RE: [xml-dev] SGML complexity

["Michael Kay" <mike@saxonica.com>]

> While there's certainly a lot of value in 2.0, there are also a number 
> of questionable items. The bar for fully implementing it is 
> just way too high to interest a lot of folks.

The bar is certainly quite high: even a basic XSLT 2.0 processor is probably
twice the size of an XSLT 1.0 processor. But from the feedback I get, there
are very few features that don't carry their weight. Of course not everyone
needs everything, and you're right of course that a great many tasks can be
coded in XSLT 1.0 without too much inconvenience (the language wouldn't have
achieved the success it has if that weren't the case). But it's also true
that XSLT 1.0 demands that you learn a lot of crazy workaround solutions for
pretty common tasks: when you've become accustomed to these quirks it's easy
to forget quite what a barrier they are for newcomers and those less skilled
in computer science.

Michael Kay
http://www.saxonica.com/