Interoperable Logging WAS: Re: [topicmapmail] Questions About XTMOccurrence Semantics

[Conal Tuohy]

Peter P. Jones wrote:

> > If I refer to the item identifier of the occurrence item 
> using a subject 
> > locator, and make that subject locator the subject 
> identifier of a topic 
> > (reifying?) do I then get to attach say, metadata, to the construct 
> > purely qua construct? [my terminological spaghetti].

If you reify an XTM occurrence element with a topic, then your topic is
really about the occurrence itself (i.e. the real occurrence
relationship), rather than the particular representation of that
occurrence in XTM syntax. As Lars Marius pointed out earlier in the
thread:

 A topic reifying a topic map construct in reality represents the
real-world thing 
 represented by that topic map construct. A topic reifying an
association really 
 represents the relationship represented by that association ...

  from http://www.isotopicmaps.org/sam/sam-model/#d0e994

> Thinking about this some more, maybe I should be asking 
> whether there is 
> anything in the way of an interoperable best practices approach to 
> 'attaching' logging, provenance and auditing information to the 
> construction, editing and merging of TMs?

I believe the standard practice for this kind of thing is to use scope.
It's quite simple to use scoping topics to attach metadata to an
occurrence. In my work, I am frequently harvesting topic maps from a
variety of sources, and merging them. I create a topic to represent a
source topic map, and all occurrences harvested from that source are
automatically scoped with that particular harvest topic. The harvest
topic then has any necessary metadata attached. See the documentation on
the XTM mergeMap element:
http://www.topicmaps.org/xtm/index.html#elt-mergeMap

Cheers

Con