--- Log opened Sat Feb 23 00:00:08 2013 | ||
brownies | kanzure: used what within the past month? | 00:12 |
---|---|---|
@kanzure | opencv | 00:20 |
brownies | ahh | 00:22 |
brownies | i actually haven't. all of my CV work was done using the in-house lib we had at... that one place. | 00:22 |
brownies | and it was a long time ago. | 00:23 |
@kanzure | oh, i thought you were doing opencv things recently. something robotic. | 00:23 |
@kanzure | or at least pretending to maybe be assembling some of those parts together. | 00:23 |
brownies | i got as far as making a parts list and picking out what software stack i'd use... then i got distracted by startup things | 00:24 |
brownies | i'm still a little shocked that Willow Garage folded | 00:25 |
@kanzure | nope, it hasn't folded | 00:26 |
@kanzure | http://www.willowgarage.com/blog/2013/02/11/willow-garage-changing | 00:26 |
@kanzure | "Willow Garage has decided to enter the world of commercial opportunities with an eye to becoming a self-sustaining company. This is an important change to our funding model." | 00:26 |
brownies | oh that is interesting. | 00:28 |
brownies | that post is still worryingly short. and it doesn't mitigate the rumors i saw of layoffs. | 00:28 |
@kanzure | the layoff rumors might still be true. | 00:28 |
brownies | kanzure: wtf, though? what was their business model before? how did they survive for the last 5 years? | 00:28 |
@kanzure | charisma and burritos? | 00:29 |
@kanzure | not sure. | 00:29 |
brownies | kanzure: wikipedia doesn't really illuminate much either | 00:31 |
brownies | my best guess is "DARPA grant money + the largesse of early google employees" | 00:31 |
nmz787 | what did willow garage do? | 00:51 |
nmz787 | did they just host opencv? | 00:51 |
nmz787 | or develop it too? | 00:52 |
nmz787 | that's all i've heard of them for, opencv downloads and docs | 00:52 |
@kanzure | they also did ROS and lots of robots | 00:56 |
brownies | openCV, ROS, PR2, and some other spinoffs | 00:58 |
-!- ezrios [~Ezrios@excarn.net] has joined ##hplusroadmap | 01:55 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 248 seconds] | 01:58 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 02:01 | |
-!- sylph_mako [~mako@103-9-42-1.flip.co.nz] has quit [Ping timeout: 248 seconds] | 02:38 | |
nmz787 | i don't know what ROS or PR2 is | 03:01 |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 03:06 | |
-!- superkuh [~superkuh@unaffiliated/superkuh] has joined ##hplusroadmap | 03:15 | |
-!- EnLilaSko [EnLilaSko@unaffiliated/enlilasko] has joined ##hplusroadmap | 03:43 | |
-!- lichen [~lichen@c-24-21-206-64.hsd1.or.comcast.net] has joined ##hplusroadmap | 03:44 | |
-!- JayDugger [~duggerjw@173.74.81.239] has left ##hplusroadmap [] | 04:43 | |
-!- joshcryer [~g@unaffiliated/joshcryer] has quit [] | 05:58 | |
-!- strophariad [~str@unaffiliated/strophariad] has quit [Read error: Connection reset by peer] | 06:47 | |
-!- strophariad [~str@unaffiliated/strophariad] has joined ##hplusroadmap | 06:47 | |
-!- archbox_ [~archie@unaffiliated/archbox] has quit [Quit: bye] | 07:04 | |
-!- strages_1ome [~strages@adsl-98-67-105-239.hsv.bellsouth.net] has joined ##hplusroadmap | 07:10 | |
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap | 07:11 | |
-!- strages_home [~strages@adsl-98-67-97-41.hsv.bellsouth.net] has quit [Ping timeout: 276 seconds] | 07:12 | |
-!- eudoxia [~eudoxia@r186-54-245-78.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 07:14 | |
eudoxia | oh shit there's a possible gunman at MIT | 07:24 |
ThomasEgi | who you gonna call.... ghost busters! | 07:27 |
ThomasEgi | or the pizza delivery service | 07:27 |
eudoxia | >Scene is clear. Call unfounded. No threat to public safety in #CambMA #MIT | 07:28 |
eudoxia | thank you cambridge police department twitter feed | 07:28 |
eudoxia | fuck i was already worried every kill would push the h+ future ~2 years into the future | 07:28 |
eudoxia | >Hearing reports of the shooter being taken down by a grad student with a "prototype plasma weapon". Improbable I know, but this is MIT after all. | 07:30 |
eudoxia | >At MIT? An ordinary firearm? Aha ha ha ha ha! Anything less that a photon laser will be vaporized by freshmen. | 07:30 |
eudoxia | hahahaha oh the comments | 07:30 |
-!- strages_1hop [~strages@38.110.18.223] has joined ##hplusroadmap | 07:37 | |
-!- strages_1hop is now known as strages_atl | 07:37 | |
-!- strages_1ome [~strages@adsl-98-67-105-239.hsv.bellsouth.net] has quit [Ping timeout: 252 seconds] | 07:55 | |
-!- strages_home [~strages@adsl-98-67-105-239.hsv.bellsouth.net] has joined ##hplusroadmap | 07:57 | |
-!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 08:03 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds] | 08:05 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 08:06 | |
-!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 256 seconds] | 08:10 | |
-!- eudoxia [~eudoxia@r186-54-245-78.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving] | 08:22 | |
-!- strages_atl [~strages@38.110.18.223] has quit [Ping timeout: 276 seconds] | 08:22 | |
-!- strages_atl [~strages@38.110.18.223] has joined ##hplusroadmap | 08:33 | |
-!- yashgaroth [~ffffff@cpe-66-27-118-94.san.res.rr.com] has joined ##hplusroadmap | 08:39 | |
-!- strages_atl [~strages@38.110.18.223] has quit [Ping timeout: 255 seconds] | 08:39 | |
-!- Mariu [Jimmy98@89.41.57.33] has quit [Quit: leaving] | 08:47 | |
-!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has joined ##hplusroadmap | 09:06 | |
-!- strophariad [~str@unaffiliated/strophariad] has left ##hplusroadmap [] | 09:42 | |
@kanzure | hrm. http://www.nirgoldshlager.com/2013/02/how-i-hacked-facebook-oauth-to-get-full.html | 09:47 |
brownies | kanzure: is that 0day? or did he responsibly disclose first? | 09:56 |
@kanzure | he disclosed it i think | 09:57 |
brownies | it's hard to tell. damn his broken english. | 09:59 |
-!- strages_atl [~strages@38.110.18.223] has joined ##hplusroadmap | 10:16 | |
-!- strages_atl [~strages@38.110.18.223] has quit [Ping timeout: 244 seconds] | 10:35 | |
-!- strages_atl [~strages@38.110.18.223] has joined ##hplusroadmap | 10:39 | |
-!- strages_atl [~strages@38.110.18.223] has quit [Ping timeout: 246 seconds] | 10:45 | |
-!- strages_atl [~strages@38.110.18.223] has joined ##hplusroadmap | 10:46 | |
@kanzure | paperbot: http://rspb.royalsocietypublishing.org/content/early/2010/03/04/rspb.2010.0001.short | 10:54 |
paperbot | no translator available, raw dump: http://diyhpl.us/~bryan/papers2/paperbot/Atmospheric%20oxygen%20level%20and%20the%20evolution%20of%20insect%20body%20size.pdf | 10:54 |
-!- padz [~not@bas5-kingston08-3096457487.dsl.bell.ca] has quit [Ping timeout: 246 seconds] | 10:56 | |
-!- strages_atl [~strages@38.110.18.223] has quit [Ping timeout: 244 seconds] | 11:04 | |
@kanzure | paperbot: http://www.springerlink.com/content/h348734770442854/ | 11:04 |
