--- Log opened Fri Apr 18 00:00:43 2014
00:02 -!- kyknos_ [~kyknos@89.233.130.143] has quit [Ping timeout: 240 seconds]
00:03 -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap
00:06 -!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap
00:43 -!- ryankarason [~rak@opensource.cse.ohio-state.edu] has joined ##hplusroadmap
00:57 -!- nsh_ [~nsh@wikipedia/nsh] has joined ##hplusroadmap
00:57 -!- petora [~asakharov@24.60.79.55] has quit [Quit: quit]
00:57 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap
00:58 -!- nsh [~nsh@wikipedia/nsh] has quit [Ping timeout: 252 seconds]
01:00 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has quit [Remote host closed the connection]
01:10 -!- nsh__ [~nsh@wikipedia/nsh] has joined ##hplusroadmap
01:12 -!- nsh_ [~nsh@wikipedia/nsh] has quit [Ping timeout: 276 seconds]
01:14 -!- Adifex is now known as Adifex|zzz
01:14 -!- Adifex|zzz is now known as Adifex
01:14 -!- Adifex is now known as Adifex|zzz
01:15 -!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 252 seconds]
01:35 -!- nsh__ is now known as nsh
01:40 < chris_99> paperbot: http://link.springer.com/chapter/10.1007/3-540-60902-4_12
01:40 < paperbot> http://diyhpl.us/~bryan/papers2/paperbot/5b8b88e46d715615386c67d9d4e4baa3.txt
01:46 -!- nsh_ [~nsh@wikipedia/nsh] has joined ##hplusroadmap
01:46 -!- nsh [~nsh@wikipedia/nsh] has quit [Ping timeout: 258 seconds]
02:07 -!- nsh_ is now known as nsh
02:11 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds]
02:25 -!- nsh [~nsh@wikipedia/nsh] has quit [Ping timeout: 240 seconds]
02:31 -!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap
02:40 -!- devrandom [~devrandom@70-36-136-78.dsl.dynamic.sonic.net] has quit [Ping timeout: 240 seconds]
02:41 -!- devrandom [~devrandom@70-36-136-78.dsl.dynamic.sonic.net] has joined ##hplusroadmap
03:37 -!- yorick [~yorick@oftn/member/yorick] has joined ##hplusroadmap
03:44 -!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has quit [Ping timeout: 240 seconds]
03:47 -!- EnLilaSko [EnLilaSko@unaffiliated/enlilasko] has joined ##hplusroadmap
03:48 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap
03:50 -!- mosasaur [~mosasaur@188.88.208.189] has joined ##hplusroadmap
03:57 -!- ElixirVitae [~Shehrazad@95.5.64.15] has joined ##hplusroadmap
03:57 -!- ElixirVitae [~Shehrazad@95.5.64.15] has quit [Changing host]
03:57 -!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has joined ##hplusroadmap
04:09 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving]
05:04 -!- ThomasEgi [~thomas@185.5.8.81] has joined ##hplusroadmap
05:04 -!- ThomasEgi [~thomas@185.5.8.81] has quit [Changing host]
05:04 -!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap
05:07 -!- FourFire [~fourfire@108-214-15.connect.netcom.no] has joined ##hplusroadmap
05:12 -!- strangewarp [~strangewa@c-67-176-51-230.hsd1.co.comcast.net] has quit [Read error: Connection reset by peer]
05:12 -!- strangewarp [~strangewa@c-67-176-51-230.hsd1.co.comcast.net] has joined ##hplusroadmap
05:23 -!- nsh [~nsh@wikipedia/nsh] has quit [Ping timeout: 250 seconds]
05:49 -!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap
05:52 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap
06:06 -!- jamesz [~ielo@c-69-255-193-198.hsd1.md.comcast.net] has joined ##hplusroadmap
06:46 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has joined ##hplusroadmap
07:12 -!- audy [~audy@unaffiliated/audy] has quit [Excess Flood]
07:15 -!- audy [~audy@unaffiliated/audy] has joined ##hplusroadmap
07:18 -!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has quit [Quit: Leaving]
07:18 -!- yorick [~yorick@oftn/member/yorick] has quit [Remote host closed the connection]
07:18 -!- ElixirVitae [~Shehrazad@95.5.64.15] has joined ##hplusroadmap
07:18 -!- ElixirVitae [~Shehrazad@95.5.64.15] has quit [Changing host]
07:18 -!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has joined ##hplusroadmap
07:30 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving]
08:13 -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap
08:26 < dingo> hey what the hell
08:30 < FourFire> ?
08:30 < JayDugger> Eh?
08:31 < dingo> this guy's got my nickname as his ident
08:31 < dingo> delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap
08:32 -!- dlfk [~dlfk@chem-179-154.chem.tamu.edu] has joined ##hplusroadmap
08:32 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap
08:33 < FourFire> dingo, is an animal
08:34  * dingo drools
08:34 < FourFire> and not some overly obscure one, like, say meercats, or sloths
08:35 < dingo> i was like 11 then, and it was before internet, i wasn't very creative
08:44 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Ping timeout: 252 seconds]
09:13 < FourFire> why did you have an IRC nick before the internet?
09:14 < juri_> hio.
09:20 -!- sandeepkr [~sandeep@117.254.216.28] has joined ##hplusroadmap
09:47 -!- yorick [~yorick@oftn/member/yorick] has joined ##hplusroadmap
09:48 -!- nsh [~nsh@wikipedia/nsh] has quit [Ping timeout: 276 seconds]
09:52 -!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap
09:54 < xmj> someone explain FourFire that nicknames exist in reallife
09:55 < FourFire> xmj ;)
10:04 <@kanzure> ugh
10:05 < ParahSailin> anyone know the deal with opal stop codon being translated as arg
10:06 < ParahSailin> http://www.ncbi.nlm.nih.gov/nuccore/KF985959.1
10:06 < ParahSailin> fucking virus how do they work
10:06 <@kanzure> freedom infection just isn't going to fly
10:07 < dingo> < FourFire> why did you have an IRC nick before the internet?
10:07 < dingo> dial-up bulletin board systems
10:08 < dingo> and it was my fidonet nick, etc.
10:08 < ParahSailin> if a fungus can infect an ants brain to make it fly to freedom, the freedom virus will work
10:09 -!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 240 seconds]
10:09 < ParahSailin> paperbot: http://www.pnas.org/content/87/22/8860.full.pdf
10:09 < paperbot> http://diyhpl.us/~bryan/papers2/paperbot/520b988bd4f53bace772caf6879c21ef.pdf
10:13 -!- nmz787_i [~nmccorkx@192.55.54.38] has joined ##hplusroadmap
10:20 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap
10:22 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 240 seconds]
10:28 -!- marciogm [~marciogm@179.126.74.112] has joined ##hplusroadmap
10:35 -!- sandeepkr [~sandeep@117.254.216.28] has quit [Quit: Leaving]
10:53 -!- Qfwfq [~Qfwfq@cm113.kappa36.maxonline.com.sg] has joined ##hplusroadmap
10:55 -!- mosasaur [~mosasaur@188.88.208.189] has quit [Quit: Leaving.]
11:00 <@kanzure> have there been any reasonable proposed alternatives to researchers writing papers?
11:01 <@kanzure> like, not papers
11:02 < xmj> you somehow need to document your research
11:02 < xmj> how would you do it?
11:02 <@kanzure> depends on what i'm doing
11:02 <@kanzure> paper-style research heavily constrains the type of work that you're expected to engage in
11:03 < xmj> to what?
11:03 <@kanzure> to work that gets written up in journal articles? duh
11:03 < xmj> you mean 'accepted by journals'
11:03 <@kanzure> not just that
11:03 <@kanzure> i don't care about the fact that rejections exist
11:03 <@kanzure> i mean, it makes sense, that if you're going to have acceptances then you're going to need rejections
11:05 < xmj> and your point is...?
11:05 <@kanzure> my point is that i do not mean "accepted by journals"
11:05 <@kanzure> please try to follow along.
11:06 < xmj> please do make a point and try not to be enigmatic.
11:06 <@kanzure> uh, i was responding to your statement
11:06 <@kanzure> this is how conversations work
11:06 < xmj> i was responding to your conversation style
11:07 <@kanzure> go on?
11:07 < nmz787_i> you're both just writing, stop this, and do some other form of communication to prove/disprove kanzure's point!
11:07 < xmj> you ask, are there reasonable proposed alternatives to papers?
11:07 < nmz787_i> all journal articles are now conference phone calls
11:07 < xmj> then you go on a tirade about what's bad about papers
11:07 < xmj> not how you'd do it by NOT using paper-style.
11:07 <@kanzure> i did not actually say anything about what's bad about papers
11:08 < nmz787_i> mo3
11:08 < nmz787_i> mp3*
11:08 < xmj> "paper-style research heavily constrains the type of work that you're expected to engage in"
11:08 < xmj> if constraining isn't bad why do you care about finding an alternative?
11:08 < nmz787_i> and really they aren't papers most times these days, they're documents
11:08 <@kanzure> constraints can be definitions, and i don't think definitions are inherently bad
11:09 < xmj> right
11:09 < xmj> again
11:09 < xmj> how else would you do it?
11:09 <@kanzure> so either i have to have a proposal, or else i think that definitions or papers are inherently bad?
11:09 < xmj> i haven't joined in this conversation to discuss your bikeshed color or whether or not papers are bad.
11:10 < xmj> you ask for a different way to do it, i'm curious about how you'd do it.
11:10 < xmj> go ahead.
11:10 <@kanzure> uh?
11:12 < FourFire> We need some way to keep score
11:12 <@kanzure> i don't know why you are saying these things to me
11:12 < FourFire> something better than papers would be nice
11:12 -!- delinquentme [~dingo@192-195-81-250.PUBLIC.monkeybrains.net] has joined ##hplusroadmap
11:12 < nmz787_i> conference calls.
11:12 < xmj> 21:00:49 <@kanzure> have there been any reasonable proposed alternatives to researchers writing papers?
11:12 < xmj> 21:01:04 <@kanzure> like, not papers
11:12 < nmz787_i> mp3s
11:12 < FourFire> what are papers opitmized for?
11:12 < xmj> Something better than papers would be nice, but I read much faster than listen to things.
11:13 < FourFire> accessibility? Ability to file away?
11:13 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap
11:13 <@kanzure> i don't understand what my question has to do with 11:10 < xmj> you ask for a different way to do it, i'm curious about how you'd do it. 11:10 < xmj> go ahead.
11:13 < xmj> I'm much more into the higher information density conveyed in papers. Audio/multimedia is slow
11:13 < nmz787_i> xmj: I am joking, I think documents are a great way to iteratively re-organize thoughts
11:13 < FourFire> it's 15th century tech, sure
11:13 <@kanzure> mp3 is a valid suggestion, but a bad one
11:13 < xmj> nmz787_i: can you try to explain kanzure what i mean?
11:13 -!- Qfwfq [~Qfwfq@cm113.kappa36.maxonline.com.sg] has quit [Ping timeout: 250 seconds]
11:14 < nmz787_i> i think its a draw!
11:14 < xmj> i'm pretty sure he's trying to not get me intentionally, this is boring :p
11:14 < FourFire> agreed, that text is higher density than, video
11:14 <@kanzure> xmj: have you considered that maybe i'm just an idiot?
11:14 < xmj> sure
11:15 < xmj> and i'm sure nmz787_i is better at talking to idiots than i am :D
11:15  * xmj chuckles
11:15 < xmj> happy holidays.
11:16 < FourFire> kanzure, what do you find suboptimal about papers ?
11:16 <@kanzure> not machine readable
11:16 < FourFire> so... you want people to file their research in binary?
11:17 < xmj> oh
11:17 <@kanzure> they already do, they transmit ascii over binary to their publishers
11:17 < xmj> brilliant
11:17 < xmj> i have an idea.
11:17 < xmj> let's make a machine-parsable xml-based paper standard.
11:17 <@kanzure> xmj: please try to be less obnoxious
11:17 <@kanzure> xmj: bad ideas like that should be cause for banning
11:18 <@kanzure> xmj: people take that shit seriously and then you end up with crap like jstor.org
11:18 < xmj> kanzure: xml is the defacto standard of machine parsability
11:18 < xmj> kdbus.
11:18 < xmj> case closed
11:18 <@kanzure> wrapping a pile of text in xml is not exactly machine readable
11:18 <@kanzure> and why should there be a pile of text in the first place? fuck you
11:19 < FourFire> you know, if that's the case then some dialect of english needs to be made by AI people, which can be understood by their AI interpreter, that is NOT a programming language, and then people can write their papers in that dialect, like simple wikipedia sort of does
11:19 <@kanzure> why should they be writing like that?
11:19 < xmj> FourFire: lisp is a programming language
11:20 <@kanzure> what is this fascination with piles of written text
11:20 < xmj> you could probably shape it into an easily AI-parsable document-style written in reverse polish notation.
11:20 <@kanzure> i asked for alternatives, not xml wrappers or AI handwaving
11:20 < FourFire> xmj but is is comphrenehsible to someone who doesn't know how to program?
11:21 <@kanzure> i don't think that programming is the only possible alternative
11:21 < nmz787_i> "FourFire: agreed, that text is higher density than, video"  waiiiitttttt a sec, I've always heard a picture is worth a 1000 words
11:21 < xmj> FourFire: does it have to be? we're discussing ideas, they don't have to be optimal
11:21 < FourFire> the point is, that the language is a tool which is mutually intelligible by unskilled people, And specially skilled computers
11:21 <@kanzure> but why should language be the tool to solve "this problem"
11:21 < FourFire> nmz787, yes, because you can pretend that it is a still, a moment in time
11:22 < FourFire> you can say things with a picture which you cannot say in the same way with words
11:22 <@kanzure> just because everyone is capable of writing doesn't mean that scientific progress should be based on writing long-form documents
11:22 < FourFire> you can unveil a brutal truth, on people who will not listen, by showing them a picture
11:22 <@kanzure> everyone is capable of singing, should we be singing our science
11:22 < FourFire> kanzure, I disagree
11:22 < FourFire> I cannot sing
11:23 <@kanzure> "singing" when broadly and ambiguously defined :)
11:23 < xmj> has anyone read neal stephenson's anathem?
11:23 < nmz787_i> hieroglyphics science!
11:23 < xmj> in that book they actually do sing mathemathic formulas.
11:23 < FourFire> how can you transfer information to a computer using a protocol based on instinctive human singing sound-rules?
11:24 <@kanzure> nmz787_i: hieroglyphics is still writting
11:24 <@kanzure> *writing
11:24 < FourFire> I mean, decades from now, we'll be inputting directly from out brains with passive EM scanners
11:24 < FourFire> but that's not now
11:25 < nmz787_i> i thought they were more drawings
11:25 < FourFire> kanzure, have you read some of simple.wikipedia.org ?
