Hmmm it seems we now have the technology to build an automated LessWrong poster. Merely program the bot to comment in heavy traffic threads where it sure to be noticed then after a fixed amount of time delete any comments that get less than 1 karma.
I hereby request that the technology be named The Automated Teacher’s Password Guesser, because your comment about the higher frequency of such comments was the first thing I thought of when I saw this.
after a fixed amount of time delete any comments that get less than 1 karma.
Of course, it’s practically a solved machine learning problem to also use those comments to learn what sorts of comments get downvoted and improve the bot.
If someone makes a really well-optimized version of Armok’s bot that pays attention to key words in comments and replies somewhat intelligently, I’ll give them the Will_Newsome LessWrong account to use for that purpose.
Alternatively I think a Will_Newsome bot that made lots of references to Leibniz and algorithmic information theory and decision theory concepts and then appealed to secret evidence would be pretty cool.
Alternatively I think a Will_Newsome bot that made lots of references to Leibniz and algorithmic information theory and decision theory concepts and then appealed to secret evidence would be pretty cool.
Awesome, that’ll go great alongside the “Ban comments by Will Newsome” bot we provide Alicorn!
I’m to the left of him and I agree, but I have to note that in our age political large-C Conservatism and being small-c conservative in practice have basically nothing to do with each other.
Hmmm it seems we now have the technology to build an automated LessWrong poster. Merely program the bot to comment in heavy traffic threads where it sure to be noticed then after a fixed amount of time delete any comments that get less than 1 karma.
Man and machine, wisdom extreme!
I hereby request that the technology be named The Automated Teacher’s Password Guesser, because your comment about the higher frequency of such comments was the first thing I thought of when I saw this.
Of course, it’s practically a solved machine learning problem to also use those comments to learn what sorts of comments get downvoted and improve the bot.
If someone makes a really well-optimized version of Armok’s bot that pays attention to key words in comments and replies somewhat intelligently, I’ll give them the Will_Newsome LessWrong account to use for that purpose.
Alternatively I think a Will_Newsome bot that made lots of references to Leibniz and algorithmic information theory and decision theory concepts and then appealed to secret evidence would be pretty cool.
Awesome, that’ll go great alongside the “Ban comments by Will Newsome” bot we provide Alicorn!
I am now reminded of M-x psychoanalyze-pinhead.
You’re too conservative Konkvistador, I think I’d rather use these powers for chaos & destruction. Bwa, ha, ha.
I’m to the right of Konkvistador and I think this is a fully general argument contra Konkvistador.
I’m to the left of him and I agree, but I have to note that in our age political large-C Conservatism and being small-c conservative in practice have basically nothing to do with each other.