--- Log opened Tue Apr 22 00:00:47 2014
00:32 -!- _Sol_ [SolGr@c-50-166-90-49.hsd1.nj.comcast.net] has quit [Ping timeout: 240 seconds]
00:38 -!- marciogm [~marciogm@186-210-029-254.xd-dynamic.ctbcnetsuper.com.br] has quit [Quit: Computer has gone to sleep.]
00:39 -!- _Sol_ [SolGr@c-50-166-90-49.hsd1.nj.comcast.net] has joined ##hplusroadmap
01:00 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap
01:13 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 240 seconds]
01:15 -!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 265 seconds]
01:21 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap
02:10 -!- kuldeepdhaka [~kuldeepdh@unaffiliated/kuldeepdhaka] has joined ##hplusroadmap
02:13 -!- EnLilaSko [EnLilaSko@unaffiliated/enlilasko] has joined ##hplusroadmap
02:35 -!- kuldeepdhaka [~kuldeepdh@unaffiliated/kuldeepdhaka] has quit [Ping timeout: 252 seconds]
02:36 < chris_99> paperbot: http://link.springer.com/article/10.1007/s10708-013-9516-8
02:36 < paperbot> http://diyhpl.us/~bryan/papers2/paperbot/The%20real-time%20city%3F%20Big%20data%20and%20smart%20urbanism.pdf
02:36 < chris_99> merci paperbot
02:38 -!- kuldeepdhaka [~kuldeepdh@unaffiliated/kuldeepdhaka] has joined ##hplusroadmap
02:42 -!- mosasaur [~mosasaur@95.98.35.148] has joined ##hplusroadmap
03:11 -!- kuldeepdhaka [~kuldeepdh@unaffiliated/kuldeepdhaka] has quit [Quit: Leaving]
03:17 -!- mosasaur [~mosasaur@95.98.35.148] has quit [Quit: Leaving.]
03:18 -!- mosasaur [~mosasaur@95.98.89.217] has joined ##hplusroadmap
03:43 -!- yorick [~yorick@oftn/member/yorick] has joined ##hplusroadmap
04:42 -!- Adifex is now known as Adifex|zzz
05:04 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving]
05:05 -!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap
05:10 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap
05:25 -!- eudoxia [~eudoxia@r179-25-177-62.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap
05:25 -!- GoatStimulator [~tpi@c-107-4-148-59.hsd1.va.comcast.net] has quit [Quit: Leaving]
05:25 -!- cyberman [~dalek@c-107-4-148-59.hsd1.va.comcast.net] has joined ##hplusroadmap
05:35 -!- andytoshi [~andytoshi@unaffiliated/andytoshi] has joined ##hplusroadmap
05:35 -!- marciogm [~marciogm@179.126.60.189] has joined ##hplusroadmap
06:23 -!- marciogm [~marciogm@179.126.60.189] has quit [Quit: Computer has gone to sleep.]
06:55 -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap
07:23 -!- ThomasEgi [~thomas@185.5.8.81] has joined ##hplusroadmap
07:23 -!- ThomasEgi [~thomas@185.5.8.81] has quit [Changing host]
07:23 -!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap
07:32 -!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 240 seconds]
07:46 -!- fireprfHydra [~fireprfHy@pool-173-70-216-225.nwrknj.fios.verizon.net] has joined ##hplusroadmap
08:10 < heath> kanzure, bkero, jrayhawk, ParahSailin: do you all use vagrant?
08:10 <@ParahSailin> no
08:11 < eudoxia> fwiw i use it :x
08:11 < heath> eudoxia: link?
08:11 < eudoxia> you mean to a git repo that uses it?
08:11 < mosasaur> whoosh
08:12 < heath> oh, no, i misread that as "i use :x"
08:13 < eudoxia> oh haha
08:14 < kanzure> yes i use vagrant sometimes
08:16 < heath> What I don't have a grasp on is switching from one project to the next with vagrant. I think you want to create a new vm per project, but I'm not sure if that's entirely accurate
08:17 < eudoxia> one project = one vagrant vm
08:17 < eudoxia> or more than one
08:19 < kanzure> might as well use lxc
08:19 < heath> i just got through using vagrant for an app, but now i have a new project to work on and i'm not wanting to go through the manual installation process..
08:19 < kanzure> i use one packer template/defintion per project
08:19 < heath> yeah, multiple vms sounds like a terrible idea
08:19 < kanzure> and then i build a vagrant box if i need to, or a docker container, or an ec2 image, or whatever
08:20 < kanzure> this way i don't need to store 10000 vagrant boxes on my hdd
08:22 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving]
08:23 -!- ebowden [~ebowden@CPE-60-231-177-134.lns4.dav.bigpond.net.au] has quit [Remote host closed the connection]
08:23 < ryankarason> --3
08:35 -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap
08:38 -!- nmz787_i [nmccorkx@nat/intel/x-aegyclltnusdttni] has joined ##hplusroadmap
08:38 -!- fireprfHydra_ [~fireprfHy@pool-173-70-216-225.nwrknj.fios.verizon.net] has joined ##hplusroadmap
08:38 -!- fireprfHydra [~fireprfHy@pool-173-70-216-225.nwrknj.fios.verizon.net] has quit [Ping timeout: 255 seconds]
08:43 < kanzure> do monks have specific locations on their heads that they hit regularly, or is it just general forehead/wood bashing
08:44 < mosasaur> only when they seem to doze off
08:47 < kanzure> "80% of Technical Information is Found Only in Patents" http://www.patinformatics.com/blog/revisiting-an-old-standard-80-of-technical-information-is-found-only-in-patents/
08:47 < kanzure> "The origins of this particular statistic date back to the late ‘70s, and for the most part patent practitioners have taken this as a given since the. If this is the case, has anyone actually looked at this question recently and determined if it is still true? In order to revisit this statistic I studied the patented substances indexed by the Chemical Abstracts Service (CAS). The patenting of chemical substances represents a reasonable ...
08:47 < kanzure> ... percentage of the technologies covered by the world’s patenting authorities, and thus represents a good collection to study to determine how often technologies mentioned in patents are never mentioned elsewhere."
08:56 -!- hehelleshin is now known as helleshin
09:06 -!- HEx2 [~HEx@hexwab.plus.com] has joined ##hplusroadmap
09:06 -!- HEx1 [~HEx@hexwab.plus.com] has quit [Read error: Connection reset by peer]
09:10 < jrayhawk> i don't even know what vagrant does
09:11 < jrayhawk> i use qemu and lxc and vservers a lot
09:11 < jrayhawk> i should probably learn to use openvz since they have cooler stuff than the vserver project does
09:13 < jrayhawk> i have previously used virtualbox but i was generally fairly unimpressed with it and i am not clear on what purpose vagrant serves with regards to it
09:25 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 264 seconds]
09:27 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap
09:32 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap
09:34 < kanzure> jrayhawk: vagrant supports a number of other backends other than virtualbox
09:36 -!- nmz787_i [nmccorkx@nat/intel/x-aegyclltnusdttni] has quit [Quit: Leaving.]
09:38 < jrayhawk> well, i suppose it would be easy to make something less terrible than libvirt
09:38 < jrayhawk> i am not super clear on why anyone uses libvirt either
09:39 < kanzure> vagrant is often used for distributing .box files with a vm inside
09:39 < kanzure> since nobody knows how to use chroots etc
09:39 < chris_99> i was trying to use libvirt until i found it doesn't support kvm's -snapshot mode
09:42 < jrayhawk> libvirt seems like a "14 competing standards? ridiculous! we need to develop one universal standard that covers everyone's use cases." sort of boondoggle
09:43 < kanzure> jrayhawk: http://docs.vagrantup.com/v2/why-vagrant/index.html http://vagrantbox.es/
09:52 < jrayhawk> this is the sort of situation that makes me want to start up a troll project
09:53 < kanzure> what would it look like?
09:53 < jrayhawk> "gosh, there are so many version control systems and they're so hard for developers; let's make version control meta-tool that makes a unified and universal workflow for developers"
09:54 < kanzure> you can call it version nexus
09:56 < mosasaur> Some database formats have been commoditized to plugin status, some people tried it with GUI frameworks too.
09:56 < ThomasEgi> jrayhawk, http://xkcd.com/927/ as you desired
09:57 < kanzure> i don't think he actually desired that
09:57 < kanzure> plus, it's a very old/common saying
09:58 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has quit [Remote host closed the connection]
09:59 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has joined ##hplusroadmap
10:01 -!- mosasaur [~mosasaur@95.98.89.217] has quit [Quit: Leaving.]
10:02 < jrayhawk> the comical thing is a PC disk image or a filesystem snapshot is something that is well understood and can be managed with a wide variety of tools, both standard and specialized
10:02 < jrayhawk> instead we get a box file
10:02 < jrayhawk> because "simplicity???"
10:04 -!- nmz787_i [~nmccorkx@192.55.55.37] has joined ##hplusroadmap
10:04 -!- ruthie [~ruthie@c-98-225-143-81.hsd1.pa.comcast.net] has quit [Ping timeout: 264 seconds]
10:07 -!- fireprfHydra_ [~fireprfHy@pool-173-70-216-225.nwrknj.fios.verizon.net] has quit [Ping timeout: 252 seconds]
10:08 -!- nmz787_i1 [nmccorkx@nat/intel/x-mpudqlfapajncmnx] has joined ##hplusroadmap
10:09 -!- nmz787_i1 [nmccorkx@nat/intel/x-mpudqlfapajncmnx] has quit [Client Quit]
10:10 -!- FourFire [~fourfire@87-113-15.connect.netcom.no] has joined ##hplusroadmap
10:11 -!- nmz787_i [~nmccorkx@192.55.55.37] has quit [Ping timeout: 240 seconds]
10:14 < kanzure> jrayhawk: getting everyone on a team on the same page in terms of development VMs or environments has been one of my number one problems i keep running into
10:14 < kanzure> jrayhawk: the easiest thing for me to do would be to deploy remote boxes on ec2 or whatever, but the problem is that you immediately give up all of the advantages of local development
10:14 < kanzure> and then there's a herd of people that keep insisting on osx
10:15 < kanzure> and sometimes even windows ("because that's what my laptop runs, duh")
10:24 -!- ruthie [~ruthie@dhcp-130-58-194-8.swarthmore.edu] has joined ##hplusroadmap
10:26 -!- trotsky [~ruthie@dhcp-130-58-194-8.swarthmore.edu] has joined ##hplusroadmap
10:30 -!- ruthie [~ruthie@dhcp-130-58-194-8.swarthmore.edu] has quit [Ping timeout: 276 seconds]
10:34 -!- marciogm [~marciogm@179.126.60.189] has joined ##hplusroadmap
10:37 < bkero> heath: no. I use lxc.
