Re: [xsl] XSLT/XPath 2.0 (was "Identifying two tags...")

[Dan Holmsand <holmsand at myrealbox dot com>]

Hi Stuart,

Stuart Celarier wrote:
> Specifications and standards are not intended to be textbooks or good
> exposition. Just try reading the ANSI/ISO C++ Standard. It is darn good
> as a specification, in that it is (generally) straightforward to tell if
> a C++ compiler conforms to the Standard; but it would be a dreadful way
> to learn to program in C++.

Well, clear writing is a difficult thing to achieve. I seem to have 
proven this myself...

I'm not trying to bash the writing in the C++ standard (I haven't even 
read this). Nor do I have any problems with the writing per se in e.g. 
the Schema spec. I really appreciate that the standard is trying to be 
terse and compact.

But I have a real problem with the complexity of the subject matter that 
these standards describe. XSLT is evolving from a fairly straightforward 
little language, that can be described in a couple of pretty 
straightforward specs, into a language that has to be described in 
hundreds of pages full of terse and compact definitions. That's my problem.

> The language of specifications is necessarily complex and specific. I
> don't think that their use of language is contrived, gratuitous, or a
> sign of some pretension to erudition. The comparison to Martin Heidegger
> isn't particularly apt, other than to say in both cases that the writing
> is dense and strives to be precise. If you see that as obfuscation, you
> may be missing the point. If you think that makes specifications hard to
> read, you're not alone, brother.

Well, I must admit that I had to look up "erudition" (English is not my 
first language, as is probably obvious). Good word.

Martin Heidegger is obviously an excellent writer. He manages to invent 
his own language, and uses it to describe his own universe, and (at 
least nearly) gets away with it.

But I don't think that XSLT has to be that hard to describe: I think 
that the dependency on the complexities of XML Schema gives me precious 
little benefit, compared with the headaches it causes me.

Also, let me just note that in spite of the darn good C++ Standard, it 
*has* taken quite a while for most compiler writers to achive full 
standard compliance. Complexity always costs.

/dan




 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list