Re: Which engine? (RE: JavaScript and XSL)

----- Original Message ----- 
From: Eric van der Vlist <vdv@xxxxxxxxxxxx>

> I reckon that for companies who cannot afford participating to the
> support of the open source products they are relying on, XT is currently
> a very bad choice.

XT is dead, I think. It has no saxon:evaluate ;-) 
 
> The real work is yet to be started and in the meantime I have though
> worth publishing some extensions developed for my own usage (running on
> the unmodified release of XT) and providing some guidance to XT users
> asking questions.
> 
> Or shouldn't I ?

If you care about not being blamed, you should not do anything.
Do nothing - and nobody will blame you.  
I think you already know that. ;-)

> > I don't think these things are bad ideas, but one person's requirement for
> > productivity is another person's toy... if XT is good enough for you
> > as-is, these things are great new features. If it's not good enough, these
> > things are annoying deviances from more productive development.
> 
> Until now, XT has been good enough for me.

For a while it was good enough for me ( for example, because 
it is Java 1.1, keeps me free of DOM e t.c. XT is elegant ).

Then XT become not good enough for me. For many reasons. 

I think,  the real story is that at some point Mike realized that XT is not 
good enough for him. Why this simple situation has to be reflected in such 
a strange discussion I don't understand. I don't understand most of the 
points Mike is making, actually. Or I'm better to think that I don't understand.

XT is still ;-) nice, robust and very well embeddable thing. Even I'm not 
using it any longer ( I'm using SAXON ) this fact does not affect the 
technical points I was making for a couple of months.

I've tried embedding SAXON ( with Trax ) and I've tried embedding XT.
The real ( not hypotetical ) situation with embeddability could 
be somehow illustrated by the latest source code of XSLScript 
version 0.7 ( based on SAXON ) and  XSLScript version 0.5 
( based on XT ). Latest distribution of XSLScript contains both things. 

Rgds.Paul.

From: Mike Brown <mike@xxxxxxxx>

> We moved to SAXON recently, in spite of the
> slight, but noticeable, performance hit, mainly because of the robustness
> of the product --in particular, its support for keys and proper HTML
> output when indent="yes", something not even the latest MSXML can
> achieve

Using the word 'robustness'  instead of  'conformance'  is nice. 
I think you've used the wrong word. Mistyping or something. Right?  

By the way - have you tried to debug  buggy stylesheets with, say,  
complex recursion with XT and SAXON ?  Do your stylesheets 
place some considerable load on extension functions written in Java ? 

SAXON is good. XT is good. Both are real stuff. 

<rant>
Politics is no good. Politics is not a real stuff.
</rant> ;-)




 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list