Re: [ubl-fpsc] Combining XPath and UN Layouts

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

At 2003-06-05 21:41 +0100, Stephen Green wrote:
>I'd appreciate comments on a few matters arose in today's UBL-FPSC call.

Thanks, Stephen, I'm glad these matters are coming up explicitly so we can 
talk about them and make sure we are all singing from the same song sheet.

>1) It was realised that by specifying XPaths for, say, each box/label in 
>the UN Layout, people could set these side-by-side with XPaths for other 
>standards and, for example, write stylesheets for multiple standards all 
>in one. This may help implementers understand the standards more than 
>would lengthy schema specs.

My choice of XPath in a formatting specification gives us a method of 
talking strictly about the XML instances governed by the UBL document 
models independent of any transformation technology.

While XPath happens to be used by XSLT, XPath is *about* XML, not *about* 
XSLT.  We can unambiguously address components of any XML document using 
XPath, and in so doing be totally unbiased about any transformation 
technology, proprietary or non-proprietary.

So, further to your point, yes once we document the "functioning" and 
"understanding" of what is expected of a particular layout and then include 
the addressing of the components of UBL instances to meet the "needs" of 
the layout, then this will be an invaluable template for anyone else with 
any other XML vocabulary to "fill in the blanks" and replace UBL XPath 
addresses with whatever XPath addresses will correctly address the 
information found in other instances.

>2) The above shows a (perhaps unintended) value of the UN Layouts in 
>harmonising standards and so it would seem important to ensure that the UN 
>Layout XPaths at least are completed (even if, say, just an Order and an 
>Invoice).

I'm hoping we will be able to do them all because I'm anticipating the 
Joinery and Office mappings to be much shorter and more quickly dispensed 
with in our schedule.

>3) All in all, I think we do need to emphasise the value of the normative 
>specifications, especially the normative XPath expressions and 'mappings' 
>to UN Layouts.

Precisely!  You've just described what I think fully encompass our 
normative specifications (modulo the other scenarios): our objective is to 
equip those who wish to format instances of UBL documents with everything 
they need to know, so they can go and use whatever it is they have to 
produce a conforming result.

>4) To me the most valuable deliverables will be the 
>UN-Layout-label-to-Xpath-expressions since these provide implementers with 
>a fully standard means of comparing standards, and this in an 
>internationally acceptable way.

Interestingly, clarification and direction were my objectives, not 
comparison, though I readily recognize that benefit we will enjoy with the 
finalized formatting specifications.

I was approached in December to write stylesheets for the 0p70 
release.  When I undertook the task, no-one was able to direct me or give 
me a formatting specification.  How could I, as an outsider, ever hope to 
create an implementation of formatting without being told, categorically, 
what was expected of the implementation?  So I tried to throw together my 
personal (and eventually proven to be incorrect) interpretation of a 
formatting specification given my flawed understanding of UBL semantics, 
and write stylesheets from there.

When then asked to formalize the work I was doing and chair a group, I only 
signed on provided our task wouldn't be "the writing of stylesheets" but 
would be "the writing of formatting specifications" from which anyone could 
write stylesheets, or Word templates, or whatever they wanted.  I wouldn't 
want to see anyone have the problems I had with 0p70.

>5) I accept though, that to raise awareness, wow people and  provide for 
>those less interested in the UN layouts we do need the other scenario 
>specs, example forms and, from CraneSoftwrights/Ken, the stylesheets.

And I'll be scrambling to try and do so.  I'm hoping other vendors with 
other technologies will also jump into the fray thus proving the 
flexibility and openness of our committee's specifications.

Thanks again, Stephen, for bringing this up.  I look forward to the 
comments from anyone regarding the above and how that fits with their 
understanding of the work of the committee.

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

--
Upcoming hands-on courses: (registration still open!)
-      (XSLT/XPath and/or XSL-FO) North America: June 16-20, 2003

G. Ken Holman                mailto:gkholman@CraneSoftwrights.com
Crane Softwrights Ltd.         http://www.CraneSoftwrights.com/o/
Box 266, Kars, Ontario CANADA K0A-2E0   +1(613)489-0999 (F:-0995)
ISBN 0-13-065196-6                      Definitive XSLT and XPath
ISBN 0-13-140374-5                              Definitive XSL-FO
ISBN 1-894049-08-X  Practical Transformation Using XSLT and XPath
ISBN 1-894049-11-X              Practical Formatting Using XSL-FO
Member of the XML Guild of Practitioners:    http://XMLGuild.info
Male Breast Cancer Awareness http://www.CraneSoftwrights.com/o/bc