11:25 <@kanzure> yes, why?
11:25 < FourFire> so, an english dialect which is in that general direction, so that it is easier to code AI to understand it
11:26 < xmj> *why*
11:26 < FourFire> as long as a concept can be decomposed into simple words, and the computer can understand the simple words
11:26 <@kanzure> why does it have to be text though
11:26 < FourFire> xmj transfer of *knowledge*
11:26 < FourFire> to computers
11:26 < FourFire> so that they can comprehend and do stuff with it
11:26 < xmj> there's been such good research on natural language processing that machines are getting better and better
11:26 < FourFire> it could be audio too
11:26 < FourFire> but that's another layer of apporximation
11:27 < xmj> wouldn't it make sense to leverage that, make it more efficient, hook it into some Friendly AI, and get going with singularity?
11:27 <@kanzure> xmj: please fuck off
11:27 < FourFire> my accent, in english for example is pretty messed up, but I reckon 95% of english speakers can still get what I'm saying
11:27 < nmz787_i> I only have like 4 or 5 senses, sight, touch, hearing, smell, taste
11:27 <@kanzure> xmj: FourFire is a moron and wont be able to tell that you're being sarcastic
11:27 < xmj> kanzure: if everything else fails, insult people
11:27 < xmj> oh, i wasn't
11:27 < xmj> NLP is going steady
11:27 <@kanzure> xmj: you legitimately believe that NLP is Friendly AI? or what
11:27 < FourFire> you were being sarcastic?
11:28 < xmj> are you an idiot?
11:28 -!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has quit [Quit: Leaving]
11:28 < nmz787_i> xmj: he is pretty bad at doing that, he has hurt my feelings before in a semi-lasting manner
11:28 < xmj> NLP is not a friendly AI
11:28 < FourFire> well I'm sorry I took your bait then, I thought you were sincere
11:28 < FourFire> (I didn't mean to!)
11:28 <@kanzure> oh "hook it into"
11:28 < xmj> and if you READ my sentences, you'd know i did not confuse them
11:28 < nmz787_i> xmj: but he seems to come around
11:28 <@kanzure> "well if it's just that simple"
11:28 < xmj> jesus
11:28 < xmj> nmz787_i: I was totally serious.
11:28 < xmj> kanzure: please fuck yourself.
11:28 <@kanzure> you really think that there are friendly ai implementations laying around?
11:28 < xmj> it might help
11:28 <@kanzure> way ahead of you
11:29 < xmj> some people have issues when their balls get too blue..
11:29 < FourFire> (ok, reading your next sentences, NLP has NOTHING to do with AI, I would have gotten it then)
11:29 < cpopell> welp, this seems like a fascinating bout of dick-fencing to wade back into
11:29 <@kanzure> often people conflate ai stuff with natural language processing
11:30 < xmj> FourFire: yes, but you'd think a wellbehaved AI could understand language if NLP was sufficiently welldeveloped.
11:30 <@kanzure> could we please not just handwave ai into our solutions
11:30 <@kanzure> it makes conversations boring if every answer is "just use ai"
11:30 <@kanzure> or "just use friendly ai"
11:30 < cpopell> how 2 get girlfriend
11:31 < cpopell> >AI
11:31 < xmj> "The best move is not to play"
11:31 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has quit [Remote host closed the connection]
11:32 < xmj> i'm still unsure about which other forms you'd prefer
11:32 < nmz787_i> xmj:  he's not an idiot, more of an autistic/aspergers person in the sense of empathetic disconnect, but also isn't shy and doesn't back down into his own mental hole... so he kinda just starts arguing... you kind of have to 'wade' as cpopell put it
11:32 < FourFire> I was trying to explore the possibilities of reducing the bar to entry for language processing, thus showing that potential AI applications of knowledge processor eiaser
11:32 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has joined ##hplusroadmap
11:32 < xmj> nmz787_i: that makes two of us, awesome.
11:32 < xmj> i see language as symbols to manipulate
11:32 < FourFire> nmz787, kanzure, xmj I sort of resent you guys talking as if I wasn't right here
11:33 <@kanzure> that was only me
11:33 < xmj> and to my mind it doesn't matter if these symbols are in english, german, math + greek symbols, lambda expressions ((lisp)), forth python or c.
11:33 <@kanzure> english is not a symbol
11:33 < xmj> LEARN TO FUCKING READ
11:33 < xmj> jesus christ
11:34 < nmz787_i> FourFire: ?
11:34 < nmz787_i> FourFire: I was referring to kanzure :P
11:34 < xmj> lol
11:34 < xmj> nmz787_i: you're evil.
11:35 < nmz787_i> being mentally different is good though, it aides innovation
11:35 < nmz787_i> it can be hard though
11:35 < nmz787_i> I still don't know what's better than text though
11:36 < xmj> depends for what
11:36 < FourFire> so, my point is that you have to start somewhere
11:36 < xmj> good luck restating formulas in plain english
11:36 < FourFire> the closest I've seen to machine comprehension from human input is something about a computer being able to play some OSS clone of civilization
11:37 < FourFire> and the bot got better at playing the game after reading the instructions
11:37 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has quit [Ping timeout: 258 seconds]
11:37 < FourFire> >30% better, or something
11:37 <@kanzure> why did this conversation turn into anything near "machine comprehension from human input"
11:37 < ParahSailin> attempto english works
11:37 < chris_99> have you guys heard of linguistic relativity, re. language?
11:37 < FourFire> kanzure, I asked you what you found to be suboptimal about scientific papers
11:38 <@kanzure> yes, and you took the opposite of my answer
11:38 < FourFire> I'm sort of for text based information, because of high information density per amount of data
11:40 < nmz787_i> really the only 'better' I can think of would be some sort of memory/experience transfer
11:40 <@kanzure> there are alternative ways of recording scientific information without invoking "machine comprehension"
11:40 < nmz787_i> which text, movies, audio are all fragments of
11:40 < FourFire> "xml is the defacto standard of machine parsability" I'm saying we make a new standard (snrk) which is layman readable, and maybe even layman writable
11:40 <@kanzure> that's [Bfour	that's still text though
11:40 <@kanzure> blah
11:40 <@kanzure> FourFire: that's still text though
11:40 < nmz787_i> but who wants to re-live a whole 5 years of trial and tribulation for some science discovery... some people only want the abstract
11:41 <@kanzure> nmz787_i: suppose that everyone had the same lab equipment api
11:41 < nmz787_i> and then human memories might not be machine usable until we get brain uploading to work
11:41 <@kanzure> nmz787_i: they would then have "software" for orchestrating the api
11:41 <@kanzure> nmz787_i: and publish the piles of software instead, which is not exactly a "pile of text" per-se
11:41 < FourFire> then scientists, might be able to write their papers in "machine-english" and, if it's possible and done correctly, computers might *understand* the content
11:41 < FourFire> and be able to do things with it
11:41 < FourFire> you people know what's disgusting?
11:41 < ParahSailin> machine english exists already and there are validating parsers for it
11:42 <@kanzure> what does it mean to "understand".. that's a hugely vague word.
11:42 <@kanzure> how about instead of arbitrary things like understand, you pick exact targets like, repeatability
11:42 <@kanzure> the majority of molecular biology is a dice toss
11:42 < FourFire> kanzure, ok, what format would you prefer?
11:42 < FourFire> sound?
11:42 < FourFire> *olfactory*
11:43 < FourFire> *tactile patterns*
11:43 <@kanzure> data can be made that isn't meant for reading/writing like books or novels
11:43 < cpopell> how do you transmit its content to brains?
11:43 < FourFire> so, visual
11:43 <@kanzure> do you really need to transmit the content to brains?
11:43 <@kanzure> you just need to be able to transmit it to other labs and people doing lab stuff
11:43 < cpopell> people doing lab stuff are brains.
11:44 <@kanzure> does your brain really know each byte of your operating system etc?
11:44 < FourFire> it has to be translatable into some form of human sense which is understandable without some odd specilized form of training
11:44 <@kanzure> lab training is already pretty specialized
11:44 < FourFire> reading/writing is pretty much standard for most people these days
11:44 <@kanzure> and odd
11:44 <@kanzure> reading/writing is not actually enough for most molecular biology training, as an example
11:44 < FourFire> "do you really need to transmit the content to brains?"
11:44 < FourFire> yes
11:44 <@kanzure> actually, ParahSailin might disagree with my last statement, although i'm not sure
11:44 < FourFire> unless you are magically going to offload all research onto machines
11:45 < ParahSailin> "and then you bang the flask on the bench like so"
11:45 < FourFire> kanzure, you neglected to make a proposal, which was superior to papers, so I made mine, which is superior in that it is (possibly) machine readable
11:45 < ParahSailin> "ding, trypsinized"
11:46 <@kanzure> "it's somewhere between a one-quarter inch and one-half inch punch"
11:46 -!- fireprfHydra [~fireprfHy@pool-173-70-216-225.nwrknj.fios.verizon.net] has joined ##hplusroadmap
11:46 <@kanzure> "make sure it swirls to the left, unless you're below the equator"
11:46 <@kanzure> ParahSailin: i think a lot of it is going to have to be thrown out
11:47 <@kanzure> ParahSailin: it's just useless
11:48 < FourFire> kanzure, by *understand* I mean, to be able to take bytes of data and then mix them in a similar way that brains work with concepts
11:48 < ParahSailin> if we throw out these die rolls, then how will we mulligan experiments until we get the desired result?
11:48 < cpopell> http://www.news.wisc.edu/22756 Fuuuuuu
11:48 < FourFire> like parsing a linguistic logic sentence
11:49 < delinquentme> Do you guys know if theres a python codebase or an API  to take a firefox screenshot of a website and return either an image or static URL?
11:49 < ParahSailin> phantomjs
11:49 <@kanzure> delinquentme: just use testling or browserling if you want something quick. if you want something that sucks you can use firefox+selenium.
11:49 <@kanzure> ParahSailin: phantomjs is not gecko/firefox unfortunately
11:49 < FourFire> oh great, I hope they don't do retarded things and actually uncover stuff like male and female brains being structurally different, or whatever it is
11:50 < ParahSailin> kanzure: i was reading into him being imprecise
11:50 <@kanzure> oh yeah i should have done that
11:50 <@kanzure> given historical evidence
11:50 < delinquentme> kanzure, specifically I need image capturing
11:50 <@kanzure> phantomjs
11:51 < FourFire> "reading/writing is not actually enough for most molecular biology training, as an example" I agree, but not all papers are molecular biology
11:51 < FourFire> maybe my idea is too ambitious
11:51 < FourFire> at best, it could lower the bar to a slightly less educated superset of people
11:51 <@kanzure> "your" idea is lacking an implementation despite thousands of people trying
11:52 < delinquentme> and phantom.js can use different browsers
11:52 <@kanzure> phantomjs is only webkit
11:52 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has joined ##hplusroadmap
11:53 <@kanzure> there's more than just moelcular biology that suffers from similar issues
11:53 <@kanzure> *suffer
11:54 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has quit [Remote host closed the connection]
11:54 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has joined ##hplusroadmap
11:56 < FourFire> I wonder if/where there are internet tests which allow you to check your proficiency, or literacy inside scientific fields
11:57 < FourFire> I, personally am not educated in anything besides basic physics, secondary biology and secondary chemistry
11:59 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has quit [Ping timeout: 276 seconds]
12:00 < FourFire> But, I feel like (confirmation bias) I'm more of a scientific oriented person in general, and there is some sort of (aura effect) knock on knowledge which allows me to understand more of scientific papers, in other areas
12:00 < FourFire> more, than what I assume (typical mind, projection fallacy) other people do.
12:02 -!- nsh [~nsh@wikipedia/nsh] has quit [Ping timeout: 245 seconds]
12:02 < nmz787_i> delinquentme: maybe printscreening a headless instance of chromium via https://code.google.com/p/cefpython/wiki/VirtualKey
12:02 < nmz787_i> delinquentme: cefpython.VK_SNAPSHOT # Print Screen key 
12:03 <@kanzure> if he wants multiple browsers then his only option is selenium or a service like browserling
12:03 < nmz787_i> ahh
12:06 < dingo> delinquentme: if you have osx, and safari is an option, i once helped someone write python code that uses the cocoa api to simply launch safari at a url and grab a png of the rendered page
12:07 <@kanzure> phantomjs works on osx and has the advantage of also being webkit (like safari is using)
12:07 < dingo> not that i have the code, but, its possible and was pretty easy at the time
12:07 < dingo> (and no actual safar window would be visible, headless, as you ask)
12:07 <@kanzure> oh hm
12:07 < xentrac> FourFire: if your level of proficiency is unknown to you, then how will you know which tests to trust?
12:08 -!- cluckj [~cluckj@cpe-24-92-63-104.nycap.res.rr.com] has quit [Ping timeout: 240 seconds]
12:08 <@kanzure> :)
12:08 < xentrac> e.g. the hacker purity test which asks, among other things, "can you build a puffer train?" is pretty awesome in some ways
12:08 < xentrac> but it's also outdated
12:09 <@kanzure> hrm i imagine that's mostly a question about whether you (1) can forge and smelt and (2) remember basic train engine structure
12:09 <@kanzure> oh. cellular automata. that's disappointing.
12:09 < xentrac> haha
12:09 < xentrac> no, it's AWESOME
12:10 < FourFire> xentrac, it's still very much a barrier to entry which cuts off everyone who can't
12:10 < xentrac> but imagine how puzzled I was about your response until I figured that you were talking to me
12:10 < nmz787_i> steam engines came to my mind when i saw that
12:11 <@kanzure> yeah, why should cellular automata be more hacker-oriented than backyarding a steam engine?
12:11 < delinquentme> you guys rock :D
12:14 < delinquentme> looks like Im getting ot use phantom.js :D
12:14 < delinquentme> http://stackoverflow.com/questions/12450868/how-to-print-html-source-to-console-with-phantomjs
12:14 < delinquentme> that little bit of code looks way too simple ... but im gonna try this out
12:15 -!- ruthie [~ruthie@dhcp-130-58-194-12.swarthmore.edu] has joined ##hplusroadmap
12:16 < nmz787_i> ha, i like the quote on the phantomjs author's website 'don't code today what you can't debug tomorrow'
12:16 < delinquentme> HAAA fuck im so happy. kanzure  +1
12:16 -!- ruthie [~ruthie@dhcp-130-58-194-12.swarthmore.edu] has quit [Remote host closed the connection]
12:21 -!- ruthie [~ruthie@dhcp-130-58-194-12.swarthmore.edu] has joined ##hplusroadmap
12:22 -!- ruthie [~ruthie@dhcp-130-58-194-12.swarthmore.edu] has quit [Remote host closed the connection]
12:22 -!- fireprfHydra [~fireprfHy@pool-173-70-216-225.nwrknj.fios.verizon.net] has quit [Quit: Leaving]
12:26 < cpopell> anyone know of a decent rss reader, paid or free, that has word search with tagging?
