2013-01-14.log

--- Log opened Mon Jan 14 00:00:30 2013
rigeloh who is @pdx.edu, and should i meet them00:00
-!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap00:01
@kanzurerigel: almost everyone. jrayhawk, nmz787, lichen, mokbor, bioguy, the other jerks.00:01
lichenhmm?00:02
* nsh hips hat at room00:02
@kanzurewell, you are close to pdx.00:02
nshand tips tat00:02
rigelyes, quite close00:02
rigeli go to dorkbot now and again00:03
lichenah, yeah, im in pdx00:03
lichenhello00:03
rigelhi00:03
lichengoing to psu as well, heh00:03
rigeli am up on the hill actually00:04
lichenah, neat00:04
lichenyou at a program at ohsu?00:04
rigelnod00:05
lichenvery cool00:05
lichenim just an undergrad still :p00:05
bkeroSon of a bitch00:05
bkeroMore pdx people?00:05
rigelwell i would be keen to meet up and chat00:05
bkerolichen: what major at psu? I have many friends there.00:05
lichenbiomedical physics00:06
bkeroright on00:06
* bkero lives in PDX, but is currently traveling at the moment. Be back sometime early next month.00:06
rigeli was a bench lab monkey before coming here00:06
lichenhehe00:06
bkerololwetwork00:06
licheni work at a warehouse doing manual labor bullshit00:06
bkeroI did that back in high school. Mini/maxi-preps, PCR, glorified dishwashing.00:06
lichenschool part time for now00:06
lichenalways on the lookout for something more related00:07
@kanzurehttp://www.nature.com/nrn/journal/v13/n12/full/nrn3399.html00:08
paperbotHTTP 401 unauthorized http://www.nature.com/nrn/journal/v13/n12/pdf/nrn3399.pdf00:08
@kanzurewhat00:09
@kanzureWHAT00:09
bkeroyupyup00:10
bkeropaywallol00:10
@kanzurewell paperbot is supposed to have access to that00:11
@kanzurepaperbot: http://www.nature.com/nrn/journal/v13/n12/pdf/nrn3399.pdf00:11
paperboterror: didn't find any pdfs on http://www.nature.com/nrn/journal/v13/n12/pdf/nrn3399.pdf00:11
paperbothttp://diyhpl.us/~bryan/papers2/paperbot/f4612cfb3251a43a730d64439905a0ec00:11
-!- qu-bit_ [~shroedngr@unaffiliated/barriers] has joined ##hplusroadmap00:22
-!- barriers [~barriers@unaffiliated/barriers] has joined ##hplusroadmap00:22
-!- qu-bit [~shroedngr@unaffiliated/barriers] has quit [Ping timeout: 252 seconds]00:23
-!- barriers_ [~barriers@121-73-87-49.cable.telstraclear.net] has quit [Ping timeout: 272 seconds]00:23
@kanzurei think discussing scraping in public is probably okay00:34
@kanzurehttps://groups.google.com/group/science-liberation-front/about00:34
@kanzuresign 'em up00:34
* nsh nods00:39
bkeroWasn't there a person in the news lately who did something like that?00:39
@kanzurei think it's better to keep discussions on this public for the moment00:39
@kanzurebecause people are angry and are going to be trying to do some possibly mistaken things00:40
@kanzureand if they talk about best methods first maybe their efforts wont be as useless00:40
-!- qu-bit [~shroedngr@unaffiliated/barriers] has joined ##hplusroadmap00:47
-!- barriers [~barriers@unaffiliated/barriers] has quit [Ping timeout: 244 seconds]00:48
-!- qu-bit_ [~shroedngr@unaffiliated/barriers] has quit [Ping timeout: 246 seconds]00:48
-!- barriers [~barriers@unaffiliated/barriers] has joined ##hplusroadmap00:49
-!- wrldpc [~wrldpc@203.105.94.33] has joined ##hplusroadmap01:10
-!- lichen [~lichen@c-24-21-206-64.hsd1.or.comcast.net] has quit [Quit: later]01:17
-!- qu-bit_ [~shroedngr@unaffiliated/barriers] has joined ##hplusroadmap01:35
-!- barriers_ [~barriers@121-73-87-49.cable.telstraclear.net] has joined ##hplusroadmap01:36
-!- barriers [~barriers@unaffiliated/barriers] has quit [Ping timeout: 255 seconds]01:37
-!- qu-bit [~shroedngr@unaffiliated/barriers] has quit [Ping timeout: 255 seconds]01:37
-!- sylph_mako [~mako@103-9-42-1.flip.co.nz] has quit [Ping timeout: 252 seconds]02:09
-!- nsh [~nsh@wikipedia/nsh] has quit [Ping timeout: 276 seconds]03:10
-!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap03:13
-!- delinquentme [~asdfasdf@c-24-3-73-35.hsd1.pa.comcast.net] has joined ##hplusroadmap03:21
archelswhat the- http://www.ncbi.nlm.nih.gov/pubmed/1919731603:35
archelsCarbon nanotubes might improve neuronal performance by favouring electrical shortcuts.03:35
superkuhEh... that seems unlikely.03:43
superkuhAt least the mechanism proposed.03:43
superkuhThe electrical shortcut is only 10nm across the cell membrane.03:44
superkuhIt's not like there are longitudinal currents.03:45
-!- nsh [~nsh@wikipedia/nsh] has quit [Remote host closed the connection]03:45
-!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap03:46
* Viper168 licks superkuh 03:54
-!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap04:09
-!- joshcryer [~g@unaffiliated/joshcryer] has quit []04:11
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 248 seconds]04:12
-!- Viper168_ is now known as ViperMeowsworth04:32
-!- yorick [~yorick@oftn/member/yorick] has joined ##hplusroadmap04:35
-!- kirka [~Kirka@95.161.252.108] has joined ##hplusroadmap04:41
kirkaHi guys.04:41
ViperMeowsworthHO GUYS04:44
ViperMeowsworth-caps04:44
kirkaActually I wanted to talk with kanzure about intricacies of running a web forum (about nanotechnology), but he is probably sleeping now, so I'll wait.04:49
-!- gene_hacker [~chatzilla@cpe-70-113-85-111.austin.res.rr.com] has joined ##hplusroadmap04:49
kirkagene_hacker Are you a real genehacker or it's just a fancy nickname?04:50
-!- eudoxia [~eudoxia@186.52.45.156] has joined ##hplusroadmap04:50
kirkaHi, eudoxia!04:50
eudoxiayo04:50
eudoxiai don't think it would take off04:51
kirkaeudoxia Jim wants to do it. None of us has web development experience though.04:52
ViperMeowsworthlol04:52
eudoxiawhich jim?04:52
ViperMeowsworthI used to run forums04:52
ViperMeowsworthbut computer related04:52
ViperMeowsworthgeneral help, software, free and cheap internet listings04:52
ViperMeowsworthetc..04:52
ViperMeowsworthwas freeaccess.com, but the domain got sold off04:52
kirkaeudoxia Last post here: https://groups.google.com/forum/?fromgroups=#!topic/sci.nanotech/Sjvz_Nr857U04:52
kirkaeudoxia I think there should be at least one forum on MNT.04:53
kirkaeudoxia To discuss design of molecular machinery, new software tools etc.04:53
-!- qu-bit [~shroedngr@unaffiliated/barriers] has joined ##hplusroadmap04:53
eudoxiayeah04:53
kirkaeudoxia btw I'm developing MNT CAD in lisp: http://rghost.ru/43023154.view04:54
kirkaThat's very early stage though.04:54
kirkaMy goal is million atoms on screen.04:54
eudoxiaand i'll port it to hylas :)04:54
kirkaHeh, that's possible, I thought about it.04:55
kirkaHylas desperately needs documentation.04:55
eudoxianeed to write some opengl binings04:55
eudoxiawell i updated the readme and wrote some docs04:55
eudoxiait's up in github04:55
eudoxiathere are some code examples04:56
kirkaYes, I have seen this.04:56
-!- qu-bit_ [~shroedngr@unaffiliated/barriers] has quit [Ping timeout: 255 seconds]04:56
eudoxiaare you using CL?04:57
kirkaI think there should be explanation of Hylas's abstractions over LLVM assembler (if tehre are any).04:57
eudoxia?04:57
eudoxiacode generation is straight to LLVM; IR04:57
kirkaeudoxia For the first prototype model viewer I used Racket. It's too slow though, so I'm switching to CL.04:58
eudoxiai see04:58
kirkaI think CL is very good language for this thing, as it demonstrates that 20 year old code (Maxima) runs well on it now.04:58
eudoxianeeds a better gui toolkit04:59
eudoxiaif you need help with CommonQt i have a tutorial on my blog04:59
kirkaSo this CAD will work 10-20 years into the future, without bit rot.04:59
kirkaeudoxia I tink Qt is a monstrous dependence.04:59
eudoxiayeah04:59
kirkaIt's changing04:59
kirkaAPI can change in a couple of years.04:59
eudoxiathe molecular cad code shouold be separate from the interface05:00
kirkaI will have several graphics backends, including parallel software render.05:00
kirkaYes, absolutely.05:00
eudoxiain NanoEngineer, well...05:00
eudoxia"User interface and code are only one, and they love each other so much they are completely glued together."05:00
eudoxia- a friend at work05:00
kirkaHehe.05:00
kirkaGraphics backend will provide simple interface for drawing model and relaying user's input into command logic module.05:01
kirkaeudoxia So Hylas is S-Exp version of LLVM IR?05:02
kirkaIt also does variable allocation (static by default),05:02
eudoxiathere isn't a huge distance between assembly and the language, no05:03
eudoxiai intended it to combine low-level shit with macros05:03
-!- EnLilaSko [~Nattzor@unaffiliated/enlilasko] has joined ##hplusroadmap05:03
eudoxiaand metaprogramming, introspection, reflection, etc.05:03
kirkaBut to program in it one should clearly understand what's done behind the scenes.05:03
kirkaeudoxia That's cool idea.05:03
kirkaI have been waiting for 2 years for such language to be created.05:04
eudoxiai won't let you down bro05:04
kirkaThat's good, heh.05:05
kirkaFor example there is a keyword "recursive", but one can write macro that infers recursiveness of function and adds this label if necessary.05:07
eudoxiathat's actually pretty good05:07
eudoxiaalso i made the cover for the docs book (docs/res/i05:09
eudoxiamg) and its pretty as shit05:09
eudoxiatits 'n stuff05:09
kirkaHeh05:09
kirkaIt would be good if you described primitives (like define, let etc) and memory allocation issues.05:10
kirkaViperMeowsworth As you can see we want to start a forum on molecular nanotechnology.05:12
eudoxiai'll write the function reference today05:12
eudoxiaif i can05:12
ViperMeowsworthyeah I'm scanning, hopping between half a dozen channels05:13
-!- Humean [~quassel@unaffiliated/humean] has quit [Ping timeout: 272 seconds]05:13
gene_hacker@kirka it's my name of course, though once the technology becomes cheap we'll see about it05:15
kirkaeudoxia There is no need to be in a hurry,05:15
kirkaeudoxia You are the creator of the language after all.05:16
eudoxiastill i want to get the docs done05:17
eudoxiaso kirka will you put your cad up on github?05:34
kirkaeudoxia I will, when it's usable05:35
kirkaeudoxia It's still just pdb-loader with sloow 3d-viewer.05:35
eudoxiaareyou using opengl or something else for the 3d?05:35
kirkaopengl05:35
kirkaBut for million of atoms, maybe I'll have to use sprites.05:36
eudoxiahm05:37
eudoxiai've heard about something like that, used, coincidentally, for a molecule viewer05:37
kirkaI also think that in very large designs parts could be abstracted away, represented as polygonal meshes with bondpoints.05:39
eudoxiai was thinking about that too05:39
eudoxiafor MNT MEMS stuff05:39
kirkaAnd to see whole atomic structure one could render them offline.05:39
kirkaYep.05:39
kirkaAlso I want to classify parts and their structural patterns by difficulty of manufacturing.05:47
kirkaAnd build complex structures with easy-to-manufacture components. (relatively "easy", of course)05:47
eudoxianeeds afm integration05:48
kirkaHeh05:48
kirkaOf course, "model structures", not "build".05:48
eudoxialol05:48
kirkaE.g. look at this sorting pump: http://www.nanoengineer-1.net/mediawiki/index.php?title=File:SortingPump1.png05:49
kirkaThere are unnecessary curved surfaces05:49
kirkaAnd sulfur atoms.05:50
kirkaThe design is cool. of course.05:50
eudoxiahttp://wiki.transhumani.com/index.php?title=Molecular_Machinery#Abstract_Sorting_Pump is much better05:52
