[sc34wg3] abstract-constraint isa topic-type?

Graham Moore gra at networkedplanet.com
Fri Jan 29 07:24:55 EST 2010


Hi

> A topic t is said to be an instance of the topic type tt if following the association type tmdm:type-instance from t produces either tt, a subtype of tt, or if tt is tmdm:subject.

This is for evaluating a tmcl constraint and it matches whats in tmdm.

As I see it, the issue on the constraint, abstract-constaint thing is
that for the TMCL for TMCL to be valid we need to explicitly state
that all constraint types are an instance of Topic Type.  If we don't
say this anyone who validates a schema against the TMCL for TMCL will
be told that any abstract-constraint topic is not allowed because its
type is not marked as a topic type.

If we do this then I think we are cool.

On 29 January 2010 13:14, Robert Cerny <robert at cerny-online.com> wrote:
>
> On Jan 29, 2010, at 12:44 PM, Lars Marius Garshol wrote:
>
>> Yes. It's always that case that if
>>
>>  a isa b
>>  c ako a
>>
>> then
>>
>>  c isa b
>>
>> This is stated here:
>>  http://www.isotopicmaps.org/sam/sam-model/#sect-subtypes
>
> I would translated the first sentence of that section into:
>
>  c ako a
>  b isa c
>
> then
>
>  b isa a
>
> But this is something else than the above.
>
> I think the following statement has to be made explicit in the meta schema:
> tmcl:abstract-constraint isa tmcl:topic-type.
>
> It also makes sense if you read:
> http://www.isotopicmaps.org/tmcl/2009-10-19/#d0e504
>
>> Could the problem be that "abstract-constraint" sounds like it should be
>> the parent type of "constraint"? It really is
>> "topic-must-be-abstract-constraint".
>
> No, i just chose abstract-constraint because it is the first one listed. But
> now that i thought that i am totally confused :-)
>
> Best,
>
> Robert
> _______________________________________________
> sc34wg3 mailing list
> sc34wg3 at isotopicmaps.org
> http://www.isotopicmaps.org/mailman/listinfo/sc34wg3
>



-- 
Graham Moore, Director, Networked Planet Limited
Editor XTM 1.0, ISO13250 (TopicMaps) -2,-3, TMCL
e: graham.moore at networkedplanet.com
w: www.networkedplanet.com
t: +44 1865 811131
m: +44 7769658611 (UK)
m: +47 45271713   (Norway)

Networked Planet Limited is registered in England and Wales, no. 5273377


More information about the sc34wg3 mailing list