12:33 -!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap
12:42 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has joined ##hplusroadmap
12:45 -!- kuldeepdhaka [~kuldeepdh@unaffiliated/kuldeepdhaka] has quit [Remote host closed the connection]
12:47 < xentrac> kanzure: that's a very interesting question: why are CAs more hackish than steam engines?
12:47 < xentrac> and I don't have a good answer
12:48 < xentrac> historical contingency? economic requirements? manual skills? risk of death?
12:49 < xentrac> the difficulty of putting the steam engine on your ftp site or posting it to Usenet?
12:50 < cpopell> lack of lifting capabilities required
12:50 < cpopell> :P
12:50 -!- jamesz [~ielo@c-69-255-193-198.hsd1.md.comcast.net] has quit [Ping timeout: 258 seconds]
12:50 < xentrac> well, that's usually resolved with chain lifts anyway
12:51 < xentrac> it could also be that steam engines are too simple (in the sense of having few parts) to be interesting to hackers who have the alternative of computers available
12:52 < cpopell> so, poked at SAP Lumira
12:52 < cpopell> Someone saw the money in Tableau
12:52 < xentrac> hmm, that sounded overly dismissive, which I didn't mean
12:52 -!- HashNuke [uid12117@gateway/web/irccloud.com/x-ivjfzfqxytqfxlww] has quit [Quit: Connection closed for inactivity]
12:52 < xentrac> some hackers are certainly gearheads
12:52 < xentrac> but not because they are hackers
12:52 < cpopell> yeah, but as a percentage
12:55 -!- marciogm [~marciogm@179.126.74.112] has quit [Quit: Computer has gone to sleep.]
13:02 <@kanzure> xentrac: what do you mean "but not because they are hackers"? for context, i would definitely say my interest in mechanical things is because of hacking, and not the other way around
13:03 <@kanzure> xentrac: meanwhile, finding any machinist that knows anything about computing is pretty difficult
13:04 < heath> hm, serf or consul
13:04 <@kanzure> serf
13:23 -!- nsh [~nsh@wikipedia/nsh] has quit [Ping timeout: 265 seconds]
13:24 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving]
13:44 -!- Adifex|zzz is now known as Adifex
14:02 -!- nmz787_i [~nmccorkx@192.55.54.38] has quit [Quit: Leaving.]
14:11 < xentrac> kanzure: I think most hackers are interested in mechanical things for hackish reasons
14:12 <@kanzure> sure
14:12 < xentrac> but unless you're wealthy, that has historically not been enough to get you into actually doing things
14:12 < xentrac> now that we have RepRaps, that is starting to change
14:12 <@kanzure> i had always ussed that it was because the idea just didn't occur to them, not because of a lack of wealth
14:12 <@kanzure> things were cheap to do even before reprap, though
14:12 <@kanzure> *always assumed
14:13 -!- Lemminkainen [uid2346@gateway/web/irccloud.com/x-uyunzbpvweepekpc] has joined ##hplusroadmap
14:13 < xentrac> well, I think the mental orientation needed to successfully rebuild a Volkswagen engine, for example, is pretty different from the mental orientation needed to successfully write a compiler
14:13 <@kanzure> i mean, it's not like hackers everywhere are sitting around thinking about mechanical hackery
14:13 < xentrac> although there are some things in common
14:13 < Lemminkainen> hie thee to an endmill, kanzure
14:14 < xentrac> while the mental orientation needed to build a puffer train is pretty similar to what's needed for a compiler
14:14 <@kanzure> hm! well it's certainly true that a volkswagen engine isn't "merely juggling a bunch of competing grammar rules and linkerstuff"
14:15 <@kanzure> although there's been lots of funny broken attempts at hardware non-electronic-related vhdl/vlsi stuff
14:16 < xentrac> yeah
14:16 < xentrac> I think for that to happen we are going to need someone who's simultaneously a good machinist and a good hacker
14:17 < xentrac> and a good mechanical engineer, I think
14:17 <@kanzure> that's basically fenn, except i stole all his ideas and claimed them as my own
14:17 <@kanzure> http://fennetic.net/
14:17 < xentrac> what happened to him?
14:17 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has quit [Remote host closed the connection]
14:17 <@kanzure> well he spent a few years in the desert in what i would call the "burning man is never gonna end, man" phase
14:18 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has joined ##hplusroadmap
14:18 < xentrac> like 2012 to 2014?
14:18 <@kanzure> yes?
14:18 < xentrac> ah
14:18 < xentrac> burning man is pretty awesome
14:19 < xentrac> I think it's contributed a lot to getting hackers into making machinery
14:19 < Lemminkainen> I've dreamt of putting metal lathes on the street corners of SF
14:20 < Lemminkainen> it would be a self-selecting skill process and might get more people into it
14:20 < xentrac> <Lemminkainen> shivs for all!
14:20 < Lemminkainen> CLASSY shivs
14:20 < Lemminkainen> with beveled edges and flat design
14:21 < xentrac> one thing about metal lathes
14:21 < xentrac> compilers don't CRUNCH when you misuse them
14:21 < Lemminkainen> or rip your knuckles to shreds
14:22 < Lemminkainen> maybe hot-glue guns and glitter is a better start for such wayward brogrammers
14:23 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has quit [Ping timeout: 258 seconds]
14:24 <@kanzure> xentrac: he might be around in a bit, i established phone contact
14:34 -!- fenn [~fenn@131.252.130.248] has joined ##hplusroadmap
14:35 < fenn> huzzah
14:35 <@kanzure> xentrac: compilers can have equally damaging security issues though
14:35 < fenn> oh it's kragen sitaker
14:35 <@kanzure> known anomaly?
14:35 < fenn> xentrac: i've been playing around with your byn script (wikipedia microprinting)
14:37 < fenn> xentrac: http://fennetic.net/irc/combat_droid_byn_500w_2c_step1.gif
14:38 < fenn> sorry i havent been around ##hplusroadmap, i've been dead for approximately one of your earth years
14:40 < fenn> i guess a library needed for irssi was updated and irssi hung for some number of months
14:43 -!- nmz787_i [~nmccorkx@134.134.139.76] has joined ##hplusroadmap
14:45 < fenn> kanzure: something you might find interesting, "ranger" a ncurses file manager
14:46 <@kanzure> looks okay
14:47 <@kanzure> http://ranger.nongnu.org/screenshots/screenshot0.png
14:47 < fenn> i never figured out how to get bitmap images in the console though (or ascii art images either for that matter)
14:48 -!- Jaakko910 [~Jaakko@cpc13-newc15-2-0-cust64.16-2.cable.virginm.net] has joined ##hplusroadmap
14:49 <@kanzure> substack did something called picture-tube
14:49 <@kanzure> https://github.com/substack/picture-tube (ascii art)
14:50 < dingo> http://1984.ws/rivermeadow.png
14:50 < dingo> my next resume will be ansi art
14:50 -!- ruthie [~ruthie@dhcp-130-58-194-12.swarthmore.edu] has joined ##hplusroadmap
14:50 <@kanzure> dingo: how bored are you
14:50 -!- Jaakko910 [~Jaakko@cpc13-newc15-2-0-cust64.16-2.cable.virginm.net] has quit [Client Quit]
14:50 < dingo> not bored, burned out
14:51 <@kanzure> didn't take long :/
14:51 <@kanzure> kinda impressive really
14:51 < dingo> that picturetube didn't uhh have any code, but it looks like it was made using libcaca, http://caca.zoy.org/wiki/libcaca
14:51 <@kanzure> oh.
14:54 -!- ruthie [~ruthie@dhcp-130-58-194-12.swarthmore.edu] has quit [Ping timeout: 240 seconds]
14:57 -!- nsh [~nsh@host217-43-194-64.range217-43.btcentralplus.com] has joined ##hplusroadmap
14:57 -!- nsh [~nsh@host217-43-194-64.range217-43.btcentralplus.com] has quit [Changing host]
14:57 -!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap
14:57 < fenn> aaview is better than libcaca as far as image quality is concerned, though frankly any ascii art package should do minimal edge detection and use the characters as something like DCT blocks, instead of simple average brightness
14:59 < fenn> er, asciiview, or aview (where did aaview come from?)
15:00 < fenn> OCR for properly displaying text is pretty important too.. all this assumes anyone actually cares about ascii as a display method, instead of the dancing bear factor
15:01 < dingo> http://nyancat.dakko.us/
15:03 < fenn> does anyone know why old monitors were amber colored?
15:03 < dingo> why? it was just nicer than green
15:03 < dingo> i had an amber wyse terminal
15:04 -!- eudoxia [~eudoxia@r186-53-101-205.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap
15:04 < dingo> white, green, or amber, that was basically the choice
15:04 < dingo> i liked amber best
15:04 < fenn> I agree wholeheartedly, but there must be some scientific reason so many things have converged on this particular shade of yellow-orange
15:04 < eudoxia> i was starting to wonder where fenn had gone
15:04 < fenn> hello eudoxia, i was dead
15:05 < eudoxia> O:
15:05 < fenn> fortunately the cryonics society was able to revive me, since we're in the future now
15:05 < eudoxia> also, #AD3108 orange is best orange
15:06 < eudoxia> fenn: don't you mean the cryonics institute, unless the ACS became some sort of cryonics underground while I wasn't paying attention?
15:06 < fenn> shhh
15:07 < fenn> reviving the dead isn't yet legal, don't go spreading it around
15:07 < fenn> i've been using #ca0 and #da0 lately
15:07 < fenn> (for html links)
15:09 < eudoxia> how do you even perceive the difference
15:09 < fenn> i guess that's #CCAA00 and #DDAA00, not #C0A000/#D0A000
15:10 -!- AshleyWaffle_ [~waffle@gateway/tor-sasl/anastasiawyatt] has joined ##hplusroadmap
15:10 < fenn> you can tell when there are some links on a page and one looks darker
15:10 < dingo> fenn: somebody did the world cup over telnet, using ascii art, and used closed captioning to provide naturally ascii subtitles
15:10 -!- AshleyWaffle [~waffle@gateway/tor-sasl/anastasiawyatt] has quit [Disconnected by services]
15:11 -!- AshleyWaffle_ is now known as AshleyWaffle
15:11 < eudoxia> what is it with people and ASCII art
15:11 < dingo> http://ascii-wm.net/
15:11 < fenn> it's like pixel-art for old fogeys
15:12 < fenn> just wait, once the majority of displays are OLED we'll have persistence of vision art
15:12 < fenn> just a single line
15:14 < fenn> moving a color picker over its own output is interesting
15:14 < xentrac> fenn: oh cool! I'm pleasantly surprised that you've found it useful
15:15 < da_shirlz_HBIC_> paperbot http://www.sciencemag.org/content/335/6070/831.short
15:15 < paperbot> http://libgen.org/scimag/get.php?doi=10.1126%2Fscience.1214081
15:15 < xentrac> Lemminkainen: I think maybe the biggest difference is that you can't revert an unwanted change to your engine
15:15 < fenn> xentrac: i want to make a downloadable wikipedia app for e-ink readers like nook. it's a big project though
15:15 < da_shirlz_HBIC_> paperbot http://www.sciencemag.org/content/335/6070/831.short
15:15 < paperbot> http://libgen.org/scimag/get.php?doi=10.1126%2Fscience.1214081
15:16 < xentrac> fenn: that combat droid looks fantastic
15:16 < fenn> the model was made by long0800
15:17 < fenn> http://cargocollective.com/long0800
15:17 < cpopell> ugh
15:17 < xentrac> as for amber, I assume that of the couple of dozen phosphors available, that was one of the few with a persistence in the optimal range
15:17 < cpopell> when friends say 'I think I have an idea on how to make AI'
15:17 < cpopell> Sigh.
15:18 < cpopell> 'first we need to write a new natural language system'
15:18 < fenn> some day i'll be able to paint like that
15:18 < xentrac> yes, a downloadable wikipedia for e-ink readers would be super awesome
15:18 < fenn> cpopell: oh you mean loglan, lojban, calxism, etc
15:19 < eudoxia> 'i think i can make an AI' is like the nerd version of those stoner theory-of-everything manifestos
15:19 < cpopell> new people in here-anyone know of an RSS with robust auto-tagging features?
15:19 < cpopell> *RSS reader
15:19 <@kanzure> calxism language was not particularly instrumental to the core concept, but nice try
15:20 < fenn> why do people use RSS? how can you possibly not have enough stuff to read already?
15:20 < cpopell> fenn: cataloguing and storing sci/tech developments I find interesting
15:21 < xentrac> fenn: once you have enough stuff to read, the next step is to find ways to avoid reading it
15:21 < fenn> but not everything is in RSS so then you need a subsuming paradigm to merge your RSS and PDF or whatever
15:21 < eudoxia> wait, what is this calxism thing
15:21 <@kanzure> eudoxia: i was indoctrinated into an internet cult at an early age
15:21 < cpopell> fenn: er, this is for short news articles, not long form reading
15:22 < cpopell> some interesting results when I google calxism
15:22 < fenn> eudoxia: leibniz said "one day we will stop arguing, instead we'll sit down with our calculating devices and calculate the logical resolution to any problem"
15:22 < eudoxia> kanzure: i saw a post by you that reads remarkably less wrongy
15:22 <@kanzure> eudoxia: i regret everything
15:22 < cpopell>  (6) What if the hokey pokey really is what it's all about?
15:22 -!- EnLilaSko [EnLilaSko@unaffiliated/enlilasko] has quit [Remote host closed the connection]
15:23 < eudoxia> it happens to the best of us kanz
15:23 < eudoxia> probably
15:24 <@kanzure> yes i'm lucky i didn't sit down writing utility functions or something
15:24  * cpopell wrote some dumb shit when he was 15
15:24 < fenn> not me, all my english class essays were top notch
15:24 < cpopell> oh, sure, my english essays were fine
15:24 < eudoxia> sometimes i-i secretly hope the internet archive burns down
15:25 < eudoxia> just a few fileservers :?
15:25 < xentrac> eudoxia: we'd been wondering who set that fire
15:25 <@kanzure> it did burn down
15:25  * xentrac dispatches the hounds
15:25 < xentrac> (hint: traceroute my IP)
15:25 < fenn> we'll just ask the NSA for a copy
15:25 < FourFire> " 'i think i can make an AI' is like the nerd version of those stoner theory-of-everything manifestos" nice quote, I'
15:25 < FourFire> 'm going to use it next time one such turns up
15:26 < fenn> FourFire: are you rastafarian?