kirkaI agree. It's strange that nobody designed atomically precise version yet.05:53
eudoxiayeah05:53
eudoxiai would if i knew how to make a bidning site05:53
kirkaAlso, there is a need in springs.05:53
eudoxiai wouldn't be so sure05:53
eudoxiaconsider, on one side, you got actual pressure from molecules to be sorted05:53
eudoxiaon the other side there's near vacuum05:53
eudoxiaprobability would allow the pump to work without moving binding sites05:54
kirkaHmph. That's a question of kT >> EnergyBarrier for sliding05:54
kirkaIf that's true then it'll work this way.05:55
eudoxiai think drexler said it wiuld05:56
eudoxiawould*05:56
eudoxiasomewhere in nanosystems05:56
kirkaYes, I remember that.05:56
kirkaBinding sites probably should be designed to approximate potential energy surface of target molecule.05:57
kirkaWe need to visualize PES first.05:59
-!- gene_hacker [~chatzilla@cpe-70-113-85-111.austin.res.rr.com] has quit [Ping timeout: 264 seconds]06:01
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap06:06
kirkaeudoxia There is a note on complimentary surfaces on p.265 of "Nanosystems".06:07
kirkaeudoxia http://www.zyvex.com/nanotech/bindingSites.html06:28
kirkaThey thought about it 20 years ago.06:32
-!- abetusk [~abetusk@cpe-24-58-232-122.twcny.res.rr.com] has quit [Quit: Leaving]06:48
-!- gene_hacker [~chatzilla@wireless-206-76-99-58.public.utexas.edu] has joined ##hplusroadmap06:59
-!- gene_hacker [~chatzilla@wireless-206-76-99-58.public.utexas.edu] has quit [Ping timeout: 240 seconds]07:04
-!- nsh [~nsh@wikipedia/nsh] has quit [Remote host closed the connection]07:22
-!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap07:22
-!- nsh [~nsh@wikipedia/nsh] has quit [Ping timeout: 276 seconds]07:55
-!- archels [~foo@sascha.esrac.ele.tue.nl] has quit [Ping timeout: 260 seconds]07:59
-!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap07:59
-!- eudoxia [~eudoxia@186.52.45.156] has quit [Ping timeout: 272 seconds]08:12
-!- eudoxia [~eudoxia@r186-52-45-156.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap08:27
-!- eudoxia [~eudoxia@r186-52-45-156.dialup.adsl.anteldata.net.uy] has quit [Read error: Connection reset by peer]08:27
-!- eudoxia [~eudoxia@r186-52-45-156.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap08:28
eudoxiathat paper is almost as old as i am08:28
-!- augur [~augur@208.58.5.87] has quit [Remote host closed the connection]08:51
@kanzuresigh yet another forum08:58
kirkakanzure Do you have better suggestion?08:59
@kanzureyes. the vast majority of the people from foresight institute all prefer mailing lists.09:01
kirkasci.nanotech is dead.09:01
kirkaLeitl's list is also half-dead.09:01
@kanzurethat's not what i'm talking about09:01
@kanzureleitl's list isn't dead.09:02
eudoxiapostbiota/nano?09:02
kirkaIt's repost of non-related news, that's all.09:02
kirka*nano09:02
@kanzurewell, then you should call him out on it09:02
kirkaThere is no mailing list of forum that's created solely for MNT-related discussions.09:03
-!- ParahSailin [~eg@unaffiliated/parahsailin] has joined ##hplusroadmap09:05
@kanzurekirka: jrayhawk created one for you09:05
@kanzurebut you seemed to never look09:05
kirkaI looked09:05
eudoxialink?09:05
@kanzureeudoxia: it was the sympa link09:05
kirkaIt was something scary on some unrealted domain.09:05
@kanzureit even has a web ui for you09:06
eudoxia<309:06
eudoxiahttp://diyhpl.us/sympa/ ?09:07
@kanzureyep09:08
kirkaActually, Jim Logajan wants to start a forum.09:08
kirkaAnd we discuss software, domain and webhosting providers.09:08
kirkakanzure Can you give an advice on choice of good forum engine?09:09
kirkaThere are hunderds of them.09:09
eudoxiaphpBB309:09
@kanzurethey are all fucking awfl09:09
@kanzurephpbb is terrible and evil, you should never use it09:10
eudoxiahahah i'm trolling09:10
kirkaI tried jForum, and it simply works.09:10
eudoxiaroll your own09:10
eudoxiac:09:10
kirkaI don't want to dig into web programming.09:10
@kanzureyou should never use ikonboard, fluxbb, invisionboard, jforum, phpbb, smf, or any of the others.09:10
kirkaHeh.09:10
@kanzurekirka: what do you have against email?09:10
eudoxiahave a git repo on a server where people write their posts as yaml files and threads are folders09:11
kirkakanzure Nothing, but participating in mailing lists is more complex than using a forum.09:11
-!- archbox_ [~archbox@unaffiliated/archbox] has joined ##hplusroadmap09:11
eudoxiai don't even know how to unsubscribe from llvm-dev09:11
kirkaJim wants a forum and I agree with him.09:11
@kanzureeudoxia: llvm-dev-unsubscribe@whatever09:11
-!- lichen [~lichen@c-24-21-206-64.hsd1.or.comcast.net] has joined ##hplusroadmap09:12
eudoxiaso i just send whatever email to that?09:12
@kanzureno, usually you have to include unsubscribe in the subject line to that address09:12
@kanzureotherwise it will complain09:12
eudoxiaoh okay09:12
eudoxiathanks kanz+09:12
@kanzurebtw, this was in the instructions when you first subscribed.09:13
eudoxiaprobablyh09:13
-!- archels [~neuralnet@unaffiliated/archels] has joined ##hplusroadmap09:19
ParahSailinkirka: make it a subreddit09:26
kirkaThere is http://www.reddit.com/r/nanotech already.09:27
kirka95% of unrelated information.09:27
ParahSailinso make another?09:27
kirkaMaybe.09:27
kirkaBut if something happens to reddit, all discussion will be lost.09:28
@kanzurei think kirka doesn't believe in the existence of email09:28
kirkaI believe.09:28
kirkaA read different mailing lists and participate in some.09:29
kirkaBut forum with web interface is simpler.09:29
eudoxiai don't even know how to post to one lol09:29
@kanzureeudoxia: you write an email.09:29
kirkaI use google groups.09:29
ParahSailinif something happens to the internet backbone, all will be lost09:29
ParahSailinclearly we need pigeons09:30
kirkaWe need backups.09:30
eudoxiadogs carrying DVDs09:30
-!- augur [~augur@129-2-129-34.wireless.umd.edu] has joined ##hplusroadmap09:30
@kanzurekirka: backups of mailing lists are very easy. you just upload your mbox file.09:30
kirkaI agree.09:30
eudoxiaor terabyte flash drives, that might have more bandwidth than most internet connections09:30
eudoxiaalthough the latency would be fucked09:31
kirkaBackups of forums are also easy. You just copy the database.09:31
kirkaThe question is "what forum engine do we choose?"/09:31
@kanzurehow about mailman09:32
@kanzureParahSailin: https://groups.google.com/group/science-liberation-front09:33
kirkakanzure Do you plan to upload a torrent with terabyte of scientific papers?09:33
@kanzureno09:34
@kanzurelibrary genesis tried that and nobody seeded09:34
ParahSailinthe last guy who did that did not have a very good day09:34
kirkaI know.09:34
@kanzureParahSailin: aaronsw did not use torrents09:34
kirkaSomeone could host them in I2P.09:35
-!- eleitl [~root@beryllium.ativel.com] has joined ##hplusroadmap09:36
ParahSailinsomeone just needs to make a pirate marketplace for papers09:36
eleitlhowdy bryan09:36
eudoxiaoh jesus it's eugene leitl09:36
* eudoxia faints09:36
eleitlhi eudoxia09:36
ParahSailinlet people run bots to sell papers for like five bitcents09:37
@kanzureeleitl: we are your official fan club09:37
eleitl:))09:37
@kanzureParahSailin: i think you should propose that to the group09:37
* eudoxia twirls a little leitl flag09:37
@kanzureParahSailin: i was thinking of letting people rent out their student ids/passwords for bitcoins09:37
@kanzureParahSailin: and make money on their student loans09:37
@kanzureeleitl: so, eudoxia and kirka have some complaints about the nanotech list09:37
eudoxiai don't just kirka09:38
@kanzureand they are too passive aggressive to tell you in person09:38
eudoxiagee thanks kanz making me look bad infront of the leitl09:38
kirkaWell that's problem of MNT in general09:38
eleitlwhat's wrong with the list, other that it doesn't get traffic?09:38
eleitlmachine-phase is dead in the water09:38
ParahSailinpeople dont like my ideas, and that silkroad for papers would be really low on my own to-do list09:38
@kanzureParahSailin: you should post it anyway, someone else might build it09:39
@kanzureParahSailin: like me09:39
kirkaIt's practiced by maybe 100 researchers in the world.09:39
eudoxia100'¿09:39
eudoxiayou nuts09:39
eudoxia5 tops09:39
kirkaOk, 2009:39
eleitllibgen is pretty good, I hope they can start accepting BTC soon09:39
eleitlkirka, yes, but it's not easy to change that09:39
eudoxiaassuming metzger plays with nanoengineer in his free time09:39
ParahSailini think you're probably about 75% of the programming productivity of that entire group, kanzure09:39
eleitlperry is working hard, but he needs to establish himself academically first09:39
@kanzureParahSailin: ah so i should tone it down..?09:40
ParahSailinso reposting idea there would be redundant09:40
@kanzureParahSailin: please? :(09:40
ParahSailinok09:40
eleitlkanzure, we're going to mix some VM-1 perfusate this weekend, for a test on a human cadaver09:40
@kanzureplease elaborate09:40
nmz787funny their page here doesn't list winter 2013 http://genefoo.com/blogs/news/6074124-launching-personalpcr09:40
@kanzurewhere have you acquired a human cadaver09:41
ParahSailinVM-1 is some kind of antifreeze?09:41
@kanzurenmz787: yeah that seems to be what macowell is doing09:41
eudoxiaCI's cryoprotectant09:41
eleitlthe human cadaver is not our own, it's a group of embalmers which klaus has dug up09:41
eleitlI don't know any of them, though I would like to attend, if feasible09:41
eleitlVM-1 is Pichugin's vitrifyable perfusate09:41
eleitlwe start with it because it's cheap, and we've been asked to do it in a somewhat urgent fashon09:42
-!- yorick [~yorick@oftn/member/yorick] has quit [Remote host closed the connection]09:42
-!- yorick [~yorick@oftn/member/yorick] has joined ##hplusroadmap09:42
eleitlto all: a group of kraut cryonicists are trying to establish a local cryonics provider09:43
-!- mode/##hplusroadmap [-b dende!*@*] by kanzure09:43
eudoxiai saw those pictures of your new lab09:43
eleitlwe've rented lab space in a startup incubator09:43
eleitlwe'll try to do some SENS and other research in there during daytime09:43
eleitlwe're supposed to get our dewars by weekend, but it will likely take a little longer09:44
nmz787eleitl: what's with all the DMSO?09:45
eleitldon't ask me, ask Pichugin. It's not my perfusate of choice.09:46
nmz787is that for the emblaming?09:46
eleitlHowever, it's easy and cheap, and if we do fracturing research, we'll also start with VM-1, and possibly VM-1 with iceblockers.09:46
eleitlNot embalming, perfusion, and then thermal descent to dry ice, and transport overseas on dry ice.09:47
eleitlVitrified, not frozen.09:47
eleitlAssuming perfect perfusion, which almost never happens with the kind of ugly cases we're getting in krautlandia.09:47
eudoxiatests or actual past cases?09:49
eleitlThere have been CI cases in Germany, and they typically do not look very good, due to very conservative criteria for death declaration.09:50
eleitlIt doesn't matter to me, since we need to be able to learn the technique, regardless where it is going to be used.09:51
@kanzurekirka: can you elaborate why you don't like eleitl's nanotechnology forum?09:53
kirkaIt's not that I don't like it.09:53
kirkaThere are some interesting posts.09:54
