Forums before death by AOL, social media and spammers... "We can't have nice things"
|    comp.ai    |    Awaiting the gospel from Sarah Connor    |    1,954 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 12 of 1,954    |
|    JXStern to All    |
|    Re: The AI Theory-of-Everything (Was: Ho    |
|    25 Jul 03 00:17:34    |
      From: JXSternChangeX2R@gte.net              On Thu, 24 Jul 2003 02:02:29 GMT, jorn@enteract.com (Jorn Barger)       wrote:       >> >> Expert systems were built around the idea of "surface knowledge" being       >> >> valuable and workable without an in-depth analysis and modelling of a       >> >> system's innards.       >> >Was this explicitly debated at the time?       >>       >> It was asserted at the time and argued afterwards, I'm not sure       >> "debate" is the appropriate term! The proof was supposed to be in the       >> pudding. A quick Google now does not show that the issue survived       >> very well, but it was fairly big then.       >       >I'd be interested to know what happened-- I think AI and       >simulation are two sides of the same coin.              Two sides of the same dodecahedron, maybe.              The thing is, starting with nineteenth century positivism, then going       into twentieth century behaviorism and verificationism, the idea came       about that you could, should, must treat things as black boxes.              Later (circa 1960) came the idea that these were wrong and you could,       should, and perhaps must produce performance models of things, with       functionalism bridging any variances in platform.              Then (circa 1980) came the idea that functionalism misses the point       somehow, and the state of the art seemed to return to logicism, where       the idea was to find the truest logical representation of things. Odd       date, that, since the last AI bubble was in the early 1980s, flying in       the face of this retreat (as I see it) to logicism.              Now I'd say that AI and simulation are related somehow or other, but a       careful discussion has to separate issues of reference,       representation, performance, competence, necessity, convention, and a       few other things. Twelve faces is probably not enough!              Joshua Stern              [ comp.ai is moderated. To submit, just post and be patient, or if ]       [ that fails mail your article to |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
(c) 1994, bbs@darkrealms.ca