DV said:
can anyone tell me or point me to somewhere what the exact differences
(in terms of when would I use what) between XPath, XLink, XPointer and
DOM are?
Does it require special software to use any of these, or can any
XML-aware software (parser/validator or xsl- enginge) use any of
these?
Are these technologies already firmly standardised or would you
consider them "moving targets" and advice caution when basing
workflows on them?
Well, all are W3C recommendations (standards) so visit the W3C web site
to find out more.
As for XPath, yes, XPath 1.0 is a recommendation (firmly standardized)
since November 1999:
http://www.w3.org/TR/xpath
and XSLT (XSL Transformations) 1.0 which uses XPath has the same date.
Work on XPath 2.0 and XSLT 2.0 however is still going on so these are so
far not firmly standardized.
As for the DOM, the W3C DOM group has so far produced three levels, see
http://www.w3.org/DOM/DOMTR
where W3C DOM Level 1 is a recommendation from 1998, while most modules
of DOM Level 2 are dated from November 2000, and where the first modules
from DOM Level 3 have also reached recommendation status in 2004.
As DOM has different modules for different aspects and some modules are
extensions it is hard to give a judgement for the whole DOM but in my
view DOM Level 2 Core is well established when you look at
implementations in web browsers or with DOM parsers in different
programming languages.
As for what the DOM is and covers see
http://www.w3.org/DOM/#what
http://www.w3.org/TR/2000/REC-DOM-Level-2-Core-20001113/introduction.html
http://www.w3.org/TR/2004/REC-DOM-Level-3-Core-20040407/introduction.html
As for XLink, XLink 1.0 is a recommendation since June 2001, see
http://www.w3.org/TR/xlink/
but it is a complex specification with many aspects and only some
features (called simple XLinks) for instance have made it into the
Mozilla browser.
As for XPointer, it is like XLink a recommendation of the W3C XML
Linking Work Group and reached recommendation status in March 2003.
As for what software supports it, any XSLT processor needs to support
XPath as XSLT makes use of XPath. Not all XSLT processor however do
probably expose an API to only use XPath.
As for the DOM there are parsers that can build a document object model
but there are also other parsers like SAX which don't do that. Some XSLT
processors can operate on DOM trees but that is not a requirement for
XSLT, it has its only tree model.
http://www.w3.org/TR/xlink/#app-reqs explains what an XLink application
is, so far I haven't worked with one, as said I only know about simple
XLinks in Mozilla.
