--- Log opened Mon Aug 03 00:00:29 2015
00:20 -!- xtalmath [~xtalmath@ip-83-134-138-51.dsl.scarlet.be] has quit [Quit: Leaving.]
00:20 -!- xtalmath [~xtalmath@ip-83-134-138-51.dsl.scarlet.be] has joined ##hplusroadmap
00:52 -!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:4cfb:b1bc:a156:e60] has quit [Quit: Leaving]
01:15 -!- Dumpster_D1ver [~Vishnu@pool-72-83-63-112.washdc.east.verizon.net] has quit [Ping timeout: 264 seconds]
01:33 -!- Dumpster_D1ver [~Vishnu@pool-72-83-63-112.washdc.east.verizon.net] has joined ##hplusroadmap
01:46 -!- xtalmath [~xtalmath@ip-83-134-138-51.dsl.scarlet.be] has quit [Ping timeout: 246 seconds]
02:02 -!- xtalmath [~xtalmath@ip-83-134-138-51.dsl.scarlet.be] has joined ##hplusroadmap
02:11 -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-uusklojkzqnjrsoj] has joined ##hplusroadmap
02:39 -!- Madplatypus [uid19957@gateway/web/irccloud.com/x-hnaapfbeefyracrm] has quit [Quit: Connection closed for inactivity]
03:04 < xtalmath> is black "pure" chocolate a kind of positive photoresist?
03:04 -!- Stskeepz [~cvm@boat.tspre.org] has joined ##hplusroadmap
03:04 -!- thundara_ [~thundara@jean.markov.codes] has joined ##hplusroadmap
03:04 -!- thundara [~thundara@jean.markov.codes] has quit [Disconnected by services]
03:05 < xtalmath> are there any simple household photoresists?
03:05 -!- mf1008_ [~mf1008@unaffiliated/mf1008] has joined ##hplusroadmap
03:05 -!- Stskeeps [~cvm@unaffiliated/stskeeps] has quit [Ping timeout: 240 seconds]
03:05 -!- mf1008 [~mf1008@unaffiliated/mf1008] has quit [Ping timeout: 240 seconds]
03:05 -!- delinquentme [~delinquen@74.61.157.78] has quit [Ping timeout: 240 seconds]
03:05 -!- mf1008_ is now known as mf1008
03:06 -!- delinquentme [~delinquen@74.61.157.78] has joined ##hplusroadmap
03:44 -!- delinquentme [~delinquen@74.61.157.78] has quit [Ping timeout: 240 seconds]
03:49 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap
03:50 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 246 seconds]
04:16 -!- delinquentme [~delinquen@74.61.157.78] has joined ##hplusroadmap
04:19 -!- Viper168_ is now known as Viper168
04:25 < fluffypony> http://www.local10.com/news/south-florida-woman-first-to-receive-bionic-eye-in-florida/34470754
04:25 -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-cxkljykapwdtcomm] has joined ##hplusroadmap
04:36 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap
04:46 -!- Quashie [~boingredd@50.14.92.17] has quit [Ping timeout: 244 seconds]
04:54 < xtalmath> is there evidence of biological backpropagation at each synapse? is inhibitory or excitatory effect determined only by the presynaptic neuron? or can the postsynaptic neuron express different receptors to interpret a fixed neurotransmitter as either inhibitory or excitatory?
04:56 < xtalmath> i.e. is the synaptic weight determined solely by the presynaptic neuron or both ( weight=w(pre,post) ) ? and specifically is the sign of the weight a function of solely presynaptic neuron |weight|=w(pre) ?
04:58 < xtalmath> I guess the postsynaptic neuron has influence on the weight by expressing more or less receptors to change the magnitude of the weight. but do neurotransmitters exist for which there are multiple receptors with opposite inhibitory & excitatory postsynaptic interpretation of the same neurotransmitter exist?
04:59 -!- thundara_ is now known as thundara
05:00 -!- proofoflogic [uid65184@gateway/web/irccloud.com/x-smtootuicnjpvgny] has joined ##hplusroadmap
05:03 < xtalmath> i.e. how can a neuron know in what sense to change its output synapse weights -say to accomplish learning- if there is no feedback from the postsynaptic neuron?
05:04 -!- delinquentme [~delinquen@74.61.157.78] has quit [Ping timeout: 265 seconds]
05:05 < archels> the classical view is that the postsynaptic neuron determines the strength of the synapse, and the presynaptic side just adjusts to match the size
05:05 < archels> of course this is a huge simplification, not to say just plain false
05:06 < xtalmath> with strength you mean magnitude?
05:06 < archels> yeah, amplitude of the EPSC/EPSP
05:06 < xtalmath> I was more curious about the sign in fact
05:07 < xtalmath> how does the presynaptic side adjust to the postsynaptic neuron? this entails flow of information in the opposite direction of classical digital neural networks?
05:07 < archels> it may help to look at it the other way around. How can a neuron know how to change its input synapse weights? Well, it can just "sample" nearby axons that might be interesting.
05:07 < archels> do you know about STDP?
05:08 < CaptHindsight> xtalmath: photoresist for what application? What chemicals does it need to hold up to be be removed by?
05:08 < CaptHindsight> or removed by
05:11 < xtalmath> archels: STDP is interesting
05:12 < xtalmath> CaptHindsight: well, I would be happy if there was some "household" hardware store chemical for which any photoresist behaviour is known
05:13 < xtalmath> (i.e. I won't be picky about poor performance, or limited processes)
05:13 < archels> you can get spray can photoresist to apply coatings
05:13 < xtalmath> yeah, you mean for PCB
05:13 < archels> if you spray it on a PCB, sure
05:14 < archels> not sure what exotic applications you had in mind =)
05:14 < xtalmath> are there like exhaustive lists of photoinitiators and heuristic chemical rules?
05:14 < CaptHindsight> emulsion type photoresists for silk screen are really cheap
05:15 < xtalmath> is that like literally artist silk screen?
05:15 < CaptHindsight> artist, poster, t-shirt etc
05:15 < xtalmath> how cheap is cheap? how "ecofriendly" are they?
05:16 -!- fluffypony [~fluffypon@unaffiliated/fluffypony] has left ##hplusroadmap ["Textual IRC Client: www.textualapp.com"]
05:17 < CaptHindsight> http://www.advancedscreenprintsupply.com/screen_printing_emulsions.htm?gclid=CNG8q93yjMcCFQataQod3UYIdQ
05:19 < xtalmath> they can be spin coated? used at micron scale resolutions?
05:20 < CaptHindsight> what do you want them to adhere to?
05:21 < CaptHindsight> xtalmath: thats why I asked the application and how they would be removed rather than play 20 questions
05:22 < xtalmath> I am sorry
05:28 -!- eudoxia [~eudoxia@r167-57-167-40.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap
05:41 -!- xtalmath [~xtalmath@ip-83-134-138-51.dsl.scarlet.be] has quit [Ping timeout: 260 seconds]
05:58 -!- xtalmath [~xtalmath@ip-83-134-138-51.dsl.scarlet.be] has joined ##hplusroadmap
05:59 < xtalmath> what was my last msg?
06:02 < archels> log url in topic
06:06 < xtalmath> archels: I don't understand how a neuron can change its input weight? and suppose it is according to some rule trying to change the sign of the weight (i.e. from "0.01" to "-0.02")? it should then somehow demand the previous neuron to use a different neurotransmitter? i.e. what exactly is inhibitory/excitatory? strictly the transmitter molecule? or also the receptor?
06:07 < xtalmath> if also the receptor, then it is very misleading to have inhibitory/excitatory neurotransmitter terminology. I believe that fundamentally there must be backpropagation at the neuronal level, I can't believe all learning to pass through some global "good/bad" brain variable.
06:07 < xtalmath> also, how does the single neuron adjust a specific input weight to react more or less inhibitory/excitatory, without equally adjusting the other input weights?
06:08 -!- eudoxia [~eudoxia@r167-57-167-40.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving]
06:09 < archels> xtalmath: look up Dale's law
06:09 < xtalmath> furthermore, only adjusting the input weights cannot cause backpropagation, since that would require postsynaptic neurons somehow influencing the current neuron's firing. Information must cross the synapse in both ways for backpropagation to work. Unless backpropagation for learning requires cyclic connections.
