Re: XSLT model not "natural"? [was Re: [ANN] FreeMarker 2.3 as an alternative to XSLT]
2003-06-27 12:41:27
Jonathan,
First off, let me state I am glad to see your product and your advocacy of
it. My challenging of its implicit assumption that XSLT is difficult was
not meant so much as a challenge of the product or its approach, so much as
of the reigning dogma that says XSLT is difficult.
At 05:11 AM 6/27/2003, you wrote:
That's fine. It suits my purposes for there to be dialogue on this. I have
not responded earlier simply because I was too busy moving house.
Cool.
First of all, you surely understand that my announcement was advocacy
material and I made my biases quite clear for the reader.
Absolutely.
Well, I can't help but make the casual comment that professors and
graduate students, i.e. academics, are much more intellectually oriented
than the general population, and I would say, more likely to be open to
stretching their minds than the average person.
That's a good point: I hadn't thought of it. On the other hand, I've met
many smart and intellectually engaged people outside academia too -- and
not a few narrow-minded academics. Though the latter, not in an XSLT
course, I concede. :->
I would guess that you are a talented teacher and trainer and are good at
conveying the core concepts to people. However, the fact that you can make
a living teaching XSLT already suggests that it is not that easy!
Heh: maybe so. Yet I don't make my living teaching the stuff, exclusively.
I also use it, and I do lots of non-XSLT work. This doesn't completely
invalidate your point, I know; in fact much of the problem with training in
high-tech is that so few trainers are also practitioners. The prevalence of
poor or mediocre training that makes XSLT seem more complex than it is
contributes to its reputation, I believe.
The programming model embodied by XSLT is perfectly "natural" when it's
understood for what it is, and not confused with some other model.
The above argument is not very convincing to me. That something is natural
once you understand it is surely true of any conceptually challenging thing....
Quite so. The scare quotes were intended to send up a flag that the concept
of "natural" might itself be a little misplaced here. There's not much
natural about any of this stuff, is there? Is Perl more natural than
Python? (If so, is "natural" so good? ;-)
take the time to come to grips with what's different
If you have to take the time to "come to grips" with something, then it
doesn't "come naturally".
Mm, maybe -- but let's not let the debate slide into a rathole about what
it means to be natural. Rather, I'd like to focus on the "difficulty"
created by XSLT's not being well-suited (not being "naturally fit" :-) for
uptake by people who are so crunched for time that they can't afford to
learn and practice with their toolkit before they have to use it.
Ironically, this may mean that a little time up front can make Barbara an
XSLT "expert", writing strong and effective code, while Bob, who is no less
intelligent and experienced, struggles with it indefinitely, becoming
convinced that XSLT is hard.
Maybe this is all to say that XSLT isn't *inherently* hard or easy at all:
rather, how easily you come to it depends on other factors besides the
concepts of the language itself.
I have witnessed more than one "non-programmer" jump out of their chairs
with excitement on discovering how "easy" XSLT is....
That somebody is discovering how easy XSLT is means that they had a
preconceived idea that XSLT was quite difficult. IOW, I sense a tacit
admission of the fact that XSLT has a reputation for being difficult.
Oh yes, I admit that -- I claim also that its reputation for being
difficult is undeserved, and that its actual difficulty (for those who find
it difficult) is due to factors extraneous to the language itself.
I don't admit, however, that all the surprise is due to discovering it's
easy in the face of preconceptions. Sometimes these are people who think
*any* kind of "programming" is hard.
This leads to the question of why XSLT has such a reputation. How did it
get such a "bum rap"? OTOH, such preconceived ideas, even obnoxious
stereotypes, typically have at least some basis in the truth. If XSLT has
a reputation for being difficult, I doubt that this is just a calumny
invented out of whole cloth.
I agree: there are reasons. One reason could be that people who consider
themselves smart have trouble with it, and aren't used to thinking "hey,
maybe I'm going about this the wrong way" but are rather more willing to
blame the language. They are then confirmed in this belief by others (such
as those of us in the business of being "XSLT experts") who have an
interest in fostering it, for whatever reason having nothing to do with the
actual case.
My belief is that there could well be interest in an easier alternative to
XSLT -- an alternative that is at least easier for the subset (however
small or large) of people who find XSLT difficult. (You do yourself
recognize the existence of this subset of people.)
Absolutely. Not only that, but I believe that there's room for more than
one tool in the box, and that XSLT can and is complimentary with other
technologies.
It shouldn't be necessary always to pose problems as either/or. Rather,
let's see more innovation.
If I were in an IT department or company that hired various web page
design types, and we needed technology for transforming XML, I have to say
that I would guffaw at the idea of trying to train typical web design
people to use XSLT. I would suspect that there are very mixed experiences
out there in this regard. An alternative that could be easier for people
to learn might well be attractive to people.
Perhaps: but you might also find it is as easy to oversell something else
as it has been to oversell XSLT -- and for similar reasons. For example,
its being broadly applied to solve problems outside its original scope --
in this way, XSLT has been a victim of its own success. People had so much
need for a powerful general-purpose transformation language, that they (we)
have pulled into service a tool designed for "down-translations", not
really general-purpose at all.
In any case, thanks for sticking up for yourself (we both get a chance to
be clearer); and good luck with FreeMarker.
Cheers,
Wendell
======================================================================
Wendell Piez
mailto:wapiez(_at_)mulberrytech(_dot_)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
|
|