If Narsese does not map naturally to RDF, then I suppose that it can be encoded in RDF. For example in my KB facts are currently crisp, having a default truth value that is not naturally exported in RDF. But by reifying the fact in RDF, then a meta assertion can be created to state something about the fact, such as its numerical truth or its Bayesian probability value. I agree that terms should mean nothing if not considered within a context of associated statements, provided that two agents can exchange terms while sharing the same context of associated statements. -Steve
Stephen L. Reed Artificial Intelligence Researcher http://texai.org/blog http://texai.org 3008 Oak Crest Ave. Austin, Texas, USA 78704 512.791.7860 ----- Original Message ---- From: Pei Wang <[EMAIL PROTECTED]> To: [email protected] Sent: Friday, January 18, 2008 6:04:08 AM Subject: Re: [agi] NARS source opened Hi Steve, I haven't tried this yet, so cannot be very sure. My guess is that in principle it can be done, but not in a straightforward way. The problem is not in syntax, but in semantics. RDF is designed mainly to represent "objective facts", while Narsese is for "subjective beliefs". Most Narsese statements cannot be taken as "facts" after their truth-values are omitted, and the meaning of most terms cannot be determined by their denotation. Furthermore, there will be the issue of compound term. In general, I guess it won't be too hard for NARS to accept input in RDF --- the system can assume a default truth value. The other way around will be much more complicated, given the uncertainty in NARS beliefs. The grammar of Narsese is at http://nars.wang.googlepages.com/NARS-Guide.html , with sample sentences in http://nars.wang.googlepages.com/NARS-Examples-SingleStep.txt and http://nars.wang.googlepages.com/NARS-Examples-MultiSteps.txt . I look forward to meeting you in Memphis. Regards, Pei On Jan 18, 2008 3:21 AM, Stephen Reed <[EMAIL PROTECTED]> wrote: > > Hi Pei, > > I looked the source code a bit. Do you think that a NARS statement can be > represented as an RDF triple? Furthermore can a NARS term have a URI to be > RDF compatible? If so, then that would facilitate statement and term > exchange between our systems. > > Nice photo of you on the mountain! > > Cheers, > -Steve > Stephen L. Reed > > Artificial Intelligence Researcher > http://texai.org/blog > http://texai.org > 3008 Oak Crest Ave. > Austin, Texas, USA 78704 > 512.791.7860 > > > > > ----- Original Message ---- > From: Pei Wang <[EMAIL PROTECTED]> > To: [email protected] > Sent: Thursday, January 17, 2008 3:35:08 PM > Subject: [agi] NARS source opened > > Hi all, > > *. The most recent NARS source code has been put at > http://code.google.com/p/open-nars/ > > *. A group for NARS-specific discussion is set up at > http://groups.google.com/group/open-nars > > *. A book draft about the ideas behind NARS is turning into a > (non-technical) E-book on intelligence at > http://nars.wang.googlepages.com/gti-summary > > Since these websites are all at their initial stage, there are lots of > things to be added into each of them, which I'll do as soon as I have > the time. > > Feedbacks on all aspects of them are welcome. > > Pei > > ----- > This list is sponsored by AGIRI: http://www.agiri.org/email > To unsubscribe or change your options, please go to: > http://v2.listbox.com/member/?&; > > > ________________________________ > Looking for last minute shopping deals? Find them fast with Yahoo! > Search.________________________________ > This list is sponsored by AGIRI: http://www.agiri.org/email > To unsubscribe or change your options, please go to: > http://v2.listbox.com/member/?&; ----- This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?&; ____________________________________________________________________________________ Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ ----- This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=8660244&id_secret=87483852-11d5cd
