[topicmapmail] graphic language for describing TopicMaps

[Murray Altheim]

Patrick Durusau wrote:
> Murray,
> 
> Murray Altheim wrote:
>>Thomas B. Passin wrote:
[...]
>>>The only real difference is that in RDF, the bnode is a Resource, just 
>>>like all the other nodes.  However, if we regard an association as a 
>>>specialized variety of topic, the two cases are nearly isomorphic.  
>>>And why not consider an association to be a specialized variant kind 
>>>of topic?  Then you wouldn't have to create a topic to reify an 
>>>association, it would be its own topic already.
>>
>>Actually, that road has been travelled. I think Steve Pepper
>>wrote a paper on everything being a Topic a few years ago. But
>>at this point, the TM paradigm wouldn't necessarily benefit
>>from this simplification, I'm certain the community wouldn't,
>>and my take on things is that it's far better to concentrate
>>on the existing semantics -- I mean, we have an ISO standard
>>behind us -- it's not like we should push for a change to that.
>
> The first reference I could quickly find with "everything being a topic" 
> was:
> 
>>2) Conformance of work done in the XTM specification group must be monitored
>>not to object in any sense to the fundamental concept of topic maps as given
>>by the ISO 13250 standard. By fundamental I mean:
>>- the core "philosophical" concepts like e.g. the relation between topics
>>and subjects, the idea of everything being a topic, the orthogonality of the
>>topic and occurrence layers and so on.
> 
> 
> From: # "Dr. Heiko Beier" <heiko.beier@moresophy.de>
> 
> http://lists.oasis-open.org/archives/topicmaps-comment/200104/msg00054.html

Well, I'm pretty sure it was Steve that wrote that paper, but I
don't have any handy references to it now. It was basically on
the idea that everything in the model could be a Topic, that we
didn't need associations, that the association was simply another
kind of topic characteristic.

> Curious as to why you don't think there is a benefit in this 
> "simplification"? (Which I think is already present in ISO 13250, but 
> obscured by later practice in particular syntaxes and proposed 
> processing models.)
> 
> I fail to see the advantage to the extra steps.

These two last paragraphs seem to be at odds with each other, so
I'm not sure how to respond. I am not an advocate of making any
substantial changes at this point simply because we have an entire
community (which is hopefully growing) based on several renditions
of an ISO standard, plus software systems and hundreds of ancillary
documents written around the current model(s). To make a profound
change as this would suggest would to me not worth the resulting
instability and confusion. The current model works as it is.

Put it this way: I don't want to have to rewrite all of my software,
and I'm certain nobody else does either, simply to alter the Topic
Map model in a way that doesn't actually provide much in the way
of benefit. If it's just looking at things through a fun house
mirror, I prefer without the mirror. (Or, if you prefer, if I'm
already looking at them through the mirror, I'll stick with the
mirror.)

Murray

......................................................................
Murray Altheim                    http://kmi.open.ac.uk/people/murray/
Knowledge Media Institute
The Open University, Milton Keynes, Bucks, MK7 6AA, UK               .

   The Resume of George W. Bush
   http://www.buzzflash.com/contributors/03/04/23_resume.html

                    The sleep of reason brings forth monsters -- Goya