[sc34wg3] SAM-issue term-scope-def

Kal Ahmed sc34wg3@isotopicmaps.org
Fri, 14 Jun 2002 10:18:14 +0000


On Friday 14 June 2002 07:37, Graham Moore wrote:
> Hi,
>
> Bernard I think I share your position here - I think this is effectivel=
y
> what I said in my small paper on Is Scope Bogus? i.e. that scope is no =
more
> than a lazy way of stating things about associations and that really to=
 be
> useful associations should be further qualified by other associations.
>
> I would like us to try and get some consensus on this idea and put
> something into the spec to reflect this. I dont think we need to ditch
> scope just make it very clear that its a sloopy way to work. I think we=
 do
> this by defining something like the scope assoc example bernard provide=
d
> and say 'this is what scope is, but as you can see its not very
> expressive'. Users of topics maps will then have all the info about wha=
ts
> really going on and if they want fairly useless bags of topics then fin=
e,
> if not we'll have much better maps.
>
> I also think this goes hand in hand with ditching the topic naming
> constraint as a MUST do and introducing typed names instead of scoped
> names.
>
> If I can give a short example about the names,
>
> topic t1 has name n1.
> n1 has scope of (t2, t3)
>
> Scope on names is used in 2 ways:
>
> 1. to type
> 2. to qualify the topic
>
> i.e.
>
> 'gra' scoped by {shortname} is a typing action
>
> 'granada' scoped by {city} is a topic qualifying action
>

I must say that I agree with both Graham and Bernard that the existing "s=
cope"=20
construct is not expressive enough for all needs (although it is suitable=
 for=20
some useful subset of all applications). I'm not sure that I agree with y=
our=20
definition of the "use" of scope. My belief is that scope is used on name=
s:

1) To turn them into context-dependent identifiers (for the TNC)
2) To restrict their use in an application (scope by language for example=
)

I believe that the topics that one would typically use to do that are, as=
 you=20
say topics that either type or in some other way qualify the named topic.=
 I=20
also believe that the same holds true for occurrences (even though they h=
ave=20
no identity semantics)

For associations, on the other hand the usage of scope is much more about=
=20
defining a context of validity and I agree with Bernard [1] that there ar=
e=20
many facets to such a context and that simply lumping them all together i=
n a=20
bag is not helpful. Perhaps the solution to this lies, as Bernard suggest=
s,=20
in treating the scope syntax as a shortcut for a much more powerful "scop=
ing=20
association". I am thinking of something simliar in nature to the existin=
g=20
class-instance association / <instanceOf> duality.

Cheers,

Kal

[1]  http://isotopicmaps.org/pipermail/sc34wg3/2002-June/000295.html

--=20
Kal Ahmed, techquila.com
XML and Topic Map Consultancy

e: kal@techquila.com
p: +44 7968 529531
w: www.techquila.com