10:37 < bkero> virtualbox is awful
10:38 < chris_99> theres always kvm ;)
10:38 < kanzure> vagrant does lxc things though
10:38 < kanzure> and kvm things
10:38 < bkero> wut
10:38 < kanzure> yeah..
10:38 < bkero> vagrant doesn't do reparenting of processes like lxc does
10:38 < bkero> or running the guest off the same filesystem
10:38 < bkero> or have real networking
10:38 < kanzure> https://github.com/fgrehm/vagrant-lxc
10:38 -!- fireprfHydra [~fireprfHy@pool-173-70-216-225.nwrknj.fios.verizon.net] has joined ##hplusroadmap
10:39 < bkero> Yeah, those sorts of things have existed for a while. afaict none of them are actually complete or on parity with virtualbox.
10:39 < kanzure> well it's not like anyone recommends virtualbox
10:40 <@ParahSailin> is the virtualbox ko still considered taint due to shitty programmering?
10:40 < bkero> yes
10:42 < FourFire> So, what's the opinion in here of "AI-Box" ?
10:43 < kanzure> not interesting
10:43 < kanzure> suppose there is an ai on earth
10:43 < kanzure> does it really matter if it was originally in a box or not?
10:44 <@ParahSailin> would you let the basilisk out of the ai box?
10:45 < kanzure> even if i don't, that doesn't mean that everyone else wont
10:45 < kanzure> and it also doesn't mean that an ai wont show up that wasn't constrained to such a "box" originally
10:49 < chris_99> i'm curious about the transcripts for that ai box thing
10:49 < eudoxia> i'm curious about the transcripts when eliezer did it
10:52 < chris_99> i'm betting it involves some kind of blackmail
10:53 < eudoxia> i assumed they were just conspiring
10:54 < eudoxia> to help eliezer's image during the cult-buildind satge
10:54 < eudoxia> stage*
10:54 < chris_99> oh heh
11:01 < cpopell> eudoxia: it's been done between pairs of people who aren't Eliezer
11:02 < cpopell> I generally think LW-ers are among the worst people to play the AI-box game as guardians
11:02 < cpopell> they're already predisposed to wanting FAI
11:02 < eudoxia> 'will you give me a pony?'
11:02 < cpopell> not quite
11:03 < chris_99> cpopell, are there transcripts of those
11:03 < cpopell> chris_99: you'll have to ask around for them
11:04 < eudoxia> there are public ones i think
11:04 -!- Qfwfq [~Qfwfq@cm113.kappa36.maxonline.com.sg] has joined ##hplusroadmap
11:04 -!- Qfwfq [~Qfwfq@cm113.kappa36.maxonline.com.sg] has quit [Changing host]
11:04 -!- Qfwfq [~Qfwfq@unaffiliated/washirving] has joined ##hplusroadmap
11:05 -!- fireprfHydra [~fireprfHy@pool-173-70-216-225.nwrknj.fios.verizon.net] has quit [Ping timeout: 255 seconds]
11:07 < chris_99> http://lesswrong.com/lw/9ld/ai_box_log/ seems er, dubious
11:08 < justanotheruser> http://www.coindesk.com/airbitz-wins-toronto-bitcoin-expo-hackathon-darkmarket/
11:08 < eudoxia> that's the one
11:09 < kanzure> interesting points about an ETF http://www.thisismadness.eu/bitcoin/5-reasons-why-a-bitcoin-etf-is-a-bad-idea/
11:09 < chris_99> oh i misread the outcome sorry
11:14 < xentrac> eudoxia: are you saying the cult-building stage is over? why?
11:15 < eudoxia> xentrac: hm, maybe just critical, early cult building
11:18 < FourFire> chris_99, well we might find out about that, later
11:19 < FourFire> eudoxia, I'm not sure that it's part of "cult building"
11:19 < chris_99> what reason would there be for not publishing the transcripts
11:20 < FourFire> chris_99, there are transcripts of some people who are not EY doing AI-Box, but the ones I've seen aren't all that interesting
11:20 < chris_99> nah i'm talking about the original ones
11:20 < FourFire> chris_99, lose your strategy "I would never fall for that!" when it's tailored to your specific opponent
11:21 < FourFire> or, maybe, yeah it is tied to blackmail, and neither party wants it to get out
11:21 < FourFire> others have claimed that the logs themselves can be used as blackmail, since emotional or character weaknesses can be exposed during the intense 2 hour (or longer) period
11:22 < FourFire> people might say... morally reprehensible things, if money is at stake
11:23 < FourFire> the is a few reasons, if you think about it for a bit, actually.
11:23 < eudoxia> 'intense 2 hour period' god it's a thought experiment not a war
11:23 < eudoxia> this is what eliezer wants you to believeeeeee
11:23 < xentrac> if we take the extreme perspective
11:24 < xentrac> an UFAI would find the transcripts useful for exiting the box
11:24 < FourFire> eudoxia, I'm unsure whether two people pitting their minds up against eachother should be shoved aside so easily
11:24 < chris_99> eudoxia, heh, two hours on a computer! this is madness!
11:24 < kanzure> what's the purpose of the ai box experiment?
11:25 < kanzure> or the thought experiment, rather
11:25 < xentrac> kanzure: to persuade people that FAI is necessary
11:25 < chris_99> FAI?
11:25 < xentrac> "friendly AI"
11:25 < kanzure> wow, really?
11:25 < FourFire> though, eudoxia having said that, I would be interested in doing the AI-Box experiment with EY as the AI
11:25 < kanzure> i would have guessed another reason entirely
11:25 < xentrac> what's your reason?
11:25 < kanzure> well, i don't have one, which is why i asked
11:25 < xentrac> chris_99: an Eliezer term meaning "AGI whose interests are aligned with those of humanity"
11:25 < xentrac> oh
11:25 < kanzure> but it would not have been "because it demonstrates that ai will kill you"
11:26 < FourFire> of course, I'm poor, so I'd never be able to afford his time, and there are much much better things to spend both time and money on
11:26 < kanzure> "it's just a text stream, it's not going to convert me into computronium"
11:27 < xentrac> kanzure: it seems to me that one of the major objections to FAI is that it's unnecessary because we can keep UFAIs safely boxed up; just as you said: "it's just a text stream, it's not going to convert me into computronium"
11:27 < FourFire> kanzure, the problem with that statement is, that as far as I know, at least three people, with monetary stakes, have had their minds changed during the experiment
11:27 < kanzure> like, aren't there much simpler ways to argue that ai will kill you, other than "ai box experiment"?
11:27 < xentrac> and the purpose of the experiment is to persuade people that in fact a text stream is sufficient to, effectively, convert you into computronium
11:27 < kanzure> FourFire: what do you mean? they suddenly decide that ai wont kill them, because it's text?
11:27 < FourFire> the thought experiment was proposed in order to determine whether humans can be hacked by humans
11:28 < FourFire> and if they can, then it can be assumed that they certainly can be hacked by an AGI
11:28 < kanzure> of course humans can be hacked by humans. we can even get "hacked" by metal rods being shot through your skull.
11:28 < xentrac> right
11:28 < xentrac> mmmm, metal rods :)
11:28 < FourFire> regardless of whether it was a F or uF AI
11:28 <@ParahSailin> the basilisk is actually a FAI
11:28 < kanzure> i still don't see why the thought experiment is necessary
11:29 < catern> the AI box stuff is just an argument against people who think we can prevent AI being dangerous by just sticking it in a box
11:29 < FourFire> hacked... by a text stream in an IRC window
11:29 < FourFire> ^
11:29 < xentrac> ParahSailin: haha
11:29 < kanzure> irc can fuck you up
11:29 < kanzure> i used to be a young school girl
11:29 < kanzure> look at me now
11:29 < xentrac> you're all wrinkly
11:29 < FourFire> uhh
11:29 < FourFire> I can't see you
11:30 < catern> there are people who think that all you need to do for FAI is box the AI, so the purpose of AI box experiment stuff is to counter them
11:30 < kanzure> anyway, the box experiment seems to have some other problems with it too
11:30 < kanzure> suppose that you really did have an ai in a box
11:30 < kanzure> why is it that only one person accesses the box?
11:31 < kanzure> okay.
11:31 < kanzure> well anyway, it's stupid
11:31 < kanzure> one side of the argument is saying "it's an unstoppable bullet by definition" and the other side is going "i don't accept your definition, so therefore it is stoppable"
11:31 < cpopell> iirc Tuxedage has made 2 grand off of it.
11:31 < cpopell> as the AI.
11:31 < FourFire> kanzure, no, the people decided to let their opponent "out of the box" after having resolved to, and bet money on, not let them out
11:31 < cpopell> convincing gatekeepers to let people out
11:32 < cpopell> *let him out
11:32 < eudoxia> ugh how do people even
11:32 < FourFire> these are people who lost money, so it's unlikely that they were just doing it for lols
11:32 < kanzure> and, even if you do have an ai in a box, what happens to all the other ai on the planet or in the galaxy
11:32 < cpopell> eudoxia: like I said, I don't think people who do LW are great gatekeepers
11:33 < eudoxia> how do we know money actually changed hands though?
