| --- Log opened Fri Nov 21 00:00:57 2014 | ||
| -!- lichen [~lichen@c-50-139-11-6.hsd1.or.comcast.net] has joined ##hplusroadmap | 00:25 | |
| -!- juri_ [~juri@vpn166.sdf.org] has joined ##hplusroadmap | 00:25 | |
| -!- Merovoth [~Merovoth@gateway/tor-sasl/merovoth] has quit [Ping timeout: 250 seconds] | 00:28 | |
| -!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has quit [Remote host closed the connection] | 02:15 | |
| -!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has joined ##hplusroadmap | 02:16 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-hbetmxohdaifxccn] has joined ##hplusroadmap | 02:28 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 02:51 | |
| -!- Beatzebub [~beatzebub@S0106b81619e8ecee.gv.shawcable.net] has quit [Read error: Connection reset by peer] | 02:53 | |
| -!- drewbot [~cinch@ec2-54-80-232-96.compute-1.amazonaws.com] has quit [Remote host closed the connection] | 03:00 | |
| -!- drewbot [~cinch@ec2-54-81-213-162.compute-1.amazonaws.com] has joined ##hplusroadmap | 03:00 | |
| -!- juri_ [~juri@vpn166.sdf.org] has quit [Ping timeout: 272 seconds] | 03:06 | |
| -!- juri_ [~juri@vpn166.sdf.org] has joined ##hplusroadmap | 03:07 | |
| -!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap | 03:57 | |
| -!- eudoxia [~eudoxia@r190-133-109-242.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 04:21 | |
| -!- eudoxia [~eudoxia@r190-133-109-242.dialup.adsl.anteldata.net.uy] has quit [Client Quit] | 04:23 | |
| -!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has quit [Ping timeout: 240 seconds] | 04:36 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-hbetmxohdaifxccn] has quit [Quit: Connection closed for inactivity] | 04:37 | |
| -!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has joined ##hplusroadmap | 04:39 | |
| -!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has quit [Ping timeout: 245 seconds] | 04:43 | |
| -!- pete4242 [~smuxi@boole.london.hackspace.org.uk] has joined ##hplusroadmap | 04:44 | |
| chris_99 | Has anyone seen any DIY flow cytometry type devices out of interest? | 04:58 |
|---|---|---|
| -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 244 seconds] | 05:03 | |
| chris_99 | without using any fancy florescent proteins | 05:04 |
| chris_99 | i'm wondering if you could slow the flow of the liquid, possibly through some kind of microfluidic slide, and do it optically through a high frame rate camera? | 05:06 |
| -!- Vutral [ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 05:08 | |
| -!- yorick [~yorick@oftn/member/yorick] has quit [Ping timeout: 245 seconds] | 05:09 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-vphbxwbezadqurkd] has joined ##hplusroadmap | 05:17 | |
| -!- yorick [~yorick@oftn/member/yorick] has joined ##hplusroadmap | 05:20 | |
| -!- eudoxia [~eudoxia@r190-133-109-242.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 05:32 | |
| eudoxia | paperbot: http://dl.acm.org/ft_gateway.cfm?id=2351489&ftid=1277049&dwn=1&CFID=601643638&CFTOKEN=40195041 | 05:33 |
| eudoxia | oh right paperbot doesn't work or something | 05:34 |
| bbrittain | paperbot: YOU BROKE? | 05:35 |
| bbrittain | kanzure: ^ | 05:35 |
| kanzure | paperbot is totally broken | 05:41 |
| kanzure | https://github.com/kanzure/paperbot | 05:42 |
| kanzure | youfix? | 05:42 |
| kanzure | hmm got an email from someone's email address hosted at chmail.ir | 05:49 |
| kanzure | about pdfparanoia | 05:49 |
| kanzure | i am gonna try sending back: | 05:53 |
| kanzure | من می خواهم برای کمک به. من خوب فارسی صحبت کنم. می تواند شما را کپی و ارسال هر گونه پیغام خطا؟ لطفا به من | 05:53 |
| kanzure | and | 05:54 |
| kanzure | بگویید مراحل شما با استفاده از. | 05:54 |
| kanzure | oh and "قیمت من پلوتونیوم است." | 05:55 |
| eudoxia | the gnusha bot doesn't like unicode | 05:57 |
