Re: [xsl] question about identity transform

Subject: Re: [xsl] question about identity transform
From: Wendell Piez <wapiez@xxxxxxxxxxxxxxxx>
Date: Thu, 02 Nov 2006 12:32:56 -0500
At 11:51 AM 11/2/2006, Dimitre wrote:
From a user's perspective, how is it helpful to make a
distinction between processing order and result order?

It is always helpful to know the truth in advance than to discover it by accident.

 If we, as users,
can't use the word "process" (because it's reserved for implementation
details) then what word do we use?

The word "process" can surely be used -- Where did I hint it couldn't?

We can say that" the nodes selected by the expression as specified the select
attribute are *processed" not in any predefined order -- in
indeterministic way" ... or "asynchronously", if the latter is more
understandable. Both of these definitions of processing mean that
there is no specified order.

I happen to agree with both sides in this debate. Dimitre is correct both about the processing model and about the helpfulness of knowing the truth in advance, and David is correct (as is Mike) to stress how this detail of the processing model has sound theoretical foundations with practical consequences; yet Evan is also correct that experience shows that this particular point is often too abstruse to be worth stressing to beginners.

On occasion I have sometimes tried to explain this whole business to XSLT newcomers only to see their eyes glaze over, as they concluded that their instructor knew far too much than was good for any mortal soul, and had lost his grip on practical realities. I was trying to innoculate them from unpleasant and inexplicable surprises in the future, and to show them a bit more of the big wonderful picture, only to have it backfire on me. ("If the processor's execution order doesn't matter to us, then why is he telling us this?") At the worst, I'm afraid I left them with the impression that XSLT is only for coneheads, which is the opposite of the message I wanted to convey.

Yet I also feel that XSL-List is a good place for beginners to see both the wonderful abstruseness and the practical usefulness of the language. Evan has a good point about what's really useful to know when; yet readers of this list also need to be willing to indulge us. I'd like to think that all of us are rewarded by letting the experts have a little fun.


Wendell Piez                            mailto:wapiez@xxxxxxxxxxxxxxxx
Mulberry Technologies, Inc.      
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

Current Thread