Subject: Re: [xsl] linkedin discussion of "can you sell an XSLT?" From: Robert Koberg <rob@xxxxxxxxxx> Date: Thu, 4 Jun 2009 19:30:39 -0400 |
On Thu, Jun 4, 2009 at 4:10 PM, Robert Koberg <rob@xxxxxxxxxx> wrote:And of course once you compile it with a specific processor you lose one of
the best things about an XSL stylesheet: the ability to run it on more than
one processor.
JVM byte-code or .NET IL code both are not processor-specific. Fortunately, there is no need for XSLT byte-code :)
--
Cheers,
Dimitre Novatchev
---------------------------------------
Truly great madness cannot be achieved without significant intelligence.
---------------------------------------
To invent, you need a good imagination and a pile of junk
-------------------------------------
Never fight an inanimate object
-------------------------------------
You've achieved success in your field when you don't know whether what
you're doing is work or play
-Rob
On Jun 4, 2009, at 6:52 PM, Dimitre Novatchev wrote:
Many XSLT processors can compile a (at least a main/primary) stylesheet to executable code. AFAIK Saxon can compile to Java byte-code or Java source code and .NET XslCompiledTransform compiles to .NET IL code. Many years ago Sun provided such "XSLT compilation.
In this way a single application written in XSLT can be offered in executable code and this is usually a significant part of the technical protection of any software application.
What is still missing is the ability to compile a set/library of stylesheets that can be used by a primary stylesheet at runtime. Though there is not any provision in the W3 Spec how to do this for XSLT (I think things are better for XQuery), a de-facto standard/implementation would be extremely useful. Not only some existing libraries such as FXSL, FunctX, ()one put here any EXSLT *implementation*, although EXSLT is just a specification, and by default the implementations are not in XSLT themselves) etc., would benefit from easier deployment (say in browsers), but this will stimulate authors to write more such libraries -- to the benefit of the whole XSLT developers community.
I would welcome if EXSLT2 specifies a mechanism for run-time binding between a main stylesheets and one or more compiled libraries of functions/templates.
--
Cheers,
Dimitre Novatchev
---------------------------------------
Truly great madness cannot be achieved without significant intelligence.
---------------------------------------
To invent, you need a good imagination and a pile of junk
-------------------------------------
Never fight an inanimate object
-------------------------------------
You've achieved success in your field when you don't know whether what
you're doing is work or play
On Thu, Jun 4, 2009 at 1:21 PM, Paul Kiel <paul@xxxxxxxxxxxxxxx> wrote:
Hi Folks,
Had an interesting discussion that I started on a Linkedin group, but
should
probably have started here. I had originally asked the question "can you
sell an XSLT?". And by that I don't mean is it possible, but would
anyone
buy one. What is the market for XSLTs as products?
The comments clearly indicated that people are willing to (and actually
do)
buy XSLTs. They tend to come in two forms. First, a specific solution
to a
specific task. A key niche need. Second, there is the idea of XSLT as a
software platform. XSLT as programming in Xml. I've certainly thought
this
because I've been working with xml for so long that I *think* in terms of
Xml (perhaps there is a 12 step program for this). So XSLT to me is the
java of Xml. Of course no offense to java users, myself being one too.
The reason I brought this up is that while I've been involved in open
source
XSLT work as well as created innumerable ones for clients individually, I
have lots of ideas about XSLTs as general products.
Ken Holman of course highlighted a key component to this topic, which is
obfuscation in order to protect intellectual property. That would be
the
difference between open source and product solutions.
So I post these comments here wondering if you have thoughts on this
topic
as well. Would love to hear them.
Cheers,
Paul Kiel
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
Re: [xsl] linkedin discussion of "c, Dimitre Novatchev | Thread | RE: [xsl] linkedin discussion of "c, Michael Kay |
Re: [xsl] linkedin discussion of "c, Dimitre Novatchev | Date | RE: [xsl] linkedin discussion of "c, Michael Kay |
Month |