| kanzure | ugh | 05:57 |
| kanzure | i have 99 bots and they're all problems | 05:58 |
| archels | .tr من می خواهم برای کمک به. من خوب فارسی صحبت کنم. می تواند شما را کپی و ارسال هر گونه پیغام خطا؟ لطفا به من | 05:59 |
| yoleaux | archels: Sorry, that command (.tr) crashed. | 05:59 |
| archels | make that 100 | 05:59 |
| archels | plutonium, haha | 06:01 |
| kanzure | i wonder if this causes time travel: | 06:08 |
| kanzure | http://www.theguardian.com/science/2014/nov/21/tapeworm-parasite-mans-brain-four-years-china | 06:08 |
| kanzure | .wik spirometra erinaceieuropaei | 06:09 |
| yoleaux | "Spirometra erinaceieuropaei is a tapeworm that infects domestic animals and humans. In humans infection is called sparganosis. The worm has an interesting lifecycle. The adult worm is present in the small intestine of cats and dogs where it may grow as long as 1.5 metres." — http://en.wikipedia.org/wiki/Spirometra_erinaceieuropaei | 06:09 |
| kanzure | yes i'll take one 1.5 meter tapeworm in my brain, please | 06:09 |
| eudoxia | ew | 06:10 |
| kanzure | "Rather than living on the brain tissue of its unknowing victim, the parasite is thought to have simply absorbed nutrients from the man’s brain" | 06:10 |
| eudoxia | this is like that thing where botflies grow in your brain | 06:10 |
| kanzure | paperbot: http://genomebiology.com/2014/15/11/510/abstract | 06:10 |
| kanzure | .title | 06:10 |
| yoleaux | Genome Biology | Abstract | The genome of the sparganosis tapeworm Spirometra erinaceieuropaei isolated from the biopsy of a migrating brain lesion | 06:10 |
| kanzure | "The patient first noticed something was wrong in 2008 when he began suffering headaches, math, memory flashbacks and strange smells." | 06:12 |
| archels | suffering math? | 06:13 |
| kanzure | hehehe "use of frog poultice, a traditional Chinese remedy where raw frog meat is used to calm sore eyes." | 06:13 |
| superkuh | http://www.superkuh.com/pictures/plif/wc083.gif (re: worms in brain, math) | 06:14 |
| kanzure | yeah i imagine that's accurate | 06:15 |
| -!- comma8 [comma8@gateway/shell/yourbnc/x-vsuutkxukgollmot] has quit [Ping timeout: 272 seconds] | 06:15 | |
| -!- comma8 [comma8@gateway/shell/yourbnc/x-iinbqhehvqhhggxb] has joined ##hplusroadmap | 06:23 | |
| bbrittain | kanzure: what is wrong with paperbot? I don't see any new issues | 06:23 |
| eudoxia | probably it's not the code that's broken, just the world around it | 06:24 |
| bbrittain | stupid world, changing on us | 06:25 |
| chris_99 | haha | 06:25 |
| -!- MutMan [~muthuri@197.237.156.120] has joined ##hplusroadmap | 06:33 | |
| MutMan | paper | 06:34 |
| MutMan | paperbot, http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6945905 | 06:34 |
| paperbot | http://libgen.org/scimag/get.php?doi=10.1109%2FTPAMI.2014.2366761 | 06:34 |
| kanzure | it's definitely paperbot itself that is broken | 06:35 |
| -!- Zinglon [~Zinglon@5351F2A8.cm-6-2d.dynamic.ziggo.nl] has joined ##hplusroadmap | 06:44 | |
| -!- snuffeluffegus [~snuff@5.150.254.180] has joined ##hplusroadmap | 06:50 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 244 seconds] | 07:04 | |
| kanzure | jrayhawk: why does git have its own transport protocol? | 07:06 |
| kanzure | or rather, | 07:06 |
| kanzure | pasky: why does git have its own transport protocol? | 07:07 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 07:12 | |
| -!- Merovoth [~Merovoth@gateway/tor-sasl/merovoth] has joined ##hplusroadmap | 07:16 | |
| pasky_ | rather than...? | 07:39 |
| pasky_ | the protocol is very simple | 07:39 |
| pasky_ | basically it exchanges refs to negotiate the list of objects to send, then packs them up the same as the on-disk pack format is and sends them | 07:40 |
