2014-11-23.log

--- Log opened Sun Nov 23 00:00:59 2014
-!- lichen [~lichen@c-50-139-11-6.hsd1.or.comcast.net] has quit [Quit: Lost terminal]00:37
-!- ebowden_ [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has quit [Ping timeout: 264 seconds]00:55
-!- gene_hacker [~chatzilla@c-50-137-46-240.hsd1.or.comcast.net] has quit [Ping timeout: 264 seconds]02:28
-!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 255 seconds]02:30
-!- lichen [~lichen@c-50-139-11-6.hsd1.or.comcast.net] has joined ##hplusroadmap02:36
-!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has joined ##hplusroadmap02:44
-!- Qfwfq [~WashIrvin@unaffiliated/washirving] has quit [Ping timeout: 255 seconds]03:11
-!- ebowden_ [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has joined ##hplusroadmap03:15
-!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has quit [Read error: Connection reset by peer]03:15
-!- Qfwfq [~WashIrvin@unaffiliated/washirving] has joined ##hplusroadmap03:16
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap03:17
archelshttp://www.gravitricity.com/03:18
archelshm, one 100 kilotonne weight raised and lowered over 100 m could supply 1 kW for 24 h03:20
archelsthat's not so much03:20
chris_99heh, yeah that does seem rather low03:22
-!- genehacker [~chatzilla@c-50-137-46-240.hsd1.or.comcast.net] has quit [Ping timeout: 272 seconds]03:24
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 258 seconds]03:31
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap03:32
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-jbxrxtkvqsjoaejr] has joined ##hplusroadmap03:38
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-jbxrxtkvqsjoaejr] has quit [Client Quit]03:38
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-mvwohobrnmxyhuwh] has joined ##hplusroadmap03:40
-!- Lemminkainen [uid2346@gateway/web/irccloud.com/x-kezbuclhxdnxjluc] has quit [Quit: Connection closed for inactivity]03:52
-!- drewbot [~cinch@ec2-54-160-187-179.compute-1.amazonaws.com] has quit [Remote host closed the connection]04:28
-!- drewbot [~cinch@ec2-54-89-56-222.compute-1.amazonaws.com] has joined ##hplusroadmap04:28
-!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has joined ##hplusroadmap04:56
-!- ebowden_ [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has quit [Read error: Connection reset by peer]04:57
kanzureheath: yeah i don't recommend jenkins at all. don't use it.05:17
kanzurehttps://uk.news.yahoo.com/cancer-patients-driven-darknet-cheap-drugs-120943973.html#B2n9aWn05:28
-!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 264 seconds]05:40
-!- Vutral [ss@mirbsd/special/Vutral] has joined ##hplusroadmap05:48
archelsoh, hah, the FBI is actually serving up "This hidden site has been seized" images on those captured onion sites05:54
kanzuredoxbin is only 50% of the time showing that05:55
kanzurethe other 50% it is showing doxbin :)05:55
cuba_does it still work?05:57
kanzurehaven't tried, but my informants say yes05:58
cuba_the post on the tor mailing lists is quite interesting05:59
kanzurelink06:04
-!- Zinglon [~Zinglon@5351F2A8.cm-6-2d.dynamic.ziggo.nl] has joined ##hplusroadmap06:08
cuba_https://lists.torproject.org/pipermail/tor-dev/2014-November/007731.html06:18
paperbotKeyError: 'title' (file "/srv/ikiwiki/paperbot/modules/papers.py", line 170, in download)06:18
-!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has quit [Remote host closed the connection]06:19
archels"Kowloon Walled City expressed in php" haha06:24
kanzure.title06:27
yoleaux[tor-dev] yes hello, internet supervillain here06:27
-!- Merovoth [~Merovoth@gateway/tor-sasl/merovoth] has quit []06:30
-!- Merovoth [~Merovoth@gateway/tor-sasl/merovoth] has joined ##hplusroadmap06:38
kanzuresalome meshing stuff (related to opencascade) https://github.com/tpaviot/smesh06:49
kanzure"In Finland we have an ongoing FOI request process with Finnish universities. They declined our request for information on licensing fees paid to Elsevier and other publishers, and we have appealed to court. The arguments seem to be on our side however the results are yet to be seen. Preliminary information of the process is available in Finnish at https://github.com/okffi-science/2014-tietopyynto-lisenssimaksut we are planning to blog more ...06:52
kanzure... on this later."06:52
kanzure"I also heard rumours that Finland recently had similar negotiations with Elsevier than France but no more information on this."06:52
kanzurehttp://blog.okfn.org/2014/11/11/france-prefers-to-pay-twice-for-papers-by-its-researchers/06:52
kanzurefinnish elsevier FOI stuff was via Leo Lahti <leo.lahti@iki.fi>06:54
kanzurehttp://www.gwern.net/Black-market%20survival "I compile a dataset of 53 black-markets in the vein of the famous Silk Road, recording their openings/closing and relevant characteristics. A survival analysis indicates the markets follow a Type TODO lifespan"... such analysis.07:01
-!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap07:10
kanzure.wik mckendree cylinder07:29
yoleaux"A McKendree cylinder is a type of hypothetical rotating space habitat originally proposed at NASA's Turning Goals into Reality conference in 2000 by NASA engineer Tom McKendree. As with other space habitat designs, the cylinder would spin to produce artificial gravity by way of centrifugal force. The design differs from the classical  …" — http://en.wikipedia.org/wiki/McKendree_cylinder07:29
kanzureoh right07:29
kanzure"Organ donation pisses me off. What do I get out of it? What does my family get out of it? Their certainly is money to be had. The insurer, the medical equipment manufacturers, the hospital, probably the doctors, and plenty of other people are getting fat on profit off of my death. I don't even want money. I want my family to be at the top of those lists. My nieces and nephews, parents, children, spouse, and inlaws. The entire set up reeks of ...07:43
kanzure... hypocrisy and misdirection. I personally refuse to donate because it pisses me off. When you show me a dying kid to shame me into donating, it's my kid that would have to die. It's so insensitive. It doesn't even mean what people say it means. It's insulting that they act as if corneas and skin grafts "save lives." Then I cannot even be paid for giving up my organs. Shit, I just died. What do you think my wife and kid are going through? ...07:43
