[sc34wg3] SAM-issue term-scope-def

[Lars Marius Garshol]

* Bernard Vatant
| 
| I'm lost here. Could you explain what practically will make the
| difference between valid and true for an application?

The application will assume that what is valid is also true, even
though we know that this need not be the case.

Basically, I was separating the notion of valid (inside the topic map)
from the notion of true (here interpreted as "correctly corresponding
with the state of the world").
 
* Lars Marius Garshol
|
| If the topic map has "X instanceOf Y in scope Z", and you set
| context scope to "W", and ask "is X an instanceOf Y" the system will
| have to say "no".
 
* Bernard Vatant
|
| Hmmm. I thought scope on "instanceOf" was not allowed, something I
| did not understand well BTW, we had a debate on that and you
| explained me why. Has it changed since?

You must have misunderstood me. The design of XTM explicitly allows
scope on instanceOf, if it is represented using associations typed
using the published subjects provided by XTM.

For a long time I was debating (partly with myself, partly with
others) whether that was acceptable or not, and in the end decided
that it *was* acceptable.

So the SAM accepts scope on instanceOf, which I hope is clear from the
text. If it is not, please say so, and we'll correct it.
 
| Anyway let's assume it. I maintain in this case that the only
| logical sustainable answer of the system there should be: "I don't
| know". Say for example "X is an instanceOf expert" in the scope
| "Markup Languages". The scope type there being the field of
| expertise. Now I switch the scope to "Astrophysics". To the question
| "is X an expert in the scope Astrophysics", I have no elements of
| answer, and there is no reason to answer "no" more than "yes".

To be flippant, it sounds as though you might just as well replace
this by a system that always replies "MAYBE", on the principle that
the underlying data might be wrong.

Prolog has the exact same problem as we face here, and there this was
resolved as follows. If the system replies "YES", it means: "yes, I
could prove your proposition from the factbase", and if it replies
"NO", that means "no, I could not prove your proposition". 

At no point does the question of truth enter the picture. Ensuring
that the factbase is truthful is the task of the user, and the
computer system can provide no help there.

Of course, one can use less loaded terms to refer to the outcome of
the process, which tolog does. If you say

  instance-of(lmg, expert)?

the system will reply with an empty result set. That means it didn't
find any matches. (Or, if you like, the statement was false.)

| [the danger of SAM inconsistencies]
|
| Absent formal model, inconsistencies unfortunately will not appear
| until you begin to use rules to build inferences, using TMQL or
| TMCL, or through any algorithm trying to drive inferences from the
| assertions expressed in a topic map. 

So the danger is that what we write in the SAM might lead us into
mistakes in the TMQL that again lead to inconsistencies in TMQL query
results. That sounds like a relatively remote contingency to me, but I
accept that it is possible. 

What can we do to ensure that there are no such problems?

| And my concern is that is the kind of things we are at the risk to
| find with the current fuzzyness on scope, or the absence of
| templates (another thread).

We'll revise the section on scope, once the dust has settled
sufficiently to allow us to see what should be done.

-- 
Lars Marius Garshol, Ontopian         <URL: http://www.ontopia.net >
ISO SC34/WG3, OASIS GeoLang TC        <URL: http://www.garshol.priv.no >