Conformance comment: Re: [ubl] First draft of UBL Customization Guidelines available for review

["Stephen Green" <stephen.green@documentengineeringservices.com>]

Congratulations on this complete draft.

A comment/question on conformance.

Section 1.2.1
Lines 196/198

Maybe this paragraph should be highlighted in some way
or does it quote a conformance clause in the UBL? I could
not see a conformance clause normatively specified in UBL
2.0 http://docs.oasis-open.org/ubl/cs-UBL-2.0/UBL-2.0.html
so is this paragraph that clause? If so (I think possibly it is),
then maybe it should be set out as such so that it can be
referenced specifically elsewhere. I would see the paragraph
as amounting substantially to two conformance clauses;
one for instance implementations (more general than just
customisations) - lines 196/7 - and the other, lines 197/8
for customisation implementations only.

Secondly, on the first sentence, lines 196/7, this seems to
mirror but slightly contradict the normative requirements in
UBL 2.0 for instances "Additional Document Constraints"
http://docs.oasis-open.org/ubl/cs-UBL-2.0/UBL-2.0.html#CONSTRAINTS
This seems to indicate that actually lines 196-8 are not to
be treated as a conformance clause but merely characterise
conformance loosely but maybe inaccurately. Obviously
there are possible instances which will validate against the schemas
but still not conform to the "Additional Document Constraints".
The spec doesn't say either "Normative" or "Informative" against
those constraints so maybe these Customization Guideline
are an opportunity to reinforce whether the additional constraints
are indeed normative and essential to UBL 2.0 conformance.

Personally I think the whole requirement for no empty elements
etc, is an onerous overhead both on implementation and on either
validation and/or conformance testing and also a bit burdensome
and even a hindrance for the customisation guidelines. For the
latter it becomes unclear whether an ASBIE can or cannot be
both empty in an instance and conformant - or for that matter
a customisation technique. If a customisation allows 'Party' say
to be empty, not inheriting the 'additional document constraints'
explicitly and therefore, by not stating them, obscuring them -
does that customisation still have the right to claim conformance?
Similarly with a technique or tool for customising UBL or a test
suite or validation method which does not test for the additional
constraints. However the guidelines do at least need to be
conformant themselves with the additonal constraints I would think.

Best regards


-- 
Stephen D. Green

Partner
SystML, http://www.systml.co.uk
Tel: +44 (0) 117 9541606
Associate Director
Document Engineering Services
http://www.documentengineeringservices.com

http://www.biblegateway.com/passage/?search=matthew+22:37 .. and voice

2008/9/4 Jon Bosak <bosak@ibiblio.org>:
> Hello UBL TC,
>
> The long-awaited first draft of the UBL 2.0 Customization
> Guidelines is now ready for your review.  You can find version
> 0.91 at
>
> http://www.oasis-open.org/committees/download.php/29194/custguide091d.pdf
>
> Note that no URL has been assigned this yet.  That can wait till
> next week.  I also hope in the next few days to get more of the
> original artwork so that we can get a little more uniformity in
> the illustrations.
>
> Please review this draft at your first opportunity and prepare to
> discuss it at next week's Pacific and Atlantic TC calls.  If you
> can't participate (or even if you can), please send your comments
> to the list, referring to line numbers as appropriate.
>
> Many thanks to the authors -- Mavis Cournane, Mike Grimley, Ken
> Holman, and Tim McGrath -- for the efforts that have put this
> remarkable document before you.
>
> Jon
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
>