kirkaBut as in other mailing lists there is little of MNT.09:54
eudoxiait's not about the medium09:54
eleitlof course, because MNT is rare as hen's teeth09:54
kirkaIt's a global problem.09:54
eudoxiathere just aren't many people interested09:54
kirkaI agree.09:54
@kanzurethen why did you want to make a separate forum?09:54
@kanzurei am very confused.09:54
kirkaMolecular manufacturing has plitical burden on it, so it's doesn't get funded.09:55
kirkakanzure Jim Logajan wants to make a replacement for sci.nanotech mailing list.09:55
eleitlI'm not married to MNT. Nano is nano. All paths lead to machine-phase, eventually.09:55
eleitlsci.nanotech is a Usenet group, not a mailing list, right?09:55
kirkaI agree, protein/foldamer engineering has very good future.09:56
kirkaeleitl Right.09:56
kirkaAnd also google group.09:56
eleitloh, haven't seen that one.09:56
kirkaBut I think that nanoparticles are a dead end.09:57
eleitlnanoparticles are not real nanotechnology09:57
eleitljust inflationary use09:57
eleitlJust subscribed to sci.nanotech, thanks09:58
eudoxiaeleitl, what do you think about Zyvex's epitaxy approach?09:58
kirkaYes, it's after creation of NNI and lobbying of Nanobusiness alliance nanoparticles research has become NNI's main focus.09:59
eleitlhaven't look at zyvex in more than a decade10:00
eudoxiahttp://www.zyvexlabs.com/Research.html10:00
eudoxiahttp://nextbigfuture.com/2010/10/what-is-optimal-bootstrapping-pathway.html10:00
eleitlthanks, I will look at it10:01
eudoxia"Von Ehr hopes to have a primitive nanotechnology system that can create blocklike objects by 2015 and rudimentary molecular manufacturing by 2020. The big game-changer to my mind is Digital Matter. For enzymes, catalysts, and increasingly even for transistors, every atom has to be in the correct place in order for the molecule or component to function. "10:01
eudoxiai don't want to get my hopes up ;_;10:01
-!- mode/##hplusroadmap [-b d3nd3!*@*] by kanzure10:02
nmz787rigel: so you're in PDX?10:02
nmz787I'm gonna go tour FEI, the e-microscope and FIB company this thurs evening10:02
kirkaThere is also a promising parallel MEMS-STM firm: http://www.icspicorp.com/10:02
nmz787http://portland-or.sites.acs.org/newsletter.pdf10:02
kirkaThey are partners of Zyvex.10:03
eleitlI wouldn't get the hopes up too far10:03
eleitlthe best hope is with DNA autoassembly10:03
eleitlwhich is not much10:03
nmz787using modified cells seems like a reasonable way to go about APM10:04
kirkaRecent synthetic DNA ion channel is good.10:04
kirkaThis one: http://www.sciencemag.org/content/338/6109/932.abstract10:04
kirkaIt's a DNA-origami structure that works as ion channel, designed from scratch.10:04
-!- eudoxia [~eudoxia@r186-52-45-156.dialup.adsl.anteldata.net.uy] has quit [Quit: leaving]10:04
eleitlnot sure I've sent that. why don't you send such papers to the nano list?10:05
kirkaI will, next time. I'm not used to mailing lists yet.10:05
eleitlHeh. What do you use? Usenet?10:06
@kanzure"I know Ariya used to work for Trolltech and that has obviously been fundamental in PhantomJS's success to date."10:06
@kanzureoh. i didn't know that.10:06
@kanzurepaperbot: http://www.sciencemag.org/content/338/6109/932.abstract10:06
paperboterror: didn't find any pdfs on http://www.sciencemag.org/content/338/6109/932.abstract10:06
paperbothttp://diyhpl.us/~bryan/papers2/paperbot/e92df00e68df7c336e7bf8fccf642b3210:06
@kanzurehmm10:06
archelseleitl: do you think it would be worthwhile to look at plastification as an alternative to cryonic preservation?10:07
@kanzurepaperbot: http://www.sciencemag.org/content/338/6109/932.full.pdf10:07
paperbotno translator available, raw dump: http://diyhpl.us/~bryan/papers2/paperbot/859715fc8cbb731cfe0429f2b19aba5d10:07
nmz787did science change their format?10:07
eleitlMike Darwin is making some very interesting noises about fixation/plastination, so I'm hanging on until he delivers.10:07
kirkaeleitl I'm from 1990's generation, so I haven't seen age of UseNet, heh.10:07
eleitloic, kirka10:07
@kanzurehttp://diyhpl.us/~bryan/papers2/paperbot/Synthetic%20Lipid%20Membrane%20Channels%20Formed%20by%20Designed%20DNA%20Nanostructures.pdf10:07
@kanzurethere you go10:08
@kanzurenmz787: i guess so :(10:08
eleitlI think fixation/plastination might be feasible, but it's very difficult to validate.10:08
eleitlthe no-feedback problem10:09
@kanzurefascinating https://github.com/laurentj/slimerjs10:09
@kanzuregecko-based phantomjs port10:09
@kanzurenot so headless though10:09
kirkaeleitl Drexler is into peptoids/foldamers these days. Probably that's a faster route than UHV-STM.10:10
eleitlyes, I've always been of the opinion that bootstrap by bio is easier10:10
-!- mode/##hplusroadmap [-b *!*@*.croy.cable.virginmedia.com] by kanzure10:10
kirkaGood protein/foldamer engineering is rare though.10:10
eleitlinverse protein folding is almost solved10:10
@kanzuremode -b *!*@*cpc10-croy17-2-0-cust245*10:10
@kanzureoh god10:10
-!- mode/##hplusroadmap [-b *!*@*cpc10-croy17-2-0-cust245*] by kanzure10:10
nmz787i think we just need to design a minimal cell that is XYZ controllable10:11
eleitlstrangely, it is easier than straight protein folding10:11
nmz787XYZ and rotate10:11
-!- d3nd3 [~dende@cpc10-croy17-2-0-cust245.croy.cable.virginmedia.com] has joined ##hplusroadmap10:11
d3nd3success10:11
d3nd3kanzure: ty10:11
kirkanmz787 But their membranes are absolutely non-stiff.10:11
eleitlwe need an enzyme which can lay down bucky10:11
eleitlthat might be possible10:11
nmz787kirka: what's stiffness have to do with moving around to place atoms?10:12
nmz787then the cell becomes the STM and STP tip with atom loader10:12
kirkanmz787 Stiffness matters because of thermal noise.10:12
nmz787isn't that why receptors exist, to lock onto a target before manipulation?10:12
-!- ViperMeowsworth [~Viper@unaffiliated/viper168] has quit [Ping timeout: 255 seconds]10:13
kirkanmz787 So the receptor is the controlled molecular machie, floating in cell?10:14
kirka*machine10:14
kirka*in cells membrane10:14
nmz787well the whole cell is the controlled machine10:14
nmz787you'd tell it export receptor A to surface10:14
kirkaeleitl is right, we need engineered enzymes.10:14
nmz787export enzyme G to surface10:14
eleitlguys, it's very interesting, but I need to catch the train home. I'll try dropping by later today, or at least I'll read the scroll tomorrow.10:14
eleitlcatch you laters. Thanks!10:15
nmz787I guess telling the cell to move 5 hydroxyls south vs north would be hard10:15
@kanzureyou could magnetically move the cell10:16
@kanzureand make it produce magnetic-responsive enzymes10:16
nmz787maybe it'd have to be magneto or light -tactic... so move X atoms in J direction10:16
nmz787well the moving X atoms is done already with actin filament moving10:16
nmz787so somtehing could be adapted for non-polymers10:17
kirkaI don't think it's possible to move cell as a whole with right precision (1 angstrem).10:17
nmz787but movement would be too coarse I think if you didn't have an enzyme doing the positioning10:17
kirkaIt's possible for components of protein/RNA/DNA machinery though.10:17
nmz787well you don't need to move the whole cell to move 1 angstom10:18
nmz787since the membrane is much larger area, you can just move within the membrane10:18
kirkaIf we had an analog of Ribosome that can manufacture more complex structures than aminoacid-chains that'd be a huge step forward.10:18
nmz787what other polymers would you want it to make?10:19
kirkaThere is already work on mutating ribosome to understand 4-word codons.10:19
kirkanmz787 Polymers with two covalent bonds for example10:19
kirkanmz787 They would be more predictable.10:20
kirkaThere are, for example peptoids: http://en.wikipedia.org/wiki/Peptoid10:21
kirkaThey are more stable than proteins.10:21
kirkaThere is a large space of possible monomers, with them it may be easier to build stable structures than with amino acids.10:24
nmz787lacking secondary structure seems pretty limiting to make anything catalytic10:25
archelskanzure: you should tell paperbot to grab the supplements as well ;)10:25
archelsthe supplement on that DNA-origami-ion-channel paper is much better than the brief main publication10:25
kirkanmz787 >Design and Conformational Analysis of Peptoids Containing N-Hydroxy Amides Reveals a Unique Sheet-Like Secondary Structure10:26
archelsI still have them on my webserver, http://www.turingbirds.com/temp/Langecker.SM.pdf10:26
archelspaperbot: http://www.turingbirds.com/temp/Langecker.SM.pdf10:26
archelslow hit rate :P http://diyhpl.us/~bryan/papers2/paperbot/10:27
-!- archbox_ [~archbox@unaffiliated/archbox] has quit [Read error: Operation timed out]10:27
kirkanmz787 The main problem as it seems to me is a lcak of good method to design foldamers.10:28
paperbotno translator available, raw dump: http://diyhpl.us/~bryan/papers2/paperbot/1e9ee8083534045e107de3879995fe4510:28
kirka*lack10:28
kirkanmz787 There is a need in foldamer design software with real-time simplified force fields.10:29
ParahSailinpaperbot: https://s3-us-west-2.amazonaws.com/droplr.storage/files/acc_15515/sK7z?AWSAccessKeyId=AKIAJSVQN3Z4K7MT5U2A&Expires=1358189301&Signature=7GvBHMfkgBjcBFpnQ79a0jENm7c%3D&response-content-disposition=attachment%3B%20filename%2A%3DUTF-8%27%271-s2.0-016378278990009X-main.pdf10:30
paperbotno translator available, raw dump: http://diyhpl.us/~bryan/papers2/paperbot/293105c6bfdc3f3d95184ef93139394710:30
-!- barriers_ [~barriers@121-73-87-49.cable.telstraclear.net] has quit [Read error: Connection reset by peer]10:30
kirkaDrexler did Pam-3 and Pam-5 into NE1 for this purpose.10:30
nmz787hmm10:30
nmz787seems like it would make a lot more sense to pour resources into learning how to fold normal proteins, since we're composed of them already10:31
-!- qu-bit_ [~shroedngr@unaffiliated/barriers] has joined ##hplusroadmap10:31
nmz787them modify the strategy from there10:31
nmz787then*10:31
kirkanmz787 Folding and designing aren't inverse problems.10:31
d3nd3so how do i view paperbots papers ?10:31
nmz787why not?10:32
kirkanmz787 Structures can be designed to be easily predictable.10:32
-!- barriers [~barriers@unaffiliated/barriers] has joined ##hplusroadmap10:32
nmz787but we care about function10:32
kirkaE.g. this protein was designed to be easily predictable: http://www.nature.com/nature/journal/v458/n7236/full/nature07841.html10:33
kirkaAnd it works.10:33
kirkaId on't think that all of natural protein's complexity is related to function.10:33
-!- qu-bit [~shroedngr@unaffiliated/barriers] has quit [Ping timeout: 255 seconds]10:33
-!- nsh [~nsh@wikipedia/nsh] has quit [Remote host closed the connection]10:34
nmz787no but it obviosuly works10:34
kirkaIt's just evolution building complex patterns in sequental steps (it adds complexity until functionality breaks).10:34
nmz787and there's a huge amount of prior art waiting to be understood and hacked10:34
kirkaI agree10:35
-!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap10:35
-!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has quit [Quit: Leaving]10:35
kirkaBut much of this complexity can be just random mutations.10:35
-!- BathWater is now known as ElixirVitae10:35
kirkaDrexler writes a lot abou this.10:35
nmz787there's still a reason that it works10:35
nmz787that isn't random10:35
