RE: Request for a poll: (was RE: Datatypes vs anarchy)

[Gavin Thomas Nicol <gtn@ebt.com>]

>Use of validating parsers

Seldom.

>Use of namespaces

Seldom.

>Use of pipeline processing techniques (e.g. xml in, xml out)
 
Yes

>Use of XPath

Yes

>Use of entity structure

Seldom.

>Use of internal document type declaration subsets

yes 

>Use of notations

Seldom

>Use of DOM levels

Level 1 core almost exclusively

>Use of SAX1/2

Only SAX2 now.

>Use of JDOM/Pyxie/etc.
 
No
 
>Use of SGML tools e.g. nsgmls

Yes

>Use of lexical parsing e.g. regular expressions
 
Yes
 
>Use of mixed content
 
Yes

>Use of ID/IDREF

Yes

>Thoughts?

Most of the things I need have existed in one form or another
for about 1-2 years. I have adopted incremental improvements.

Namespace awareness (rather than just using Qname) has been
a great source of pain.