paperbot | no translator available, raw dump: http://diyhpl.us/~bryan/papers2/paperbot/395b02db5127aa6cf2fe78ae28a7d2d4.txt | 11:04 |
@kanzure | pfft. | 11:04 |
-!- padz [~not@69.158.79.132] has joined ##hplusroadmap | 11:06 | |
-!- augur [~augur@208.58.5.87] has quit [Remote host closed the connection] | 11:12 | |
-!- strages_atl [~strages@38.110.18.223] has joined ##hplusroadmap | 11:16 | |
-!- padz [~not@69.158.79.132] has quit [Read error: Connection reset by peer] | 11:19 | |
-!- padz [~not@69.158.79.132] has joined ##hplusroadmap | 11:20 | |
-!- Charlie_ [~quassel@64.31.59.70] has quit [Remote host closed the connection] | 11:25 | |
-!- Charlie [~quassel@64.31.59.70] has joined ##hplusroadmap | 11:26 | |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Read error: No route to host] | 11:30 | |
-!- Charlie [~quassel@64.31.59.70] has quit [Ping timeout: 252 seconds] | 11:31 | |
-!- strages_atl [~strages@38.110.18.223] has quit [Ping timeout: 244 seconds] | 11:33 | |
-!- augur [~augur@129-2-129-33.wireless.umd.edu] has joined ##hplusroadmap | 11:53 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 264 seconds] | 11:57 | |
-!- nsh [~nsh@wikipedia/nsh] has quit [Ping timeout: 276 seconds] | 11:58 | |
yashgaroth | so kanzure how would you feel about doing a prototype dna synthesizer with a liquid handling robot and multi-well plates instead of microfluidicz | 12:01 |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 12:04 | |
@kanzure | yashgaroth: pretty good | 12:04 |
yashgaroth | ok cool, because we'll have liquid handling robots and/or converted 3d printer assemblies at the lab, and 8000 wells isn't unforseeable with 384-well plates | 12:05 |
@kanzure | not sure how i feel about microwellarrayplates though. you can be a lot more sloppy with just 8x8 test tubes or something. | 12:05 |
yashgaroth | sure but when you've got thousands of wells...and even 3d printing heads have at least a millimeter of resolution | 12:06 |
@kanzure | accurate liquid handling on a microarray might be slightly more difficult, is my point. | 12:06 |
yashgaroth | oh not microarrays, these are well plates | 12:06 |
@kanzure | ok. | 12:07 |
yashgaroth | like dis http://www.thermoscientificbio.com/uploadedImages/Products/PCR_Related_Products/PCR_Plastics/ab-1384.jpg | 12:07 |
yashgaroth | I mean for super-initial testing of the enzymes and reactions, sure we can use a few ep tubes and mix by hand | 12:08 |
yashgaroth | but if we end up having to wait for microfluidics to catch up, this might work well enough | 12:08 |
@kanzure | so uh, i know this is a tough topic for you, but have you inventoried the liquid handlers you were given yet? | 12:10 |
yashgaroth | heh well we have two mendelmax's, and two real liquid handlers...I don't know how to assay them for functionality, and one seems to be a semi-homemade setup | 12:10 |
yashgaroth | I can't just go wander around the lab yet, since jojack has The One Key and is always traveling | 12:11 |
@kanzure | this is why you should learn lock picking | 12:11 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 12:11 | |
yashgaroth | the parking lot is where local police like to hang out for donut breaks though | 12:11 |
@kanzure | it would be nice if you could either get the model names and numbers, or any sort of barcodes or version numbers, from the bodies of the devices. soemtimes these are on the back, or inside a stupid panel. | 12:12 |
-!- Cat4D [182bc9d2@gateway/web/freenode/ip.24.43.201.210] has joined ##hplusroadmap | 12:12 | |
@kanzure | *sometimes | 12:12 |
yashgaroth | now that I'm on the 'equipment team' it will be my first priority in general, next time we have a meeting I'm doing that | 12:12 |
yashgaroth | it's not like the handler needs to be super fast or anything, each reaction will take a few minutes, so even sterilizing the head in between, all it needs is <1mm repeatable resolution | 12:13 |
yashgaroth | anyway we have jcline we can do anything | 12:14 |
@kanzure | he will try to convince you to use perl, you must resist this as much as possible | 12:15 |
@kanzure | nah not really, let him write it. | 12:15 |
yashgaroth | works for me | 12:16 |
@kanzure | http://dsp.stackexchange.com/questions/374/river-detection-in-text | 12:22 |
-!- Charlie [~quassel@64.31.59.70] has joined ##hplusroadmap | 12:26 | |
-!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has quit [Read error: Connection reset by peer] | 12:28 | |
-!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has joined ##hplusroadmap | 12:28 | |
-!- sheena2 [~home@174.4.134.159] has quit [Ping timeout: 252 seconds] | 12:30 | |
-!- augur [~augur@129-2-129-33.wireless.umd.edu] has quit [Remote host closed the connection] | 12:41 | |
-!- radivis [~quassel@pD9F8AD83.dip.t-dialin.net] has joined ##hplusroadmap | 13:10 | |
-!- Cat4D [182bc9d2@gateway/web/freenode/ip.24.43.201.210] has quit [Ping timeout: 245 seconds] | 13:13 | |
-!- Grognor [~broken_co@174.141.212.1] has joined ##hplusroadmap | 13:14 | |
@kanzure | Grognor: hi | 13:22 |
-!- sylph_mako [~mako@103-9-42-31.flip.co.nz] has joined ##hplusroadmap | 13:23 | |
Grognor | I perceive that this channel has an ethos of doing things rather than just writing all the time, but I don't know if there is anything I can do that is actually useful | 13:24 |
Grognor | so I'm sort of requesting suggestions | 13:25 |
yashgaroth | what are you good at | 13:27 |
Grognor | my resources are: lots of time, that's about it. I'm homeless and jobless at the moment, so I don't really have money or | 13:27 |
Grognor | I don't have very many skills | 13:27 |
Grognor | http://www.newgrounds.com/portal/view/594240 this is the closest thing I can come to impressiveness, currently | 13:27 |
Grognor | it'd be nice to know if there is a skill I can acquire from a (series of) textbook(s) that would inevitably be useful enough to get a job, or not, I don't think I necessarily need a job, but something, you know? | 13:28 |
@kanzure | do you have a computer? | 13:28 |
Grognor | yes | 13:28 |
Grognor | I haven't lurked/logspotted enough to know this channel's consensus opinion on pirating but I at least know about libgen.info | 13:29 |
@kanzure | write software. | 13:29 |
Grognor | I guess that's the default option for people who use irc, but it's not available to me | 13:31 |
@kanzure | because you don't have a computer? | 13:32 |
Grognor | because my brain has no innate programming skill module and refuses to install one and I know you're not going to believe me and you're probably going to say something about how I haven't tried hard enough but for god's sake don't because I've had that conversation a thousand times and it really gets on my nerves by now | 13:33 |
@kanzure | no i happen to not believe in trying hard enough | 13:33 |
@kanzure | because http://archive.autistics.org/library/inertia.html | 13:34 |
