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

 


Help: OASIS Mailing Lists Help | MarkMail Help

tag message

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


Subject: Re: [tag] belated regrets


Title: Re: [tag] belated regrets

Sorry about missing todays call. Work crisis and being short staffed today, I was asked to help.

Lynne.
--------------------------
Sent from my BlackBerry

----- Original Message -----
From: stephen.green@systml.co.uk <stephen.green@systml.co.uk>
To: tag@lists.oasis-open.org <tag@lists.oasis-open.org>
Sent: Wed Nov 21 14:54:36 2007
Subject: Re: [tag] 'Co-requisites' (co-constraints) cf 'pre-requisites'

Interestingly, the example of coded values is quite a nice example
of where a precondition would qualify an IUT: the IUT could be an
invoice instance and the precondition could be that it contains at
least one coded value.

UBL doesn't express TAs as such so it isn't an example of prior
art but you could say that its XML Schema and Schematron artefacts
list most of it's TAs: in which case the example below might be
one of just a few additional ones necessary to have a TA list
covering sufficient of its requirements. (In case we make it an
example, I'd better add that there is a caveat that Schematron
doesn't have to be used - I think so anyway, best check on that.)

> In regard to following up the comment/question I made on the
> call:
>
> What I'm asking is - are there sometimes 'co-requisites' rather
> than pre-requisites - that is, TAs which have to be taken
> together to produce an outcome without any requirement about
> which has to be tested first?
>
> An example that springs to mind from UBL is the need to use
> two passes of validation of coded values in an instance. Both
> have to be passed for there to be any outcome. They are both
> using different artefacts (an XML Schema and a Schematron schema
> as it happens but that isn't relevant here). It isn't necessarily
> the case that one test/validation has to be performed before
> the other, as long as both are performed. (In this example there
> is one of the tests which is meaningless if the other isn't
> tested with a pass but that is not helpful to the example -
> it could have been that that was not the case and that each is
> irrelevant if the other isn't passed.) This is one of the main
> UBL TAs if not THE main TA. There are other examples such as
> the need to pass two sets of validation (no requirement which
> goes first) when there is a second set of constraints applied
> to a subset instance. Again both Schematron and XML Schema
> would be typical artefacts to be applied to the IUT to
> determine the outcome. Neither would be relevant without the other
> in some situations. If the two TAs can't be combined in some way
> (and making just one TA isn't really appropriate) then there would
> be a need to add TA-like information to a conformance clause which
> probably didn't belong there but is really TA material.
>
> Other examples relate to things like tax requirements and are
> more semantic than technical but they are still important in
> specifying document vocabularies like UBL. If and only if an invoice
> has any sales or value added tax it must also have a tax
> registration number: for which there might be one TA to see
> that it has tax correctly applied and another to see whether
> or not it has a tax reg number:
> if both then pass
> if one only then fail
> if neither then pass
>
> Thanks
>
> --
> Stephen Green
>
> Partner
> SystML, http://www.systml.co.uk
> Tel: +44 (0) 117 9541606
>
> http://www.biblegateway.com/passage/?search=matthew+22:37 .. and voice
>
>
>
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  You may a link to this group and all your TCs in OASIS
> at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php




---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail.  You may a link to this group and all your TCs in OASIS
at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php




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