15:26 < xentrac> fenn: in a sense my ideal for the internet is kind of like a better version of the public library
15:26 < eudoxia> wow it did actually burn down once
15:26 < fenn> xentrac: naw that's missing out on 99% of the functionality
15:26 < FourFire> "i saw a post by you that reads remarkably less wrongy" eudoxia linky?
15:27 < eudoxia> but I wasn't in the US at the time (or ever) so it wasn't some split personality
15:27 < fenn> xentrac: your nook should be the public library tho
15:27 < xentrac> right
15:27 < eudoxia> FourFire: https://www.livebusinesschat.com/smf/index.php?topic=61.0
15:27 < xentrac> fenn: specifically when thinking about information overload though
15:27 < FourFire> fenn, I'm not sure, so that's a no
15:27 < xentrac> there's far more text in a good public library than you'll ever read
15:27 < fenn> there's far more on my nook than i'll ever read too
15:27 <@kanzure> eudoxia: your stalkpower is increasing
15:28 <@kanzure> i don't think you should be expected to read the entire public library
15:28 < fenn> actually, there's probably more on my nook than in a public library
15:28 < xentrac> but you don't have the urgency to read it
15:28 <@kanzure> piles of text is not mechanically useful to me
15:28 <@kanzure> it's like asking me to write or rewrite all of debian's packages
15:28 < xentrac> I've often thought about doing that
15:28 <@kanzure> there should be some functional form to our civilization content
15:28 <@kanzure> and not just dna that folds into shitty proteins
15:28 < eudoxia> kanzure: that was on the first page of google
15:28 < dingo> i've noticed public libraries becoming more media-oriented -- audio cd's, movies, etc... its sad
15:29 < eudoxia> i think my best internet stalking was when me and kirka tracked down that guy, uh, markus krunnemacker i think
15:29 <@kanzure> well, media should be archived too
15:29 < cpopell> An old friend taught me how to internet-stalk.
15:29 < cpopell> he made me find his house number
15:30 <@kanzure> 1) discover the heartbleed vulnerability 2) read the memory of whatever server you think has the data
15:30 < Lemminkainen> 3) have apple pie
15:30 < FourFire> ideology and cult in the same post...
15:31 <@kanzure> FourFire: i'm over it, don't worry
15:31 < FourFire> yeah, taking into consideration that it's been nearly a decade since you wrote that
15:31 < fenn> cult is just short for culture
15:31 <@kanzure> that doesn't mean culture is good
15:31 < cpopell> fenn: cult is short for cultus
15:31 < cpopell> or 'worship'
15:32 < eudoxia> "culture is for bootlickers im so libertarian i don't even talk to ppl"
15:32 <@kanzure> and apparently use abbreviations
15:32 < FourFire> "will you shoot me?"
15:33 <@kanzure> FourFire: i don't recommend reading into it
15:33 < FourFire> :D
15:33 < Lemminkainen> http://i.imgur.com/fEqegwZ.gif
15:35 < eudoxia> 'oil rig ROV camera footage' should be its own genre of horror
15:35 < Lemminkainen> coming soon to Oculus Rift
15:35 <@kanzure> because life is horrible?
15:35 < eudoxia> kanzure: i hope your stalking log keeps some sort of record of my hilarious one liners
15:36 < Lemminkainen> life isn't so much horrible as horribly perverted and persistent
15:36 <@kanzure> i think that log is the sort of document that gets kids in school locked up for admitting to
15:42 <@kanzure> fenn: it's inconvenient for a large civilization to have everything trapped in text
15:42 < Lemminkainen> what's your alternative, kanzure?
15:42 < eudoxia> YAML files
15:42 <@kanzure> a billion yaml files
15:42 < Lemminkainen> media isn't asychronous
15:43 < Lemminkainen> I don't think a billion would be enough
15:43 <@kanzure> neither is text
15:43 < Lemminkainen> I consume text asychronously
15:44 < xentrac> to this discussion of "media" I would like to point out that paper and ink are media.
15:44 <@kanzure> enki1 certainly did- he was experimenting with 4 to 5 simultaneous audio streams (mostly fiction)
15:45 < eudoxia> reminds me of that guy who used a text-to-speech thing real fast
15:45 <@kanzure> the best kind of knowledge is the sort that handles itself
15:46 <@kanzure> like compiler infrastructure
15:47 < eudoxia> how do compilers handle themselves
15:47 < eudoxia> because they compile themselves?
15:47 < eudoxia> (sometimes)
15:47 < fenn> why is that even a question? it's just a program like any other program
15:49 < fenn> something something godel completeness theorem something
15:49 <@kanzure> strange thing is that i think it's still actually possible to read all wikipedia articles
15:49 <@kanzure> only 5M right?
15:49 < fenn> but what order do you read them? they will be different by the time you are done downloading them all
15:50 < Lemminkainen> does reading them have to be that high fidelity?
15:50 < fenn> (clicking "random wikipedia page" was one of the most depressing things I've ever done, there's so much useless crap that will never matter)
15:51 < Lemminkainen> if you have some mutation in the text as you go, then I see no problem with it
15:51 < Lemminkainen> <some small level of background mutation that could be measured>
15:52 < fenn> check it out, useless knowledge: http://en.wikipedia.org/wiki/Triathlon_at_the_2006_Asian_Games_%E2%80%93_Men%27s_individual
15:52 < fenn> why do the deletionists even bother i wonder
15:53 < eudoxia> apparently it's called "random article" now
15:53 < eudoxia> two-sentence articles that are just an infobox about some middle-of-bumfuck-nowhere town should probably be in a separate site
15:53 <@kanzure> the data arbitration sentience entity disagrees.
15:54 < fenn> i think many of those location pages were moved to toolserver.org
15:54 < eudoxia> also the endless, endless collection of articles about sport events
15:54 < eudoxia> millions of them
15:54 <@kanzure> crap i'm starting to cross nagaru tanigawa and vernor vinge
15:55 < fenn> (arbitration?)
15:55 < eudoxia> next up kanzure's novel: The deep fire upon Haruhi Suzumiya
15:56 < fenn> a brony fan fic
15:56 < fenn> exploring the methods of rationality
15:56 <@kanzure> fenn: http://lesswrong.com/lw/jgz/aalwa_ask_any_lesswronger_anything/ap84 "I do have some early role models. I recall wanting to be a real-life version of the fictional "Sandor Arbitration Intelligence at the Zoo" (from Vernor Vinge's novel A Fire Upon the Deep) who in the story is known for consistently writing the clearest and most insightful posts on the Net. And then there was Hal Finney who probably came closest to an actual real-life ...
15:56 <@kanzure> ... version of Sandor at the Zoo, and Tim May who besides inspiring me with his vision of cryptoanarchy was also a role model for doing early retirement from the tech industry and working on his own interests/causes."
15:56 <@kanzure> fenn: from why-wei-dei-is-probably-satoshi-nakamoto.txt
15:58 < eudoxia> guys maybe satoshi kon is satoshi nakamoto
15:58 < eudoxia> (1) he's japanese (2) he's named satoshi (3) he disappeared around the time satoshi disappeared
15:58 < eudoxia> (4) he's into anime so he's probably into computers (reverse causality?)
15:58 < Lemminkainen> impeccable logic
15:59 < Lemminkainen> I hereby peer review it and approve it for submission to Bloomberg
15:59  * xentrac peccs the logic
15:59 < fenn> the only net personality i remember is "twirlip of the mists" http://web.hexapodia.org/
15:59 <@kanzure> that sounds like an awful nethack name
15:59 < eudoxia> we're through the looking glass here people
15:59 <@kanzure> actually i think that one is from the book too
16:00 < xentrac> Hexapodia as the key insight
16:00 < Lemminkainen> welcome home eudoxia
16:00 < fenn> xentrac: you have to admit, it explains quite a few things!
16:00 < fenn> for example the fondness for hexagons
16:01 < eudoxia> another data point: he wrote part of katsuhiro otomo's Memories. another episode of that anthology is about a future dystopian perpetual-war society
16:01 < eudoxia> the facts just keep piling up
16:01 < xentrac> they do.  to the point that libraries have to start using microfilm and CD-ROM.
16:02 < xentrac> what's the cheapest way to make a single copy of an N-bit string that could plausibly survive a millennium?
16:02 < xentrac> Laser printers are my current best.
16:02 < eudoxia> xentrac: carve it into a big big stone
16:02 < xentrac> how cheaply can you carve stone?
16:02 < eudoxia> *shrug* i hear those sand blasters are pretty good
16:03 < xentrac> sand blasters?
16:03 <@kanzure> atomic force microscopes are surprisingly easy and cheap to build [depending on a personal preference for tolerances and accuracy]
16:03 < fenn> there is even a major organized religion "hexagonalism" http://hexnet.org/about
16:03 < fenn> xentrac: DNA
16:03 < eudoxia> kanzure: alternating atoms on a surface to encode information can't be too long lived
16:03 <@kanzure> DNA doesn't actually survive that long
16:03 < fenn> a millennium is nothing
16:03 <@kanzure> the half-life is only a few ten thousand years
16:03 < eudoxia> a few stray cosmic rays and poof
16:04 < xentrac> fenn: how cheap is DNA?
16:04 < fenn> kanzure: not naked DNA, in a thing
16:04 < fenn> uh, a living thing
16:04 <@kanzure> oh right, 1000 years
16:04 <@kanzure> yes dna can plausibly survive 1000 years
16:04 < xentrac> a 1200dpi laser printer can print an A4 page for about 6 cents, which is a sixteenth of a square meter. that's about 2.2 gigabits per US$
16:04 < xentrac> can you order 1.1 gigabases of novel DNA for US$1?
16:05 < fenn> if it's the right sequence, yes
16:05 < xentrac> an acid-free A4 page.  Otherwise it costs 3 cents but will only last a few decades at best
16:05 < xentrac> the right sequence?
16:05 < fenn> "you can get any color you like, as long as it's human-colored"
16:05 < xentrac> human DNA isn't novel
16:06 < fenn> what! every new human has a different sequence than all the others
16:06 < xentrac> sure, you have maybe 25 to 100 bits of novelty
16:06 <@kanzure> moving around gold or carbon on a surface would be cheaper than de novo dna synthesis at this point
16:06 < Lemminkainen> put a retrotransposon into the germline near an important metabolic gene
16:06 < xentrac> how cheap is moving around gold or carbon on a surface?
16:06 < Lemminkainen> let the information encoded therein descend through time through procreation
16:06 <@kanzure> well you can build an afm in parts for <$1k
16:06 < fenn> xentrac: what is "cheap"
16:07 < xentrac> 2.2 gigabytes per US$, as I said
16:07 < fenn> no, i mean the definition
16:07 < xentrac> gigabits, sorry
16:07 < xentrac> cheap: costing little money
16:07 < eudoxia> hm for that cost you'd probably need an AFM with many tips
16:08 < xentrac> kanzure: how high is the bit rate of the AFM?
16:08 < xentrac> and how long does it last before it breaks?
16:08 < fenn> so, i have in my pocket a $20 supercomputer with high resolution color display, multi band network radio, sensors out the wazoo
16:08 < fenn> why is it "cheap"?
16:08 < xentrac> fenn: mostly because of mass production
16:09 < eudoxia> xentrac: http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.106.136101
16:09 <@kanzure> paperbot: http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.106.136101
16:09 < paperbot> http://diyhpl.us/~bryan/papers2/paperbot/475108cbb2f5ec2d30bd470770be3dde.txt
16:09 <@kanzure> :(
16:09 < xentrac> fenn: I explored using the processes used to produce your pocket supercomputer for archiving information
16:09 < eudoxia> here it is http://nanomanipulation.ut.ee/abs/Sweetman_oral.pdf
16:10 < fenn> sure nmz787 was playing around with FIB'ing the bible on the head of a pin, or something
16:10 <@kanzure> photolithography?
16:10 < xentrac> yeah
16:10 < FourFire> wei dei satoshi?
16:10 < FourFire> lel
16:10 < xentrac> it turns out that MOSIS is a couple of orders of magnitude more expensive than a laser printer
16:10 <@kanzure> i guess etchants are easier to work with than afm calibration
16:10 < fenn> you know what's in a FIB? mostly nothing
16:10 < fenn> building huge arrays of them in orbit would be relatively straightforward (it's getting to orbit that's the problem)
16:10 < xentrac> eudoxia: that's just the abstract
16:11 < eudoxia> oh top lel
16:11 < xentrac> FIB?
16:11 < eudoxia> focused ion beam something
16:11 < xentrac> ah, that makes sense
16:11 < xentrac> either that or an AFM might be a better approach than using a laser printer
16:12 < xentrac> but I don't have a good sense of how much they cost in the end
16:12 < fenn> what's the purpose of this endeavor? don't CD-roms last long enough in reasonable storage conditions?
16:12 < eudoxia> some dude made an AFM for 100 USD
16:13 < eudoxia> i don't think it was sub-micron
16:13 < fenn> the thing is you need a computer to read a CD-rom, whereas a human can read microprint with a paperclip and a drop of water
16:13 < xentrac> CD-Rs only last a few years, usually; archival-quality ones are supposed to last 300 years
16:13 -!- FourFire [~fourfire@108-214-15.connect.netcom.no] has left ##hplusroadmap ["Leaving"]
16:13 < fenn> right but nobody really knows
16:14 < fenn> DNA has been around for billions of years
16:14 < eudoxia> like those space shuttle screws that had impossibly-low-to-test failure rates
16:14 < xentrac> however, they aren't really that much cheaper: US$0.10 for a 700MB CD-R is only 56 gigabits per US$
16:14 < fenn> that's pretty cheap
16:15 < xentrac> which is better than 2.2 but that's at the cost of degrading in a few short years
16:15 < xentrac> and, yes, also needing a computer
16:15 < xentrac> I think that you could probably use AFMs or ion beams to print holograms that people could read with a magnifying glass
16:16 < xentrac> dependent on the light angle
16:16 < Lemminkainen> I would hide it in ant genomes
16:16 < eudoxia> that's a good plot for a sci fi novel
16:16 < xentrac> Lemminkainen: how much do genomes cost?  if the cost is higher than US$1 per 1.1 gigabases then laser printers win
16:16 < eudoxia> kind of like the message in pi from Contact
16:17 < fenn> there's a tree in yuma arizona that is about 10,000 years old
16:17 < xentrac> Jaron Lanier proposed using cockroaches: http://www.jaronlanier.com/roach.html
16:17 < fenn> they accidentally killed the oldest one by measuring its age
16:17 < xentrac> fuck
16:18 < xentrac> at the time it was US$0.60 per base pair, which compares very poorly with laser printers.  but that was 1999
16:18 < fenn> they didnt know it was the oldest one until they measured though, so it was kinda unavoidable
16:18 <@kanzure> dna synthesis and dna sequencing have different costs
16:18 <@kanzure> synthesis is still pretty expensive and nobody is really putting pressure on the price
16:18 < xentrac> this was synthesis
16:18 < fenn> you actually want to do gene synthesis </pedantry>
16:19 < Lemminkainen> xentrac I guess I care more about the format being clever than cheap
16:19 < xentrac> Lemminkainen: DNA is clearly very clever, and it is also quite replicable
16:19 < xentrac> DNA is probably the most cheaply replicable format we have
16:19 <@kanzure> under very specific circumstances
16:19 < fenn> xentrac: we've had hundreds of discussions around here about dna synthesis, so that's a known bias
16:20 <@kanzure> pcr master mix is not as cheap as it could be
16:20 -!- Baube [~Baube@65.95.14.15] has joined ##hplusroadmap
16:20 <@kanzure> unless you mean in vivo dna stuff
16:20 < Lemminkainen> aye, I do
16:20 < xentrac> yeah, in vivo
16:20 < xentrac> any estimate on bit rate of AFM or FIB recording?