| archels | hmm, not sure if this means that Singularity University is active in The Netherlands or whether they just held an event here http://www.singularityuniversity.nl | 07:54 |
| archels | meetup of "Amsterdam Futurists Society" next Wednesday http://www.meetup.com/Amsterdam-Futurists-Society/ | 07:55 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Read error: Connection timed out] | 07:56 | |
| kanzure | or domain squatting | 07:58 |
| archels | lolwat http://www.brainiacdating.com/ | 08:00 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 08:00 | |
| -!- snuffeluffegus [~snuff@5.150.254.180] has quit [Quit: Leaving] | 08:08 | |
| -!- MutMan [~muthuri@197.237.156.120] has quit [Ping timeout: 255 seconds] | 08:09 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Remote host closed the connection] | 08:10 | |
| kanzure | they will soon be hearing from http://upload.wikimedia.org/wikipedia/en/7/7b/Brainiac(STAS).jpg | 08:11 |
| -!- nmz787_i [~nmccorkx@134.134.139.72] has joined ##hplusroadmap | 08:37 | |
| -!- eudoxia [~eudoxia@r190-133-109-242.dialup.adsl.anteldata.net.uy] has quit [Quit: Lost terminal] | 09:06 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 09:07 | |
| -!- pete4242 [~smuxi@boole.london.hackspace.org.uk] has quit [Remote host closed the connection] | 09:26 | |
| -!- Zinglon [~Zinglon@5351F2A8.cm-6-2d.dynamic.ziggo.nl] has quit [Ping timeout: 265 seconds] | 09:41 | |
| * heath waves good evening | 09:41 | |
| kanzure | "yuri: i am expert in shadowy field of many things" | 09:42 |
| chris_99 | is it possible to pump fluid through a Hemocytometer somehow, or is that a dumb idea for some reason? | 09:49 |
| nmz787_i | if the top is sealed, sure it would work | 09:51 |
| nmz787_i | doesn't seem like a dumb idea to me if you're interested in counting cell concentrations or flow dynamics | 09:52 |
| nmz787_i | (i did some opencv to track cells passing through a microfluidic years ago) | 09:52 |
| nmz787_i | it was basically a hemocytometer without the grid etched on the bottom | 09:53 |
| chris_99 | nice, yeah i'd like to count yeast cells during fermentation | 09:53 |
| chris_99 | continuously | 09:53 |
| chris_99 | i assume you can't optically tell live from dead cells right? without dyeing | 09:54 |
| nmz787_i | you likely can, but idk how simple it is | 09:55 |
| nmz787_i | when i did those kind of counts we used dye | 09:55 |
| chris_99 | gotcha | 09:56 |
| nmz787_i | yeah trypan blue was one of them | 09:56 |
| chris_99 | theres another one that's less toxic too | 09:56 |
| nmz787_i | if the tank is well mixed, you could just steal a small amount, add dye, check, then discard | 09:56 |
| chris_99 | mmm true | 09:56 |
| chris_99 | that could be done automatically too | 09:56 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Read error: Connection timed out] | 10:03 | |
| heath | paperbot: http://www.sciencedirect.com/science/article/pii/0304397594902321 | 10:04 |
| paperbot | http://diyhpl.us/~bryan/papers2/paperbot/%0A%20Asynchronous%20automata%20versus%20asynchronous%20cellular%20automata%0A%20.pdf | 10:04 |
| @ParahSailin_ | optical density | 10:05 |
| @ParahSailin_ | why do you need to microscopically inspect yeast cells for growth curve | 10:05 |
| @ParahSailin_ | yeast cells gunk up flow cytometers anyway | 10:05 |
| @ParahSailin_ | great way to ruin expensive machine | 10:06 |
| chris_99 | ParahSailin_, what other ways are there to count them? | 10:06 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 10:08 | |
| @ParahSailin_ | OD600 | 10:08 |
| chris_99 | hmm interesting, would that work even with things where the main colour could be black | 10:09 |
| nmz787_i | you might be able to do impedance spectroscopy | 10:14 |
| chris_99 | looking at the wiki for that now, thanks | 10:17 |
| nmz787_i | chris_99: you could also look up coulter counter | 10:18 |
