[
Lists Home |
Date Index |
Thread Index
]
> >But in the comparison between XQuery and XSLT people are comparing the
> >non-XML textual representation of XQuery with XSLT.
>
> I actually think a keyword syntax for XSLT might be more accessible for many programmers.
There have actually been a number of candidates, but they seem to have acquired few enthusiasts. Perhaps this is because of the preprocessor problem - you want error messages to relate directly to the source you are editing.
The biggest conceptual hurdle faced by newcomers to XSLT (which can be a much bigger hurdle for novice programmers than for non-programmers) is the lack of procedural variables. XQuery of course shares this characteristic. The next hurdle is perhaps the use of template rules, which are not used in XQuery. You don't need to use them in XSLT either, though everyone will tell you that it's good practice. Recursion, discussed in recent messages on this thread, is needed only for more advanced applications, and again is used equally in both XSLT and XQuery. In fact the need for recursive programming is considerably reduced in XSLT 2.0 and XQuery 1.0 because of the availability of constructs such as "for" expressions. I do agree with the view that many people find recursion difficult to get to grips with - but then people once found the "set-at-a-time" approach of SQL equally difficult. It takes practice.
Michael Kay
-----------------------------------------
Email provided by http://www.ntlhome.com/
|
