Re: Announce: Beginnings of a new XSL stylesheet library for UBL 1.0 documents

["G. Ken Holman" <gkholman@CraneSoftwrights.com>]

At 2004-05-20 20:04 -0400, david.lyon@tradalogs.com wrote:
>I've had a look at your resources and they seem to look quite good.

Thank you ... I'm hoping to complete them soon and am looking for feedback.

>May I ask what general direction you plan on taking these resources? If 
>you plan
>to add additional resources to possibly do other things like downloading to
>make some sort of workable business system. That is, download messages from a
>site and then print them. or maybe update a database.

I cannot begin to guess the myriad of ways I suspect people will choose to 
use UBL ... I am so excited about how popular this vocabulary is going to be.

My "general direction" is to keep what I do as modular and as reusable as 
possible so that when people do downloading, or updating a database, or 
whatever they want with UBL messages, that when it comes time to visualize 
the message that my stuff might come in handy.

>Definitely, you have a good start, but what I am sure a business would be 
>asking is how to fit them in to some existing system.

But I don't know how the existing business systems are working (I'm just an 
XML geek) so the best I can do is make this stuff modular enough that when 
someone who *does* know business systems needs to visualize a UBL document 
and the visualization they need happens to be a UN Layout Key kind of 
layout, then the committee has developed the formatting specifications and 
Crane Softwrights has implemented a sample set of stylesheets.

By releasing these stylesheets publicly, integrators who need to tweak them 
can do so without having to wait for me to make any changes to Crane's library.

The FPSC work reveals a model of what trading partners will probably need 
to do when they need to visualize their UBL messages: write formatting 
specifications.

And I suspect that trading partners will probably wish to write and 
implement (or have written and implemented) their own formatting 
specifications because they will probably not wish to use the UN Layouts.

Remember the work of FPSC is informative and not normative ... one is not 
obliged to use the UN layouts ... we did the work in FPSC as a model to 
show an example of how to write formatting specifications so that trading 
partners can follow that model when it comes time for them to write their 
own formatting specifications.

I've delivered two conference presentations on this:

   http://www.idealliance.org/papers/dx_xmle04/papers/04-01-01/04-01-01.html

>I think this would be the natural development path for the project.

The natural development for the FPSC committee project that I chair is to 
move on to look at more human computer interface issues related to 
UBL.  We've illustrated one way to approach print, now what can we do about 
input?

The natural development for Crane's stylesheet library project is to 
further some research on modularity and reusability, hopefully making 
useful resources available to everyone at the same time.

A quick comment about this library: this is my second attempt, from 
scratch, to write formatting stylesheets for UBL.  I had and implemented an 
annotation idea called LiterateXSLT(tm) for the first set of stylesheets 
and it succeeded far beyond my expectations but had some drawbacks.  But, 
within a week of me deploying those stylesheets, I got a question from a 
participant in another standards community (EAN-ICC) about utilizing my 
stylesheets in their environment.  Of course that makes sense, but my 
annotation methodology hardwired all of the effort I put into my 
stylesheets to UBL.

So when UBL 1.0 came out this gave me the opportunity to address how to 
make a modular stylesheet fragment both useful for UBL and useful for other 
groups who want to use the UN layouts.  And after much noodling, I started 
again from scratch and I think I've accomplished just that with 
ResultXSLT(tm) and without the drawbacks of the first bit of research.  I'm 
going to finish the UBL interfaces right away due to the recent release of 
1.0 and then prove to myself that I can support other document models of 
other projects using the same stylesheet fragments.  I've never seen this 
kind of stylesheet synthesis or exploitation done before my initial UBL 
work 16 months ago so I think it is covering new ground and it has been fun 
and interesting to explore.

BTW, both LiterateXSLT(tm) already is and ResultXSLT(tm) will soon also be 
free resources from our web site collection of free developer 
resources.  Follow the link in the marginalia on the right of our home page 
noted below.

Back to your "how to fit them in to some existing system?" question: I'm 
not sure ... are existing systems modular enough to accommodate XSLT 
engines and XSL-FO engines in order to exploit my freely-available 
stylesheets?  I don't know.

Perhaps we can start a thread here on UBL-Dev regarding existing systems 
and their support of print and display.  I'm guessing that existing systems 
probably use proprietary technologies for rendering rather than "plug and 
play" XSL stylesheets.

I hope this helps.

.............. Ken

--
Public courses: Spring 2004 world tour of hands-on XSL instruction
Next: 3-day XSLT/XPath; 2-day XSL-FO - Birmingham, UK June 14,2004

World-wide on-site corporate, govt. & user group XML/XSL training.
G. Ken Holman                 mailto:gkholman@CraneSoftwrights.com
Crane Softwrights Ltd.          http://www.CraneSoftwrights.com/u/
Box 266, Kars, Ontario CANADA K0A-2E0    +1(613)489-0999 (F:-0995)
Male Breast Cancer Awareness  http://www.CraneSoftwrights.com/u/bc
Legal business disclaimers:  http://www.CraneSoftwrights.com/legal