Grognor | it's not a bad suggestion inherently, just inapplicable to me. i'd appreciate others, even other obvious ones because I have not thought of any at all | 13:36 |
jrayhawk | if it helps, i despise writing software and thus resort to supporting roles for programmers | 13:38 |
jrayhawk | Quantified Self-type stuff is also reasonably productive. | 13:38 |
Grognor | maybe I should use that excuse in the future | 13:38 |
jrayhawk | Hypothesis formation/experimentation/publishing a la Seth Roberts and friends. | 13:39 |
Grognor | I signed up for the first round of "Soylent" trials | 13:39 |
jrayhawk | oh neat | 13:39 |
Grognor | things like drugs and Zeos cost money, which isn't really one of my available resources | 13:39 |
Grognor | that's worth thinking about anyway, though, I think | 13:40 |
nmz787 | Zeos? | 13:40 |
nmz787 | kanzure: i can't find a definition for the term bsearching | 13:41 |
nmz787 | brute force searching? | 13:41 |
@kanzure | probably means binary searching | 13:41 |
nmz787 | it was in context of hacing | 13:41 |
nmz787 | hacking* | 13:41 |
nmz787 | ahh | 13:41 |
@kanzure | nobody calls it bsearching | 13:41 |
jrayhawk | kanzure: what other support roles besides hardware/networking/system administration are useful to programmers? | 13:41 |
jrayhawk | Bug triage/knowledge base/community management stuff, maybe. | 13:42 |
Grognor | the thing that measures your brain while you sleep http://www.myzeo.com/sleep/ | 13:42 |
jrayhawk | huh, neat | 13:42 |
@kanzure | jrayhawk: photoshop wizards, or the better kind that can figure out how to convert psds to svg | 13:43 |
Grognor | gwern's drug experiments used them, usually | 13:43 |
@kanzure | jrayhawk: cad modeling stuff. there's a lot of stuff that just doesn't have brlcad/opencascade/openscad/implicitcad representations yet. | 13:43 |
nmz787 | kanzure: what are ACs in that article? | 13:43 |
@kanzure | nmz787: which article? | 13:44 |
nmz787 | the inertia | 13:44 |
@kanzure | autistic something, i'm sure. | 13:44 |
jrayhawk | Seems like we should have a list of these sorts of skills somewhere for people who accuse you of being hostile towards non-programmers. | 13:44 |
nmz787 | Grognor: what/who is gwern? | 13:45 |
jrayhawk | Oh yeah, statistics and signal analysis are also useful, but those are bordering on programming skills. | 13:45 |
@kanzure | gwern is just some dude that hangs out in #lesswrong posting quotes from blogs | 13:45 |
@kanzure | he also likes anime and eliezer fanfic or something. | 13:45 |
@kanzure | ladies, he's available. | 13:45 |
@kanzure | jrayhawk: another thing that is useful is consolidating information in http://diyhpl.us/~bryan/papers2/ | 13:46 |
Grognor | he writes extremely long essays that quantitatively analyze topics that he figured wouldn't pass the notability test on wikipedia and also he is a fucking asshole who I'd probably kill given the chance | 13:46 |
klafka | LOL | 13:47 |
klafka | math and statistics | 13:47 |
klafka | biotech | 13:47 |
@kanzure | if anything the reason to kill gwern is because of his refusal to do anything but write, not because of any asshole trait. | 13:47 |
Grognor | he does self-experimentation, which jrayhawk just said is a useful thing | 13:47 |
@kanzure | i suppose he pops a lot of pills, which is not writing. but those pills don't seem to cause much except more writing? | 13:47 |
klafka | self experimentation of what? | 13:48 |
@kanzure | nootropics. same old same old. | 13:48 |
Grognor | but unlike most people he had the mathematical chutzpah to do statistics on the data he acquired from subjective assessments and zeo stuff | 13:49 |
@kanzure | jrayhawk: i think cad stuff is a pretty good answer. | 13:49 |
Grognor | and managed to double-blind himself | 13:49 |
@kanzure | chutzpah is not how math works | 13:49 |
klafka | but um 1 person isn't a big enough sample size | 13:49 |
klafka | also you introduce new variables when you 'double blind' yourself | 13:49 |
@kanzure | klafka: "n=1 but it's the only 1 that matters" haha. reading his article is probably better if you want to criticize his math, since you can look at his claims directly. | 13:50 |
klafka | LOL | 13:50 |
klafka | i have enough things to waste time on | 13:50 |
klafka | thanks | 13:50 |
nmz787 | CAD is pretty artsy | 13:57 |
-!- Grognor_ [~broken_co@174.141.212.82] has joined ##hplusroadmap | 14:05 | |
-!- Grognor [~broken_co@174.141.212.1] has quit [Ping timeout: 246 seconds] | 14:06 | |
-!- Grognor_ is now known as Grognor | 14:06 | |
jrayhawk | Testing hypotheses has to start somewhere. | 14:07 |
@kanzure | there are some hypotheses that are easier to test than others | 14:09 |
@kanzure | "consumption of whatchamafuckit will increase my blood pressure" is pretty easy to test. | 14:09 |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 255 seconds] | 14:10 | |
-!- Urchin[emacs] [~user@unaffiliated/urchin] has joined ##hplusroadmap | 14:11 | |
EnLilaSko | paperbot: http://www.nature.com.sci-hub.org/nrn/journal/v11/n7/full/nrn2867.html | 14:12 |
paperbot | http://diyhpl.us/~bryan/papers2/paperbot/myescience.pdf | 14:12 |
@kanzure | paperbot: http://www.nature.com/nrn/journal/v11/n7/full/nrn2867.html | 14:12 |
paperbot | HTTP 401 unauthorized http://www.nature.com/nrn/journal/v11/n7/pdf/nrn2867.pdf | 14:12 |
EnLilaSko | I love you for creating paperbot kanzure | 14:13 |
@kanzure | sadly we don't have access to this paper | 14:13 |
EnLilaSko | Meh, still an awesome bot | 14:14 |
jrayhawk | klafka: The important part is not the generalizability of the result, but the verifiability/falsifiability of the result. Suboptimal data is still data; we don't need to wait for perfect data to learn something. | 14:16 |
klafka | you can't reject a null hypothesis without a statistically significant sample | 14:18 |
klafka | not without the confidence interval being huge | 14:18 |
Grognor | even a shitty dataless experiment with results like "I tried it and it made me feel better" can at least point to things worth trying | 14:19 |
klafka | so you're saying that anecdote can drive experimentation | 14:20 |
klafka | sure | 14:20 |
jrayhawk | My life has been vastly improved by N=1 experiments. Empiricism is not stochastic. | 14:23 |
@kanzure | http://diyhpl.us/~bryan/papers2/paperbot/The%20unreasonable%20effectiveness%20of%20my%20self-experimentation.pdf | 14:24 |
Grognor | I'm really saying "If you don't have maximum statistical rigor anything you do is useless and should be rejected as evidence" is bullshit | 14:25 |
jrayhawk | Or, rather, my life has been vastly improved by reports of N=1 experiments from others, N=1 experiments which I was able to reproduce thanks to their weak, subjective, "not big enough" data. | 14:28 |
-!- st0461 [46248632@gateway/web/freenode/ip.70.36.134.50] has joined ##hplusroadmap | 14:31 | |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving] | 14:34 | |