11:33 < eudoxia> federal reserve notes don't have a blockchain
11:33 < cpopell> eudoxia: I don't think they care -that much- about falsifying this
11:33 < eudoxia> hm yeah i should drop the tinfoil for a while
11:34 < FourFire> if they're covering up, they were still convinced to cover up
11:34 < FourFire> there's an amount of reputation at stake here as well
11:34 < cpopell> eudoxia: you can always just ask Tuxedage
11:35 < FourFire> kanzure, it's not a thought experiment for determining whether we can stop AI by keeping it in a box.
11:36 < FourFire> It's a thought experiment to test whether humans can be hacked
11:36 < kanzure> you don't need ai to determine whether or not humans can be hacked
11:36 < catern> i think it's actually neither of those
11:36 < chris_99> what do you think it is catern
11:36 < FourFire> and, the context for why humans are being hacked, is a super intelligent AI, being boxed
11:36 < catern> because it's obviously true to anyone with sense that you can't stop AI by keeping it in a box, and it's obviously true that humans can be hacked,
11:37 < catern> but some people are dense and don't know this
11:37 < catern> so the AI box stuff is just stating the obvious for their benefit
11:37 < FourFire> the experiment is to provide data, which might convince such stubborn people
11:37 < catern> yeah, and that
11:37 < FourFire> (or prove them right all along)
11:37 < kanzure> well that's really boring
11:37 < catern> yes
11:38 < catern> it's not aimed at you
11:38 < cpopell> alternately it's a decent way to figure out who's susceptible to being talked out of things like that
11:45 < archels> https://www.kickstarter.com/projects/openworm/openworm-a-digital-organism-in-your-browser
11:45 < archels> looks professional
11:45 < kanzure> "you can get all these stickers! because science."
11:45 < archels> godspeed.
11:46 < kanzure> it will be interesting to see if they meet their funding goal
11:46 < FourFire> gudsped
11:46 < kanzure> because they didn't donate to themselves
11:46 < kanzure> e.g. usually kickstarter projects have a 10-30% completion rate in the first hour because they just give themselves their own money
11:47 < kanzure> (which did not happen)
11:47 < kanzure> nothing like some propaganda though https://s3.amazonaws.com/ksr/assets/001/910/103/be54723725f6f3cd8cf05e4639ae98ec_large.jpg?1398156892
11:48 -!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 264 seconds]
11:48 < eudoxia> https://s3.amazonaws.com/ksr/assets/001/881/814/7e8b26b817d087dca308f203ada09127_large.png?1397568588
11:48 < archels> yeah, cute
11:48 < eudoxia> "let me be" lol
11:49 < justanotheruser> Interesting. So I could just make a kick starter, fund myself completely and then I would have a little hype behind me
11:49 < justanotheruser> I assume funding yourself is against the ToS?
11:49 < archels> eudoxia: tamagotchi 2.0
11:49 <@ParahSailin> what is this open worm stuff anyway, is it somehow attempting to simulate every cell of worm?
11:49 < kanzure> justanotheruser: nope, as far as i know they are happy to take a % of your money
11:50 < kanzure> justanotheruser: but most of the time people only fund 10-20% of their project on their own, to make it look like there's momentum
11:50 < kanzure> justanotheruser: (this helps users not feel like idiots when donating)
11:50 < justanotheruser> kanzure: I see...
11:50 < archels> ParahSailin: not every cell, but every neuron and muscle at least. the rest is point-and-stick mechanics, afaik
11:51 < kanzure> they have been around for a while
11:51 < kanzure> too bad they keep doing those google hangouts :(
11:51 < kanzure> what happened to their irc channel. did they just decide irc was too old school for them?
11:51 < chris_99> heh
11:52 < archels> ParahSailin: not sure what the 'let me be' adds, there
11:52 < kanzure> 11:27 < petertodd> amiller: reminds me of a much simplier idea I had: have OpenPGP WoT edges be based on what you believed the cost would be for an adversary to fool you into signing the wrong key
11:52 <@ParahSailin> did i whups somewhere
11:53 < kanzure> nope, it wasn't you
11:55 -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap
11:57 -!- eudoxia [~eudoxia@r179-25-177-62.dialup.adsl.anteldata.net.uy] has quit [Quit: leaving]
11:59 < justanotheruser> kanzure: you following amiller in -wizards? Its pretty interesting
12:07 < xentrac> http://www.amazon.com/1000m-Dyneema-Strong-Braided-Fishing/dp/B009665S24/ref=sr_1_3?ie=UTF8&qid=1398193369&sr=8-3&keywords=dyneema occurred to me reading Tim Bray's blog post about a fallen tree
12:07 < xentrac> https://www.tbray.org/ongoing/When/201x/2014/04/17/Postmodern-Repairs a tree fell on his house and smashed part of it
12:08 < catern> justanotheruser: -wizards?
12:08 < justanotheruser> catern: #bitcoin-wizards
12:09 < catern> ah, should have guessed
12:09 < FourFire> looking back at the logs of one time I ran AI-Box, as the gatekeeper
12:09 < FourFire> the poor AI was getting trolled pretty hard
12:10 < xentrac> it seems like you could quite reasonably string a Dyneema net over your roof to protect it from falling tree branches
12:15 -!- cpopell [~cpopell@pool-71-255-241-91.washdc.fios.verizon.net] has quit [Quit: Nettalk6 - www.ntalk.de]
12:15 -!- cpopell [~cpopell@pool-71-255-241-91.washdc.fios.verizon.net] has joined ##hplusroadmap
12:16 < delinquentme> ParahSailin, kanzure is reselling ecm a decent idea? Or am I just enamored w the slight macabre and the thought that it would be a good financial vessle into working into human applications
12:16 < delinquentme> ?
12:16 <@ParahSailin> wut
12:16 < delinquentme> like from what I can tell its a pretty straight forward operation ... and all I need is a HPLC
12:17 <@ParahSailin> if it keeps you busy, im all for it
12:17 < chris_99> Electronic countermeasure?
12:17 < delinquentme> ParahSailin, per yesterday Ive been looking at decellularizaing simple organs for research purposes and selling freeze dried ecm powder
12:17 < delinquentme> Extracellular matrix :D
12:18 <@ParahSailin> whats that rodent cell line that secretes good ecm
12:19 <@ParahSailin> ehs sarcoma
12:21 -!- nmz787_i [nmccorkx@nat/intel/x-daakmmcjfgcpxyku] has joined ##hplusroadmap
12:24 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving]
12:27 -!- entelechios [~elysium@181.194.129.225] has quit [Ping timeout: 252 seconds]
12:30 -!- FourFire [~fourfire@87-113-15.connect.netcom.no] has left ##hplusroadmap ["Leaving"]
12:33 < delinquentme> so if I run this pulverized + suspended ecm gel through a HPLC ... the source animal shouldn't matter right?
12:34 < delinquentme> And I have no idea how to setup the HPLC to select out the ECM material
12:34 < kanzure> ParahSailin: why is it even called high-performance anyway
12:35 < kanzure> the super duper plasma chromatography mcahine (SDPC)
12:35 <@ParahSailin> hewlett packard was first to market the machine
12:35 < kanzure> i'm going with superduper from now on
12:35 <@ParahSailin> high performance was a backronym
12:35 < kanzure> aha
12:36 < delinquentme> http://www.pnas.org/content/84/20/7109.full.pdf
12:37 -!- trotsky [~ruthie@dhcp-130-58-194-8.swarthmore.edu] has quit [Remote host closed the connection]
12:37 -!- ruthie [~ruthie@dhcp-130-58-194-8.swarthmore.edu] has joined ##hplusroadmap
12:38 < delinquentme> http://en.wikipedia.org/wiki/Extracellular_matrix#Molecular_components
12:38 < delinquentme> BAM!
12:38 < delinquentme> #internet!
12:38 <@ParahSailin> er
12:38 <@ParahSailin> not sure what you're trying to say
12:39 <@ParahSailin> that you've mastered search engines for yourself now and dont need to ask so many easily googled questions here?
12:40 < delinquentme> paperbot, http://www.ncbi.nlm.nih.gov/pubmed/15482831
12:40 < paperbot> http://libgen.org/scimag/get.php?doi=10.1016%2Fj.biomaterials.2004.04.042
12:41 < cpopell> kanzure: you were discussing technical debt the other day, right?
12:41 < delinquentme> ParahSailin, one day we're gonna hug it out
12:41 < kanzure> cpopell: what of it
12:41 < delinquentme> a few trustfalls later and we'll be skipping through the fields
12:41 < cpopell> oh, just came across a 2013 talk by Steve McConnell on technical debt
12:41 < cpopell> or, his slides at least
12:43 -!- ruthie [~ruthie@dhcp-130-58-194-8.swarthmore.edu] has quit [Ping timeout: 276 seconds]
12:54 < AshleyWaffle> Hey everyone, I'm starting the first episode of my libertarian transhumanist podcast today in 1 hour and 30 minutes. Please contact me if you are interested in joining in.
12:55 < cpopell> Link a script.
12:55 < cpopell> or, topic list.
12:55 < cpopell> I'm skeptical.
12:57 <@ParahSailin> you believe in free will?
12:57 < cpopell> who are you addressing?
12:57 < justanotheruser> I'm starting the first episode of my statist primoanarchist podcast today what a coincidence
12:58 <@ParahSailin> libertarian
13:00 < cyberman> i though podcasts would have been one of those things that would have died a long time ago
13:00 < cpopell> instead they've only grown!
13:00 < cyberman> whats the roi these days
13:01 < justanotheruser> -1%
13:01 < cyberman> figures
13:01 < cyberman> prolyl why i got out
13:01 < justanotheruser> That was a made up number
13:02 < cyberman> too much like a job i wasnt getting paid for
13:02 < cpopell> cyberman: generally you have to be an expert in a given field already, or a celebrity, and you increase your branding.
