Re: XSLT 2.0 *and* XSLT 1.0 validation -- how to?

On Sun, 24 Oct 2004 23:47:20 +0100, David Carlisle <davidc@xxxxxxxxx> wrote:
> 
> > So better to have the stylesheets as 1.0, but individual templates as
> > 2.0?
> 
> Not necessarily better, just different. You get different kinds of
> errors and/or maintenance problems in the two cases, which is best
> depends on various factors.
> 
> Basically the underlying 2.0 model of un-nested ordered sequences poluted
> with XSD types is sufficiently different from that of 1.0 that there
> will always be behavioural differences between a 1.0 and 2.0 processor,
> even in backward compat mode. Given where they chose to start from I
> think they've made a pretty good job of compatibility, but still they
> could have started from somewhere else...
> 
> I think I'd view the 1.0 compat features in 2.0 as a way of easing
> transition from one to the other not something that you'd really want to
> build on as a long term feature of the design of the stylesheets.


I'm trying to think how such mixing can help me learn XSLT 2.0 and I'm
definitely against it -- it is much more confusing to use two
languages (mixed in unspecified proportions) than just one albeit new
language.

Also, implementing this feature would most probably put considerable
strain on potential developers and even completely discourage them
from going on with this undertaking.

I wonder if there are XSLT 1.0 programmers, who find this mixed
language feature useful?

Cheers,

Dimitre.