RE: Generating high-level formatting output

Subject: RE: Generating high-level formatting output
From: "Sebastian Rahtz" <sebastian.rahtz@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>
Date: Fri, 18 Jun 1999 22:18:19 +0100 (BST)
Didier PH Martin writes:

 > I see it [XSL] as a new language based on "<>" markups. There are
 > others like DSSSL based on a different syntax (scheme based)
sorry, i think this flies in the face of history. XSL was invented by
a committee, largely DSSSL people, who wanted to make a more
acceptable syntax face on DSSSL. Yes, technically its a new language,
in practice its the DSSSL we know and love, recast.

 > genetically improved DSSSL. This is what some would us to believe. DSSSL can
 > become what _we_, as users, decide it will be. There is always ways to
 > propose to ISO (but this time with experiments) the result of a community
 > work.
I suspect thats harder than you think, changing an ISO standard... is
the ISO DSSSL committee meeting at all?

 > Actually the reality of the XSL world is that several individuals or groups
 > are creating their own XSL implementation. To keep inter-operability, they
 > have to stick to the "recommendations". 
thats unfair. there *is* no XSL Recommendation yet, just a draft, so
its hardly surprising if test implementations differ.

 > The case for OpenJade is different. There is only one DSSSL implementation:
 > Jade and its new successor OpenJade.
thats a bit hard on Ralph Ferris, so say there is only one DSSSL
implementation! not that Jade *is* an implementation, technically. its 
a partial implementation, it would hardly get ISO approval.

 > bazaar. This is not yet the case of any XSL processor. In fact, XSL
 > processors are more controlled by players like Microsoft, Lotus, IBM and Sun
the XSL processor I suspect most people have used is by James Clark!

 > than it is based on the common effort of a community. OpenJade can become
 > what we want it to become and DSSSL to.
no. you cannot have it both ways. the argument for DSSSL is that it is 
a published formal standard. start changing it, and it isnt
DSSSL. OpenDSSSL, maybe. We can beaver away at OpenJade, great, I am
all in favour, but be wary of extending DSSSL

 > the best of their knowledge. I cannot say the same thing of the XSL universe
 > dominated by Microsoft, IBM and Sun (java). In OpenJade, we try to keep the
 > core code transportable and adapted to other platforms like the different
 > flavor of Linux, Unix, windows. What is common in this diversity is
 > OpenJade.
I dont really see this. XSL is, of course, like DSSSL, portable and
open. controlled by the (fairly evil) W3C members, of course, which is 
worrying. and the implementations in Java are also, of course,
portable (Java also being pretty open)

Sorry to sound like a XSL nerd, but all this talk reminds me of the
other world I live in, TeX. The characteristic attitude there is also
"we are small but we eat organic food and know we are right". I am not 
sure its healthy.

Anyway, maybe we should stop this thread. As you know, I am not
anti-DSSSL or Jade, far from it, so best to just get on with OpenJade!

sebastian


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


Current Thread