@kanzure | jrayhawk: i'm not convinced that this article is really the source of knowledge that breakfast cereals cause early awakenings.. surely we knew that earlier? | 14:37 |
@kanzure | hrm this article might just be a marketing stunt for http://sethroberts.net/science/ | 14:39 |
EnLilaSko | kanzure: So his theory is kinda that different flavours and what you think of the food affects the set-point? | 14:43 |
@kanzure | seems so; i don't know what biological mechanism he is proposing though. | 14:43 |
EnLilaSko | Iirc, ghrelin and leptin plays a fairly big role in the set-point theory, you can influence ghrelin by just thinking | 14:44 |
EnLilaSko | So he might be partially right | 14:44 |
jrayhawk | There's also cortisol-cycle-dependent adaptations to insulin spikes. | 14:47 |
@kanzure | a hemostatis-related explanation would seem more plausible to me | 14:47 |
@kanzure | erm, wait. | 14:48 |
@kanzure | that's very much not the word i wanted to use.. | 14:48 |
EnLilaSko | Homeostasis maybe? | 14:48 |
@kanzure | hey that one looks much better | 14:48 |
EnLilaSko | Lyle McDonald suggests people to stay at the bodyfat they "want" for as long as possible with refeeds once a week or so | 14:49 |
EnLilaSko | And Martin Berkhan suggests IF (very slowly) with some refeeds | 14:50 |
jrayhawk | There are also also metabolic adaptations to carb starvation, which I suppose is probably more important, here. | 14:51 |
@kanzure | surely there was some carb starvation sleep wakening theories out a long long time ago? | 14:51 |
@kanzure | Bed-time food supplements and sleep: effects of different carbohydrate levels (1981) | 14:52 |
@kanzure | Inhibition of glycolytic metabolism and sleep-waking states in cats (1973) | 14:53 |
-!- radivis [~quassel@pD9F8AD83.dip.t-dialin.net] has quit [Remote host closed the connection] | 14:53 | |
@kanzure | "In cats, injections of 2-deoxy-D-glucose, a glucose antimetabolite, produces dose-dependent increases of slow wave sleep and decreases of REM sleep. Accordingly, variations of glycolytic metabolism may participate in the control of sleep-waking behavior." | 14:53 |
EnLilaSko | So does that mean carbs at night = good or bad for sleep? | 14:54 |
jrayhawk | Depends on the context. | 14:55 |
jrayhawk | Sleep can be maintained while melatonin dominates blood sugar, but you can also panic the body by having *too little* blood sugar and wake up at 3AM ravenous. | 14:56 |
jrayhawk | Similarly, you may produce too little melatonin to safely eat carbs and still attempt to sleep. | 14:57 |
EnLilaSko | So it depends on the individuals genetics and such (melatonin production and such)? | 14:58 |
jrayhawk | Yeah, though 'genetics' is kindof a loaded term and I don't want it to carry implications of biological determinism, here. | 14:58 |
-!- pwang_ [~pwang@unaffiliated/parahsailin] has joined ##hplusroadmap | 15:00 | |
jrayhawk | The cortisol cycle is supposed to wake you up in the morning through gluconeogenesis, but it's dependant on timing cues and hormone profiles that might be completely wrong, and meal timing and composition does seem to have some effect on that, but it's hard to make generalizable conclusions about due to all the variable context. | 15:01 |
-!- ParahSail1n [~pwang@unaffiliated/parahsailin] has quit [Ping timeout: 276 seconds] | 15:04 | |
EnLilaSko | I just realised that I kinda like this chat a lot, no idea why I usually skip reading it | 15:04 |
-!- pwang__ [~pwang@unaffiliated/parahsailin] has joined ##hplusroadmap | 15:05 | |
jrayhawk | All the context the reason rigorous n=1 experiments are so important; giving you the tools to reproduce good results is brilliant even if you can only get positive results out of ten percent of them. | 15:06 |
jrayhawk | "Hey, that guy's problems sound just like mine!" | 15:06 |
-!- pwang_ [~pwang@unaffiliated/parahsailin] has quit [Ping timeout: 264 seconds] | 15:09 | |
-!- OldCoder_ [~OldCoder_@c-69-181-140-134.hsd1.ca.comcast.net] has quit [Remote host closed the connection] | 15:11 | |
klafka | jrayhawk: that's fine but it's not science | 15:13 |
jrayhawk | haha what? | 15:14 |
klafka | or rather quantifiable science | 15:14 |
klafka | which would be a hypothesis test of a falsifiable hypothesis based on empirical data | 15:15 |
klafka | I mean - I guess you could do it but your margin of error would be rather drastic | 15:15 |
jrayhawk | Empiricism is not stochastic. | 15:16 |
EnLilaSko | Fuck the brain is too complex to write about, lol | 15:17 |
Grognor | I guess in most vernaculars the word "science" still means exactly what it did in Popper's day | 15:20 |
-!- docl [~Freenode@unaffiliated/docl] has joined ##hplusroadmap | 15:24 | |
-!- eudoxia [~eudoxia@r186-52-166-62.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 15:35 | |
-!- Inurdaes [ba34a63e@gateway/web/freenode/ip.186.52.166.62] has joined ##hplusroadmap | 15:36 | |
-!- Inurdaes [ba34a63e@gateway/web/freenode/ip.186.52.166.62] has quit [Client Quit] | 15:36 | |
-!- docl [~Freenode@unaffiliated/docl] has quit [Remote host closed the connection] | 15:41 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 244 seconds] | 15:45 | |
-!- st0461 [46248632@gateway/web/freenode/ip.70.36.134.50] has left ##hplusroadmap [] | 15:52 | |
-!- augur [~augur@c-68-55-204-212.hsd1.md.comcast.net] has joined ##hplusroadmap | 15:53 | |
-!- EnLilaSko [EnLilaSko@unaffiliated/enlilasko] has quit [Quit: - nbs-irc 2.39 - www.nbs-irc.net -] | 16:01 | |
-!- Grognor [~broken_co@174.141.212.82] has left ##hplusroadmap [] | 16:04 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 16:07 | |
-!- eudoxia [~eudoxia@r186-52-166-62.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving] | 16:17 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 252 seconds] | 16:57 | |
-!- augur [~augur@c-68-55-204-212.hsd1.md.comcast.net] has quit [Remote host closed the connection] | 16:59 | |
-!- docl [~docl@unaffiliated/docl] has joined ##hplusroadmap | 17:02 | |
-!- OldCoder_ [~OldCoder_@c-69-181-140-134.hsd1.ca.comcast.net] has joined ##hplusroadmap | 17:04 | |
-!- augur [~augur@c-68-55-204-212.hsd1.md.comcast.net] has joined ##hplusroadmap | 17:08 | |
-!- augur [~augur@c-68-55-204-212.hsd1.md.comcast.net] has quit [Remote host closed the connection] | 17:09 | |
docl | I so want a better way to input text on my android phone. You'd think by now there would be ways to use the camera to type on a superimposed virtual keyboard or some such. | 17:10 |
docl | Heck, with the front facing camera, there could be an eye tracker. | 17:12 |
@kanzure | docl: https://play.google.com/store/apps/details?id=com.volosyukivan | 17:13 |
@kanzure | docl: or just use a bluetooth keyboard. | 17:13 |
ThomasEgi | eye tracking? with a vga-camera from half a meter distance? | 17:13 |
ThomasEgi | in a shaky hand? | 17:13 |
