xml-dev - RE: Schemata are not just constraints [was: "RDF + Topic Maps" = TheFutu

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

Yes, that is the case.  The concept of 
authority is based on trust, the consent, 
in this instance, of the governed.  For 
the practical operational issues, that 
is all that is necessary in the broad 
perspective of global operations.  In more 
local operations, it isn't enough because 
the pockets of non-cooperators can be 
large enough to disrupt the system's 
efficiency to the degree that the 
broad operational result is gridlock.  
As put by an Internet pundit on the 
subject of bipartisan cooperation, 
"Sure, I'll pet your elephant and 
you can kiss my a**."  Not a pretty 
future and perhaps inevitable given the 
historical scenario but we need not 
debate it here for the example.

When we look at integrating ontologies, 
we have to consider that, taking the 
Topic Map stance, we are integrating 
opinions.  As pointed out in earlier 
emails, the problems of lots of little 
noisy systems which contribute to the 
binary decision result in a control 
adjudicating the decision when the 
votes are unacceptably close.  

That control has authority by legitimate 
or illegitimate means, and establishing 
legitimacy can be a very tough dogfight 
on the WWW.  The US Supreme Court has it 
by proclamation but with one decision 
just lost much of its credibility in 
the point of view of a large number.  
So, the ability of the ontology to 
advise credibly becomes a behavioral 
issue and that behavior has a statistical 
characteristic with regards to the 
ontological commitment.  We can make  
statistical assertions about that and 
that is all.

Again, the ontological commitment 
is to the authority of the ontology 
that the relationships expressed are 
credible and useful.   Operational 
means for creating the ontology are 
part of the authoritative credential 
of the owner of the ontology.

Len Bullard
Intergraph Public Safety
clbullar@ingr.com
http://www.mp3.com/LenBullard

Ekam sat.h, Vipraah bahudhaa vadanti.
Daamyata. Datta. Dayadhvam.h


-----Original Message-----
From: Martin Bryan [mailto:mtbryan@sgml.u-net.com]

Len.
>
> It seems reasonable to test the existence of
> assertions.   We can look at multiple ontologies
> (contexts of assertions) and discover that multiple
> sources have the same assertion so establish
> evidence by multiple sources.  We can never
> formally prove the assertion unless we both
> agree to a test of fact and commit to behave
> accordingly in a testable way.

And who will test the testers? One man's proven assertion is another man's
obvious mistake, as adherents of any two religions will tell you, volubly.
The real key is "What proportion of a community are willing to identify the
same stated meaning as being valid to their understanding of a term/phrase.
If 75% of Americans agree that "George Bush is the legitimate president of
the US" is or is not that enough to make the phrase "George Bush, US
President" a meaningful one from 20th January 2001, even if it is not true
today, and will not be true on 20th January 2020. Creating a test for the
truth of the statement will not help if the validity of the choice of judges
is challenged, as was the case in the Florida and Supreme Court rulings.