Re: [ubl] BBIE consistency problems

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

At 2013-02-13 14:56 -0500, Jon Bosak wrote:
> Some of these look like they were made ..1 because it wouldn't
> make sense (or be advisable) to have multiple language versions.
> "Value", "name", and "type" all suggest this.

Yet there are uses of "Value", "Name" and "Type" that have been 
validated by committee members as requiring multiple language 
versions ... here are a couple of examples for "Name":

   http://lists.oasis-open.org/archives/ubl/201301/msg00013.html
   > "Declaration. Name" (new): 0..n
   Prose
   > "Tender. Contract Name. Text" (new): 0..n
   Prose

There is an example of "..n" for each of "Value" and "Type":

   "Item Property. List Value. Text" (new): 0..n
   "Service Provider Party. Service Type. Text" (new): 0..n

If those are supposed to be "..1" instead of "..n" then I will adjust 
the cardinalities for those two entries and then take those two nouns 
out of the list.  All will then be "..1" for "Value" and "Type".

If these two nouns are examples of nouns allowed to have both "..1" 
and "..n", then I can accommodate that by changing the checker to not 
report inconsistencies for those two nouns.

Same with "Name" ... in tendering, there are occurrences of Name that 
I've been directed require multiple languages.  I've been using 
"prose" as a short form for "requiring multiple languages" because 
that is the predominant use case for requiring multiple languages.

. . . . . . . Ken

> Jon
>
> G. Ken Holman wrote:
> > As part of my cleanup from this post:
> >    http://lists.oasis-open.org/archives/ubl/201302/msg00009.html
> > ... I took all of the property nouns that people have told me over 
> > the past week are prose, in that, they justified the "..n" 
> > cardinality to me for each item.
> > As part of due diligence, I then added to the list of prose nouns 
> > all of the properties they told me about in order to clean up my 
> > reports ... here is today's list of prose BBIE nouns:
> >
> > > Bid Comments Content Contract Name Conditions Description Domain 
> > > Expression Frequency Information Instructions Justification Name 
> > > Note Phase Program Purpose Qualifier Reason Remarks Requirements 
> > > Resolution Restrictions Scope Situation Statement Risk Terms Type Value Weight
> > So I plugged those into my model checker today and, sure enough, I 
> > reduced to zero the number of "problems" with 0..n.
> > However, the new list of nouns now produces "problems" with other 
> > uses of those nouns that are "..1".  Both in UBL 2.0 and in the 
> > proposed UBL 2.1.  If these nouns truly represent elements that are 
> > prose, then they need "..n" for multiple languages and not "..1".
> > In fact I see things like "Contract Name" that are now "..n" 
> > elsewhere in UBL from edits of last week, but, apparently, are 
> > still "..1" twice in the list below.
> > Please advise ASAP which of the items below need to be changed to "..n".
> > Honestly, I'm not trying to delay things here.  I'm just trying to 
> > make the model consistent before we publish it for review.  The 
> > model is not consistent now, and the reason for my mechanical 
> > checks is to find the inconsistencies.  Getting feedback on any of 
> > these during the upcoming review would have triggered an even 
> > longer delay by restarting the review process over again.
> > If some nouns are allowed to be "..1" in some places and "..n" in 
> > others, then I will modify my model checker to allow for both for 
> > those nouns (though that won't help with any consistency if there 
> > is no consistency).
> > Thank you for your patience, and for your quick assistance with the 
> > list below.
> > . . . . . . . . Ken
> > *..1 Cardinalities found on prose text BBIEs: 34 (11 old; 23 new)
> > "Activity Property. Value. Text" (new): 1
> > "Attached Document. Document Type. Text" (old): 0..1
> > "Awarded Notification. Contract Name. Text" (new): 0..1
> > "Certificate Of Origin Application. Certificate Type. Text" (old): 1
> > "Certificate. Certificate Type. Text" (new): 1
> > "Classification Category. Code Value. Text" (old): 0..1
> > "Classification Scheme. Agency Name. Text" (old): 0..1
> > "Communication. Value. Text" (old): 0..1
> > "Consumption Correction. Correction Type. Text" (new): 0..1
> > "Consumption Point. Subscriber Type. Text" (new): 0..1
> > "Consumption Report Reference. Consumption Type. Text" (new): 0..1
> > "Consumption Report. Consumption Type. Text" (new): 0..1
> > "Consumption Report. Heating Type. Text" (new): 0..1
> > "Consumption Report. Residence Type. Text" (new): 0..1
> > "Contract. Contract Type. Text" (old): 0..1
> > "Contracting Activity. Activity Type. Text" (new): 0..1
> > "Contracting Party Type. Party Type. Text" (new): 0..1
> > "Document Distribution. Print_ Qualifier. Text" (old): 1
> > "Document Reference. Document Type. Text" (old): 0..1
> > "Item Property Range. Maximum_ Value. Text" (new): 0..1
> > "Item Property Range. Minimum_ Value. Text" (new): 0..1
> > "Item Property. Value. Text" (old): 0..1
> > "Meter Property. Value. Text" (new): 0..1
> > "Meter Reading. Meter Reading Type. Text" (new): 0..1
> > "Meter. Meter Name. Text" (new): 0..1
> > "Person. Birthplace Name. Text" (new): 0..1
> > "Price. Price Type. Text" (old): 0..1
> > "Telecommunications Supply. Telecommunications Supply Type. Text" (new): 0..1
> > "Transport Equipment Seal. Sealing Party Type. Text" (old): 0..1
> > "Unawarded Notification. Contract Name. Text" (new): 0..1
> > "Utility Item. Consumption Type. Text" (new): 0..1
> > "Utility Item. Current_ Charge Type. Text" (new): 0..1
> > "Utility Item. One Time_ Charge Type. Text" (new): 0..1
> > "Utility Item. Subscriber Type. Text" (new): 0..1
> > <end/>


--
Public XSLT, XSL-FO, UBL and code list classes in Europe -- Apr 2013 |
Contact us for world-wide XML consulting and instructor-led training |
Free 5-hour lecture: http://www.CraneSoftwrights.com/links/udemy.htm |
Crane Softwrights Ltd.            http://www.CraneSoftwrights.com/o/ |
G. Ken Holman                   mailto:gkholman@CraneSoftwrights.com |
Google+ profile: https://plus.google.com/116832879756988317389/about |
Legal business disclaimers:    http://www.CraneSoftwrights.com/legal |