06:09 < archels> backprop is just a theoretical construct. It (very very likely) does not happen in biology.
06:10 < kanzure> weights can be changed by adding and removing ampa receptors
06:10 < kanzure> and the amount of expression is some obnoxious methylation thing or transcription cofactor thing
06:11 < kanzure> also regulation of expression of nmda receptors plays an important role in modulation of plasticity and synapse weighting
06:13 -!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has quit [Ping timeout: 265 seconds]
06:15 < xtalmath> archels: well I can still imagine "backpropagation" mechanism to work at a very low level, for those circuits that contain cyclic feedback connections. but that does not explain learned patterns in feedbackless parts of the brain (if they even exist? perhaps the neurons most closely connected to input sensors, like close the retina).
06:16 < archels> the retina is an exceedingly intricate neural tangle
06:17 < archels> you're right, but only for very simple peripheral sensors such as mechano-, pain receptors
06:17 < xtalmath> kanzure: to be honest I don't see how transcription (co)factors and gene regulation can target a specific input weight among the many of this same neuron cell?
06:17 < kanzure> it's not a specific input weight
06:17 < kanzure> it's just more/less of the same
06:19 < xtalmath> kanzure: so according to you there is backpropagation at neuronal level?
06:20 < xtalmath> I don't even understand how this is still a question of debate with all these fluorescence microscopy and neuron chips...
06:21 < kanzure> http://diyhpl.us/~bryan/papers2/neuro/Review%20-%202007%20-%20AMPA%20receptor%20trafficking.pdf
06:22 < xtalmath> thx
06:26 -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-cxkljykapwdtcomm] has quit [Quit: Connection closed for inactivity]
06:40 < xtalmath> are there publicly available large datasets of neuron fires and synaptic protein levels for a bunch of connected neurons (say on chip or in culture)?
06:48 < xtalmath> "In particular, neurons haveevolved specific pathways to transport mRNAout  into  dendrites,  where  subsequent  localtranslation can occur" ! thx again kanzure
06:49 < xtalmath> so it is probably translation modulation instead of transcription modulation
06:51 < xtalmath> and somehow local (because of long dendrite distance, I assume the concentrations at different synapses can be very different, while having low concentration gradients along the long dendrites...)
06:52 < xtalmath> they are almost seperate cells
06:55 < xtalmath> one could speculate that different synaptic weights of the same neuron are correlated by how close they are as leaves on dendrite branches...
07:03 -!- Guest97788 is now known as HEx1
07:31 -!- proofoflogic [uid65184@gateway/web/irccloud.com/x-smtootuicnjpvgny] has quit [Quit: Connection closed for inactivity]
08:14 -!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap
08:16 -!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 272 seconds]
08:19 < kanzure> https://soundcloud.com/progressivehouseworldwide/sets/phw-elements-radio-digitally
08:26 < JayDugger> Anyone know 1) what relationship Rob Rinehart has with Soylent, and 2) what's this about "Soylent 2.0"?
08:28 < JayDugger> See here: http://robrhinehart.com/?p=1331
08:28 < kanzure> recipe is at http://robrhinehart.com/?p=424
08:29 < kanzure> see http://gnusha.org/logs/2013-02-20.log
08:31 -!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has joined ##hplusroadmap
08:33 < JayDugger> Thank you.
08:33 -!- augur_ [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has joined ##hplusroadmap
08:38 -!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has quit [Ping timeout: 256 seconds]
08:38 -!- sandeep [~sandeep@111.235.64.135] has joined ##hplusroadmap
08:39 -!- sandeep is now known as Guest97176
08:39 -!- Guest97176 [~sandeep@111.235.64.135] has quit [Read error: Connection reset by peer]
08:39 -!- sandeep_ [~sandeep@111.235.64.135] has joined ##hplusroadmap
08:43 -!- nmz787_i [~ntmccork@192.55.54.38] has joined ##hplusroadmap
08:46 -!- the8thbit [~8bit@66.186.100.194] has quit [Remote host closed the connection]
08:53 -!- Houshalter [~Houshalte@oh-71-50-58-200.dhcp.embarqhsd.net] has joined ##hplusroadmap
08:54 -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap
09:01 < ParahSailin_> so he claims that the algae oil is more efficient than plant oil
09:01 < ParahSailin_> you'd picture the algae being harvested from a solar pond of some sort right?
09:01 < ParahSailin_> nope "Solazyme grows microalgae in the dark, inside huge stainless-steel containers. The company's researchers feed algae sugar, which the organisms then convert into various types of oil. The oil can be extracted and further processed to make a range of fuels, including diesel and jet fuel, as well as other products
09:02 < kanzure> yes, solar doesn't work with dense algae ponds without lots and lots of mixing
09:02 < ParahSailin_> corn
09:02 < nmz787_i> omg like no one has figured out mixing
09:02 -!- the8thbit [~8bit@66.186.100.194] has joined ##hplusroadmap
09:03 < ParahSailin_> soylent guy is pretty delusional
09:03 < chris_99> how bad does soylent taste out of interest
09:04 < ParahSailin_> 1.3 tastes like oatmeal
09:04 < ParahSailin_> he should have stuck to the high oleic sunflower
09:08 -!- sandeep_ [~sandeep@111.235.64.135] has quit [Ping timeout: 244 seconds]
09:23 -!- nmz787_i [~ntmccork@192.55.54.38] has quit [Quit: Leaving.]
09:53 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 255 seconds]
09:54 -!- nmz787_i [ntmccork@nat/intel/x-pqtysdkdisxeouws] has joined ##hplusroadmap
09:57 -!- nmz787_i1 [~ntmccork@134.134.137.71] has joined ##hplusroadmap
09:58 -!- nmz787_i [ntmccork@nat/intel/x-pqtysdkdisxeouws] has quit [Ping timeout: 250 seconds]
10:13 -!- Madplatypus [uid19957@gateway/web/irccloud.com/x-apjrgwgiyqtickng] has joined ##hplusroadmap
10:17 -!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 264 seconds]
10:18 -!- crescendo [~mozart@unaffiliated/crescendo] has quit [Ping timeout: 252 seconds]
10:19 -!- crescendo [~mozart@unaffiliated/crescendo] has joined ##hplusroadmap
10:20 -!- FourFire [~FourFire@185.7.192.138] has quit [Ping timeout: 246 seconds]
10:32 -!- drethelin [drethelin@71-87-115-157.dhcp.mdsn.wi.charter.com] has joined ##hplusroadmap
10:34 -!- eudoxia [~eudoxia@r167-57-167-40.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap
10:41 -!- xtalmath [~xtalmath@ip-83-134-138-51.dsl.scarlet.be] has quit [Quit: Leaving.]
10:52 -!- sandeep_ [~sandeep@111.235.64.135] has joined ##hplusroadmap
10:55 -!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 240 seconds]
11:02 -!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap
11:18 -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-uusklojkzqnjrsoj] has quit [Quit: Connection closed for inactivity]
11:22 -!- FourFire [~FourFire@43-153-11.connect.netcom.no] has joined ##hplusroadmap
11:24 < nmz787_i1> juri_: the SEM I've got is a JSM-T200
11:29 -!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap
11:29 -!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 264 seconds]
11:32 -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 252 seconds]
11:34 < juri_> nmz787_i1: i've got a cambridge stereoscan 200 here.
11:36 -!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap
11:38 -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap
11:42 -!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 264 seconds]
11:43 -!- nmz787_i1 [~ntmccork@134.134.137.71] has quit [Read error: Connection reset by peer]
11:45 -!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap
11:49 -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 240 seconds]
11:54 -!- delinquentme [~delinquen@74.61.157.78] has joined ##hplusroadmap
12:00 < FourFire> nmz787, If you grow algae in the dark you don't need to tye up so much land surface area
12:01 < FourFire> and there are issues with grime buildup on the inside of the glass tubes
12:03 -!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Read error: Connection reset by peer]
12:03 -!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap
12:10 -!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 264 seconds]
12:11 -!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap
12:16 -!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 264 seconds]
12:18 < Houshalter> FourFire, well the sugar has to come from somewhere.