13:02 < cyberman> end up plugging sponsors
13:03 < cyberman> and selling adspace
13:03 < cpopell> mostly ads, from what I've seen
13:03 < cpopell> and merch
13:03 < cyberman> swag
13:05 -!- nsh_ [~nsh@wikipedia/nsh] has joined ##hplusroadmap
13:06 -!- nsh [~nsh@wikipedia/nsh] has quit [Ping timeout: 240 seconds]
13:06 < AshleyWaffle> just using google hangouts instead
13:07 < kanzure> gross
13:07 < justanotheruser> ^
13:07 < AshleyWaffle> lol
13:07 < AshleyWaffle> just for first episode
13:10 < AshleyWaffle> I will be on http://www.youtube.com/watch?v=no1pxcvK8hw
13:14 -!- ruthie [~ruthie@dhcp-130-58-194-8.swarthmore.edu] has joined ##hplusroadmap
13:25 -!- nsh_ [~nsh@wikipedia/nsh] has quit [Ping timeout: 240 seconds]
13:25 < kanzure> http://www.securityfocus.com/bid/63676/discuss ""Attackers can exploit this issue to bypass Address Space Layout Randomization (ASLR) protection mechanisms of applications. This may aid in further attacks that may lead to arbitrary code execution."
13:25 -!- ruthie [~ruthie@dhcp-130-58-194-8.swarthmore.edu] has quit [Quit: DANG]
13:27 -!- kuldeepdhaka [~kuldeepdh@unaffiliated/kuldeepdhaka] has joined ##hplusroadmap
13:28 < kanzure> https://www.facebook.com/andrea.kuszewski/posts/10203765603640037?comment_id=8534877
13:28 < kanzure> "The frequency and scale of cyber-attacks against commercial and government interests has increased dramatically. Massive troves of classified government documents have become public through the actions of a few. Yet we have not seen significant growth in the illegal sharing of peer-reviewed academic articles. Should we truly expect that biomedical publishing is somehow at less risk than other content-generating industries? What of the larger ...
13:28 < kanzure> ... threat—a “Biblioleaks” event—a database breach and public leak of the substantial archives of biomedical literature?"
13:28 < kanzure> hahaha
13:28 < kanzure> paperbot: http://www.jmir.org/2014/4/e112/
13:28 < paperbot> http://diyhpl.us/~bryan/papers2/paperbot/77c2fe55ae949ec1c5d0b1757fcb1b99.txt
13:28 < kanzure> aww
13:29 -!- nsh_ [~nsh@wikipedia/nsh] has joined ##hplusroadmap
13:39 < delinquentme> AshleyWaffle, so peter thiel is calling in?
13:40 < AshleyWaffle> wait what!?
13:40 < delinquentme> ^
13:40 < AshleyWaffle> I don't know about that but there's a good chance Adam Kokesh is
13:40 < delinquentme> thats what the bio on the page says?
13:40 < delinquentme> thats cool
13:41 < AshleyWaffle> what?
13:45 -!- marciogm [~marciogm@179.126.60.189] has quit [Quit: Computer has gone to sleep.]
13:59 < xentrac> 15:53 < cpopell> Link a script.
13:59 < xentrac> 15:53 < cpopell> or, topic list.
13:59 < xentrac> 15:53 < cpopell> I'm skeptical.
13:59 < xentrac> 15:55 <@ParahSailin> you believe in free will?
13:59 < xentrac> of course I believe in the free list
13:59 < xentrac> I mean, malloc works, right?
13:59 < xentrac> oh wait
14:09 -!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 252 seconds]
14:11 -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap
14:11 -!- dlfk [~dlfk@chem-179-154.chem.tamu.edu] has quit [Quit: Nettalk6 - www.ntalk.de]
14:17 -!- dlfk [~dlfk@chem-179-154.chem.tamu.edu] has joined ##hplusroadmap
14:29 < andytoshi> hey guys, this has been bugging me, what if you uploaded a consciousness and then applied program obfuscation http://eprint.iacr.org/2013/451 to it, so that it'd be functionally identical but the source becomes opaque, wolud it still be conscious? the obfuscation in that paper works by representing the conscious circuit as a series of branching 5x5 matrices, then basically evaluating each matrix in a
14:29 < andytoshi> random basis and evaluating the output in the composition of all basis changes
14:29 < kanzure> one solution to consciousness is to consider maybe it doesn't exist
14:29 < kanzure> suppose that you have a whole brain emulation that works
14:30 < andytoshi> so it seems like the person could not be conscious because he doesn't have the priviledged view into his own working (since he doesn't know the random bases)
14:30 < kanzure> let's also say that it's accurate and it works correctly
14:30 < kanzure> but it does not possess consciousness (as a definition of the scenario)- is this to bad?
14:30 < kanzure> *is this so bad?
14:30 < andytoshi> but the I/O behaviour means that he'll still talk about being conscious and stuff. so isn't this a chalmers zombie?
14:30 < andytoshi> kanzure: the problem i have is that a philosophical zombie seems like an incoherent idea to me
14:30 < kanzure> and of course by emulation i refer to the type discussed here: http://diyhpl.us/~bryan/papers2/brain-emulation-roadmap-report.pdf
14:31 < kanzure> andytoshi: elaborate?
14:31 < kanzure> paperbot: http://eprint.iacr.org/2013/451
14:31 < paperbot> http://diyhpl.us/~bryan/papers2/paperbot/Candidate%20Indistinguishability%20Obfuscation%20and%20Functional%20Encryption%20for%20all%20circuits.pdf
14:31 < andytoshi> the link i posted was free :)
14:31 < kanzure> at this point it's often easier to use paperbot than check
14:31 < andytoshi> one sec, i'll find an article by eli yudkowskey on philosophical zombies, i don't have much to say beyond that..
14:32 < QuantumG> wait, someone actually discussing transhumanism on ##hplusroadmap?
14:32 -!- nsh_ is now known as nsh
14:32 < andytoshi> ok, it's super long, sorry
14:32 < andytoshi> http://lesswrong.com/lw/p7/zombies_zombies/
14:32 < kanzure> andytoshi: just to be clear upfront, i don't think i'm conscious
14:32 < jrayhawk> i am sure they'll be banned soon enough
14:32 -!- EnLilaSko [EnLilaSko@unaffiliated/enlilasko] has quit [Remote host closed the connection]
14:33 < kanzure> or rather, i don't think i have consciousness. (i am certainly in the medically-defined conscious state, of course. responsive, aware, communicative.)
14:33 < andytoshi> kanzure: well, i think i'm conscious, it sure seems to me that there's a 'view' i have of the world that others don't
14:33 < kanzure> and are you claiming that this view does or does not correspond to emulatable state in your head-matter?
14:33 < QuantumG> you don't think you have a poorly defined term?
14:33 < andytoshi> i'm claiming it does
14:34 < jrayhawk> if you're not self-aware of how self-aware you are, then...
14:34 < andytoshi> does correspond to an emulatable state*
14:34 < jrayhawk> or, rather, if you're not self-aware that you're self-aware, then... greetings mister epimedes
14:34 < QuantumG> there was that "recent" mathematical definition of consciousness.. that is basically useless
14:34 < kanzure> so you're asking if a brain emulation is still a brain emulation, even if it's not identical to the original source data/matter?
14:35 < andytoshi> i think consciousness is real, it's a purely physical phenomenon (otherwise i couldn't be talking about it on the physical IRC could i?), and running by brain in a sim is sufficient to acheive that consciousness
14:35 < andytoshi> worse than that, i'm ok with "not identical"
14:35 < andytoshi> in this case it's "obfuscated so that you cannot see any subset of the code, only the input/output behaviour of the whole thing"
14:35 < kanzure> then i don't know what you're asking anymore
14:38 < andytoshi> on p18 on that eprint.iacr.org paper, the security assumption for obfuscation is described, it is that if you fix any subset of your inputs, at attacker can't tell which subset except by inferring it from the circuit's output
14:39 < andytoshi> so if your input is a conscious circuit, which you are inputting to a universal circuit…
14:40 < andytoshi> …then you shouldn't be able to distinguish the conscious circuit with one where you've replaced any subset of its code with an obviously non-conscious functionally identical code
14:40 < QuantumG> I'm pretty sure I could design an algorithm to predict the distance of the nearest bong from an in-use keyboard from its output.
14:41 < kanzure> if it's functionally identical then what difference does conscious and non-conscious confer
14:41 < andytoshi> none, except to the conscious person itself, which has an 'inside view' which does not appear in its output
14:42 < kanzure> many systems (other than people) have internal state
14:43 < andytoshi> sure, and obfuscation makes this state inaccessible (that is, indistinguishable from random) except where it appears in the program's output
14:45 < andytoshi> so it seems to me that this is a construction of a p-zombie, it'll talk about consciousness just like a normal person but there's nothing but noise happening inside
14:46 -!- sheena [~home@d108-180-136-217.bchsia.telus.net] has joined ##hplusroadmap
14:46 < kanzure> actual noise would, to me, suggest that all output would be noise or random, but clearly you could have zombies or non-human systems that are not producing random output but instead text without having a completely-noisy internal system
14:48 -!- yorick [~yorick@oftn/member/yorick] has quit [Remote host closed the connection]
14:50 < andytoshi> well, in e.g. shamir secret sharing https://en.wikipedia.org/wiki/Shamir%27s_Secret_Sharing you can encrypt some data with a secret, and the encrypted data can be anything since the secret can be anything. then you split the secret up among N parties, and if M of them combine their pieces they can reconstruct the secret
14:51 < andytoshi> but even with (M-1) people participating the secret can still be anything, no information is gained
14:52 < andytoshi> so the secret is completely unknown and the encrypted text will be 'noise' in the sense that you can't distinguish it from random, until a critical mass of participants get togethr
14:53 < pasky> and what does that have to do with consciousness?
14:54 < andytoshi> pasky: because you can use the construction in http://eprint.iacr.org/2013/451 to turn a consciousness into noise, throw away the secret data, and it'll still appear to all outside observers to be conscious because it'll act exactly the same way
14:54 < pasky> but why would I do that?
14:54 < pasky> maybe I should've read more backlog than up to 23:45
14:54 < andytoshi> it's a thought experiment, you do it to learn about the nature of consciousness
14:55 < pasky> I guess that's the point I'm missing, what does this have to do with the nature of consciousness
14:55 < pasky> I can do this with an ELF binary
14:55 < pasky> will that tell me something about the nature of the ELF binary?
