OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

ubl-csc message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]


Subject: Re: FW: [ubl-csc] Format for Library Content Review


We discussed this on the LC call today and felt that the issue would best be dealt with by the Chairs group (strategic issues).  

however, we did resolve that LC could use the template Jon has developed as a basis for our current normative material, as it is something we would have to design anyway.  


Probert, Sue wrote:
FW: [ubl-csc] Format for Library Content Review

Oops. Sorry, Tim, I inadvertently forgot to add your name to this circulation list. Maybe my recent agenda request will make more sense now!

regards

Sue

Sue Probert
Senior Director, Document Standards
Commerce One
Mobile: +44 7798 846652
Tel: +44 1425 275117 or +44 1753 483000
email: sue.probert@commerceone.com


-----Original Message-----
From: Probert, Sue
Sent: 23 December 2002 08:41
To: 'jon.bosak@sun.com'; Mark CRAWFORD (E-mail)
Cc: Michael Adcock (E-mail); Gunther Stuhec (E-mail); Stig Korsgaard
(E-mail); Burdett, David
Subject: RE: [ubl-csc] Format for Library Content Review


Hi Jon

The challenges you raised with regards to presentation of the library spreadsheet remind me to raise a serious concern which Mike, Gunther, Stig and myself wished to raise after the Christmas break. We remember well the arguments for and against the ISO fast-tracking of the UN/EDIFACT standards.

Whilst we didn't in that case have the context methodology component we certainly did have the syntax rules component similar to our UBL ndrs and we did have the directories part equivalent to our library component. The UN/EDIFACT syntax has gone through four versions and therefore four ISO fast-track publication cycles over the past ten years or more. But it was recognised a long time ago that this ISO publication process did not suit the directories for some very significant reasons which we believe are still very much valid today.

Firstly, it is the issue of technical assessment and change request procedures. The library aspect must change much more fequently in order to address newly emerged business requirements and b) to correct any bugs. In the case of the UN/EDIFACT directories they have released new versions twice yearly since 1990. This enables individual trading commmunities to choose the specific directory version which most suits their requirements. Upwards compatability (met in every case except an infamous Q/C process in 1991) is a key ingredient as are the rigorous quality controlled change request and technical assessment procedures. These have been enhanced and developed by enormous voluntary contributions over the years until today they are incredibly streamlined and efficient and for sure the crown jewels of the CEFACT process. I do not believe that this could have been achieved within an ISO framework. Also these same procedures are already being adapted to support core components and XML. Having been personally involved in these tech assess and change request procedures for many years we believe it would be crazy not to take advantage of the availability of this hardwon experience to progress and manage the UBL library content within UN/CEFACT itself.

Secondly, it is the very issue of government and international organisation support for ISO standards that was and still is the largest concern. Because international organisations and particularly governments and their agencies prefer ISO standards trading communities often adopt their recommendations. Sounds like that is actually what we want for UBL but over time the different organisations and governments could choose different versions of the ISO standard which could encourage divergence. Also because they wish to encourage stability of implementation they tend to fix their ISO version recommendations for long periods - in the case of EDIFACT , for example, CEN tried in 1993 to make a blanket recommendation for Europe for three years to use a particular directory which proved impossible to follow for most european trading communities as they needed the improvements added in the 1994 and 1995 directories. Following this experience it has never been tried again and consequently the discussion of moving the UN/EDIFACT directories to ISO standards has never been raised again either.

As my head is full of the Christmas presents I still need to buy and wrap as well as the food that I still need to buy and prepare, this is just an introduction to a subject that the four of us would like to suggest is debated further in the New year.

In the meantime best wishes to all for the upcoming celebrations.

regards

Sue

Sue Probert
Senior Director, Document Standards
Commerce One
Mobile: +44 7798 846652
Tel: +44 1425 275117 or +44 1753 483000
email: sue.probert@commerceone.com


-----Original Message-----
From: Jon Bosak [mailto:Jon.Bosak@sun.com]
Sent: 22 December 2002 21:35
To: ubl-csc@lists.oasis-open.org
Cc: eduardo.gutentag@sun.com; lseaburg@midsouth.rr.com;
lseaburg@aeon-llc.com; mavis.cournane@cognitran.com
Subject: [ubl-csc] Format for Library Content Review


As promised in the LCSC meeting last week, I have been attempting
to figure out the document structure of the next Library Content
(and NDR) review.