nmz787so PDX has this, so this /should/ work paperbot: http://www.nature.com/nature/journal/v458/n7236/full/nature07841.html10:36
kirkaYep, core structure, functional folds, etc.10:36
kirkaBut look at that O2 transport protein - it doesn't contain dpecial folds that natural one has.10:37
kirkaAnd it works.10:37
nmz787yeah but all that says is that evolution either had a reason to keep that fold, or it's just prehensile10:38
nmz787err no10:38
nmz787s/prehensile/hanging around/10:38
nmz787o10:39
nmz787oO big snow flakes outside :D10:39
-!- sylph_mako [~mako@103-9-42-1.flip.co.nz] has joined ##hplusroadmap10:39
@kanzureParahSailin: what kinda link was that10:39
kirkanmz787 http://metamodern.com/2009/03/30/a-revolution-in-de-novo-protein-engineering/10:39
kirkaThe point is that evolation adds complexity until something breaks.10:40
kirka*evolution10:40
@kanzurebrownies: "January Events at Genspace - Glow Your Mind!"..... you got what you wanted, are you happy now?10:40
nmz787"Protein engineering is often approached as if it were part of biology. Imagine approaching aerospace engineering as if it were part of ornithology: "10:41
nmz787isn't a bird still more efficient than a plane?10:41
kirkaActually, no.10:41
kirkaThough that's complex to measure10:41
nmz787yeah esp if you are talking mass being moved, or just a single being moving aroun10:42
nmz787like person per fuel or mass per fuel spent in the air10:42
kirkaThere is an article about this: mb-soft.com/public3/birdeff.html 10:43
brownieskanzure: not until i have a glowing cat of my own10:43
kirkaBut looks like it's down.10:43
brownieskanzure: but i do applaud their innovative focus10:43
nmz787kanzure: is this list any good diyresearch-bounces@inventati.org]10:44
nmz787?10:44
@kanzurekirka: http://web.archive.org/web/20120407104406/http://mb-soft.com/public3/birdeff.html10:44
@kanzurenmz787: not sure. it's just diyresearch@inventati.org10:44
kirkakanzure Thanks.10:44
@kanzurenmz787: it's probably people stalking diy people, and pertending it's research10:44
-!- EnLilaSko [~Nattzor@unaffiliated/enlilasko] has quit [Ping timeout: 272 seconds]10:49
-!- EnLilaSko [~Nattzor@unaffiliated/enlilasko] has joined ##hplusroadmap10:52
kirkabtw I haven't found any good book on protein engineering.10:52
ParahSailinbooks are for topics that are 30+ years old10:55
kirkaYes, looks like I have to dig papers on the subject. I'm doing it anyway.10:56
nmz787kanzure: none of the PDX people you told me to email got back :(10:57
nmz787no one wants to be my for realz friend10:57
@kanzurewhaat10:57
@kanzurebunch of jerks10:57
nmz787it's probably my fault for being so weird10:58
@kanzureno they are the weirdos10:59
@kanzurehuman decency is to at least reply10:59
-!- eudoxia [~eudoxia@r186-53-129-15.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap10:59
rigelnmz787: yes10:59
nmz787rigel: have you ever gone to The Circuit climbing gym?11:00
rigelno11:01
rigeli am fat11:01
@kanzurefat monkeys are also capable of climbing11:02
nmz787if it weren't for my epigenetics (I think) I'd probably be fat too11:02
nmz787my mother was bulemic when she was pregnant, and I think that might have something to do with me not adding weight11:03
nmz787my body thinking i'm born to starve or something11:03
rigeldoubtful11:03
rigelprobably the other direction11:03
rigellimited resources, fetus works hard to keep as many calories etc as possible11:04
nmz787i dunno, i read something about this regarding starving during pregnancy causing the kids to be insensitive to insulin11:04
nmz787meaning they wouldn't store sugar11:04
rigelits all speculative horseshit anyway11:04
rigeli am in a foul mood11:04
nmz787plus is that if you're really starving you brain gets all the sugar it can, since sugar is never stored where the brain cant get it11:04
rigeli probably do not need to be having this conversation right now. yes i am in portland.11:04
bkerooh dear11:05
nmz787downside if you're well fed, you can easily become too sweet and start to damage organs11:05
nmz787rigel: what brings you here?11:09
nmz787i see you do something at OHSU11:09
rigelaaron swartz11:11
nmz787so the sanyasin (where OSHO the rolls royce guru stayed in E Oregon) ranch is now a christian youth ranch11:11
nmz787even though the Sheela lady commited bioterrorism, I think the christian camp is probably worse overall11:11
nmz787well i take that back11:12
rigelthats kind of bigoted11:12
@kanzure"Identifying the DNS server used to fulfill a HTTP request" http://5f5.org/ruminations/dns-debugging-over-http.html11:13
nmz787i googled bigtoed11:13
rigellots of different kinds of religious folks, some good some bad11:13
nmz787well neither creed is mine, so i don't think i'm a bigot11:13
rigelof course you dont11:14
-!- rigel [~pi@c-76-105-237-98.hsd1.or.comcast.net] has left ##hplusroadmap []11:14
nmz787by technical definition i'm not a bigot in this case11:15
nmz787rigel are you interested in touring FEI with me this Thurs evening?11:15
@kanzurenmz787: https://groups.google.com/group/science-liberation-front11:15
nmz787kanzure: wasn't the first message in there 'can we change the name of this group to avoid being targeted'11:16
@kanzuresure11:17
eudoxiait's okay guys11:17
@kanzurenobody suggested an alternative name11:17
eudoxiawe're all already in multiple watchlists11:17
@kanzure"OH MY GOD WE'RE TALKING ABOUT READING?"11:17
eudoxiajust for the record I think the DHS provides many useful services like their C vulnerabilities database11:17
eudoxiaAmerica <311:17
nmz787C vulnerabilites eg?11:18
nmz787eh?11:18
nmz787I was reading about CODIS a few weeks ago11:18
nmz787they don't have protocols listed11:18
@kanzure00:36 <+nsh> ARCHIVISTS OF THE WORLD UNITE! YOU HAVE NOTHING TO LOOSE BY YOUR INSTITUTIONAL ACCESS!11:18
nmz787so I called their office, but they didnt get back to me11:18
-!- nsh [~nsh@wikipedia/nsh] has quit [Ping timeout: 276 seconds]11:25
kirkaeudoxia So, for C2H2 buckytube (9,0) seems to be a good pocket.11:32
eudoxiayep11:32
eudoxiait's in nanosystems too11:32
eudoxiaactuall i think the picture from that old paper was a figure in NS11:33
kirkaeudoxia To make a srting pump we should think how to bond it with diamond rotor (without inducing too much strain).11:33
eudoxiabuckytube can be grown on some diamond surfaces, and has been grown11:33
nmz787wasn't someone saying nanoengineer isn't realistic at all?11:34
eudoxiahttp://img405.imageshack.us/img405/1253/nanotube60onc111.png11:34
-!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap11:34
kirkaeudoxia That;s good, but we need large diameter and hole for rod to slide.11:35
kirkanmz787 In what sense "unrealistic" ?11:35
eudoxia"I was reading the Minimal Toolset paper, on the section explaining the synthesis of graphene through mechanosynthesis, where is says “nanotubes with n=6, 8, 10, 12, and 18 are readily mated to the C(111) surface, but nanotubes with n = 7, 9, 11, 13, and 15 are geometrically incompatible.”. I decided to try out a (6,0) nanotube. I feel kinda sad for that little Hydrogen atom in the middle though."11:35
nmz787like not applicable to reality11:35
nmz787kanzure: i think it was that guy who started at MIT when he was a kid11:36
eudoxiait's all a conspiracy by Big Nano11:36
eudoxialel11:36
kirkaBig NanoBusiness11:36
eudoxiathey killed jfk11:37
kirkaCool nano-sunglasses - serious business11:37
nmz787no i'm being serious11:38
kirkaThen you have to elaborate11:38
nmz787he gave specific reasons regarding it not taking a bunch of physics into account11:39
nmz787well kanzure didnt get back11:39
kirkaWho?11:39
nmz787he had a double acronym in his handle11:39
nmz787err11:39
nmz787double letter11:39
nmz787his first and last name had the same starting letter11:39
nmz787and he was some child prodigy at MIT i think11:39
eudoxiaDavid11:39
eudoxiaDarlymple11:39
nmz787yes11:39
nmz787what's his handle, I'll grep the logs11:40
eudoxiadon't know about his criticism of MNT, though I know he graduated from MIT in biophysics at 1611:40
eudoxiaor something11:40
@kanzuredavidad11:41
-!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap11:41
eudoxiathis channel sure has seen its fair share of awesome people11:45
@kanzurepaperbot: http://www.nature.com/nature/journal/v493/n7431/full/493159a.html11:45
paperbothttp://diyhpl.us/~bryan/papers2/paperbot/Altmetrics%3A%20Value%20all%20research%20products.pdf11:45
@kanzureaww yeah11:45
nmz787hmm i'm not good enough at cmd line searching11:47
@kanzuregrep davidad *.log11:47
nmz787i guess it wasnt davidad11:47
nmz787i want to grep for ne1, then grep for another query within +-50 lines of that line11:49
nmz787or +-100 lines11:50
eudoxiagrep ne1 *.log -b 25 -a 25 | grep davidad ?11:50
eudoxianot sure11:50
@kanzurejust write a regular expression that accepts 50 surrounding lines11:50
nmz787eudoxia: it looks like that's almost correct except -B and -A11:53
eudoxiaman grep seems to side with me11:54
eudoxiatry uppercase11:54
nmz787yeah thats what I just sent in the chat11:55
kirkaInteresting to hear concrete criticism. I don't understand yet if he criticises software (NE1), molecular dynamics, or whole MNT as a field.11:55
nmz787kanzure: cool, I'm more valuable to the NSF already!11:57
kirka"Science Liberation Front" sounds cool12:00
nmz787ahh i guess this was it https://groups.google.com/forum/?fromgroups=#!topic/diybio/1H2MBrMAwUU12:00
kirka>The constructions also don't appear to be chemically relevant i.e. you just place random atoms next to each other without any care for charge or interaction.12:01
kirkaHehe12:01
nmz787that sounds like a pretty serious flaw if true12:02
kirkaThat's typical biology guy's criticism.12:03
nmz787if we could form a working group, with scientists and mathematicians and engineers and programmers, we could do this for real12:03
kirkaNanoDynamics-1 is a specialised MD engine for simulationg molecular machinery in vacuum.12:04
nmz787kanzure: didnt that russian guy say he got NE1 working in late-version ubuntu natively?12:04
kirkaIt has custom force field.12:04
@kanzurenmz787: kirka IS that russian guy12:04
kirkaHehe12:04
kirkaYes I did12:04
nmz787kanzure: really?12:04
@kanzuresigh12:04
nmz787kanzure: isn't there another russian guy?12:04
nmz787:D12:05
kirkaThere could be one12:05
nmz787lol12:05
nmz787ok12:05
nmz787well12:05
eudoxiai think there was12:05
nmz787if you did that, then you know the guy I'm talking about12:05
nmz787i thought he had more broken/rough english12:05
nmz787fitzzsim12:07
nmz787sorry fitzsim12:08
kirkaWell, nmz787 at most that counts as criticism of NanoDynamics-1 - custom MD engine that NE1 uses.12:08
@kanzurethomas fitzsimmons12:08
kirkaThere are reasons to state that for stiff covalent structures it gives results accurate enough.12:08
@kanzureor you might be thinking of theirix12:08
nmz787nah i think i was just crossing neurons12:10
kirkaAnd for "chemical relevance" that's his opinion as biomolecular guy. Who knows, maybe he doesn't know macromolecules except DNA, RNA and proteins.12:10
nmz787well if the bonds aren't right, that's major regardless12:10
eudoxiai think the criticism was of things like the Planetary Gear12:10
kirkaPlanetary Gear is speculative indeed.12:10
eudoxiaie bullshit pretty colors things12:10
eudoxiabatshit retarded12:11
