Jeni Tennison writes:
 > Hi Mike,
 > 
 > > As for the division between XPath and XSLT, I think the best way of
 > > rationalising the split is that XPath is for selecting information
 > > from the source document, XSLT is for constructing the result
 > > document. It should never be necessary to use multiple instructions
 > > at the XSLT level in order to extract a single piece of information
 > > from the source document.
 > 
 > I'm beginning to understand that part of that separation is also
 > between simple types (strings, decimals, dates etc.) and complex types
 > (XML). For example, even in XPath/XSLT 1.0, you can construct a string
 > from within XPath with the concat() and format-number() functions. So
 > FLWR expressions in XPath 2.0 would be used for generate sequences of
 > simple values whereas xsl:for-each constructs in XSLT 2.0 would be
 > used to generate complex values.
 > 
 > ... but then what are the functions like xf:copy() and xf:shallow()
 > doing in XPath? They seem to be about constructing the result
 > document. Aren't they?
 > 
 > >> Can you (or someone) reassure me that there will still be automatic
 > >> conversions between value types such that we won't have to be
 > >> casting/constructing specific data types all over the place?
 > >
 > > This is a matter of intense debate. I wish I could give you such an
 > > assurance but you'll have to wait and see what gets published.
 > 
 > OK Mike, I'll trust you to fight the good fight.

Is there some way that we should make it clear to the working group
how big Mike's "constituency" is?




 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list