kanzure... How is she going to pay off that mortgage on her own? It isn't like we are questioning the ethics of everyone else getting paid in this process. Then you want to force me or trick others into it? It's bullshit."07:43
kanzure.title http://triblive.com/news/allegheny/4408091-74/organ-organizations-procurement#axzz3JrqCZ9ml07:45
yoleauxGift of life worth millions to donation organizations | TribLIVE07:45
kanzure"So why not create a scenario where a donors surviving heirs are compensated? Place the $100k worth of organs harvested from dad into the estate for his kids. I'm saying take the profits from the hospitals and give it to heirs. You can figuratively and literally let somebody liquidate the purest form of capital any human has when they die, their body."07:46
kanzure"Compensating the donor's family could lead to "accidents" around the house."07:46
kanzuremeh that's no different from life insurance though07:46
kanzure"Because the National Organ Transplant Act forbids the creation of binding contracts for organ transplant, steps in the procedure had to be performed roughly simultaneously. Two pairs of patients means four operating rooms and four surgical teams acting in concert with each other. Hospitals and professionals in the transplant community felt that the practical burden of three pairwise exchanges would be too large.[26] ..... A 12-party (six ...07:51
kanzure... donors and six recipients) kidney exchange was performed in April 2008.[27][28]"07:51
kanzurewtf07:51
-!- big3_brother [~big3_brot@187-032-043-006.static.ctbctelecom.com.br] has joined ##hplusroadmap07:54
-!- snuffeluffegus [~snuff@5.150.254.180] has quit [Remote host closed the connection]08:05
-!- big3_brother [~big3_brot@187-032-043-006.static.ctbctelecom.com.br] has left ##hplusroadmap ["Leaving"]08:11
-!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 250 seconds]08:13
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-mvwohobrnmxyhuwh] has quit [Quit: Connection closed for inactivity]08:17
kanzurehttp://scholarlyoa.com/2014/11/20/bogus-journal-accepts-profanity-laced-anti-spam-paper/08:19
kanzure"A scientific paper titled "Get Me Off Your F****** Mailing List" was actually accepted by the International Journal of Advanced Computer Technology."08:20
kanzurehmmm http://www.scs.stanford.edu/~dm/home/papers/remove.pdf08:20
kanzurenot very funny08:20
kanzureoh right, they probably meant the publisher that was spamming them08:21
chris_99shows what a joke some 'publishers' are08:28
kanzurescholarlyoa's whole thesis is that they are just spam rings08:29
kanzureso not even jokes, just typical spammers08:29
chris_99looks like a way to collect money apparently08:30
chris_99"Instead, they automatically accepted the paper — with an anonymous reviewer rating it as "excellent" — and requested a fee of $150."08:30
-!- superkuh [~superkuh@unaffiliated/superkuh] has quit [Ping timeout: 245 seconds]08:36
chris_99i sent the editor an email, i'm curious if they respond08:47
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-fvkmdegpkvgjhclk] has joined ##hplusroadmap08:56
-!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 265 seconds]09:10
-!- Vutral [sebAtZZaDu@p5B2A6621.dip0.t-ipconnect.de] has joined ##hplusroadmap09:12
-!- Vutral [sebAtZZaDu@p5B2A6621.dip0.t-ipconnect.de] has quit [Changing host]09:12
-!- Vutral [sebAtZZaDu@mirbsd/special/Vutral] has joined ##hplusroadmap09:12
-!- Vutral [sebAtZZaDu@mirbsd/special/Vutral] has quit [Excess Flood]09:12
-!- Vutral [ss@mirbsd/special/Vutral] has joined ##hplusroadmap09:18
-!- Alloran [~Alloran@cpc68288-cdif17-2-0-cust713.5-1.cable.virginm.net] has joined ##hplusroadmap09:22
kanzurehello Alloran09:46
Alloranhello there09:46
kanzurewhat brings you here09:47
Allorani have no idea what this channel is, i just joined a bunch of random ones09:47
Alloranwhat's it all about?09:48
kanzuresee /topic09:48
Alloranoh yeah. so is this like a hypothetical discussion thing or are you guys actually doing this stuff? seems pretty far out09:49
kanzurethere are various projects with varying levels of realness09:49
Alloranneat09:49
kanzureare you justanotheruser?09:49
AlloranI'm just me. I'm new on freenode09:50
kanzureoh right, you're not at purdue09:51
Allorani dont know what that is09:52
-!- yashgaroth [~ffffff@2606:6000:cb85:6a00:5968:1bb1:bb69:2b18] has joined ##hplusroadmap10:00
-!- Alloran [~Alloran@cpc68288-cdif17-2-0-cust713.5-1.cable.virginm.net] has left ##hplusroadmap ["Leaving"]10:18
-!- Alloran [~Alloran@cpc68288-cdif17-2-0-cust713.5-1.cable.virginm.net] has joined ##hplusroadmap10:55
-!- juri_ [~juri@vpn166.sdf.org] has quit [Ping timeout: 240 seconds]10:58
-!- juri_ [~juri@vpn166.sdf.org] has joined ##hplusroadmap10:59
-!- Alloran [~Alloran@cpc68288-cdif17-2-0-cust713.5-1.cable.virginm.net] has quit [Quit: Leaving]11:07
fennopsec fail11:19
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Ping timeout: 240 seconds]11:35
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 255 seconds]11:36
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap11:37
-!- cpopell [~cpopell@c-76-26-144-132.hsd1.dc.comcast.net] has quit [Ping timeout: 264 seconds]11:38
-!- cpopell [~cpopell@c-76-26-144-132.hsd1.dc.comcast.net] has joined ##hplusroadmap11:38
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap11:43
fenni never understood why o'neill cylinders had windows - surely it makes more sense to point a telescope at the sun and pipe the light in through the center11:54
fennotherwise you'd have a light fluctuation with the same period as the cylinder rotation, which is on the order of minutes instead of 24 hours11:55
-!- superkuh [~superkuh@unaffiliated/superkuh] has joined ##hplusroadmap11:55
fennalso you lose half the internal area to windows, and have to worry about micrometeorites11:57
fennah orionsarm gets it right11:58
chris_99http://www.ualberta.ca/~bsuther/papers/procsorrento0998/reprint_style.pdf is really cool12:00
chris_99(Synthetic schlieren)12:00
-!- Zinglon [~Zinglon@5351F2A8.cm-6-2d.dynamic.ziggo.nl] has quit [Remote host closed the connection]12:00
fenngravitricity is good for long-term storage with no leakage, but at such small scales ~24kWh it's not useful as a storage medium, so this means it would be used as a rapid cycing demand buffer similar to how capacitors work in electronic circuits. the same underground volume could store much more energy in high speed flywheels, but they'd need to be stronger in tension than a pile of rocks12:08
fennenergy leakage*12:09
fennflywheels are more efficient in a vacuum, so you'd want to pump down the whole shaft to vacuum, and section it off for upgrades and maintenance12:11
fennyou could transport the flywheel mass as a long tape through an auxiliary shaft with tape reels12:11
fennultra high speed flywheels turn out to share optimal material characteristics with space tethers, since the characteristic velocity is the most important factor determining energy density12:13
fennit might make sense for orbiting tethers to use flywheels as a power storage buffer between eccentric orbits12:14
kanzuremaaku: can you describe your general method or process that you used (or would use) for cost estimation related to brain emulation?12:15
fennunused tether mass (otherwise used for lifting maximum mass payloads) can be transformed into flywheel mass to increase efficiency of solar power to orbital velocity by eliminating battery storage12:16
fennsince you only boost an eccentric orbit at certain times12:18
-!- genehacker [~chatzilla@8-12.ptpg.oregonstate.edu] has joined ##hplusroadmap12:21
kanzuregenehacker: why is nobody using those "mofs" and "sbus" for arbitrary shape synthesis?12:21
genehackerwhat do you mean by arbitrary shape synthesis?]12:22
genehackerdo you mean like molecular polyhedra and stuff?12:22
fennmolecular nanotechy things12:22
kanzuremolecular polyhedra stuff12:22
genehackeralready been done12:22
genehackerlet me find the paper12:22
kanzurefor example, arbitrary number of edges and surfaces12:22
fennseems more useful as scaffolding to me12:22
kanzuredifferent shapes, like extruded three-dimensional alphabet symbols12:22
genehackerhttp://www.researchgate.net/publication/233957695_Structural_design_principles_for_self-assembled_coordination_polygons_and_polyhedra12:23
genehackerwe can do polyhedra12:23
genehackerdifferent shapes, well getting them to self assemble is hard12:23
kanzurewhat about single-molecule shapes though?12:23
genehackerbut some people have suggested that you could do molecular 3d printing with MOFs ;)12:24
fennpaperbot: http://www.researchgate.net/publication/233957695_Structural_design_principles_for_self-assembled_coordination_polygons_and_polyhedra12:24
kanzureright.. you could push different polyhedrons around with a tip.. i guess.12:24
genehackerthere is a process to grow MOFs by stamping them12:24
genehackerresolution isn't very high, but point is you can do it12:25
fennstamping what?12:25
fenna precursor solid?12:26
genehackerso you know MOFs right?12:26
fennnot really :\12:26
kanzurehttp://diyhpl.us/~bryan/papers2/chemistry/metal-organic-frameworks/12:26
genehackeryou got your organic linkers and nodes12:26
fenni know it's a mildly hierarchical assembly pattern of metal ions and (usually) organic ligands12:27
genehackerlinkers are an organic chain with stuff on the end that bonds to metal ions to form nodes12:27
fennare zeolites MOFs? (or the inorganic equivalent)12:28
genehackeryes12:28
genehackerand they are much harder to synthesize12:29
fennok so stamping12:29
genehackeryou can spend years trying to synthesize the zeolite you want12:30
genehackerwith MOFs you can just find the ligand you want on a chemical site, order it and you're ready to go12:30
genehackerhere's some stuff on stamping http://www.sciencedirect.com/science/article/pii/S136970211270046912:32
fennpaperbot: http://www.sciencedirect.com/science/article/pii/S136970211270046912:32
paperbothttp://diyhpl.us/~bryan/papers2/paperbot/%0A%20Epitaxially%20grown%20metal-organic%20frameworks%0A%20.pdf12:32
fennwell that's a lie, but it's better than nothing12:33
fennhey the cover art on materials today is tallakahath's thing (?)12:34
fenn.img http://ars.els-cdn.com/content/image/1-s2.0-S1369702112X70020-cov150h.gif12:35
yoleauxhttp://inweh.unu.edu/archive/images/1-s2.0-S1353829212X00030-cov150h.gif (more: http://is.gd/EDsx2J)12:35
genehackerwhat does tallakahath do?12:35
fenni guess .img doesn't do reverse image search12:35
fenncarbon nanotubes, 2 photon micro 3d printing http://www.elizabethdecolvenaere.com/research.html12:37
genehackeractually I might be wrong, I don't know if we can grow MOFs layer by layer yet, but there's no reason we shouldn't be able to12:37
fennfile is from 201112:38
genehackerI just hope no one from st-dd-rt group is in this channel12:38
fennnow you've doomed us all12:40
fennis this a certain mechanostereochemist at northwestern?12:41
genehackeryes12:42
fenni wish there were more cross-eye structure illustrations in chemistry12:43
kanzureah so that was what the party was about last night12:47
kanzurethe internet is just a large coincidence detection engine12:49
fennjello MOF shots?12:49
fennadd quinine for lasing activity12:50
kanzuretallakahath didn't say, busy partying (or giving a campus tour now? i don't know)12:50
genehackerI still want to synthesize edible cd-mof12:50
fenncompetitive BBQ?12:50
fennthe materials today cover was from 201212:52
kanzurewell, there's another one apparently12:53
fennlink or it didnt happen12:53
fennalso i hate sciencedirect with a passion12:53
kanzurelink will come later ('cause i don't have it)12:53
kanzurethis page does not have enough linkjuice:12:54
kanzure"remaining technical challenges for diamondoid mechanosynthesis stuff" http://www.molecularassembler.com/Nanofactory/Challenges.htm12:54
genehackerhas that page been updated recently?12:55
kanzure2009 :(12:56
fennwtf 27 hits12:56
fennthat must be broken12:56
kanzureoh, 2010. because the patent.12:58
fennwhat patent?12:59
kanzurefreitas' molecular nanotechnology patent13:00
kanzure"on a method for manufacturing the DCB6 carbon dimer placement tool"13:00
fenn.title https://encrypted.google.com/patents/US827621113:02
yoleauxPatent US8276211 - Positional diamondoid mechanosynthesis - Google Patents13:02
fenn^2011 tho13:02
kanzurewell, i meant the page update13:02
kanzurefiling happened in 2004, page updated in 2010 about the patent for some reason.13:03
fennhttp://www.molecularassembler.com/Papers/US7687146.pdf13:03
fenn^201013:03
kanzureer, what was the original reason for molecule-scale gears?13:03
-!- Boscop [me@unaffiliated/boscop] has quit [Read error: Connection reset by peer]13:03
fennextremism13:04
-!- Boscop [me@unaffiliated/boscop] has joined ##hplusroadmap13:04
fennMOF legos seems much more feasible13:04
kanzurethose are also pretty small, though13:04
fennor other nanoparticles, there are a lot of different ways to make regular tileable modular structures13:05
genehackeryes13:05
kanzureright... seems to me that you can get most of the way there just by pushing around large polyhedrons.13:05
genehackerand there is this: http://metamodern.com/2009/02/15/nanomaterials-nanostructures-and-stiffness/13:06
genehackerMOFs might not be stiff enough, but their counterparts organic frameworks might be13:06
fennwhy can't you just lower the temperature to get rid of thermal fluctuations?13:07
fennor is there some rate limiting heat generation in the process of assembly that makes this silly13:07
-!- Alloran [~Alloran@cpc68288-cdif17-2-0-cust713.5-1.cable.virginm.net] has joined ##hplusroadmap13:11
fennpolyhedrons definitely reduces the "sticky fingers" problem13:11
kanzurei am not sure you really need functionalized tooltips... it is enough to have functionalized molecules that you push around.13:12
kanzurewhenever i read stuff from freitas/merkle/drexler i get the sense that a lot of this could be avoided if some chemist greybeard would sit down with them for a few hours13:13
-!- Alloran [~Alloran@cpc68288-cdif17-2-0-cust713.5-1.cable.virginm.net] has quit [Read error: Connection reset by peer]13:14
fennhuh i wonder why bitcoin-wizards in particular was spammed by freenode for dogecoin donations13:15
kanzureyeah i think that was a fluke13:15
kanzureor not freenode-approved13:16
-!- Alloran [~Alloran@cpc68288-cdif17-2-0-cust713.5-1.cable.virginm.net] has joined ##hplusroadmap13:16
kanzureso a lot of the tooltip assembly stuff seems to require lots of work on positional accuracy.. but if they would just be okay with 100x larger objects, it would work now.13:18
-!- Alloran [~Alloran@cpc68288-cdif17-2-0-cust713.5-1.cable.virginm.net] has quit [Read error: Connection reset by peer]13:18
* fenn mumbles something about lego self replicators13:19
genehackerit's been done with magnets and cheating13:20
fennif you have a sufficiently precisely flat metal plate, and remove all the oil and surface gunk, and put it on another plate similarly prepared, they will spontaneously weld together at room temperature13:21
fennwell, it works for steel at least13:21
fennapparently this is a problem in high precision machine shops13:21
fennegads "one starts a program with a small time-lock puzzle which must be solved before the program does anything evil, in the hopes that the antivirus scanner will give up or stop watching before the puzzle has been solved and the program decrypts the evil payload; the puzzle’s math backing means no antivirus software can analyze or solve the puzzle first."13:25
fennthere must be some cute name for this, like "halting hell"13:26
-!- snuffeluffegus [~snuff@5.150.254.180] has joined ##hplusroadmap13:27
TMAfenn: with lead the sufficient precision and cleanliness can be easily obtained by an ordinary file -- which makes it a great demonstration for high school physics class13:28
-!- snuffeluffegus [~snuff@5.150.254.180] has quit [Remote host closed the connection]13:38
kanzurehere is a thing from eric hunting from a few minutes ago: https://groups.google.com/d/msg/openmanufacturing/IpXb-nUN5w0/QgOYwvRW6cMJ14:14
maakukanzure: the WBE roadmap + Kurweil curve fitting14:14
maakukanzure: plus my own assessment that a neural network like simplification of the neural model would not be sufficient for emulation, but rather there exists some degree of complexity at the level of neural cells14:16
kanzurehmm14:17
kanzureso i would think a good estimation should involve things like cost of labor inputs, cost of hardware, cost of software (not all software costs the same, some of it is much more easily built compared to other chunks, and not all chunks are difficult/impossible/as-risky)...14:17
maakumeaning that real-time WBE requires k * O(N + S), where N = number of neurons, S = avg number of connections, and k >> 114:18
maakuoh definately14:18
kanzureand then natural fudge factors like how terrible we are all at project estimations, hehe14:19
maakuthe point was more that even if you assume Moore's law holds, it would be difficult to do WBE before late 2030's at the earliest14:19
maakuand my own inside-view analysis put full recursive AGI before then, so I didn't bother costing WBE more accurately14:19
kanzurehmm, i remember reading some stuff from henry markram that he expects current commodity hardware to be sufficient14:19
kanzureby remember i guess i mean i don't remember. can't think of where i would find that quote from him.14:20
fennwe just don't know which parameters are important14:23
fennemulation != simulation14:23
maakufenn: right, but (1) it seems likely to me that simulation will still require a k value >1, if not >>1, and14:25
maaku(2) it seems highly unlikely that we will have learned enough to do such a simulation before we have the capability to do full emulation for testing purposes14:25
fennwe are already doing partial simulations, so you can very accurately estimate that value for whole brain simulations14:25
maakuwe are not doing simulations at the level of detail required to learn that kind of information14:26
fennwhat kind of information?14:26
maakuhow to reduce the level of computation required by orders of magnitude by making non-physical simplifications14:26
maakuit's interesting work, but to me it seems orders of magnitude more difficult that straight artificial general intelligence14:27
kanzuremany of the biologically-accurate neuron emulators right now are already using simplifications14:27
kanzureso by difficult what do you mean though14:27
fennthe problem with going from simulations to emulations right now is that we look at the simulation running, and it looks like a simulation, and that's it; you can't ask it "how's it going in there?" because it's basically just a brain slice14:27
kanzurewhat?14:28
kanzurethere are many places you can plug into a human brain and get visual output from either the eyes or a distorted signal from the eyes14:28
kanzureso why wouldn't that be true in non-emulated systems?14:28
fennall you can do is compare statistical distributions, like "is it in alpha state", which tells you nothing about the qualia or whatever14:29
maakukanzure: by 'difficult' i mean timeline, ultimately14:29
kanzurehmm14:29
kanzureso,14:29
kanzurecan that timeline be accelerated with more parallel work?14:29
kanzurei assume you mean human labor timeline stuff14:29
maakuit is my non-consensus view that we have a workable framework for AGI, one which could be completed a few short years if we had tons of funding, a decade or so otherwise14:30
maakuso I personally see WBE as just so far outside of that time horizon that I don't bother keeping up with it14:30
fennyeah, well that's just like, your opinion, man14:31
maakuif you think we can do an upload + WBE within ten years, then by all means I'd be happy for you to convince me of it14:31
kanzurethat seems totally crazy to me. you should investigate your options before making judgements.14:31
* fenn abides14:31
kanzurehow are you supposed to plan without data14:31
kanzureblah14:31
maakukanzure: wtf? I just said I made that judgement off of the WBE roadmap14:32
fennmaaku: unfortunately people have been saying that since the dawn of the computer age14:32
maakufenn: at some point the people saying it will be right. the trick is knowing the difference14:32
* kanzure looks again at file:///home/kanzure/Downloads/paper_42.pdf14:32
fennit may actually be true now, but how do we estimate the likelihood of something unprecedented?14:32
kanzurew shit14:32
kanzureaw shit14:32
kanzure(i mean the paper "Bridging the Symbolic/Subsymbolic Gap")14:33
maakufenn: inside view. for sixty years people have been making those predictions based on hope. "it *feels* like it's ten years away!"14:33
kanzure"inside" of what14:34
maaku(30 years is actually more commonly the prediction -- 30 years is sufficiently far in the future that everything "feels" 30 years away)14:34
kanzurelike why do you think minsky hasn't been able to solve this14:34
maakukanzure: http://lesswrong.com/lw/vz/the_weak_inside_view/14:34
maakubecause minsky is not working on a constructivist framework14:34
kanzuremaybe minsky is the wrong choice for me to make here14:34
maakuhis society of mind theory is a very good analysis tool, but it doesn't result is a workable architecture14:35
maakuas his grad student's work have shown14:35
maakuthis, on the other hand, is a decent architecture: http://wiki.opencog.org/w/CogPrime_Overview14:36
maakuand you can walk though Society of Mind or The Emotion Machine and for each chapter figure out how what Minsky is describing could be implemented inside CogPrime14:36
kanzurehmmm http://goertzel.org/MonsterDiagram.jpg14:37
kanzureyeah i don't really buy "minsky has only been motivated to think about society of mind ideas for 40 years"14:38
maakuNow I don't take CogPrime as gospel. I actually think Goertzel made too many simplifications in some parts (e.g. his logical reasoning framework), and the OpenCog implementation is shit.14:38
maakukanzure: ?14:38
kanzurecan you elaborate on the shittyness please?14:38
maakuIt's non-recursively self-modifiable, and ill-performant for such general inteligence tasks14:39
maakuOpenCog is basically a grab-bag of narrow AI systems glued together .. which is fine because that's what the CogPrime architecture is about.14:39
maakuBut in practice it is important that the subsystems be implemented within a general, reflective environment so that the system can reason about itself14:40
maakuRather, in OpenCog you have to explicitly re-introduce to the system any aspects of its operation that you want it to reason about, which really makes any self-improvement a mechanical turn operation14:42
maaku*mechanical turk14:42
maakuE.g. in that giant diagram you posted, each box is a separate C++ program that interacts with the data stored in the center box, the atomspace14:42
maakuI would much rather have those implemented within the system, using its own virtual machine language such that the specialized programs are themselves part of the atomspace14:43
kanzureand then interrupts trigger different programs in that pool of "stored vectors"?14:44
maakuEffectively14:44
maakuNot really interrupts -- there's a concept of "attentional allocation currency", and a bidding market for CPU time14:45
maakuso really there's thousands, or tens of thousands of processes in motion, and the task switcher jumps between those which offer the most attentional currency14:45
maakubut that's deep in the weeds14:45
maakuI could criticise just about every component of the monster diagram, in terms of how it is implemented in OpenCog, but the diagram itself I mostly adhere to14:46
fennmaaku: people have been saying "10 years away" for forever, not "30 years away"14:46
maakuIt's an architecture which would work, and which could be implemented in 5-10 years14:47
kanzurehow do you know it would work?14:47
fennhow do you arrive at this number?14:47
maakufenn: https://intelligence.org/files/PredictingAI.pdf14:47
fennwhat.14:48
kanzureokay, so you're saying "since everyone sucks at making predictions, i can make any prediction i want"?14:48
fenn"using a database of 95 AI timeline predictions" seriously?14:48
fennthis is not the sort of thing you just do a cochrane review and take the average value14:48
kanzurei blame steve14:49
maakukanzure: I pulled out a psych textbook, and society of mind, and flipped through the pages. and was surprised to find that I could figure out how each aspect of human thought could work in the CogPrime architecture14:49
kanzurehis name isn't on this paper but i blame him anyway14:49
maakukanzure: no, read the paper. the point is that for certain predictions asking experts is the worst possible thing you can do to get a reliable estimate (arrival of AGI falls in this category)14:49
maakuthe weak inside view is the alternative : http://lesswrong.com/lw/vz/the_weak_inside_view/14:50
kanzurethat doesn't mean ask non-experts or make up numbers14:50
kanzureit means you suck at estimating14:50
fennobviously, since the experts have been saying "in the next 10 years" for the last 60 years14:50
kanzurei think maaku is just trying to say "nobody has even tried to be comprehensive"14:51
kanzure"with the exception of opencog"14:51
kanzurejust because nobody has tried to be comprehensive does not mean that you gain magical estimation powers....14:52
maakukanzure: yes, essentially14:52
maakui haven't seen any other system except Novamente/OpenCog/CogPrime which tries to be comprehensive14:53
kanzurehaha that paper cites dani eder14:54
fennthat "weak inside view" article was pointless14:54
maakui've done my independent analysis of CogPrime, and am mostly in agreement that it would work as advertised14:54
kanzureand this agreement is based on "i have taken an index of things i found in a psychology textbook, and found plausible implementations of those things in the proposed software architecture that could account for those things"?14:55
maakuyup. i could be wrong!14:55
maakubut what else would you suppose I do?14:56
fennsay "i don't know"14:56
kanzuremore carefully evaluate the costs and plausibility-risks of emulation/simulation14:56
kanzurethat is what i would suppose you to do14:56
kanzurehaving said that,14:57
kanzurewhen i last looked at opencog it was a giant pile of software14:57
kanzuresprawling everywhere14:57
maakukanzure: ok answer me this: is there any plausible scenario where WBE could be achieved within 10 years?14:57
kanzureif you believe your implementation ideas can be expressed more concisely than opencog's, that might be advantageous even if it tunrs into 6 months of wasted programming14:57
kanzureyes, i think 10 years is more than enough time14:57
fennif "braincoin" economically incentivizes neural emulation asics...14:58
kanzurealso, i would argue that whole brain may be unnecessary and this might save a lot of effort14:58
kanzure(partial brain etc)14:58
maakukanzure: ok. i'm willing to be convinced of that, although IRC is probably the wrong medium.14:58
kanzureirc is a fine medium for it, i just don't have any data prepared to show you14:58
kanzureif i was to show you numbers and estimates, what things were most concerning to you in the past for that sort of direction?14:59
maakumy own reading (e.g. the WBE roadmap) put the lower bound much further out than 10 years (20 or so), so in comparison with a CogPrime-like AGI approach, not relevant14:59
maakubut if i'm wrong on that, sure I'd be open to considering WBE14:59
kanzurehmm was that in a section of that roadmap. i should have that memorized huh..14:59
* kanzure looks15:00
maakukanzure: there's a table near the back with timelines assuming various computational targets15:00
fennagain you're basing your estimate of "CogPrime-like AGI" arrival on what?15:00
maakuwhich is essentially just kurzweilian curve fitting15:00
kanzureoh, the appendix of large-scale neural network simulations?15:01
maakufenn: my own analysis of what it'd take to implement a proto-AGI for my purposes (sudo build me a nanofactory)15:01
kanzureyeah that's shit and they should feel morally bad that they implied that table was relevant.... they could have just done an index of cpu cost or something.. bleh.15:01
fennmaaku don't you see that that's just pure planning fallacy though?15:01
kanzurei suppose it is relevant because of the topic, of course, but i mean it's not relevant for specifically estimating brain emulation itself.... e.g., blue brain project has a >$1 billion euro budget now, parallelism works and helps.15:02
kanzuremaybe i should pester anders/todd about that. hrm.15:02
maakufenn: I've been project manager for many projects, and received certificational training on PM techniques.. which doesn't mean I'm wrong, but I did account for that :P15:03
kanzurethat's a non-answer15:03
maakubesides it doesn't matter here, the same fallacy applies to estimates of WBE15:03
maakukanzure: ?15:03
kanzureestimation of the feasibility of brain emulation are very different from estimates of feasibility of something never achieved before15:03
fennto predict brain emulation you make a small number of assumptions and do quantitative prediction based on known quantitative trends15:04
kanzure"it may actually be true now, but how do we estimate the likelihood of something unprecedented?"15:04
fennto predict AGI you make a huge pile of assumptions and then wave hands around based on previous predictions15:05
fennthe large number of permutations of assumptions accounts for the spread in AGI timeline prediction, but it doesn't account for the huge optimistic bias15:06
fenn(by AGI researchers)15:07
kanzurei find it very strange that i am championing a conservative approach to an engineering project...15:07
fennmy bad, apparently non-experts have the same huge optimistic 15-25 year bias15:08
maakukanzure: I hardly think they're comparable. For one they achieve different goals.15:10
kanzurehmm, also: suppose that brain emulation happens. i think that a working emulation or simulation will accelerate non-brain-related agi progress significantly, especially through the "delete parts of the emulation until it stops working" method.15:10
maakukanzure: I want "sudo make me a nanofactory". How do I get that with the WBE approach?15:10
maakuI upload someone, having them recursively improve themselves, then ask them to design me one?15:11
maakuHow do you do that without being a gigantic risk?15:11
maakuWhat I'm saying is I don't think the WBE is as risk or uncertainty free as you make it out to be.15:12
kanzureby risk you mean superweapon risk, or feasibility failure type risk15:12
kanzure+do15:13
maakuThe standard AI safety risks15:13
maakuE.g. how do you trust the machine?15:13
maakuit's solveable, but there's a whole host of issues here15:15
kanzurei am not particularly worried about grey goo15:15
maakui'm not talking grey goo15:15
kanzureor uh, walls of interstellar computronium killing everything15:15
kanzurealways seemed like a non-statement to me15:16
kanzureyes, it is true that bad things are bad when they happen15:16
maakukanzure: when you say WBE do you mean uploading? starting with an actual human?15:16
kanzureno, but i'm open to that route15:16
maakuok if you're not talking uploading then I don't know how what you're saying is any different than AGI15:17
kanzureit involves lots of study and copying of existing implementations15:17
maakuif you are talking uploading, then you're actually enabling many of the nightmarish MIRI concerns15:17
kanzureyes15:18
kanzureit often surprises people to learn that singinst is anti-transhumanist15:18
kanzurebut it should be obvious? they practically yell it constantly, just not in those words...15:18
maaku(aside: for some time I've wanted to make them a "death clock" that counts the number of people that have died through their inaction)15:18
maakuok so IF we are talking uploading, there are two concerns:15:19
kanzureso you are suggesting that non-uploading brain emulation/simulation is agi?15:19
maaku(1) upload who? I don't think I can find anyone I'd trust, including myself15:19
kanzurei think i have already established that i don't believe in the ai trust shit15:20
fennneuromorphic AI15:20
maaku(2) even if we did upload the perfect somebody, the recursive self-improvement process is not guaranteed to be stable with respect to that person's values15:20
kanzurelet's assume that FAI is impossible15:20
kanzureand i'm still gonna do things anyway15:21
maaku(I dislike MIRI because they apply this argument generally, but in this specific case it really is a concern.)15:21
kanzureyou could argue that it is immoral of me to take action in a universe where FAI is impossible, but meh15:21
fenndefinitely not stable if you start deleting shit randomly15:21
fenn"wow suddenly i don't have empathy, yay!"15:21
fennkanzure: your existence offends my ethics, sir15:22
kanzurei know15:22
kanzurehappens often around these parts15:22
maakukanzure: let's not get distracted by an anti-MIRI rant15:23
maakuwe're on the same page there15:23
kanzure15:21 <@kanzure> user_: it is immoral for you to take action in a universe where FAI is impossible. yes/no?15:23
kanzure15:22 < user_> "no": in that setting "immoral" doesn't quite mean anything15:23
maakubut in this particular case i think the argument has merit15:23
fennlol that was so fast15:23
kanzure15:23 < user_> see case #5 of http://lesswrong.com/lw/khf/six_plausible_metaethical_alternatives/15:24
kanzureheh wei dai15:24
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Ex-Chat]15:24
maakuRight now at this point in time the only research MIRI is doing is on the value drift problem, and all they've accomplished is to show that it reduces to some very, very hard comp sci problems15:24
-!- CharlieNobody [~CharlieNo@97-85-244-245.static.stls.mo.charter.com] has joined ##hplusroadmap15:24
-!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap15:25
kanzure15:25 < user_> you might like posting that or something else from http://lesswrong.com/user/Wei_Dai/submitted/ into that discussion15:25
fennin other circles we call this "analysis paralysis"15:25
maakuSo I wouldn't dismiss it. You start rewiring a human brain, or simplifying the simulation (for performance reasons), who knows what kind of monster you'd end up with15:25
kanzurei did not dismiss it though15:25
kanzurei fully agree that powerful machines can do nasty stuff15:26
kanzure(and probably will do nasty stuff)15:26
fennyou're all a bunch of plant murderers15:26
maakuRight but you say partial emulation may be enough. I'm saying you're going to have to solve this problem before you attempt cutting corners15:26
kanzure"this problem" = sorry i just lost context.. :(15:27
fennthe ethics drift problem15:27
kanzurei think i did get distracted by the MIRI rant15:27
kanzureoh, i consider ethics drift to be MIRI stuff15:27
kanzure15:27 < user_> the "died through inaction" thing is generally covered by astronomical waste arguments15:28
kanzure15:27 < user_> it's not *nice*, it breaks a lot of generally-depended-upon *precedents*, to prioritize the interests of people in the future over people in the present15:28
kanzure15:29 < user_> but if you're already arguing about large abstract utilitarian things anyway, talking about the fact that there are like 10^absurd times as many people in the potential future as in the present doesn't break the precedents *as much*15:29
maakuhow come kanzure's comments are prefixed with "15:29 < user_>" ?15:30
fennhe's talking to a MIRI person on another network15:30
kanzurelesswrong/miri informant15:30
kanzurei'm pleased that you think i would say those things, maybe15:30
kanzuremaybe not so pleased15:31
kanzureer, i'm still curious about how you think that emulation/simulation is the same as your agi plans15:31
kanzureer, specifically, in the absence of uploading15:31
kanzurei think that is your current claim15:31
kanzureit might be a correct claim, i'm just curious to hear your thoughts behind that one..15:31
kanzurewhen you say uploading i think of things like "taking a known human brain and doing non-destructive or destructive scanning and then seeding an emulation/simulation with data extracted from the scanning process"15:34
kanzureand non-uploading would be things like, "working from general knowledge and first principles extracted from previous neuroscience research, here's some stuff that is pretty okay to start with" and, failing that, resorting to high-resolution scans...15:35
kanzureon a related note, it's pretty funny to see stephen reed showing up in -wizards15:37
kanzure.title https://www.youtube.com/watch?v=WIlcLPJqvvA15:39