docl | Hmm, good point | 17:15 |
docl | Might need an actual eye accessory for that idea to work. | 17:15 |
@kanzure | another way is that you can send yourself an sms from twilio | 17:25 |
@kanzure | or load a web page with the content you want | 17:25 |
@kanzure | wifi keyboard works pretty well | 17:25 |
docl | Huh? It seems to require another computer entirely. | 17:27 |
docl | The whole point of using the Droid is to get away from that. | 17:29 |
ThomasEgi | how bout tapping in morse. utilizing the accelerometer sensordata | 17:30 |
ThomasEgi | you could just tap against the back of your phone that way | 17:30 |
docl | Maybe. Any apps for that? | 17:31 |
ThomasEgi | i only know that there's one that can guess your phone pin/gesture this way | 17:31 |
ThomasEgi | guess it shouldnt be too hard to write one | 17:32 |
ThomasEgi | which gives roughly 40% hitratio. so guess morsing should be pretty reliable that way | 17:33 |
@kanzure | https://play.google.com/store/search?q=morse+code+input | 17:36 |
@kanzure | docl: i don't see what's wrong with a keyboard. | 17:40 |
@fenn | docl: you could learn stenography if your phone supports truhttp://www.newgrounds.com/portal/view/594240e multitouch | 17:40 |
@fenn | gah | 17:40 |
@kanzure | docl: also, there's somme really cheap kyocera phones on ebay that have usable keyboards attached. | 17:41 |
@kanzure | *some | 17:41 |
@kanzure | http://www.ebay.com/sch/Cell-Phones-Smartphones-/9355/i.html?_from=R40&_nkw=Kyocera+Rise+C5155 | 17:42 |
nmz787 | docl: I think kanzure sent me that same link a while ago when i bitched about android input! | 17:42 |
nmz787 | I was confused | 17:42 |
@kanzure | wow their price has increased. they were closer to $50 a while ago. | 17:42 |
@kanzure | nmz787: what's confusing about it.. it works very well. | 17:42 |
nmz787 | i would actually like better audio interfacing | 17:42 |
@fenn | docl: also dasher is interesting, and there's an android input whatsit for it already | 17:42 |
nmz787 | i hate texting while driving | 17:42 |
nmz787 | but i get bored in the car | 17:43 |
nmz787 | or i have a question about somethign i see | 17:43 |
nmz787 | kanzure: confusing because it wasn't helpful for the situation | 17:43 |
@kanzure | fenn: $200 myvu? http://www.ebay.com/itm/Myvu-iPod-Personal-Media-Viewer-/270763945293?pt=LH_DefaultDomain_0&hash=item3f0ac9ed4d | 17:44 |
nmz787 | res? | 17:44 |
@fenn | you probably want the "universal" not the ipod connector | 17:44 |
docl | Hmm. Morse code could be good. There are several touch-based ones at least. No punctuation though. | 17:45 |
@kanzure | you guys really really hate keyboards | 17:45 |
@fenn | morse code is ... slow | 17:45 |
@kanzure | i have never done anything productive with dasher. sometimes on dasher i like to pretend i am typing something that i wont immediately delete, but it's always a lie. | 17:47 |
@fenn | you've summed up the last five years of my life | 17:47 |
docl | Keyboards are horribly non-portable, usually. | 17:48 |
nmz787 | i need HUD + eye tracking and better voice I/O | 17:48 |
@kanzure | i used to strap a keyboard to my pants, i wanted a rectractable drawstring though.. | 17:48 |
nmz787 | then a few simple click scroll buttons would be enough | 17:48 |
@fenn | how about a nintendo power glove | 17:49 |
docl | Glove could be good. | 17:49 |
@kanzure | fenn: what happened to your myvu? | 17:49 |
nmz787 | how are those neurosky things for simple XY locating? | 17:49 |
ThomasEgi | i _love_ my keyboard.. what i hate are the commonly used layouts :) | 17:49 |
@fenn | it's in storage in sf, and the battery needs fixing | 17:49 |
@kanzure | neurosky is overhyped. so is emotiv epoc. they just bring you lots of disappointment. | 17:49 |
nmz787 | like are there drawing or item selection apps for those things? | 17:50 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 17:50 | |
@kanzure | emotiv epoc has a proprietary api and you're supposed to submit your software to the company to get it signed or something, and then you can use it locally. | 17:50 |
@kanzure | daeken, myself and qdot (teledildonics extraordinaire) somehow managed to figure out most of the protocol for that device. | 17:51 |
@kanzure | there were like <10 applications on emotiv's site when i last checked.. | 17:52 |
@kanzure | http://emotiv.com/store/apps/applications/ | 17:52 |
@kanzure | oh brother "EmoLens automatically indexes Flickr photos by emotions so you can search for them later not only with keywords but by your feelings too." | 17:53 |
@kanzure | great now instead of only worrying about the cosmic implications of which tag i choose i have to worry about the implications of my feelings about each item. | 17:53 |
nmz787 | are there clear CMOS or CCD sensors? | 17:56 |
ThomasEgi | what do you mean with "clear" ? | 17:56 |
ThomasEgi | transparent? | 17:56 |
nmz787 | see-through | 17:56 |
nmz787 | aren't there see-through LCDs? | 17:57 |
nmz787 | for doing HUD | 17:57 |
nmz787 | or is that my imagination | 17:57 |
ThomasEgi | there are see through lcd's | 17:57 |
ThomasEgi | but no ccd or cmos | 17:57 |
nmz787 | is google glass supposed to do eye tracking? | 17:58 |
@fenn | what would be the point of an image sensor that lets the light through, it would have 0% sensitivity | 17:58 |
ThomasEgi | pretty much true | 17:58 |
nmz787 | it could be neutral density | 17:58 |
ThomasEgi | if you need semi-transparent image recording you are best off with a conventional cmos and a half-mirror | 17:58 |
@fenn | glass seems to be driven by speech recognition rather than direct manipulation or eye tracking | 17:59 |
@fenn | it might do hand tracking | 17:59 |
nmz787 | hmm | 17:59 |
nmz787 | i wonder if glass might be good for driving | 17:59 |
ThomasEgi | for eye tracking you don't need a transparent cmos chip either | 17:59 |
ThomasEgi | you can just mount a tiny camera module out of the fov and still track the eye | 17:59 |
@fenn | also you can track infrared or use an infrared splitter | 18:00 |
ThomasEgi | also possible. my guess would be a separate camera is cheaper | 18:00 |
@fenn | certainly | 18:01 |
@fenn | EOG is interesting too | 18:01 |
@fenn | http://en.wikipedia.org/wiki/Electrooculography | 18:02 |
ThomasEgi | yeah. but requires a setup more difficult than just putting on regular glasses | 18:02 |
@fenn | might be as simple as four dry electrodes | 18:02 |
@fenn | out of my realm of expertise i'm afraid | 18:02 |
docl | Brow tracking or something like that might also work. | 18:03 |
ThomasEgi | EEG stuff usualy requires well made circuitry, lots of shielding , reference electrodes etc.. | 18:03 |
@fenn | what you do is take your homunculus and put it in a jar full of levers and buttons and whatnot | 18:03 |
ThomasEgi | i'd rather suggest something like a one-hand-keyboard that fits into your pocket and can be cmofortably hold. | 18:05 |
ThomasEgi | with 2 rows of button for each finger. | 18:05 |
ThomasEgi | with a bit of binary inputs you get 2x5 bit | 18:05 |
@fenn | it's called a twiddler | 18:06 |
@fenn | or perky | 18:06 |
ThomasEgi | hm.. yeah something like that.. just a lot less bricky | 18:06 |
yashgaroth | what about a chorded keyboard | 18:06 |
ThomasEgi | somethin thats actually comfortable to use | 18:06 |
@fenn | http://homepage2.nifty.com/perky/belt.htm | 18:08 |
ThomasEgi | that's already a lot closer to what i had in mind | 18:09 |
@fenn | http://homepage2.nifty.com/perky/grip.htm | 18:11 |
ThomasEgi | if you'd have a touch sensitive backside on your mobile. you could use that to lay out a keyboard too. | 18:11 |
ThomasEgi | totoroki! :D that's like.. the most awesome thing following pizza | 18:12 |
@fenn | i didnt really get how it would work | 18:13 |
@fenn | i guess it could be a gyro mouse | 18:14 |
-!- Juul [~Juul@c-76-21-28-218.hsd1.ca.comcast.net] has joined ##hplusroadmap | 18:20 | |
docl | http://m.io9.com/5985558/temporary-tattoos-could-make-electronic-telepathy-and-telekinesis-possible | 18:27 |
@kanzure | please don't read io9.com | 18:28 |
klafka | http://onlinelibrary.wiley.com/doi/10.1111/j.1476-5381.2012.01998.x/abstract;jsessionid=99E2123EEB63B21F6A823B616C100A9E.d02t02 | 18:29 |
klafka | paperbot: http://onlinelibrary.wiley.com/doi/10.1111/j.1476-5381.2012.01998.x/abstract;jsessionid=99E2123EEB63B21F6A823B616C100A9E.d02t02 | 18:29 |
docl | Sorry. | 18:29 |
paperbot | no translator available, raw dump: http://diyhpl.us/~bryan/papers2/paperbot/799549f98745afcfa432442402c4b881.txt | 18:29 |
klafka | dammit | 18:29 |
klafka | is there something better i can do to make paperbot work kanzure? | 18:29 |
@kanzure | in this case the problem is that paperbot does not have access to http://onlinelibrary.wiley.com/doi/10.1111/j.1476-5381.2012.01998.x/pdf | 18:30 |
@kanzure | https://github.com/kanzure/paperbot | 18:30 |
klafka | aah | 18:30 |
@kanzure | the python parts of paperbot need to be rewritten. that would be very helpful. | 18:30 |
klafka | does anyone in chan have access? | 18:31 |
@kanzure | i've been making incremental changes but they are piling up and will become unmaintainable. | 18:31 |
klafka | aah i see | 18:31 |
klafka | i don't really have time to undertake that - i owe people two mobile apps already | 18:31 |
klafka | :( | 18:31 |
klafka | also a mobile gaming ad algorithm | 18:31 |
@kanzure | slacking is the best way to get stuff done. | 18:32 |
klafka | right now i'm catching up on my dancesafe stuff | 18:32 |
klafka | since i run the bay area chapter >_> | 18:32 |
klafka | paperbot: http://link.springer.com/article/10.1007%2Fs13181-010-0018-5 | 18:33 |
paperbot | error: HTTP 500 http://diyhpl.us/~bryan/papers2/paperbot/Recreational%20Use%20of%20Mephedrone%20%284-Methylmethcathinone%2C%204-MMC%29%20with%20Associated%20Sympathomimetic%20Toxicity.pdf | 18:33 |
klafka | interesting 500 error | 18:34 |
@kanzure | the 500 error is from translation-server. | 18:34 |
@kanzure | https://github.com/kanzure/paperbot#readme | 18:34 |
-!- jmil [~jmil@hive76/member/jmil] has joined ##hplusroadmap | 18:47 | |
@kanzure | jmil: howdy | 18:48 |
jmil | o/ kanzure | 18:49 |
jmil | how goes it | 18:49 |
nmz787 | klafka: interesting, do you process random samples? | 18:49 |
nmz787 | like on hplc or do you have some 'friend' who 'tests' stuff? | 18:50 |
klafka | umm we have reagent tests that we can do on random samples | 18:50 |
nmz787 | jmil: long time! | 18:50 |
klafka | we or you can also send samples to ecstasydata.org and they will gc/ms them for a fee | 18:50 |
jmil | o/ nmz787 | 18:50 |
jmil | i been traveling | 18:50 |
jmil | has been awesome | 18:50 |
jmil | tryin to launch a career | 18:50 |
jmil | :D | 18:50 |
nmz787 | klafka: who tests their stuff? who is a good enough dealer to do that? | 18:50 |
klafka | a number test it via reagent tests | 18:51 |
klafka | which are obviously not perfect | 18:51 |
nmz787 | jmil: cool | 18:51 |
klafka | but better than not | 18:51 |
nmz787 | i thought they were all about profit | 18:51 |
klafka | nmz787: have you known many drug dealers? | 18:51 |
nmz787 | jmil: where do you want to take your career? | 18:51 |
nmz787 | klafka: I guess not | 18:51 |
jmil | nmz787: want to be a professor with a research lab. will have a whole wing for repraps :D | 18:52 |
nmz787 | jmil: so you're looking for a tenured prof postions? | 18:52 |
klafka | well I have known a lot at various levels, and a number do very much care about what they are giving to people both because they don't want their shit to cause someone to od/freak out /die and also beacuse people want to buy the good shit and also because they do this because they believe in the drugs that they sell | 18:52 |
nmz787 | or do you still need to do 85 years of post doc frist? | 18:53 |
nmz787 | :P | 18:53 |
Juul | 99 years of post doc on the wall | 18:54 |
Juul | hey klafka do you or anyone you know have small satellite dishes (like directv) they don't need? | 18:55 |
Juul | we're looking for a bunch for the east bay mesh | 18:55 |
nmz787 | Juul: will this support a predominantly yuppie community or more likely an impoverished community? | 18:57 |
@kanzure | nmz787: why would that matter. | 18:58 |
@kanzure | klafka: also you can test samples on science exchange i think. | 18:58 |
klafka | of potentially illegal substances | 18:59 |
Juul | nmz787, we're definitely trying to interface with a diverse community. | 18:59 |
Juul | most of the active nodes are in east oakland right now | 18:59 |
klafka | Juul: i don't have any | 19:00 |
klafka | i'll post in my loft facebook group | 19:00 |
Juul | klafka, nice | 19:00 |
nmz787 | kanzure: I'd be more motivated to help bring access to a community that was otherwise internet-limited | 19:01 |
nmz787 | kanzure: others share similar feelings, if I came across someone with a bunch of dishes, it's a question they might ask | 19:01 |
klafka | ok Juul | 19:02 |
klafka | also nmz787 if most of the nodes are in east oakland - impoverished community | 19:02 |
klafka | or lower income | 19:02 |
nmz787 | yeah that's kinda what i was thinking | 19:02 |
Juul | nmz787, we're interested in using it to create a resilient (censorship / disaster) network for mapping out local resources and communicating spatially relevant information in a decentralized way | 19:02 |
nmz787 | Juul: what are the community tie-ins? | 19:02 |
nmz787 | are you advertising it? | 19:02 |
nmz787 | like 'get mesh in your community by voting here ' | 19:03 |
nmz787 | or in schools or something? | 19:03 |
jmil | nmz787: yes tenure track faculty position | 19:03 |
nmz787 | jmil: RIT in Rochester has a pretty new biomed dept/college | 19:03 |
Juul | nmz787, we're just beginning to set it up, and we're mostly trying to gain a sense of how well the different kinds of technology perform, both software and hardware, and what kind of needs and interests the local communities have | 19:03 |