16:20 < Lemminkainen> cost of 1 TALENs synthesis, 1 in vivo transfection
16:21 <@kanzure> if you get down to torr^-9 then you can do single atom manipulation stuff
16:21 <@kanzure> what are the dimensions of bit rate in this context? sorry
16:23 < xentrac> ideally bits per dollar, but I was thinking that bits per second would provide a floor for bits per dollar
16:23 < xentrac> a lower bound
16:23 < xentrac> since your AFM has a limited lifetime
16:24 < fenn> so how big is wikipedia in A4 sheet with tiny text?
16:24 < fenn> vs how many hard drives (how big is it now anyway?)
16:24 <@kanzure> one of the main factors to estimate is the cost per tungsten tip or the cost to build each tungsten tip (there's a do-it-yourself method but it requires some motor skills?)
16:25 < xentrac> I don't think the cost per tungsten tip is significant
16:25 < xentrac> fenn: https://en.wikipedia.org/wiki/Wikipedia:Size_of_Wikipedia has some estimates
16:26 < xentrac> https://en.wikipedia.org/wiki/Wikipedia:Size_in_volumes says 2625 million words
16:27 < xentrac> not counting pictures
16:27 < xentrac> 21 GB
16:27 < fenn> so i'm seeing something like 44GB for the database (9GB compressed) and 23TB of images
16:27 < xentrac> uncompressed
16:28 < xentrac> I think it would be reasonable to focus just on the Vital Articles
16:28 < fenn> oh there is such a thing? good
16:28 < xentrac> for a start
16:28 < xentrac> https://en.wikipedia.org/wiki/Wikipedia:Vital_articles
16:29 < xentrac> there's a top-ten list, a top-100 list, and a top-1000 list that is currently at 988 articles
16:30 < fenn> hey Culture came in above Earth
16:30 < fenn> huh it's all dead english white guys
16:30 < xentrac> the top-10 list isn't ranked :)
16:30 < xentrac> Avicenna, Nagarjuna, and Qin Shi Huang are dead english white guys?
16:31 < fenn> i was looking at http://en.wikipedia.org/wiki/Culture
16:31 < xentrac> I mean there's a clear Occidental bias but it's hardly so overwhelming as that
16:31 < xentrac> oh
16:32 < fenn> the "number 1 article" apparently
16:32 < xentrac> yeah, sure is
16:32 < xentrac> no, it's just one of the top 10
16:32 < fenn> maybe the top 0 should just be a list of articles
16:33 < xentrac> the first one in https://en.wikipedia.org/wiki/Wikipedia:Vital_articles/Level/2 is Architecture
16:33 < xentrac> the top 10 are in alphabetical order
16:33 < xentrac> Culture comes before Earth because C is before E
16:33 -!- yorick [~yorick@oftn/member/yorick] has quit [Remote host closed the connection]
16:33 <@kanzure> no love for protoculture
16:34 <@kanzure> oops that's a real word, weird
16:34 <@kanzure> there we go http://en.wikipedia.org/wiki/Protoculture_(Macross)
16:35 < fenn> so those petroglyphs are 10,000 years old, what's the bitrate of spraying mud on rock
16:35 < xentrac> kanzure: any idea if AFM engraving in, I don't know, wax is on the order of 10 bits per second, or 100, or 1000, or 10k, or 100k, or 1M, or 10M, or 100M?
16:35 < xentrac> it depends on how you're spraying it
16:35 < fenn> um, via inkjet
16:35 <@kanzure> no, sorry, i don't have a good model of afm tip stuff
16:35 < xentrac> ah, potentially very high
16:36 <@kanzure> all i know is that everyone crashes their tip
16:36 < xentrac> my experience making neo-petroglyphs suggests that the bit rate of carving stone is like 1bps
16:36 < xentrac> by hand
16:36 < fenn> what about just spraying tiny quantities of hydrofluoric acid
16:36 < fenn> (ignoring the dissolving inkjet problem)
16:37 < xentrac> you can do that, or you can print something that you use for resist later, like toner-transfer PCBs
16:37 < xentrac> I wrote about this at
16:37 < xentrac> http://lists.canonical.org/pipermail/kragen-tol/2005-May/000777.html
16:37 < fenn> hmm that's a smarter way of doing it
16:37 < xentrac> but still haven't tried it
16:37 < xentrac> I'm pretty sure it's practical
16:38 < fenn> it's practical for making circuit boards at least
16:38 < xentrac> well I did try it in one sense
16:38 < xentrac> I tried it with Coca-Cola
16:38 < xentrac> didn't work
16:38 < fenn> vinegar, hydrogen peroxide, and table salt
16:39 < fenn> takes a while to get started but it works
16:39 -!- ruthie [~ruthie@dhcp-130-58-194-12.swarthmore.edu] has joined ##hplusroadmap
16:39 < xentrac> neat, I didn't know that
16:39 < xentrac> for copper, you mean?
16:39 < fenn> uh, on copper i mean
16:39 < xentrac> what color is the resulting solution?
16:39 < fenn> green, but the tank fills up with crud
16:41 < xentrac> I wonder if you can use that to get http://members.optusnet.com.au/eseychell/PCB/etching_CuCl/index.html
16:41 < fenn> vinegar + sodium chloride yields HCl, cupric chloride is formed by reaction between the copper and HCl, and in the presence of oxygen CuCl dissolves the copper
16:41 < fenn> right
16:42 < xentrac> neat
16:42 < fenn> you know if they're just going to stockpile all that gold they might as well engrave wikipedia on it
16:42 < xentrac> it's easier to get hydrogen peroxide than hydrochloric acid here in Argentina
16:42 < xentrac> I should take a lithography class so I can learn how to etch stone
16:42 < fenn> well HCl is just vinegar plus salt (minus the sodium acetate)
16:44 < fenn> there are a number of different chemical compounds known as "rock" or "stone", but you can dissolve silica (glass) with hydrofluoric acid
16:44 -!- nmz787_i [~nmccorkx@134.134.139.76] has quit [Quit: Leaving.]
16:44 < fenn> it's nasty stuff, causes lung damage and forms tiny needles in your blood, so it's probably a good thing that it's hard to get
16:44 -!- nmz787_i [~nmccorkx@134.134.139.76] has joined ##hplusroadmap
16:45 < Lemminkainen> you're not supposed to inhale the HCl or etchant
16:45 < Lemminkainen> don't any of you heathens have fume hoods?
16:45 -!- nmz787_i [~nmccorkx@134.134.139.76] has quit [Client Quit]
16:46 < fenn> whether a chemical is hard to get doesn't matter if we have fume hoods or not
16:46 < fenn> sometimes I have trouble with english
16:47 < xentrac> hmm, I think that if you mix vinegar and salt, you end up with vinegar and salt, not HCl and sodium acetate
16:47 < xentrac> there are a lot of stones that are easier to etch than glass
16:48 < fenn> sure, "rock salt" is water soluble
16:49 < fenn> how easy to etch do you want it to be?
16:49 < fenn> i suggested gold because it doesn't dissolve in hardly anything
16:50 < fenn> oh fwiw glass is a ... "glass" or amorphous crystalline matrix, and it will flow like a liquid over thousands of years
16:51 < xentrac> glass doesn't actually flow like liquid
16:52 < fenn> but, i was taught in gradeschool!
16:52 < xentrac> I know, me too
16:52 < xentrac> gold is indeed pretty stable
16:52 < eudoxia> so does glass flow like liquid or not
16:52 < xentrac> rock salt *does* flow like liquid when under pressure, which I think makes it unsuitable
16:52 < xentrac> no
16:52 < eudoxia> :c
16:53 < fenn> i think there are multiple misunderstandings around the "does glass flow" issue
16:54 < xentrac> https://en.wikipedia.org/wiki/Glass#Structure talks about them
16:54 < eudoxia> is ice a slow moving liquid as long as it's above the vitrification point of water?
16:55 < fenn> a plate of glass a meter tall and a centimeter thick was placed in an upright position at room temperature, the time required for the glass to flow down so as to thicken 10 angstrom units at the bottom (a change the size of only a few atoms) would theoretically be about the same as the age of the universe: close to ten billion years.
16:55 < xentrac> eudoxia: no, ice crystallizes almost completely
16:56 < eudoxia> oh man another cool shower thought that dies on contact with reality
16:56 < eudoxia> like my dreams and hopes
16:56 < xentrac> however, like salt, it does flow under pressure!
16:56 < xentrac> glass has some great benefits as an archival medium
16:56 < xentrac> it won't burn because it's already ash
16:56 < xentrac> except in, say, a fluorine atmosphere
16:57 < xentrac> and it
16:57 < xentrac> it's extremely isotropic and uniform, down to the atomic level
16:57 < xentrac> so you can write very, very small on it
16:57 < xentrac> and it's extremely cheap
16:58 < fenn> you could make a fluoride glass that wouldn't burn even in a fluorine atmosphere
16:59 < xentrac> yes, I imagine that you could
17:00 < xentrac> the solid fluorine compounds I know about tend to be pretty crystalline, but that's probably solvable
17:00 < fenn> i would be remiss if in this context i didn't even mention the silliness of "storing" anything; probably you've read this article http://kk.org/thetechnium/2008/12/movage/
17:00 < xentrac> I mean window glass is mostly silica, and almost nothing loves to crystallize more than silica does
17:01 < xentrac> that article is a good statement of the problem I want to solve
17:01 < xentrac> if all you need to erase a fact forever is to ban it for five years, humanity is in danger
17:02 < fenn> i found some old 3d models i made in mac OS 7 and have no idea how to get them into a modern format
17:03 < xentrac> yeah
17:03 < xentrac> on the other hand, textual STL files from the 1980s are still importable into modern software
17:03 < fenn> the whole "resource fork" thing is bad enough, but apparently there's a "desktop database" that keeps track of all the files and their names
17:03 < xentrac> and even if they weren't, you could hack something up pretty easily
17:04 <@kanzure> except that stl is useless
17:04 < fenn> I can export as DXF but it looks awful
17:05 < fenn> the models are cubic bezier lofts, and dxf just uses the control points as vertices of the mesh
17:05 < xentrac> yeah, stl kind of sucks
17:05 < xentrac> aw :(
17:05 < fenn> basically i just took a screenshot and said "welp guess that's that"
17:06 < fenn> lesson: only use free software
17:07 < xentrac> yeah, free software helps a lot
17:07 < xentrac> but also media that someone will be able to read in 300 or 1000 years will help a lot too
17:08 < fenn> kevin kelly's iMovie bitrot problem wasn't actually about data at all, it's about being locked into shitty proprietary formats
17:08 < fenn> xentrac: but will anyone understand english in 1000 years?
17:08 < xentrac> really? I think it's about using unstable heat-sensitive organic dyes to record your data
17:08 < fenn> (personally I hope not)
17:08 < xentrac> the problem isn't about iMovie
17:09 < xentrac> it's about the DVDs he used for the backups
17:09 < fenn> article subsection 1) Formats change.
17:09 < xentrac> it seems likely that if there is anyone in 1000 years, they will be able to understand English
17:10 < xentrac> I mean, what were the major languages 1000 years ago?  Arabic, Chinese, Sanskrit/Hindi, Latin, and Greek
17:10 < xentrac> more or less, right?
17:11 < xentrac> and those are still the major languages today, with the addition of English
17:11 < fenn> xentrac: let's make a bet: if, in 1000 years, the median sentient agent can parse 20th century english into a semantic tree, i will give you 1/1000 of my personal whuffie, and vice versa
17:11 < xentrac> no, I don't care about the median
17:11 < xentrac> the median sentient agent can't do that with Latin today
17:11 < fenn> exactly
17:12 < xentrac> but things written in Latin 1000 years ago are hardly unparsable today
17:12 < xentrac> it's just that most of them are lost because there weren't enough copies
17:12 -!- eudoxia [~eudoxia@r186-53-101-205.dialup.adsl.anteldata.net.uy] has quit [Quit: leaving]
17:12 < fenn> how many standard deviations do you need until you can find someone who reads ancient latin (and understands all the idioms and allusions, probably none)
17:12 < xentrac> we might hit a singularity in the next few decades that will render the whole discussion moot
17:13 < fenn> i haven't read much of the long now stuff, but i did read "anathem"
17:13 < xentrac> doesn't really matter though; I can muddle my way through Cicero myself, and my friend Seth can read it straightforwardly
17:13 < fenn> it seems like the big problem is people using your archive material as shack liner material or firewood
17:13 < xentrac> there are at least hundreds of people in my own city who can read it straightforwardly
17:13 < xentrac> that's one big problem
17:14 < fenn> similar discussions about the nuclear waste "problem"
17:14 < xentrac> but before you get to that problem you have to deal with the DVDs that fade spontaneously in 5 years
17:14 <@kanzure> maybe the archival material should give humans cancer
17:14 < fenn> hey bingo just use nuclear waste as the archive material, it's a glass
17:14 < xentrac> high-energy radiation is a way to make anything unstable, though
17:14 <@kanzure> radiation poisoning is probably more obvious, cancer is 3-5 months and you're scratching your head
17:15 <@kanzure> yeah but maybe it's just directed radiation
17:15 < fenn> if the grain size is smaller than 100nm (random ass-guess) the molecules in the crystal spontaneously rearrange and negate any radiation damage
17:16 < fenn> i forget how that works exactly but it is a real technology used in nuclear reactor cores
17:16 < xentrac> you only get that benefit if your medium is crystalline
17:16 < xentrac> and it just attenuates the damage
17:17 < xentrac> at grain boundaries you can still get it recrystallizing differently
17:17 < xentrac> but it attenuates the damage a *lot*
17:17 < fenn> kanzure: if the archive material were a DNA virus that integrated into the genome... it would probably cause cancer in some small percentage of infected hosts
17:18 < fenn> now I am thinking about that movie Prometheus
17:18 < fenn> what if you just landed it on the moon
17:19 < fenn> there are way too many free variables in this calculation
17:19 <@kanzure> ParahSailin wants to make a freedom virus to infect north korea with
17:20 -!- ruthie [~ruthie@dhcp-130-58-194-12.swarthmore.edu] has quit [Read error: Connection reset by peer]
17:21 < xentrac> the other day some friends of mine finally got around to printing out http://canonical.org/~kragen/bible-columns.png
17:22 < fenn> that link should have a warning about its size
17:22 < xentrac> sorry. it's about ten megs
17:22 < xentrac> more relevantly it's a lot of megapixels
17:22 < fenn> it's mostly the megapixels
17:24 < fenn> the project gutenberg copyright headers/footers are a little out of control
17:26 < fenn> if any man shall add unto these things, God shall add unto him the plagues that are written within this book
17:26 < fenn> the biblical copyright license, not OSI approved
17:29 < fenn> Infect a population with a harmless biological virus that makes them susceptible to suggestions.