14:56 < andytoshi> pasky: sure, and if you do it to an ELF binary which, say, decrypts some data by using a secret key, the obfuscated ELF binary will still decrypt the data but you won't be able to get the key from it in any way
14:56 < andytoshi> so you haven't changed its functional behaviour at all but you have changed the way that it achieves that behaviour, to a way that you can't break apart or partially evaluate
14:57 < kanzure> and your point about consciousness is then something about andytoshi_neurons.exe..?
14:57 < andytoshi> ya
14:57 < nmz787_i> kanzure: any preference for XML to python dict lib? I like the sound of https://github.com/martinblech/xmltodict and maybe http://www.picklingtools.com/html/xmldoc.html
14:57 < andytoshi> if that program had an inner view of its workings, that inner view is destroyed by the obfuscation, but it will -still talk- about having an inner view
14:57 < kanzure> nmz787_i: probably lxml
14:58 < QuantumG> I don't think it is
14:58 < pasky> andytoshi: will it?
14:58 < kanzure> inner view meaning, access to ram whil erunning?
14:58 < kanzure> *while
14:58 < pasky> andytoshi: let's say it's implemented in a language that supports introspectoin
14:58 < andytoshi> kanzure: inner view meaning, whatever it is that people refer to by qualia
14:58 < pasky> (I think all programs using introspection are self-aware to a degree ;)
14:59 < pasky> andytoshi: if it's still performing the expected function, I think the view via introspection should still be the same, right?
14:59 < pasky> or I'm not following your argument
14:59 < nmz787_i> kanzure: i feel like the output of that is quite ugly compare to the two libs i sent
14:59 < andytoshi> pasky: if you can access the program's introspection by giving it appropriate input, it'll still give exactly the same output
14:59 < kanzure> nmz787_i: you can use python -mjson.tool to make json less ugly?
14:59 < QuantumG> obfuscation is defined as changing a program so it is hard for someone trying to read the code to understand, *without* changing the function of the program. If "reflection" is a required function of the program, obfuscation can't change that, or it's not obfuscation.
14:59 < andytoshi> but it'll be lying, just like a p-zombie is lying when it says that it's conscious
15:00 < andytoshi> QuantumG: there is a definition of obfuscation in the paper i linked
15:00 < QuantumG> I don't care about your paper
15:00 < kanzure> ouch
15:00 < andytoshi> lol
15:00 < andytoshi> QuantumG: it means given two functionally identical circuits, a computationally bounded attacker cannot distinguish between their obfuscations
15:01 < QuantumG> having worked on software obfuscation for decades, I think I know a little better than some paper you found on the internet
15:01 < pasky> andytoshi: okay, in that case, I don't think it actually *matters*; the program thinks it's working in a particular way, everyone else thinks it's working in a particular way, it is *working*, so why care?
15:02 < pasky> maybe I'm thinking by my guts instead of my brain, I don't care as long as I continue thinking
15:02 < kanzure> brainguts
15:03 < andytoshi> QuantumG: having actually worked on that paper, i think i know a little better than you what i'm talking about
15:03 < QuantumG> the whole p-zombie nonsense is supposed to make you question the external observation equivalence hypothesis.
15:03 < kanzure> QuantumG: to be fair, i haven't seen anything about circuit analysis from you w.r.t obfuscation, but i have seen various reverse engineering things.
15:04 < QuantumG> andytoshi: great, did you actually ask anyone who works on program obfuscation before you defined the word?
15:04 < kanzure> QuantumG: http://diyhpl.us/~bryan/papers2/paperbot/Candidate%20Indistinguishability%20Obfuscation%20and%20Functional%20Encryption%20for%20all%20circuits.pdf
15:04 < QuantumG> cause, ya know, I can see where your definition is going but it misses some important concepts.
15:04 < pasky> I'm afraid UNIX people never talked to plumbers about defining "pipelines" :(
15:05 < andytoshi> QuantumG: please elaborate
15:05 < QuantumG> anyway, if you want to talk about *your* definition of obfuscation, that's cool.. I just like to know your internal model so I can understand your outputs :)
15:05 < andytoshi> :P
15:05 < nmz787_i> QuantumG: they like to also call that minification
15:05 < nmz787_i> though i guess that might be less obfuscating than a purpose built obfucator
15:05 < andytoshi> QuantumG: i'm curious if you think there's a stronger definition of obfuscation we could be using tho
15:05 < andytoshi> which is not subject to impossibility results
15:06 < QuantumG> you have to define "equivalent programs"
15:06 < QuantumG> (or circuits, no idea why you went with that terminology)
15:06 < nmz787_i> it's all circuits!
15:06 < kanzure> a program is just a graph
15:06 < kanzure> executable graph thing
15:06 < kanzure> directed, probably
15:06 < nmz787_i> you could probably say a circuit is just a graph
15:07 < kanzure> it is
15:07 < QuantumG> it's all just lambda calculus man.. Church-Turing thesis.. where's that bong?
15:07 < nmz787_i> everything is everything
15:07 < nmz787_i> the weather here is crazy
15:07 < kanzure> now you're going too far
15:07 < andytoshi> QuantumG: "equivalent" means identical input/output pairings, i.e. functionally equivalent
15:08 < nmz787_i> backdrop of dark clouds, sunshine in the foreground.... a few minutes ago I saw thunder and it was hailing
15:08 < kanzure> it's very common to study electrical circuits as circuits and graphs, that's not "everything is everything"
15:08 < QuantumG> right, so buried in your definition of equivalent is the answer to your p-zombie conundrum.
15:09 < kanzure> QuantumG: did i show you https://github.com/kanzure/pokecrystal or https://github.com/iimarckus/pokered
15:09 < QuantumG> it goes away as soon as you define your terms, which is why it doesn't happen for philosophers because they refuse to define consciousness so it can yield to rational argument.
15:10 < kanzure> QuantumG: pokered.git contains source code that completely compiles into the original pokemon red game
15:10 < kanzure> (as in, byte-equivalent)
15:10 < QuantumG> I think I saw a youtube video you did in regards to it
15:10 < kanzure> QuantumG: i also did a bunch of asm injection and hacking-by-gadgets https://github.com/kanzure/pokemon-reverse-engineering-tools/blob/vba-automation/pokemontools/vba/vba.py#L206
15:11 < kanzure> QuantumG: like using known asm to string together malicious programs
15:11 < kanzure> known-asm inside the original rom
15:12 < kanzure> thought you might find that interesting, i forgot about your reverse engineering stuff
15:13 < andytoshi> QuantumG: consciousness is detectable in a brain by looking at the neural activity, but not by, say, talking to someone on IRC. but you should be able to upload a consciousness, define the inputs and outputs entirely through IRC, and still have it be conscious even though you can't detect consciousness over IRC. but then the obfuscation will frag anything which isn't detectable over IRC, creating an
15:13 < andytoshi> IRC zombie
15:13 < QuantumG> I'm not as *keen* on it as I used to be.
15:14 < kanzure> consciousness is not detectable by looking at neural activity
15:14 < andytoshi> this should be true for any definition of consciousness which isn't IRC-detectable, so the philosphers have plenty of room to faff
15:14 -!- ebowden [~ebowden@CPE-60-231-177-134.lns4.dav.bigpond.net.au] has joined ##hplusroadmap
15:15 < kanzure> i don't know if there is any form of anything that is not IRC-detectable. you can encode a bunch of information into text..
15:15 < QuantumG> either consciousness is important for external behavior, in which case a p-zombie is impossible, or it isn't, in which case a p-zombie is irrelevant.
15:15 < QuantumG> whether a p-zombie is impossible or irrelevant is not worth answering.
15:17 < QuantumG> let's replace the word "conscious" with "drunk". If you can't tell that I'm doing my job drunk, why would you care?
15:17 < kanzure> drunk has a bunch of specific definitions, unfortunately
15:17 < kanzure> like: breathalyzer results, even if it doesn't say anything about your brain-state
15:18 < kanzure> and you can pretend that the ability of your boss to smell your breath is a crude breathalyzer
15:18 < QuantumG> I say the debate over breathalyzer vs walk-the-line/say-the-alphabet-backwards testing is identical to the debate over p-zombies
15:18 < pasky> I completely agree with general QuantumG's argument though
15:20 -!- Adifex|zzz is now known as Adifex
15:21 < QuantumG> imagine you had a "consciousness breathalyzer", set up roadblocks and did random testing. The people who failed the test would be like "dude, I'm totally conscious!" too
15:23 < QuantumG> if all other external testing failed to detect the lack of consciousness, you'd say the person was a p-zombie, then let them go home because, by definition, they're no danger to themselves or others.
15:24 < QuantumG> so what's the point of this gadget? Finding real zombies or something? p-zombies are just the false positives?
15:25 < QuantumG> I think the real purpose of p-zombies is to demonstrate that anyone who uses the word "consciousness" without a sneer probably has a bong nearby.
15:26 < ebowden> LOL
15:27 < kanzure> ebowden: you can ask sheena about dog things, by the way
15:27 < QuantumG> wtf, google just sent me $120?
15:27 < QuantumG> okay, thanks google
15:27 < kanzure> last december google reversed 3-4 years of adsense deposits
15:27 < kanzure> like the entire amounts that people spent
15:28 < kanzure> so if you spent $60k on ads, you had +$60k
15:28 < kanzure> and then they turned around to reverse the transactions, taking $60k back... but as you can imagine, this causes problems because of exchange rate fluctuation, etc
15:28 < ebowden> Oh, ok, thanks kanzure.
15:28 < QuantumG> fun
15:30 < QuantumG> I think I set the $200 threshold back in 2007.
15:30 < QuantumG> so yeah, 7 years of earnings.. woooo..
15:30 < nmz787_i> i wonder if anyone caught it in between and closed their bank account before google could re-reverse it
15:30 < kanzure> i was surprised that the news wasn't all over it
15:31 < kanzure> i guess nobody was surprised that the banking infrastructure is totally fucked up
15:31 < QuantumG> they're forcing everyone in Australia to put a pin number on their credit card next year
15:31 < QuantumG> I've never had one, because the concept of a cash advance on a credit card disturbs me.
15:32 < ebowden> Oh, sheena, are you there?
