[sc34wg3] Disentangling 'scope', 'context' and 'applies'

[Marc de Graauw]

* Martin Bryan

| Just a few (carefully edited) reminders from the definition of scope in
| 13250:
|
| Scope
| The extent of the validity of a topic characteristic assignment:
| the context in which a name or an occurrence is assigned to a given topic;
| the context in which topics are related through associations.

Quite right. One thing I started wondering about is whether those two parts are
the same or distinct.
The 'extent of validity' suggests an exact, formal notion: within the scope an
assignment is valid, outside the scope not (or: not known to be valid). The
'context in which ...' suggests a much looser intention: contextual information,
meta-information, extra information that might be of relevance...

To give an example wholly outside the realm of Topic Maps:
1) The extent of validity of my parking permit is my neighborhood. Inside it is
valid; outside it is not.
2) Remark: "X said that Topic Maps are useless and RDF is great." Answer: "Yes,
but look at the context, X was drunk at the time." Here the fact that X said
something is not true only when he is drunk, and false (or unknown) when he is
sober. The answer suggests X would possibly not adhere to the view he expressed
when he is sober, but it says nothing definitive about it, it only gives
information which is relevant in judging X's words.

It would be very strange to say:
1) The context wherein my parking permit is valid is my neighborhood.
2) The extent of validity of X's remark is his drunkenness.

So in natural language contexts (outside Topic Maps), there is a difference in
use of 'extent of validity' and 'context'.
ISO 13250:2000 seems to say those two things are equivalent due to the colon
between the definitions. I just wonder whether there is not a certain tension
between the two definitions, and whether this does not lead to two uses of scope
which are not entirely reconcilable. As several people have already noted, scope
is used for very different things in practive. As you can see, I am not very
certain on these issues and would like any feedback. You were there. Did you
(and the others) mean that 'extent of validity ...' is equivalent to 'the
context in which ...', or should we not read this so strict, more as: scope can
be used as an extent of validity, or as a context in which ...

It is certainly true that I have taken the 'extent of validity' notion as a
starting point.

<skipped>
| Marc uses the term Perspective for "the way a user wishes to look at a Topic
| Map", and then equates this to "the set of themes which apply". To me this
| against the spirit of 13250 in the sense that it seems to require all
| controls of the subsetting of topic maps to be defined formally as topics
| and applied formally as scopes.  This does not seem to allow sufficient
| power to concepts such as roles, which are equally applicable to managing
| "the way a user wishes to look at a Topic Map".
|

True. So the proposed definition of 'perspective' could not stand as it is.

Marc