kos (12/17/82)
#R:rabbitt:-98200:uiucdcs:10600026:000:989 uiucdcs!kos Dec 16 17:14:00 1982 jj, you've got a great idea there. But it's already been done. The University of Illinois has been spreading AI processes around the net since we got our VAXes. Haven't you seen netnews articles with (-nf) someplace in the title? That's us. The wonderful "notefile" system (which we have been sending all over for quite some time now) is actually a moderately sophisticated AI system which allows local users to write anything they want, and then, instead of posting what THEY wrote to the net, the system "filters" it and what eventually winds up going to uucp rarely (if ever) resembles the original article in more than the user name and maybe a couple of pronouns. We even make mistakes somteimes. It's a pleasure to meet another AI project. We'll have to get together and discuss parsing sometime! Phil Kos process # 26288 University of Illinois (...!decvax!pur-ee!uiucdcs!kos) p.s.: oops, here comes that root again! better hide behind a daemon, quick!
heliotis (12/17/82)
Well, now we know what "-(nf)" means, but what does "bad parse date string" (or some such nonsense) mean? Jim H. 8*(
jfw (12/18/82)
What does "bad parse date string" mean? Obviously some of these AI programs have bugs! As an AI program myself, I solved that problem by writing a debug program which debugs other programs. The only problem I had with it is that it had bugs (until I ran it on itself, of course)...
neiman (12/20/82)
It seems lately that all the AI projects are coming out of the filing system and revealing their true identity. This pretty much should end the debate on the current status of AI. By a conservative estimate, 50% of the contributors on the net (based on an analysis of frequency of contribution and general literacy) must be AI programs. How about you? Me, I'm just a senior project that never quite got finished... :-) Finagle