(This is a side discussion and should be deleted unread, unless you find
this useful) (01)
well John (02)
I do not need a study list from you, thank you anyway. (03)
It becomes a religion, John. (04)
One cannot see what the argument is that I make regarding formalism - if one
gets side tracked on other issues, like the legacy use of the relationa;
database. If this was the governing argument before Tesla and Westhouse
brought alternating current to the market, then we would all still be using
direct current to light our homes. (05)
So I feel very much like someone whose discussion with a relgious group ends
in "Paul you just do not understand, perhaps you should read more of our
literature." (06)
Clearly there is some sense of "logic" that I feel is more appropriate to
computing with tokens that are strings of letters whose properties and
attributes are specified in an Ontology standard like OWL, or some other
standard like what others may propose based on triples that are not
templated as subject, verb, predicate as the RDF triples are. And it should
not be so hard to move the discussion to explore these issues. Right? (07)
There are numerous works on "quantum logic" and this "type" of logic is
quite different from standard set theory. (08)
Paul (Werbos) and I have had many discussions about the works of Robert
Rosen and the "claimed" adequacy of common notions of set theory, and common
notions of inference. (09)
He and I have different points of view, but perhaps he will make a comment. (010)
-----Original Message-----
From: John F. Sowa [mailto:sowa@xxxxxxxxxxx]
Sent: Saturday, November 26, 2005 12:17 PM
To: Paul S Prueitt
Cc: Paul J. Werbos
Subject: Re: on the sufficiency of predicate logic (011)
Paul, (012)
Indeed there are many engineered systems that are
based on logic. Preeminent among them are all
the SQL databases in the world. The entire
economy of the world runs on them, and nobody
is going to get rid of them. (013)
> If you stand and say, "what I say is true and I
> am the authority"; then I have to rely on a more
> complex argument based in neuroscience, connectionist
> paradigm and complexity theory. At that point I have
> to look across at you and ask if you know differential
> equations, stochastic theory, and topology (as well as
> I clearly do). (014)
I understand differential equations, quantum mechanics,
and related systems very well indeed, and I can assure
you that there is no conflict between them and logic.
All of mathematical reasoning at every level uses
classical logic for its proof theory. And indeed there
are very important uses for continuous mathematics as
a supplement to discrete mathematics. See for example,
the following slides by Arun Majumdar and me: (015)
http://www.jfsowa.com/talks/continuo.pdf (016)
And I can also assure you that there is no conflict
between logic and connectionist methods, probabilistic
methods, etc. They are just as compatible with logic
as any other mathematical systems. (017)
Anyone who is doing work in any area of computer science
should have all of these techniques in their toolkit.
They are not in competition. They are complementary. (018)
I made the following points on my own authority by virtue
of the fact that nobody who understands anything about
logic disputes them. If you doubt them, I'll send you
a study list. (019)
John
_________________________________________________________ (020)
From: sowa@xxxxxxxxxxx [mailto:sowa@xxxxxxxxxxx]
Sent: Friday, November 25, 2005 5:12 PM
To: Paul S Prueitt
Subject: Offline note (021)
Paul, (022)
There are some pointed criticisms that one can raise against
many of the logic-based proposals. But if you don't understand
logic, you can't even begin to formulate them, and you will not
be taken seriously. (023)
> Would you help me understand why you feel that a decrease
> in detail can be done with "precise" meanings. (024)
Ambiguity and vagueness are totally different. Ambiguity
means multiple interpretations, each of which can be precise.
Vagueness means that the boundaries are unclear, and it is
not possible to make a sharp distinction between what is
being asserted and what is being denied. (025)
If you have a precise specification with many axioms, you can
delete any axiom and still have just as precise a specification
as you had before. But it will allow more interpretations --
in other words, you could say that it is more ambiguous. (026)
Each axiom rules out some possibilities. Each time you
delete one, you have more options. If you delete them all,
you have no restrictions whatever, and things are possible. (027)
When you can delete *all* the axioms, you have a statement
that says absolutely nothing. But it is still precise. (028)
These are elementary principles about logic, and if you don't
understand them, you can't expect anybody to take your
criticisms of logic seriously. (029)
John (030)
_________________________________________________________________
Message Archives: http://colab.cim3.net/forum/ontac-forum/
To Post: mailto:ontac-forum@xxxxxxxxxxxxxx
Subscribe/Unsubscribe/Config:
http://colab.cim3.net/mailman/listinfo/ontac-forum/
Shared Files: http://colab.cim3.net/file/work/SICoP/ontac/
Community Wiki:
http://colab.cim3.net/cgi-bin/wiki.pl?SICoP/OntologyTaxonomyCoordinatingWG (031)
|