We have what appear to be these two main requirements:

 - The document set has to be conformant (so far as it can be at
   this stage) with ISO guidelines.

 - The document has to be easy for us and our reviewers to work
   with.

A couple of days spent trying to construct a traditional document
conforming to ISO/IEC Directives, Part 3 (Rules for the structure
and drafting of International Standards) with additional reference
to ISO/IEC JTC 1/SC 2/WG 3 N 505 (Swedish view on ISO/IEC 8859
document layout) have convinced me that this it is possible to
achieve both objectives, but not at the same time.

The biggest problem is the spreadsheets.

 - They contain hidden annotations that don't show up in a
   printout.  We will have to solve this editorially by breaking
   the annotations out into text that will precede the printed
   version of the spreadsheets.

 - They use colors in a semantically meaningful way.  This doesn't
   work so well on paper unless you have a color printer, which
   most people don't.  I don't have a good solution for this; I
   guess people using the printed version will just have to cope
   with the grayscale approximations.

 - They want to be displayed sideways on pages that use a separate
   framework for page numbering and TOC generation.  This can be
   accomplished by creating JPEG images of each spreadsheet page,
   rotating the images, and pasting them into the document, but
   the process is not easy and results in insanely large files.
   One obvious solution is to convert everything to PDF and use
   Acrobat to assemble the document, but I will need some time to
   explore Acrobat's capabilities with regard to headers, footers,
   and TOC generation.  I also believe (on general principles)
   that this problem can be solved by using the openoffice suite
   to convert everything to openoffice xml formats and working
   with those using emacs and perl scripts, but I will need some
   help from the openoffice team in figuring out how to do this.

So while I believe that it is possible to create a printed ISO/IEC
conformant document when we're all done and ready to move UBL into
the international standardization process, it's going to take some
further work to figure out how to do this, and it seems to me
unlikely that the solution we finally adopt is one that we and our
reviewers will find easy to work with during the design and review
phases.

Until we get to the place where we're ready to release UBL as a
Committee Specification, therefore, I suggest that we publish the
NDR and LC sets as hypertext documents that conform as closely as
possible to ISO/IEC document organization guidelines but are
published using HTML together with spreadsheet and drawing formats
that can be worked with using free tools.  The Excel spreadsheet
format qualifies under this heading because the equivalent program
from openoffice.org can work with it.

In conformance with the ISO/IEC naming guidelines, I suggest the
following official names for the pieces of this effort:

   Universal Business Language -- Part 1. Naming and Design Rules
   Universal Business Language -- Part 2. Library Content
   Universal Business Language -- Part 3. Context Methodology

Mockups of website-resident and local versions of Part 2
(part2.htm and part2loc.htm, the latter in a zip file) are
attached below.  They are identical except that the
website-resident version points to specific URLs on the UBL LCSC
part of the site while the local version points to files bundled
with the document.  The URLs are fictional at this point, of
course.  I've used the XSD and XLS files from Gunther Stuhec's
LCSC message of 20 December to partially populate the local
version so that you can see how this works.  I've tried for
wording that enables the local version of the HTML file to be
generated from the website-specific version with a single global
search-and-replace on the site-specific part of the URLs.

We could use the same approach to include stylesheets and case
studies, but since we're not planning on releasing them
synchronously with the schemas in this review cycle, I haven't
bothered to include hooks for them in these mockups.

Please look this over and tell me whether you think it's heading
in the right direction.  It would also be helpful if someone
subscribed to the LCSC list could forward to the LCSC a URL to
this message in the CSC archive in time for the LCSC
conference call tomorrow (Monday 23 December).

Jon


-- 
regards
tim mcgrath
fremantle  western australia 6160
phone: +618 93352228  fax: +618 93352142 



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]


Powered by eList eXpress LLC