Re: Application Design

[Paul Tchistopolskii <pault12@pacbell.net>]

<Manos Batsis>
> I would add that xslt:
>  
> 1) it's not only serverside
> 2) it's the best at what it does
<snip/>
>  I'm not an xslt wizard or something
</Manos Batsis>

Becoming one may ( will ) change your 
point of view on XSLT. I estimate 
learning time of one year. Have you read the 
XSLT programming reference by Michael Kay?
It is 'must read'. Helps to understand how many 
things are hidden. To call himself xslt wizard, 
one should learn much more than just 35 xslt 
keywords.

What XSLT 'does' is :

<quote source=http://www.w3.org/TR/xslt>

Abstract 

XSLT is designed for use as part of XSL, which is a stylesheet 
language for XML.

...

XSLT is also designed to be used independently of XSL. 
However, XSLT is not intended as a completely general-purpose 
XML transformation language. Rather it is designed primarily for 
the kinds of transformations that are needed when XSLT is used 
as part of XSL.

</quote>

Sure, XSLT is good as part of XSL, which is stylesheet language for XML.
Sure it is the best at what it does. There is no other language designed as 
part of XSL. Will XSL survive -  that's another story.

<Don Park>
If I am right, XML consultants would be doing a disservice to our clients 
if we pushed XSLT-based solutions.  I have not qualified my statement 
with 'unless they have in-house XSLT expertise' because, if they had 
XSLT expertise in-house, they are not likely to need XML consultants 
in the first place.

Sad but I see what I see.
</Don Park>

XSLT *has not been designed* for many things that  people do 
with XSLT.  

In my opinion:

- XML consultants, who build on complex and 'neat' XSLT stylesheets, 
instead of trivial perl scripts, are sometimes doing a disservice to 
their clients.

- XML consultants, who build on  brutal XSLT stylesheets, are 
providing a good service to their clients.

Rgds.Paul.