| --- Log opened Sun Mar 22 00:00:29 2015 | ||
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 00:16 | |
| -!- DumpsterD1ver [~loki@pool-151-200-31-189.washdc.btas.verizon.net] has quit [Ping timeout: 255 seconds] | 01:10 | |
| -!- soylentbomb [~k@d53-64-122-232.nap.wideopenwest.com] has quit [Quit: leaving] | 01:15 | |
| -!- delinquentme [~delinquen@74.61.157.78] has quit [Quit: Leaving] | 02:14 | |
| archels | nmz787: sadly not at this time, due to the way in which the money is flowing here | 02:38 |
|---|---|---|
| -!- drewbot [~cinch@ec2-54-166-239-86.compute-1.amazonaws.com] has quit [Remote host closed the connection] | 03:08 | |
| -!- drewbot [~cinch@ec2-54-146-45-108.compute-1.amazonaws.com] has joined ##hplusroadmap | 03:08 | |
| -!- Shannon [~Shannon@unaffiliated/shannon] has joined ##hplusroadmap | 03:30 | |
| -!- TK_ [~TK@85.253.73.199.cable.starman.ee] has joined ##hplusroadmap | 03:34 | |
| -!- Shannon is now known as Sh-Arch | 04:21 | |
| -!- Sh-Arch is now known as Shannon | 04:22 | |
| -!- Qfwfq [~WashIrvin@unaffiliated/washirving] has quit [Ping timeout: 264 seconds] | 04:26 | |
| -!- Qfwfq [~WashIrvin@unaffiliated/washirving] has joined ##hplusroadmap | 04:33 | |
| -!- Shannon [~Shannon@unaffiliated/shannon] has quit [Quit: Leaving] | 04:33 | |
| -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 256 seconds] | 04:43 | |
| -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 05:05 | |
| -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Excess Flood] | 05:09 | |
| -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 05:12 | |
| -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Excess Flood] | 05:13 | |
| -!- Qfwfq [~WashIrvin@unaffiliated/washirving] has quit [Ping timeout: 244 seconds] | 05:19 | |
| -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 05:20 | |
| -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Excess Flood] | 05:21 | |
| -!- narwh4l [~michael@unaffiliated/thesnark] has joined ##hplusroadmap | 05:21 | |
| -!- Qfwfq [~WashIrvin@unaffiliated/washirving] has joined ##hplusroadmap | 05:26 | |
| fenn | heath: sorry yesterday i was busy at the most awesome garage sale ever, acquiring hardware. no plans for today though so i'm down for whateva | 05:31 |
| -!- nsh [~lol@wikipedia/nsh] has quit [Quit: ZNC - http://znc.in] | 05:34 | |
| -!- nsh [~lol@wikipedia/nsh] has joined ##hplusroadmap | 05:36 | |
| kanzure | boop | 05:44 |
| kanzure | strongly tempted to design gundam part with solvespace | 05:46 |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 05:49 | |
| kanzure | here is some evidence that others read papers https://news.ycombinator.com/item?id=9245467 | 05:49 |
| -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 06:09 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Quit: Leaving] | 06:13 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 06:16 | |
| -!- TK_ [~TK@85.253.73.199.cable.starman.ee] has quit [Ping timeout: 244 seconds] | 06:20 | |
| -!- zadock [~zadock@81.180.210.87] has quit [Quit: Leaving] | 06:34 | |
| -!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has joined ##hplusroadmap | 06:36 | |
| JayDugger | Which part of which gundam? | 06:45 |
| kanzure | unspecfied. why do you ask? | 06:45 |
| JayDugger | Sounded interesting but vague. | 06:50 |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 06:51 | |
| fenn | do it. | 07:02 |
| fenn | do it! | 07:02 |
| kanzure | "this is the 5th left angle bracket" | 07:04 |
| fenn | lol https://en.wikipedia.org/wiki/How_to_Read_a_Book#How_to_Read_a_Book_Video | 07:05 |
| kanzure | "For unknown reasons sometime after their original publication, these videos had been lost for many years." | 07:06 |
| fenn | the lost ancient wisdom desperately needed by the world's children | 07:06 |
| fenn | step 1: look at the cover | 07:06 |
| kanzure | step 2: ask someone else to help | 07:07 |
| fenn | step 3: post your own video of you reading a book | 07:08 |
| kanzure | that probably already exists. probably a bunch of bored parents reading children's books on youtube. | 07:09 |
| cluckj | there's a book about how to not read books | 07:09 |
| cluckj | I bought it for my mom after I said I didn't want to read some garbage book she sent to me | 07:10 |
| cluckj | http://www.amazon.com/Talk-About-Books-Havent-Read/dp/1596915439 | 07:12 |
| -!- Qfwfq [~WashIrvin@unaffiliated/washirving] has quit [Ping timeout: 256 seconds] | 07:12 | |
| kanzure | hmm so my yesterday idea for brain emulation doesn't make sense because there's no particular reason that a data center of 10k individual neural tissue cultures would add up to even dog brain ability or human brain ability. | 07:16 |
| -!- Qfwfq [~WashIrvin@unaffiliated/washirving] has joined ##hplusroadmap | 07:17 | |
| cluckj | there's a lot of useless stuff the human brain does that you don't need to have cells do in a dish | 07:18 |
| cluckj | like...breathe or poop | 07:19 |
| kanzure | 1 million tissue culture devices (each with 1k neurons per tissue culture)) at $1k per device is already $1 gigabucks | 07:32 |
| kanzure | i dunno why i picked $1k | 07:33 |
| kanzure | if you are going to make a million then you would probably get the cost per tissue culture down lower than $1 | 07:33 |
| kanzure | lower than $1k | 07:33 |
| archels | kanzure: that's the same as saying that 10k insect brains aren't going to add up to a rat brain | 07:35 |
| archels | that doesn't make studying and emulating insect brains any less valid | 07:35 |
| kanzure | hmm | 07:36 |
| kanzure | but nobody in their right mind would spend $1B on insect brain cultures? i mean that's not even an emulation yet, that's the prep work to start working on emulation :-). | 07:37 |
| archels | $1B might be a little over-the-top, but look at all the interest in OpenWorm/Si Elegans, and that beast only has 302 neurons | 07:38 |
| kanzure | another thing is that this system looks an awful lot like "a giant neural network in a simulation".. except it's physical instead.. if izhikevich's pile of neurons don't do anything, why would mine? | 07:38 |
| archels | ^ best argument against this line of study | 07:39 |
| kanzure | huh? | 07:39 |
| kanzure | that's not a good argument at all | 07:39 |
| archels | i.e. the whole "let's throw a bunch of neurons together in a culture and hook them up to sensors/effectors" | 07:39 |
| kanzure | no, you're supposed to tell me "simulations aren't anywhere near as detailed as the real thing" | 07:40 |
| archels | that's not necessarily true at all | 07:40 |
| kanzure | well they aren't. | 07:40 |
| archels | of course how deep the rabbit hole goes is a huge open question, but we can get pretty detailed as it standds | 07:41 |
| kanzure | if you are trying t oargue to me that our simulations/emulations are exactly equivalent to human brain operation, we so far have empirical evidence that you are wrong (we do not get human-brain-like results from our simulations) | 07:44 |
| archels | no but now you're lifting it up to the whole-brain level; I was talking about a single-cell level | 07:45 |
| fenn | i bet you could get the cost of a petri dish with optogenetic interface down to $1 if you make some drastic reductions in bandwidth and specificity for any particular neuron | 07:48 |
| kanzure | you don't need single-neuron interfacing.... just some inputs and just some outputs. anything in between can go fuck itslef. | 07:49 |
| archels | need for what though | 07:49 |
| kanzure | ? | 07:50 |
| fenn | light projection tech is still kinda expensive | 07:50 |
| archels | fenn: any idea---ballpark figure---what a 2-photon setup would actually cost? | 07:50 |
| archels | can't seem to find any figures via Google | 07:50 |
| fenn | huh? | 07:50 |
| fenn | $100k | 07:50 |
| kanzure | what was wrong with electrodes, again? | 07:50 |
| fenn | that's the science research markup though | 07:50 |
| archels | ouch | 07:51 |
| fenn | i'm just talking about like, a magnifying glass and dumb-phone screen :P | 07:51 |
| archels | precision (what are you stimulating? what are you recording?) mostly, also durability | 07:51 |
| archels | also this is tech from like 2 centuries ago, c'mon | 07:51 |
| kanzure | you don't care what you are stimulating (just that you are in the same location) | 07:51 |
| kanzure | and you don't care about what you are recording (as long as you are recording something) | 07:51 |
| fenn | i thought electrodes tended to kill the cells you are interested in | 07:51 |
| kanzure | a certain amount of hardware depreciation is acceptable. just have a replacement schedule. | 07:52 |
| fenn | it's on the order of a couple months | 07:52 |
| archels | fenn: don't think it's quite that bad, optogenetics also has some toxicity associated with it | 07:52 |
| fenn | yeah but blue light is worse than, say, infrared | 07:52 |
| fenn | another engineering tradeoff... | 07:53 |
| kanzure | having 1 million different cultures churn over the course of 4 months would be bad, maybe if it was like 1-10% churn/month | 07:53 |
| archels | nah I mean with the proteins itself, and with the ions that the optically stimulated channels let in/out of the cell | 07:53 |
| archels | but yeah light itself also | 07:53 |
| fenn | kanzure: pixels are super duper cheap compared to electrode array elements | 07:54 |
| kanzure | electrode array elements are like, typical mems stuff | 07:54 |
| fenn | i mean a 2 MILLION pixel camera is like $0.50 | 07:54 |
| archels | oh, middle ground is CMOS array used as electrode array | 07:54 |
| archels | not sure how far along that tech is by now | 07:54 |
| fenn | like integrated plasmonics? or just a ram chip? | 07:54 |
| archels | camera chip, more like | 07:55 |
| fenn | right | 07:55 |
| fenn | can that detect optical emissions from neurons? | 07:55 |
| archels | afaik they were studying it more in the context of measuring capacitively coupled electrical signals | 07:56 |
| kanzure | btw i don't expect the cost of culture devices/chambers to be stimulation and recording, but rather things like tissue media food flow | 07:56 |
| fenn | it seems like it should detect near-field light, but having something that close to the element makes all the angles wrong | 07:56 |
| kanzure | (and networking equipment) | 07:56 |
| fenn | food, seriously? that's the cheapest part | 07:57 |
| kanzure | food itself is cheap | 07:57 |
| archels | dunno, MEAs can be a couple hundred euros easily | 07:57 |
| archels | and if you need to throw them out every 3 months... | 07:57 |
| kanzure | these aren't just black boxes that you initialize and then lock down | 07:57 |
| fenn | oh noes i have to connect a tube | 07:57 |
| kanzure | i don't think it's just a tube | 07:57 |
| kanzure | not sure | 07:57 |
| fenn | two tubes | 07:57 |
| kanzure | hah | 07:58 |
| kanzure | you just doubled the tube budget! | 07:58 |
| fenn | totally tubular man | 07:58 |
| archels | also grad students are cheap | 07:58 |
| archels | =) | 07:58 |
| fenn | yeah just hook them up to some tubes and yer good to go | 07:58 |
| archels | haha | 07:59 |
| kanzure | i wonder if liquid cooling racks are directly compatible with this | 08:01 |
| kanzure | or do people just submerge entire racks? | 08:01 |
| fenn | liquid cooling is pretty rare in data centers | 08:05 |
| fenn | only fancy mainframes like cray do immersion | 08:05 |
| kanzure | was not aware that craystuff did that | 08:05 |
| fenn | http://spectrum.ieee.org/computing/it/is-there-a-liquid-fix-for-the-clouds-heavy-energy-footprint | 08:06 |
| fenn | "Because liquids are denser than gases, they are a more efficient medium to transport and remove unwanted heat." is not exactly true. steam has a higher enthalpy per unit volume than water | 08:08 |
| kanzure | a friend mentioned to me once that air is cheaper and more efficient anyway | 08:08 |
| kanzure | and also doesn't require special equipment compared to alternatives | 08:08 |
| kanzure | soo... air. | 08:08 |
| fenn | but water has a smaller boundary layer so it's more efficient to pump lots of it around in small pipes | 08:08 |
| fenn | heat pipes are pretty cool (so punny) | 08:09 |
| fenn | most laptops have them now | 08:10 |
| -!- narwh4l [~michael@unaffiliated/thesnark] has quit [Quit: Leaving] | 08:22 | |
| -!- ParahSailin [~parahsail@unaffiliated/parahsailin] has joined ##hplusroadmap | 08:23 | |
| kanzure | hm | 08:24 |
| ParahSailin | brain transplant of bouncer | 08:24 |
| ParahSailin | its a p-zombie now im afraid though, no qualias | 08:24 |
| kanzure | irc bouncer or physical bouncer? what are we talking about | 08:25 |
| ParahSailin | irc | 08:26 |
| kanzure | what is an insect brain good for anyway. | 08:32 |
| kanzure | "Thomas DeMarse at the University of Florida used a culture of 25,000 neurons taken from a rat's brain to fly a F-22 fighter jet aircraft simulator.[100] After collection, the cortical neurons were cultured in a petri dish and rapidly began to reconnect themselves to form a living neural network. The cells were arranged over a grid of 60 electrodes and used to control the pitch and yaw functions of the simulator. " | 08:33 |
| kanzure | oh, i thought it was like a <500 neuron culture | 08:33 |
| kanzure | ( http://neural.bme.ufl.edu/page13/assets/NeuroFlght2.pdf ) | 08:34 |
| fenn | but did it fly | 08:34 |
| kanzure | The typical performance of the network’s control was within 10 degrees of desired for both pitch and roll. However, as the high frequency stimulations increase the weights used to fly the aircraft these weights will eventually become too large resulting in over corrections from even the smallest in errors. Moreover, the rate at which pitch and roll channels change (as a result of the high frequency stimulation) over time can lead to ... | 08:36 |
| kanzure | ... differences in control. This over correction is apparent in the data for the roll angles whose flight weights caused the aircraft to begin to oscillate." | 08:36 |
| fenn | but did it fly | 08:37 |
| kanzure | (figure 4) | 08:38 |
| fenn | step 4 ask someone else to read the paper for you | 08:38 |
| kanzure | "After, several minutes of flight, the network will slowly begins to correct for any errors in the flight path resulting in straight and level flight." | 08:38 |
| JayDugger | Flying within 10 degrees of error on two axes means crashing. | 08:38 |
| fenn | wow it's really tiny | 08:38 |
| kanzure | oh leave it to the aviation person to ruin the party | 08:39 |
| kanzure | hmm. | 08:39 |
| fenn | i'm actually kinda surprised nobody has managed to make commercially viable biological information processing units yet | 08:41 |
| JayDugger | I apologize. You can always walk away from crash in a simulator. | 08:41 |
| fenn | compared to fabbing silicon the costs are really low per therblig or whatever it is they're doing | 08:42 |
| fenn | even if you don't have electrodes stuck to every square micron, the neural culture can still do useful stuff | 08:43 |
| fenn | it's just harder to debug | 08:43 |
| kanzure | perhaps not, if that culture didn't really fly | 08:43 |
| fenn | i would also think that directing the cell growth radially or with some other structured pattern would improve the i/o characteristics | 08:44 |
| JayDugger | What does "therblig" mean here? | 08:44 |
| fenn | .wik therblig | 08:44 |
| yoleaux | "Therbligs are 18 kinds of elemental motions used in the study of motion economy in the workplace. A workplace task is analyzed by recording each of the therblig units for a process, with the results used for optimization of manual labor by eliminating unneeded movements." — http://en.wikipedia.org/wiki/Therblig | 08:44 |
| JayDugger | Yeah, did that. I still don't follow. | 08:44 |
| fenn | some kind of poorly defined information processing task | 08:44 |
| JayDugger | Okay. Got it. Thank you. | 08:44 |
| fenn | like for controlling a robot, say | 08:45 |
| fenn | silicon gates are so cheap now i doubt commercial biological information processing will ever happen in the future | 08:46 |
| fenn | but it totally could have happened | 08:46 |
| fenn | also, supply chains and hardware malware | 08:47 |
| fenn | silicon is cheap but you have to trust TSMC to make exactly what you tell them | 08:48 |
| fenn | or <insert billion dollar fab here> | 08:48 |
| kanzure | same with neurons though | 08:48 |
| fenn | why? | 08:48 |
| kanzure | there might be other hardware before/after | 08:49 |
| fenn | afaik nobody knows how to implant latent malware into a neuron | 08:49 |
| fenn | and if they did, you have a lot more control over the tissue replication process than you do over chip manufacturing | 08:49 |
| fenn | yeh there will be other hardware but it's less fab intensive | 08:50 |
| fenn | anyone here do macro photography? | 08:51 |
| -!- Qfwfq [~WashIrvin@unaffiliated/washirving] has quit [Ping timeout: 245 seconds] | 08:51 | |
| fenn | i am wondering about correcting for chromatic abberation in software and RGB LEDs for illumination | 08:52 |
| kanzure | demarse's earlier paper seems more comprehensive http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2440704/ | 08:53 |
| fenn | .title | 08:53 |
| yoleaux | The Neurally Controlled Animat: Biological Brains Acting with Simulated Bodies | 08:53 |
| fenn | why is it always simulated though | 08:53 |
| fenn | warwick at least drove a toy robot around | 08:53 |
| kanzure | "Cortical tissue (from day 18 embryonic rats) is dissociated (Potter and DeMarse, 2001) and cultured on a 60 channel multi-electrode array (Multichannel Systems) shown in Fig. 2. Each electrode can detect the extracellular activity (action potentials) of several nearby neurons and can stimulate activity by passing a voltage or current through the electrode and across nearby cell membranes (e.g., +/−600 mV 400 μs, biphasic pulses). ... | 08:54 |
| kanzure | ... Dissociated neurons begin forming connections within a few hours in culture, and within a few days establish an elaborate and spontaneously active living neural network. After one month in culture, development of these networks becomes relatively stable (Gross et al., 1993; Kamioka et al., 1996; Watanabe et al., 1996) and is characterized by spontaneous bursts of activity. This activity was measured in real-time and used to produce ... | 08:54 |
| fenn | simulated bodies miss out all the squitchy bugs that hardware bodies have and silicon can't deal with | 08:54 |
| kanzure | ... movements within a virtual environment." | 08:54 |
| fenn | s/silicon/symbolic control systems/ | 08:54 |
| fenn | one month eh | 08:55 |
| cluckj | my blood glucose has a website now :D | 08:56 |
| cluckj | blood in the cloud | 08:56 |
| fenn | hook it up to your web toaster, web vending machine, and web coffee maker | 08:56 |
| -!- Qfwfq [~WashIrvin@unaffiliated/washirving] has joined ##hplusroadmap | 08:56 | |
| cluckj | display my blood sugar on every web-enabled device in the house!!!!!!!!!!!!!!!!!!! | 08:59 |
| kanzure | hook it up to taskrabbit or instacart and get cookies delivered | 08:59 |
| cluckj | hahahahahaha | 08:59 |
| fenn | that's what i meant | 08:59 |
| cluckj | there is a cookie delivery service here | 09:00 |
| fenn | "your blood sugar seems a little low, wouldn't you like to try a nice refreshing mountain dew?" | 09:00 |
| kanzure | now you know what you must do | 09:00 |
| cluckj | yessss | 09:01 |
| cluckj | https://insomniacookies.com/ | 09:02 |
| fenn | cookies that never sleep | 09:02 |
| kanzure | use the postmates api, http://blog.postmates.com/post/104856354257/powering-on-demand-logistics | 09:02 |
| kanzure | docs https://postmates.com/developer | 09:02 |
| kanzure | er i meant https://postmates.com/developer/docs | 09:03 |
| cluckj | nice | 09:04 |
| fenn | hexagons! and cubes! | 09:05 |
| fenn | blue and white CSS! | 09:05 |
| -!- Quashie [~boingredd@TOROON0240W-LP140-01-2925095627.dsl.bell.ca] has joined ##hplusroadmap | 09:09 | |
| fenn | "you can enter any store or address, specify an item and have it not only delivered but also purchased at the pick-up" | 09:09 |
| fenn | we waited until 2015 to have a guy go down the street and buy a candy bar? | 09:10 |
| fenn | surely this must have existed in the 1999 dot com bubble | 09:10 |
| -!- Quashie_ [~boingredd@TOROON0240W-LP140-03-845478924.dsl.bell.ca] has quit [Ping timeout: 264 seconds] | 09:10 | |
| kanzure | webvan | 09:10 |
| fenn | .wik webvan | 09:11 |
| yoleaux | "Webvan was an online "credit and delivery"[clarification needed] grocery business that went bankrupt in 2001. It was headquartered in Foster City, California, USA, in Silicon Valley." — http://en.wikipedia.org/wiki/Webvan | 09:11 |
| * fenn wonders what's at 20 mcallister street | 09:13 | |
| fenn | oh it's the fortress of failed economies | 09:13 |
| fenn | .wik hibernia bank | 09:14 |
| yoleaux | "Hibernia is the Classical Latin name for the island of Ireland." — http://en.wikipedia.org/wiki/Hibernia_Bank | 09:14 |
| fenn | hm no that's 1 jones street | 09:15 |
| fenn | maybe it has 2 addresses | 09:15 |
| kanzure | "We thank C. Michael Atkin, Gray Rybka, and Samuel Thompson for early programming on the Animat and neural interface and MultiChannel Systems (http://www.multichannelsystems.com) for their gracious technical support; Sami Barghshoon and Sheri McKinney for help with cell culture. This research was supported by a grant from the National Institute of Neurological Disorders and Stroke, RO1 NS38628 (SMP) and by the Burroughs-Wellcome/Caltech ... | 09:15 |
| kanzure | ... Computational Molecular Biology fund (DAW)." | 09:15 |
| kanzure | also things weren't very mobile at the time | 09:23 |
| kanzure | so people would have to go home to get new instructions | 09:23 |
| kanzure | or develop a phone system for that | 09:23 |
| kanzure | which may be too cumbersome for the delivery people to bother with | 09:23 |
| fenn | it seems like mixing huge vats full of mineral oil and high powered electronics would be a big fire hazard | 09:25 |
| fenn | didnt they have pagers? that's totally 90s | 09:26 |
| fenn | surely someone could figure out how to send pages from a computer | 09:26 |
| kanzure | you have such high expectations | 09:28 |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-bzgauiaspnmkbmbw] has joined ##hplusroadmap | 09:32 | |
| kanzure | win 55 | 09:34 |
| kanzure | whoops | 09:34 |
| -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 244 seconds] | 09:38 | |
| -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 09:40 | |
| -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Excess Flood] | 09:41 | |
| -!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has quit [Read error: Connection reset by peer] | 09:43 | |
| CaptHindsight | kanzure: what's the current state of the patents on the first gen DNA synthesizers? | 09:51 |
| -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 10:01 | |
| kanzure | expired | 10:06 |
| CaptHindsight | any idea how fast they are currently able to read base pairs? (referring to DNA a linear recording) | 10:11 |
| kanzure | they don't read | 10:11 |
| kanzure | those are dna sequencers | 10:11 |
| kanzure | which also have lots of expired patents (especially around sanger sequencing, i think) | 10:12 |
| kanzure | however, there's lots of recent fancypants dna sequencing technology that is much more efficient | 10:12 |
| CaptHindsight | I know I'm talking about reading now vs printing/sequencing | 10:12 |
| kanzure | i would pester ParahSailin about dna sequencing things, he seems to know various hiseq machines on the market at the moment | 10:12 |
| ParahSailin | solexa sequencing has been around for about 8 years now | 10:13 |
| CaptHindsight | feed DNA though an aluminum oxide pore, detect and record sequences as they pas through the end of the pore | 10:14 |
| ParahSailin | is that how minion works? | 10:15 |
| kanzure | are you asking a question | 10:15 |
| kanzure | there are many nanopore sequencing methods | 10:15 |
| CaptHindsight | should be able to read an entire length in minutes | 10:16 |
| ParahSailin | this is your theory? | 10:17 |
| CaptHindsight | I'm just taking a fresh look at reading and synthesizing | 10:17 |
| kanzure | CaptHindsight: here's some stuff, | 10:17 |
| kanzure | http://diyhpl.us/~bryan/papers2/bio/Cheap%20third-generation%20sequencing%20-%20nanopores%20-%20cyclodextrin%20-%20hemolysin%20-%202009.pdf | 10:17 |
| CaptHindsight | someone made a single chip reader, I forget the bandwidth | 10:17 |
| kanzure | http://diyhpl.us/~bryan/papers2/bio/Physical%20approaches%20to%20DNA%20sequencing%20and%20detection%20-%202007.pdf | 10:17 |
| kanzure | http://diyhpl.us/~bryan/papers2/bio/Progress%20in%20sequencing%20DNA%20with%20an%20AFM.pdf | 10:17 |
| kanzure | http://diyhpl.us/~bryan/papers2/bio/Rapid%20Sequencing%20of%20Individual%20DNA%20Molecules%20in%20Graphene%20Nanogaps%20-%202008.pdf | 10:18 |
| kanzure | http://diyhpl.us/~bryan/papers2/bio/Rapid%20sequencing%20of%20individual%20DNA%20molecules%20in%20graphene%20nanogaps.pdf | 10:18 |
| kanzure | http://diyhpl.us/~bryan/papers2/bio/Real-time%20DNA%20sequencing%20from%20single%20polymerase%20molecules.pdf | 10:18 |
| kanzure | http://diyhpl.us/~bryan/papers2/bio/Single-molecule%20sequencing%20of%20an%20individual%20human%20genome.pdf | 10:18 |
| kanzure | http://diyhpl.us/~bryan/papers2/bio/The%20potential%20and%20challenges%20of%20nanopore%20sequencing%20-%20Schloss.pdf | 10:18 |
| kanzure | that's all i remember at the moment | 10:18 |
| kanzure | oh also http://diyhpl.us/~bryan/papers2/bio/Nanopore%20Unzipping%20of%20Individual%20DNA%20Hairpin%20Molecules%20-%202004.pdf | 10:18 |
| t12 | nanopore has a bad rep | 10:19 |
| t12 | from 20 years of being promised the world | 10:19 |
| ParahSailin | minion is actually available | 10:19 |
| kanzure | you're thinking of nanotech | 10:19 |
| kanzure | different subject | 10:19 |
| t12 | no, nanopore sequencing | 10:19 |
| t12 | oxfort et al | 10:19 |
| t12 | oxford | 10:19 |
| t12 | the most imperssive sequencing is prolly pac bio's single molecule | 10:20 |
| t12 | if you have a spare million dollars | 10:20 |
| ParahSailin | you can actually buy minion... | 10:20 |
| t12 | buying a thing and a thing working for practical use | 10:20 |
| t12 | are very diffnt | 10:20 |
| CaptHindsight | whats the problem with nanoopre by MinION? | 10:20 |
| ParahSailin | it has error rate, but its a fine complementary sequencing tech | 10:20 |
| kanzure | pacbio's stuff was from back in 2008 and ignores the other 6 years of sequencing | 10:21 |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Ping timeout: 244 seconds] | 10:21 | |
| CaptHindsight | recall how high the error rate is and why? | 10:22 |
| t12 | capt: last review i read said essentially | 10:22 |
| t12 | beta testers genomes that they tried to push through it | 10:22 |
| t12 | couldn't successfully assemble | 10:22 |
| ParahSailin | so we actually have one and have actually used it | 10:22 |
| t12 | my assumption is due to big messed up chunks it cant handle well | 10:22 |
| kanzure | right now i'm like 100% sure that t12 is full of shit | 10:22 |
| ParahSailin | pretty high indel rate | 10:22 |
| t12 | i know i'm going against hype religion with hating on nanopore sequencing here | 10:22 |
| t12 | and it may be the future | 10:23 |
| t12 | but it doesnt seem to be there yet | 10:23 |
| kanzure | is your position that ParahSailin is lying? | 10:23 |
| kanzure | please be more specific | 10:23 |
| ParahSailin | but if you are sequencing microsatellite stuff, its a decent complementary technology | 10:23 |
| CaptHindsight | I'm wondering if the problem with MinION is just the wrong people making it | 10:23 |
| ParahSailin | solexa will probably always be the best | 10:24 |
| t12 | i guess i'm saying that | 10:24 |
| t12 | try and replace a modern illumina sequencer with a nanopore sequencer | 10:24 |
| kanzure | t12: and more speciifcally, you believe that he hasn't actually used minion? | 10:24 |
| t12 | and failure is likely | 10:24 |
| t12 | sequencer performance is very contexted on what your task is | 10:24 |
| ParahSailin | i wouldnt use minion for general use | 10:24 |
| t12 | for some things some kinds of errors do/dont matter | 10:25 |
| t12 | for some things read length does/doesnt matter | 10:25 |
| t12 | for some things some kinds of sequences/samples preps do/dont work | 10:26 |
| CaptHindsight | http://www.technologyreview.com/news/530746/radical-new-dna-sequencer-finally-gets-into-researchers-hands/ | 10:26 |
| kanzure | CaptHindsight: the first rule is never read the news | 10:26 |
| CaptHindsight | a list of commercially available readers would be handy to review | 10:26 |
| CaptHindsight | lest philosophy more science | 10:27 |
| CaptHindsight | lest/less | 10:27 |
| t12 | theres also ion torrent | 10:27 |
| t12 | i think they're mainly fast | 10:27 |
| kanzure | the second rule is "no philosophy" | 10:27 |
| t12 | but its kinda immensely complex | 10:27 |
| ParahSailin | ion torrent is not that useful for anything | 10:27 |
| t12 | solids are on the way out | 10:27 |
| ParahSailin | we just sold ours | 10:28 |
| CaptHindsight | I want to see what these people are having trouble with and why they can't seem to make them work | 10:28 |
| t12 | you can get solids pretty cheap | 10:29 |
| kanzure | sequencing is a very competitive market right now. synthesis is where you should be applying your effort. | 10:29 |
| t12 | like banged up ones for <10k on ebay | 10:29 |
| t12 | getting it working, getting reagents | 10:29 |
| t12 | suffering through the prep | 10:29 |
| t12 | another story | 10:29 |
| CaptHindsight | need both sides | 10:29 |
| kanzure | sequencing works | 10:29 |
| CaptHindsight | how long does it take? | 10:29 |
| t12 | sequencing? | 10:29 |
| CaptHindsight | yes | 10:30 |
| -!- Souljack [souljack@shell.xshellz.com] has quit [Ping timeout: 250 seconds] | 10:30 | |
| t12 | for illumina stuff | 10:30 |
| ParahSailin | illumina's marketing material is accurate | 10:30 |
| t12 | 24hr-1week depending | 10:30 |
| ParahSailin | wrt performance | 10:30 |
| t12 | oh model, run type, etc | 10:30 |
| t12 | read length | 10:30 |
| CaptHindsight | I'm talking about reading an entire sequence in minutes | 10:30 |
| t12 | what does entire sequence mean | 10:31 |
| -!- delinquentme [~delinquen@74.61.157.78] has joined ##hplusroadmap | 10:31 | |
| t12 | generally you chop stuff up and read alot in parallel | 10:31 |
| ParahSailin | if you want benchtop rimmediate results, maybe minion or iontorrent | 10:31 |
| t12 | then try and stitch the data back together | 10:31 |
| CaptHindsight | 1.8 meters of DNA | 10:32 |
| ParahSailin | but youd be surprised how infrequent that use case is | 10:32 |
| kanzure | even if you are doing single-molecule sequencing, you still use lots of molecules in parallel | 10:32 |
| ParahSailin | waiting for accurate illumina sequence in a day is never your time bottleneck | 10:32 |
| t12 | generally sample prep+data analysis | 10:33 |
| t12 | become limits pretty quick | 10:33 |
| t12 | unless you've really really got it dialed in | 10:33 |
| t12 | also just straight up failing runs | 10:33 |
| t12 | illumina will refund your chemistry costs if you get <50% spec i think it is | 10:34 |
| t12 | otherwise its acceptable loss for example | 10:34 |
| CaptHindsight | most of the time these dev teams have the wrong people working on the tech | 10:35 |
| t12 | which dev teams | 10:36 |
| CaptHindsight | I suspect that the case here again with the sequencers like minION | 10:36 |
| -!- yashgaroth [~ffffff@2602:306:35fa:d500:9dff:6269:2f1d:71ab] has joined ##hplusroadmap | 10:36 | |
| CaptHindsight | I'll take a look at it later | 10:37 |
| CaptHindsight | thanks for all the links! | 10:37 |
| t12 | i'm sure the standard science world caveats apply to the reviews | 10:37 |
| t12 | ie: PI gets one, gives it to overworked postdoc says give me something i can write up about this | 10:37 |
| t12 | postdoc has no paper motive to put much effort into it, etc | 10:38 |
| kanzure | postdoc is just going to pawn it off on gradstudentbot | 10:38 |
| -!- bugcatcher [~bugcatche@131.252.130.250] has joined ##hplusroadmap | 10:40 | |
| delinquentme | heh | 10:40 |
| -!- bugcatcher [~bugcatche@131.252.130.250] has quit [Remote host closed the connection] | 10:41 | |
| CaptHindsight | I was working with a group making synthetic DNA | 10:42 |
| ParahSailin | idt? | 10:42 |
| CaptHindsight | they can easily mass produce it in bulk | 10:42 |
| -!- gradstudentbot [~gradstude@131.252.130.250] has joined ##hplusroadmap | 10:42 | |
| -!- gradstudentbot [~gradstude@131.252.130.250] has quit [Remote host closed the connection] | 10:42 | |
| CaptHindsight | for ID and anti-counterfeit use | 10:43 |
| kanzure | there's no reason to do single molecule dna synthesis | 10:43 |
| CaptHindsight | I'd like to obsolete the tech and also in doing so have it available for biomed | 10:44 |
| kanzure | if your solid support has ~1000 copies on it, it's really not problematic | 10:44 |
| kanzure | but also, one method that people have used is cmos-based patterning of the initial components so that strands have a very specific spacing between each other, i think | 10:45 |
| delinquentme | "solid support" ? | 10:49 |
| kanzure | microbeads, larger beads, etc | 10:49 |
| kanzure | solid phase synthesis | 10:49 |
| kanzure | .wik solid-phase synthesis | 10:49 |
| yoleaux | "In chemistry, solid-phase synthesis is a method in which molecules are bound on a bead and synthesized step-by-step in a reactant solution; compared with normal synthesis in a liquid state, it is easier to remove excess reactant or byproduct from the product. In this method, building blocks are protected at all reactive functional groups." — http://en.wikipedia.org/wiki/Solid-phase_synthesis | 10:49 |
| kanzure | also i'm not sure i've heard of a reasonable single molecule dna synthesis proposal | 10:53 |
| kanzure | there's some nanopore ideas that were floated around in the enzymatic dna synthesis group | 10:53 |
| kanzure | https://groups.google.com/group/enzymaticsynthesis | 10:53 |
| kanzure | so i suppose that would be single molecule... | 10:54 |
| t12 | capt: twist? | 10:58 |
| kanzure | there's tunneling tips that have been used for chemistry http://www.physics.mcgill.ca/~peter/534A/rieder.pdf | 11:00 |
| CaptHindsight | are there signs that people already have this all working quickly and they are just keeping it quiet or are they having trouble getting it working? | 11:02 |
| kanzure | yes, the groups that are interested in throughput and timeliness just use highly parallel setups | 11:03 |
| kanzure | e.g., million-pixel LCDs and microarrays | 11:03 |
| t12 | the synthesis is not so bad | 11:04 |
| kanzure | even venter paid >$40M for a genome | 11:05 |
| t12 | its the purity vs length, throughput, etc issues | 11:05 |
| ParahSailin | cambrian genomics probably has real progress | 11:05 |
| kanzure | (well, $400k for the final synthesis...) | 11:05 |
| kanzure | hm "Synthesis of 3d metallic single-molecule magnets" | 11:06 |
| CaptHindsight | whats their turn around time to read and make copies? | 11:06 |
| kanzure | dna copying takes like <5 minutes | 11:06 |
| CaptHindsight | heh, that just gave me an idea | 11:07 |
| kanzure | at least one of the known polymerases operates at 20,000 bp/sec | 11:07 |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has joined ##hplusroadmap | 11:08 | |
| CaptHindsight | not sure what the error rate is for the nanopore | 11:08 |
| ParahSailin | 15% | 11:08 |
| CaptHindsight | wow, do they scan in parallel and average the results? | 11:09 |
| ParahSailin | no | 11:09 |
| t12 | you can sorta eliminate errors in data analysis | 11:11 |
| t12 | if you have lots of repeated copies of something | 11:11 |
| CaptHindsight | I guess you can with biological vs synthetic | 11:12 |
| CaptHindsight | the whole point of the synthetic is to make it not compatible with anything in nature (well it you're acting responsibly) | 11:13 |
| CaptHindsight | it/if | 11:13 |
| CaptHindsight | it also tends to be much shorter stands | 11:14 |
| -!- Souljack [souljack@shell.xshellz.com] has joined ##hplusroadmap | 11:18 | |
| kanzure | "not compatible with anything in nature" ?? | 11:54 |
| -!- balrog [~balrog@discferret/developer/balrog] has quit [Remote host closed the connection] | 11:57 | |
| -!- balrog [~balrog@discferret/developer/balrog] has joined ##hplusroadmap | 12:05 | |
| -!- zadock [~zadock@81.180.210.87] has joined ##hplusroadmap | 12:06 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Quit: Leaving] | 12:13 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 12:17 | |
| kanzure | "Science as a vocation" (1918) http://anthropos-lab.net/wp/wp-content/uploads/2011/12/Weber-Science-as-a-Vocation.pdf | 12:27 |
| delinquentme | kanzure, keeping gunicorn running as application server post disconnect from server | 12:36 |
| delinquentme | supervisor? | 12:36 |
| delinquentme | something else? IE: " What handles instantiating the application server ?" | 12:36 |
| kanzure | "disconnect"? | 12:38 |
| kanzure | um, yes supervisor is fine here | 12:39 |
| kanzure | also normal things like runit, systemd, sysvinit, whatever | 12:39 |
| kanzure | supervisord is probably more your style (screw the standards!) | 12:39 |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Ex-Chat] | 12:42 | |
| delinquentme | kanzure, perhaps theres not a good reason to use supervisor? | 12:44 |
| delinquentme | i've just used it before ... where these ones you've mentioned, I've got no exp with | 12:45 |
| kanzure | the others are often conventionally used to boot your computer | 12:45 |
| -!- FourFire [~FourFire@2.149.64.29.tmi.telenormobil.no] has joined ##hplusroadmap | 13:05 | |
| -!- CaptHindsight is now known as not-zeeshan | 13:14 | |
| -!- not-zeeshan is now known as CaptHindisght | 13:15 | |
| -!- zadock [~zadock@81.180.210.87] has quit [Quit: Leaving] | 13:18 | |
| -!- Zinglon [~Zinglon@ip565f6f48.direct-adsl.nl] has quit [Ping timeout: 245 seconds] | 13:21 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-bzgauiaspnmkbmbw] has quit [Quit: Connection closed for inactivity] | 13:33 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 246 seconds] | 13:38 | |
| -!- CRM114 [~chatz@unaffiliated/urchin] has joined ##hplusroadmap | 13:41 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-equtdsflzcndfdqq] has joined ##hplusroadmap | 13:51 | |
| -!- delinquentme [~delinquen@74.61.157.78] has quit [Remote host closed the connection] | 14:07 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 14:21 | |
| kanzure | "We just have developed an interactive substructure search engine that runs over the wikipedia data" http://www.cheminfo.org/wikipedia/ | 14:29 |
| kanzure | .title http://www.jcheminf.com/content/7/1/10/abstract | 14:29 |
| yoleaux | Journal of Cheminformatics | Abstract | Wikipedia Chemical Structure Explorer: substructure and similarity searching of molecules from Wikipedia | 14:29 |
| -!- Souljack [souljack@shell.xshellz.com] has quit [Ping timeout: 244 seconds] | 15:12 | |
| -!- Souljack [souljack@shell.xshellz.com] has joined ##hplusroadmap | 15:14 | |
| maaku | .title http://www.implicitcad.org/ | 15:16 |
| yoleaux | ImplicitCad.org | 15:16 |
| kanzure | maaku: use instead http://solvespace.com/index.pl | 15:17 |
| nmz787 | kanzure: why are you still recommending things other than BRLCAD? | 15:20 |
| nmz787 | maaku: implicitCAD isn't very good for getting really really nice models if they have high dynamic-range in my experience | 15:22 |
| nmz787 | (dynamic range of scale of features) | 15:22 |
| kanzure | brlcad's implementation of nurbs stuff is based on opennurbs (which lacks tests... rhino keeps those private), and some custom stuff brlcad had to implement (which is nowhere near as cleanly implemented as in solvespace) | 15:22 |
| nmz787 | if it's nurbs, should the tests be the same pretty much, other than the APIs? | 15:23 |
| nmz787 | s/should/shouldn't/ | 15:23 |
| nmz787 | and can you really trust one developer to implement things well versus a team of people at a company specifically focusing on CAD? | 15:24 |
| kanzure | well what else are you going to test, minor variations in step output? that's not relevant | 15:24 |
| kanzure | yes, a single person is 10000x more trustable | 15:24 |
| nmz787 | well NURBs is math | 15:24 |
| nmz787 | so shouldn't the tests all be math-based tests | 15:24 |
| kanzure | you're welcome to try that, but i wouldn't be surprised if even very basic operations (like "create a sphere" or "intersect a box with a sphere") produce dramatically different results between libraries | 15:25 |
| nmz787 | well I haven't really done anything with NURBs since the math is still a bit beyond me | 15:26 |
| kanzure | http://en.wikipedia.org/wiki/Non-uniform_rational_B-spline#General_form_of_a_NURBS_curve | 15:27 |
| -!- FourFire [~FourFire@2.149.64.29.tmi.telenormobil.no] has quit [Quit: Leaving] | 15:28 | |
| nmz787 | I need an animated cartoon video version of that | 15:28 |
| nmz787 | also, why don't these smart people who know NURBS contributing to something like BRLCAD which has tons of eyes looking at the code, and tons of file converters | 15:30 |
| kanzure | http://upload.wikimedia.org/wikipedia/commons/3/3d/B%C3%A9zier_2_big.gif | 15:30 |
| nmz787 | why don't all these people work together instead of splintering and leaving half-useful code lying around | 15:30 |
| nmz787 | ah yeah that gif is a good one | 15:31 |
| nmz787 | this was a good one too http://geometrie.foretnik.net/files/NURBS-en.swf | 15:31 |
| kanzure | realistically i think that this should not necessarily be internal to brlcad. there should be a separate nurbs kernel, and then brlcad should consume that library, and even develop on it; but it should not be a sublibrary of brlcad. | 15:35 |
| -!- zadock [~zadock@81.180.210.87] has joined ##hplusroadmap | 15:35 | |
| kanzure | there are lots of subtle and annoying reasons why people will tend to create new libraries, and most of the reasons are bad | 15:35 |
| kanzure | unfortunately it is often easier to write something new than to read existing source code, because you might have to fix the existing source code in things that seem totally unrelated to what you are actually doing (this is just the nature of large software projects...) | 15:36 |
| kanzure | *of poorly planned large software projects | 15:36 |
| kanzure | i believe that technically the nurbs surface-surface intersection stuff in brlcad is working, although i don't know which edge cases it is presently failing on, or which ones they have surpassed. there was a gsoc writeup on their wiki explaining the state in 2012 and again in 2013. | 15:38 |
| nmz787 | yeah there was someone in the kicad chatroom mentioning some converter that BRLCAD already had, and reading their code was the only reason I remember them giving for not wanting to use that. | 15:43 |
| kanzure | brlcad's converters are alright | 15:44 |
| nmz787 | friggin, I pay for BRLCAD with my taxes! | 15:45 |
| * nmz787 goes to demand things in #BRLCAD | 15:46 | |
| kanzure | i think they would be happy to integrate a library like solvespace if it was cleaned up a little bit (and if it had a more permissive license) | 15:47 |
| kanzure | (iirc, they can't integrate gpl stuff) | 15:47 |
| nmz787 | even tax credits for patches would be pretty cool | 15:48 |
| chris_99 | http://www.renishaw.com/en/tonic-linear-encoders--10186 - down to 1nm resolution | 15:56 |
| -!- CRM114 [~chatz@unaffiliated/urchin] has quit [Remote host closed the connection] | 15:58 | |
| -!- Shannon [~Shannon@unaffiliated/shannon] has joined ##hplusroadmap | 15:58 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Ping timeout: 255 seconds] | 16:02 | |
| kanzure | https://github.com/genworks/gendl | 16:16 |
| kanzure | "Gendl allows for high-level declarative, object-oriented problem solving and application development, including but not limited to the ability to generate and manipulate 3D geometry. To solve a problem in Gendl, you formulate it using the define-object operator, which allows you to specify inputs, outputs (computed-slots), and child objects, which then gives the ability to generate a "tree" of objects, useful for decomposing complexity." | 16:16 |
| kanzure | "has you describe parts and sub-assemblies as objects, and then generating geometry commands from the object graph." | 16:17 |
| -!- delinquentme [~delinquen@74.61.157.78] has joined ##hplusroadmap | 16:33 | |
| delinquentme | http://www.implicitcad.org/ | 16:33 |
| kanzure | -_- | 16:36 |
| kanzure | delinquentme: this can't possibly be the first you're learning about implicitcad | 16:36 |
| kanzure | first mentioned in hplusroadmap logs back in 2011 (2011-10-01) and also at least nmz787, fenn and maybe even juri_ have squawked about using it... | 16:39 |
| kanzure | but also, check http://gnusha.org/logs/2015-03-22.log | 16:39 |
| -!- delinquentme [~delinquen@74.61.157.78] has quit [Remote host closed the connection] | 16:39 | |
| nmz787 | what a troll | 16:39 |
| kanzure | right? | 16:40 |
| -!- heath [~heath@unaffiliated/ybit] has quit [Quit: leaving] | 17:12 | |
| -!- heath [~heath@unaffiliated/ybit] has joined ##hplusroadmap | 17:12 | |
| -!- Quashie [~boingredd@TOROON0240W-LP140-01-2925095627.dsl.bell.ca] has quit [Ping timeout: 244 seconds] | 17:19 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 17:45 | |
| justanotheruser | whoah implicitcad is cool | 17:58 |
| -!- helleshin [~talinck@66-161-138-110.ubr1.dyn.lebanon-oh.fuse.net] has quit [Read error: Connection reset by peer] | 17:59 | |
| -!- helleshin [~talinck@66-161-138-110.ubr1.dyn.lebanon-oh.fuse.net] has joined ##hplusroadmap | 17:59 | |
| nmz787 | it's easy to design stuff in, but wasn't usable quality-wise for me :( | 18:04 |
| kanzure | hooray i helped a cadperson https://news.ycombinator.com/item?id=9248640 | 18:05 |
| nmz787 | ugh, why aren't there some minions for me around here? | 18:10 |
| nmz787 | i mean people to hang out with and discuss things on whiteboards | 18:10 |
| nmz787 | light-emitting display screens are so last decade/century | 18:11 |
| -!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has joined ##hplusroadmap | 18:13 | |
| cluckj | oh shit is reflecting ambient light back in? | 18:14 |
| kanzure | nmz787: maybe recruit someone from diybio list to join the channel | 18:15 |
| kanzure | nmz787: did you ever meet lichen? | 18:20 |
| nmz787 | i didn't | 18:20 |
| nmz787 | we chatted briefly though | 18:20 |
| kanzure | what about that tyler gilleries person | 18:20 |
| nmz787 | erhmmm | 18:21 |
| nmz787 | who? | 18:21 |
| kanzure | i dunno, some person | 18:21 |
| kanzure | i don't have any data | 18:21 |
| nmz787 | are pair-programming and peer-programming different things? | 18:32 |
| nmz787 | (which are presumably both different than pear-programming) | 18:32 |
| -!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap | 18:36 | |
| kanzure | pair programming is where two programmers fight for control over a keyboard | 18:38 |
| kanzure | and when they get infuriated with one another for solving problems in different orders | 18:39 |
| nmz787 | but when it's >2 programmers, then it's peer-programming? | 18:41 |
| kanzure | n oidea | 18:41 |
| kanzure | *no idea | 18:41 |
| kanzure | also: someone is interested in making us an hplusroadmap electronic/progressive track/mix | 18:41 |
| kanzure | e.g., we can pick some samples/quotes from youtube videos | 18:42 |
| kanzure | i'm thinking we should pick quotes from talks by aubrey de grey, anders sandberg, carl sagan (when he's not being totally retarded), ralph merkle, etc. | 18:42 |
| cluckj | lol | 18:44 |
| nmz787 | hmm, $500 but a far drive... I think something like this would be like $500 to get shipped http://www.ebay.com/itm/Amray-1200C-Scanning-Electron-Microscope-SEM-/191542207330?_trksid=p2054897.l4275 | 18:44 |
| kanzure | .title https://www.youtube.com/watch?v=l9uWyR_2j8U | 18:44 |
| yoleaux | Ralph Merkle - The Contributions of Robert Freitas to Molecular Nanotechnology - YouTube | 18:44 |
| kanzure | so many good quotes | 18:44 |
| kanzure | .title https://www.youtube.com/watch?v=FJJvzTsIITY | 18:49 |
| yoleaux | Part 3: Ralph Merkle on the State of the Art of Cryopreservation - YouTube | 18:49 |
| kanzure | .title https://www.youtube.com/watch?v=XfTqXL0d9Ls&t=14s | 19:02 |
| yoleaux | Quest for immortality - YouTube | 19:02 |
| kanzure | excellent quotes | 19:02 |
| kanzure | ooh i should get some jojack quotes | 19:02 |
| -!- AmbulatoryCortex [~Ambulator@173-31-9-188.client.mchsi.com] has quit [Read error: Connection reset by peer] | 19:04 | |
| cluckj | yes, do it | 19:07 |
| -!- Shannon [~Shannon@unaffiliated/shannon] has quit [Read error: Connection reset by peer] | 19:08 | |
| kanzure | hmm | 19:15 |
| kanzure | yesssss https://www.youtube.com/watch?v=XfTqXL0d9Ls&t=1m58s | 19:17 |
| kanzure | and this george church quote https://www.youtube.com/watch?v=Q_zzFRjeGRI&t=2m10s | 19:28 |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-equtdsflzcndfdqq] has quit [Quit: Connection closed for inactivity] | 19:33 | |
| kanzure | lrp5/bruce willis quote https://www.youtube.com/watch?v=Q_zzFRjeGRI&t=7m2s | 19:33 |
| kanzure | "After ImplicitCAD, I worked with Rob Gilson on a SVG/WebGL constraint-based modelling tool, mech.ly. We got some neat stuff working, but never got to a point where we wanted to release things. I'm pretty persuaded this is the right direction to go." | 20:18 |
| kanzure | (dead domain) | 20:19 |
| kanzure | "I abandoned ImplicitCAD to write a constraint-based CAD program with my friend Rob Gilson. (I handled constraint solving with gradient descent, which worked surprisingly well, and a dab of grobner bases here and there.) We got some neat stuff working, but eventually got pulled away to other projects." | 20:19 |
| kanzure | you guys are boring | 20:30 |
| kanzure | is that SEM working | 20:34 |
| kanzure | hm | 20:34 |
| kanzure | "maybe" | 20:34 |
| kanzure | sketchtastic: "I am selling an Amray 1200C SEM (Scanning Electron Microscope), I bought it not to long ago from a guy who said it worked, it does turn on but I can't seem to get a image to appear" | 20:37 |
| kanzure | "I accidentally got some cheetos stuck in one of the vacuum tube, and I think the capacitors have turned to sand. But other than that, yeah it works." | 20:37 |
| nmz787 | i didn't look up the specs on it to see if it would be very nice or not | 20:41 |
| nmz787 | this looks like a good deal, says its working, but it's in Massachusetts http://www.ebay.com/itm/FEI-611-FIB-Focused-Ion-Beam-system-/261815650373?pt=LH_DefaultDomain_0&hash=item3cf56dc845 | 20:45 |
| nmz787 | don't see any dual-beams on ebay | 20:45 |
| -!- Proteus1 [~Proteus@97-121-112-41.omah.qwest.net] has joined ##hplusroadmap | 20:49 | |
| nmz787 | "The SEMs have too many options. The little Vega3-SBH starts at under $100k and the Lyra 3 with 1 GIS and liftout is about $630k but can be dolled up to $1.5m or so without too much trouble." | 20:50 |
| -!- Proteus [~Proteus@unaffiliated/proteus] has quit [Ping timeout: 272 seconds] | 20:51 | |
| nmz787 | http://diyhpl.us/~nmz787/pdf/TESCAN_VEGA3_SB.pdf | 20:53 |
| nmz787 | diyhpl.us/~nmz787/pdf/LYRA_GM.pdf | 20:56 |
| CaptHindisght | http://www.ebay.com/itm/AMRAY-SCANNING-ELECTRON-MICROSCOPE-MODEL-1830D-/310478306270 $6k | 20:56 |
| nmz787 | interestingly my school has one of these https://wiki.rit.edu/display/smfl/Amray+1830+SEM | 21:00 |
| nmz787 | and the manuals there | 21:00 |
| nmz787 | http://www.smfl.rit.edu/pdf/manual/Manual_Amray_1830_SEM_2.pdf | 21:00 |
| nmz787 | 'a manual' | 21:00 |
| nmz787 | precursor to autofocus algorithms "Press Wobbler to check the alignment of the final aperture. A properly aligned aperture will produce an image that pulses in and out of focus when the Wobbler is engaged. If the image shifts up and down or left and right, the final aperture needs alignment" | 21:01 |
| t12 | ems are quite a bit of trouble to service/maintain | 21:02 |
| t12 | i guess sems are somewhat easier than tems | 21:02 |
| nmz787 | yeah no one in the DIY community outside the xray folks and maybe tesla-coil folks do much with High VOltage | 21:02 |
| nmz787 | maybe neon lighting folks | 21:03 |
| nmz787 | http://www.hssemgroup.com/high-school-sems | 21:04 |
| nmz787 | .title | 21:05 |
| yoleaux | High School SEMs - HS_SEM (High Schools with Scanning Electron Microscopes) | 21:05 |
| t12 | also cost quite a bit of money to keep on | 21:05 |
| nmz787 | huh, no wonder bergen county has one (geohot went there) | 21:05 |
| nmz787 | that Vega3 runs off a 15A 110 | 21:06 |
| nmz787 | 100V | 21:06 |
| nmz787 | 110V | 21:06 |
| t12 | thats still like 1k/mo | 21:06 |
| t12 | at baseline residential | 21:06 |
| nmz787 | $126 bux a month around here | 21:06 |
| nmz787 | .11 per kWh * (110V * 15A / 1000V per kV) * 24 hours * 30 days | 21:07 |
| nmz787 | .11*1.6*24*30 == 126.72 | 21:07 |
| nmz787 | err 130.68 | 21:08 |
| t12 | ahh yeah sorry | 21:09 |
| t12 | i always fail at arith | 21:09 |
| kanzure | "optimist prime" | 21:11 |
| kanzure | "I like both the Suffused and Blake Baltimore collections. Both have parts that are excellent for productive work. I also originally thought about mixing in Aubrey, Craig Venter, myself and Peter Diamandis into Aging is a terminal disease track, but it turned out to be too distracting and gave negative impression, so I decided to minimize the distraction and add a female voice with a simple phrase and then playing with the amplitude and ... | 21:29 |
| kanzure | ... wavelengths to go through the various "phases" of life. However, when you are sponsoring a track, you can set any specifications. A track may cost your anywhere from $10 to $10,000 depending on the quality and fame of the DJ and whether you are looking to integrate existing work or produce it de-novo. [....] He wold charge about 2,000 euro for three weeks of work on a track and would work through an agent. Let me know if you would ... | 21:29 |
| kanzure | ... like me to connect you. For your project I can chip in as well if you like. There are many DJs, who would be willing to work on contract. But the quality is very important, so it is best to sample a few. Here I can also contribute. Personally, or even through Insilico." | 21:29 |
| nmz787 | paperbot: http://link.springer.com/article/10.1007%2Fs00300-004-0612-6 | 21:36 |
| paperbot | http://diyhpl.us/~bryan/papers2/paperbot/New%20record%20of%20moss%20and%20thermophilic%20bacteria%20species%20and%20physico-chemical%20properties%20of%20geothermal%20soils%20on%20the%20northwest%20slope%20of%20Mt.%20Melbourne%20%28Antarctica%29.pdf | 21:36 |
| -!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 246 seconds] | 22:01 | |
| -!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap | 22:27 | |
| -!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 272 seconds] | 22:36 | |
| -!- zadock [~zadock@81.180.210.87] has quit [Quit: Leaving] | 23:06 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 256 seconds] | 23:48 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 23:56 | |
| --- Log closed Mon Mar 23 00:00:30 2015 | ||
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!