12:26 < FourFire> yes, but cane fields are a mature technology, and cheap
12:27 < FourFire> but building out glass tubes and maintaining them is probably way more epxnsive if you have to do it spread out over many square meters, instead of in an efficent, cubical building
12:27 < Houshalter> algae is more efficient at photosynthesis
12:27 < Houshalter> by like a lot
12:28 < delinquentme> why no go up?
12:28 < Houshalter> in 20 years most sugar will be produced by algae
12:28 < delinquentme> algae in vertical could be mush cheaper no?
12:28 < FourFire> delinquentme, shadows
12:28 < delinquentme> Houshalter, in 20 years well be on silicon :
12:29 < delinquentme> FourFire, penetration
12:29 < FourFire> also, water pressure?
12:29 < Houshalter> i doubt that
12:31 < delinquentme> arrange it that darkest growth is at bottom
12:33 < FourFire> delinquentme, I doubt it (silicon)
12:33 < delinquentme> ::D
12:34 < FourFire> maybe in 40, a large minority will be, but at significant personal disadvantage
12:34 < FourFire> ok, clear with me, how tall do you picture these vertical tubes?
12:35 < FourFire> Houshalter, I'm sure at larger scale, the incresed energy efficiency will make the initial capital investment of vast fields of glass worthwhile, but not before
12:36 < FourFire> a large company with loads of capital and the will to spend it makes it happen
12:37 < ParahSailin_> open ponds seem more worthwhile
12:38 < delinquentme> ParahSailin_, stop being so contrarian
12:38 < ParahSailin_> glass is pretty expensive
12:39 < ParahSailin_> can you imagine glass on a cash crop scale?
12:39 < delinquentme> FourFire, also if you aerate from below, you can adjust aeration rate to get  a nice density gradient upwards
12:40 < FourFire> ParahSailin_, that's my point
12:40 < FourFire> but open ponds waste water
12:40 < ParahSailin_> yeah thats problem
12:41 < FourFire> we donæt have that much water unless we've solved energy and can just desalinate
12:41 < ParahSailin_> theres algae that grow in saline
12:41 < FourFire> you don't want to salt that much land
12:42 < ParahSailin_> theres lots of australia thats just wasteland
12:43 < ParahSailin_> the chinese at one point had a plan to turn some of the desert into an inland sea
12:43 < ParahSailin_> with massive amounts of pumping
12:44 < delinquentme> do you meant that semiconductor wastewater oasis?
12:44 < delinquentme> theres no salt! just a few heavy metals !
12:47 -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap
12:48 < Houshalter> FourFire, maybe, sugar is quite expensive.
12:49 < Houshalter> FourFire, not everyhwere in the world has a water shortage
12:49 < Houshalter> i live in a former swamp and we have more water than we know what to do with
12:50 < FourFire> no, but if we begin to actually use our resources efficiently in order to develop the amount of infrastructure we actually need, we will quickly run short of anything we don't produce
12:51 -!- FourFire [~FourFire@43-153-11.connect.netcom.no] has quit [Read error: Connection reset by peer]
12:51 < delinquentme> FourFire, that can be said for the sun.
12:51 -!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 265 seconds]
12:51 -!- FourFire [~FourFire@43-153-11.connect.netcom.no] has joined ##hplusroadmap
12:52 < FourFire> (non/slowly renewable ground and surface freshwater being one of those resources)
12:52 < Houshalter> water is not a finite or even particularly scarce resource. If you are using it for algae, it doesn't even need to be clean.
12:52 < FourFire> sure, but how much oil are we using?
12:52 < FourFire> how much oil will we need to power the world in 10, 20 years
12:53 < FourFire> we're going tp have to cover the majority of the petroleum supply with bio alternatives, because of the peak
12:53 < drethelin> peak oil is not a thing
12:54 < drethelin> people talk about the available supply running out
12:54 < FourFire> Houshalter, what's the billion words dataset?
12:54 < drethelin> but that's because people only spend the minimum amount possible ot maintain an available supply
12:54 < Houshalter> FourFire, what how did you find out about that?
12:54 < drethelin> talking about only having enough oil to last us a few years is like talking about grocery store food only lasting us a few days
12:54 < FourFire> drethelin, we aren't going to exhaust available oil, we'll just use up all the priftable oil
12:54 < drethelin> that's a nonsensical statement
12:55 < drethelin> "profitable at what price"
12:55 < Houshalter> FourFire, it's not the amount of oil that needs produced ,it's whether it's even economical at all
12:55 < drethelin> not "profitable oil"
12:55 < drethelin> consider this: it costs money to find oil
12:55 < delinquentme> kanzure, surely someones algae'd in here before
12:55 < drethelin> this money needs to be spent to find it
12:55 < drethelin> if you have a large stockpile
12:55 < drethelin> why would you spend money finding oil
12:55 < Houshalter> we will worry about running out of fresh water to grow algae when we are already producing thousands of gallons a minute in a swamp somewhere
12:55 < drethelin> before you have to?
12:55 < kanzure> delinquentme: i worked on an algae biofuel project and an alge insulin project, what do you want
12:55 < FourFire> drethelin, about opportunity cost
12:56 < Houshalter> nothing wrong with using sea water either
12:56 < delinquentme> I want dougnuts
12:56 < FourFire> if you need to provide X energy per time in order to have Y productivity otherwise it will cost you Z strategic position
12:56 < delinquentme> but these guys are algaeing or brainstorming on it
12:56 < FourFire> then you pay for X
12:57 < Houshalter> FourFire,  https://www.kaggle.com/c/billion-word-imputation
12:57 < FourFire> of course, I'd prefer that we all just put solar panels on our roofs and found a better method for electrical energy storage and upgraded out logistics netowrk to electric tech
12:57 < FourFire> but that's not going to happen in time
12:57 < drethelin> in time for what
12:58 < drethelin> you're getting your predictions from status-whoring fearmongers
12:58 < drethelin> the other thing is
12:58 < drethelin> there is no discontinuity
12:58 < drethelin> we won't "suddenly" run out of oil
12:58 < FourFire> drethelin, in time to provide for rising demand as supply shortfalls
12:58 < drethelin> there's not a due date
12:59 < drethelin> there will be no supply shortfalls
12:59 < FourFire> of course not, we're gradually lowering production
12:59 < drethelin> 'I hope next year's crops come in in time to make up the supply shortfalls from last year!"
13:00 < drethelin> it's an incoherent position
13:00 -!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap
13:00 -!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Client Quit]
13:00 -!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap
13:00 < FourFire> ok, the catastrophe I am implying is a dramatic lowering in median wealth, because the cost of using, energy, on a personal basis, and indirectly resources which require energy to be produced, will be relatively higher over time rather than lower, as the trends have been these past... what 300+ years?
13:00 < drethelin> yes
13:00 < drethelin> and Im saying that will never happen
13:01 < drethelin> it will not be dramatic
13:01 < drethelin> technological improvements are dramatic changes
13:01 < FourFire> ok, what reason leads you to believe it?
13:01 < drethelin> because markets are a thing, and aggregation is a thing
13:02 < drethelin> there is not some single big pool of oil that will suddenly run out
13:02 < Houshalter> drethelin, yes oil prices have never increased rapidly before and it's totally impossible they could ever do that to a greater degree
13:02 < FourFire> Ok, I admit that my mental state is unstable today
13:02 < FourFire> today has been a pessimistic day from my perspective, due to IRL events
13:02 < drethelin> Househalter: oil prices have both increased and decreased rapidly
13:02 < FourFire> Houshalter, you will have noticed this from our earlier discussion
13:02 < drethelin> but not from some cosmic or environmental reason causing us to run out of oil
13:03 < FourFire> I didn't say anything about cosmic
13:04 < FourFire> The model behind the concept of peak oil is based off a well regarded geologist's research
13:04 < Houshalter> they are pumping it out of the ground much faster than it's economically rational to do
13:04 < drethelin> what does that mean
13:04 < FourFire> I mean, sure, I'm hopeful that this time Fusion isn't just 20 more years away
13:04 < drethelin> why would they do that?
13:05 < drethelin> Fourfire: so is every model that has failed to come true
13:05 < FourFire> that the price of solar panel production and installation will fall below the price of purchasing firing oil, or gas
13:05 < drethelin> there are THOUSANDS of well regarded scientists in almost every field
13:05 < drethelin> some of them will ALWAYS predict disasster
13:05 < Houshalter> who's they? You have a bunch of oil in the ground and you need to sell it. In the long run it would make way more sense to keep it in the ground and wait till prices rise, but who is going to take that bet? With who's money?
13:05 < drethelin> people thought there was going to be a giant population crash or an ice age or mass starvation
13:06 < Houshalter> drethelin, oh yes someone in the past made a bad prediction therefore all predictions are wrong and bad things can never happen
13:06 < FourFire> yeah we stalled the ice age bac in the bronze age, due to deforestation
13:06 < FourFire> population crash, we invented fertilizers and other agriculture stuff
13:06 < drethelin> Houshalter: predicting the end of the world makes you well known and noticed. It should be no surprise that predictions of the end of the world are more common than world ending events
13:06 < FourFire>  we learned how to make a world that sort of worked, where <5% of the population produced the food for everyone
13:07 < FourFire> and we did it all by burning petroleum
13:07 < FourFire> drethelin, I'm not predicting the end of the world, just a slowing down of technological development, which is personally fatal
13:10 < FourFire> hey I'm all for looking forward to the best futre, where technology accelerates just fst enogyh, where we platau at above average human artifical general intelligences and fix all the coordination problems, then launch into space, and then AI takes off due to massive amounts of computational resources made avilible by permitting recycling of the earth's constituent atoms (once we've recorded it, of course)
13:11 < delinquentme> ooo theres an interesting storage problem.
13:11 < delinquentme> if we wanted to 'store' the arrangement of atoms on the earth. How large would that storage medium need to be
13:11 < Houshalter> FourFire, that is a rather specific prediction of future events and depends on a lot of things going just right
13:12 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap
13:12 < FourFire> yeah, a lot could go wrong
13:12 < FourFire> there could be antagonistic social warfare if tech progresses too quickly, too visibly
13:12 < Houshalter> I just count on one thing going right: FAI. Everything else is a distraction
13:12 < FourFire> we could completely lose control to half formed artificial intelligences
13:13 < drethelin> the point is not that something ccan't go disastrously wrong
13:13 < drethelin> the point is that any given disaster that someone is pushing is probably more self-serving
13:13 < drethelin> than it is informative
13:13 < drethelin> and that for a LOT of disasters we have good evidecne that they are not in fact imminent
13:13 < FourFire> which is the most likely thing to go wrong, besides things we already know for sure, climate change, etc.
13:13 < drethelin> climate change is a paper tiger
13:14 < FourFire> it's no danger to pretty much anyone right now, but if we donæt do anything about it, then a lot of people are going to be dying/moving in the 50s, onwards
13:14 < eudoxia> the way i see it, there's two competing trends: ephemeralization (leads to MNT) and resource exhaustion (leads to war etc.), and i don't really know which trend will overtake the other
13:14 < Houshalter> " list of major governments in 1900 would probably put the Ottoman Empire or Austria-Hungary well ahead of the relatively young United States.  Citing the good track record of the US alone, and not all governments of equal apparent stability at the start of the same time period, is purest survivorship bias."
13:14 < Houshalter> http://lesswrong.com/lw/hy/riskfree_bonds_arent/
13:14 < eudoxia> although to be fair the world3 business-as-usual scenario has more or less been on track with reality and leads to mass die-offs
13:15 < drethelin> I'm with taleb and scott adams
13:15 < drethelin> slow moving predicted risks are not what you should be worreid abou
13:18 < kanzure> "lose control" you never had control
13:18 < eudoxia> heh
13:19 < FourFire> eudoxia, the problem with ephemeralization is that a lot of the gains are lost to https://en.wikipedia.org/wiki/Jevons_paradox
13:20 < FourFire> suddenly people can do really expensive stuff, for the same cost as doing earlier stuff
13:20 < FourFire> and sometimes that expensive stuff is wasteful or even harmful to other resources
13:22 < FourFire> drethelin, is AI "slow moving"?
13:23 < FourFire> meh I'm giving it up for the day, not in a contructive mood
13:23 -!- FourFire [~FourFire@43-153-11.connect.netcom.no] has left ##hplusroadmap ["Leaving"]
13:23 < drethelin> AI is not slow moving
13:23 < drethelin> neither is a bio engineered pandemic or an asteroid impact
13:26 < juri_> lets get working on AI, improved biology, and leaving the planet then.
13:26 < drethelin> "improved biology'
13:27 < drethelin> the problem there is working on improved biology is exactly the sort of thing that would let someone make SmallAidsPox
13:28 < juri_> AI -> drone warfare && rockets -> ICBMs. we've survived those.
13:31 -!- rigel_ [~yourmom@c-24-21-55-24.hsd1.or.comcast.net] has joined ##hplusroadmap
13:32 < archels> http://www.nrel.gov/news/features/feature_detail.cfm/feature_id=19532
13:32 < archels> "C02"
13:33 -!- c0rw|timetravel [~c0rw1n@91.176.95.206] has joined ##hplusroadmap
13:35 < kanzure> c0rw|timetravel: greetings
13:35 < c0rw|timetravel> hello kanzure
13:35 < c0rw|timetravel> hello transhumanists also
13:36 < eudoxia> hello
13:38 < kanzure> are you defecting
13:40 < c0rw|timetravel> who, me?
13:41 < c0rw|timetravel> defecting against what ?
13:45 < kanzure> lesswrong
13:45 < c0rw|timetravel> ... they don't like h+ ?
13:46 < kanzure> it's an existential risk to them
13:48 < c0rw|timetravel> from here i don't see how (yet)
13:49 < drethelin> do you mean they view it as an existential risk
13:49 < kanzure> yes
13:49 < drethelin> or that the H+ cocmmunity is an existential risk to the lesswrong community
13:49 < kanzure> no
13:50 < kanzure> because both communities are equally ineffective
13:50 < drethelin> sure but communities that share mindspace are probably the biggest threats to each other's xontinued existence
13:50 < c0rw|timetravel> no
13:50 < c0rw|timetravel> subculture archipelago says they're not
13:52 < c0rw|timetravel> nothing prevents anyone from being in both, and more or less involved with either
13:53 < drethelin> it's not about prevention
13:53 < kanzure> er, right, i am just teasing you
13:53 < c0rw|timetravel> incompatible goals?
13:54 < kanzure> yes kinda
13:55 < c0rw|timetravel> i don't see how, but then i just arrived here
13:56 < kanzure> aren't you c0rwin?
13:56 < c0rw|timetravel> yes
13:56 < kanzure> excuse me, i mean c0rw1n
13:57 < c0rw|timetravel> i use c0rw|timetravel for when i might respond in channels to things said hours ago
13:57 < kanzure> time travel is pretty dangerous
13:58 < c0rw|timetravel> LOL
13:58 < drethelin> https://www.youtube.com/watch?v=hG5I-54Yu4o'
13:59 < drethelin> https://www.youtube.com/watch?v=hG5I-54Yu4o
13:59 < kanzure> .title
13:59 < yoleaux> I *WILL* MESS WITH TIME - Rick and Morty Season 2 Episode 1 - YouTube
13:59 < kanzure> i was thinking more like https://www.youtube.com/watch?v=KJRh-37H4fA
14:02 < kanzure> or https://www.youtube.com/watch?v=k-e640rn8S4
14:06 -!- c0rw|timetravel is now known as c0rw1n
14:13 -!- BlueLobster [medea@unaffiliated/fellini] has quit [Ping timeout: 246 seconds]
14:14 -!- BlueLobster [medea@persephone.darkness-reigns.net] has joined ##hplusroadmap
14:17 -!- delinquentme [~delinquen@74.61.157.78] has quit [Ping timeout: 246 seconds]
14:18 < c0rw1n> ... what is it that h+ wants to do that lesswrong would find an x-risk ?
14:18 < kanzure> intelligence augmentation
14:18 < kanzure> nootropics
14:18 < c0rw1n> *blink* *blink* how do they NOT want that ?
14:18 < kanzure> recursive self-improvement
14:19 < kanzure> what? they think that's the fucken end of the world.
14:19 < drethelin> kanzure as always you are being a jackass
14:19 < kanzure> go on
14:19 < drethelin> in general Lesswrong approves of nootropics and intelligence augmentation
14:19 < drethelin> IN HUMANS
14:20 < drethelin> because the timescales for recursive self improvement are long enough that if something crazy starts happening you can react
14:20 < kanzure> nobody at lesswrong strongly believes that recursive self-improvement is particularly safe in humans but not ai
14:20 < drethelin> really?
14:20 < drethelin> I would best most people would if you asked them
14:20 < drethelin> bet
14:20 < kanzure> because humans are friendly....?
14:20 < drethelin> because humans are SLWO
14:20 < drethelin> SLOW
14:21 < kanzure> so something like "humans are so slow that no amount of recursive self-improvement would fix that slowness"?
14:21 < drethelin> humans are not capable of recursive self improvement to the same extent as digital entities
14:21 < c0rw1n> .. yet ...
14:21 < drethelin> "yet"
14:21 < drethelin> there is no yet about it
14:21 < drethelin> at the point where humans can recursively self improve as fast as a totally digital intelligence they will no longer be humans
14:21 < c0rw1n> ( this is *begging* to go into "is an Em an human" isn't it )
14:21 < kanzure> that's just semantics
14:21 < drethelin> like talking about a cheetah that can outrun a rocket
14:21 < c0rw1n> yeah
14:21 < kanzure> nobody cares if they are human
14:22 < kanzure> and why can't my rocket cheetah count?
14:22 < drethelin> because it's the rocket that's fast
14:22 < kanzure> ethics board said no, that's why i left to start my super zoo
14:22 < drethelin> if you strap a human to a recurisvely improving computer
14:22 < drethelin> it's the computer you should be scare of
14:22 < kanzure> strapped a rocket to a cheetah but then this happened http://vignette2.wikia.nocookie.net/familyguy/images/1/19/Chester_Cheetah.jpg/revision/latest?cb=20090907190317
14:22 < Houshalter> you can tie a rocket to the back of a cheetah, but it kills the cheetah
14:22 < drethelin> consider how long it takes to run ONE experiment on human improvement
14:22 < c0rw1n> in the space of possible goals to have, humans generally cluster around a pretty tiny place
14:23 < drethelin> humans can't install extra ram
14:23 < c0rw1n> machines, well, mumble murgle clippy
14:23 < drethelin> they can't control multiple bodies simultaneously
14:23 < drethelin> or network
14:23 < c0rw1n> yet
14:23 < kanzure> drethelin: i think that you seem to have a broken definition of improvement
14:23 < drethelin> kanzure: ?
14:24 < kanzure> small improvements can lead to other improvements
14:24 < drethelin> yes
14:24 < drethelin> and?
14:24 < kanzure> perhaps your argument is "humans would take so long tha ti expect that ai stuff will happen before anything interesting happens"?
14:24 < kanzure> *that i expect
14:24 < drethelin> yeah pretty much
14:25 < drethelin> human generations are slow
14:25 < kanzure> you don't need human generations
14:25 < drethelin> you don't ALWAYS need them
14:25 < kanzure> did i ever mention my nootropics plans
14:25 < drethelin> I assume they're absurd
14:25 < drethelin> but right now our most promising avenue for understand intelligence and intelligent enhancement is genetic
14:25 < drethelin> I think
14:26 < drethelin> in which field you need generations to test innovations
14:26 < kanzure> microbial nootropics
14:26 < drethelin> even if we learned to clone von neumann
14:26 < drethelin> he'd still have to grow up
14:26 < c0rw1n> nootropics can  only marginally augment flesh capabilities, even if it's like one or two OOMs, digital would go so so much further
14:26 < kanzure> selection for which microbes migrate to certain brain regions
14:26 < kanzure> c0rw1n: yes but we don't have human brain emulations yet, so.....
14:26 < c0rw1n> inorite ;_;
14:27 < kanzure> i'm also not convinced about upper limits of flesh capabilities
14:27 < kanzure> there's a lot of stupid brain mass
14:27 < drethelin> so my belief is basically, there are two main "steps" to recursive intelligence improvement
14:27 < drethelin> hardware and software
14:27 < drethelin> code optimization and hardware improvements
14:27 < drethelin> in both of these humans are harder to work in than digital entities
14:28 < kanzure> so far digital entities haven't been made yet
14:28 < kanzure> be practical
14:28 < drethelin> ok fair point
14:28 < drethelin> but code optimization is still a thing we sometimes do
14:28 < kanzure> see http://diyhpl.us/wiki/projects/human-like-cognitive-abilities/
14:28 < drethelin> (when we're not relying on having more processing power than we know what to do with)
14:28 < drethelin> but a human's ability to access and modify his code is negligible
14:28 < drethelin> compared to a theoretical AI's
14:29 < kanzure> yes compared to some imaginary bullshit, it can be whatever
14:29 < kanzure> this doesn't help me
14:29 < drethelin> I'm not trying to help you?
14:29 < kanzure> no, i mean comparison to AIs doesn't help me
14:29 < drethelin> again
14:29 < drethelin> I'm not ttying to help you
14:29 < drethelin> I'm trying to explain why lesswrong mostly doesn't give a shit about recursive human self improvement
14:29 < kanzure> you were trying to argue about why self-modifying humans could never go fast enough
14:30 < kanzure> why wouldn't a self-modifying human be faster at making accurate human brain emulations than a non-self-modifying human?
14:30 < kanzure> hmph
14:30 -!- drewbot [~cinch@ec2-54-159-162-239.compute-1.amazonaws.com] has quit [Remote host closed the connection]
14:30 < drethelin> ??
14:30 < drethelin> because a human can't self-modify fast
14:30 -!- drewbot [~cinch@ec2-54-224-113-37.compute-1.amazonaws.com] has joined ##hplusroadmap
14:30 < c0rw1n> why emulate a human brainin a human brain ...
14:30 < kanzure> c0rw1n: because emulating a mouse brain is less useful
14:30 < drethelin> if a human comes up with an improvement to the humman equivalent of RAM architecture
14:30 < kanzure> drethelin: i don't think that's true; the design-test-build cycle can be significantly shortened.
14:30 < drethelin> he still can't exactly slot that into his own brain saffely
14:31 < drethelin> the way a computer can
14:31 < kanzure> that's true, you need sandboxing, but so what
14:31 < c0rw1n> yeah, would need backups
14:31 < drethelin> so that slows things down A LOT!
14:31 < kanzure> compared to something that doesn't exist yet :-)
14:31 < drethelin> neither does your theoretical bullshit enhanced human
14:31 < drethelin> so cut the sass
14:31 < kanzure> nootropics exist (although they are all weak)
14:32 < drethelin> weak enough that we only have anecdotes for net beneficial effects
14:32 < kanzure> so you're not too hot on the microbial nootropics plan, eh?
14:32 < drethelin> no I like it
14:32 < drethelin> I'm in favor of human enhancement
14:32 < kanzure> hm.
14:32 < drethelin> I'm just not worired about it going recursive and creating someone smart enough to take over the world
14:32 -!- delinquentme [~delinquen@74.61.157.78] has joined ##hplusroadmap
14:33 < kanzure> i think a person could already take over the world, without the assistance of a nootropic
14:33 < c0rw1n> ... 'finil and 'cetam seem to work preeetty well doe't they ( but ok, no order-of-magnitude improvement there though )
14:33 < c0rw1n> i t'm really not sure
14:33 < c0rw1n> *i'm
14:33 < drethelin> depends on your definition of take over
14:33 < kanzure> c0rw1n: my definition of a minimally-viable nootropic is a nootropic that endows you with the ability to make a better nootropic
14:33 < kanzure> it's an escape-velocity nootropic thing
14:33 < drethelin> people clearly can do stuff like be Stalin or Mao
14:33 < drethelin> hmm
14:34 < c0rw1n> heh, then feed finil and cetam to a bunch of pharma researchers?
14:34 < drethelin> who are the biggest names who use nootropics?
14:34 < kanzure> hell yeah
14:34 < kanzure> drethelin: erdos
14:34 < drethelin> heh
14:34 < drethelin> more like erdosseral
14:34 < drethelin> adderalldos
14:34 < kanzure> oh come on be nice
14:34 < drethelin> hmm
14:34 < kanzure> sounds like a children's story
14:34 < drethelin> anyway I said use not used
14:35 < drethelin> does musk or tao use modafinil?
14:35 < c0rw1n> who knows
14:35 < kanzure> musk just never sleeps instead
14:35 < drethelin> imagine if people are right about how important sleep is
14:35 < drethelin> and he would actually justs be wway smarter
14:35 < drethelin> if he slept 8 hours
14:36 < kanzure> i don't know if it's smartness that he needs more of
14:36 < drethelin> fair
14:36 < kanzure> i guess it would always be nice eh
14:36 < drethelin> who's the elon musk of biotech
14:36 < kanzure> craig venter :-(
14:38 < delinquentme> me
14:38 < kanzure> you don't have the cheeks for it
14:39 < delinquentme> someone touch muh butt?
14:39 < delinquentme> roflmao
14:39 < delinquentme> sorry idk why my OKC persona crossed w the transhumanist
14:39 < kanzure> nah that's how i setup mine
14:40 < drethelin> hmm
14:40 < drethelin> I wonder if I should be more openly transhumanist on okcupid
14:40 < kanzure> my profile is... definitely transhumanist.
14:40 < kanzure> http://web.archive.org/web/20150625162512/http://www.okcupid.com/profile/kanzure
14:44 -!- Houshalter [~Houshalte@oh-71-50-58-200.dhcp.embarqhsd.net] has quit [Ping timeout: 265 seconds]
14:50 < delinquentme> I want to buy a cell line . but not pay the commercial cost
14:50 < kanzure> erdos seems like an alright role model
14:50 < delinquentme> how do I do this?
14:50 < delinquentme> or at least ... I'd like to defer the cost
14:50 < drethelin> you could go to a university or somethin
14:50 < drethelin> ask a grad student to breed you some cells
14:50 < kanzure> how can you not afford a cell line, i thought you had a $200k budget
14:50 < c0rw1n> how are the commercial ones manufactured ?
14:50 < drethelin> pay for the growth media and so on
15:00 < justanotheruser> kanzure: why do you like on your profile?
15:01 < justanotheruser> *lie
15:01 < drethelin> gives romantic partners the fun of uncovering the truth
15:01 < c0rw1n> lulz
15:01 -!- mgin [~mgin@unaffiliated/mgin] has quit [Quit: mgin gonna mgin]
15:01 < justanotheruser> Drugs Never
15:02 < drethelin> probably because the standard definition of Drugs
15:02 < drethelin> is like "marijuana and cocaine"
15:02 < drethelin> and not "noopept and theanine"
15:03 < kanzure> because "i have 200 empty adderall bottles" wasn't any of the answer options
15:03 < chris_99> haha
15:04 < justanotheruser> do you put them on a shelf like a frat boy puts liquer bottles in a shelf?
15:05 -!- eudoxia [~eudoxia@r167-57-167-40.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving]
15:05 < juri_> with all that's wrovg with me, i now store smaller parts of laptops (screws and such) in perscription bottles.
15:05 < kanzure> no clue
15:05 < c0rw1n> that's a good idea
15:05 < c0rw1n> prevents loss of screws and such
15:07 < kanzure> you can also just use sticky tape
15:07 < drethelin> prescription bottles are actually a really good little container
15:07 < drethelin> transparent and they won't all open accidentally if you drop em
15:09 < c0rw1n> yeah, seems like a perfectly good storage method
15:10 < c0rw1n> nothing wrong with it, nor with you for using it
15:10 < delinquentme> http://www.dorkly.com/post/75600/watch-ronda-rousey-gush-about-her-cartoon-crush-vegeta # off topic ish
15:10 < kanzure> vegeta is completely on topic
15:11 < c0rw1n> how so ?
15:12 < justanotheruser> vegeta is transhuman
15:13 < kanzure> https://www.youtube.com/watch?v=lgaYv37stMk
15:13 < kanzure> he's hilarious
15:13 < drethelin> vegeta isn't a huma1
15:13 < drethelin> human
15:13 < kanzure> aww it didn't show the gorilla part
15:15 < justanotheruser> drethelin: Yeah he's not human, he's transhuman, that's what I'm saiyan
15:15 < c0rw1n> >.<
15:15 < drethelin> ughghghghghghghghg
15:17 < kanzure> plus his scouter
15:17 < drethelin> google glass power level over 1500$
15:17 < c0rw1n> BWAHAHA
15:43 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 255 seconds]
15:59 -!- nmz787_i [~ntmccork@192.55.54.36] has joined ##hplusroadmap
16:02 < nmz787_i> $48k for post-halcyon-tech portable SEM (small enough to fit in large backpack)
16:07 -!- eudoxia [~eudoxia@r167-57-102-197.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap
16:07 < eudoxia> oh my god
16:07 < eudoxia> i was reading old logs
16:07 < eudoxia> http://gnusha.org/logs/hplusroadmap-logs/2012-07-19.log
16:07 < eudoxia> kanzure: was the guy who contacted you about pyphantomjs this guy https://github.com/san-mate
16:07 < c0rw1n> yeah, what timestamp pls ?
16:08 < c0rw1n> oh, not for me
16:08 < c0rw1n> 'k
16:09 -!- drewbot_ [~cinch@ec2-54-158-204-187.compute-1.amazonaws.com] has joined ##hplusroadmap
16:13 < kanzure> not sure
16:14 -!- drewbot [~cinch@ec2-54-224-113-37.compute-1.amazonaws.com] has quit [Ping timeout: 246 seconds]
16:15 < eudoxia> well, i did not know him in 2012
16:15 < eudoxia> but if he was that person then he was employed where i work
16:15 < eudoxia> what a small world huh
16:28 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Remote host closed the connection]
16:33 < nmz787_i> so ibidi makes a disposable fludic that almost would work for the TdT synthesis...
16:33 -!- eudoxia [~eudoxia@r167-57-102-197.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving]
16:34 < nmz787_i> http://ibidi.com/xtproducts/en/ibidi-Labware/Flow-Chambers/m-Slide-Membrane-ibiPore-Flow
16:34 < nmz787_i> simpore says they could help me build a similar device using their nanoporous product instead of the microporous one
16:36 < nmz787_i> (i've talked to simpore about this before, years ago, since they were in Rochester)
16:49 < kanzure> acetaminophen-producing bacteria https://news.ycombinator.com/item?id=10000026
16:50 < CaptHindsight> nmz787: do you know the range of pore sizes of the nanopore versions?
16:51 < CaptHindsight> is there more than just the 0.5um?
16:53 -!- sandeep_ [~sandeep@111.235.64.135] has quit [Ping timeout: 240 seconds]
16:54 -!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap
16:54 -!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap
16:54 < nmz787_i> weel the 0.5 is a microporous
16:54 < nmz787_i> they can tune the size from like 5nm up to 100s I think
16:54 < nmz787_i> with variation in size increasing with increased target pore size
16:55 < nmz787_i> so if we wanted a 20nm pore on avg... we'd get a spread from probably 10nm to 30nm in range
16:55 -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 240 seconds]
16:55 < nmz787_i> so well within oligo diameter, and well away from enzyme size
16:56 -!- nmz787_i [~ntmccork@192.55.54.36] has quit [Quit: Leaving.]
16:57 -!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Read error: Connection reset by peer]
16:57 -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap
17:01 -!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Quit: Leaving.]
17:01 -!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap
17:16 -!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap
17:17 -!- xtalmath [~xtalmath@ip-83-134-138-137.dsl.scarlet.be] has joined ##hplusroadmap
17:20 -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 252 seconds]
17:21 -!- Beatzebub [~beatzebub@d162-156-106-225.bchsia.telus.net] has joined ##hplusroadmap
17:25 -!- delinquentme [~delinquen@74.61.157.78] has quit [Ping timeout: 250 seconds]
17:25 -!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has joined ##hplusroadmap
17:45 -!- delinquentme [~delinquen@74.61.157.78] has joined ##hplusroadmap
17:53 -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap
17:54 -!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 264 seconds]
18:00 < fenn> so steve said something worrying about my complaining about the mouse brain ATUM/EM scan thing in nature the other day
18:01 < fenn> i said "it's not very useful because there isn't even an entire cell in the scanned volume" and he was like "good. that means more time to develop FAI theory in the small window between developing enough credibility to raise funding and the emergence of neuromorphic AI'
18:01 -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 272 seconds]
18:02 < fenn> thus putting ${people who want to understand the brain} into conflict with ${people who want to save the world from AI}
18:03 < c0rw1n> "save the world FROM ai" group is already a roadkill of history, like feudalists and communists.
18:03 < c0rw1n> Might be getting somewhere on "save the world USING AI" , yes.
18:05 < fenn> i don't get what these math people think they are doing, when they have no direct experience with how an AI works
18:06 < fenn> (because there aren't any in existence afaik)
18:06 < fenn> there is a long roadkill of history of people pretending they know what AI will be like
18:06 < c0rw1n> looks like a standard fallacy there fenn
18:06 < c0rw1n> we already have a lot of AI tech
18:06 < c0rw1n> not AGI yet, sure
18:07 < fenn> it looks just as bad as the boundless optimism of mccarthy in the 70s with regards to symbolic ai
18:08 < fenn> "we don't understand how motivation works but let's make lots of predictions about it anyway"
18:08 -!- mgin [~mgin@unaffiliated/mgin] has joined ##hplusroadmap
18:09 -!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap
18:10 < c0rw1n> don't patternmatch to that. It'll be neuromorphic AI. Maybe not with the sort of NNs we have now, maybe we'll need finer values than just one input and one output numbers per neuron per tick, but it'll happen and it'll happen there.
18:10 < fenn> uh 1960s i mean
18:11 < fenn> pre-1969-singularity :P
18:12 < fenn> Moravec explains, "Their initial promises to DARPA had been much too optimistic. Of course, what they delivered stopped considerably short of that. But they felt they couldn't in their next proposal promise less than in the first one, so they promised more."
18:12 < c0rw1n> lol @ the dance of incentives
18:13 -!- yashgaroth [~ffffff@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has joined ##hplusroadmap
18:14 < fenn> i'm pretty impressed with the capabilities of RNN and deep belief networks, but i don't see anyone combining them into agents that try to actually do anything at all, they are just used for prediction and classification
18:14 < c0rw1n> nah, there is much scarier than that
18:15 < c0rw1n> tiling NNs in cortical-like structures
18:15 < fenn> what's the difference between that and a deep belief network
18:16 < c0rw1n> could well make a full brain with those, idk that deep belief networks are that neuropmorphic
18:17 < c0rw1n> but then idk deep belief nets so idk
18:17 < fenn> i think they are pretty similar in layout, it's just a matter of parameters
18:18 < c0rw1n> oh. ok then
18:18  * c0rw1n goes google for DBNs
18:18 < fenn> there is a ton of other biological squishy stuff that probably matters too, but apparently it's not as important as the basic architecture because deep belief networks actually work
18:18 < fenn> it's just a multilayer neural network
18:19 < c0rw1n> nah, i mean .. ok you know how the visual cortex is a bunched-up plane of hypercolumns of neurons
18:20 < c0rw1n> and RNNs doing visual works like classification are not
18:20 < fenn> like i said, it's just parameters
18:20 -!- nmz787_i [~ntmccork@192.55.55.41] has joined ##hplusroadmap
18:20 < fenn> you can set your minicolumn size to equal the human brain or not
18:21 < c0rw1n> i think the structure itself is what's important (and i might well be completely, spectacularly wrong)
18:22 < fenn> sure it's important but that doesn't mean only one parameterization will be able to do things in the world
18:22 < c0rw1n> oh, agreed
18:22 < fenn> supposedly autistic people have smaller minicolumns in larger numbers, and some of them are extremely talented
18:23 < c0rw1n> 's just, we have a model that we know already works (natural brains), so let's begin by replicating that and then see how to optimize it
18:23 < c0rw1n> also i want muh upload ^_^
18:24 < fenn> sure i agree that's the best approach because the machine will at least be understandable to humans
18:24  * fenn looks at http://minicolumn.org/publications/autism-pamphlet.html
18:26 < nmz787_i> .tell fourfire why not just use a salt-water algae?
18:26 < yoleaux> nmz787_i: I'll pass your message to fourfire.
18:26 < fenn> the ethylene thing is cute because it sidesteps the "how do we collect all these tiny things that clog filters" problem
18:26 -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap
18:27 < fenn> removing water from the algae culture was the hardest part of the process
18:27 -!- sh [~sh@89.101.222.213] has quit [Read error: Connection reset by peer]
18:27 < fenn> nmz787_i: cyanobacteria already grow in salt water
18:27 < fenn> fourfire was just doom-mongering
18:28 < nmz787_i> fenn: the obvious answer is to engineer a lipid membrane-pump (system)
18:29 -!- superkuh [~superkuh@unaffiliated/superkuh] has quit [Excess Flood]
18:30 -!- superkuh [~superkuh@unaffiliated/superkuh] has joined ##hplusroadmap
18:30 < nmz787_i> so there was an xray sensor guy at the conference (i spent all afternoon in the vendor area, wow, I wish the local mall had a tech wing like that)
18:30 < nmz787_i> and I saw the opamp next to the Silicon pin-diode;
18:30 < nmz787_i> and he told me they use peltier to cool the sensor and FET, but nothing else
18:30 < nmz787_i> which is pretty cool
18:31 -!- sh [~sh@89.101.222.213] has joined ##hplusroadmap
18:32 < nmz787_i> and then SPI chem suppliers had a self-contained battery-powered microfluidic pump with a SiNi window over the microchannel (in PDMS) and after bonding the window to the microchannel, they put a 3 micron hole in the SiNi window... then put the whole thing in an SEM/TEM
18:32 < nmz787_i> I guess SEM
18:32 < fenn> you can do x-ray spectroscopy with that sensor?
18:32 < nmz787_i> and he said the surface tension was enough to keep the water/liquid from escaping, and that evaporation wasn't enough to foul the e-beam or imaging/xray analysis
18:33 < nmz787_i> fenn: oh yeah, PIN diodes have a thickness corresponding with the energy of xrays
18:33 < nmz787_i> I didn't think of a peltier to cool it though
18:34 < nmz787_i> the older alternatives were SiLi which have been superseded by SDD (silicon drift detector)
18:34 < nmz787_i> but PIN are common for handheld xrf spectrometers
18:34 < nmz787_i> (he said)
18:34 < nmz787_i> just slower
18:34 < nmz787_i> and don't /need/ cooling with LN2
18:34 < nmz787_i> (as SiLi and SDD do)
18:35 < nmz787_i> kanzure: wanna buy me a $48k mochii SEM?
18:36 < nmz787_i> (USA made)... or a 70k / 120k (comes with edx) larger korean one?
18:36 < nmz787_i> 2 week lead time on either
18:36 < nmz787_i> http://www.coxem.com/ds1_1.html
18:38 < nmz787_i> the mochii is small enough to fit in a backpack, while the coxem is not
18:38 < nmz787_i> but it could fit into a compact car
18:54 -!- Dumpster_D1ver [~Vishnu@pool-72-83-63-112.washdc.east.verizon.net] has quit [Ping timeout: 264 seconds]
18:57 -!- Dumpster_D1ver [~Vishnu@pool-72-83-63-112.washdc.east.verizon.net] has joined ##hplusroadmap
18:57 < nmz787_i> .title http://www.google.com/patents/US8598527
18:57 < yoleaux> Patent US8598527 - Scanning transmission electron microscopy - Google Patents
18:57 < nmz787_i> they immediately start to talk about DNA sequencing
19:03 < nmz787_i> "Larger acceptance angle may improve resolution. Because of this relationship between the acceptance angle and resolution of STEM 10, the acceptance angle can be selected based on the desired resolution. For example, if 1 Ångström resolution at 100 kilovolts is desired, it may be desirable to have at least about 30 milliradians acceptance half-angle, or even at least about 40 milliradians acceptance half-angle. In one example, single-atom
19:03 < nmz787_i> resolution—namely resolution at least as good as about 0.3 nanometers and in some instances at least as good as about 0.15 nanometers—may be desirable for a DNA-sequencing application. However, with an angular range that is unnecessarily high, current may be wasted undesirably. Once a suitable accelerating voltage is chosen, the desired resolution may determine the acceptance angle of objective lens100."