17:29 < fenn> Now you can control their minds by sending suggestions via a TV program, email, etc!
17:30 < fenn> "ya gotta believe me" virus, also by vernor vinge
17:31 < fenn> uff google fail
17:32 < fenn> http://fennetic.net/irc/vernor_vinge_rainbows_end.pdf
17:32 < xentrac> yes
17:32 < xentrac> didn't vernor post a text version on his own web site?
17:34 < fenn> tracker works better for finding text in books
17:34 < fenn> so, uh, maybe?
17:35 < fenn> http://vrinimi.org/
17:35 < xentrac> yeah, that's the one
17:35 < xentrac> except I don't see the link to the full text
17:35 < fenn> yeah i thought it was there, oh well
17:36 < xentrac> strangely it's in the WABAC
17:36 < xentrac> /rainbowsend.html
17:36 < xentrac> http://web.archive.org/web/20071124035416/http://vrinimi.org/rainbowsend.html
17:36 < fenn> probably his publisher complained
17:38 < xentrac> it was there for quite a while I think
17:39 < xentrac> at least until 2008-07-04
17:39 < xentrac> I guess that's just over a year
17:39 < fenn> the median internet page lifetime
17:39 < xentrac> sigh
17:39 < xentrac> btw, Vinay's NaNoWriMo product is kind of Rainbows-End-ish
17:40 < xentrac> mother of hydrogen
17:40 < xentrac> http://files.howtolivewiki.com/MOTHER_OF_HYDROGEN_NOVEL/
17:43 < fenn> added to reading list
17:43 < fenn> i wish NaNoWriMo were named something else
17:44 < fenn> i keep seeing it, thinking "that sounds familiar, is it about nanotechnology?" and being disappointed
17:44 < xentrac> heh
17:44 < xentrac> this particular novel has a fair bit of nanotechnology in it
17:44 < xentrac> in particular the political consequences of its limited prohibition
17:45 < fenn> ah, doctorow's "the siege" was pretty intense
17:46 < fenn> "after the siege" http://craphound.com/?p=1674
17:48 < fenn> sigh another redacted fulltext http://fennetic.net/irc/cory_doctorow_after_the_siege.pdf
17:49 -!- ruthie [~ruthie@dhcp-130-58-194-12.swarthmore.edu] has joined ##hplusroadmap
17:50 < fenn> oh it was linked as "infinite matrix"
17:51 < fenn> some of doctorow's work hits a little too close to home, like when you remember sitting at the desk being described in the story, working on the 3d printer described in the story
17:52 < fenn> and then the main character gets hauled off into some CIA torture cell because he was a the wrong place and time
17:53 -!- ruthie [~ruthie@dhcp-130-58-194-12.swarthmore.edu] has quit [Ping timeout: 258 seconds]
17:53 -!- jamesz [~ielo@c-69-255-193-198.hsd1.md.comcast.net] has joined ##hplusroadmap
17:55 < fenn> wow mupdf is fast, where have you been all these years
17:58 < fenn> great, now I can chuck out all that crap i was keeping around just to run evince
18:02 -!- rayston [~rayston@ip68-106-242-42.ph.ph.cox.net] has joined ##hplusroadmap
18:08 -!- nmz787_i [nmccorkx@nat/intel/x-cvvcszmgedzcnhny] has joined ##hplusroadmap
18:17 < fenn> Is this the beginning of digital Darwinism? To survive, information has to grab attention so as to be moved to the next media type. Information which can’t do this, effectively dies.
18:19 <@kanzure> yawn
18:19 < fenn> are you suffering a modafinil deficiency
18:19 <@kanzure> "scarcity exists! things that get copied tend to have more copies of themselves in existence!"
18:20 < fenn> people have been talking about "memes" for a while, but not literally in the context of data storage
18:21 < fenn> anyway the "beginning" was in ancient sumeria
18:21 < dingo> 01:17 < fenn> Is this the beginning of digital Darwinism? To survive,
18:21 < dingo> i worked at portico.org specificly in digital preservation
18:22 < dingo> its not easy when fuckers are using closed-source document formats, like MSWord
18:22 <@kanzure> yes dingo has suffered enough cumulative xml-hours for an entire country
18:22 < dingo> whose to say you can still use MS Word in 25 years?
18:22 < dingo> much less computers that can do machine-dependent data types
18:22 < fenn> dingo: ah then you must know all about emulators
18:22 < fenn> dingo: how do i preserve my old macintosh documents?
18:23 < dingo> start with an era macintosh and begin working your way up, losing formatting along the way
18:23 < fenn> i can't seem to copy them correctly
18:23 < fenn> i mean the files go invisible and weird shit like that
18:23 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 258 seconds]
18:23 < dingo> dunno i only began macintosh with the release of osx, my first mac had osx 10.0.1
18:24 < fenn> bah that's just unix, it works properly, no challenge there
18:24 < dingo> well /usr/bin was /Usr/bin that release, so it didn't work so well then, hehe
18:24 < dingo> and tcsh shell without bash
18:24 < fenn> well at least a file is a file
18:25 < fenn> not an entry in two databases and a file and a resource fork
18:25 < dingo> indeed; for older macs you have to deal with the classic concept of Resources, that might be where you're losing things, where a file isn't a file
18:25 < dingo> new macs still have these, but the resources innards are files, again
18:26 < dingo> ahh well all chat makes for no code :-) ttyl fenn
18:26 < fenn> cheers
18:31 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap
18:42 -!- drewbot [~cinch@ec2-50-17-42-27.compute-1.amazonaws.com] has quit [Remote host closed the connection]
18:42 -!- drewbot [~cinch@ec2-54-196-84-253.compute-1.amazonaws.com] has joined ##hplusroadmap
18:44 -!- rayston [~rayston@ip68-106-242-42.ph.ph.cox.net] has quit [Ping timeout: 240 seconds]
18:54 -!- delinquentme [~dingo@192-195-81-250.PUBLIC.monkeybrains.net] has quit [Ping timeout: 245 seconds]
19:06 -!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Remote host closed the connection]
19:11 -!- sheena1 [~home@d75-155-198-81.bchsia.telus.net] has quit [Ping timeout: 240 seconds]
19:12 -!- rayston [~rayston@ip68-106-242-42.ph.ph.cox.net] has joined ##hplusroadmap
19:12 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has joined ##hplusroadmap
19:26 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has quit [Ping timeout: 240 seconds]
19:26 -!- kyknos_ [~kyknos@89.233.130.143] has joined ##hplusroadmap
19:30 -!- kyknos__ [~kyknos@89.233.130.143] has quit [Ping timeout: 240 seconds]
19:30 -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap
19:31 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has joined ##hplusroadmap
19:35 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has quit [Ping timeout: 245 seconds]
19:43 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has joined ##hplusroadmap
19:43 <@kanzure> http://zentasrobots.com/2014/04/17/outdoor-footage-of-morphex-and-a-servo-failure/
19:48 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has quit [Ping timeout: 276 seconds]
19:53 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has joined ##hplusroadmap
19:56 -!- cluckj [~cluckj@cpe-24-92-63-104.nycap.res.rr.com] has joined ##hplusroadmap
20:00 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has quit [Ping timeout: 276 seconds]
20:06 -!- HashNuke [uid12117@gateway/web/irccloud.com/x-tjpbrpvgkxfgyxge] has joined ##hplusroadmap
20:09 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has joined ##hplusroadmap
20:10 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has joined ##hplusroadmap
20:21 -!- nmz787_i [nmccorkx@nat/intel/x-cvvcszmgedzcnhny] has quit [Quit: Leaving.]
20:25 -!- jamesz [~ielo@c-69-255-193-198.hsd1.md.comcast.net] has quit [Ping timeout: 245 seconds]
20:25 -!- AshleyWaffle [~waffle@gateway/tor-sasl/anastasiawyatt] has quit [Remote host closed the connection]
20:25 -!- AshleyWaffle [~waffle@gateway/tor-sasl/anastasiawyatt] has joined ##hplusroadmap
20:47 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has quit [Ping timeout: 265 seconds]
20:51 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has joined ##hplusroadmap
20:55 -!- delinquentme [~dingo@74.61.157.78] has quit [Remote host closed the connection]
21:07 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has quit [Ping timeout: 240 seconds]
21:10 -!- Baube [~Baube@65.95.14.15] has quit [Read error: Connection reset by peer]
21:10 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has joined ##hplusroadmap
21:12 -!- Baube [~Baube@65.95.14.15] has joined ##hplusroadmap
21:13 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has quit [Client Quit]
21:16 < juri_> Muhahaha!
21:16 < juri_> fenn: you're alive!
21:17  * juri_ just created a new channel, for kanzure to forward people to: ##hminusroadmap
21:19 <@kanzure> that doesn't even need a roadmap, just blow up the planet
21:20 < juri_> i can just parody the dumb advice i see here.
21:20 -!- ebowden [~ebowden@CPE-121-223-135-8.lns1.bat.bigpond.net.au] has joined ##hplusroadmap
21:21 < juri_> "take lots of drugs! never sleep! homeopathy is the winrar!"
21:21 <@kanzure> that's your impression of this channel?
21:21 < juri_> "i will become immortal through repeated application of bacon!"
21:21 < juri_> no, that's my impression of the people you kick off of this channel. ;)
21:22 < ebowden> LOL. You guys get homeopaths?
21:22 <@kanzure> not many
21:22 < cpopell> nah
21:22 < cpopell> the people he kicks off are usually
21:22 < cpopell> RESURRECT MY MOTHER
21:22 < ebowden> I should hope so.
21:22 < cpopell> YOU TERRIFYING MONSTER AI WILL KILL US ALL
21:22 < cpopell> AND UPLOADING IS THE DEVIL
21:22 <@kanzure> or my favorite, "kanzure is an evil ai that is going to turn the world into computronium"
21:22 < cpopell> <sends Kanzure 600 pages of spam a day for 3 years straight>
21:23 < juri_> what about "i heard on reddit you can resurect my mother by uploading the memories i have of her. can you do that for $100, and a windows 95 CD?"
21:24 < ebowden> ...
21:24 < ebowden> What?
21:25 < juri_> ebowden: sorry, i'm parodying the horrible people that get kicked from here. that's bound to be confusing.
21:25 < ebowden> Oh, I know you are parodying.
21:25 < ebowden> It's just confusing.
21:25 < ebowden> Lack of context.
21:26 <@kanzure> confusion is the primary function of irc
21:26 < ebowden> So, I'd imagine everyone here would have heard of ISRIB and curcumin loaded nanoparticles by now?
21:27 < cpopell> don't assume people have heard of a specific thing
21:27 < cpopell> generally the safest
21:27 < cpopell> although I'm sure Lemminkainen has
21:27 < juri_> everyone here has their own specialty, and its not necissarily anything that overlaps. i'm a crazy 3d printer person.
21:29 <@kanzure> you're a crazy xml person
21:29 < juri_> xslt!
21:29 < juri_> (looking at xslt right now.)
21:30 < ebowden> Oh, who here does neuroscience, and are there any molecular biologists?
21:31 <@kanzure> yes
21:31 < ebowden> Neat.
21:32 < ebowden> Are you the molecular biologist?
21:33 <@kanzure> i like to pretend to be
21:33 < ebowden> I seem to remember Juri_ mentioning that the OP of this channel was in some scientific field.
21:33 < ebowden> Ah, ok.
21:33 < ebowden> Well, I only recently turned 18, so, you are very likely pretending a lot less than I.
21:34 < ebowden> You heard of the whole myeloperoxidase breaking down CNTs thing?
21:34 <@kanzure> i don't know what field i fall into
21:34 < ebowden> Oh?
21:35 <@kanzure> i don't see how your age is relevant
21:35 < ebowden> Well, whatever field it is, you can thank your lucky stars it aint bioethics.
21:36 -!- mode/##hplusroadmap [-o kanzure] by kanzure
21:36 < ebowden> Well, my level of knowledge is relavent.
21:37 < ebowden> And someone not yet out of high school is unlikely to have the same level of knowledge as an academic in his own field.
21:37 < kanzure> very few people in here qualify as an academic
21:37 < kanzure> one dude is in school, but that's his own fault
21:38 < ebowden> Well, I didn't say I thought there would be many.
21:38 < ebowden> That aside, what did you mean when you said you weren't sure what field you fell into?
21:39 < kanzure> well what's a field?
21:39 < kanzure> anyway, have some things:
21:39 < kanzure> http://diyhpl.us/cgit
21:39 < kanzure> http://diyhpl.us/~bryan/papers2/neuro/
21:39 < kanzure> http://diyhpl.us/~bryan/papers2/bio/
21:40 < cpopell> I qualified as an academic for a couple years and loathed it.
21:40 < cpopell> jmiller is an academic
21:40 < ebowden> Ah, ok.
21:40 < ebowden> So, what do you do cpopell?
21:41 < cpopell> cry about not having done CS and work on shitty freelance projects til I find a job
21:41 < cpopell> uh
21:41 < cpopell> ...I'm a freelance market analyst, formerly startup focusing on emerging techs, formerly nanoprinting masters student.
21:41 < ebowden> Ah.
21:42 < cpopell> I basically spent 7 years in school figuring out a few things-that all of my degrees were in fields I didn't like, that I should have started programming -way- sooner, and that I really like business strategy.
21:42 < ebowden> Ok.
21:42 < cpopell> Right now I'm trying to find a job in BI
21:42 < ebowden> BI?
21:42 < cpopell> business intelligence
21:42 < ebowden> And what's Jmiller's field?
21:42 < ebowden> Oh, ok.
21:43 < cpopell> he does advanced manufacturing
21:43 < ebowden> Oh?
21:43 < kanzure> no
21:43 < kanzure> tissue cultures
21:43 < ebowden> Ok.
21:43 < cpopell> http://amrinstitute.org/build_the_future..html he's the head of this group
21:44 < cpopell> is chevbird out of academia yet? Hmmm
21:45 < ebowden> What's chevbird do?