kirkaWe son't know about stability until we do quantum chemistry on these models.12:11
kirka*won't12:11
-!- rigel [~pi@c-76-105-237-98.hsd1.or.comcast.net] has joined ##hplusroadmap12:13
nmz787i'd like to see DNA origami become a little easier12:13
rigelsorry i was so crabby earlier12:13
nmz787that's something I could help with on the synthesis side12:13
eudoxiai guess we're going to have to be the ones to do that12:13
kirkaI don't think it's impossible to build stable gear-like structure on that scale.12:13
nmz787rigel: no worries12:13
eudoxiabecause merkle and freitas sure won't do anything12:13
eudoxiale sigh12:13
kirkaeudoxia You think so?12:14
eudoxiaoh totally bro12:14
rigeli have no skills to try and implement this but are fpgas useful for solving protein folding problems?12:14
nmz787i think a PS3 would be better12:15
rigeli have not seen that approach taken in the lit12:15
nmz787or a GPU12:15
kirkarigel I don't think so.12:15
kirkaGPUs ahve better GLOPS/$12:15
kirka*have12:15
rigelwell gpus would be better than x8612:15
kirkaFPGA have just fixed point math12:15
nmz787rigel: check out DE Shaw12:15
rigelbut i was figuring you might be able to program a "processor" per functional unit or something12:15
kirkaYou can optimise MD algorithms for fixed-point (D.E. SHaw & co did this) and run it on FPGA, but it's world class work.12:16
rigelmaybe fpu isnt as necessary?12:16
nmz787rigel maybe folding@home12:16
kirkaIt isn't12:16
kirkaBut you have to carefully design MD algoeithm.12:16
kirkaAh, you have no skills12:16
rigelcorrect12:17
rigeli was just musing12:17
ParahSailinwut.  < kirka> FPGA have just fixed point math12:17
kirkaI think it'll be useful (for you in the first place) to learn about programming, numeric methods and MD.12:17
rigelpeople turn fpgas into gpus, so maybe you could take that even further12:17
kirkaParaSailin Yes, they don't have hardware FPUs12:18
nmz787well it seems that if you're in the quantum domain, you should be able to use fixed for a helluva lot12:18
kirka>quantum domain12:18
rigelyeah, i am in no position to learn any of that shit12:18
nmz787rigel: but fpga is more general purpose12:18
chris_99that's not ture kirka, you use a softcore to to Floating point arithmetic ;)12:18
chris_99*true12:18
rigelno time, hate school12:18
kirkachris_99 I understand12:18
kirkachris_99 But it'll be slow12:19
ParahSailinoh, how common fpgas have 18bit multipliers?12:19
ParahSailingotcha12:19
kirkaYes12:19
kirkaFixed point.12:19
nmz787rigel:  what major?12:19
rigeli am in med school right now12:19
rigelhating life. may quit12:19
ParahSailineasy enough to use that to do fp ops12:19
nmz787rigel: congrats12:19
rigelits fucking garbage12:19
nmz787o12:19
nmz787were you an EMT before joining?12:20
rigelmemorize shit to regurgitate it12:20
kirkarigel I didn't thought about school when I made my propoosition. Nowadays there is a lot of wonderful books for free on the internet.12:20
rigeli am not that guy12:20
rigelkirka, yeah but no time12:20
nmz787rigel: can you switch to MD/PhD?12:20
rigelnmz787: no, bench scientist before this12:20
kirkaParaSailin My proposition is this "With soft IP FPUS FPGA will make less FLOPS/$ than GPU".12:21
rigeli am 36 and do not want to be in school til im 5012:21
rigelwhich would be mdphd12:21
nmz787ahh, my buddy went from EMT to Paramedic and is now starting med school in fall12:21
ParahSailinah ok12:21
nmz787yea i agree with kirka12:22
nmz787you want a platform with as many cores as possible12:22
nmz787or virtual cores12:22
kirkaParahSailin For fixpoint-optimized algs like DSP they may be better. They are also much more compact. Military like this and use FPGAs a lot.12:22
nmz787you want the pipeline to be as wide as possible12:22
nmz787to do parallel operations12:22
rigelthats sort of what i figured12:23
rigeland limited instruction set per core12:23
nmz787being more generalized means more space per operation12:23
nmz787physical nanometer space12:23
nmz787on the chip[12:24
kirkaI'd like to write some special-purpose MD engine for e.g. HD7870 GPUs (in OpenCL).12:24
chris_99nmz787, what are you after? a small chip for electronics stuff or something very hefty12:24
kirkaBut that's a distant goal.12:24
nmz787FPGAs and CPLDs are often used to test and iron out logic designs that often go on to ASIC designs12:24
kirkaYes, they are.12:25
kirkahttp://www.dinigroup.com/new/DN7020k10.php12:25
kirkaThis for example <312:25
nmz787chris_99: i don't know how to program FPGAs, even some of ARM and DSP chips are huge learning curves to get up and running12:26
nmz787so i am sticking to off-the-shelf processing wise12:26
kirkaI did some FPGA programming in the past.12:26
chris_99nmz787, i just bought a little FPGA board recently, it's fun to play with it's got VGA, serial etc. on it and pretty easy to program12:26
nmz787cool12:27
ParahSailinwhat board?12:27
nmz787sparkfun has some $15 CPLDs i was looking at12:27
nmz787for doing glue logic12:27
nmz787ahh not sparkfun12:28
nmz787http://www.seeedstudio.com/depot/index.php?main_page=advanced_search_result&search_in_description=0&keyword=cpld&x=26&y=1112:28
chris_99ParahSailin, http://www.ebay.co.uk/itm/290790893721?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1497.l264912:28
-!- Humean [~quassel@unaffiliated/humean] has joined ##hplusroadmap12:29
nmz787i wanna make a 1080p B/W line graph on a propeller using a parallel to DVI chip12:30
kirkaEP2C5 is old12:30
-!- OldCoder_ [~OldCoder_@c-69-181-140-134.hsd1.ca.comcast.net] has quit [Ping timeout: 276 seconds]12:30
chris_99yeah i know12:30
kirkaI created a board for Ep3C10 3 years ago12:31
kirkaDid some fun things with that board12:31
kirkaAnd moved to other things12:31
chris_99i'm going to experiment with PUFs on it12:31
-!- Humean [~quassel@unaffiliated/humean] has quit [Ping timeout: 260 seconds]12:33
nmz787this looks cool, too long to read now http://genomeinformatician.blogspot.de/2012/05/dna-compression-reprise.html12:34
-!- Hu_Meanan [~quassel@199.48.197.18] has joined ##hplusroadmap12:43
nmz787fenn: you around?13:03
kirkaExotic chemistry question: http://en.wikipedia.org/wiki/Pentacene - are long chains of benzene rings (this is 5-chain) stable/possible?13:04
nmz787long being how many?13:04
kirka100 monomers13:05
nmz787most likely13:05
kirkaI can't find anything13:05
kirkaLooks like it's exotic.13:05
xxalright i have a question13:06
ParahSailini imagine difficult to synthesize, but not impossible/unstable13:06
xxanimal manipulations and implants based on our vision of them13:06
nmz787polycyclic aromatics are common13:06
xxthats anthropocentric right?13:06
nmz787anytime biomass is burnt13:06
xxor anthropomorphic?13:07
nmz787well lava will cause biomass to burn13:07
nmz787so not totally anthro13:07
nmz787oh that wasnt connected13:07
kirkaParahSailin nmz787 Do you think that tensile strength of such chain will be higher then e.g. Polyethylene?13:08
kirkaThere are two bonds in chain's section, so tensile strength should be higher.13:09
nmz787kirka: https://www.princeton.edu/mae/people/faculty/carter/ecdocs/EAC-147.pdf13:09
kirkaBut that's a good question.13:09
ParahSailinwell, that sort of thing could be considered a graphene13:09
kirkaYes, extremely thin piece of graphene13:09
nmz787ParahSailin: that's what I was thinkin13:09
nmz787not too much diff13:09
kirkanmz787 Thanks!13:09
nmz787prob similar modelling13:09
kirkaeudoxia https://www.princeton.edu/mae/people/faculty/carter/ecdocs/EAC-147.pdf13:09
nmz787i google scholared decacene polymer aromatic13:09
nmz787i dunno if its right but i figured decacene might be a 100mer benzene13:10
nmz787or 1013:10
kirkaThat's a good  component for rod logic signal transmission.13:10
nmz787http://en.wikipedia.org/wiki/Acene13:10
nmz787ahh13:10
kirkaHere it is13:10
nmz787that should be linked from the PAH wiki page13:11
ParahSailinah, gets more reactive as you get bigger13:11
kirkaThat's bad.13:11
nmz787ParahSailin: why do you say that?13:12
kirkaWell, we'll think on another alternatives.13:12
ParahSailinsays that on page you linked13:12
nmz787princeton or wiki?13:12
ParahSailinwiki13:12
-!- G0VERNMENT [~G0VERNMEN@unaffiliated/g0vernment] has joined ##hplusroadmap13:13
jrayhawkcheese it! it's the fuzz!13:13
nmz787ahh I see13:13
G0VERNMENThola que tal?13:14
jrayhawkgesundheit13:14
eudoxiaguten tag bro13:14
kirkaGuten Abend13:14
eudoxiaeleitl what do you think of this list of cryopatients i compiled http://wiki.transhumani.com/index.php?title=Cryonics#Patients13:15
@kanzurewhy not just commit that to diyhpluswiki ahhhhhh13:17
eudoxia>:(13:18
kirkaLooks like Polyacenes are unstable and have complex electronics properties.13:18
nmz787oligoacenes is a good keyword13:27
nmz787i like it13:27
@kanzure"Accelrys has acquired its long-time partner Vialis AG, a systems integrator based in Liestal, Switzerland. Vialis’ experience implementing and supporting paperless laboratory solutions strengthens Accelrys’ position in the laboratory informatics software market and expands the company’s capabilities in the downstream analytical development"13:32
@kanzurehttp://accelrys.com/about/news-pr/0113-announcement.html13:32
-!- qu-bit_ is now known as qu-bit13:45
kirkahttp://allofusnow.com/joelstream/wp-content/uploads/2010/07/Cheetah_tech.png <313:45
eudoxia:313:47
-!- wrldpc [~wrldpc@203.105.94.33] has quit [Quit: wrldpc]13:50
-!- Juul [~Juul@50-0-83-116.dsl.static.sonic.net] has joined ##hplusroadmap14:04
-!- d3nd3 [~dende@cpc10-croy17-2-0-cust245.croy.cable.virginmedia.com] has left ##hplusroadmap []14:15
kirkakanzure Is it possible to make a backup of Nanoengineer-1 wiki? There is valuable information.14:27
-!- EnLilaSko [~Nattzor@unaffiliated/enlilasko] has quit [Remote host closed the connection]14:33
-!- G0VERNMENT [~G0VERNMEN@unaffiliated/g0vernment] has quit [Ping timeout: 260 seconds]14:37
-!- G0VERNMENT [~G0VERNMEN@unaffiliated/g0vernment] has joined ##hplusroadmap14:38
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Ping timeout: 252 seconds]14:39
@kanzurei have a backup.15:00
@kanzureJuul: join us or don't https://groups.google.com/group/science-liberation-front15:01
kirkakanzure Is it possible to create a local copy (wikimeda engine etc) ?15:01
@kanzureyes, but i hate mediawiki15:02
kirkakanzure Can I download this backup somewhere?15:02
@kanzureno15:02
@kanzuremaybe i will dump it into diyhpluswiki.git15:02
@kanzureif you would find this useful?15:02
kirkaThere are tutorials on design of molecular machinery and mathematics behind force fields, ao I find it useful15:03
kirkaBut I'm fine with local copy15:03
@kanzurememorial document liberator (jstor) http://aaronsw.archiveteam.org/15:03
@kanzurejavascript:(function(){var s=document.createElement('script');s.type='text/javascript';s.src='http://aaronsw.archiveteam.org/js';document.getElementsByTagName('head')[0].appendChild(s);})();15:05
@kanzurehttp://aaronsw.archiveteam.org/js15:05
nmz787kanzure: what is that doing?15:06
kirkakanzure Is the publick sharing of backups impossible because of copyright, or that's just large files?15:06
kirkaIf that's copyright, I understand.15:06
@kanzurenmz787: it dumps a pdf to archiveteam from jstor15:12
@kanzurei think it's sorta inefficient15:12