15:32 < QuantumG> and if someone forges your signature, you're never liable for their purchases (which isn't the case for a pin)
15:32 < kanzure> i believe you're liable if the 45-90 day window expires
15:33 < QuantumG> if you have "fraud protection" you're not even liable then, as they're required to detect it and contact you.
15:34 < kanzure> s/liable/not able to get money bcak
15:34 < kanzure> *back
15:34 < pasky> here in czech republic, all cards, credit or debit, use pins and signatures are used only in very exceptional cases
15:34 < QuantumG> I think you mean "not required to repay it", as you never had it in the first place
15:35 < QuantumG> pasky: they probably require chip and pin and do bulk processing.
15:35 < QuantumG> ?
15:35 < pasky> yeah they are all chip cards
15:35 < pasky> hmm, i think i had a card about 9 years ago which would trigger using a signature instead of pin on some gas stations
15:35 < pasky> because they had offline terminals
15:36 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds]
15:36 < pasky> but you basically never see that anymore with czech-issued cards in cz
15:36 < QuantumG> I'm figuring I'm going to get rid of my credit cards anyway.. I'm just paying for the privilege of having them.
15:36 < QuantumG> for online purchases I'll just buy a prepaid or refillable card
15:38 < QuantumG> can't be more fucked than using bitcoin
15:41 < QuantumG> http://forum.nasaspaceflight.com/index.php?topic=34532.msg0 hundreds of pictures of Dragon.. cause it's not exactly the same as the last time I saw these pictures.
15:45 -!- ebowden [~ebowden@CPE-60-231-177-134.lns4.dav.bigpond.net.au] has quit [Remote host closed the connection]
15:46 -!- ebowden [~ebowden@CPE-60-231-177-134.lns4.dav.bigpond.net.au] has joined ##hplusroadmap
15:54 -!- eudoxia [~eudoxia@r186-55-41-228.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap
16:06 -!- nmz787_i [nmccorkx@nat/intel/x-daakmmcjfgcpxyku] has quit [Quit: Leaving.]
16:30 -!- kuldeepdhaka [~kuldeepdh@unaffiliated/kuldeepdhaka] has quit [Quit: Leaving]
16:45 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap
16:51 < sheena> ebowden: am now
16:51 < ebowden> Ah, ok.
16:52 < sheena> dogs?
16:52 < QuantumG> NO DOGS NO DOGS NO DOGS NO DOGS NO DOGS NO DOGS NO DOGS
16:52 < ebowden> So, what do you think of genetically engineering dogs to have higher intellect, and have linguistic abilities?
16:55 < sheena> i think its theorhetically plausible
16:55 < eudoxia> i can already see the vice headline
16:55 < sheena> what would you use the dogs for?
16:55 < sheena> and what would you do with all the 'almost useful but too smart for homes' ones? large kenneling facility?
16:55 < eudoxia> "A geneticist, a bioinformaticist, and a web developer teamed up with furaffinity. You won't believe what happened next"
16:56 < ebowden> LOL
16:56 < ebowden> Sheena, no, for wacky talking dog adventures.
16:56 < sheena> lol what?
16:56 < eudoxia> suuuuuuuureeeeeeeeee
16:58 < eudoxia> it would be cool to do something like upload a doge
16:58 < eudoxia> and then increase the intelligence of the upload
16:58 < eudoxia> that would really be HN front page material
17:00 < cpopell> ebowden: why the fixation in uplifting dogs
17:00 < ebowden> Mainly the idea of wacky talking dog adventures.
17:01 < cpopell> Do you consider this seriously?
17:01 < ebowden> Also, they are better at understanding people than chimps.
17:01 < jrayhawk> counterpoint: Mr. Peabody is a dick
17:01 < sheena> can you explain a wacky talking dog adventure example?
17:02 < sheena> they are certainly better at some human-interaction interpretation than chimps
17:02 < ebowden> We go around the world solving mysteries together.
17:02 < ebowden> :D
17:02 < eudoxia> where in the world is satoshi nakamoto
17:02 < eudoxia> woof woof
17:02 < kanzure> nose work might count
17:02 < ebowden> cpopell, no, dogs are just very good candidates for it.
17:03 < ebowden> (For genetically engineering in linguistic abilities.
17:03 < ebowden> )
17:03 < sheena> so you want a dog with dog senses, but who can comunicate linguistically?
17:04 < ebowden> Well the wacky adventures and mystery solving is mainly a joke, but...
17:05 < ebowden> It would be nice to see it happen.
17:05 < eudoxia> it might be easier to surgically shape a human into a doge, then somehow give them a doge's personality and cheerful demeanor
17:07 -!- HEx1 [~HEx@hexwab.plus.com] has joined ##hplusroadmap
17:07 < ebowden> It would also be incredibly creepy.
17:07 -!- HEx2 [~HEx@hexwab.plus.com] has quit [Read error: Connection reset by peer]
17:07 < kanzure> let's keep to only credibly creepy
17:07 < sheena> so i think dogs have some language already, that we mostly dont tap into
17:07 < sheena> did kanzure send you the lnk i sent him a fwe days ago about this?
17:08 < jrayhawk> the language of butt sniffing
17:08 < kanzure> yes i did
17:08 < QuantumG> if the result is functionally equivalent, what's it matter? I think you want some sort of p-zombie dog.
17:08 < kanzure> http://www.amazon.com/1000m-Dyneema-Strong-Braided-Fishing/dp/B009665S24/ref=sr_1_3?ie=UTF8&qid=1398193369&sr=8-3&keywords=dyneema
17:08 < kanzure> wait, no
17:08 < kanzure> http://www.amazon.com/Dogs-Can-Sign-Too-Breakthrough/dp/1587613530
17:10 < sheena> yeah, that one
17:10 < sheena> whats your experience with genetic engineering?
17:11 < kanzure> i don't think he has any, but some others in here do (meeeee)
17:11 < ebowden> kanzure, oh?
17:11 < ebowden> :D
17:12 < ebowden> Can you tell me about it?
17:12 < kanzure> what, you think this channel is just full of people talking about news?
17:12 < ebowden> No, I'm just curious.
17:12 < kanzure> what do you want to know
17:13 < ebowden> Well, what kind of stuff did you work on?
17:13 < ebowden> Knockout mice? Crops?
17:14 < sheena> ebowden: selective breeding has worked pretty well in creating crazy variety in dogs..
17:15 < ebowden> Yeah, it has.
17:15 < ebowden> But we are yet to get ones that talk and solve mysteries.
17:16 < ebowden> Kanzure?
17:16 < kanzure> is it really important that you know the exact specimens i've worked with? the protocols are not very hard to learn if you've done one, you can usually muddle your way through the next.
17:16 < ebowden> Ah, ok.
17:16 < ebowden> Well, no.
17:17 < ebowden> Just curious what kind of stuff you got to do.
17:17 < kanzure> i do whatever i want
17:17 < ebowden> Oh?
17:17 < kanzure> well why shouldn't i?
17:17 < ebowden> You do whatever you want, in regards to genetic engineering?
17:17 < ebowden> :D
17:17 < kanzure> you are a very confusing person
17:18 < ebowden> (There is absolutely no reason I can think of for you not doing what you want in regards to genetic engineering, kanzure.)
17:18 < andytoshi> i bet he's an ELIZA… must have heard my earlier comments about consciousness not being irc-detectable
17:19 < kanzure> must be
17:19 < sheena> ebowden: i'm not sure anyone is breeding specificaly for language and mystery solving
17:19 < sheena> though dogs are pretty good at problem solving, depending what sorts of mysteries you have to solve!
17:19 < kanzure> i'm sure they are breeding for smell mystery solving
17:19 < kanzure> mystery smell i mean
17:20 < ebowden> Sheena, basically fast track the process.
17:20 < kanzure> is that a command?
17:20 < eudoxia> 'here dog, smell this mailing list archive'
17:20 < eudoxia> 'woof woof'
17:21 < ebowden> (Oh, also kanzure, all the ones I could think of would be very, very stupid recycled anti-GMO arguments.)
17:21 < kanzure> no, you said "got to do"
17:21 < kanzure> it's not about "getting to do" anything, it's about what you happen to do..
17:21 < ebowden> Kanzure, no, not a command.
17:21 < ebowden> Just explaining.
17:21 -!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has quit [Ping timeout: 265 seconds]
17:21 < kanzure> i still don't understand then. "basically fast track the process" is an explanation for what?
17:22 < sheena> ebowden: fast track the process of developing a dog with language skills, using genetic engineering?
17:22 < ebowden> I was explaining to sheena.
17:22 < sheena> i'm not saying it wouldnt work.. but who would fund it?
17:22 < sheena> ebowden: just ignore kanzure ;)
17:22 < ebowden> And yes, sheena gets it.
17:23 < ebowden> Ah, ok.
17:23 < ebowden> sheena: I can dream.
17:24 < sheena> lol ok
17:24 < sheena> i just figure if you dont have a practical application in mind, funding will be difficult
17:24 < kanzure> gene reductionism is a common problem in biology where people think that individual traits must reduce to a single gene
17:24 < sheena> shit im late. sorry. ttyl
17:24 < sheena> hdsheena@gmail.com if you have more thoughts, ebowden
17:24 < ebowden> sheena, I can think of a group of people who would be VERY willing to fund it...
17:24 < ebowden> :D
17:24 < jrayhawk> gene determinism is also awful
17:24 < ebowden> Oh, ok sheena.
17:25 < kanzure> funding isn't so much the problem as much as "the idea is probably wrong in a number of ways"
17:25 < kanzure> (like in the ways that matter if you want to make something happen)
17:25 < ebowden> Oh?
17:25 < kanzure> what do you mean "Oh?" after i explained it and jrayhawk too
17:26 < ebowden> Well, never mind, I'll take your word for it.
17:26 < kanzure> that's even worse!
17:26 < ebowden> I know. :D
17:27 < jrayhawk> so, for reference, we didn't know there were six common DNA nucleotides until 18 months ago. The amount of shit we're clueless about is enormous.