21:45 < cpopell> he worked on next generation sequencing
21:46 < cpopell> we have a lot of people who do research out of academia though
21:46 < ebowden> Like normal sequencing, but next generation.
21:46 < ebowden> Huh.
21:46 < ebowden> Ok.
21:46 < ebowden> So, what was different about the sequencing he was working on?
21:46 < Lemminkainen> ebowden:
21:47 < Lemminkainen> nanoparticles across blood-brain-barrier is not easy
21:47 < ebowden> Yes?
21:47 < ebowden> Ah.
21:47 < ebowden> Lemminkainen, you do nanoparticles?
21:47 < Lemminkainen> astrocyte endfeet form very tight junctions
21:47 < Lemminkainen> aye, of a sort
21:48 < ebowden> Oh?
21:48 < ebowden> Lipid nanoparticles?
21:48 < ebowden> Some other BBB crossing material?
21:50 < Lemminkainen> lipid nanoparticles have a lot of problems
21:50 < ebowden> Oh?
21:50 < Lemminkainen> to be frank, I don't yet know how to cross the BBB with NP
21:50 < Lemminkainen> it's a still-open question and we have a few theories, but it'll be a few months before we have data
21:51 < ebowden> I thought it'd been done.
21:51 < ebowden> Huh.
21:51 < ebowden> I was mistaken.
21:51 < Lemminkainen> there've been some academic papers showing 0.5-1.5% delivery efficacy
21:52 < ebowden> Interesting to see you can still do a lot with just that.
21:52 < Lemminkainen> clinically meaningless, though
21:53 < ebowden> Well, I'm even more wrong.
21:54 -!- Baube [~Baube@65.95.14.15] has quit [Ping timeout: 265 seconds]
21:54 < ebowden> So, size and lipophilicity are not the only factors?
21:54 < ebowden> What other ones have been identified?
21:55 < Lemminkainen> surface chemistry matters a LOT
21:55 < Lemminkainen> and liposomal formulations are going to accumulate in the liver pretty much no matter what
21:56 < ebowden> Ah.
21:56 < ebowden> So, what about the surface chemistry?
21:56 < Lemminkainen> that's a broad question
21:57 < ebowden> I know.
21:57 < ebowden> Just asking for some examples, things that could be worked on.
21:57 < Lemminkainen> well there've been some attempts to put MAb in them
21:57 < Lemminkainen> there've been a couple long charged polymer variations
21:57 < ebowden> MAb?
21:57 < Lemminkainen> BIND Therapeutics is doing interesting things
21:57 < Lemminkainen> MAb = monoclonal antibodies
21:58 < ebowden> Sorry, not too familiar with the nomenclature.
21:58 < ebowden> Google to the rescue.
21:59 < kanzure> acronyms are dumb anyway
21:59 < ebowden> Well, it saves a little keyboard time.
22:00 < kanzure> does it? http://www.seanwrona.com/typeracer/profile.php?username=kanzure
22:00 < Lemminkainen> your name doesn't stand for Kicking Antlers Non-Zodiac Utopian Recursion Efflux, kanzure ?
22:01 < kanzure> well that's a unique one
22:02 < cpopell> Not surprised antlers showed up tho
22:02 < ebowden> So, do you know exactly what BIND are doing lemminkainen?
22:02 < Lemminkainen> MAb-targeted docetaxil delivery
22:03 < Lemminkainen> lung and prostate tumors
22:04 < ebowden> Well, I din't mean "do you know", obviously you do. I meant to ask if you could tell me. Sorry if that sounded rude.
22:04 < ebowden> Which you did.
22:04 < Lemminkainen> don't worry about tone
22:04 < Lemminkainen> you're curious and that's neat
22:05 < ebowden> Good to see that kind of attitude.
22:05 < ebowden> (As in, welcoming curiosity.)
22:11 < ebowden> Well, this is one of the better places to go interneting. Lemminkainen, who here would be able to assist me in answering the question about the formation of random neural networks?
22:11 < ebowden> Well, this is one of the better places to go interneting. Lemminkainen, who here would be able to assist me in answering a question about the formation of random neural networks?
22:12 < fenn>  "What are you doing?", asked Minsky.
22:12 < fenn> "I am training a randomly wired neural net to play Tic-tac-toe", Sussman replied.
22:12 < fenn> "Why is the net wired randomly?", asked Minsky.
22:12 < fenn> "I do not want it to have any preconceptions of how to play", Sussman said.
22:12 < fenn> Minsky then shut his eyes.
22:12 < fenn> "Why do you close your eyes?" Sussman asked his teacher.
22:12 < fenn> "So that the room will be empty."
22:12 < fenn> At that moment, Sussman was enlightened.
22:12 < ebowden> What's that from?
22:13 < fenn> the Jargon File
22:13 < ebowden> Ok.
22:13 < ebowden> Fenn, you do neuroscience?
22:14 < fenn> i keep reading your name as ed boyed, or erik's bowden extruder
22:15 < fenn> derp. s/ed boyed/ed boyden/
22:15 < ebowden> My name is Eliot Pascal Bowden-Ferguson.
22:15 < cpopell> are your parents mathemeticians?
22:16 < fenn> apparently a bowden extruder is named for http://en.wikipedia.org/wiki/Bowden_cable whose invention is "popularly attributed to Sir Frank Bowden, founder and owner of the Raleigh Bicycle Company"
22:17 < ebowden> No, my father is a 72 year old impressionist artist, who's father, Phillip Bowden, was a scientist, and my mother is the EO of TABIS.
22:17 < cpopell> the whole pascal thing, I was curious.
22:17 < fenn> i had a similar reaction upon meeting tatiana gelfand
22:17 < fenn> and ruza markov
22:18 < fenn> that was an interesting desert trek
22:19 < fenn> ebowden: all of my neuroscience has been learned by necessity, not choice
22:19 < ebowden> Oh?
22:19 < fenn> i guess i have an interest in psychophysics
22:20 < ebowden> Psycophysics?
22:20 < kanzure> yeah, i don't think you should bother with randomly wired neural nets
22:20 < kanzure> go play around with yale.edu's NEURON
22:20 < kanzure> http://www.neuron.yale.edu/neuron/
22:20 < kanzure> "for empirically-based simulations of neurons and networks of neurons"
22:21 < kanzure> "NEURON is a simulation environment for modeling individual neurons and networks of neurons. It provides tools for conveniently building, managing, and using models in a way that is numerically sound and computationally efficient. It is particularly well-suited to problems that are closely linked to experimental data, especially those that involve cells with complex anatomical and biophysical properties."
22:21 < ebowden> Neat.
22:22 < kanzure> "The default integration method is implicit Euler, which provides robust stability and first order accuracy in time (sufficient for most applications)."
22:22 < kanzure> "There is also a Crank-Nicholson method that provides second order accuracy at little additional computational cost. However, this is prone to numerical oscillations if dt is too long, voltage clamps are present, or system states are described by algebraic equations."
22:23 < ebowden> Kanzure, I mainly want to find out if anyone has ever used particularly strong excititory tCMS to cause the formation of random networks.
22:23 < kanzure> "User-defined mechanisms such as voltage- and ligand-gated ion channels, diffusion, buffering, active transport, etc., can be added by writing model descriptions in NMODL, a high-level programming language that has a simple syntax for expressing kinetic schemes and sets of simultaneous algebraic and/or differential equations. NMODL can also be used to write model descriptions for new classes of artificial spiking cells. These model ...
22:23 < kanzure> ... descriptions are compiled so that membrane voltage and states can be computed efficiently using integration methods that have been optimized for branched structures. A large number of mechanisms written in NMODL have been made available on the WWW by the authors of published models; many of these have been entered into ModelDB which makes it easy for users to find and retrieve model source code according to search criteria such as author, ...
22:23 < kanzure> ... type of model (e.g. cell or network), ionic currents, etc.."
22:23 < fenn> psychophysics roughly is quantification of subjective perception, performance testing of the senses, basically attempting to determine the transfer function of the human signal transduction path
22:24 < ebowden> And if these networks are actually random.
22:24 < ebowden> Oh, ok fenn.
22:25 < fenn> i'm willing to bet you don't know what a transfer function is
22:25  * cpopell should review his control systems theory.
22:25 < ebowden> And you bet correctly.
22:25 < fenn> arr my chrome url bar is so dyslexic, it hurts
22:26 < ebowden> cpopell, what's your control systems theory?
22:26 < cpopell> the subject of control systems theory?
22:26 < fenn> note to self: wait until next LTS release is ready before "upgrading"
22:26 < cpopell> http://en.wikipedia.org/wiki/Control_theory
22:27 < kanzure> if servo is passing acid tests, does that mean it is generally usable for basic web surfing?
22:27 < fenn> a browser can be horribly slow and crash constantly and still pass ACID
22:27 < fenn> heh check out "surf"
22:28 < kanzure> i think i'm okay with those things, since it's embeddable and that provides a bunch of other options
22:28 < ebowden> So, kanzure, are networks formed by overly strong excititory tCMS actually random?
22:28 < fenn> i really doubt it
22:29 < fenn> the brain is strongly hierarchical due to cell growth patterns that have nothing to do with learning
22:30 < fenn> you will never get a hopfield network either since the cells take up physical space and have a certain maximum length
22:30 < ebowden> I would be interested to see, say, if we could induce an increase in functional connectivity between two regions by that method.
22:30 < ebowden> Oh, ok fern.
22:30 < fenn> it's "fenn"
22:30 < ebowden> (The overly strong tCMS.)
22:30 < ebowden> Oh, oops.
22:30 < ebowden> Sorry.
22:31 < ebowden> So, has anyone ever tried to see if they could do that?
22:32 < fenn> ebowden: do you know what a cortical column is? (a cortical minicolumn)
22:32 < kanzure> you can make neurons grow in certain directions by marking surfaces with uh, sugar or something
22:32 < ebowden> Please enlighten me.
22:32 < kanzure> maybe not sugar. some protein.
22:32 < ebowden> Oh, thanks kanzure.
22:33 < fenn> well, you see.. "they are made out of meat"
22:33 < ebowden> ...
22:33 < ebowden> Could you be a little more detailed?
22:34 < fenn> as an embryo grows its neural tube inflates like a balloon. the surface of the balloon self-patterns with many little hexagons about 0.1mm thick, these are the columns. as the balloon stretches it buckles and develops the wrinkles that people think of as the surface of the brain
22:34 < ebowden> Ok.
22:35 < fenn> 99% of the connections are between cells in the column and the neighborhing columns
22:35 < ebowden> Right.
22:35 < fenn> and there are a few long-range connections that are what differentiate autistic people from neurotypicals, or something
22:36 < ebowden> Oh?
22:36 < fenn> i am really the wrong person to learn about neuroanatomy from
22:36 < fenn> the point is that the brain is not randomly wired, not nowhere nohow
22:37 < ebowden> Oh, I know that.
22:37 < fenn> brain cancer maybe
22:38 < ebowden> Just asking if you could really call networks formed by 'too strong' excitatory tCMS truly random.
22:38 < ebowden> And you guys answered that quite well, thankyou.
22:38 < fenn> why are you interested in randomly wired neural nets? (btw "neural net" generally refers to the AI technique of weighted multilayer analog classifiers)
22:39 < ebowden> Well, I have heard some reports of seizure induced synaesthesia.
22:40 < fenn> heh
22:40 < fenn> what caused the seizure
22:40 < fenn> the synaestheizure!
22:40 < kanzure> if you can hit the head just right..
22:40 < ebowden> LOL
22:41 < kanzure> the one-quarter inch synesthesia-inducing punch
22:41 < fenn> you are already dead
22:41 < kanzure> my entire life, man
22:42 < ebowden> Mainly interested in seeing if it could be used to functionally increase the connectivity in or between certain areas.
22:42 < fenn> I tear apart my psyche, but still the conclusion grows clearer, the resolution sharper.
22:42 < fenn> Myself, constructing the simulator. Designing those defense structures gave me the perspective
22:42 < kanzure> increased connectivity might not bestow any cognitive side-effects
22:43 < kanzure> but also, targetting two particular neurons separated by multiple inches in dense brain tissue is highly problematic
22:43 < fenn> No time. All I can do is metaprogram myself over randomly, at a furious pace. An act of desperation, possibly crippling. Milliseconds pass. My death passes before my eyes. I absorbed the fatal insights before I had any defenses raised.
22:44 < ebowden> Oh, two adjacent areas kanzure.
22:45 < fenn> "Understand." At first I don't. And then, horrifyingly, I do.
22:46 < fenn> It's a memory
22:46 < fenn> trigger: the command is made out of a string of perceptions, individually harmless, that he
22:46 < fenn> planted in my brain like time bombs. The mental structures that were formed as a result of those
22:46 < fenn> memories are now resolving into a pattern, forming a gestalt that defines my dissolution.
22:46 < kanzure> "If I could find courage, I wondered, what would I see? Would I be ashamed of the arrangement of my programs - of my very self - beyond my control? Ah, but what if I could write new metaprograms, controlling this arrangement of programs"
22:46 < kanzure> using shame as a narrative device is sort of annoying though
22:47 < fenn> well he's a good jewish boy
22:48 < kanzure> sandor at the zoo only sent like two emails in the whole novel, how can wei dei claim to be inspired by that.
22:48 < kanzure> and why would 20 civilizations be sitting around constructing emails?
22:48 < fenn> it's "wei dai"
22:48 < kanzure> i mean, as a single entity
22:48 < kanzure> oops yes
22:48 < kanzure> for some reason i knew about him 2008
22:49 < kanzure> i'm not sure why you know about him?
22:49 < fenn> 2008 wasn't that long ago
22:49 < kanzure> *him in
22:49 < fenn> er, from my perspective at least
22:49 < fenn> having being dead and all
22:49 < kanzure> well you're perpetually in a weird time warp
22:49 < kanzure> so you don't count
22:49 < fenn> i mean what happened between 2008 and now
22:50 < kanzure> you ran into him before?
22:50 < fenn> i don't know
22:50 < kanzure> well they're all the same people anyway, i guess
22:50 < fenn> chinese names are impossible to keep track of
22:51 < fenn> and then academic journals have the gall to abbreviate their names to an initial
22:51 < fenn> "by w. dai et al."
22:51 < ebowden> Kanzure, basically, what I'm wondering is this, can we reliably induce synaesthesia in mice with localised tCMS?
22:51 < ebowden> Kanzure, basically, what I'm wondering is this, could we reliably induce synaesthesia in mice with localised tCMS?
22:51 < fenn> oh great that could be one of FIFTY MILLION PEOPLE
22:52 < kanzure> fenn: well, this one: http://www.weidai.com/
22:52 < fenn> ebowden: how do you know the mouse is experiencing synaesthesia?
