Re: philosophical question about occurrences

Hi Knud,

At 01:23 20.02.2003 +0100, Dunker wrote:
> While writing about occurrences, something struck me: the
> literature usually says that occurrences ARE the resources that topics
> refer to...
>
> However, occurrences can be scoped...
>
> I hope this wasn't all to confusing and meaningless. Maybe the question
> is totally irrelevant...

It certainly isn't irrelevant. It is right on target. Before I explain,
let me state up front that

(1) occurrences are NOT resources
(2) lots of places claim that they ARE resources

What you have discovered is a remnant of the evolution of the topic
map paradigm! At the time of the publication of the topic map standard
in early 2000, much of our thinking was clear, but not all of it, as
we have since discovered.

Back then we tended to regard the resources themselves as being the
occurrences. For example, the ISO standard defines "topic occurrence"
as "Information that is specified as relevant to a given subject".[1]
Later on, section 5.2.3 states that the <occurs> element "references
information (one or more 'occurrences') that is relevant...". There is
also a note stating that "Topic occurrences may be offline resources".

This usage is (still, unfortunately) reflected in the language used in
my paper, "The TAO of Topic Maps", originally written in 2000.[2]

We started to refine this thinking during the development of XTM 1.0
as it became clear that it is the *relationship* between the topic and
the resource that is of interest; it is that relationship that is subject
to scoping (not the resource itself), and it is the relationship that
should be regarded as the "occurrence". You can see this reflected in
the Conceptual Model in Annex B of the XTM spec (see section B.7).[3]

Unfortunately, full clarity arrived after publication and the XTM spec
is thus still a bit inconsistent in this regard. (For example, section
2.2.3 again talks about occurrences as if they were resources.)

All of this is now being cleared up in a restatement of ISO 13250. The
place where it will be most apparent is in the description of the data
model, currently known as the SAM (Standard Application Model). This
defines occurrence as "a relationship between a subject and an
information resource" (see section 3.7)[4].

Under this view it becomes clear that an occurrence really is just a
special kind of association (a binary association in which one role
player is a topic and the other is a resource).

I really should fix this in The TAO of Topic Maps, I know, but I
simply haven't had time to write version 2.0. (However, please note
that a partly revised version exists at the URL below.) In the meantime,
please be sure to study the SAM specification, which is by far the
most consistent exposition of the TAO model that we currently have.

Steve

[1] http://www.y12.doe.gov/sgml/sc34/document/0129.pdf
[2] http://www.ontopia.net/topicmaps/materials/tao.html
[3] http://www.topicmaps.org/xtm/1.0/#conceptualmodel
[4] http://www.y12.doe.gov/sgml/sc34/document/0356_files/0356.html#d0e1162


--
Steve Pepper, Chief Executive Officer 
<pepper-76mms6M3oqTR7s880joybQ@xxxxxxxxxxxxxxxx>
Convenor, ISO/IEC JTC1/SC34/WG3  Editor, XTM (XML Topic Maps)
Ontopia AS, Waldemar Thranes gt. 98, N-0175 Oslo, Norway.
http://www.ontopia.net/ phone: +47-23233080 GSM: +47-90827246