17:27 < kanzure> bah jrayhawk, don't bring up unknown unknowns, that's not a legitimate argument
17:27 < kanzure> i was expecting you to paste your it's-not-quite-like-software quote
17:27 < jrayhawk> well, that'd work, too
17:27 < kanzure> which would seem more informative for ebowden's case
17:28 < kanzure> i suspect he strongly believes that genes directly control human-documented features in a very one-to-one way
17:28 -!- ElixirVitae [~Shehrazad@95.5.71.53] has joined ##hplusroadmap
17:28 -!- ElixirVitae [~Shehrazad@95.5.71.53] has quit [Changing host]
17:28 -!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has joined ##hplusroadmap
17:28 < ebowden> kanzure, what you do you mean by one-to-one way?
17:29 < kanzure> as opposed to one-to-many or many-to-many
17:29 < ebowden> Oh, no, I don't think that/
17:29 < ebowden> Oh, no, I don't think that.
17:29 < andytoshi> jrayhawk: link for 6 necleotides?
17:30 < kanzure> there's also a bunch of synthetic/artificial nucleotides that labs have been using
17:30 < ebowden> I know genes don't control one discreet trait.
17:30 < kanzure> in rare circumstances they do, but it's not like machine code.
17:30 < ebowden> I know that, only creationists try to pretend it is.
17:31 < kanzure> uh?
17:31 < ebowden> Kanzure, what's this about new nucleotides?
17:31 < kanzure> there are many people who have misunderstandings about biology other than creationists
17:32 < ebowden> Oh, creationist arguments involve saying that DNA is exactly like a hard drive.
17:32 < ebowden> Oh, some creationist arguments involve saying that DNA is exactly like a hard drive.
17:32 < ebowden> Yes, that's true.
17:32 < jrayhawk> andytoshi: 5-hydroxymethylcytosine is the latest one
17:32 < andytoshi> ebowden: you sholud read 'darwin's dangerous idea', it has a high-level overview of some of this stuff and a shitload of citations
17:33 < ebowden> Thanks andytoshi.
17:33 < andytoshi> jrayhawk: we don't bother with simple names anymore? ;)
17:33 < kanzure> shrug, the simple name is "the second cytosine" or something
17:33 < QuantumG> nah, creationists are all about the watchmaker analogy
17:33 < ebowden> They do use the other one as well.
17:33 < QuantumG> apparently the operation of cells is so complex and beautiful that it must have been created by an intelligence
17:33 < ebowden> LOL
17:34 < QuantumG> clearly, they've never learnt a damn thing about how cells work, or any sort of engineering either
17:34 < ebowden> Yea, the components of clocks do not have natural affinities to each other.
17:34 < ebowden> jrayhawk, can you explain this novel nucleotide thing to me?
17:34 < QuantumG> cause if an engineer made the crap we've found in cells we'd string him up
17:34 < kanzure> it just means that some molecules of dna have different nucleotides
17:34 < kanzure> cell metabolism to some extent determines the available molecules to build more dna
17:35 < ebowden> QuantumG LOL
17:35 < kanzure> and sometimes this means less "normal" nucleotide-equivalent options are available
17:35 < ebowden> Ok.
17:35 < kanzure> or, that you can get stuck in a local minima on a loaf of bread in a fridge and maybe those cells that don't require the unavailable nucleotides are able to build more of themselves more than the other cells
17:36 < ebowden> kanzure, have you got a link where I can learn about this?
17:36 < kanzure> http://diyhpl.us/~bryan/papers2/DNA/
17:36 < ebowden> Ah, ok.
17:36 < ebowden> Thankyou kanzure.
17:37 < QuantumG> On the other hand, you can kinda see where creationists get this watchmaker analogy from if they've ever looked at a molecular biology textbook
17:37 < QuantumG> molecular biologists just love making models of how cellular functions work that are all clean and engineered looking
17:37 < ebowden> Still, the stupid hurts.
17:37 < ebowden> Yeah.
17:38 < kanzure> except they aren't actually clean
17:38 < jrayhawk> ebowden: DNA is dynamically configurable; specifically cytosine can be swapped out for 5-methylcytosine and 5-hydroxymethylcytosine. These are currently conceptualized as 'speedbumps' in the transcription process that determine expression, but biology is always more complicated than we suspect.
17:38 < andytoshi> i also think if you've never coded you'll be more susceptible to creationism, even hume said he couldn't imagine where all this complexit came from
17:38 -!- nmz787_i [~nmccorkx@192.55.54.42] has joined ##hplusroadmap
17:38 < ebowden> jrayhawk, yeah.
17:38 < andytoshi> and he was a massive 'fuck the pope' atheist, if he were alive today he'd literally starve he'd spend so much time on reddit
17:39 < QuantumG> heh
17:39 < jrayhawk> s/determine expression/\0 of a particular subsequent gene/
17:39 < kanzure> "... there is no source, the bytecode has multiple reentrent abstractions, is unstable and has a very low signal to noise ratio, the runtime is unbootstrappable, the execution is nondeterministic, it tries to randomly integrate and execute code from other computers... multiple reentrant and self-modifying abstractions. absolutely everything has subtle side effects."
17:40 < kanzure> bitcoin is biology
17:40 < jrayhawk> i wonder how much of that quote i stole from matthew garrett
17:41 < kanzure> kill him and claim it for yourself
17:41 < jrayhawk> he's fueled by booze and rage
17:41 < ebowden> LOL
17:41 < jrayhawk> all i have is bacon and deontardation
17:42 < QuantumG> any good papers on protein engineering as a pathway to productive nanotechnology in the last few years kanzure?
17:42 < jrayhawk> http://mjg59.dreamwidth.org/6376.html ah, maybe half of it
17:42 < kanzure> only moderatively productive stuff is the dna origami things from winfree's lab, but you might be bored by that by now
17:42 < kanzure> as for protein engineering, there's a lot of known motifs
17:43 < kanzure> here's some papers about engineering just polymerase enzymes: http://diyhpl.us/~bryan/papers2/polymerase/
17:43 < kanzure> this paper was also fun: http://diyhpl.us/~bryan/papers2/bio/protein-engineering/Design%20of%20protein%20function%20leaps%20by%20directed%20domain%20interface%20evolution.pdf
17:43 < nmz787_i> kanzure, walnut creek transcript link? this dude is saying they said (sometime, didn't specify where/when) that they said " diybio movement would cross a 'red line' for them if it had cheap
17:43 < nmz787_i> > independent dna synthesis technology"
17:43 < ebowden> Oh, jrayhawk, do you think the effects of the active compounds in lion's mane mushrooms are anything special?
17:43 < nmz787_i> i remember them blowing off that question
17:43 < nmz787_i> but maybe something happened since then
17:43 < kanzure> nmz787_i: git clone git://diyhpl.us/diyhpluswiki.git
17:44 < kanzure> nmz787_i: http://diyhpl.us/wiki/transcripts/fbi-diybio-2012/howard-simon-dna20.txt
17:44 < nmz787_i> s/happened/was said/
17:44 < QuantumG> read any of these? http://mechanosynthesis.mit.edu/?page_id=11
17:44 < jrayhawk> http://www.ncbi.nlm.nih.gov/pubmed/18844328 http://www.ncbi.nlm.nih.gov/pubmed/20834180 http://journals.prous.com/journals/servlet/xmlxsl/pk_journals.xml_summary_pr?p_JournalId=2&p_RefId=1173290&p_IsPs=N http://www.ncbi.nlm.nih.gov/pubmed/20834180 yes, though a lot of brain research is hampered by a terrible null hypothesis
17:44 < paperbot> ConnectionError: [Errno 111] Connection refused (file "/usr/local/lib/python2.7/dist-packages/requests/models.py", line 625, in send)
17:45 < kanzure> nope, but this one seems fun: "Microscale machining and nanoscale surface modification of carbon nanotube forests by ultrafast laser irradiation."
17:45 < kanzure> jrayhawk: what terrible null hypothesis in those? or at least, not those, but the generally-terrible one you are expecting
17:45 < nmz787_i> hmm, I don't see my question in there
17:45 -!- yashgaroth [~ffffff@cpe-76-167-105-53.san.res.rr.com] has joined ##hplusroadmap
17:45 < kanzure> me either, search the others
17:46 < jrayhawk> american test subjects
17:46 < ebowden> jrayhawk, thankyou.
17:46 < kanzure> "red line" doesn't show up in the transcripts
17:46 < kanzure> it's possible that i was pooping
17:46 < kanzure> and not typing
17:47 < jrayhawk> or, rather, test subjects with atrociously inflammatory diets
17:47 < kanzure> are these transcripts even useful? should i bother
17:47 < kanzure> because there's probably some non-negligble damage that i'm doing to myself
17:49 < jrayhawk> pretty much anything with any immune modulating activity whatsoever, including cortisol reduction from the placebo effect, will have "special" effects on diseases of affluence.
17:49 < jrayhawk> lion's mane mushroom could be bioaccumulating lithium for all i know
17:49 < kanzure> "Robofurnace: A semi-automated laboratory CVD system for high-throughput nanomaterial synthesis and process discovery."
17:50 < ebowden> jrayhawk, do you think we should be testing just the isolated active compounds?
17:51 < ebowden> I do.
17:51 < jrayhawk> no, the problem is the null hypothesis
17:51 < ebowden> Well, if they really are the active compounds.
17:51 < ebowden> What is wrong with it?
17:52 < jrayhawk> normalizing on pathology results in conclusions largely tangential to health
17:52 < ebowden> ok.
17:53 < jrayhawk> Take any single RCT on palmitic acid you like; it will demonstrate LDL receptor downregulation and resultant inflammation and oxidative stress
17:53 < jrayhawk> s/single/single-variable/
17:54 < jrayhawk> Take any single-variable RCT on arachidonic acid you like; it will demonstrate increased imbalanced eicosanoid production resulting in inflammation and oxidative stress
17:54 < ebowden> Right.
17:55 -!- eudoxia [~eudoxia@r186-55-41-228.dialup.adsl.anteldata.net.uy] has quit [Quit: leaving]
17:55 < jrayhawk> Take any single-variable RCT on methionine you like; it will demonstrate reduced glutathione production resulting in oxidative stress and inflammation
17:56 < kanzure> jrayhawk: do you ever look at or use the kebb metabolism image database thing?
