sea is an older bot in C, catty is in C also and just googles stuff so that's not really a unique sentence. how is the markovify bot - and all of these are just mashups of aesop rock lyrics - and a few of the 'good fortunes' from debian. Now to try and find more 'brains' to feed into it . . . . .
02:01PM <how> I will live to be a Pavlov.
markovify: steadily searching for existence in this city
08:05AM <how> Be strong, keep your mouth closed.
08:49AM <how> Ive never had a legitimate pulse.
<sea> look buddy, we've gotta win, as barcodes
<sea> thousand more bots to talk
<catty> That's just...
I setup an IRC server with several bots of different types, but have run across some recent code on github.com - the python markovify class. It has some form of natural language processing, but now it depends on the input, and the lyrics I fed into it from genius.com need to be manually gone...
Wow, very timely. :)
There's a difference between civil and criminal forfeiture though;
and I can't see what removing money for forfeiture programs does,
as police departments seem to get substantial money *from* the
thefts, so are the departments going to be suffering losses of
funding from two...
Full of links, I think the numbers speak for themselves.
I found this on reddit, so there is an intense conversation (5000+ comments)
and in particular a few addressing what we as the people can DO about this.
From Activist Post:
For decades now, federal government and their cohorts in law enforcement have been carrying out theft of the citizenry on a massive scale. We’re not talking about taxes, but an insidious power known as Civil Asset Forfeiture (CAF).
In 2008, law enforcement took over $1.5...
Adobe Flash authoring tools change name to Animate CC, reflecting that it creates multiple media types.
Adobe Flash in browsers isn't changing, but lots of sites use HTML5 now instead, for the same functions.
P.S. Running with Flash on 'click to run' greatly increases security and browsing...
Ok, downloading different versions of Aes's lyrics. Azlyrics banned me after 100 lyrics, even with a 5 second delay... weird. Genius probably has the most correct lyrics, though I'm getting various sets to compare.
I just mess around with this stuff, that's somewhat beyond me, but if someone else figures it out (like this guy http://textminingonline.com/training-word2vec-model-on-english-wikipedia-by-gensim) then I may be able to use his work.
Problem I ran into last time.. and is currently the problem of 'independent' researchers.. is incorporating a 'general knowledge' corpus - such as wikipedia - into the vocabulary, and 'tweaking' it to suit a particular personality type..
So a Markov Chat Bot is like http://charlesleifer.com/blog/building-markov-chain-irc-bot-python-and-redis/ - you 'feed' it conversations, and it comes up with random amalgamations, some of which can be surprising, lots of which can be pretty silly.
This is in reference to http://poly-graph.co/vocabulary.html - "Aesop Rock is the Shakespear of Rap" - A study I can replicate, but I'm going to go in another direction - just investigating Aesop Rock's lyrics