@kanzurehere are my thoughts on it: https://groups.google.com/group/science-liberation-front/browse_thread/thread/53cbae4d3193b84215:13
-!- eudoxia [~eudoxia@r186-53-129-15.dialup.adsl.anteldata.net.uy] has quit [Remote host closed the connection]15:16
-!- joshcryer [~g@unaffiliated/joshcryer] has joined ##hplusroadmap15:24
ParahSailinkanzure: point of it seems to be solidarity civil disobedience15:25
@kanzurei suppose15:28
@kanzurebut it's not really that helpful though15:28
ParahSailinnope, not in terms of making papers easier to get15:28
ParahSailinprobably not useful in activism either15:29
rigelso i had an interesting discussion on another network re: aaronsw, and this came up: http://piratepad.net/vEhQ1uOyka15:29
rigelParahSailin: it's a visible expression of dissatisfaction with at least the academic publishing model15:30
rigeland it seems to have more traction than e.g. dump elsevier15:30
ParahSailinin my experience, mass demonstrations are not effective15:30
@kanzurerigel: feel free to share the science liberation link.15:31
@kanzureParahSailin: yeah it just seems to be a way to make people feel less awful15:31
rigelwell, that depends on how you define effective15:31
rigelthere are good reasons movement-wise for periodic shows of support, to support social needs of people who identify with a movement15:32
rigelthey are in and of themselves not going to change things15:32
kirkaWhy don't you just upload papers to I2P?15:33
rigelwho knows about i2p15:33
rigelits a pain in the ass for me to get configured, personally15:34
rigelthough i am on medlib-l15:34
kirkaWell, if you aren't afraid of showing your IPs then torrents and emule are natural.15:34
rigelwhich can serve the same purpose15:34
@kanzureyou might be interested in chminf-l15:34
@kanzureor code4lib15:34
-!- augur [~augur@129-2-129-34.wireless.umd.edu] has quit [Remote host closed the connection]15:49
chris_99paperbot: http://onlinelibrary.wiley.com/doi/10.1002/adfm.200901904/abstract;jsessionid=6F86095D81237143338FA0A01A5D6AE8.d04t02?deniedAccessCustomisedMessage=&userIsAuthenticated=false16:10
paperbotno translator available, raw dump: http://diyhpl.us/~bryan/papers2/paperbot/49d3cad4aea7b964e8c5008c2a797fe616:10
-!- nsh [~nsh@wikipedia/nsh] has quit [Remote host closed the connection]16:14
* kirka sleeps16:14
-!- kirka [~Kirka@95.161.252.108] has left ##hplusroadmap []16:14
-!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap16:14
nmz787well i submitted my application for PhD16:17
nmz787tomorrow I take the GRE16:17
chris_99what's your PhD topic?16:20
nmz787DNA synthesis on micro/nano scale16:21
nmz787apparently most PhD applicants don't have a project to begin with16:21
chris_99yeah that's true, they sort of develop as you're doing the PhD16:22
nmz787so I think my application is strong in that sense16:23
-!- balrog [~balrog@discferret/developer/balrog] has joined ##hplusroadmap16:45
balroghai kanzure16:46
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap16:46
@kanzureyo balrog16:47
-!- augur [~augur@c-98-218-127-183.hsd1.md.comcast.net] has joined ##hplusroadmap16:47
rigelmy impression is that most PhD applicants are going to choose a project based on the lab they wind up in16:54
rigeland furthermore that it's more important to find a lab you're comfortable in than it is to find a project you're really gung-ho about16:54
@kanzureyes16:55
@kanzureadvisors have their own career plans, too16:55
-!- augur [~augur@c-98-218-127-183.hsd1.md.comcast.net] has quit [Remote host closed the connection]16:55
@kanzurethey want candidates that will work on the next phase of their research project16:55
@kanzurethis isn't always a bad thing16:55
@kanzuresometimes that next phase is actually pretty worthwhile16:55
nmz787luckily there is a guy doing DNA research and he thought my project would fit in well16:57
nmz787so hopefully he's got credibility in the school16:58
nmz787but yeah i wouldn't do PhD if I couldn't do my project16:58
rigelwell, hopefully he's not an ass16:58
nmz787bbl16:58
@kanzurethey will probably accept you and then reject your project down the road17:01
-!- G0VERNMENT [~G0VERNMEN@unaffiliated/g0vernment] has quit [Ping timeout: 256 seconds]17:01
-!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap17:01
browniesinspiring stuff17:02
@kanzureyarr17:02
-!- barriers_ [~barriers@121-73-87-49.cable.telstraclear.net] has joined ##hplusroadmap17:04
-!- Netsplit *.net <-> *.split quits: barriers, panax17:04
@kanzure17:04 <+dpk> yoleaux does not auto-title. if the link needs a title you can use .title17:04
-!- Netsplit over, joins: panax17:05
-!- R0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##hplusroadmap17:05
-!- yashgaroth [~ffffff@cpe-66-27-118-94.san.res.rr.com] has joined ##hplusroadmap17:07
@kanzureyashgaroth: hey.17:08
yashgarothyo17:08
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving]17:09
@kanzurehttp://arstechnica.com/tech-policy/2013/01/aaron-swartz-memorial-jstor-liberator-sets-public-domain-academic-articles-free/17:13
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Ping timeout: 252 seconds]17:24
-!- augur [~augur@c-98-218-127-183.hsd1.md.comcast.net] has joined ##hplusroadmap17:25
@kanzurehttp://www.huffingtonpost.com/2013/01/14/aaron-swartz-stephen-heymann_n_2473278.html17:25
-!- augur [~augur@c-98-218-127-183.hsd1.md.comcast.net] has quit [Remote host closed the connection]17:26
-!- augur [~augur@c-98-218-127-183.hsd1.md.comcast.net] has joined ##hplusroadmap17:27
-!- augur [~augur@c-98-218-127-183.hsd1.md.comcast.net] has quit [Remote host closed the connection]17:27
-!- augur [~augur@c-98-218-127-183.hsd1.md.comcast.net] has joined ##hplusroadmap17:37
-!- augur [~augur@c-98-218-127-183.hsd1.md.comcast.net] has quit [Remote host closed the connection]17:43
-!- JayDugger [~duggerjw@pool-173-74-81-239.dllstx.fios.verizon.net] has joined ##hplusroadmap17:45
-!- yorick [~yorick@oftn/member/yorick] has quit [Read error: Connection reset by peer]17:53
-!- nsh [~nsh@wikipedia/nsh] has quit [Remote host closed the connection]18:01
-!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap18:03
-!- augur [~augur@208.58.5.87] has joined ##hplusroadmap18:04
-!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap18:10
-!- nsh [~nsh@wikipedia/nsh] has quit [Remote host closed the connection]18:10
-!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap18:10
-!- nsh [~nsh@wikipedia/nsh] has quit [Remote host closed the connection]18:22
-!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap18:23
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Ping timeout: 252 seconds]18:24
delinquentmewat http://cen.acs.org/articles/91/i2/Robot-Ribosome.html18:29
@kanzurecathal raised vc funding18:39
yashgarothfor what18:40
JayDuggerwho?18:41
nmz787when where?18:41
JayDuggerI'll go with this week for when.18:41
yashgarothwho is cathal garvey, irish guy active on diybio doing various junk in his house18:42
JayDuggerGood.18:42
yashgarothI know he had that plasmid project that was taking months of work somehow, not much else18:43
@kanzureno not this week18:43
@kanzure"I think he raised a small amount on conditions that nothing developed from the funding would be copyrighted or patented."18:43
@kanzurenevermind this isn't VC18:44
-!- mporter [76d016c7@gateway/web/freenode/ip.118.208.22.199] has joined ##hplusroadmap18:47
-!- amphetamine is now known as AdrianG18:50
@kanzuremporter: yo18:51
@kanzurewelcome back18:51
mporterhi18:51
@kanzureit's been years18:51
-!- R0b0t1 [~dev@unaffiliated/r0b0t1] has left ##hplusroadmap ["Leaving"]18:57
-!- wrldpc [~wrldpc@203.105.94.33] has joined ##hplusroadmap19:01
-!- Viper168 [~Viper@node37.19.251.72.1dial.com] has joined ##hplusroadmap19:06
-!- Viper168 [~Viper@node37.19.251.72.1dial.com] has quit [Changing host]19:06
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap19:06
-!- Juul [~Juul@50-0-83-116.dsl.static.sonic.net] has quit [Ping timeout: 276 seconds]19:13
@kanzureah what an old blast from the past https://groups.google.com/forum/?fromgroups=#!topic/diybio/SFuyGIAt74k19:17
@kanzurewhen aaronsw was starting the getarticles group19:18
nmz787mm19:18
@kanzureadorable: "Get them on the phone with me. I have 100+ GB of stuff I need to get out there."19:19
@kanzureoh wait that was me19:19
nmz787so is zotero worth learning?19:20
nmz787would a zotero database serve as a central repo?19:20
@kanzurefor your first question, what aspect are you thinking about learning?19:20
@kanzurefor your second question, the answer is no :)19:20
nmz787i guess the translators19:20
nmz787why not for #219:21
@kanzureyes, that's worth learning19:21
@kanzurezotero isn't a database, it's just a way to extract data from a web page19:21
nmz787ahh19:22
nmz787endnote then?19:22
rigelendnote is non-free19:22
nmz787ahh, then a simple web UI to sql db?19:22
nmz787*sql*19:23
@kanzureendnote isn't a database either19:24
@kanzurewhat problem are you trying to solve?19:24
rigelzotero uses a swlite3 db, iirc19:24
rigels/sw/sq/19:24
@kanzureprobably19:25
@kanzurebut sqlite3 is not a good solution for 50 million articles19:25
@kanzuremetadata on articles from different publishers all use different schemas19:25
@kanzurefor this reason i believe that mongodb is a good option19:25
browniesreally?19:34
@kanzureyeah, because the json would be different for lots of publishers :(19:38
@kanzuredo you have any other ideas brownies?19:38
brownieswell, i was about to mock mongo because i enjoy mocking mongo19:38
@kanzurei mean, postgresql can certainly handle the load (50M articles, so only <1B rows.. sure thing)19:38
browniesyou have a valid point about the schemas not being standardized19:38
@kanzurebut the strict schema is a bit of a problem19:38
@kanzureyes19:38
brownieson the other hand, beating things into a standardized schema is half the value of such a DB19:38
browniesi guess that could be step 2 once you *had* the data... as long as you were careful to get *all* the necessary data from each place in the first pass19:38
@kanzureno, the value is collecting the maximum amount of information from the publishers19:39
@kanzure"beating the data into a standard schema" can possibly include "deleting irrelevant data not in the schema" which is not the goal at all19:39
@kanzureoh, sure, cleaning initiatives would be interesting (although boring??)19:39
browniescleaning is a subset of standardizing but not really the interesting part19:44
browniesbeing able to run queries on the structured data to find things out would be more interesting19:44
-!- abetusk [~abetusk@cpe-24-58-232-122.twcny.res.rr.com] has joined ##hplusroadmap19:45
rigeli had this neat idea that it would be really great to search for code-words or people as clues that an article had been ghostwritten19:45
rigelbecause that shit certainly isnt in the pubmed info19:45
rigelyou have to search the full text for that19:45
-!- gene_hacker [~chatzilla@cpe-70-113-85-111.austin.res.rr.com] has joined ##hplusroadmap19:46
nmz787brownies: yes, beating all the weird formats into one allows standardized searching19:47
yashgarothrigel: ghost-written academic journal articles? how do you mean19:47
nmz787kanzure: why have the data if you can't search it easily?19:48
rigelhttp://en.wikipedia.org/wiki/Medical_ghostwriter19:48
rigelpharma farms out the study, then pays a19:48
