpreted with the IBM
plug-in, we would have the same ease of use. Even better, printing from the
plug-in. If JadeTex would have its output directly interpretable with the
IBM tex plug-in we would have for a XML document:
a) create the DSSSL script and (for example) include a style-sheet
processing instruction in the XML document such as: `<?xml-stylesheet
type="text/dsssl" href="myscript.dsl" media="screen,tex"?> (note the media
attribute)
b) if you are in the Document explorer environment you just click on the
document icon in the browsing view to get the document displayed in the
document view area with the IBM plug-in
c) print it form there

What we gained, simply ease of use.



 > Sebastian said:
 > do I, as a user, care what the implementation language is? no....
 >
 > Didier says:
 > but you may care about the speed no?

Sebastian said:
minimally.

Didier said:
Its OK then, if speed is not important most XSL engine are more than OK. For
me it is important because I use more interactive tools like the document
explorer and I like to just click on the document and see the rendering. The
faster I get the rendering the best it is. If only I could get instant
rendition I would be happy. It is only a question of personal taste. I
respect your taste.


Sebastian said:
good lord, XSL is still in _draft_, and you talk about learning from
its flaws????? why not get in there and make XSL right????

Didier says:
I did, just look at all my intervention in the XSL list. You'll see that I
insisted on the XSLT separation from the XSL specs, mentioned several flaws
or non-precise statements in the specs, did some experiments, etc... So, I
did my part. But I am also interested by DSSSL-2 so I choose my camp. Also,
because XSL will be mostly what Microsoft and AOL will decide it will be :-)
Also, because I found that there is enough room for other alternative
languages. Also, because OpenJade (and Thanks to James) allows us to do
things that people rushing to XSL forgot it can do. Also because the
OpenJade team is composed of great people. Also, because DSSSL is useful and
if we can (with the OpenJade team) make it even more useful and practical,
why not. Also, because I have learned form XSL and CSS. Its maybe time to
bring back this knowledge to DSSSL-2 :-)

regards
Didier PH Martin
mailto:martind@xxxxxxxxxxxxx
http://www.netfolder.com


 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist