| --- Log opened Sat May 30 00:00:26 2015 | ||
| -!- Beatzebub [~beatzebub@d75-155-236-222.bchsia.telus.net] has quit [Read error: Connection reset by peer] | 00:12 | |
| -!- Beatzebub [~beatzebub@d75-155-236-222.bchsia.telus.net] has joined ##hplusroadmap | 00:12 | |
| -!- drewbot [~cinch@ec2-54-146-0-110.compute-1.amazonaws.com] has quit [Remote host closed the connection] | 00:41 | |
| -!- drewbot [~cinch@ec2-54-159-44-236.compute-1.amazonaws.com] has joined ##hplusroadmap | 00:41 | |
| JayDugger | Rain. | 00:42 |
|---|---|---|
| -!- Beatzebub [~beatzebub@d75-155-236-222.bchsia.telus.net] has quit [Quit: Beatzebub] | 00:54 | |
| -!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap | 01:22 | |
| -!- jdolan [~jdolan@c-50-188-243-111.hsd1.fl.comcast.net] has joined ##hplusroadmap | 01:56 | |
| -!- jdolan_ [~jdolan@c-50-188-243-111.hsd1.fl.comcast.net] has quit [Read error: Connection reset by peer] | 01:56 | |
| -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 02:10 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 276 seconds] | 02:12 | |
| -!- Viper168_ is now known as Viper168 | 02:13 | |
| -!- FourFire [FourFire@2a02:270:2015:cafe:8429:dedb:4489:979c] has joined ##hplusroadmap | 02:49 | |
| -!- wrldpc [~ben@baicb85d74e.bai.ne.jp] has joined ##hplusroadmap | 03:02 | |
| -!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Quit: austerity chic brand destruction] | 03:09 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 03:16 | |
| -!- Urchin[emacs] [~user@unaffiliated/urchin] has joined ##hplusroadmap | 03:18 | |
| Urchin[emacs] | hi | 03:19 |
| FourFire | hello Urchin[emacs] | 03:19 |
| FourFire | Do you know much about compiler optimization for modern instructionsets Urchin[emacs] ? | 03:19 |
| FourFire | I found this: http://www.agner.org/optimize/blog/read.php?i=49#209 and wondered whether the open source option was still that bad | 03:20 |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-rlrmscvobaldfgzs] has joined ##hplusroadmap | 03:27 | |
| Urchin[emacs] | no, didn't know much about the subject | 03:28 |
| TMA | FourFire: isn't that a library problem? | 03:34 |
| FourFire | TMA I'm not sure, I am not very versed in the intricate mechanics of software | 03:41 |
| Urchin[emacs] | TMA: it pretty much says it is | 03:42 |
| TMA | FourFire: In my understanding the "benchmark" you have linked (compute cosine of eight single precision floats in a vector) can be written in different ways | 03:42 |
| TMA | variant (1) standard ISO C: | 03:43 |
| TMA | float result[8], data[8]; for (int i = 0; i < 8; ++i) result[i] = fcos(data[i]); | 03:44 |
| TMA | variant (2) elements per vector = 4 | 03:45 |
| TMA | float result[8], data[8]; for (int i = 0; i < 8; i+=4) cos_4_at_time(result+i,data+i); | 03:46 |
| TMA | variant (3) elements per vector = 8 | 03:46 |
| TMA | float result[8], data[8]; cos_8_at_time(result+i,data+i); | 03:47 |
| TMA | FourFire: a compiler optimization is concerned exclusively with transforming variant (1) into one of the other variants automatically | 03:48 |
| TMA | FourFire: the library should make the computation (fcos, cos_4_at_time, cos_8_at_time functions) as fast as possible | 03:49 |
| TMA | the problem described is that all the libraries fail to deliver the speed in their implementations (for different reasons -- intel chooses slow version for CPUs that do not report GenuineIntel -- it dispatches on the vendor, not on the capabilities) | 03:52 |
| TMA | I gather that glibc does not have the SIMD routines, it does not use them anyway | 03:53 |
| TMA | FourFire: the library implementation can be made faster by anyone who knows the processors involved; that is tricky in itself | 03:57 |
| FourFire | TMA why is this? | 03:57 |
| TMA | however, if the proble is with the compiler (i.e. the benchmark was written in the style of variant (1) and it was _not_ transformed), then the person needs to be intimate with the compiler and optimizations which is another skillset entirely | 03:59 |
| TMA | the compiler author takes the library routine for granted and rewrites the loop into one or two calls to the routine | 04:00 |
| TMA | [...granted and makes the compiler rewrite the loop...] | 04:01 |
| -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 04:01 | |
| TMA | FourFire: I cannot tell. I have not seen the benchmark source, I have not examined glibc source. | 04:02 |
| FourFire | Thanks for clarifying | 04:02 |
| FourFire | I'm interested in HPC and a possible 19x optimisation is very desirable to me | 04:03 |
| TMA | All I am saying is a somewhat educated guess. I have written C++ compiler (as a school project with several colleagues), I have some basic understanding of assembly language. | 04:03 |
| TMA | FourFire: could you please expand the HPC acronym for me? | 04:04 |
| TMA | Hewlett-Packard Compiler? probably not | 04:04 |
| FourFire | High Powered Computing | 04:06 |
| FourFire | http://en.wikipedia.org/w/index.php?title=High-performance_computing&redirect=no | 04:06 |
| FourFire | performance* | 04:06 |
| TMA | oh, sorry, it is obvious in hindsight :) | 04:07 |
| FourFire | TMA I was geared up about a theoretical configuration for my theoretical workload which would result in a reduction of runtime by about 74% | 04:12 |
| FourFire | 19x reduction in runtime, even for a function which is only used 1% of the time, is crazy | 04:12 |
| TMA | FourFire: my recomendation is not to worry prematurely. You can gain much more by (1) solving the right problem (i.e. no needles computation) (2) solving the problem right (i.e. no slow algorithms) (3) watching the memory accesses (a cache miss might slow you down thousand times more than avoiding it) (4) optimizing just the hotspots | 04:14 |
| FourFire | yeah I don't know much about memory contraints, both capacity and bottlenecks, or caching, yet. | 04:15 |
| TMA | it is only 0.94 % faster overall even if you really gain 19x speedup | 04:15 |
| TMA | nowadays (2) and (3) are intertwined | 04:17 |
| TMA | an optimal algorithm can be completely wrecked by cache misses | 04:18 |
| FourFire | yes, though I suppose if such operations are poorly optimized then surely other operations (in the open source compiler) are more probably also unoptimized compared to proprietary compilers | 04:19 |
| TMA | (1) to (3) are almost completely independent of the compiler (1) and (2) completely, (3) can be influenced by the compiler misbehaving | 04:21 |
| -!- wrldpc [~ben@baicb85d74e.bai.ne.jp] has quit [Quit: wrldpc] | 04:22 | |
| TMA | FourFire: what are you trying to accomplish with HPC? | 04:22 |
| -!- wrldpc [~ben@baicb85d74e.bai.ne.jp] has joined ##hplusroadmap | 04:42 | |
| FourFire | Molecular Dynamics simulation | 04:47 |
| FourFire | TMA^ | 04:47 |
| FourFire | I'm interested in running a virtual gene through a genetic algorithm where the fitness function measures how the protein derived from the gene performs in a simulated cellular environment | 04:48 |
| Urchin[emacs] | sounds interesting | 04:49 |
| FourFire | I don't yet know what will be required, but I estimate that each "generation" will consist of around 1Mil simulations each | 04:50 |
| FourFire | so, HPC | 04:50 |
| FourFire | Urchin[emacs], I want to direct evolution in order to evolve synthetic genes which can be useful for that end goal thing that is sometimes talked about in that other channel | 04:50 |
| -!- rk[1] [~rak@opensource.cse.ohio-state.edu] has quit [Ping timeout: 265 seconds] | 04:52 | |
| Urchin[emacs] | that's the goal of this channel | 04:53 |
| Urchin[emacs] | this channel is very biotech oriented | 04:53 |
| -!- fredox [~chatzilla@c27-253-11-127.brodm4.vic.optusnet.com.au] has quit [Read error: Connection reset by peer] | 05:05 | |
| TMA | FourFire: it definitely will need a supercomputer, even just folding the protein will | 05:07 |
| FourFire | TMA I am aware | 05:10 |
| FourFire | I fully expect that the computing hardware for this kind of thing will not exist for at least another 20 years | 05:10 |
| -!- fredox [~chatzilla@c27-253-11-127.brodm4.vic.optusnet.com.au] has joined ##hplusroadmap | 05:11 | |
| maaku | FourFire: it exists now, if you have the money | 05:12 |
| TMA | if the memristor memory and related technologies will materialize it might be a huge leap forward. the cache misses (3) might become irrelevant as the memory/disk distinction | 05:22 |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 05:26 | |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Ping timeout: 265 seconds] | 05:38 | |
| Urchin[emacs] | FourFire: have you considered colaborating with one of the protein folding distributed projects? | 05:51 |
| Urchin[emacs] | like folding@home | 05:51 |
| Urchin[emacs] | or that other one I forgot what it was called | 05:51 |
| Urchin[emacs] | also, do you need the simulation of actual folding | 05:51 |
| FourFire | Urchin[emacs], actually I only really *need* the basepair sequence of the gene I want | 05:54 |
| FourFire | but how to get to it | 05:54 |
| FourFire | I could spend the rest of my life trying to engineer the protein from scratch, I could do mass IRL testing with single celled organisms | 05:54 |
| FourFire | but if I can get this method I mentioned first to work, then any scientist could use it to obtain custom gene sequences | 05:55 |
| FourFire | any bioinformatician at least | 05:56 |
| -!- jdolan_ [~jdolan@c-50-188-243-111.hsd1.fl.comcast.net] has joined ##hplusroadmap | 05:57 | |
| -!- jdolan [~jdolan@c-50-188-243-111.hsd1.fl.comcast.net] has quit [Read error: Connection reset by peer] | 05:57 | |
| FourFire | Urchin[emacs], I could, but I predict that I will require something on the order of tens of Zettaflops in processing power in order to run the entire evolutionary algorithm until I get a functional gene | 06:05 |
| FourFire | but perhaps my proof of concept test with the human tertiary structure of histone could be done with computing infrastructure existing this decade | 06:06 |
| FourFire | or perhaps between now and then I find a series of individually trivial optimizations which reduce my computational requirements by 4-6 orders of magnitude | 06:07 |
| Urchin[emacs] | it's really not my field, all I know about it is from what I've picked up from seminars on biophysics when I was at a very biophysics heavy uni | 06:08 |
| FourFire | yeah | 06:08 |
| FourFire | I'm not in university yet | 06:09 |
| FourFire | next semester if they accept me at UiO | 06:09 |
| Urchin[emacs] | UiO? | 06:12 |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 06:54 | |
| kanzure | 05:54 < FourFire> Urchin[emacs], actually I only really *need* the basepair sequence of the gene I want | 06:58 |
| kanzure | 05:54 < FourFire> but how to get to it | 06:58 |
| kanzure | uh, you look it up on ncbi and entrez | 06:58 |
| Urchin[emacs] | kanzure: I think he wants to backwards engineer it from an appropriate protein upon which genetic algorithm operates | 06:59 |
| fredox | thats kind of like saying 'justgoogle it' | 07:00 |
| kanzure | fredox: blast | 07:02 |
| fredox | tldr hes talking about from scratch engineering (which we've all dreamed of for decades) | 07:02 |
| kanzure | there's a limited subset of amino acids that are easier to work with and predict things for | 07:02 |
| fredox | but still processing power for such is basically non-existant. i have done some work on custom chips for the problem, but to actually get it produced is $$$$ | 07:04 |
| fredox | which i dont have | 07:04 |
| kanzure | er, use an fpga | 07:04 |
| fredox | nah, need 1000's x their power | 07:06 |
| kanzure | fredox: what was your implementation | 07:06 |
| -!- poppingtonic [~poppingto@unaffiliated/poppingtonic] has joined ##hplusroadmap | 07:13 | |
| poppingtonic | beep boop | 07:13 |
| FourFire | kanzure, I'm almost certain the gene I need doesn't exist in nature | 07:14 |
| fredox | kanzure: its basically a cheaper way to make to supercomputers | 07:14 |
| kanzure | fredox: that's not what i asked >::( | 07:15 |
| FourFire | fredox, neat, what do you mean by 1000x FPGA power? | 07:16 |
| FourFire | lots of gates? | 07:16 |
| @fenn | a protein-based computer substrate? | 07:16 |
| fredox | FourFire: yeah, it needs to be massively parallel | 07:17 |
| fredox | mostly regular chip production tech | 07:18 |
| kanzure | i like how you avoid my questions | 07:19 |
| kanzure | this is so boring | 07:19 |
| kanzure | was it just an implementation of ab initio stuff? | 07:20 |
| -!- soylentbomb [~k@unaffiliated/soylentbomb] has joined ##hplusroadmap | 07:20 | |
| fredox | oh, no talking about whole molecular simulation of protien folding | 07:21 |
| kanzure | why no talking? | 07:22 |
| kanzure | argh | 07:22 |
| fredox | from an 'attack this substrate' point of view | 07:22 |
| fredox | still a long way off | 07:22 |
| fredox | like lay the groundwork now and maybe we'll have it in 10-15 years | 07:23 |
| fredox | so its somewhat moot, unless you know someone willing to spring $10M+ for a chip production facility... | 07:26 |
| kanzure | i'm bored by broad generalizations and you should switch to actual details | 07:27 |
| kanzure | seriously though- don't waste people's time, yo | 07:28 |
| -!- wrldpc [~ben@baicb85d74e.bai.ne.jp] has quit [Quit: wrldpc] | 07:28 | |
| @fenn | kanzure> was it just an implementation of ab initio stuff? <- yes, i think | 07:29 |
| @fenn | anyway i doubt fourfire even knows that much | 07:29 |
| @fenn | it's good to see young people passionate about their interests :P | 07:29 |
| fredox | in other words, you don't want to do any of the work, you just want other people to give you the answers | 07:31 |
| @fenn | engineered protein design is easier than arbitrary folding simulations | 07:31 |
| @fenn | computationally | 07:31 |
| FourFire | fenn, I certainly have much to learn, indeed I'm quite daunted by the sheer volume of my ignorance | 07:31 |
| @fenn | you can constrain the amino acid sequence to known structural motifs | 07:32 |
| FourFire | sure, but that can be implemented into the genetic algorithm/ simulation script | 07:33 |
| @fenn | unfortunately, not many people are doing rational protein design, they just throw up their hands and say "oh noes we'll never solve the folding problem" | 07:33 |
| @fenn | i doubt it even requires a computer | 07:34 |
| FourFire | fenn sometime between now and 2035 I may well learn enough about proteins and how they work to take a crack at designing some things, at least the basic structural features and then it's just brute forcing which exact basepairs go in the gaps | 07:34 |
| FourFire | that would massively reduce the search space and might even be doable by current day F@H | 07:35 |
| -!- wrldpc [~ben@baicb85d74e.bai.ne.jp] has joined ##hplusroadmap | 07:35 | |
| @fenn | yep | 07:35 |
| FourFire | s/basepairs/aminoacids | 07:35 |
| @fenn | you could even try your sequences out in reality - cryoEM is getting cheap enough to be accessible by students | 07:36 |
| kanzure | fredox: it is morally wrong of you to claim that i don't want to do work. you're an idiot of the worst kind; i specifically asked you for details and instead i get shit about "wait 10 years". how the fuck do you expect me to make contributions to software and hardware projects if you don't inform me of them? | 07:37 |
| FourFire | I'm just not betting on that happening, I'm preparing for the worst case scenario where this is no shortcut and it just has to be done the hard way (I might also just be too stupid to design proteins, but I might meet someone significantly more intelligent than me, interested in the same problem and get them to design whatever) | 07:37 |
| FourFire | that way in the worst case, it still might be possible to solve this problem after 20-25 years | 07:37 |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Remote host closed the connection] | 07:37 | |
| -!- eudoxia [~eudoxia@r167-57-33-149.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 07:38 | |
| kanzure | fredox: for future reference, something like "hur dur i was using CHARMM" is the correct response (or something along those lines) | 07:38 |
| FourFire | kanzure, insulting people you want to get reliable information out of isn't a sustainable strategy | 07:39 |
| kanzure | i already gave him a chance | 07:39 |
| FourFire | but I understand your frustration | 07:39 |
| kanzure | i insulted him after he turned out to be a moron | 07:39 |
| kanzure | not before. | 07:39 |
| @fenn | fourfire i think you are underestimating the current rate of growth of computational power per dollar | 07:39 |
| FourFire | fenn I'm not sure that I'm underestimating my ability to learn GPU programming | 07:39 |
| kanzure | 07:32 <@fenn> you can constrain the amino acid sequence to known structural motifs | 07:40 |
| kanzure | i think that is a good plan in general | 07:40 |
| @fenn | it doesn't take 25 years to larn ye a CUDA | 07:40 |
| FourFire | but I could just bribe someone else to do the coding for me, except that would be really unwieldy because they won't be inside my head and know what I mean by what I want | 07:40 |
| kanzure | i'm a little concerned about the lack of progress in the literature about (restricted) amino acid structures tho | 07:40 |
| kanzure | or rather, the lack of anyone attempting or reporting that | 07:40 |
| @fenn | kanzure: yeah well there are like 4 people actually doing anything | 07:41 |
| kanzure | oh yeah | 07:41 |
| @fenn | everyone else is off tilting at the arbitrary sequence windmill | 07:41 |
| FourFire | fenn, well it's taken me about eight months from learning about gromacs to actually learn how to and run a simulation, and I still haven't managed that yet | 07:41 |
| kanzure | here's an interesting project once sequencing gets cheaper, | 07:41 |
| FourFire | still I have a box with a low end CUDA 5.0 card in it | 07:42 |
| kanzure | you could have a batch of bacteria and then you look at structural motifs, then select for genomes where they use more stereotypical motifs and less amino acids (or something) | 07:42 |
| kanzure | also allowing for unnatural amino acids in this scenario is fine | 07:42 |
| kanzure | as long as the number of variations of components is reduced overall | 07:42 |
| FourFire | kanzure, you mean IRL bruteforcing? | 07:42 |
| kanzure | bacteria are extremely good at bruteforcing | 07:42 |
| kanzure | it's like sending out billions of minions into battle | 07:43 |
| kanzure | well wait, i mean they don't bruteforce on their own. you have to apply the selective pressure. | 07:43 |
| @fenn | hmm that's not what i meant at all | 07:43 |
| FourFire | I'm uncertain, because I don't know all the legal issues I'd have to overcome | 07:43 |
| kanzure | which one did you mean? | 07:43 |
| kanzure | FourFire: legal issues....? | 07:43 |
| FourFire | my plan is to overcome them by avoiding them entirely by only running sims in silico, at first | 07:44 |
| FourFire | GMO blah blh | 07:44 |
| kanzure | .... | 07:44 |
| FourFire | I live in Europe | 07:44 |
| kanzure | look, if we already knew what to simulate, there wouldn't be a problem left | 07:44 |
| FourFire | and one of the more conservative european countries to boot | 07:44 |
| FourFire | uhh | 07:44 |
| FourFire | I know what to simulate | 07:44 |
| FourFire | it's just going to require a fucking 50 ZettaFlop supercomputer | 07:44 |
| TMA | FourFire: sooner or later you have to delegate, even though the person delegated to is not in posession of a perfect knowledge of your idea | 07:45 |
| @fenn | by "restrict the sequence to known motifs" i meant only using gly-gly-ala-ala-gly-gly-ala-ala or whatever to build your alpha helices, then this part of the protein is computationally fixed and has a high melting point and you just basically ignore it for the rest of the evolutionary search because it is an indivisible unit | 07:45 |
| FourFire | TMA ok well I'd like to have made a proof of concept work at least first | 07:45 |
| kanzure | TMA: he's no where near the delegation stage. learning CUDA is important, don't discourage him. | 07:45 |
| FourFire | that way the thing I delegate to can't fail in the most obvious ways | 07:45 |
| kanzure | fenn: and then what was the thing i said? | 07:45 |
| FourFire | I hope to end up having the proof of concept ready sometime in the next five years | 07:46 |
| @fenn | kanzure: and you said something like, "make bacteria do the search for you" but they don't know how to self-mutate only in specific places | 07:46 |
| kanzure | uh, a cuda implementation of protein simulations is something i would expect yo uto have done over a weekend | 07:46 |
| kanzure | not 5 years | 07:46 |
| kanzure | fenn: oh, well i wsa thinking mutate everywhere in large quantities. | 07:46 |
| FourFire | if I get accepted to UiO this semester I can do it along with my Masters Degree in Cellular Biology & Molecular Chemistry | 07:47 |
| @fenn | a weekend is too optimistic, but a year is not | 07:47 |
| kanzure | there are already cuda implementations of similar things or the exact same thing, so that's why i think a weekend would be a good idea | 07:47 |
| FourFire | kanzure, well like I said, it's taken me 8 months so far, and I haven't even run a single sim properly | 07:47 |
| kanzure | http://www.nvidia.com/content/cuda/spotlights/gpu-accelerated-molecular-dynamics.html | 07:47 |
| kanzure | FourFire: i haven't seen you ask any questions related to this, at all | 07:47 |
| kanzure | here's one, run this http://lammps.sandia.gov/ | 07:47 |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 07:48 | |
| FourFire | kanzure, I didn't ask in here while you were online I think, I've been asking #bioinformatics mostly | 07:48 |
| FourFire | cool, I'll try it tonight | 07:48 |
| kanzure | here are some pre-made simulation examples, | 07:49 |
| kanzure | http://git.lammps.org/git/?p=lammps-ro.git;a=tree;f=examples;h=41ea99eb9813f77708102401db2cb9cea4432070;hb=HEAD | 07:49 |
| kanzure | here's some configuration for a peptide simulation, | 07:49 |
| kanzure | http://git.lammps.org/git/?p=lammps-ro.git;a=tree;f=examples/peptide;h=3d0863ae4d37224ae6f65e3ce5611ba2d9616199;hb=HEAD | 07:49 |
| FourFire | kanzure, is lammps superior to gromacs? | 07:49 |
| FourFire | because gromacs is what I've already installed | 07:49 |
| kanzure | it doesn't matter, just use at least one of the two | 07:50 |
| kanzure | gromacs is more popular last i checked | 07:50 |
| FourFire | I've read somewhere it was the most optimized, but that's probably out of date | 07:50 |
| kanzure | http://repo.or.cz/w/gromacs.git/tree/HEAD:/src/gromacs | 07:51 |
| kanzure | let's see... | 07:51 |
| FourFire | I'm also interested in running sims entirely on the GPU, gromacs 4.6 could do this, but something about it only supporting a certain level of detail or something | 07:51 |
| @fenn | heh i don't even know what half of those lammps example simulations are intended to do | 07:51 |
| kanzure | their file formats suck | 07:52 |
| kanzure | the hell is this bullshit http://git.lammps.org/git/?p=lammps-ro.git;a=blob_plain;f=examples/friction/in.friction;hb=HEAD | 07:52 |
| @fenn | .g hugoniostat | 07:52 |
| yoleaux | http://link.aps.org/doi/10.1103/PhysRevB.70.014103 | 07:52 |
| fredox | kanzure: dude, if the intellectual level of your contribution is of the level 'just use an FPGA' it is you who are the ignorant condescending fuck who thinks you are 10x as intelligent as you are. good luck with your delusions of immortality. | 07:53 |
| -!- fredox [~chatzilla@c27-253-11-127.brodm4.vic.optusnet.com.au] has left ##hplusroadmap [] | 07:53 | |
| -!- fredox [~chatzilla@c27-253-11-127.brodm4.vic.optusnet.com.au] has joined ##hplusroadmap | 07:54 | |
| @fenn | fredox: why do you think it costs $10M to make an ASIC? | 07:55 |
| -!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has quit [Ping timeout: 240 seconds] | 07:56 | |
| -!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has joined ##hplusroadmap | 07:56 | |
| -!- fredox [~chatzilla@c27-253-11-127.brodm4.vic.optusnet.com.au] has left ##hplusroadmap [] | 07:56 | |
| @fenn | meh | 07:56 |
| @fenn | "The Bitfountain company said: in China, the cost is ~150k USD for 130nm, and ~500k USD for 65nm. ... cost at the TSMC foundry were 200-300k USD for their 110nm ASIC." | 07:57 |
| @fenn | of course that's just for one chip, so you better get it right the first time | 07:58 |
| @fenn | i'm sure hiring the relevant expertise to get it right isn't cheap, but not $9.5M | 07:59 |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Ping timeout: 245 seconds] | 07:59 | |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 08:02 | |
| -!- augur [~augur@c-73-46-94-9.hsd1.fl.comcast.net] has joined ##hplusroadmap | 08:11 | |
| @fenn | http://lammps.sandia.gov/images/polyethylene.jpg | 08:11 |
| @fenn | i've been reading a 1969 book about manufacturing materials and how alloys are designed and used. it's surprising how much is in common between alloys of different base metals | 08:12 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 256 seconds] | 08:13 | |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Ping timeout: 256 seconds] | 08:19 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 08:21 | |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 08:22 | |
| kanzure | .to fredox believe it or not, an fpga is perfectly compatible with that idea. it would just be slow. | 08:32 |
| yoleaux | kanzure: I'll pass your message to fredox. | 08:32 |
| kanzure | i'm not very sure what that polyethylene image is showing | 08:33 |
| kanzure | why is there a cube? | 08:33 |
| kanzure | and why are there things not inside the cube? | 08:33 |
| kanzure | i wonder if i should tell fredox he's wrong about me thinking i'm intelligent. that's like the exact opposite of what i suspect. how could he misinterpret that so hugely? | 08:34 |
| kanzure | "Geometry of the structure of viruses" http://www.liga.ens.fr/~deza/Sem-FullCCirmVirusSpFull/VirusConf.pdf | 08:46 |
| -!- wrldpc [~ben@baicb85d74e.bai.ne.jp] has quit [Quit: wrldpc] | 08:49 | |
| * heath looks into http://www.shippable.com/ | 08:58 | |
| heath | .title https://www.youtube.com/watch?v=Y_OhxogNCtI | 08:58 |
| yoleaux | Gaming AWS with Docker : 70% Cost Reduction With 2x More Scale - YouTube | 08:58 |
| @fenn | calling someone a moron tends to have that effect | 09:00 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 250 seconds] | 09:01 | |
| heath | i like kanzure's last link better than mine | 09:02 |
| @fenn | http://lammps.sandia.gov/movies/brittle.avi simulation of shear fault in a brittle solid | 09:02 |
| @fenn | needs more frames | 09:03 |
| kanzure | morons can't call each other out on it? | 09:06 |
| kanzure | neat video | 09:06 |
| -!- poppingtonic [~poppingto@unaffiliated/poppingtonic] has quit [Read error: Connection reset by peer] | 09:07 | |
| -!- FourFire [FourFire@2a02:270:2015:cafe:8429:dedb:4489:979c] has quit [Ping timeout: 276 seconds] | 09:07 | |
| -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 09:14 | |
| -!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 246 seconds] | 09:22 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-rlrmscvobaldfgzs] has quit [Quit: Connection closed for inactivity] | 09:23 | |
| -!- FourFire [~FourFire@host-37-191-195-228.lynet.no] has joined ##hplusroadmap | 09:29 | |
| -!- erasmus [~esb@unaffiliated/erasmus] has quit [Quit: Namaste] | 09:44 | |
| -!- jdolan [~jdolan@c-50-188-243-111.hsd1.fl.comcast.net] has joined ##hplusroadmap | 09:56 | |
| -!- jdolan_ [~jdolan@c-50-188-243-111.hsd1.fl.comcast.net] has quit [Read error: Connection reset by peer] | 09:56 | |
| -!- wrldpc [~ben@baicb85d74e.bai.ne.jp] has joined ##hplusroadmap | 09:59 | |
| -!- wrldpc [~ben@baicb85d74e.bai.ne.jp] has quit [Ping timeout: 240 seconds] | 10:18 | |
| -!- jdolan [~jdolan@c-50-188-243-111.hsd1.fl.comcast.net] has quit [Ping timeout: 246 seconds] | 10:27 | |
| -!- jdolan [~jdolan@2601:3:8200:11a7:58e6:d898:a591:bcf0] has joined ##hplusroadmap | 10:27 | |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Ping timeout: 246 seconds] | 10:50 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-julgueplspuejtfb] has joined ##hplusroadmap | 10:59 | |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 11:00 | |
| -!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has quit [Read error: Connection reset by peer] | 11:27 | |
| -!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap | 11:33 | |
| -!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 256 seconds] | 12:00 | |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Ping timeout: 246 seconds] | 12:31 | |
| -!- jdolan [~jdolan@2601:3:8200:11a7:58e6:d898:a591:bcf0] has quit [Ping timeout: 256 seconds] | 12:49 | |
| -!- jdolan [~jdolan@c-50-188-243-111.hsd1.fl.comcast.net] has joined ##hplusroadmap | 12:49 | |
| -!- FourFire [~FourFire@host-37-191-195-228.lynet.no] has quit [Ping timeout: 244 seconds] | 12:59 | |
| -!- Narp [~blargh@unaffiliated/narp] has joined ##hplusroadmap | 13:04 | |
| kanzure | Narp: hi | 13:13 |
| Narp | hey how's it goin | 13:13 |
| -!- Madplatypus [uid19957@gateway/web/irccloud.com/x-hsvaydxuvdhfyxzn] has quit [Quit: Connection closed for inactivity] | 13:23 | |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 13:33 | |
| Narp | Looking for a good future economics book. | 13:38 |
| eudoxia | kanzure: so is paperbot kill | 13:41 |
| -!- eudoxia [~eudoxia@r167-57-33-149.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving] | 13:56 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Ping timeout: 240 seconds] | 13:59 | |
| Narp | I guess I'll get this book http://www.amazon.com/Zero-Marginal-Cost-Society-Collaborative-ebook/dp/B00HY09XGQ/ | 14:00 |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 14:08 | |
| kanzure | huh, eliezer finally relented: | 14:12 |
| kanzure | "And for purposes of posing the question of the intelligence explosion, we may be better off with “Intelligence is that sort of smartish stuff coming out of brains, which can play chess, and price bonds, and persuade people to buy bonds, and invent guns, and figure out gravity by looking at wandering lights in the sky; and which, if a machine intelligence had it in large quantities, might let it invent molecular nanotechnology; and so ... | 14:12 |
| kanzure | ... on.” To frame it another way, if something is powerful enough to build a Dyson Sphere, it doesn’t really matter very much whether we call it “intelligent” or not. And this is just the sort of “intelligence” we’re interested in—something powerful enough that whether or not we define it as “intelligent” is moot. This isn’t to say that definitions are forbidden—just that further definitions would stake the ... | 14:12 |
| kanzure | ... further claim that those particular definitions were apt for carving reality at its joints, with respect to accurately predicting an intelligence explosion." | 14:12 |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Excess Flood] | 14:15 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 14:16 | |
| heath | https://www.youtube.com/watch?v=8G2tQ_0AiqY | 14:24 |
| heath | .title | 14:24 |
| yoleaux | MoMA PS1's Mushroom Tower | Hy-Fi by The Living - YouTube | 14:24 |
| heath | http://www.archdaily.com/521266/hy-fi-the-organic-mushroom-brick-tower-opens-at-moma-s-ps1-courtyard/ | 14:24 |
| kanzure | ( https://intelligence.org/files/IEM.pdf ) | 14:24 |
| kanzure | "I would start out by saying that the evolutionary history of hominid intelligence doesn’t show any signs of diminishing returns—there’s no sign that evolution took ten times as long to produce each successive marginal improvement of hominid brains. (Yes, this is hard to quantify, but even so, the anthropological record doesn’t look like it should look if there were significantly diminishing returns. See section 3.6.) We have a ... | 14:30 |
| kanzure | ... fairly good mathematical grasp on the processes of evolution and we can well approximate some of the optimization pressures involved; we can say with authority that, in a number of important senses, evolution is extremely inefficient (Yudkowsky 2007). And yet evolution was able to get significant cognitive returns on point mutations, random recombination, and non-foresightful hill climbing of genetically encoded brain architectures. ... | 14:30 |
| kanzure | ... Furthermore, the character of evolution as an optimization process was essentially constant over the course of mammalian evolution—there were no truly fundamental innovations, like the evolutionary invention of sex and sexual recombination, over the relevant timespan." | 14:30 |
| heath | http://www.studio360.org/story/the-ethics-of-synthetic-biology/ | 14:32 |
| heath | endy mentions a project of using shrooms + synbio to create mobile phones, any idea where this project is located? | 14:33 |
| -!- eudoxia [~eudoxia@r167-57-33-149.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 14:33 | |
| heath | wait, the podcast was http://www.studio360.org/story/designing-life-art-meets-synthetic-biology/ | 14:34 |
| -!- eudoxia [~eudoxia@r167-57-33-149.dialup.adsl.anteldata.net.uy] has quit [Client Quit] | 14:36 | |
| heath | this one was a little more fun than the typical media on the topic | 14:42 |
| -!- strangewarp_ [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 15:00 | |
| -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Ping timeout: 256 seconds] | 15:00 | |
| kanzure | "More broadly, another way of framing this is to ask what the world should look like if there were strongly decelerating returns to evolutionary optimization of hominids." | 15:10 |
| -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 250 seconds] | 15:15 | |
| kanzure | "The possibility of increased selection pressures should mainly lead us to suspect that there are huge cognitive gaps between humans and chimpanzees which resulted from merely linear returns on cumulative optimization—there was a lot more optimization going on, rather than small amounts of optimization yielding huge returns. But we can’t have a small cognitive gap between chimps and humans, a large amount of cumulative selection, and ... | 15:15 |
| kanzure | ... fitness returns on individual mutations strongly diminishing, because in this scenario we wouldn’t get much evolution, period. The possibility of increased rates of genetic change does not actually imply room for cognitive algorithms becoming “harder to design” or “harder to improve upon” as the base level grows more sophisticated. Returns on single positive mutations are lower-bounded by the logic of natural selection." | 15:15 |
| kanzure | "To wedge diminishing cognitive returns on evolution into this model—without contradicting basic evolutionary points about how sufficiently small fitness advantages take huge amounts of time to fixate, or more likely don’t fixate at all—we would have to suppose that small cognitive advantages were somehow providing outsize fitness advantages (in a way irrelevant to returns on cognitive reinvestment for AIs trying to improve ... | 15:18 |
| kanzure | ... themselves). To some degree, “inflated fitness advantages” occur in theories of runaway sexual selection (where everyone tries to mate with whoever seems even nominally smartest). To whatever extent such sexual selection was occurring, we should decrease our estimate of the sort of cognitively produced fitness advantage that would carry over to a machine intelligence trying to work on the protein folding problem (where you do not ... | 15:18 |
| kanzure | ... get an outsized prize for being only slightly better)." | 15:18 |
| @fenn | did you just quote-wall yudkowsky? | 15:19 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 15:19 | |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Read error: Connection reset by peer] | 15:19 | |
| @fenn | even a marginally better algorithm would be selected for in a protein folding competition | 15:21 |
| @fenn | or genetic algorithm, linear optimization, whatever | 15:21 |
| -!- FourFire [~FourFire@44-126-11.connect.netcom.no] has joined ##hplusroadmap | 15:22 | |
| @fenn | i guess i missed the point, but i still don't know what the point is | 15:23 |
| kanzure | "Natural selection did use vast amounts of computational brute force to build humans. The “naive” estimate is that natural selection searched in the range of 10^30 to 10^40 organisms before stumbling upon humans (Baum 2004). Anthropic considerations (did other planets have life but not intelligent life?) mean the real figure might be almost arbitrarily higher (see section 3.8)." | 15:32 |
| kanzure | fenn: well, he was trying to estimate or think about local minimas or maximas in cognitive-ability-related fitness landscapes | 15:33 |
| kanzure | for some reason he doesn't call them minimas or maximas or "getting stuck" which i find a little unusual (i'll attribute this to him not getting enough exposure to biology?) | 15:33 |
| -!- Adlai [~Adlai@unaffiliated/adlai] has joined ##hplusroadmap | 15:33 | |
| kanzure | anyway one of the things i like about this document is that it offers a rather simple retort about brain size enthusiasts (that the total change in brain size has only been 4x and could have changed much more easily; that hips were not a constraint because larger skulls happened in other hominid subspecies) | 15:35 |
| -!- rasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap | 15:35 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 240 seconds] | 15:37 | |
| -!- FourFire [~FourFire@44-126-11.connect.netcom.no] has quit [Ping timeout: 264 seconds] | 15:39 | |
| -!- erasmus [~esb@unaffiliated/erasmus] has quit [Ping timeout: 255 seconds] | 15:39 | |
| -!- rasmus is now known as erasmus | 15:39 | |
| @fenn | .wik encephalization quotient | 15:41 |
| yoleaux | "Encephalization Quotient (EQ), or encephalization level is a measure of relative brain size defined as the ratio between actual brain mass and predicted brain mass for an animal of a given size, which is hypothesized to be a rough estimate of the intelligence or cognition of the animal." — http://en.wikipedia.org/wiki/Encephalization_quotient | 15:41 |
| @fenn | i still think brain size is important, but there's obviously something to EQ also | 15:41 |
| @fenn | oh you said hominid. is that actually true? hominids with larger skulls than h. sapiens? | 15:42 |
| kanzure | if brain size was the primary driver then we would have much larger brains, methinks | 15:42 |
| @fenn | there are other constraints, energy being a big one | 15:43 |
| Adlai | ehhhhhh if EQ were such a driver, wouldn't we see | 15:43 |
| Adlai | oops | 15:43 |
| kanzure | whoops, it was neanderthals with larger cranial volume, not necessarily skull er.. size.. what? | 15:43 |
| * Adlai isn't certain what we'd see | 15:44 | |
| @fenn | oh nice there's a summary http://en.wikipedia.org/wiki/Brain_size | 15:44 |
| Adlai | ultimately, the body mass taken up by the appendix and the tailbone is probably not contributing too much to intelligence; but there's still a lot to be said for smarter software regardless of the hardware | 15:45 |
| @fenn | 1600cc in neanderthals is not much bigger than 1500cc (largest modern brain size) | 15:47 |
| @fenn | that was only 30k years ago | 15:47 |
| -!- jdolan [~jdolan@c-50-188-243-111.hsd1.fl.comcast.net] has quit [Remote host closed the connection] | 15:47 | |
| kanzure | "It’s not clear to me how much we should conclude from brain sizes increasing by a factor of only four—whether we can upper-bound the returns on hardware this way. As I understand it, human-sized heads lead to difficult childbirth due to difficulties of the baby’s head passing the birth canal. This is an adequate explanation for why we wouldn’t see superintelligent mutants with triple-sized heads, even if triple-sized heads could ... | 15:50 |
| kanzure | ... yield superintelligence. On the other hand, it’s not clear that human head sizes are hard up against this sort of wall—some people have above-average-sized heads without their mothers being dead. Furthermore, Neanderthals may have had larger brains than modern humans (Ponce de León et al. 2008).48 So we are probably licensed to conclude that there has not been a strong selection pressure for larger brains, as such, over very ... | 15:50 |
| kanzure | ... recent evolutionary history.49" | 15:50 |
| kanzure | note 49 is "It is plausible that the marginal fitness returns on cognition have leveled off sharply enough that improvements in cognitive efficiency have shifted the total resource cost of brains downward rather than upward over very recent history. If true, this is not the same as Homo sapiens sapiens becoming stupider or even staying the same intelligence. But it does imply that either marginal fitness returns on cognition or marginal ... | 15:51 |
| kanzure | ... cognitive returns on brain scaling have leveled off significantly compared to earlier evolutionary history." | 15:51 |
| kanzure | "We can upper bound the fitness returns on larger brains by observing that Homo sapiens are not hard up against the wall of head size and that Neanderthals may have had even larger brains. This doesn’t say how much of that bound on returns is about fitness returns on cognition versus cognitive returns on brain size." | 15:51 |
| kanzure | and... "Do variations in brain size within Homo sapiens let us conclude much about cognitive returns? Variance in brain size correlates around 0.3 with variance in measured IQ, but there are many plausible confounders such as childhood nutrition or childhood resistance to parasites. The best we can say is that John von Neumann did not seem to require a brain exponentially larger than that of an average human, or even twice as large as ... | 15:52 |
| kanzure | ... that of an average human, while displaying scientific productivity well in excess of twice that of an average human being of his era. But this presumably isn’t telling us about enormous returns from small increases in brain size; it’s much more likely telling us that other factors can produce great increases in scientific productivity without requiring large increases in brain size. We can also say that it’s not possible that a ... | 15:52 |
| kanzure | ... 25% larger brain automatically yields superintelligence, because that’s within the range of existing variance." | 15:52 |
| @fenn | saying "people have larger than average heads, so therefore larger brains aren't better" is bad logic | 15:54 |
| @fenn | also, in very recent history and many parts of the modern world, women die in significant numbers in childbirth | 15:55 |
| kanzure | i wonder if that has been increasing recently, or if any increase is hard to see because of medical improvements (like handwashing, sigh) | 15:56 |
| @fenn | increase in what? | 15:56 |
| Adlai | also, in recent history and many parts of the modern world, reproductive fitness has become rather uncorrelated from intelligence [past a certain level] [citation needed] | 15:56 |
| kanzure | childbirth deaths due to skull size | 15:56 |
| @fenn | Adlai: negatively correlated... :\ | 15:57 |
| kanzure | Adlai: arguably there is still some generality that is being selected for, even in the case of whatever you're referencing | 15:57 |
| kanzure | the genetic material to "make it" in an inner city rap gang is probably pretty similar to things that would be construed as intelligence-related | 15:58 |
| Adlai | but the top percentile of intelligence do not seem, by my observations, to have any better reproductive fitness than the next several percentiles | 15:58 |
| Adlai | so even if EQ were correlated with intelligence - it seems there's less selection pressure on intelligence, and thus on EQ | 15:59 |
| kanzure | i don't care | 15:59 |
| kanzure | why are you bringing this up? | 15:59 |
| * Adlai could /part | 16:00 | |
| kanzure | huh? | 16:00 |
| Adlai | just throwing words on the discussion. if they're not helpful, i can shut up | 16:00 |
| kanzure | my point was that pregnancy records might show a trend of increasing birthing deaths if we were to look over the past thousand years | 16:01 |
| * Adlai had no point and was basically rephrasing previously pasted quotes, to verify his own understanding | 16:01 | |
| @fenn | it might be more of a population level effect; populations of higher intelligence outcompete those of lower intelligence | 16:01 |
| @fenn | this is contradicted by the neanderthal exctinction though | 16:02 |
| kanzure | err wasn't that disputed recently | 16:02 |
| kanzure | "recently" | 16:02 |
| @fenn | assuming skull size = intelligence, that is | 16:03 |
| kanzure | wasn't there the whole "whoops neanderthals were not wiped out, my bad" thing floating around a while back | 16:03 |
| kanzure | http://en.wikipedia.org/wiki/Neanderthal_extinction#Interbreeding | 16:03 |
| kanzure | neat "Among the genes shown to differ between present-day humans and Neanderthals were RPTN, SPAG17, CAN15, TTF1 and PCD16.[31]" | 16:04 |
| @fenn | obligatory mention of "the clan of the cave bear" book series | 16:06 |
| kanzure | it's too bad the cranial volume is so close | 16:07 |
| kanzure | wikipedia says... "starting from about 600 cm3 in Homo habilis up to 1600 cm3 in Homo neanderthalensis, which was the hominid with the biggest brain size.[5] The increase in brain size topped with neanderthals. Since then, the average brain size has been shrinking over the past 28,000 years.[6] The cranial capacity has decreased from around 1,550 cm3 to around 1,440 cm3 in males while the female cranial capacity has shrunk from around ... | 16:09 |
| kanzure | ... 1,500 cm3 to around 1,240 cm3.[7] " | 16:09 |
| @fenn | i also liked "quest for fire" | 16:10 |
| kanzure | "Other sources with bigger sample sizes of modern Homo sapiens find approximately the same cranial capacity for males but a higher cranial capacity of around 1330 cm3 in females.[8]" | 16:10 |
| kanzure | "In recent years, experiments have been conducted drawing conclusions to brain size in association to the gene mutation that causes MCPH, a neural developmental disorder that affects cerebral cortical volume.[9]" | 16:10 |
| kanzure | whoops that's just microcephalin | 16:10 |
| @fenn | http://www.dailymotion.com/video/xsx5an_quest-for-fire-1981-720p-bluray-dts-x264-1_webcam | 16:10 |
| kanzure | "A number of studies have found correlation between variation in brain size in cranial capacity and geographic ancestry in humans.[10][11] This variation in cranial capacity is believed to be primarily caused by climatic adaptation that favor large round heads in colder climates because they conserve heat and slender heads in warm climates closer to the equator (See Bergmann's rule and Allen's rule).[12]" | 16:11 |
| kanzure | "The largest study done on the subject of geographic variation in brain size is the 1984 study Brain Size, Cranial Morphology, Climate, and Time Machines. The study found that human brain size varied with latitude of biogeographic ancestry.[10] The relationship between latitude and cranial size is described in the study as an example of Bergmann’s principle that crania are more spherical in cold climates because mass increases relative ... | 16:11 |
| kanzure | ... to surface area to conserve core temperatures and behaves independently of "race"." | 16:11 |
| kanzure | hmph... | 16:11 |
| @fenn | heh the subtitles are pretty unnecessary | 16:11 |
| jrayhawk | selection pressures have sortof gone haywire during the neolithic anyway, so if we're aiming for naturalistic intuitions, might be prudent to avoid that | 16:13 |
| kanzure | cool, "Adult twin studies have indicated high heritability estimates for overall brain size in adulthood (between 66% and 97%). The effect varies regionally within the brain, however, with high heritabilities of frontal lobe volumes (90-95%), moderate estimates in the hippocampi (40-69%), and environmental factors influencing several medial brain areas." | 16:13 |
| kanzure | from http://en.wikipedia.org/wiki/Brain_size | 16:14 |
| kanzure | i wonder if "environmental factors" is just the shit bin | 16:15 |
| @fenn | so and so thinks it was the dog-human alliance that beat neanderthals | 16:16 |
| kanzure | i wouldn't know how to estimate how much effort it should be to outcompete neanderthals entirely | 16:16 |
| kanzure | and by jrayhawk's law that happened in the last 40k years so i guess we can't say | 16:17 |
| -!- erasmus [~esb@unaffiliated/erasmus] has quit [Read error: Connection reset by peer] | 16:17 | |
| -!- Adlai` [~Adlai@unaffiliated/adlai] has joined ##hplusroadmap | 16:18 | |
| kanzure | hmm wasn't aware there was a wikipedia article on this, http://en.wikipedia.org/wiki/Parieto-frontal_integration_theory | 16:20 |
| -!- Adlai [~Adlai@unaffiliated/adlai] has quit [Ping timeout: 246 seconds] | 16:21 | |
| jrayhawk | http://www.slideshare.net/ancestralhealth/ahs13-will-lassek-md and yes, encephalization is apparently an enormous selection pressure | 16:22 |
| -!- Adlai` [~Adlai@unaffiliated/adlai] has quit [Quit: There will be FUD] | 16:24 | |
| kanzure | those slides really fizzled out at the end, what a disappointment | 16:34 |
| jrayhawk | Hmm, he didn't have the hibernating bear fat percentage on there. | 16:35 |
| jrayhawk | the baboon was 2%, the bear was 33%, the undergrad was 32% | 16:40 |
| kanzure | fenn: this was also interesting, http://diyhpl.us/~bryan/papers2/neuro/language/Language%20evolution%20and%20human%20development.pdf | 16:40 |
| kanzure | i suppose the same goes for jrayhawk. | 16:40 |
| @fenn | what was the point of that slideshare | 16:41 |
| @fenn | there were a bunch of contradictory things and then... no attempt at explanation | 16:41 |
| jrayhawk | first of all that we bear enormous anti-selective burdens in order to support encephalization | 16:42 |
| jrayhawk | at least, that's what i wanted to point out | 16:42 |
| kanzure | the eye tracking study was disappointing; i would want to see more evidence than just from college frat bros. | 16:42 |
| jrayhawk | his point was moreso that first births are apparently basically test runs | 16:42 |
| kanzure | well what else would they be? | 16:43 |
| @fenn | i'm not surprised that the eye fixates on the centroid of a body, but that doesn't mean it's the most attractive feature of a woman | 16:43 |
| jrayhawk | http://www.omgwallhack.org/home/jrayhawk/vid/ahs/AHS13_Will_Lassek_MD_Why_Women_Need_Fat_Three_Evolutionary_Puzzles.mp4 if you want to watch the whole thing | 16:43 |
| jrayhawk | the conclusion is wonky to me, but the data is all interesting | 16:44 |
| @fenn | there was a conclusion? | 16:44 |
| jrayhawk | i just explained the conclusion | 16:44 |
| @fenn | oh | 16:44 |
| kanzure | i'm not sure what the competing alternative that i was supposed to previously believe was? | 16:44 |
| @fenn | that first-borns are better or something | 16:44 |
| jrayhawk | that first-borns are just as precious and beautiful as subsequent babies | 16:45 |
| jrayhawk | which would be the expectation given the enormous investments in time and energy and existential risk | 16:46 |
| kanzure | that was not clear at all from any of the slides | 16:46 |
| jrayhawk | Yeah, I didn't care about that part, which is why I just linked the slides. | 16:46 |
| jrayhawk | Fenn was asking about it, though. | 16:46 |
| kanzure | would it be practical to give every girl artificial hips before her 18th birthday | 16:47 |
| * fenn mumbles something about artificial wombs | 16:47 | |
| -!- wrldpc [~ben@hcccbcad653.bai.ne.jp] has joined ##hplusroadmap | 16:47 | |
| kanzure | i'm not really sure why brain development wasn't pushed further into after-pregnancy | 16:50 |
| kanzure | er, after-birth | 16:50 |
| jrayhawk | It's pushed more than in any other mammal, apparently. | 16:50 |
| -!- wrldpc [~ben@hcccbcad653.bai.ne.jp] has quit [Quit: wrldpc] | 16:55 | |
| kanzure | some commentary from gmaxwell http://www.reddit.com/r/Bitcoin/comments/37vg8y/is_the_blockstream_company_the_reason_why_4_core/ | 16:56 |
| -!- Madplatypus [uid19957@gateway/web/irccloud.com/x-oamoxzfewojwchsk] has joined ##hplusroadmap | 16:59 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Remote host closed the connection] | 17:06 | |
| kanzure | i've been having lots of trouble finding citable evidence of the millimeters-of-brain-matter-only human person that was discovered | 17:31 |
| @fenn | well they may not actually exist | 17:32 |
| kanzure | so.. wrong universe? | 17:32 |
| @fenn | the last hydrocephaly thing i investigated it turned out her brain had a roughly normal cortical volume, it was just stretched thin because her cranial volume was larger | 17:32 |
| kanzure | that is truly disappointing | 17:34 |
| kanzure | how about, | 17:34 |
| kanzure | http://www.newscientist.com/article/dn12301-man-with-tiny-brain-shocks-doctors.html#.VWpXXlIX3RY | 17:34 |
| kanzure | which refers to, | 17:37 |
| kanzure | http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(07)61127-1/abstract | 17:37 |
| -!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap | 17:38 | |
| kanzure | oh it's just a single page | 17:38 |
| kanzure | http://www.thelancet.com/pdfs/journals/lancet/PIIS0140-6736(07)61127-1.pdf | 17:39 |
| kanzure | how did they not study this >:( | 17:39 |
| kanzure | also, what does that shape mean for routing | 17:40 |
| kanzure | "Brain death actually occurs at the time of resuscitation. As it wakes individual cells kill themselves as a result of a kind of "self check." If you cool the brain before resuscitation you can hopefully delay the self check to the point where the cells are correctly oxygenated and healthy when it happens." | 17:46 |
| erasmus | you should try it | 17:52 |
| -!- erasmus [~esb@unaffiliated/erasmus] has quit [Quit: Namaste] | 17:53 | |
| kanzure | "Ultimately, if the rules of Bitcoin can be freely rewritten by some popular push without a careful understanding of the implications, in spite of near unanimous concern from the technical experts; then perhaps the system cannot endure in the long term-- it would be too vulnerable to being usurped." | 18:03 |
| kanzure | http://www.reddit.com/r/Bitcoin/comments/37pv74/gavin_andresen_moves_ahead_with_push_for_bigger/crq4hqr | 18:10 |
| -!- yoleaux [~yoleaux@xn--ht-1ia18f.nonceword.org] has quit [Quit: made me do it!] | 18:20 | |
| -!- yoleaux [~yoleaux@xn--ht-1ia18f.nonceword.org] has joined ##hplusroadmap | 18:20 | |
| jrayhawk | why are they talking about consensus among users when the miners are what actually matter | 18:25 |
| jrayhawk | and why would anyone expect miners to mine *less* compatible blocks | 18:25 |
| jrayhawk | -core blocks will be compatible with -XT, but -XT blocks will be incompatible with -core | 18:26 |
| jrayhawk | so -XT may as well not exist among miners | 18:26 |
| jrayhawk | there's no way from the outset that this campaign will achieve anything but the sewing of discontent | 18:28 |
| -!- jdolan [~jdolan@c-50-188-243-111.hsd1.fl.comcast.net] has joined ##hplusroadmap | 18:44 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 18:44 | |
| -!- yoleaux [~yoleaux@xn--ht-1ia18f.nonceword.org] has quit [Quit: dpk made me do it!] | 18:45 | |
| -!- yoleaux [~yoleaux@xn--ht-1ia18f.nonceword.org] has joined ##hplusroadmap | 18:45 | |
| -!- jdolan [~jdolan@c-50-188-243-111.hsd1.fl.comcast.net] has quit [Remote host closed the connection] | 18:53 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-julgueplspuejtfb] has quit [Quit: Connection closed for inactivity] | 19:13 | |
| kanzure | jrayhawk: miners will mine less compatible blocks if it is in their interest to squeeze out the rest of the network from existence | 19:26 |
| kanzure | and yes, discontent (and hardforking) is the result | 19:27 |
| kanzure | jrayhawk: and also; miners can be legally compelled if they are operating as a business. governments and law enforcement may compel large miners (operating as companies) to attack the network in certain ways. | 19:27 |
| jrayhawk | oh, that's an interesting point | 20:09 |
| -!- wrldpc [~ben@hcccbcad653.bai.ne.jp] has joined ##hplusroadmap | 20:21 | |
| -!- wrldpc [~ben@hcccbcad653.bai.ne.jp] has quit [Quit: wrldpc] | 20:31 | |
| -!- wrldpc [~ben@hcccbcad653.bai.ne.jp] has joined ##hplusroadmap | 20:54 | |
| -!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 21:14 | |
| -!- TheoryCat [wye_naught@unaffiliated/wye-naught/x-8734122] has joined ##hplusroadmap | 21:21 | |
| -!- wrldpc [~ben@hcccbcad653.bai.ne.jp] has quit [Quit: wrldpc] | 21:21 | |
| -!- night [~Adifex@unaffiliated/adifex] has quit [Remote host closed the connection] | 21:27 | |
| kanzure | hi TheoryCat | 21:43 |
| TheoryCat | hi kanzure | 21:44 |
| -!- TheoryCat [wye_naught@unaffiliated/wye-naught/x-8734122] has quit [Quit: Goodbye] | 22:00 | |
| -!- wrldpc [~ben@hcccbcad653.bai.ne.jp] has joined ##hplusroadmap | 22:08 | |
| -!- erasmus [~esb@unaffiliated/erasmus] has joined ##hplusroadmap | 22:52 | |
| wrldpc | Does anybody know how to install bootstrap? do I use yum or rpm? do i need to add a repository or? | 23:50 |
| kanzure | the css theme? | 23:51 |
| kanzure | use bower or webcompoinents | 23:51 |
| kanzure | or npm + browserify | 23:51 |
| wrldpc | on a redhat or cent os? | 23:57 |
| wrldpc | got it | 23:57 |
| wrldpc | ty | 23:57 |
| --- Log closed Sun May 31 00:00:27 2015 | ||
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!