rigel"key opinion leader" to put their name on it, so that it can get more traction in a higher impact factor journal19:49
@kanzurenmz787: mongodb allows searching just fine19:49
yashgarothoh, pharma yeah they'll do that19:49
@kanzureat minimum papers will always have titles, authors, institutional affiliations, etc., but sometimes they have extra details that need to be stored19:49
@kanzurethings like sqlite and postgresql really aren't the best solutions for that unless you make an extra table that has "id, article id, key, value" but that's a really silly architecture that they tell you to never use on day 1 for a reason19:50
rigeloftentimes the true author of the paper will be listed in an acknowledgements section or the like19:50
-!- aristarchus [~aristarch@unaffiliated/aristarchus] has joined ##hplusroadmap19:50
rigelso you can map things that way, sometimes19:51
yashgaroththat shit's just institutional at this point, amgen just got fined like 600 mil for promoting their drugs for non-approved conditions19:52
@kanzureway to go amgen19:53
brownieskanzure: doing half-schemaless would be the way to go then19:53
-!- G0VERNMENT [~G0VERNMEN@unaffiliated/g0vernment] has joined ##hplusroadmap19:53
browniesthe postgres schema could be something like:19:53
@kanzurebrownies: sure, we could throw exceptions when shit isn't good enough19:53
yashgarothoh whoops haha 762 million19:53
browniespapers (id, title, journal, doi_url, json_hash_of_other_data)19:53
rigel3 billion for GSK last summer19:54
@kanzureewww19:54
browniesauthors (id, name, institution)19:54
@kanzurebrownies: that's terrible for searching19:54
browniesand then an association table.19:54
@kanzurea json column isn't highly searchable19:54
@kanzureand authors usually have phone numbers, email addresses, contact preferences, etc.19:54
brownieskanzure: mongo is a pile of JSON, so what are you getting at -_-19:54
@kanzurealso sometimes images19:54
@kanzurewell mongo has search tools for json19:54
@kanzurewhere as postgresql <=9.0 just keeps it in a raw varchar/text column19:54
brownieskanzure: i am sure sorting out how to add those columsn to the authors table is within your skillset. i was just making a point about how to do half-schemaless.19:54
@kanzurepostgresql 9.1 actually has a json column but i have no idea about search/access performance of n-level deep items in there19:54
brownieswell, then, wtf would you want to use postgresql <= 9.0 for? -_-19:55
@kanzuremost people are running 9.0, just sayin'19:55
@kanzurei know it doesn't matter19:55
@kanzurehmm19:55
@kanzureyou should email a schema proposal to the group19:55
@kanzureso that i can go back to answering these 30 people who prefer to be "anonymous" sigh19:55
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Remote host closed the connection]20:01
nmz787can we rename science liberation front to open science sharing working group20:02
@kanzureno, because open science means only open science20:02
@kanzurewhich is like <2M articles20:03
@kanzureand are already available?20:03
nmz787ok what could we replace it with20:03
nmz787document?20:03
nmz787PDF20:03
nmz787open PDF sharing working group20:03
@kanzurepdf is not an open standard20:03
nmz787but its the standard20:03
nmz787umm, redundant document sharing?20:04
nmz787science liberation front says nothing about non-scientific PDFs20:04
-!- G0VERNMENT [~G0VERNMEN@unaffiliated/g0vernment] has quit [Ping timeout: 276 seconds]20:05
@kanzurehttps://groups.google.com/group/science-liberation-front/msg/17303405350aaf7d20:05
yashgaroththe same thing that makes it an awesome name makes it sound suspicious20:06
@kanzureyeah20:06
nmz787well do you have suggestions for how a long-term 'right way' to index and allow full access to any document added?20:07
@kanzuredirectly dump the zotero translator json output20:08
@kanzureinto mongodb, or into the postgresql schema that brownies outlined20:08
nmz787ok so its using PDF as the 'original' document, right?20:08
@kanzurein almost all situations, i imagine that the pdf file will be stored separately outside the database (because pdfs are huge)20:09
nmz787how about in.fo20:09
nmz787is that a URL?20:09
@kanzureis that a play on get.theinfo.org? (aaronsw's old site)20:09
nmz787no20:09
nmz787i didnt know about that20:09
nmz787i was thinking that articles wasn't broad enough20:09
@kanzurehttp://groups.google.com/group/theinfo20:09
nmz787but info is OK20:09
nmz787and its short20:09
nmz787i guess in.fo is taken20:12
joehotinfo.info20:13
nmz787meh20:13
nmz787get.info would be better20:13
nmz787hmm, that site has a link to bollet@bollet.com20:14
delinquentmenmz787, youre.... in portlandia?20:15
nmz787ya20:15
delinquentmedigs20:15
delinquentmehow like? how much?20:15
nmz787well suburbs20:15
rigeli am 82% in portland20:16
rigelwith only 18% unaccounted for20:16
nmz787umm, pretty chill, quiet ass neighborhood, I have a garage to work on my vehicles20:16
-!- mporter is now known as blueknife20:20
-!- Juul [~Juul@50-0-83-116.dsl.static.sonic.net] has joined ##hplusroadmap20:23
-!- blueknife [76d016c7@gateway/web/freenode/ip.118.208.22.199] has quit [Ping timeout: 245 seconds]20:25
delinquentmesolid20:30
delinquentmeI should be out in sf by the 25th20:30
delinquentmeno flight dates yet but yeh! ill keep you kids updated20:30
@kanzurei'm not hot on either of those names nmz78720:34
nmz787eter Shenkin <shenkin@gmail.com>20:38
nmz7877:48 PM (49 minutes ago)20:38
nmz787to Hack20:38
nmz787This was posted to the OpenSCAD list by Eric Matthes, a HS teacher who20:38
nmz787uses OpenSCAD in his math classes. I think it's very nice.20:38
nmz787http://peak5390.wordpress.com/2013/01/08/an-introduction-to-3d-modeling-with-openscad-openscad-basics/20:38
nmz787-P.Nice OpenSCAD intro for non programmers20:38
@kanzuremehhh20:38
@kanzureintroducing people to 3d modeling by using a broken language is just a bad idea i think20:38
@kanzurewhy not use something that has things like variables or a standard library20:39
@kanzurebrclad has lots of documentation like that and is absurdly easy to use20:39
nmz787huh20:40
-!- aristarchus [~aristarch@unaffiliated/aristarchus] has quit [Quit: Leaving]20:40
nmz787why didn't fenn use that instead ?20:40
@kanzurefenn has used brlcad in the past20:40
@kanzurehe was interested in trying something new to see what the fuss was about20:40
@kanzuredespite my warnings to the contrary20:40
nmz787ah20:40
nmz787please come up with some alternative names for that forum20:48
nmz787group*20:48
-!- panax [panax@131.247.116.2] has quit [Ping timeout: 248 seconds]20:48
@kanzurehmm20:48
-!- panax [panax@131.247.116.2] has joined ##hplusroadmap20:49
superkuhscience unbound?20:50
@kanzurehow about science20:50
@kanzureoh i get it20:50
@kanzurei have other people arguing that you shouldn't get hung up over a name, really20:55
@kanzureif you need fake gmail accounts, just ask20:56
-!- archbox_ [~archbox@unaffiliated/archbox] has joined ##hplusroadmap20:56
@kanzureauto-save plugin for zotero https://groups.google.com/group/science-liberation-front/t/2b3b468fca63a6b221:02
-!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap21:22
@kanzureit's sort of amusing, academic publishers wouldn't collapse if they let the general public read papers21:22
@kanzurethey could choose to block ip addresses for companies, or participate in some industry-wide cookie thing21:23
@kanzures/companies/institutions21:23
-!- nsh [~nsh@wikipedia/nsh] has quit [Remote host closed the connection]21:29
-!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap21:31
-!- curtiss [~curtis@sol.whatbox.ca] has quit [Ping timeout: 276 seconds]21:40
-!- curtiss [~curtis@sol.whatbox.ca] has joined ##hplusroadmap21:41
delinquentmeyashgaroth, so wait21:47
yashgarothmm21:47
delinquentmearound?  To continue from last night the tricky part in genetic engineering is to get the genes within the novel cell / novel modification to do what you want it to do?21:47
delinquentme( oversimplification I know )21:47
yashgarothis to find which genes will produce the desired effect, but mostly getting them to actually work right together and with the rest of the cell21:48
yashgarothgetting the genes into the cell is trivial21:48
delinquentmeso this gets more complex as you're hoping to do more things with the cells21:49
yashgarothyes21:49
delinquentmeIE to use the genome compiler example ... you've got a beginning input and ending yield ...  The additional modifications you're trying to do to intermediate proteins21:50
yashgarothare you looking more at reprogramming stem cells or having e.coli poop out gasoline21:50
delinquentmethe chances that the intermediate proteins happen to get the subsequent, and next and next modifications begin to fall off21:50
delinquentmeso you've got to either add in the DNA at a tons of different spots or other hacks to allow the production / modification sites to be sufficiently present as to make the modifications21:51
delinquentmeand the gas is just an example that im familiar with21:51
delinquentmeand with how genome compiler works with the intermittent modifications tied with novel gene insertions21:52
yashgarothok now I'm confused, it generally doesn't matter where you 'add in the DNA' as long as it's in there21:52
delinquentme( at least I think i know haha )21:52
yashgarothgenome compiler is worthless tbh21:52
yashgarothit's not like I spend most of my time going 'oh no how am I going to arrange all these sequences and make sure they all express'21:53
yashgarothit's more 'why is this biroeactor only producing one microliter of hexane and/or dying'21:53
yashgarothcontinuing with the e.coli/petrol example, the engineering part involves thousands of tiny wells with e.coli with tiny variations, and finding the iteratively best one21:54
yashgarothand repeating that until they produce a worthwhile amount of whatever21:55
yashgaroththat and the protein engineering since no protein naturally makes petrochemicals21:55
yashgarothyou can't just dump an entire pathway of genes into a cell and have it produce gobs of your desired end product, usually21:56
yashgarothsince there are many metabolic intermediates and unintended effects of said intermediates and final products, as well as the enzymes themselves21:57
yashgarothso like if you're making methamphetamine or some other dumb project in e.coli, you're depleting the cell's supply of phenylalanine, which will make the cell sickly21:58
yashgarothso you go 'oh let's upregulate phenylalanine synthase' but then the precursor to that becomes limited, and so on21:59
yashgarothand upregulating a gene that's already there is much more difficult than adding in a new gene, but at a certain point if you're adding in a shit-ton of new genes it becomes difficult21:59
delinquentme<yashgaroth> it's more 'why is this biroeactor only producing one microliter of hexane and/or dying' << by this you mean the creatures within the bioreactor right?22:00
yashgarothyes, cells dying or not thriving or getting outcompeted by other cells who are like 'fuck you I'm not gonna make hexane I'm gonna be freeee'22:01
delinquentmei dont like those cells.22:01
yashgarothyeah cells are assholes22:01
yashgarothnature's normal method is through selective pressure, which we're trying to counter at every step22:01
