Re: New/old pattern syntax, why can't we have both ?

Subject: Re: New/old pattern syntax, why can't we have both ?
From: Scott Lawton <scott@xxxxxxxxxxxxxxxxxxx>
Date: Tue, 25 Aug 1998 10:56:00 -0400
Quick reply to the thread's originator: if I were designing XSL as a
commercial product, I *would* support two syntax options since I agree that
both are useful.  But XSL is supposed to be a neutral interchange format so
two variations are counter-productive.

On to the recent thread....

>Mark_Overton@xxxxxxxxx wrote:

>> This may be the case now because of the lack of tools, but it absolutely
>> will not be in the long run.  You will view XSL through some sort of
>> abstraction.  If this doesn't happen, then XSL is dead.

Paul Prescod replied:

>XSL has to catch on long before there are widespread GUI tools for it, in
>order for it to have critical mass enough to make the GUI tools feasible.

>Anyways, why should we force (or "encourage") anyone into using a GUI tool
>if they don't want to?

How about text-based tools?  Wouldn't it make sense to put the task of
writing an extra parser (to convert the short form into XML) onto a few
tools developers rather than to every XSL developer?

>I must admit, I am bothered by the moral absolutism of the complainers.

I respectfully submit that there's more to it than that.

>I didn't mind the old element-based syntax. It was nice how it reflected
>the structure of the document section being matched. If someone wanted to
>make usability arguments like that, I would be very receptive and might
>well support the element-based syntax.

Ah, excellent.  OK, start with a blank slate.  Let's say we want to
generate tags in the output.  Wouldn't it be nice if we could just include
the literal tags?


Well, we can!  (Thanks to the xsl: namespace.)

Now, what's the simplest way to match a set of tags?  Wouldn't it be nice
if we could just include the literal tags?

This "query by example" makes easy cases easy.  Yes, complex queries are
more complex (though no more so than the original submission) but I think
there's incredible value in starting from a simple foundation.  And, it's
nice to make the query syntax parallel to the generation syntax (or
whatever the term).


 XSL-List info and archive:

Current Thread