On 10/8/2010 7:48 PM, John Bottoms wrote:
> Can we say:
> FOL is a subset of HOL.
> HOL is a small subset of English stated in a different notation. (01)
Yes. But the point I was trying to make is that there is an open
ended range of different levels of expressive power. Most programmers
never worry about those issues, and they work quite happily at the
upper end of the expressive range. (02)
On 10/8/2010 8:10 PM, Cecil Lynch wrote:
> I would say you [Ken] are asking a bit more from skos than it was
> designed to do. You are describing constraints best expressed in owl. (03)
I agree with the first sentence. But I would say that those constraints
can be better expressed in controlled English: (04)
Every object has a length, a weight, and a color.
For every object x, the color of x is either red or yellow or blue. (05)
Why should we have to design a different notation for every class of
problems? There are good reasons for having diagrams to supplement
linear notations (e.g., UML). There are also good reasons for using
spreadsheets or mathematical formulas for some kinds of knowledge.
But all the notations and diagrams should be integrated into a *single*
knowledge representation system. (06)
For any particular problem, a well-designed computer system can decide
which algorithm and which inference engine(s) to use for each problem.
That's what Cyc does. That's also what the high-performance theorem
provers for the tptp.org problems do. There are "knowledge compilers"
that can extract axioms from systems like Cyc and repackage them for
different purposes and different algorithms. See (07)
http://www.jfsowa.com/pubs/fflogic.pdf
Fads and fallacies about logic (08)
I also have some criticisms of Cyc. But I definitely do *not* recommend
forcing SMEs to worry about computational complexity before they can
describe what they know about their subject. Computers can make those
tradeoffs much more effectively than the SMEs. (09)
John Sowa (010)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (011)
|