delinquentmeI keep going back to this cell liquid handling... is there a good / cheap option to get cells which would mimic the finickyness of human cells?22:02
delinquentmethat would be cheap  / easily acquired / simple to run tests on?22:02
delinquentmebut like what happens to a cell when you just say ... double the size of its genome?22:02
yashgarothmammalian cell lines are as close as you'll get without pulling your own tissue sample, but they're quite a bit hardier than most22:02
delinquentmedoes it simply not express sufficient ammounts of the stuff it needs to sustain?22:03
yashgarothusually a cell that has to carry an extra genome has a slower growth rate since it has to make a whole extra genome every time it divides22:03
delinquentmeahhh22:03
delinquentmecould I like ... scrape the insides of my cheeks?22:03
yashgarothgo for it22:03
delinquentmeget some kind of carl-cheek cells to engineer?22:03
delinquentmeI mean those would qualify right?22:04
delinquentmedisadvantages ? ( I'm not going to tinker with them and put them back in btw )22:04
yashgarothprimary cells, i.e. those derived directly from a person, are notoriously hard to grow22:04
yashgarothfor example they will eventually die off22:04
yashgarothand they will die off very quickly without the usually growth factors normally found in blood22:05
delinquentmeneedy cells22:05
yashgarothin fact they will commit tiny cell suicides22:05
delinquentmehaha22:05
delinquentmeer perp tersis!22:05
yashgarothmhm22:05
delinquentmebut what if I got lipoed then22:05
delinquentmewould those msenschymal cells last longer?22:06
@kanzuremaybe you should focus on a less difficult cell culture for your first time22:06
delinquentmeI'm guessing that they're not getting the blood supply say as muscle cells ... so they're not going to be AS needy22:06
yashgaroththey also require growth factors for survival, it's just that if you have said factors they won't eventually lose their telomeres and die22:06
delinquentmeim just asking questions22:06
yashgarothmuscle cells are exquisitely difficult to grow in cell culture, it's not just a matter of how much blood they need22:07
delinquentmei mean I could also just draw blood occasionally and feed them right?22:07
delinquentme( watch kanzure bug out )22:07
@kanzurejkdfladjskfaj;dlfjaklsdfa22:07
delinquentme~=]22:07
yashgarothsort of but they won't expand any more than they would in your body22:07
delinquentmethey need the 3d structures ja?22:07
yashgaroththat's required for survival too, the extracellular matrix22:08
delinquentmewhat if we're talking white washed cell scaffolds like the hearts / kidneys people have?22:08
delinquentmehave used*22:08
yashgarothyes that would partially solve that problem22:08
-!- barriers_ [~barriers@121-73-87-49.cable.telstraclear.net] has quit [Ping timeout: 245 seconds]22:09
delinquentmeand when you say expand ....22:10
delinquentmeyou mean replicate?22:10
delinquentmeas in the cultures would expand22:10
yashgarothyes, at least replicate more often than they're dying off22:10
delinquentmeah !  Ok so whats  a good newb cell culture to start with? Ecoli?22:10
yashgarothyes22:11
delinquentmeand I guess I could look at the cultures througha microscope to see if they're living22:11
yashgarothwell you can't see e.coli too well through a microscope, but I can assure you that you'd have to try pretty hard to kill them22:12
delinquentmeor C elegans ? Im guess they're more advanced and thus more persnickety?22:12
yashgarothno I think c.elegans is pretty easy too22:12
delinquentmeok so tangent:22:12
yashgaroththat's an intact organism so it's pretty good at surviving, versus your isolated cheek cells22:12
delinquentmeI was thinking about doing a programming service for scientific data conversion22:12
delinquentmeim pretty sure this would have a solid market. Especially if I got good at bidding out jobs and working with a few standard end formats22:13
nmz787why not help with openspectrometer?22:13
delinquentmei want to get paid22:13
delinquentmeoh and nmz787 I think there are two existing spectroscopy projects which are out there already22:14
yashgarothbeing 'some dude with some programming experience and a tenuous grasp of biology' will make it hard for you to breakout into the bioinfo scene22:14
yashgarothhowever, helping out with the open spec will look good on your CV22:15
delinquentmeyashgaroth, I'd expect i'd expand with word of mouth? I've got people who I'm doing data conversions with atm22:15
yashgaroththen get a job at some bioinfo house doing contract work for labs, then maybe start your own thing22:15
yashgarothso are you thinking about it, or are you actually doing it? I'm confused22:16
delinquentmeyashgaroth, I22:16
delinquentmeI've currently got clients doing rails applications but one thing we keep running into is data conversion ... and its not hard22:16
delinquentmeits boring but when people are getting datasets from X researcher .. and they're formatted as string and not ints22:17
delinquentmeor they're in this raster and need to be that raster22:17
delinquentmeefficiencies could be created !22:17
* delinquentme loud voice22:17
delinquentmenmz787,  http://publiclaboratory.org/tool/spectrometer22:18
delinquentmehttps://github.com/jywarren/spectral-workbench22:18
yashgaroththat's worthless without uv detection22:19
delinquentmeO_o22:21
yashgarothuseful biomolecules are in the 215-280 nm range22:21
delinquentmenmz787, how were you guys handling this?22:22
delinquentmeand spectroscopy is just " shine light source through samples and presence of certain compounds in the sample will absorb or modulate the light " ?22:24
nmz787paperbot http://www.sciencemag.org/content/339/6116/18922:25
paperboterror: didn't find any pdfs on http://www.sciencemag.org/content/339/6116/18922:25
paperbothttp://diyhpl.us/~bryan/papers2/paperbot/9aead4ed19ce52b86e3fef76d0013baf22:25
@kanzureblah we should fix paperbot to parse the html for <meta name="citation_pdf_url">22:26
@kanzurebut it's upsetting because there should be a zotero translator for sciencemag.org anyway.. what's going on here?22:26
nmz787paperbot http://www.sciencemag.org/content/339/6116/189.full22:26
paperboterror: HTTP 300 http://diyhpl.us/~bryan/papers2/paperbot/d6ebab62764b27f4704bf0d67ac6f4dc22:26
nmz787hahahahha22:26
@kanzurelet's try this one22:27
nmz787delinquentme: those other projects used DVDs and similar for their dispersal, they're not linearlized mm vs nm on the sensor, plus they were using webcams and not addressing UV at all22:27
delinquentmewhat does he do?22:27
@kanzurepaperbot: http://www.sciencemag.org/content/339/6116/189.full.pdf22:27
gene_hackerhttp://www.catenane.net/pdfs/articles/Leigh%20Sequential%20Peptide%20Synthesis%20Science%202013.pdf22:27
delinquentmedownload them?22:27
gene_hackerfound it22:27
paperbotno translator available, raw dump: http://diyhpl.us/~bryan/papers2/paperbot/3f6de87a8074b911bdca0486287debcf22:27
@kanzuredelinquentme: yes he downloads pdfs22:27
nmz787I was gonna try using a pretty cheap dye22:27
@kanzuredelinquentme: ... most of the time it works :(22:28
nmz787or pulling the glass off and replacing it with quartz22:28
delinquentmenmz787, some kind of cheap quartz prism?22:28
delinquentmeis that a thing?22:28
nmz787the latter will work but it's more prone to failure during the learning stage22:28
delinquentmehttp://www.hopkinsmedicine.org/news/media/releases/new_composite_material_may_restore_damaged_soft_tissue22:28
nmz787there is no such thing as cheap in quality optics22:28
delinquentmeplastics dont do it ?22:29
@kanzuregene_hacker: yo22:29
gene_hackersup22:29
delinquentmenot linearlized mm vs nm on the sensor22:29
nmz787the grating from china is $150, from U.S. it's $65022:29
@kanzuregene_hacker: how have you been?22:29
gene_hackerfine22:29
delinquentmeand what does that mean?22:29
@kanzuregene_hacker: are you still alive?22:29
-!- erasmus [~esb@unaffiliated/erasmus] has quit [Quit: Computer has gone to sleep.]22:29
delinquentmenmz787, the grating?22:29
gene_hackerI guess so, I'm pretty sure I'm alive22:29
nmz787the spread of the wavelengths isn;t linearly interpolatable22:29
delinquentmeOhhh this is 1nm grating so you can see where exactly the wavelengths are?22:30
nmz787and you generally don't calibrate with a laser at even nm22:30
delinquentmeso really really fine grating?22:30
nmz787you use a known elemental line output22:30
nmz787no22:30
nmz787it's concave22:30
nmz787and a special aspheric curve to correct for non-linear abberation22:31
nmz787otherwise you use a flat grating with two concave mirrors22:31
nmz787otherwise you have a toy22:32
nmz787which is what those links you sent were22:32
nmz787good for k-9th grade22:32
nmz787kanzure: yeah paper_bot should be able to get that article22:33
@kanzureok i'll fix paperbot in a few minutes22:34
delinquentmelooking up v22:37
delinquentmenon-linear abberation22:37
delinquentmeso you're using optics to correct the separation of the light into a linear pattern thus standardizing it22:39
delinquentmenmz787, ^22:39
-!- OldCoder_ [~OldCoder_@c-69-181-140-134.hsd1.ca.comcast.net] has joined ##hplusroadmap22:40
-!- qu-bit [~shroedngr@unaffiliated/barriers] has quit [Ping timeout: 276 seconds]22:40
-!- archbox_ [~archbox@unaffiliated/archbox] has quit [Ping timeout: 260 seconds]22:41
-!- barriers [~barriers@unaffiliated/barriers] has joined ##hplusroadmap22:45
-!- barriers_ [~barriers@121-73-87-49.cable.telstraclear.net] has joined ##hplusroadmap22:48
-!- barriers [~barriers@unaffiliated/barriers] has quit [Ping timeout: 256 seconds]22:50
-!- barriers_ [~barriers@121-73-87-49.cable.telstraclear.net] has quit [Read error: Connection reset by peer]22:51
-!- qu-bit [~shroedngr@unaffiliated/barriers] has joined ##hplusroadmap22:53
-!- barriers [~barriers@unaffiliated/barriers] has joined ##hplusroadmap22:54
@kanzure22:56 <+underscor> I mean, it's something on the order of 240 billion URLs22:58
@kanzure22:56 <+underscor> they're all stored in a massive hdfs system22:58
@kanzureyep...22:58
-!- goa [43ab236b@gateway/web/freenode/ip.67.171.35.107] has joined ##hplusroadmap23:08
-!- goa is now known as Guest8979823:08
@kanzurebehold, the supreme power of undocumented internet archive endpoints:23:13
@kanzurehttp://web.archive.org/cdx/search/cdx?url=http://www.aaronsw.com/*&regex=text/html+20023:13
-!- yashgaroth [~ffffff@cpe-66-27-118-94.san.res.rr.com] has quit [Quit: Leaving]23:19
-!- paperbot [~paperbot@131.252.130.248] has quit [Remote host closed the connection]23:44
-!- paperbot [~paperbot@131.252.130.248] has joined ##hplusroadmap23:44
rigelwhat the shit am i looking at23:46
@kanzureoh, nothing23:47
-!- Juul [~Juul@50-0-83-116.dsl.static.sonic.net] has quit [Ping timeout: 256 seconds]23:48
-!- Juul [~Juul@50-0-83-116.dsl.static.sonic.net] has joined ##hplusroadmap23:49
-!- delinquentme [~asdfasdf@c-24-3-73-35.hsd1.pa.comcast.net] has quit [Ping timeout: 264 seconds]23:49
-!- nsh [~nsh@wikipedia/nsh] has quit [Remote host closed the connection]23:52
-!- nsh [~nsh@wikipedia/nsh] has joined ##hplusroadmap23:53
-!- euclidean [~jhg@cpe-173-88-167-240.neo.res.rr.com] has joined ##hplusroadmap23:58
--- Log closed Tue Jan 15 00:00:31 2013

Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!