[sc34wg3] TM Data Model issue: prop-subj-address-values

[Graham Moore]

Kal,=20

I'm well aware of the distinction between subj ind and subj address, =
what I
was trying to get at was there is non-determinism in both aspects of
identity assignment, primarily because comparison within the model is =
only
ever done with locators.

gra

----------------------------------------------------------------
Graham Moore, Ontopian            moore@ontopia.net
GSM: +47 926 82 437           http://www.ontopia.net


-----Original Message-----
From: sc34wg3-admin@isotopicmaps.org =
[mailto:sc34wg3-admin@isotopicmaps.org]
On Behalf Of Kal Ahmed
Sent: 03 November 2003 15:22
To: sc34wg3@isotopicmaps.org

On Mon, 2003-11-03 at 14:07, Graham Moore wrote:
> Kal wrote:
>=20
> >> FACT: 1 topic can only represent one subject
>=20
> No, 1 topic *should* represent one subject but as we know the SLUO is=20
> not achievable.
>=20

So we should abandon the only remaining heursitic in XTM that comes =
close to
the SLUO ?

> >> FACT: 1 resource is one subject
>=20
> Indeed but that doesn't matter.
>=20

It does when taken together with the preceding fact.

> Both subject identifiers and subject addresses are means of=20
> establishing identity. There is no guarentee of correctly determining=20
> (when the URIs are
> different) whether they are representing the same subject. So I think=20
> the comparison is a useful one and gives good guidance as how we =
should
proceed.
>=20

No, a subject indicator has a completely different semantic to a subject
address. A subject indicator tells you that a resource describes the =
subject
of a topic, not that it is the subject of a topic. They follow a =
heuristic
which says that "well constructed and well chosen subject indicators =
will
always convey exactly the same subject to the intended audience of the =
topic
map" - otherwise they don't work. Subject address on the other hand says
"the resource at this address *is* the subject".
There is no concept of interpretation of the resource in the semantic of
subject address. The only similarity is that both properties use =
locators.
Thats it. They use locators in different ways with the addressed =
resources
have different relationships to the subject - so I don't see how any =
useful
comparison can be made.

Cheers,

Kal


> ----------------------------------------------------------------
> Graham Moore, Ontopian            moore@ontopia.net
> GSM: +47 926 82 437           http://www.ontopia.net
>=20
>=20
> -----Original Message-----
> From: sc34wg3-admin@isotopicmaps.org=20
> [mailto:sc34wg3-admin@isotopicmaps.org]
> On Behalf Of Kal Ahmed
> Sent: 03 November 2003 14:54
> To: sc34wg3@isotopicmaps.org
>=20
> On Mon, 2003-11-03 at 10:39, Graham Moore wrote:
> > Ok, heres my take:
> >=20
> > FACT: 2 different URIs can reference the same resource.
> >=20
> > When merging topics based on, for example subject identifiers, the=20
> > current draft states that it is an exception if the topics being=20
> > merged have different values in the subject address property. From=20
> > above it follows that this is an unsafe assumption. Thus to support=20
> > this fact, this non-determinism, subject address should be a =
collection.
> >=20
>=20
> FACT: 1 topic can only represent one subject
> FACT: 1 resource is one subject
>=20
> So I think you have an even worse conflict if you let subject address=20
> be a collection.
>=20
> > Let me also position things in terms of a more general principle=20
> > taken from the use and expectations of subject identifiers. There is =

> > no reason why two subject identifiers cannot indicate the same=20
> > subject via different indicators. It is desirable to have one and=20
> > only one, but that is unlikely no matter how hard we try. But we=20
> > dont say in the standard that to have two subject indicators is an=20
> > error. i.e. we err on the side of flexibility because we can't be=20
> > sure. If two topics are merged because a name match has triggered it =

> > or some application code has triggered it then we dont throw an=20
> > exception if the two topics have different subject identifiers =
because
thats the way it works.
> > Thus, I dont see a big distinction between supporting a scheme we=20
> > have defined (Subj Inds) and a scheme that exists
> > (URI) as being different when in both cases there is a degree of=20
> > ambiguity/ flexibility.
> >=20
>=20
> I think you are comparing apples and oranges here. Subject indicators=20
> are URIs of resources that describe subjects, and a subject can be
> *described* by many resources. So there is, in my mind, no parallel=20
> here at all.
>=20
> Cheers,
>=20
> Kal
> --
> Kal Ahmed, Techquila
> Standards-based Information Management
> e: kal@techquila.com
> w: www.techquila.com
> p: +44 7968 529531
>=20
> _______________________________________________
> sc34wg3 mailing list
> sc34wg3@isotopicmaps.org
> http://www.isotopicmaps.org/mailman/listinfo/sc34wg3
>=20
> _______________________________________________
> sc34wg3 mailing list
> sc34wg3@isotopicmaps.org
> http://www.isotopicmaps.org/mailman/listinfo/sc34wg3
--
Kal Ahmed <kal@techquila.com>
techquila

_______________________________________________
sc34wg3 mailing list
sc34wg3@isotopicmaps.org
http://www.isotopicmaps.org/mailman/listinfo/sc34wg3