What will be the future improvements of XSLT?

Hi !

As a newbie in the world of XML, I've spent a full week trying to catch the XSLT "way of thinking". My first impression is that XSLT is the most interesting part of the XML technology: the concept of standard data representation coupled with standard data transformation/processing is extremely powerful.

However, I’m very disappointed by the ‘weaknesses’ of the current version of XSLT, mainly the following two:

• the non-mutability of variables (xsl:assign is not part of the standard),
• loops and templates can’t work on, nor return, result tree fragments (forbidding multiple passes, at least without dirty hacks).

Having practiced other declarative languages, such as Prolog, I join Michael Leventhal’s opinion in finding XSLT difficult (but not ugly ;-). It would be self-lying not to recognize the fact that pure declarative style, whatever the language, is hard to write, except for two-line sources of course. I want to mention that successful distributions of these ancestors included such kind of features to allow a mixture of declarative and procedural styles.

I know the ‘S’ of XSLT stands for ‘stylesheet’ but providing these features would greatly improve the range of usability of this language. I’m sure the goal of the XSL WD is not just to issue another document formatting language but to help the web community to build highly evolutive and polymorph sites.

I’ve read with great interest the discussion between Oren Ben-Kiki and James Clark on the weaknesses of XSLT and I would very much appreciate a summary/conclusion on these points to know what features I can expect to be present in future releases of XSLT.

Cordially,

Tangi VASS (tangivass@xxxxxxxxxxx)