[xtm-wg] Reification - why it came back in the CM

["Bernard Vatant" <universimmedia@wanadoo.fr>]

Following some exchanges on that subject between members of the "Paris CM
subgroup", I second Steve P. suggesting to bring it back to the public
forum.

I'll try to explain for those who were not in the CM discussion in Paris
why and how "reification" was brought back.

The very meaning given for that word * in XTM scope* is very simple : it's
the process through which a Topic is created.
Before reification, there is "something", addressable or not, resource or
not, inside or outside the system, inside or outside the Topic Map itself,
candidate to be a good subject - like *any* thing whatsoever.
After reification, you have a brand new Topic in the system, of which the
above "something" is the subject.
No more. That's dreadly simple. If somebody claims it can't be that simple,
that it's a very more subtle concept I widely misunderstood, please explain
to me where I missed the point.

This very action of creating a Topic is a fundamental process, whether it's
made by an human author, creating the Topic "from scratch", or a system
process, for instance creating a Topic out of any association in the TM, or
any data in a data base etc.
Given this process is a basic concept, it has to be pointed to and given a
name, if only to show that Topics don't exist to begin with - as such - in
the middle of nowhere, but only subjects are there, which have to be "made
real", that is processed into definite computable objects of a definite
type.

Well, why the choice of "reification" ? Because it's the very meaning of
the word : make something "real" for the system, in the sense it can handle
it, compute it, merge it etc ... What other word would fit : "topicization"
? Gads ! Other ideas ?

Another debate :
RDF uses it. OK. with another meaning. OK. That point was considered too.
What was said is that reification defined in XTM scope and reification
defined in RDF scope could be considered later as subtypes of a more
general class of process, which, IMO, could be defined by something like
the following :

"Reification is a process through which a computable/addressable object - a
resource - is created in a system, as a proxy for a non
computable/addressable object"

This definition is a "relative" one, meaning by that the definition of
addressable/computable depends on the system and the nature of the objects
it is able to handle. But my hunch is it fits both RDF and XTM particular
"reification(s)". Maybe some RDF guy around there coud tell if that makes
sense or not.

Well ... that was my 0.02 Euros for the "Grand Semantic Unification" <:o)

Bernard

---------------------------------------
Bernard Vatant
bernard@universimmedia.com
www.universimmedia.com
"Building Knowledge"
---------------------------------------



To Post a message, send it to:   xtm-wg@eGroups.com

To Unsubscribe, send a blank message to: xtm-wg-unsubscribe@eGroups.com