Altova Mailing List Archives
>xml-dev Archive Home
>Thread Prev - XSLT 2.0 and XSL(-FO) 2.0
>Thread Next - Re: [xml-dev] XSLT 2.0 and XSL(-FO) 2.0
RE: [xml-dev] XSLT 2.0 and XSL(-FO) 2.0
Date: 5/12/2002 7:11:00 PM
> Since XSLT 2.0 will inevitably impact XSL-FO 2.0 then > concerns similar to > those we have been discussing arise there. > Since XSLT and XSL-FO are developed by the same working group it's likely, if anything, that XSL-FO users have more influence on XSLT development than other groups with an equal claim. I'm not sure what point you're trying to make? Frankly, Andrew, I think you've got things totally out of proportion. The principal concepts of XSLT and XPath were laid down with the 1.0 specs. Anyone who has mastered that is going to have very little difficulty picking up the new features. People coming to the language new are not going to need to learn the difficult programming tricks that were needed in 1.0 to do common tasks like grouping, sorting dates, or computing subtotals. The only thing that is conceptually new is the integration with XML Schema, and by the time we've finished the hope is that you won't need to understand that if you don't want to use it. There's a lot more formal specification this time round, but that's intended for implementors, not for users: users, as always, will pick it up by sitting next to Nellie. The language family has grown, but it's still incredibly small compared with SQL or Java. If we've got it wrong, then (as happened with SQL) the user and vendor community will stick with 1.0. I don't personally believe that will happen: you only need to see how many of the difficult coding problems raised every day on xsl-list become trivially easy with 2.0 to see why. Yes, we all have our favorite list of features in XPath 2.0 that we think no-one needs, and hopefully some of them will go, but they all have their champions, so throwing things out is not easy. There are even people who like schemas. It doesn't actually worry me unduly if there's the odd function in the function library that no-one uses. It does worry me if we leave out features that mean people have to continue writing arcane things like count($x | $y) = count($x) to test if $x and $y are the same node. Michael Kay Software AG home: Michael.H.Kay@n... work: Michael.Kay@s...