22:52 < kanzure> ebowden: what's the test for mouse synaesthesia anyway
22:53 < kanzure> oh that's right i forgot that fenn is on my same wavelength, ugh
22:53 < kanzure> aren't there like ham radio protocols for this shit or something
22:53 < fenn> well if the mouse starts rearranging its habitat into hexagonal geometry that's a clue
22:53 < kanzure> "subject appears to be hexagonally cooperative today"
22:53 < fenn> kanzure: it's borg whispernet subspace communications
22:53 < ebowden> Well, that's probably not the ideal thing to try and test for, unless you expect them to respond differently to stimuli.
22:54 < ebowden> But there must be certain testable wires we could cross.
22:54 < kanzure> you can feed images directly into visual cortex brain matter if you want
22:54 < fenn> those voices in the darkness, is it me? is it you?
22:54 < kanzure> and you could convert audio signals on your end before you feed it into v1-v6
22:55 < kanzure> the advantage of this is that visual cortex tissue is somewhat easy to access, iirc
22:55 < ebowden> Kanzure, what would be the most testable wires we could cross, do you think?
22:55 < kanzure> uh depends on what you mean by cross
22:56 < fenn> this is all pointless, the cortical algorithm is generic. it's how stroke survivors learn to walk, [rant rematurely terminated]
22:56 < kanzure> not entirely pointless, i hadn't considered the possibility of using molecular markers to retarget synapses
22:57 < ebowden> I'm not sure how easy that would be on the macroscale.
22:57 < kanzure> but how would you get a distant axon to grow in a certain direction
22:57 < fenn> take an unmodified mouse, feed audio into V1, it will learn to "hear" again
22:57 < kanzure> on a petri dish this is trivial
22:57 < kanzure> for values of trivial that molecular biology and neurophysiology people enjoy
22:57 < fenn> still unclear (to me) if this is how the flanagan neurophone worked
22:59 < ebowden> Kanzure, are there adjacent regions in the brain that are not normally well connected, but when connected, would give obvious results?
22:59 < fenn> a seizure is pretty obvious
22:59 < kanzure> http://diyhpl.us/~bryan/papers2/neuro/Electrical%20stimulation%20of%20the%20human%20brain:%20perceptual%20and%20behavioral%20phenomena%20reported%20in%20the%20old%20and%20new%20literature.pdf
23:00 < fenn> "le grand mal" means the whole brain is involved
23:00 < kanzure> "obvious results" are hard to figure out
23:01 < ebowden> Oh?
23:01 < ebowden> Well, easy to test for.
23:01 < fenn> but it's a reliable indicator that something is happening
23:01 < fenn> mice don't normally have seizures
23:01 < kanzure> anselm spent a lot of time just on his light-induced "make the mouse run around in a circle" neuron
23:02 < kanzure> http://diyhpl.us/wiki/transcripts/hplus-summit-2009/anselm-levksaya/
23:02 < ebowden> Ok.
23:04 < kanzure> anyway the question of "obvious results" also applies to transcranial stimulation
23:04 < kanzure> the idea is that some amount of stimulation to some region of the brain could hypothetically improve some psychometric or psycho-measurable performance
23:04 < ebowden> Ok.
23:04 < kanzure> then the problem becomes a question of which region to stimulate or what has a greater chance of producing any effect
23:05 < kanzure> most of the stuff mentioned in that pdf is boring effects of electrical stimulation, like "tightness in chest"
23:05 < ebowden> Ah.
23:06 < kanzure> someone did motor cortex stimulation of a mouse to get whisker and forearm twitches
23:06 < kanzure> weren't the nazis or soviets supposed to do this shit for me? sigh
23:07 < ebowden> LOL
23:07 < ebowden> Out of curiosity, what did Nazi research contribute?
23:08 < JayDugger> Generally or in the context of cortex stimulation?
23:08 < ebowden> Well, can you think of any notable contributions to science that came from nazi research.
23:09 < fenn> pretty sure there were a number of "how far can we push a human along this axis until they die" type stuff
23:09 < ebowden> Yeah, there was a lot of that.
23:10 < JayDugger> Feel free to look up the history of aerospace, say 1946-1969.
23:11 < fenn> oh the germans came up with all kinds of things. the first computer was built by a german
23:12 < fenn> under the nazi regime. however it's unclear if that counts as "nazi research" since the bureaucrats didnt see the point until halfway through the war
23:12 < ebowden> Oh?
23:13 < JayDugger> Konrad Zuse, right?
23:14 < fenn> there's an often cited "oh the nazis did all kinds of valuable foundational research on human physiology that we could never get away with today" but i don't know off the top of my head what exactly they discovered/measured
23:14 < fenn> JayDugger: yeah, the Z1 was much more elegant than ENIAC or whatever
23:14 < JayDugger> Survival rates for hypothermia, low atmospheric pressure, I think.
23:14 < fenn> i mean BCD? what the fuck is that
23:17 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has joined ##hplusroadmap
23:18 < ebowden> So, has anyone actually achieved good bbb penetration with NPs?
23:18 < kanzure> stop with the acronyms
23:19 < ebowden> On sorry.
23:19 < ebowden> Oh, sorry.
23:19 < kanzure> it's almost as bad as cpopell saying BI everywhere
23:19 < jrayhawk> man i am totally calling you fern from now on
23:20 < fenn> my non-polynomial algorithms have successfully penetrated the better business bureau, now i can defraud consumers to my heart's content!
23:20 < ebowden> So, has anyone actually achieved good blood-brain-barrier penetration with nano particles?
23:20 < ebowden> Fenn, what?
23:21 < fenn> okay jayrayhawk
23:21 < jrayhawk> oh god, the tables have turned!
23:22 < fenn> to everything there is a season
23:22 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has quit [Ping timeout: 258 seconds]
23:24 < ebowden> Kanzure, how long has research on using nanoparticles to get drugs to cross the blood-brain barrier been going on without clinically significant results?
23:24 < cpopell> ebowden: if you want, i can hook you up with some people who did academic research on this
23:24 < kanzure> there's actually a site for clinical trial results
23:24 < cpopell> since I think Lemminkainen went back to work
23:25 < fenn> ebowden: could you insufflate instead of going through the blood?
23:25 < ebowden> Oh?
23:25 < ebowden> Insufflate?
23:25 < fenn> the olfactory bulb shares some ducts with the brain proper
23:25 < ebowden> Ok.
23:26 < fenn> hum. "Nasal insufflation (snorting) is commonly used for many psychoactive drugs because it causes a much faster onset than orally, and bioavailability is usually, but not always, higher than orally. This bioavailability occurs due to the quick absorption of molecules into the bloodstream through the soft tissue in the mucous membrane of the sinus cavity and portal circulation bypass."
23:27 < fenn> what is "systemic circulation"
23:27 < fenn> is that just the bloodstream?
23:28 < ebowden> Well, I'm not exactly an expert on neuroanatomy.
23:28 < ebowden> By not exactly, I mean in no way.
23:29 < fenn> okay, so vasopressin is a huge peptide, and people snort it for nefarious purposes. how does it get into the brain?
23:29 < ebowden> It certainly gets in.
23:30 < fenn> well, maybe not. it seems to interact with the hypothalamus, which is outsie the blood brain barrier
23:31 < ebowden> Oh.
23:31 < ebowden> Well damn.
23:31 < kanzure> i wonder what the microbe content of a sinus cavity is
23:32 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has joined ##hplusroadmap
23:32 < fenn> the other thing to consider is that there isn't such a thing as a perfect blood brain barrier, sometimes it develops cracks, and is more porous in some people than others
23:32 < fenn> kanzure: over 9000
23:32 < ebowden> LOL
23:32 < fenn> it hurts
23:32 < ebowden> Yeah, there are diseases that break it down.
23:33 < kanzure> you have an unhealthy appreciation of age
23:33 < sheena> we see this in dogs for ivermectin use. MDR1 gene
23:34 < kanzure> "22,23-dihydroavermectin B1a"
23:34 < fenn> trade name Scabo 6, for scabies! (yum)
23:35 < kanzure> "Field studies have demonstrated the dung of animals treated with ivermectin supports a significantly reduced diversity of invertebrates, and the dung persists longer."
23:35 < kanzure> brought to you by dungfacts
23:35 < ebowden> Oh, are there substances, which, when in the bloodstream, will temporarily increase the permeability of the blood-brain barrier?
23:35 < kanzure> what is your goal
23:36 < kanzure> and don't say random wiring again
23:36 < ebowden> Oh, this is a different goal.
23:36 < fenn> "Since drugs that inhibit CYP3A4 enzymes often also inhibit P-glycoprotein transport, the risk of increased absorption past the blood-brain barrier exists when ivermectin is administered along with other CYP3A4 inhibitors." (reads slowly)
23:38 < ebowden> I remember, trimethyl chitosan, increased the absorption of drugs through the gastrointestinal tract.
23:38 -!- sheena [~home@d75-155-198-81.bchsia.telus.net] has quit [Ping timeout: 276 seconds]
23:38 < fenn> P-glycoprotein 1 (permeability glycoprotein) also known as multidrug resistance protein 1 (MDR1) is an ATP-dependent efflux pump with broad substrate specificity.  in the capillary endothelial cells comprising the blood–brain barrier and blood-testis barrier, where it pumps [toxins, drugs] back into the capillaries.
23:38 < ebowden> When used to nano-encapsulate the same drugs, it increased absorbtion a lot more, and with fewer side effects.
23:39 < fenn> so basically the stuff about CYP3A4 was irrelevant, ivermectin inhibits MDR1
23:40 < ebowden> Perhaps we could use things that increase the permeability of the blood-brain-barrier in nanoparticles.
23:41 < Lemminkainen> Lemminkainen didn't go back to work, cpopell
23:41 < Lemminkainen> it's Friday night and therefore time for hedonism
23:41 < cpopell> kk. saw your fb message.
23:41 < cpopell> I'm back to working on my S.I. unit extraction script
23:41 < ebowden> Kanzure, you get my goal?
23:41 < fenn> interesting, some of these MDR1 inhibitors are used in nutraceuticals: Piperine, Quercetin,
23:42 < cpopell> plus some JoJo's Bizarre Adventure...
23:42 -!- bertm [~bertm@remote.pacificlighttech.com] has quit [Ping timeout: 245 seconds]
23:42 < ebowden> Ah, Piperine.
23:43 < fenn> cpopell: perhaps this may be of some use to you https://github.com/kanzure/skdb/blob/master/core/units.py
23:43 -!- bertm [~bertm@remote.pacificlighttech.com] has joined ##hplusroadmap
23:44 < fenn> oh my god i dont remember using sympy at all
23:44 < cpopell> this is so much more advanced than what I have
23:44 < kanzure> probably my fault
23:44 < kanzure> sorry
23:44 < cpopell> probably because this is sort of babby's first project
23:44 < fenn> anyway GNU units does a pretty good job of parsing units
23:44 < kanzure> there's no way you would have let me mess around with unit stuff though, so maybe it's not my fault
23:45 < cpopell> https://github.com/cpopell/jeweler/blob/master/prefix.py
23:46 < fenn> yeah there's no reason to do that when you have the "units" command already
23:46 < fenn> also 2/13cm^3 is pretty ambiguous
23:46 < cpopell> fenn: that's the point
23:46 < fenn> is it supposed to throw an AmbiguousUnitError?
23:47 < cpopell> no, I'm trying to figure out how to deal with bad science journalism
23:47 < cpopell> and inconsistent units
23:47 < kanzure> if i was going to bother i would probably do it with some shitty string grammar
23:47 < kanzure> i mean, if i was going to avoid using units
23:47 < kanzure> GNU units
23:47 < fenn> i could never get parser generator grammar stuff to work, it always fails in mysterious ways
23:47 -!- bertm [~bertm@remote.pacificlighttech.com] has quit [Ping timeout: 250 seconds]
23:48 < fenn> to ultra meta metatron's cube, optimus prime!
23:48 < kanzure> that's probably just the same weirdness as crond scheduling
23:48 < kanzure> "what's the point of a scheduler that never works?"
23:49 < cpopell> fenn: my unfortunately overly-optimistic goal is building a foraging system that hits 60% of stuff
23:49 < kanzure> "are we paying for this?"
23:49 < fenn> fwiw "units" says Definition: 153846.15 / m^3
23:49 < kanzure> foraging for what
23:49 < kanzure> you can most likely safely avoid all science news, so please please don't say science news
23:50 < fenn> whereas units 2/13*cm^3 Definition: 1.5384615e-07 m^3
23:50 < cpopell> not foraging for science news, but I'd like to forage from science news for general trend tracking
23:50 < ebowden> How much of science news is bollocks?
23:50 < kanzure> :(
23:50 < cpopell> a lot of it
23:50 < kanzure> approximately all of it
23:50 < ebowden> LOL
23:50 < fenn> 90% of science news is crap, but then 90% of everything is crap
23:51 < cpopell> my general assumption is that if you see something in science news you're 5 years from commercialization, if that.
23:51 < kanzure> science news is especially crap
23:51 < ebowden> You know, Lumosity hires people to go to comments sections and spread neurobollocks.
23:51 < kanzure> suppose you're designing a way to inform researchers about shit that's going on
23:51 < kanzure> science news the way it is would not be what you design
23:51 < fenn> since 90% of science is crap, and 90% of science news gets the science wrong, multiply and we get 99% overall crap factor
23:52 < kanzure> cpopell: fuck that shit, you should go straight to the source ("office of technology commercialization") at each university
23:52 < cpopell> there's a lot of shit that doesn't make it to OTCs, and there's a lot of universities that don't update their OTCs
23:53 < kanzure> evidence?
23:53 < kanzure> isn't that illegal
23:53 < ebowden> Isn't what illegal?
23:53 < kanzure> mumble mumble something bayh-dole
23:53 < fenn> hm units moved the .dat files to /usr/share/units/
23:54 < kanzure> "must be because i'm not subscribed to the gnu units mailing list"
23:54 < kanzure> "i set it in digest mode" </nerd>
23:54 < ebowden> Kanzure, have you seen those Lumosity ads on YouTube?
23:54 < kanzure> no, but i'm aware of lumosity
23:54 < fenn> so have there been any more papers on dual n-back?
23:55 < fenn> or really anything claiming to improve "fluid intelligence" (whatever that is)
23:55 < cpopell> kanzure: I do plan on going better places, but right now just writing a text scraper, tagger, and unit interpreter is a big enough challenge to push  my skill levels up several ticks.
23:56 < ebowden> Are most of the claims to improvements in working memory from practice with Dual-n-back neurobollocks?
23:56 < cpopell> http://www.gwern.net/DNB%20FAQ
23:57 < fenn> it really gets me when people write "foot-pound" and i'm like "no, that's a minus sign! and it's poundals!!!" but they never listen
23:58 < ebowden> Thanks cpopell.
--- Log closed Sat Apr 19 00:00:07 2014