>Nathan Whitehorn wrote: >> Of course, it's an even better idea to let them know aht they've >> done right, as I said in an earlier e-mail. I've gone *years* >> without hearing any feedback on several of my programs, despite >> the fact that quite a number of people use them. And that, more >> than anything else, is extremely discouraging. > >Or it could mean something else - bug free programs! Still, it's >nice to hear from people using your software. Or even from >people not using it (currently I get a fair bit of interest >on BeShare in getting my spam detector finished, which makes >me more enthusiastic about working on it). But part of that >is speaking up so that people using your software know that >you're around. Though that's not a concern with you, Nathan :-). You do realise that the link below is not actually a Bayesian technique :-) Have a read of http://radio.weblogs.com/0101454/stories/2002/09/16/spamDetection.html >P.S. Blatant plug, hope to have a version on BeBits in a week or two: >http://members.rogers.com/agmsmith/beos/AGMSBayesianSpamScreenShot3.png >And an explanation of how it works: http://www.paulgraham.com/spam.html -- Cheers David What this country needs is a dime that will buy a good five-cent bagel.