XSLS: parallel worlds

Kip Hampton <[email protected]> wrote:

> Broadly speaking, XSLT is best for transforming deeply nested 
> document-centric XML (docbook, XHTML) while Perlish, programmatic 
> solutions (SAX, Twig, XML::Simple, etc) are best for shallower 
> data-centric XML.

That sounds right to me as a general rule of thumb. Also, I think it's 
important to understand that there are actually two different groups of 
people who are likely to use XML in these different ways. They are: 1) 
those who regard XML as a programming tool, as a way of structuring data 
or as a communications protocol that is an incidental part of some 
software they are writing, and 2) those for whom XML is _content_; they 
use XML to structure and add metadata to large documents (possibly 
containing graphics and multimedia) so that those documents can be 
transformed into whatever output is needed. These two groups seem to be 
largely oblivious of each other. 

I once signed up for an "advanced" XML course in which the instructor not 
only did not know how you can use XML to structure large documents and 
output the same content as a book or as online help by manipulating it 
with XSLs, he didn't believe it could be done! He was strictly an "XML is 
for programmers" guy. Luckily, we were talking face-to-face, and 
eventually wound up with a better understanding of each other's worlds; 
had we been exchanging email in a forum like this, we would certainly have 
found ourselves in a flame war very quickly. 

Peter 
_______________________________________________
Perl-XML mailing list
[email protected]
To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs