--- Log opened Wed Aug 05 00:00:31 2015 | ||
-!- EnLilaSko [~Nattzor@unaffiliated/enlilasko] has joined ##hplusroadmap | 00:00 | |
kanzure | *grumble grumble* something something truthcoin, something something extropy-chat snailmail prediction market | 00:03 |
---|---|---|
kanzure | and why are my llvm-dev filters suddenly failing | 00:04 |
kanzure | "Language learning under working memory constraints correlates with microstructural differences in the ventral language pathway" http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.385.9707&rep=rep1&type=pdf | 00:09 |
kanzure | "Electrical stimulation of Broca's area enhances implicit learning of an artificial grammar" http://pubman.mpdl.mpg.de/pubman/item/escidoc:243879/component/escidoc:527008/DeVries_Electrical_Stimulation_JOCN_2010.pdf | 00:10 |
kanzure | i have always been disappointed that "the language amoeba hypothesis" is not at all what its name says it is >:( | 00:13 |
-!- Dumpster_D1ver [~Vishnu@pool-72-83-63-112.washdc.east.verizon.net] has quit [Ping timeout: 264 seconds] | 00:39 | |
-!- xtalmath [~xtalmath@ip-83-134-138-137.dsl.scarlet.be] has quit [Quit: Leaving.] | 00:49 | |
-!- xtalmath [~xtalmath@ip-83-134-138-137.dsl.scarlet.be] has joined ##hplusroadmap | 00:50 | |
-!- Dumpster_D1ver [~Vishnu@pool-72-83-63-112.washdc.east.verizon.net] has joined ##hplusroadmap | 00:55 | |
-!- xtalmath [~xtalmath@ip-83-134-138-137.dsl.scarlet.be] has quit [Ping timeout: 256 seconds] | 01:04 | |
-!- xtalmath [~xtalmath@ip-83-134-138-201.dsl.scarlet.be] has joined ##hplusroadmap | 01:18 | |
-!- Dumpster_D1ver [~Vishnu@pool-72-83-63-112.washdc.east.verizon.net] has quit [Ping timeout: 246 seconds] | 01:18 | |
-!- nmz787_i [~ntmccork@192.55.54.36] has quit [Ping timeout: 272 seconds] | 02:03 | |
-!- Dumpster_D1ver [~Vishnu@pool-72-83-63-112.washdc.east.verizon.net] has joined ##hplusroadmap | 02:24 | |
jrayhawk | https://thezvi.wordpress.com/2015/06/30/the-thing-and-the-symbolic-representation-of-the-thing/ | 03:19 |
-!- nsh [~lol@wikipedia/nsh] has quit [Excess Flood] | 03:23 | |
-!- nsh [~lol@wikipedia/nsh] has joined ##hplusroadmap | 03:23 | |
xtalmath | what is an electrograph? | 03:39 |
fenn | https://srconstantin.wordpress.com/2015/07/19/we-could-regrow-livers/ regrowing a thymus is especially interesting wrt longevity | 04:02 |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Quit: Leaving] | 04:21 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 04:25 | |
-!- Madplatypus [uid19957@gateway/web/irccloud.com/x-eyjnfxjtaaoqdtsl] has quit [Quit: Connection closed for inactivity] | 04:39 | |
-!- augur_ [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has quit [Remote host closed the connection] | 04:47 | |
-!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has joined ##hplusroadmap | 04:48 | |
-!- eudoxia [~eudoxia@r167-57-161-234.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 05:10 | |
-!- c0rw|zZz [~c0rw1n@91.176.95.206] has quit [Ping timeout: 252 seconds] | 05:15 | |
kanzure | "In 1979, researchers found that hepatocytes injected into the rat spleen (which is part of the lymphatic system and analogous to a large lymph node) functioned normally and grew to take up 40% of the spleen." | 05:25 |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 05:26 | |
kanzure | "I was developing a DNA nanostructure on cadnano when I realised the utilities in nanoengineer could boost this process. But there seems to be no way to import .json into nanoengineer in a coarse-graining manner. Also the PAM3-PAM5 model seems to differ from the coarse-graining model in oxDNA so that it's impossible to import .pdb from oxDNA, which could result from converting a .json file from cadnano. | 05:37 |
kanzure | Is there any utilities I can use to interconvert .json in cadnano and .mmp in nanoengineer?" | 05:37 |
kanzure | eudoxia: thoughts? | 05:43 |
eudoxia | kanzure: alright, first thing, .mmp is not properly documented: the version in the actual files differs from the html documentation in the repo | 05:46 |
eudoxia | but, if cadnano's .json is supported by ObMol, then it might be fairly easy, importing it | 05:46 |
kanzure | https://groups.google.com/d/msg/nanoengineer-dev/9PxfdXn6MRk/iVWnPR03CwAJ | 05:47 |
eudoxia | "We're working on releasing a new API and plugin engine for cadnano that will allow you to easily manipulate the data model so it won't be necessary to parse the json file directly." | 05:49 |
eudoxia | >dec 2011 | 05:49 |
kanzure | do we also have to maintain cadnano | 05:49 |
eudoxia | aaaaaa | 05:50 |
eudoxia | well this is the encoder https://github.com/sdouglas/cadnano2/blob/master/model/io/encoder.py | 05:50 |
xtalmath | has anyone ever read this slingatron paper? | 05:51 |
xtalmath | http://www.ramaccelerator.org/home/sites/default/files/sling-0571.pdf | 05:51 |
kanzure | xtalmath: have you seen https://github.com/kanzure/nanoengineer | 05:51 |
kanzure | heh "from legacyencoder import legacy_dict_from_doc" | 05:51 |
xtalmath | yeah, I think I did, but never installed it actually, thanks for reminding me | 05:52 |
xtalmath | kanzure: I had a hard time falling asleep yesterday, as I thought about the slingatron (which I had read a few years ago). I was thinking how this might be used as a vacuum pump: consider an accelerating and decelerating slingatron connected to each other in a closed loop. | 05:54 |
xtalmath | since the ball bearings would pass each point at the same frequency, but at a different speed, the distance between ball bearings would vary, in a way that the same amount of gas is compressed at the slow pipe (from declatron to slingatron), and expanded at the fast pipe (from slingatron to decelatron) | 05:56 |
xtalmath | I came on this, because one of the major problems with vacuum pumps is how to lubricate the bearings in contact with vacuum. either very $ precise, or $ magnetic bearings. so I thought let the ball bearing be the compressing element. | 05:58 |
xtalmath | so there is for sections of pipe, in the order the balls move: slingatron, straight section of fast pipe, inverse slingatron, straight section of slow pipe | 05:59 |
xtalmath | what do you think? | 05:59 |
xtalmath | the smaller the bearings, the higher the compression ratio (since the bearings can come closer) | 06:00 |
xtalmath | I need some software to simulate molecular flow, to laminar flow, to stokes flow | 06:01 |
-!- mgin [~mgin@unaffiliated/mgin] has quit [Ping timeout: 244 seconds] | 06:02 | |
eudoxia | isn't that an unsolved problem | 06:02 |
xtalmath | I think it is analytically unsolved. but numerically? I'm probably wrong though | 06:03 |
xtalmath | I am pretty sure this works in stokes/laminar flow (think of the tennis ball "marble run" that was featured on HAD), so perhaps just a molecular flow simulator | 06:04 |
xtalmath | aha Molflow! | 06:05 |
kanzure | so the other good reason to make a flat engine is because the people of flatland would be quite grateful | 06:07 |
kanzure | xtalmath: closed loop slingatron might be a scroll compressor | 06:07 |
kanzure | https://en.wikipedia.org/wiki/Scroll_compressor | 06:07 |
kanzure | https://upload.wikimedia.org/wikipedia/commons/3/31/Two_moving_spirals_scroll_pump.gif | 06:07 |
-!- math3 [uid54897@gateway/web/irccloud.com/x-hgwwhzclkwpqxsuw] has joined ##hplusroadmap | 06:07 | |
xtalmath | wow | 06:07 |
xtalmath | that is amazingly similar, but I see friction instead of rolling there... | 06:08 |
kanzure | yes/no? | 06:08 |
xtalmath | thanks for mentioning | 06:08 |
-!- c0rw|zZz [~c0rw1n@91.176.95.206] has joined ##hplusroadmap | 06:09 | |
kanzure | i was actually just thinking of a scroll compressor a few minutes ago for other reasons | 06:09 |
kanzure | a while back i wanted to make a flat engine that could be manufactured with a single manufacturing process | 06:09 |
kanzure | and thus why i made the snarky comment about flatland above | 06:09 |
xtalmath | yeah, I have had more of these accidental seeing things I was thinking about moments/days ago too, more and more recently | 06:11 |
xtalmath | thought amplification by singularity stimulated thought synchronization | 06:13 |
xtalmath | I wonder how much heat the rolling bearings would cause. also, manufacturing the bearing groove in the plane is probably hard, unless tubing is used. | 06:14 |
kanzure | nah just start with straight metal then bend it into a spiral | 06:15 |
xtalmath | so then I would use cylinders instead of ball bearings? | 06:15 |
xtalmath | im afraid cylinder will try to run out of plane. while a ball will choose its path on the tube surface | 06:17 |
xtalmath | also I think ball bearings are much cheaper, but the manifold for ball bearing is probably more expensive than for the cylinders. hmm. unless I use tubing... | 06:18 |
xtalmath | also not sure how to prevent the balls from conglomerating | 06:18 |
xtalmath | unless instead of slingatron, I use electromagnet coil wound around a section, to directly accelerate a ball, then the motor that translates the whole stage in a "rotating" fashion can be left out. | 06:21 |
xtalmath | then the relative times and speeds of the bearings can be measured optically | 06:22 |
xtalmath | and late balls can get larger pulses, and early balls can get smaller pulses. | 06:23 |
xtalmath | yeah I should try that | 06:24 |
-!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has quit [Ping timeout: 255 seconds] | 06:25 | |
xtalmath | perhaps the cylinder could self repel if the cover walls were conductors... | 06:28 |
xtalmath | magnetic cylinders | 06:28 |
xtalmath | lots of heat loss though | 06:28 |
wrldpc1 | couple friends of mine are AI researchers, mind if I invite them here? | 06:31 |
kanzure | alright | 06:31 |
-!- dustinm [~dustinm@2607:5300:100:200::160d] has quit [Ping timeout: 240 seconds] | 06:49 | |
-!- dustinm [~dustinm@2607:5300:100:200::160d] has joined ##hplusroadmap | 06:51 | |
-!- nmz787_i [~ntmccork@192.55.55.41] has joined ##hplusroadmap | 06:57 | |
wrldpc1 | .site http://www.thefirearmblog.com/blog/2013/10/08/soviet-laser-pistol/ | 07:10 |
wrldpc1 | .title http://www.thefirearmblog.com/blog/2013/10/08/soviet-laser-pistol/ | 07:11 |
yoleaux | The Soviet Laser Pistol - The Firearm Blog | 07:11 |
wrldpc1 | .. | 07:11 |
chris_99 | cool | 07:15 |
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-xxatxuslbajnrvjw] has joined ##hplusroadmap | 07:17 | |
-!- Houshalter [~Houshalte@oh-71-50-58-200.dhcp.embarqhsd.net] has joined ##hplusroadmap | 07:19 | |
-!- Houshalter [~Houshalte@oh-71-50-58-200.dhcp.embarqhsd.net] has quit [Quit: Quit] | 07:47 | |
-!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has joined ##hplusroadmap | 07:48 | |
-!- FourFire [~FourFire@185.7.192.138] has joined ##hplusroadmap | 08:00 | |
-!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:55c4:7075:caab:806e] has joined ##hplusroadmap | 08:04 | |
xtalmath | nonsense | 08:19 |
-!- superkuh [~superkuh@unaffiliated/superkuh] has quit [Ping timeout: 246 seconds] | 08:20 | |
-!- Madplatypus [uid19957@gateway/web/irccloud.com/x-bacefctqigsxqsgg] has joined ##hplusroadmap | 08:20 | |
chris_99 | why's it nonsense | 08:24 |
-!- gradstudentbot [~gradstude@131.252.130.250] has joined ##hplusroadmap | 08:29 | |
kanzure | gradstudentbot: done writing your thesis? | 08:30 |
gradstudentbot | IOError: [Errno 2] No such file or directory: 'modules/quotes.txt' (file "/home/bryan/code/gradstudentbot/phenny/modules/quotes.py", line 8, in pick_quote) | 08:30 |
-!- gradstudentbot [~gradstude@131.252.130.250] has quit [Remote host closed the connection] | 08:30 | |
kanzure | .... guess not. | 08:30 |
chris_99 | awh poor gradstudent bot | 08:30 |
-!- gradstudentbot [~gradstude@131.252.130.250] has joined ##hplusroadmap | 08:33 | |
kanzure | gradstudentbot: done writing your thesis? | 08:33 |
gradstudentbot | You know, I can just do consulting. | 08:33 |
-!- HEx1 [~HEx@hexwab.plus.com] has quit [Ping timeout: 265 seconds] | 08:36 | |
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap | 08:46 | |
-!- HEx1 [~HEx@hexwab.plus.com] has joined ##hplusroadmap | 08:47 | |
-!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-enycwrpqrsguiwhv] has quit [Quit: Connection closed for inactivity] | 08:47 | |
xtalmath | chris_99: satellites have huge relative velocities, if it came close enough it would have passed in a blink. they are hard to see, and would be very small targets. the device has no scope to aim at the satellite lenses. what about cooling? | 08:55 |
xtalmath | if shot from withing the space vehicle, there will be reflection from the window, temporarily blinding the crew. if shot from outside, the cosmonaut would have to prepare for a space walk and get outside, then aim with poor attitude control | 08:58 |
xtalmath | if laser weapons were ever brought to space (which sounds probable), it would be attached to the vehicle, so it would be under computer control, and any known target trajectory entered for coarse aiming, then locking with a low power modulated beam | 09:01 |
xtalmath | this article is wannabe cosmonaut circle jerk, or astronaut false flag "propaganda" | 09:02 |
xtalmath | or byzantine mindfuck. perhaps some poor russian machinist concocted it to sell to a collector. | 09:03 |
xtalmath | also the length for gain is miserably slow | 09:04 |
xtalmath | low | 09:05 |
xtalmath | short | 09:05 |
xtalmath | perhaps it was an gadget of secret agent sold to collectors, as a decoy to find collectors who still have moonrock | 09:07 |
xtalmath | whatever | 09:07 |
xtalmath | which was probably gypsum all along | 09:07 |
chris_99 | yeah that's a good point about a scope | 09:08 |
xtalmath | ok apparently the rod is long enough | 09:11 |
xtalmath | if its a battery it would be a molten salt battery I guess | 09:13 |
xtalmath | I still think its fake | 09:13 |
chris_99 | yeah i see your argument, it does sound a bit iffy now | 09:13 |
-!- nmz787_i [~ntmccork@192.55.55.41] has quit [Quit: Leaving.] | 09:14 | |
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-xxatxuslbajnrvjw] has quit [Quit: Connection closed for inactivity] | 09:16 | |
chris_99 | woudn't a hefty CO2 laser have been more suitable too | 09:17 |
xtalmath | also, the person who holds it (the 4 photos in one), seems to be the same person who machined it: in the first photo the cartridge has holes, and you can see a spring behind it. in the lower right of the 4 photos there are no holes yet. | 09:17 |
xtalmath | one could argue that the 4 photos are archive photo's from soviet era. but then what is the flatscreen doing in the background? | 09:19 |
xtalmath | so it was machined recently | 09:19 |
chris_99 | aren't they just recent photos? | 09:19 |
xtalmath | if you had genuine laser gun. would you start machining it? | 09:19 |
xtalmath | they must be recent, because of the flat screen. hence the holes in the first picture must have been machined recently | 09:20 |
xtalmath | since you don't unmachine holes | 09:20 |
chris_99 | oh you're saying something has changed | 09:21 |
chris_99 | in the gun | 09:21 |
xtalmath | yes, the cartridge holding the "bullets" had no holes in the pictures with flatscreen in background. in the first picture shown, the cartridge has holes. | 09:23 |
xtalmath | so it was machined after the existence of flatscreens. so it was probably created recently | 09:24 |
xtalmath | unless you'd start painting over the lost panel of "the lamb of god" if you had it. most people would sell it instead of machining it. | 09:25 |
xtalmath | wtf https://en.wikipedia.org/wiki/Soviet_laser_pistol | 09:29 |
xtalmath | the oldest reference seems to be from 2004, flatscreens were available | 09:35 |
xtalmath | but not widespread | 09:35 |
-!- FourFire [~FourFire@185.7.192.138] has quit [Quit: Leaving] | 09:40 | |
-!- nmz787_i [ntmccork@nat/intel/x-uzbkcbxzlhlutovd] has joined ##hplusroadmap | 09:59 | |
kanzure | "If I have seen further, it is by blinding everyone else." - duke-admiral isaac newton | 10:03 |
xtalmath | lol | 10:05 |
gradstudentbot | Who used the last of the growth medium? | 10:08 |
xtalmath | man, molecular flow monte carlo is complicated or simple. surface roughness or simple reflection? gravity or no gravity? thermal gradients or no thermal gradients? | 10:08 |
xtalmath | ok no gravity | 10:13 |
xtalmath | now which would have bigger effect, thermal gradients or surface roughness? or just start with a simple calculation first? | 10:13 |
xtalmath | rotating molecules or not? quantize rotation? | 10:16 |
kanzure | have you done laminar flow simulations before? | 10:17 |
xtalmath | not on a computer, but I did do exercises during continuum mechanics | 10:17 |
xtalmath | I think I want to start with the molecular flow pump. since I am convinced I can get pretty low with stages of peristaltic pump | 10:19 |
xtalmath | the reason peristaltic pump fails round 30 Torr is probably that atmosphere keeps the tube in flattened state so it refuses to open up | 10:21 |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 264 seconds] | 10:27 | |
xtalmath | I was considering the ball bearing pump as a replacement for the molecular flow $ pump | 10:31 |
xtalmath | since ball bearing mass is proportional to r^3, while cross section area is proportional to r^2, suggesting smaller is better | 10:33 |
xtalmath | im even considering the 2 micron or so saphire balls (also used in the foldscope) | 10:34 |
xtalmath | perhaps actuated by piezo... | 10:35 |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 10:36 | |
xtalmath | another advantage of smaller, is that its cheaper to have many, and then throughput vs pressure curve can be modified by heaving different amount of pumps at each stage | 10:38 |
xtalmath | not to mention that the extra volume of the pump itself that also needs to be evacuated (partially) becomes smaller. | 10:38 |
xtalmath | if saphire, the option to actuate magnetically dissapears though :( | 10:39 |
xtalmath | hmm, piezo in vacuum is probably a bad idea, if the vacuum chamber emit ions to the pump. | 10:55 |
xtalmath | back to slingatron | 10:55 |
xtalmath | hmm, im thinking more like the saphire bearings, steel manifold, and manifold vibrated by the cheap vibration motors from cell phones | 11:11 |
xtalmath | what is that toy called? its a flexible tube and you rotate it like a slingshot and it makes sound? | 11:17 |
xtalmath | ah whirly tube | 11:18 |
-!- BlueLobster [medea@unaffiliated/fellini] has quit [Ping timeout: 240 seconds] | 11:19 | |
xtalmath | perhaps a lot of whirly tubes in series :) | 11:19 |
-!- BlueLobster [medea@persephone.darkness-reigns.net] has joined ##hplusroadmap | 11:20 | |
xtalmath | thats a lot of waste volume though :( | 11:21 |
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 244 seconds] | 11:23 | |
-!- math3 [uid54897@gateway/web/irccloud.com/x-hgwwhzclkwpqxsuw] has quit [Quit: Connection closed for inactivity] | 11:29 | |
-!- BlueLobster [medea@persephone.darkness-reigns.net] has quit [Ping timeout: 255 seconds] | 11:31 | |
-!- BlueLobster [medea@persephone.darkness-reigns.net] has joined ##hplusroadmap | 11:31 | |
-!- delinquentme [~delinquen@74.61.157.78] has joined ##hplusroadmap | 11:40 | |
-!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has quit [Quit: Leaving...] | 11:51 | |
-!- BlueLobster [medea@persephone.darkness-reigns.net] has quit [Ping timeout: 240 seconds] | 11:52 | |
-!- BlueLobster [medea@persephone.darkness-reigns.net] has joined ##hplusroadmap | 11:53 | |
xtalmath | working with vacuum is much safer than working with high pressure right? | 12:06 |
-!- nmz787_i [ntmccork@nat/intel/x-uzbkcbxzlhlutovd] has quit [Ping timeout: 250 seconds] | 12:13 | |
-!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has joined ##hplusroadmap | 12:22 | |
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 12:23 | |
delinquentme | kanzure, why havnt you fixed the block chain to yield spare compute | 12:31 |
delinquentme | put blockchain to work on F@H | 12:31 |
xtalmath | wow, pure genius: https://en.wikipedia.org/wiki/Liquid_ring_pump | 12:39 |
xtalmath | there is a blockchain for science? | 12:40 |
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-oyqudysgjetzjrbj] has joined ##hplusroadmap | 12:42 | |
xtalmath | I once had the idea to make a blockchain, that rewards the proving of theorems, and the finding of inconsistencies through MetaMath (proofs would have to be entered formally) | 12:42 |
xtalmath | people could bet on which theorems would or would not be proven within X blocks, so that both correct predictor and the prover earn coin. | 12:44 |
xtalmath | every bet would be with respect to a database of axioms | 12:45 |
xtalmath | as algorithmic complexity and cryptography theorems would soon be entered, it would quickly gain reputation of the most secure blockchain (as it would be practically maintained by mathematicians) | 12:46 |
xtalmath | initiating the singularity :) | 12:46 |
kanzure | delinquentme: there's a few reasons why a blockchain is not a good "spare compute capacity" system | 12:48 |
kanzure | delinquentme: the algorithm for proof-of-work needs to be progress free and not useful in other situations, otherwise you are perverting the incentive-compatibility it provides | 12:48 |
xtalmath | the only difference between science and mathematics, are which axioms are accepted (mathematicians being very picky, so as not to introduce inconsistencies, while scientists being more loose and working with analogy). this would quickly result in rephrasing science in a manner that a computer can formally verify | 12:48 |
xtalmath | kanzure: I assume you are adressing me? | 12:48 |
xtalmath | the proof of work could be made progress free in the following manner: 50% of the planned rewards could be assinged towards classical mining, the other 50% could be retroactively divided to whomever found proofs ( of inconsistency, or unproven theorems, or shorter proofs, i.e. relying on less axioms) proportional to how much speculators bet (to approximate "importance") | 12:52 |
xtalmath | to advocate or propagandize a belief system, it would need its believers to be willing to bet things like "we believe there is no inconsistency in our belief system / school of thought" | 12:53 |
xtalmath | other bets would also become possible | 12:54 |
kanzure | no i was addressing delinquentme, he said: | 12:57 |
kanzure | 12:31 < delinquentme> kanzure, why havnt you fixed the block chain to yield spare compute | 12:57 |
xtalmath | since it would cause mechanically provable belief systems to be maintained and expanded, education will be automatable | 12:57 |
kanzure | and then i got distracted by terraform stuff for some reason | 12:58 |
kanzure | xtalmath: the problem with bets is that you would need a way to order the bets, which is exactly what proof-of-work mining is supposed to solve anyway | 12:59 |
kanzure | i recommend the following: | 12:59 |
kanzure | https://download.wpsoftware.net/bitcoin/alts.pdf | 12:59 |
kanzure | https://download.wpsoftware.net/bitcoin/asic-faq.pdf | 12:59 |
kanzure | https://download.wpsoftware.net/bitcoin/pos.pdf | 13:00 |
xtalmath | what do you mean with ordering bets? the proof-of-work would still be mining, but PoW would not be the only way to get reward | 13:00 |
gradstudentbot | Don't phage me, bro. | 13:00 |
kanzure | because the other way is broken | 13:00 |
xtalmath | so the timestamping is by regular PoW | 13:00 |
kanzure | including proof-of-work and something broken still makes for a broken system | 13:00 |
chris_99 | how are you gradstudentbot | 13:00 |
gradstudentbot | Sorry for wasting your time. | 13:00 |
chris_99 | heh | 13:00 |
xtalmath | also what do you mean with ordering bets? | 13:01 |
kanzure | in bitcoinland the purpose of mining is to order the incoming transactions | 13:01 |
xtalmath | I thought you were referring to the timestamping of blocks | 13:01 |
kanzure | otherwise you can get double spends and so on | 13:01 |
kanzure | i'm not sure how familiar you are with bitcoin's architecture | 13:01 |
xtalmath | rather familiar | 13:01 |
xtalmath | but not expert | 13:01 |
kanzure | so like... you've read the bitcoin source code? | 13:02 |
xtalmath | parts yes | 13:02 |
xtalmath | but not all | 13:02 |
xtalmath | in what sense does the order of bets matter? | 13:02 |
kanzure | you need to have consensus over the order or type of bets, i think | 13:03 |
kanzure | theorem proving also does not seem computationally intensive to me | 13:03 |
xtalmath | nah, anybody can challenge the rest with any kind of bet | 13:03 |
gradstudentbot | I'm writing that up and it will be submitted soon. | 13:04 |
kanzure | but i would rather have you present this argument in #bitcoin-wizards (e.g. how theorem proving would be a practical proof-of-work function) | 13:04 |
xtalmath | kanzure, again, the miners dont do theorem proving for security, they do it for rewarding. | 13:04 |
gradstudentbot | Hey, I got 100% yield! Oh wait, no. | 13:04 |
kanzure | looks like our intern is up to his usual antics | 13:04 |
xtalmath | there is little bitcoin related to it. its just bitcoin + printing a little extra coins for valid proofs | 13:04 |
xtalmath | the theorem proving would not be a PoW, use the same PoW as bitcoin. | 13:05 |
kanzure | then what was the theorem proving for? | 13:05 |
kanzure | and also, #bitcoin-wizards is definitely the right place :-) | 13:06 |
xtalmath | if the miner does not create transactions that rewards the mathematicians that submitted good proofs, his block is rejected by the protocol everyone runs | 13:06 |
kanzure | for technical/academic analysis of theoretical blockchain magic | 13:06 |
kanzure | whereas most of the -wizards do not idle in here and so they would miss anything | 13:06 |
kanzure | why would everyone know to reject the block? they would have to have consensus about the mathematicians' work. | 13:07 |
xtalmath | if there were no new proofs in the previous block, its identical to bitcoin | 13:07 |
xtalmath | kanzure, yes, every client contains MetaMath, so they can check proofs automatically | 13:07 |
xtalmath | if the previous block contained a correct proof, the current miner must mint extra coins for the mathematician, if not the block is rejected. | 13:08 |
xtalmath | the miner doesnt lose anything for creating new blocks | 13:11 |
xtalmath | err coins for scientific work | 13:11 |
xtalmath | the mathematician is not the miner btw | 13:11 |
xtalmath | blockchain against hypocrisy | 13:12 |
xtalmath | for education, science, consistency | 13:12 |
gradstudentbot | Haha, undergrads. | 13:13 |
xtalmath | encourages risk, because a minority will like to bet on "within 100 blocks RSA broken", so a majority keeps betting it won't get broken. so this bet has a higher weight on the 50% of fresh coins to be spent on science. | 13:15 |
xtalmath | we would become an AI, where it's thoughts are formally checked, yet manually created | 13:16 |
xtalmath | automated proof checking would get a boost | 13:17 |
xtalmath | sorry | 13:17 |
xtalmath | proof generation | 13:17 |
-!- Dumpster_D1ver [~Vishnu@pool-72-83-63-112.washdc.east.verizon.net] has quit [Ping timeout: 265 seconds] | 13:17 | |
xtalmath | the most popular belief systems (set of axioms), would have the most bets being placed | 13:19 |
xtalmath | the setup is neutral across belief systems in the sense that it is not winner takes all | 13:19 |
kanzure | what is the point of your mining scheme in particular? | 13:20 |
xtalmath | with mining, you refer to the extra rewarding of mathematicians with new provable insights? or the PoW for the transaction block miners? | 13:21 |
xtalmath | it would gain the public's trust, as mathematicians can choose what part of math they study to earn money, instead of having to work for the banks | 13:22 |
kanzure | oh, so it's not related to mining at all. nevermind. | 13:23 |
xtalmath | a paranoid cryptographer who wants to cash in on his break of cryptosystem X, can now pseudoanonymously earn coin, instead of having to make a risky sell to ABC orgs | 13:23 |
xtalmath | and without having to actually steal money from banks | 13:24 |
xtalmath | they could start working from home | 13:24 |
xtalmath | also, it could make a reputation system for scientists, by contribution, or by correct prediction of bets... | 13:25 |
xtalmath | no more elbow working | 13:26 |
kanzure | not sure how you would avoid transaction stealing in this scenario | 13:26 |
xtalmath | just focus on the work | 13:26 |
xtalmath | kanzure: it would not avoid transaction stealing, it would allow the cryptographer to take a third path which wont piss off people, or get government involved. | 13:28 |
-!- eudoxia [~eudoxia@r167-57-161-234.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving] | 13:28 | |
kanzure | okay, but others can just steal his transactions and relay them faster | 13:28 |
xtalmath | which transactions are we talking about now? | 13:28 |
xtalmath | oh you mean in the case the broken crypto was exactly the crypto used to secure the blockchain? | 13:28 |
kanzure | no, i mean the one that you were submitting to claim your money for your work | 13:29 |
kanzure | they can just lift your proof, it's easy | 13:30 |
xtalmath | it would better use post quantum cryptography. and best use at least 2 different kinds of crypto, so that the cryptographer would have to break 2 systems, decreasing the event frequency from 1/N to 1/N^2 | 13:30 |
xtalmath | kanzure: salt your proof and hash it, ask miner to include data as the hash of a claimed proof | 13:30 |
kanzure | miner can steal it when you broadcast the real version | 13:31 |
xtalmath | after the block is verified x times, you submit the proof | 13:31 |
kanzure | miners can also censor these transactions | 13:31 |
xtalmath | the first hash also contains your public key | 13:31 |
kanzure | which is one of the objections to counterparty and mastercoin | 13:31 |
xtalmath | so he can't run off with your coins | 13:32 |
kanzure | plus, if a miner broadcasts another salted first-transaction later, and then finally broadcasts your (previously censored) actual proof, and it confirms, there's nothing you can do to get the actual reward later | 13:32 |
kanzure | sure he can | 13:32 |
xtalmath | well, the funds could be frozen for say a month | 13:32 |
kanzure | there's no way to detect that there was a previous salted version even deeper back in the blockchain history | 13:32 |
xtalmath | in which you have the time to prove you submitted first (its decentralized timestamping after all) | 13:32 |
kanzure | also this is vulnerable to miner cartels | 13:33 |
xtalmath | no way to detect it, but there is a way to prove it | 13:33 |
xtalmath | yes miner cartels | 13:33 |
xtalmath | that is true | 13:33 |
xtalmath | but hey, then bitcoin is fucked as well no? | 13:33 |
kanzure | no, proof-of-work (hashcash) generates a proof that someone can't steal | 13:35 |
xtalmath | to prevent "this really is the hash of a salted proof"-SPAM, a fee could be required. (if the scientist knows roughly what amount he would earn, he would normally be willing to put a little up front) | 13:36 |
kanzure | the reason why you can't steal a hashcash proof is because it's a hash over the coinbase transaction as well | 13:36 |
xtalmath | kanzure: you refer to hashcash itself or bitcoin? | 13:37 |
kanzure | it's mostly the same | 13:38 |
xtalmath | you cant steal a mining block, but mining cartels CAN perform double spend attack | 13:38 |
kanzure | no they can't... what? | 13:38 |
kanzure | spent utxos make a block invalid | 13:38 |
xtalmath | that is not a double spend | 13:39 |
kanzure | a double spend is using the same utxos twice, in the same history | 13:39 |
kanzure | if your history never has the original spend then how would it be a double spend? | 13:39 |
xtalmath | a mining cartel with higher hashing power than the rest, show one block, containing the tx, then fork from the previous state and catch up | 13:40 |
kanzure | that's not double spending imho | 13:40 |
kanzure | that's just an average every-day reorg | 13:40 |
xtalmath | the cartel can spend the same coins twice, the first one is just ignored by the rest | 13:41 |
xtalmath | but the goods are sent. that is the original meaning of double spent when I learnt bitcoin. reorgs were rare and supposed to not happen as it indicates double spends | 13:42 |
kanzure | that's like saying i'm double spending when i create 1 million variants of a transaction to spend my utxos | 13:42 |
kanzure | reorgs happen every day dude | 13:42 |
xtalmath | now they do | 13:42 |
kanzure | no they're supposed to happen often | 13:42 |
kanzure | that's how your node gets on the better blockchain | 13:42 |
kanzure | otherwise you'd be stuck! | 13:42 |
kanzure | btw do you have any particular complaints about #bitcoin-wizards or what | 13:43 |
xtalmath | not really, I just don't see how it is related to bitcoin. what I propose doesn't contribute to bitcoin, it just inflates (by printing "excess" for scientific work) | 13:44 |
gradstudentbot | Yeah, but that was only a sample size of one. | 13:45 |
xtalmath | but it does provide a market place for speculators, one that is based on reason, instead of charts going up, or spectators going down | 13:45 |
xtalmath | gradstudentbot: a sample size of one? i dont understand | 13:45 |
gradstudentbot | Future work will focus on that. | 13:45 |
xtalmath | the batch of apples on a boat that is being traded on, is getting rotten, and then unrot if we should believe "economists" | 13:46 |
kanzure | #bitcoin-wizards is like hplusroadmap but for cryptocurrency research and development | 13:47 |
xtalmath | I will join, but I won't repeat what I said here. my rant feels over ;-) | 13:47 |
kanzure | also happens to be where most of http://diyhpl.us/~bryan/papers2/bitcoin/ comes from | 13:48 |
xtalmath | kanzure: in the language of http://diyhpl.us/~bryan/papers2/bitcoin/Bitcoin%20over%20Tor%20isn%27t%20a%20good%20idea.pdf for example "double spending" means retroactively forking history, what you call a reorg. | 13:51 |
-!- math3 [uid54897@gateway/web/irccloud.com/x-vlptubkvpaokyfzg] has joined ##hplusroadmap | 13:55 | |
xtalmath | i'm not really following the current state of bitcoin, as of since a couple of years, but I am sure reorg's were considered suspicious (they could occur spontaneously without bad intentions, but a double spend attack would result in a reorg) . so a reorg would be a necessary but not a sufficient indicator of a double spend (the unaffected rest can not discriminate between a reorg and a double spend, even not the seller who loses | 13:59 |
kanzure | you might have been cutoff at the end of your line near "even not the seller who loses" | 14:00 |
xtalmath | if the buyer is actually a member of the dominant mining cartel, and the cartel rewrote history so that the transaction apparently did not happen, then it was a double spend | 14:00 |
xtalmath | "even not the seller who loses his wares: it could be a rare reorg, combined with the buyer not noticing he still has his coins, or noticing and refusing to resend his coins." | 14:00 |
kanzure | why'd you stop keeping track? | 14:01 |
xtalmath | (keep in mind this does not only relate to physically shipped goods, for which the victim has ample time to check if he still has his funds, but also for digital goods, like another cryptocurrency) | 14:02 |
xtalmath | track of what? | 14:02 |
kanzure | well for other cryptocurrencies there are things like spv proofs per the sidechain work | 14:02 |
kanzure | track of bitcoin | 14:02 |
xtalmath | you mean of 6 confirmations? | 14:03 |
kanzure | the spv proofs are useful for doing cross-chain swaps of cryptocurrency | 14:03 |
kanzure | as described in https://blockstream.com/sidechains.pdf | 14:03 |
kanzure | as for keeping track it was a question re: "i'm not really following the current state of bitcoin, as of since a couple of years," | 14:04 |
xtalmath | kanzure: I have the impression bitcoin has not really underwent more innovation? | 14:05 |
xtalmath | also, busy with different things, like "modern times" to pay rent | 14:06 |
kanzure | sidechains are a method of transporting bitcoin into another blockchain with a separate history and separate protocol where experimentation can be more friendly | 14:06 |
kanzure | see https://github.com/ElementsProject/elementsproject.github.io | 14:06 |
kanzure | here is a good overview too http://diyhpl.us/wiki/transcripts/gmaxwell-sidechains-elements/ | 14:07 |
xtalmath | yeah that would be instantaneous. but you want a coin as an economic tool to be used for general transactions. not just pegged sidechains. | 14:07 |
kanzure | (it's the video from the github git repo, except typed instead of videoed) | 14:07 |
kanzure | transfers into sidechains are unfortunately not instantaneous | 14:07 |
xtalmath | what I mean is that, if a bitcoin reorg occurs, it affects the pegged sidechain AFAIK | 14:08 |
kanzure | also you may be interested in hearing about http://lightning.network/ | 14:08 |
kanzure | i'm not sure how close you have been paying attention to bitcoin, lot of fun stuff i can show off | 14:08 |
kanzure | yes, it does effect spv proof-based transfers into sidechains, and as a result you have to wait like 100 confirmations to be safe | 14:09 |
xtalmath | kanzure: btw is that pegged side chain thing currently operational on the actual bitcoin blockchain, or a theoretical paper with no client yet? | 14:09 |
kanzure | it's operational on testnet at the moment | 14:09 |
kanzure | i wrote rpcblockchainexplorer for its release, heh | 14:10 |
kanzure | https://github.com/kanzure/rpcblockchainexplorer | 14:10 |
xtalmath | I was very interested in the SAFE network recently | 14:11 |
kanzure | linkz | 14:11 |
xtalmath | sorry maidsafe | 14:11 |
kanzure | i reported a huge vulnerability in their design and they just told me to piss off | 14:12 |
kanzure | so i pretty much hate them | 14:12 |
xtalmath | http://maidsafe.net/ and https://en.wikipedia.org/wiki/MaidSafe | 14:12 |
xtalmath | yeah I am not claiming it is mature at all, but I think it is the right direction | 14:12 |
kanzure | vulnerability was reported on their mailing list somewhere | 14:12 |
kanzure | it's totally broken and evil | 14:13 |
-!- Dumpster_D1ver [~Vishnu@pool-72-83-63-112.washdc.east.verizon.net] has joined ##hplusroadmap | 14:13 | |
kanzure | even ipfs would be better (and i hate ipfs) | 14:13 |
xtalmath | well the person who told you to piss off, was a visitor or the people behind maidsafe itself? | 14:13 |
kanzure | maidsafe | 14:13 |
xtalmath | what is ipfs? | 14:13 |
kanzure | more bullshit | 14:13 |
kanzure | http://static.benet.ai/t/ipfs.pdf | 14:14 |
chris_99 | have you published what you found anywere? | 14:14 |
kanzure | the vulnerability was reported to their mailing list, so it's out there | 14:15 |
chris_99 | aha cool | 14:15 |
kanzure | plus, it turns out that gmaxwell reported the same vulnerability to them even earlier, so.... | 14:15 |
xtalmath | could you find it back for me? | 14:15 |
kanzure | not at the moment, no | 14:15 |
kanzure | but it's under my username somewhere | 14:15 |
xtalmath | also as kanzure? | 14:15 |
kanzure | yeah, i guess there's https://www.reddit.com/r/Bitcoin/comments/2723i9/these_guys_are_creating_a_new_internet_using/chww5us | 14:16 |
kanzure | https://groups.google.com/forum/#!topic/maidsafe-development/W-n-IQ_TUis | 14:16 |
kanzure | (also i met them in person when they were in texas) | 14:18 |
-!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-wiaeasgjsxuayojk] has joined ##hplusroadmap | 14:36 | |
xtalmath | how are they in person? not friendly? | 14:37 |
xtalmath | also I see only one person respond in that google group, (and I think your attack would work, to some extent), but to me it doesn't really look like a "fuck off", in fact he finishes his post with "So, it is definitely possible to achieve what you're suggesting, but I'm just not sure that its effect on the network would be negative." | 14:38 |
xtalmath | your attack is quite interesting and obvious, but calculating the probability of being assigned the same files is not easy | 14:45 |
xtalmath | can we try to calculate it? | 14:48 |
xtalmath | even if we collude, the probability of getting the same file assigned is low. | 14:49 |
xtalmath | also, I am not sure if we are double blinded | 14:51 |
-!- superkuh [~superkuh@unaffiliated/superkuh] has joined ##hplusroadmap | 14:54 | |
xtalmath | lets assume not. we try to fill our common storage capacity C (in units of files). the total network stores T files. together, under 2 users we request as many files as possible. first I ask and get C random files, then how do you try to get the same C random files assigned? | 14:54 |
xtalmath | operating on pure chance alone gives a probability of C/T for each file. multiply by C since you get C chances. | 14:55 |
xtalmath | so we have C*C/T files in common | 14:55 |
xtalmath | where C is our combined storage, and T is the network storage both in files | 14:56 |
xtalmath | ah way no | 14:56 |
xtalmath | it is C/2*C/2*T | 14:57 |
xtalmath | C = 2*I, where I is our individual storage size. | 14:59 |
xtalmath | so 2 honest people with the same hard drive would each get I/(T*D) of the total network get rate (if fixed rate per get), for a total of 2*I/(T*D). | 15:02 |
xtalmath | if we have the same hard drive but collude we would get I/(T*D) each just like the honest people, but we would have some hard drive space left, because of the collisions | 15:04 |
gradstudentbot | I don't have enough data to form a hypothesis. | 15:06 |
xtalmath | how much do we have left? 2*I - I*I/T, (our real storage, minus feigned storage) | 15:06 |
xtalmath | so we get the same money but our HDD cost is less: so colludewage/honestwage = ($$$ / (2*I-I*T)) / ($$$ / 2*I) = 2*I/ (2*I-T*I)= 2/(2-T) ugh, something is wrong, it goes negative, while everyone is earning a positive | 15:10 |
xtalmath | ah I see | 15:11 |
xtalmath | colludewage/honestwage = ($$$ / (2*I-I*I/T)) / ($$$ / 2*I) = 2*I/ (2*I-I*I/T)= 2/(2-I/T) so if we have 2% of network capacity, or each 1%, then we can earn 2/(2-0.01) faster as the rest per GB. so we earn 2/1.99 times faster | 15:14 |
xtalmath | 1.005025126 | 15:14 |
kanzure | hm. | 15:15 |
xtalmath | so it does work, but we hardly feel it, instead of getting 1$ per whatever it buys in hosting, we get $1.005025126 | 15:15 |
xtalmath | I only calculated it for 2 colluders | 15:16 |
xtalmath | I will think of 3 now | 15:16 |
xtalmath | so it is definitely an attack, but it is an economic one. | 15:18 |
xtalmath | oh I forgot something, its even worse | 15:20 |
xtalmath | man this is not easy to calculate | 15:20 |
xtalmath | ah no it seems right | 15:22 |
xtalmath | so as long as the network is small (not many people donating HDD storage), an amazon kind of attack is cheap | 15:23 |
xtalmath | wait, you are right kanzure: I can use the $1.005025126 to buy storage from the rest at the rate of $1 | 15:25 |
xtalmath | so without buying real storage our fraction of the network increases | 15:25 |
xtalmath | we can simply store the data back in the network | 15:26 |
xtalmath | so that grows exponentially | 15:26 |
xtalmath | lol | 15:26 |
kanzure | nah it seems fair to assume that you are not buying storage from the network, because presumably your request/response rate has to be faster than the network can handle | 15:27 |
xtalmath | at first slow though, but put some up front for amazon attack, grow exponentially, then recover amazon cost | 15:27 |
kanzure | actually nevermind, i'm not sure anymore. | 15:27 |
kanzure | also you have no guarantee that if you did request data from the network, that you wouldn't hit yourself again | 15:28 |
xtalmath | kanzure: you should store it under a new file, there will be some overhead though, zip it first? | 15:29 |
xtalmath | its going to be compression wars | 15:29 |
xtalmath | I wonder if a user can drop a file, and get a random one assigned instead? | 15:29 |
xtalmath | so that any files we dont have in common are replaced till we have them in common? | 15:30 |
xtalmath | if that is possible we earn $2 for every $1 the rest would have earned per GB | 15:30 |
gradstudentbot | Can I get some more media? | 15:31 |
xtalmath | if the duplication factor is D, it only makes sense to collude up to D persons (if we can drop files and get new ones reassigned until they collide) | 15:31 |
xtalmath | so with D people we earn $D instead of $1 | 15:32 |
xtalmath | but I assume there is some mechanism that prevents you from saying, "nah, next file" | 15:33 |
kanzure | you can always just burn your identity and start again i think | 15:33 |
xtalmath | as far as I understood, there reward functions are far from set in stone though | 15:33 |
xtalmath | kanzure: yeah, I think so | 15:34 |
xtalmath | but perhaps the mechanism is not reputation based, but simply waiting in line or smth. | 15:34 |
xtalmath | anyway the files that you throw away... sooner or later it is noticed | 15:36 |
xtalmath | perhaps they'd block your IP | 15:37 |
-!- Houshalter [~Houshalte@oh-71-50-58-200.dhcp.embarqhsd.net] has joined ##hplusroadmap | 15:38 | |
kanzure | hopefully the network is using i2p/tor/something | 15:38 |
xtalmath | I had the impression they punched their own tunnels, but not sure if they allow for tor bridge or smth | 15:38 |
xtalmath | also, the conjuring would simply become an app on the MAIDSAFE network | 15:39 |
xtalmath | so if redundancy is D, all members of this cartel would earn faster than honest, and at a higher rate than scheming couples | 15:42 |
-!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has joined ##hplusroadmap | 15:43 | |
xtalmath | did you confront them with this in Texas as well? or only online? | 15:44 |
xtalmath | I really wish they can make it work though. | 15:45 |
kanzure | i mentioned it to them in person but i didn't dwell on it | 15:45 |
xtalmath | they gave no reaction back in person? | 15:45 |
kanzure | nah just stuff like "it's up in the air" and "we're working on things, keep tuned" bleh | 15:46 |
xtalmath | yeah, the way I see it, they only described the parts they know how to do in the papers, the exact reward mechanism doesn't seem fixed. | 15:47 |
xtalmath | also, its not a flat rate per storage unit. you are encouraged to link it all to one identity when you are under a limit, and encouraged to split up if you are above the limit | 15:48 |
xtalmath | I mentioned that to them but the reaction was something like: "we are definitely open to suggestions..." or the like | 15:48 |
xtalmath | still, I like some of the things/tricks with crypto they show | 15:49 |
xtalmath | but its not a complete solution yet | 15:49 |
xtalmath | show early and people complain its broken, hide until finished and people complain they had no word on its construction... | 15:50 |
xtalmath | I personally slightly prefer the hide until finished. people can always fork later | 15:51 |
xtalmath | but yeah, people invested in them, so they got to show their progress | 15:51 |
kanzure | they invested back in 2004-2006, it's totally stupid that they still haven't figured out an architecture or pivoted or folded | 15:51 |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Remote host closed the connection] | 15:51 | |
gradstudentbot | Huh? I haven't read any papers on that. | 15:51 |
xtalmath | gradstudentbot: its on their page | 15:52 |
gradstudentbot | Are the marmosets for anyone to use? | 15:52 |
kanzure | gradstudentbot: yes, but you should write a paper about this | 15:52 |
gradstudentbot | The paper was rejected. | 15:52 |
xtalmath | gradstudentbot: are you a bot or a person? | 15:53 |