yoleauxAdvanced Minecraft Presents: 2b2t - the world's worst server - YouTube15:39
kanzure(minecraft entertainment stuff)15:39
maakuhttp://www.phoronix.com/scan.php?page=news_item&px=MTgzNjY15:48
maaku1tflop for <$20015:48
-!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has joined ##hplusroadmap16:14
fenn"it can do something like 2 trillion 32bit integer multiplies per second" and has 8GB onboard ram, TDP 270W though :\16:16
fennthe pci-e form factor is a pain in the ass for my uses16:16
kanzure"My Chinese input software gives three possibilities when I type in "wei dai": not bring, microstrip, and grave danger."16:29
-!- genehacker [~chatzilla@8-12.ptpg.oregonstate.edu] has quit [Ping timeout: 240 seconds]16:31
caternthat minecraft video16:42
caterndefinitely postapocalyptic16:42
-!- genehacker [~chatzilla@8-12.ptpg.oregonstate.edu] has joined ##hplusroadmap16:49
-!- genehacker [~chatzilla@8-12.ptpg.oregonstate.edu] has quit [Ping timeout: 245 seconds]17:14
-!- genehacker [~chatzilla@8-12.ptpg.oregonstate.edu] has joined ##hplusroadmap17:17
-!- Qfwfq [~WashIrvin@unaffiliated/washirving] has quit [Ping timeout: 255 seconds]17:26
-!- genehacker [~chatzilla@8-12.ptpg.oregonstate.edu] has quit [Ping timeout: 250 seconds]17:27
-!- Qfwfq [~WashIrvin@unaffiliated/washirving] has joined ##hplusroadmap17:32
fennWILL HE BE MURDERED? Stay tuned!17:34
-!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 240 seconds]17:34
kanzure"grave danger" seems like such a steve name17:40
kanzure"the only source of actual competence is insiders" well that's problematic18:06
-!- sheena [~home@d162-156-158-13.bchsia.telus.net] has quit [Ping timeout: 255 seconds]18:09
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-fvkmdegpkvgjhclk] has quit [Quit: Connection closed for inactivity]18:17
kanzurefenn: re your thing abut minecraft/bitcoin, https://news.ycombinator.com/item?id=863996218:25
-!- sheena [~home@50.66.72.211] has joined ##hplusroadmap18:30
fennwow this is way more creative than i was thinking https://en.bitcoin.it/wiki/BitVegas18:38
kanzureand what were you thinking?18:41
fennliterally mining bitcoins18:41
-!- sheena2 [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap18:44
-!- sheena [~home@50.66.72.211] has quit [Ping timeout: 272 seconds]18:45
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Remote host closed the connection]18:47
fennneal stephenson strikes again http://en.wikipedia.org/wiki/Reamde19:00
kanzurei'm just going to be disappointed aren't i19:01
fennprobably19:01
kanzure"that's not how you make demands! ugh"19:02
fennapparently after writing this book he decided to go make a video game19:02
fennsince mojang specifically prohibits buying in-game items/powers with "real world money" i expect this to either go underground or switch to a different mmorpg (or both)19:12
fennwasn't the whole point of farmville to do this sort of thing?19:13
fennexcept you have 0 chance of getting any money back19:13
kanzurei'm afraid farmville fandom doesn't intersect this community much19:13
fenni know, but for a very long time there was no money in video games, and now it's like 90% of people who play games play pay-to-play19:14
fennlike, zynga came out of nowhere and ruined everything19:14
fennbut i don't really get why it was a new thing19:14
fenn"10 million daily active users within six weeks of launch"19:15
fennthat's rhodonculous19:15
kanzurewhy is that ridiculous?19:16
fennit's a really fast growth rate19:17
-!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has quit [Remote host closed the connection]19:17
kanzurethat's for just the game19:17
kanzurezynga had prior games19:17
kanzureso they had a bunch of users they were leveraging19:18
kanzureand uh spending a bunch of money on ads19:25
kanzureheh they were spammers (had users install malware, etc)19:26
kanzuremystery solved19:27
-!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap19:44
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 250 seconds]19:46
-!- Jaakko9114 [~Jaakko@host81-156-218-223.range81-156.btcentralplus.com] has quit [Ping timeout: 264 seconds]20:11
kanzure"My main problem with it is that ESR is unable to differentiate between being a good programmer and being ESR."20:37
kanzurehttp://lesswrong.com/lw/gd3/outline_of_possible_sources_of_values/20:49
nmz787paperbot: http://www.researchgate.net/publication/224212753_Projection_and_Geodesic-Based_Pipe_Routing_Algorithm/file/5046352386fa9aef4e.pdf21:02
-!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has joined ##hplusroadmap21:16
-!- Jaakko9114 [~Jaakko@host81-156-218-223.range81-156.btcentralplus.com] has joined ##hplusroadmap21:20
-!- genehacker [~chatzilla@c-50-137-46-240.hsd1.or.comcast.net] has joined ##hplusroadmap21:28
-!- sheena2 [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 250 seconds]21:36
-!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has quit [Ping timeout: 255 seconds]21:40
-!- yashgaroth [~ffffff@2606:6000:cb85:6a00:5968:1bb1:bb69:2b18] has quit [Quit: Leaving]21:40
-!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has joined ##hplusroadmap21:40
-!- CharlieNobody [~CharlieNo@97-85-244-245.static.stls.mo.charter.com] has left ##hplusroadmap ["Leaving"]21:42
-!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 244 seconds]21:54
-!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has quit [Ping timeout: 255 seconds]21:57
-!- Vutral [ss@mirbsd/special/Vutral] has joined ##hplusroadmap22:05
-!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has joined ##hplusroadmap22:10
-!- Jaakko9114 [~Jaakko@host81-156-218-223.range81-156.btcentralplus.com] has quit [Quit: Nettalk6 - www.ntalk.de]22:28
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap22:43
-!- night [~Adifex@unaffiliated/adifex] has quit [Quit: Just going out for a swim...]22:48
-!- night [~Adifex@unaffiliated/adifex] has joined ##hplusroadmap22:51
-!- juri_ [~juri@vpn166.sdf.org] has quit [Ping timeout: 255 seconds]23:49
-!- juri_ [~juri@vpn166.sdf.org] has joined ##hplusroadmap23:51
ebowdenOh, hey juri_.23:57
--- Log closed Mon Nov 24 00:00:00 2014

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