Hello,



From my point of view, the prolog is just useful for parsing :


  version="1.0" : just in case another version would exist in the
future (multiple roots, multiple relationships,...)
  encoding="..." : so the parser can interpret correctly and
convert the following characters !

  

DOM is just concerned by the document in memory where just one encoding
is used (UTF-32 for example).



XPath expressions are written as strings which could use a different
encoding...



The result tree of DOM API calls or of XSL transformations is usually
saved or transfered : another encoding is necessary so the prolog has
to be written when serializing.



Alain COUTHURES

<agenceXML>

http://www.agencexml.com

Bordeaux, France



Mukul Gandhi a &eacute;crit :
7870f82e0809200524j372de030o9346bf7789124ec4@m..."
 type="cite">
  Hi all,
   I wrote the following blog post. This is perhaps a weird thought,
for people having worked with XML for long.

http://mukulgandhi.blogspot.com/2008/09/should-we-include-version10-in-xpath.html

This is a short description of the points I have in mind.

Could you please let us know, if my logic have any merit.