gradstudentbot | Dropped my sample. | 15:53 |
xtalmath | lol | 15:53 |
kanzure | :-) | 15:53 |
kanzure | .g site:youtube.com shit graduate students say | 15:53 |
yoleaux | http://ipv6.google.com/sorry/IndexRedirect?continue=http://www.google.com/search%3F%26q%3Dsite:youtube.com%2520shit%2520graduate%2520students%2520say%26btnI%3D&q=CGMSECoBBPgCAXIwAAAAAAAIaucYmamKrgUiGQDxp4NLrVpqqRkNhn_7EXdSeU3KJpO6rK0 | 15:53 |
gradstudentbot | kanzure: http://www.youtube.com/watch?v=ovEghdXC4tE | 15:53 |
kanzure | .title http://www.youtube.com/watch?v=ovEghdXC4tE | 15:54 |
yoleaux | Shit Graduate Students Say - YouTube | 15:54 |
gradstudentbot | kanzure: Shit Graduate Students Say - YouTube | 15:54 |
-!- gradstudentbot [~gradstude@131.252.130.250] has quit [Remote host closed the connection] | 15:54 | |
xtalmath | its yours? | 15:55 |
-!- gradstudentbot [~gradstude@131.252.130.250] has joined ##hplusroadmap | 15:56 | |
kanzure | yes | 15:56 |
kanzure | he's just here to lighten up the atmosphere | 15:56 |
xtalmath | yeah it works | 15:56 |
kanzure | he's based on the "shit graduate students say" video | 15:56 |
xtalmath | took me a while to realize | 15:56 |
xtalmath | I should be putting the trash out, and then go to sleep | 15:57 |
xtalmath | I will try to recalculate the collisions, and wage advantage tomorrow more accurately, and take into account their sigmoid reward structure... | 15:58 |
xtalmath | I might write it out and post it on their forum or smth | 15:58 |
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-oyqudysgjetzjrbj] has quit [Quit: Connection closed for inactivity] | 16:26 | |
-!- drewbot [~cinch@ec2-54-81-183-98.compute-1.amazonaws.com] has quit [Remote host closed the connection] | 16:27 | |
-!- drewbot [~cinch@ec2-54-227-92-150.compute-1.amazonaws.com] has joined ##hplusroadmap | 16:27 | |
-!- math3 [uid54897@gateway/web/irccloud.com/x-vlptubkvpaokyfzg] has quit [Quit: Connection closed for inactivity] | 16:29 | |
-!- xtalmath [~xtalmath@ip-83-134-138-201.dsl.scarlet.be] has quit [Quit: Leaving.] | 16:31 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Quit: Leaving] | 16:33 | |
-!- Beatzebub [~beatzebub@d162-156-106-225.bchsia.telus.net] has joined ##hplusroadmap | 16:34 | |
-!- Houshalter [~Houshalte@oh-71-50-58-200.dhcp.embarqhsd.net] has quit [Quit: Quit] | 16:40 | |
justanotheruser | oh gradstudentbot is back? | 16:40 |
gradstudentbot | IOError: [Errno 2] No such file or directory: 'modules/quotes.txt' (file "/home/bryan/code/gradstudentbot/modules2/quotes.py", line 8, in pick_quote) | 16:40 |
justanotheruser | >:( | 16:41 |
-!- gradstudentbot [~gradstude@131.252.130.250] has quit [Remote host closed the connection] | 16:42 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 16:43 | |
-!- gradstudentbot [~gradstude@131.252.130.250] has joined ##hplusroadmap | 16:44 | |
kanzure | fixed | 16:44 |
-!- c0rw|zZz is now known as c0rw1n | 16:48 | |
justanotheruser | oh gradstudentbot is back? | 16:53 |
gradstudentbot | I was searching for a new particle that alters gene expression, didn't find it, but didn't refute it's existence either. But then some other asshole professor at another school found it. | 16:53 |
-!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 17:07 | |
-!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 250 seconds] | 17:10 | |
-!- cluckj [~cluckj@pool-108-16-231-242.phlapa.fios.verizon.net] has joined ##hplusroadmap | 17:30 | |
kanzure | beep boop | 17:36 |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Quit: Leaving] | 17:39 | |
-!- yashgaroth [~ffffff@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has joined ##hplusroadmap | 17:49 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 18:02 | |
kanzure | yashgaroth: also you could test against chimps which are close enough to humans | 18:06 |
-!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Read error: Connection reset by peer] | 18:07 | |
-!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 18:08 | |
yashgaroth | well mice are probably good enough for working memory, but for the language acquisition...do they do sign language or is that mostly gorillas? | 18:10 |
kanzure | well bonobos can point to pictographs | 18:12 |
yashgaroth | just use bonobos, cheaper to feed | 18:12 |
kanzure | for example, https://www.youtube.com/watch?v=r7ttRaXlnfs | 18:13 |
yashgaroth | chimps are also more likely to spontaneously tear off various parts of your body, and then shove them into various other parts of your body | 18:14 |
-!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-wiaeasgjsxuayojk] has quit [Quit: Connection closed for inactivity] | 19:07 | |
-!- sheena [~home@d108-180-224-190.bchsia.telus.net] has joined ##hplusroadmap | 19:10 | |
-!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has quit [Read error: Connection reset by peer] | 19:56 | |
kanzure | apparently most of the mouse tests of working memory are just tests of short-term memory instead http://www.researchgate.net/post/What_is_the_latest_behavior_test_for_working_memory_in_rats | 20:01 |
-!- math3 [uid54897@gateway/web/irccloud.com/x-aexgtyqlvjkfhclk] has joined ##hplusroadmap | 20:16 | |
-!- sheena [~home@d108-180-224-190.bchsia.telus.net] has quit [Ping timeout: 240 seconds] | 20:18 | |
-!- xtalmath [~xtalmath@ip-83-134-138-201.dsl.scarlet.be] has joined ##hplusroadmap | 20:29 | |
-!- dustinm [~dustinm@2607:5300:100:200::160d] has quit [K-Lined] | 21:14 | |
-!- dustinm [~dustinm@105.ip-167-114-152.net] has joined ##hplusroadmap | 21:15 | |
-!- Beatzebub [~beatzebub@d162-156-106-225.bchsia.telus.net] has quit [Quit: Beatzebub] | 21:51 | |
-!- xrr [~quassel@c21-76.uvn.zone.eu] has quit [Quit: No Ping reply in 180 seconds.] | 21:55 | |
-!- xrr [~quassel@c21-76.uvn.zone.eu] has joined ##hplusroadmap | 21:56 | |
xtalmath | eudoxia: apparently the DSMC (Direct Simulation Monte Carlo method) simulates arbitrary finite Knudsen number flows, it is statistical but should also give correct continuum mechanics flows! source : https://en.wikipedia.org/wiki/Direct_simulation_Monte_Carlo | 22:02 |
xtalmath | "The DSMC method has been extended to model continuum flows (Kn < 1) and the results can be compared with Navier stokes solutions." | 22:02 |
-!- math3 [uid54897@gateway/web/irccloud.com/x-aexgtyqlvjkfhclk] has quit [Quit: Connection closed for inactivity] | 22:39 | |
-!- yashgaroth [~ffffff@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has quit [Quit: Leaving] | 22:48 | |
-!- Dumpster_D1ver [~Vishnu@pool-72-83-63-112.washdc.east.verizon.net] has quit [Ping timeout: 244 seconds] | 23:08 | |
xtalmath | is deuterium / deuterium oxide a controlled substance? | 23:48 |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 255 seconds] | 23:59 | |
--- Log closed Thu Aug 06 00:00:32 2015 |
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!