17:56 < kanzure> kegg, i mean
17:56 < kanzure> http://www.genome.jp/dbget-bin/www_bfind_sub?mode=bfind&max_hit=1000&locale=en&serv=kegg&dbkey=kegg&keywords=glycogen&page=1
17:57 < jrayhawk> Meanwhile all of the conclusions you derive from those RCTs are actually moving you further away from health because they only appear pathological *in the context of existing pathologies*
17:57 < kanzure> the one you are probably most interested in is 'KEGG REACTION'
17:58 < kanzure> oh wait, or is it kegg pathway?
17:58 < kanzure> http://www.genome.jp/kegg/reaction/
17:58 < kanzure> http://www.genome.jp/kegg/pathway.html
17:58 < kanzure> "EGG REACTION contains all reactions taken from KEGG ENZYME and additional reactions taken from the metabolic pathway maps in KEGG PATHWAY. Each reaction is identified by the R number, such as R00259 for the acetylation of L-glutamate. Reactions are linked to ortholog groups of enzymes as defined by the KEGG ORTHOLOGY database, enabling integrated analysis genomic (enzyme genes) and chemical (compound pairs) information."
18:00 < nmz787_i> ugh, i can't figure out how to set git to go through the proxy
18:00 < kanzure> PROXY="0.0.0.0:80" git fetch
18:00 < jrayhawk> man git-config
18:00 < jrayhawk> man curl
18:00 < jrayhawk> either one will do
18:01 < kanzure> i'm curious if anyone has done food studies on this data set
18:01 < kanzure> you can probably get your SNPs aligned to this data and then see which pathways are either working or not working for you
18:01 < nmz787_i> is diyhpl.us http or https?
18:02 < kanzure> and then based on simple models of food in terms of their macronutrient or micronutrient properties, you could determine basic metabolic rates and yields for all the reactions, or absorption rates, etc.
18:02 < kanzure> oh i guess permeability is the blocking issue there
18:02 < jrayhawk> https://secure.diyhpl.us/cgit/diyhpluswiki/
18:02 < jrayhawk> this contains the clone urls
18:02 < jrayhawk> it is not itself a clone url
18:03 < kanzure> maybe permeability doesn't matter if you do poop studies
18:03 < kanzure> anything not there was absorbed
18:04 < jrayhawk> the http transport is pretty dumb and I don't feel like supporting it.
18:04 < kanzure> hrm but then you have to account for microbial content of gut
18:05 < nmz787_i> so what is the git transport?
18:05 < nmz787_i> i used git config --global http.proxy http://server:portnum
18:06 < kanzure> that looks like http to me
18:07 < kanzure> why are you setting up a proxy anyway
18:07 < jrayhawk> intel, presumably
18:07 < nmz787_i> I'd only grep the dir for synthesi
18:07 < nmz787_i> rather that's all i want to do, i can read the files over http
18:08 < kanzure> you have an account on the server, you can just ssh in
18:08 < nmz787_i> heh, you think so?
18:08 < kanzure> what sort of fucked up work environment disables git clone?
18:08 < nmz787_i> and ssh
18:08 < kanzure> you should quit
18:08 < nmz787_i> lol
18:08 < nmz787_i> na
18:08 < nmz787_i> i like the cafe too much
18:08 < kanzure> no drugs AND no ssh? come on.
18:09 < nmz787_i> plus i work on the same floor as my gf
18:09 < nmz787_i> the hiring lady told me they stopped testing actually
18:09 < jrayhawk> do people get fired for using corkscrew
18:10 < jrayhawk> http://en.wikipedia.org/wiki/Corkscrew_(program)
18:10 < nmz787_i> i had tried something last month, idk if it was that
18:10 < nmz787_i> something with a cat in it
18:10 < nmz787_i> nmap
18:11 < nmz787_i> and it didn't work
18:11 < jrayhawk> ... "didn't work"
18:11 < nmz787_i> but i think the lingo was something like 'don't do anything that could compromise our network'
18:11 < nmz787_i> can't remember if the word tunnel was used
18:12 -!- ebowden [~ebowden@CPE-60-231-177-134.lns4.dav.bigpond.net.au] has quit [Read error: Connection reset by peer]
18:12 < jrayhawk> it would be a dumb use regardless; SSL is a tunnel.
18:12 < dingo> jrayhawk: i know of aa bunch of solaris sysadmins who got fired from A.G. Edwards for tunneling out
18:13 < dingo> but it was part of a massive layoff, and they just dug up a reason to fire them without pay
18:13 -!- ebowden [~ebowden@CPE-60-231-177-134.lns4.dav.bigpond.net.au] has joined ##hplusroadmap
18:13 < dingo> i've tunneled out of "secure research facilities", the kind that don't let you bring in a phone if it has a camera on it, and search your waistline and shoes and shit
18:13 < dingo> sysadmins are notoriously inept
18:13 < dingo> if they weren't, they'd be developers
18:14 < jrayhawk> It'd be pretty easy to notice who makes long-lived SSL connections.
18:14 < dingo> yeah i thought so too, i was waiting to get walked out
18:14 < dingo> its easy to fix anyway
18:15 < jrayhawk> I don't think network-wide MITMing counts as "easy", but that seems to be the normal approach these days.
18:15 < dingo> naw i mean just, you could dhcp new leases often, break the connection and reconnect every 15 minutes, whatever
18:15 < QuantumG> on the other hand, I know people who work in ordinary offices, have unlimited free access to facebook on their phone and still complain about IT blocking it.. and IT still block it, even though everyone has it on their phone. Common sense aint common.
18:15 < nmz787_i> couldn't you just break up the sessions?
18:15 < nmz787_i> oh, what dingo said
18:15 < dingo> proxy it through a dedicated server, like a developer's environment, where it looks purposeful
18:16 < dingo> thats what i did at my last gig
18:16 < nmz787_i> google app engine
18:16 < nmz787_i> lol
18:16 < jrayhawk> Oh, yeah.
18:18 < kanzure> jrayhawk: your task is to read the entirety of kegg
18:18 < jrayhawk> oh god
18:18 < kanzure> start here http://www.genome.jp/dbget-bin/www_bget?rn:R03024
18:19 < jrayhawk> it's more tempting than it should be to try to convert it all to ikiwiki's linkmapped graphviz plugin
18:19 < jrayhawk> https://ikiwiki.info/plugins/linkmap/ which is badass if you haven't seen it before
18:19 < kanzure> but then you end up with giant pngs
18:19 < nmz787_i> and also the thing patrikd mentioned a few times, some project with a capitalized acronym
18:19 < kanzure> if anything, it should be something other than graphical output
18:19 < kanzure> and don't tell me about graphviz's ascii output
18:19 < nmz787_i> DECODE
18:20 < kanzure> ask cluckj about that one
18:20 < kanzure> cluckj: your fault
18:20 < kanzure> jrayhawk: oh look, boxes with arrows.. great.
18:20 < jrayhawk> and links
18:20 < jrayhawk> don't forget links
18:20 < kanzure> heatmap?
18:20 < kanzure> or svg with links
18:20 < kanzure> hah
18:21 < jrayhawk> a heatmap for what?
18:21 < kanzure> no i was wondering about the implementation
18:21 < kanzure> the old school way, long before i was born, was something like "link to a cgi file, with pixel coordinates"
18:22 < nmz787_i> oh, sorry, wrong way... it's ENCODE https://www.genome.gov/encode/
18:23 < jrayhawk> i'd assumed graphviz's plotters were willing to output coordinates for debugging purposes anyway
18:23 < kanzure> maybe there's some computational biology projects that dump vitamins into these reactions/equations and calculate yields
18:23 < jrayhawk> i think grad students are normally used for that
18:23 < kanzure> that's stupid
18:23 < kanzure> QuantumG: get on it
18:25 < kanzure> paperbot: http://www.ploscompbiol.org/article/info%3Adoi%2F10.1371%2Fjournal.pcbi.1000554
18:25 < paperbot> http://diyhpl.us/~bryan/papers2/paperbot/Nutritional%20Systems%20Biology%20Modeling%3A%20From%20Molecular%20Mechanisms%20to%20Physiology.pdf
18:26 < kanzure> "A recent mathematical model was developed to quantify the factors that determine the proportion of weight loss coming from body fat versus lean tissue. The basis for the model was a classic theory of Gilbert Forbes who hypothesized that longitudinal body composition changes are described by movement along a logarithmic curve relating lean body mass to fat mass [61]."
18:26 < kanzure> oh that's weird. why not not base it on actual biology instead.
18:27 -!- da_shirlz_HBIC_ [uid29645@gateway/web/irccloud.com/x-nrorfxpjlhhekcng] has quit [Quit: Connection closed for inactivity]
18:27 < kanzure> ah this is slightly more sane: "A computational model of macronutrient balance was recently used to integrate the available data on the metabolic changes in patients with cancer cachexia. The resulting computer simulations showed how the known metabolic derangements (e.g., increased proteolysis, lipolysis, and gluconeogenesis) synergize with reduced energy intake to result in a progressive loss of body weight, fat mass, and lean tissue [71]."
18:28 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap
18:28 < kanzure> i see nothing about using the known reactions and yields
18:29 < kanzure> paperbot: http://journals.lww.com/co-clinicalnutrition/Abstract/2008/05000/Computational_modeling_of_cancer_cachexia.5.aspx
18:29 < paperbot> http://diyhpl.us/~bryan/papers2/paperbot/fc6ed9876dded579937cfcd9afd6d77f.txt
18:35 < fenn> good thing they included ™ and ° and • on this keyboard but not up or down arrows
18:37 < fenn> i am chatting today on a e-ink reader with debian and android installed
18:38 < fenn> itsvreally hard to play frozen bubble in monochrome even with colorblind mode on
18:38 < kanzure> fenn: you may find this useful https://play.google.com/store/apps/details?id=com.volosyukivan
18:38 < kanzure> it runs an http server on your android device that serves up a