jmil | nmz787: cool. they weren't hiring this year :) | 19:04 |
nmz787 | jmil: they've got a lot of $$$ that they do stupid things with, so someone with brains and a title could clean up there, or be pretty comfy | 19:04 |
jmil | aww | 19:04 |
klafka | jmil: are you a graduating PHD | 19:04 |
nmz787 | Juul: cool | 19:04 |
klafka | nmz787: oh man RIT really doesn't hire that many people unfortunately | 19:05 |
nmz787 | klafka: he was just a post doc | 19:05 |
klafka | because they have no research budget | 19:05 |
nmz787 | at upenn | 19:05 |
klafka | oh | 19:05 |
Juul | so finding rooftops where we can mount antennas and mapping out obstructions and interference. testing software (such as TidePools) and developing some of our own. we have a meetup at the sudo room hackerspace every thursday evening. | 19:05 |
nmz787 | klafka: not true re: biomed engi | 19:05 |
klafka | how much grant money do they have? | 19:05 |
nmz787 | klafka: they are just completing a brand new building for biomed engi | 19:06 |
klafka | right but that was donated and via a grant from the bioX fund right? | 19:06 |
nmz787 | klafka: dunno | 19:06 |
klafka | that's grants for building | 19:06 |
klafka | not for research | 19:06 |
nmz787 | engi has a lot more $$$ than sciences there though | 19:06 |
jmil | klafka: finishing a postdoc | 19:06 |
Juul | it looks like the directv-style satellite dishes with the receiver heads replaced by weatherproofed usb wifi donges are the best bang for the buck for high-gain outdoor directional antennas | 19:06 |
klafka | nmz787: really though imaging science has the most | 19:06 |
klafka | honestly | 19:06 |
klafka | per professor | 19:07 |
nmz787 | heh, yeah that's tru | 19:07 |
nmz787 | i worked there with a prof for a while | 19:07 |
nmz787 | he had a nice gig | 19:07 |
nmz787 | building remote sensors and going out to the wilds to plant them | 19:07 |
klafka | ah cool | 19:08 |
klafka | i knew a phd student with the lidar professor | 19:08 |
nmz787 | building crazy spectral imagers to fly over forest fires | 19:08 |
klafka | ah neat | 19:08 |
klafka | but remember the engineering building had no phd programs | 19:08 |
klafka | except microe | 19:08 |
nmz787 | yeah that sounds about right | 19:08 |
klafka | so i mean they can onyl do so much research | 19:08 |
nmz787 | there's a sustainability research prog | 19:08 |
klafka | that's a separate program from eng | 19:09 |
klafka | iirc | 19:09 |
nmz787 | i mean phd | 19:09 |
nmz787 | yeah | 19:09 |
nmz787 | in CIMS | 19:09 |
klafka | they are actually in a building in CIMS | 19:09 |
nmz787 | Center ... Manufacturing... i think | 19:09 |
klafka | my advisor went to a couple talks for people we were interviewing for it | 19:09 |
klafka | they were pretty crazy in a good way | 19:09 |
nmz787 | i used a machine shop there a few times | 19:09 |
nmz787 | it was nice to just walk into places like you owned them, flashing the student badge around | 19:09 |
nmz787 | the microe stuff had some red tape to get into but I ended up in there a bunch | 19:10 |
@kanzure | which one of you knew zacharycohn | 19:11 |
-!- Lucas_ [4a6fa682@gateway/web/freenode/ip.74.111.166.130] has joined ##hplusroadmap | 19:11 | |
nmz787 | me | 19:12 |
nmz787 | worked with him in tech support at RIT | 19:12 |
ThomasEgi | decentrelaised networking infrastructure? | 19:32 |
ThomasEgi | or what was that a bit ago | 19:32 |
klafka | yeah mesh networks in oakland | 19:36 |
ThomasEgi | http://twibright.ronja.com/ not the fastest thing in the world. but it's a technology the user has controll over. and unlike wifi it's ways more resistant to jamming/frequency overuse | 19:40 |
ThomasEgi | oh wait | 19:41 |
ThomasEgi | http://ronja.twibright.com/ | 19:41 |
ThomasEgi | other way round :D | 19:41 |
-!- Viper168 [~Viper@node236.19.251.72.1dial.com] has joined ##hplusroadmap | 19:55 | |
-!- Viper168 [~Viper@node236.19.251.72.1dial.com] has quit [Changing host] | 19:55 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 19:55 | |
Juul | ThomasEgi, yeah i think we'll eventually try one of those links to see how it compares in rainy / foggy conditions | 20:02 |
ThomasEgi | rain is no problem | 20:03 |
ThomasEgi | but with fog.. you get a problem | 20:03 |
Juul | also looking at low bandwidth long range fallback links using CB radio, DECT, homebuilt transmitters + hdtv tuners as SDR receivers, softmodem over walkie-talkie, etc. | 20:04 |
ThomasEgi | always depends on your requirements | 20:13 |
ThomasEgi | low bandwith optical communication is rather easy | 20:13 |
ThomasEgi | and.. how long, "long range" actually is | 20:14 |
Juul | one of the problems we foresee with line of sight directional links is potential misalignment after an earthquake | 20:17 |
Juul | we're going to do some research and testing on how much misalignment is likely to happen and how much the different technologies can tolerate | 20:18 |
ThomasEgi | it's not like you need tons of research | 20:18 |
ThomasEgi | each antenna/transmitter/receiver has a nice diagrams showing reception gain for certain angles | 20:19 |
ThomasEgi | the wider your angle, the more robust against misalignment | 20:19 |
ThomasEgi | and the wider the angle, the less gain you have, thus the shorter the range | 20:20 |
Juul | yes, but theory and real world conditions are different, and we don't necessarily have good info on our satellite dish + wifi usb stick solution yet | 20:20 |
ThomasEgi | mostly depends how well you can position your original wifi stick in the focal point | 20:21 |
Juul | and how much we shield the receiver from off-focal-point interference and how much that shielding causes the transmitted signal to bounce around and interfere with itself before it hits the dish | 20:23 |
Juul | we're going to try a bunch of different things and see what gives the best signal | 20:23 |
ThomasEgi | dishes usualy have a pretty good directional characteristics | 20:24 |
Juul | yeah definitely | 20:24 |
ThomasEgi | with a high gain towards the direction you point them to. | 20:24 |
Juul | best gain per $ | 20:24 |
ThomasEgi | it has very low gain directly behind the dish,too. | 20:24 |
ThomasEgi | the parts that point from the focal point along the edge does have a bit tho | 20:24 |
ThomasEgi | for an optical system. you only have the line of sight. | 20:26 |
ThomasEgi | a very narrow beam. with no pickup from other directions. | 20:26 |
Juul | yeah, but usb sticks and old satellite dishes are so accessible and cheap | 20:26 |
ThomasEgi | indeed | 20:26 |
ThomasEgi | it depends on your environment tho. | 20:27 |
ThomasEgi | there are only a limeted number of frequency bands for radio communication | 20:27 |
Juul | yeah, interference may become a very limiting factor | 20:27 |
ThomasEgi | also, relay stations usualy cut bandwith in half for wifi systems | 20:28 |
Juul | do you have a good idea for a battery powered 2.4 ghz interference generator? | 20:28 |
ThomasEgi | a microwave oven isnt exactly battery powered. but pretty much garanteed to interfere crap out of your environment :D | 20:29 |
ThomasEgi | other than that. there are open source wirleless jammers on the net | 20:29 |
ThomasEgi | neither of those are legal to use of course. | 20:30 |
Juul | hah, yeah the problem with microwave ovens is that wifi is built specifically to work with that interference present | 20:30 |
ThomasEgi | btw. what are your definitions of lowbandwith and long range? | 20:30 |
Juul | 9600 baud or greater over a few miles would be nice, but it's very important to us that it's using very cheap and accessible technology, so you don't need to have money or many tech skills to set it up | 20:32 |
ThomasEgi | for such low baud rates you can easily design an optical transmittion system | 20:32 |
ThomasEgi | you could use big sensor areas (like small solar panel pieces), and entire led-arrays or a high power led for transmittion. | 20:33 |
ThomasEgi | laserpointers would work just aswell | 20:33 |
Juul | hehe, that might be interesting | 20:33 |
Juul | a laserpointer modulated by the headphone output or some such | 20:34 |
ThomasEgi | that's low tech enough to debug with just a soundcard as scope-repacement | 20:34 |
ThomasEgi | optical phones are really easy to do,too | 20:34 |
ThomasEgi | but led's work a lot better for them | 20:34 |
ThomasEgi | they simply have a lot more output power. | 20:34 |
ThomasEgi | lasers are nice because most come with an optical system buildin. | 20:34 |
ThomasEgi | for an led, you need to source some cheap lenses | 20:35 |
Juul | yeah, it's not quite as easy to do as we'd like | 20:35 |
ThomasEgi | but really . 9600 baud is a piece of cake. | 20:35 |
Juul | the challenge is making it simple enough that the devices can be built using parts that are available in the community | 20:36 |
Juul | with little technical skill | 20:36 |
Juul | but we may decide that that's too much of a restriction | 20:36 |
ThomasEgi | well what i can tell you is that with just 2 transistors you can already build a fully working optical phone. | 20:37 |
Juul | yeah, but the problem is with the optics needed for a long range link | 20:37 |
ThomasEgi | a simple lens. | 20:38 |
ThomasEgi | they can be found everywhere. in every tool shop, ebay, china,.. | 20:38 |
Juul | hm | 20:39 |
Juul | maybe it _is_ simpler than i think | 20:39 |
ThomasEgi | as i said. i once build an optical transmittion system for audio. on just an led, one transistor and a small solar panel i sourced from an old solar-powered calculator. | 20:39 |
ThomasEgi | and an earpiece. | 20:39 |
ThomasEgi | it made pretty ok sound. above 8kHz it had a -20db slope . but regular telephone got cut of at 4khz already so who cares. | 20:40 |
ThomasEgi | and that was made from scrap stuff i had around | 20:41 |
-!- yash [~ffffff@cpe-66-27-118-94.san.res.rr.com] has joined ##hplusroadmap | 20:42 | |
ThomasEgi | Juul, best of all. it's fun to play around with , and might even teach you a good deal of electronics | 20:42 |
Juul | yeah it sounds fun | 20:43 |
-!- yashgaroth [~ffffff@cpe-66-27-118-94.san.res.rr.com] has quit [Ping timeout: 252 seconds] | 20:43 | |
-!- yash is now known as yashgaroth | 20:43 | |
ThomasEgi | Juul, got any tools to deal with electronics around? soldering iron, lab supplies , multimeter , scopes? | 20:44 |
Juul | Thomas42, yep, everything under the sun | 20:45 |
ThomasEgi | in that case.. building an optical phone should be like 10 minutes worth of work | 20:46 |
ThomasEgi | want me to draw you a simple shematic? | 20:46 |
ThomasEgi | using a software modem you may even be able to send some digital data around. not exactly much but maybe 300baud or so using simple 2 frequency keying | 20:47 |
Juul | ThomasEgi, a simple diagram would be great! | 20:48 |
ThomasEgi | gimme a second | 20:53 |
ThomasEgi | ok gimme 2. i'll make this a bit more supply-voltage independent :D | 20:56 |
Juul | cool | 20:57 |
@kanzure | at one point the ronja person was in here.. kendoka? | 21:09 |
ThomasEgi | Juul, are you fine with a postscript file? | 21:10 |
Juul | ThomasEgi, definitely | 21:10 |
ThomasEgi | kanzure, i think clock was in here once or so | 21:10 |
ThomasEgi | Juul, can i netcat you the file? | 21:11 |
ThomasEgi | Juul, or dcc transfer? or upload somewhere? | 21:12 |
Juul | ThomasEgi, hehe, sure | 21:15 |
ThomasEgi | sure which of those :D? | 21:15 |
Juul | priv msg | 21:15 |
Juul | oh sorry, totally autocompleted wrong | 21:16 |
ThomasEgi | np | 21:16 |
ThomasEgi | i love netcat. | 21:16 |
ThomasEgi | the only part that sorta matters is the resistor R4. it should be picked according to the formular. other than that, you can pretty much use any led you can get hands on. | 21:17 |
ThomasEgi | and any general-purpose diodes. | 21:17 |
ThomasEgi | and pretty whatever npn-transistor you have around that can handle the LED current. | 21:18 |
ThomasEgi | R1 and R2 are pretty non-critical too. | 21:19 |
ThomasEgi | i'd say anywhere between 300 and 10k should do for those. | 21:19 |
-!- sheena1 [~home@174.4.134.159] has joined ##hplusroadmap | 21:20 | |
ThomasEgi | supply voltage.. given those values. anywhere between 5 and 16V. | 21:22 |
ThomasEgi | as for the receiver. connecting a solar cell to a small speaker should give you first results already (try not to use your 200$ headphones as the solar panel also outputs a good deal of DC, which is not very healthy) | 21:25 |
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Remote host closed the connection] | 21:56 | |
-!- Gwyxx [~Gwyxx@isd01.hq.networklayer.com] has joined ##hplusroadmap | 22:01 | |
-!- Lucas_ [4a6fa682@gateway/web/freenode/ip.74.111.166.130] has quit [Quit: Page closed] | 22:18 | |
-!- Urchin[emacs] [~user@unaffiliated/urchin] has quit [Ping timeout: 260 seconds] | 22:20 | |
-!- yashgaroth [~ffffff@cpe-66-27-118-94.san.res.rr.com] has quit [Quit: Leaving] | 22:46 | |
-!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has quit [Remote host closed the connection] | 22:48 | |
-!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has joined ##hplusroadmap | 22:49 | |
-!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has quit [Ping timeout: 264 seconds] | 22:53 | |
-!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has joined ##hplusroadmap | 23:00 | |
sheena1 | i'm looking to talk to someone who has successfully found chinese manufacturers for a electronic gadget that they've funded on kickstarer.. anyone here know anyone wh would talk to me? | 23:26 |
-!- augur [~augur@208.58.5.87] has joined ##hplusroadmap | 23:55 | |
--- Log closed Sun Feb 24 00:00:09 2013 |
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!