[topicmapmail] philosophical question about occurrences

[Murray Altheim]

Dunker wrote:
> Hi everyone,
> 
> I'm in the process of writing my MA-thesis, which has a good deal to do  
> with topic maps. In one chapter I present the paradigm's major  
> concepts. While writing about occurrences, something struck me: the  
> literature usually says that occurrences ARE the resources that topics  
> refer to. So, I thought, resources are called (or turned into)  
> occurrences once they are referenced by a topic. Steve Pepper indicates  
> this in his TAO-Article: "A topic may be linked to one or more  
> information resources that are deemed to be relevant to the topic  
> insome way. Such resources are called occurrences of the topic.". The  
> same is indicated by the XTM-specification: "An occurrence is any  
> information that is specified as being relevant to a given subject."
> 
> However, occurrences can be scoped. The same resource  - let's say a  
> fictional piece of text - could be occurrence for two or more topics -  
> let's say two characters appearing in it. Character Paul is the main  
> character in the story, while Peter is a side character. I could scope  
> the occurrence with "main character" for topic 'Paul' and with "side  
> character" for topic 'Peter' (The example might not be a good one, but  
> technically this is possible). But, what I have actually done, is  
> scoping the _reference_ to the resource, not the resource itself. Which  
> scope I choose is dependent on the topic the resource is referred from.  
> So, if we say we can scope occurrences, are occurrences really the  
> resources themselves, or are they references to the resources? The same  
> problem appears in connection with occurrence types: If I type an  
> occurrence as 'mention', I really type its reference. Holger Rath  
> indicates this in an article(1) from 2001: "An occurrence is the link  
> to an information [...] - it connects the topic domain with the  
> resource domain."

Essentially, an occurrence is considered a *characteristic* of a
topic, i.e., that is its relation to the topic. This relation is
governed by *scope*, or the context in which this relation is true
or relevant. The semantics of this are straightforward, so long as
you keep in mind we are talking here about an abstract model of
topic maps and not any implementation.

I believe your difficulty is not in these semantics but in mixing
levels of interpretation. In terms of the semantics of the relation
between a topic and its occurrences, you can effectively ignore issues
of resources and references, i.e., what occurrences are, how they're
connected (mechanisms of reference), how many layers of reference
are used to make that connection, etc. -- these operate at lower levels
than the abstract model, something we might call an "implementation"
level.

In the "topic map graph" -- the abstract structure of the topic
map -- the arc connecting a topic node to an occurrence node
ignores the behind-the-scenes implementation; it's simply seen as
a connection. You could implement a topic map using strings and
rocks, or with a pencil and paper. This kind of thing has been
done publicly with ribbons, tape, and human bodies. You aren't
scoping the reference, you're scoping the occurrence -- the
reference is simply the mechanism by which the occurrence is
"physically" (implementationally) connected.

If you think about it, everything we use to communicate topics
(and more generally, everything we use to communicate, full stop),
from our use of spoken language, the glyphs of its representation
in written language, the use of XML syntax in representing topic
maps, all are simply references -- in semiotics they're considered
"signs". I don't mean to sound philosophical, it's just that once
one begins to decompose modes of reference, it's a spiral right
down to the lowest possible decomposition, and I'm not sure there's
any point in bringing that into your thesis unless you intend to
discuss the semiotics of topic maps, i.e., how topic maps
distinguish and map "subjects" (which is what TM topics actually
reference), and what this means, which is an interesting topic in
itself, but likely a different thesis.

Murray

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

     "In Las Vegas Mr Gates also demonstrated a prototype
      fridge magnet which can be programmed to receive traffic
      reports, sports results and advertisements from local
      restaurants using the same FM signal as the wristwatch."
                                  -- The Guardian, 10 Jan 2003.