karp@SUMEX-AIM.STANFORD.EDU (Peter Karp) (05/23/89)
In a recent paper, I argue that the meaning of the term "first principles" is at best highly sensitive to the particular domain of discourse, and at worst, vaccuous. (See "An Analysis of the Distinction Between Deep and Shallow Expert Systems", by Karp and Wilkins in an upcoming International Journal of Expert Systems). If someone would like to offer a precise definition of first versus other types of principles, I would be very interested. I actually don't think the message that we're all refering to (written by Dan Davison?) was meant to inspire such serious debate -- but perhaps it should have. I for one would not mind seeing a tighter definition of the goals of The Matrix; such a definition should probably not include words like "evangelize." Although I am not a physicist, my understanding is that we need not envy the "first principles" of physicists, because they are often computationally intractable -- consider the three-body problem and the fact that (I think) quantum mechanics is too complex to be applied to single large atoms. Biological first principles, such as those used to compute protein structures, may also have this property. Thus we turn to approximations and abstractions. Peter Karp CS Dept Stanford