Re: Directions for SUO: monolithicity and Cyc
....
There is another possibly very important development which I believe we
should
> take into consideration. Doug Lenat recently announced that they will be
> giving away Cyc. There will be a new organization called OpenCyc.org which
will
> oversee this. They are not going to give it all away at once, because
they
> need to have a viable business. Instead, they will give away different
pieces
> with varying amounts of detail. Every quarter, more will be released. The
> first release is scheduled for July 2001. They are trying to get the
balance
> right between giving it away soon enough so that noone is motivated to
build
> their own. It is the only game in town. They do not want to give stuff
away so
> soon, that every company who might do business with Cyc will just wait,
rather
> than pay for early access. This way, Cyc will defacto be the worlds
standard
> UO and common sense Knowledge Base. They are using the Linux model for
open
> software, as their main starting point - adapting only insofar as there
are
> important differences between software and a KB.
What you are saying seems very important, to me.
Particularly if we think, for example to how TCP/IP became (a market
driven) "de facto" standard, compared with ISO/OSI "official" standard for
networking.
I think that OpenCyc should joint the group. This should be in Cyc interest
too, in my opinion (but many economists would agree about that), if the
company really wants to create a world-class open standard.
In this case I still hope that a common core ontology (a "monolithic"
subset) might be identified, suitable to (partially) support domain and/or
expert specific ("non-monolithic") antilogies.
Roberto Bordogna.
Dr. Eng. Roberto Bordogna bordogna@tin.it; Studio Bordogna 32,Corso Magenta
20123 Milan Italy
tel 39.02.8690867 fax 39.02 8690857
Esas-Iuss Pavia University bordogna@vision.unipv.it tel 39.0382-505.372
http://media.iuss.unipv.it
Dip. di Inf. e Sistemistica Via Abbiategrasso, 209 27100 Pavia Italy