RE: [xsl] Wishes for XSL revisions ...

[Wendell Piez <wapiez at mulberrytech dot com>]

Speaking as a "non-programmer", but one who writes quite a bit of XSLT:

At 07:41 AM 12/27/01, Andrew wrote:
>1.  Is there ever a good time to use a for-each? From reading this list for
>a while these seem to be the root of most problems and are very rarely
>included in any solutions... are they really necessary? (Ive read in places
>they were intended to make the language easier to learn for beginners!)

Yes, there is. Even the most die-hard of us who disparage it will admit 
this, I'll warrant. When used properly, it can make stylesheets simpler to 
read, write and maintain. It also provides useful workarounds in edge cases 
(to momentarily change the context node inside a template, for example). 
When used improperly, it soon has the opposite effect. When it is 
improperly used it is almost always because the stylesheet designer is 
unaware of or still uncomfortable with the available alternatives.

Whether they are "necessary" depends on what you mean by "necessary".

I have never seen it argued that it was "intended" to make the language 
easier for beginners, and personally would doubt this. (If you or any 
reader can quote this argument directly, with any evidence for it, it might 
give somebody something to rain on. :-) I have seen it argued that it does 
this -- but personally I think this argument is specious, not too unlike 
saying that a cake mix is a good way to start learning to bake. It is not 
that you can't learn XSLT that way -- indeed, some proponents of this point 
of view certainly know the language and are good teachers, tuned in to how 
people learn. But I would argue that experience shows that xsl:for-each is 
most dangerous in the hands of a beginner who does not have the advantage 
of direct guidance by an experienced practitioner. It might provide an easy 
way in if you are so guided -- for about an hour or so. Beyond that, they 
better take you to templates or you'll get in trouble. (My $0.02!)

But all this is just generalization: specific cases are strongly affected 
by the experience, background, expectations, temperament and style of the 
learner as well as the nature of the data, the requirements for the 
transformation, and other unknowables such as whether you're ever going to 
write a second stylesheet, and what *that* one will have to do.

>2.  If a variable cannot be changed after it has been assigned, why call it
>a variable?  Ive read the reasoning behind this, and yes strictly it is a
>variable, but a more appropriate name would make sense.

It makes sense if (a) you know the term "variable" from algebra, not Perl 
(the extremely naive case), or (b) you are aware that technical 
terminologies are by nature somewhat slippery when you try to carry them 
across domains (the very sophisticated case) and don't set too much store 
by *any* name.

I remember in ninth grade feeling that the BASIC statement "x=x+1" made no 
sense, but I was willing to bend my brain to live with it. :-) (I later 
learned to make the distinction between assigning a value, and claiming an 
equivalence. This distinction is actually quite useful in the study of 
human rhetorics -- and maybe in understanding XSLT.)

>The above makes XSLT non-intuitive, and therefore harder to learn.  Is there
>another language where you cannot change the value of a variable?

I'd submit that there are cases where "harder to learn" also means "better 
to learn". Paradigm shifts and all that. I'm not arguing that this is one 
of those cases. As to your question, that depends on what you mean by 
"variable" -- a predefined thing (let's see the definition), or whatever 
the language in question calls a "variable"? I believe there are languages 
that have something named a "variable" with properties similar to the thing 
XSLT calls a variable.

Thanks for your comments, Andrew.

Regards,
Wendell


======================================================================
Wendell Piez                            mailto:wapiez@mulberrytech.com
Mulberry Technologies, Inc.                http://www.mulberrytech.com
17 West Jefferson Street                    Direct Phone: 301/315-9635
Suite 207                                          Phone: 301/315-9631
Rockville, MD  20850                                 Fax: 301/315-8285
----------------------------------------------------------------------
   Mulberry Technologies: A Consultancy Specializing in SGML and XML
======================================================================


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