19:04 < nmz787_i> Owner name: MOCHII, INC. (D/B/A VOXA), WASHINGTON
19:05 < nmz787_i> hrmm, i can't find any further info... I almost want to go back tomorrow
19:07 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap
19:12 < nmz787_i> .wik precession electrondiffraction
19:12 < yoleaux> nmz787_i: Sorry, that command (.wik) crashed.
19:12 < nmz787_i> .wik precession electron diffraction
19:12 < yoleaux> "Precession Electron Diffraction (PED) is a specialized method to collect electron diffraction patterns in a transmission electron microscope (TEM). By rotating (precessing) a tilted incident electron beam around the central axis of the microscope, a PED pattern is formed by integration over a collection of diffraction conditions." — https://en.wikipedia.org/wiki/Precession_Electron_Diffraction
19:20 -!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 260 seconds]
19:25 < nmz787_i> .title https://www.youtube.com/watch?v=MoUPJSaQdDc
19:25 < yoleaux> SemGlu - a vacuum compatible, electron-cured adhesive. - YouTube
19:26 < nmz787_i> now that is a cool video
19:27 -!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:d826:9c85:3bd9:90fb] has joined ##hplusroadmap
19:32 < kanzure> fenn: i don't think you were really that surprised that steve wants to get ai first
19:42 < kanzure> is that really surprising
20:09 -!- Madplatypus [uid19957@gateway/web/irccloud.com/x-apjrgwgiyqtickng] has quit [Quit: Connection closed for inactivity]
20:11 < nmz787> here are some papers http://diyhpl.us/~nmz787/M&M%202015/72437/TOC.html
20:11 < nmz787> 1227 to be exact
20:12 < c0rw1n> so much to learn, so little time ;_;
20:13 < nmz787> voxa/mochii http://diyhpl.us/~nmz787/M&M%202015/7337/0009.pdf
20:14 < c0rw1n> ... but then if they're all like 2pgs long, that list would not take a lifetime to read :3
20:14 < nmz787> "This low voltage microscope has features that bring accessible and on-demand EM imaging into fields and laboratories where EM was previously hindered by form factor, complexity, and cost. Among these features are small size and light weight (0.25m tall, light enough to carry in a suitcase); user-friendly native wireless tablet interface; multi- and distance-user capabilities (connection to unlimited client nodes), exceedingly low power ...
20:15 < nmz787> ... consumption (by virtue of lowpower magnetic-electrostatic optics), and an integrated metal evaporator for easy sample preparation (only one pump-down cycle to image). We expect the cost to own and operate a MochiiTM microscope to be a fraction of the cost of typical EM’s with similar imaging performance due to its low power consumption, simple design, and commoditized user-replaceable consumables."
20:16 < nmz787> see that one seems to actually be split over a couple at least http://diyhpl.us/~nmz787/M&M%202015/7337/0157.pdf
20:18 < nmz787> http://diyhpl.us/~nmz787/M&M%202015/7337/1073.pdf
20:18 < nmz787> "Democratizing the Micro-Scale: A Simplified, Miniaturized SEM for K-12 and Informal Student Scientists"
20:18 < nmz787> they all went to Evergreen :O
20:19 < nmz787> ?
20:19 < c0rw1n> ( huh who what where ? )
20:19 < nmz787> 3/4
20:19 < nmz787> last one
20:19 < nmz787> huh, and a HiveBio mention
20:20 < c0rw1n> so that's at least 3 papers on the Mochii™ ?
20:20 < nmz787> I think that is/was the seattle biohackerspace
20:20 < nmz787> yeah
20:26 -!- erasmus [~esb@unaffiliated/erasmus] has quit [Quit: Namaste	]
20:27 < nmz787> huh, LS Own is on hivebio.org/people/
20:33 -!- xtalmath [~xtalmath@ip-83-134-138-137.dsl.scarlet.be] has quit [Quit: Leaving.]
20:33 -!- xtalmath [~xtalmath@ip-83-134-138-137.dsl.scarlet.be] has joined ##hplusroadmap
20:34 -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-wcqfudmyaeedhswp] has joined ##hplusroadmap
20:38 < xtalmath> how much does that mochii cost?
20:38 < nmz787> supposedly $48k
20:38 < xtalmath> I wonder how it works without vacuum?
20:39 < nmz787> it has 2 vac pump
20:39 < nmz787> 1 turbo inside, and 1 rough outside
20:39 < xtalmath> ok
20:40 < xtalmath> how much vacuum is needed for SEM in fact? what level of vacuum did the first SEM's have?
20:40 -!- proofoflogic [uid65184@gateway/web/irccloud.com/x-minuiqogrlhjpeih] has joined ##hplusroadmap
20:41 < nmz787> they used higher accelerating voltages in the early days before lens design got better i guess
20:41 < nmz787> but they had oil diffusion back then, which is just hot and oil can deposit into the sample chamber after a lot of use
20:53 -!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:d826:9c85:3bd9:90fb] has quit [Quit: Leaving]
20:55 -!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap
21:03 < xtalmath> http://www.directorypatent.com/GB/511204-a.html
21:06 < xtalmath> I wonder if something like kelvin water dropper + millikan experiment like combination could be used to build a "electron droplet" microscope in atmosphere... ( https://en.wikipedia.org/wiki/Kelvin_water_dropper , it amplifies the voltage difference, until a saturation point where the droplets bounce upwards and seek out the droplets of the other side...
21:07 < xtalmath> there is variations with a recirculating pump
21:19 < xtalmath> might be fun to make a slow droplet accelerator, and lets the droplets collide?
21:57 < xtalmath> how does an electron microscope work on insulating materials? http://www.toyo.co.jp/file/pdf/spm/files/agi/5991-0736EN.pdf
21:58 < xtalmath> is that like pure reflection mode?
22:05 -!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap
22:15 -!- rigel_ [~yourmom@c-24-21-55-24.hsd1.or.comcast.net] has quit [Read error: Connection reset by peer]
22:15 -!- rigel_ [~yourmom@c-24-21-55-24.hsd1.or.comcast.net] has joined ##hplusroadmap
22:29 -!- c0rw1n is now known as c0rw|zZz
22:34 -!- wrldpc1 [~ben@235.106.1.110.ap.yournet.ne.jp] has quit [Quit: wrldpc1]
22:42 -!- erasmus [~esb@unaffiliated/erasmus] has quit [Quit: Namaste]
22:45 -!- yashgaroth [~ffffff@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has quit [Quit: Leaving]
22:55 -!- BlueLobster [medea@persephone.darkness-reigns.net] has quit [Changing host]
22:55 -!- BlueLobster [medea@unaffiliated/fellini] has joined ##hplusroadmap
22:59 -!- nmz787_i [~ntmccork@192.55.55.41] has quit [Quit: Leaving.]
23:06 -!- dustinm [~dustinm@105.ip-167-114-152.net] has quit [Ping timeout: 240 seconds]
23:09 -!- dustinm [~dustinm@2607:5300:100:200::160d] has joined ##hplusroadmap
23:14 -!- Madplatypus [uid19957@gateway/web/irccloud.com/x-cgsvslmpgurdclnx] has joined ##hplusroadmap
23:15 -!- delinquentme [~delinquen@74.61.157.78] has quit [Ping timeout: 265 seconds]
23:30 -!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has quit [Read error: Connection reset by peer]
23:31 -!- superkuh [~superkuh@unaffiliated/superkuh] has quit [Remote host closed the connection]
23:32 -!- superkuh [~superkuh@unaffiliated/superkuh] has joined ##hplusroadmap
23:33 -!- superkuh [~superkuh@unaffiliated/superkuh] has quit [Remote host closed the connection]
23:34 -!- wrldpc1 [~ben@161.131.169.216.client.dyn.strong-sf94.as22781.net] has joined ##hplusroadmap
23:36 -!- superkuh [~superkuh@unaffiliated/superkuh] has joined ##hplusroadmap
23:44 -!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 246 seconds]
23:53 -!- dustinm [~dustinm@2607:5300:100:200::160d] has quit [Ping timeout: 252 seconds]
23:55 -!- dustinm [~dustinm@2607:5300:100:200::160d] has joined ##hplusroadmap
--- Log closed Tue Aug 04 00:00:30 2015