| chris_99 | that sounds rather neat | 10:19 |
| nmz787_i | "Alternating current measurements are sometimes used in clinical hematology instruments, due to the special nature of cell membranes. At low frequencies (below 500 kHz), alternating and direct current measurements behave essentially the same way. At intermediate frequencies (500 kHz - 6 MHz), the plasma membrane of cells can become polarized, leading to a decreased capacitance of the measurement systems. However, at high | 10:19 |
| nmz787_i | frequencies (6-20 MHz), the cell membrane loses its polarization, and the electrical pulses provide information about the cell cytoplasm" | 10:19 |
| nmz787_i | hmm, this only goes up to 100kHz but it is a complete solution for signal generation and sensing http://www.analog.com/en/digital-to-analog-converters/direct-digital-synthesis-dds/ad5933/products/product.html | 10:22 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Read error: Connection timed out] | 10:23 | |
| chris_99 | cool | 10:23 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 10:25 | |
| chris_99 | i love the look of these | 10:27 |
| chris_99 | http://www.ebay.co.uk/itm/Coulter-Counter-Model-ZM-Control-Particle-Count-With-Dust-Cover-/380391701942?pt=LH_DefaultDomain_0&hash=item58911cbdb6 | 10:27 |
| chris_99 | they look very retro | 10:27 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Max SendQ exceeded] | 10:28 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 10:30 | |
| nmz787_i | oh, they have a whole section with seemingly similar (checking) parts http://www.analog.com/en/digital-to-analog-converters/direct-digital-synthesis-dds/products/index.html#Direct_Digital_Synthesis | 10:32 |
| chris_99 | hmm i think they're just DDS chips | 10:33 |
| chris_99 | so they might not sample an input | 10:33 |
| nmz787_i | yeah that's what I'm seeing, they don't have the ADC like the former does | 10:35 |
| chris_99 | mmm | 10:35 |
| nmz787_i | hmm https://github.com/arachnidlabs/ad983x | 10:39 |
| nmz787_i | .title | 10:39 |
| yoleaux | arachnidlabs/ad983x · GitHub | 10:39 |
| nmz787_i | 'Arduino library for interfacing with AD9833, AD9834 and AD9838' | 10:39 |
| nmz787_i | so maybe it wouldn't be too hard to throw a decent ADC next to one of those (or use the arduino's if it would be sufficient, idk) | 10:40 |
| chris_99 | you mean to do the coulter technique? | 10:40 |
| nmz787_i | huh, authored 3 days ago | 10:40 |
| nmz787_i | yeah | 10:40 |
| chris_99 | yeah it sounds v. interesting | 10:40 |
| nmz787_i | hmm https://www.tindie.com/products/arachnidlabs/circuit-patterns-trading-cards-full-deck/ | 10:44 |
| kanzure | http://blog.coinalytics.co/visualizing-silk-road-20-relationships | 10:48 |
| nmz787_i | "On November 6, 2013 the "... is that date correct? | 10:49 |
| nmz787_i | only a month after v1 got shutdown? | 10:49 |
| -!- helleshin [~talinck@66-161-138-110.ubr1.dyn.lebanon-oh.fuse.net] has quit [Ping timeout: 264 seconds] | 10:50 | |
| kanzure | they mean 2014 | 10:50 |
| -!- nmz787_i [~nmccorkx@134.134.139.72] has quit [Quit: Leaving.] | 10:52 | |
| -!- HEx1 [~HEx@hexwab.plus.com] has quit [Ping timeout: 245 seconds] | 11:23 | |
| -!- nmz787_i [nmccorkx@nat/intel/x-esjyfkctowipchgy] has joined ##hplusroadmap | 11:29 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Ping timeout: 272 seconds] | 11:35 | |
| -!- HEx1 [~HEx@hexwab.plus.com] has joined ##hplusroadmap | 11:38 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 11:38 | |
| nmz787_i | dang 8GHz amp for ~$3 on ebay http://www.minicircuits.com/pdfs/ERA-1SM+.pdf | 11:50 |
| nmz787_i | found via http://www.qsl.net/n/n5ib//DDS/ | 11:51 |
| jrayhawk | kanzure: re: git://: because git is a tree of objects and not deltas, it's useful to be able to run dynamic compression on whatever objects are being sent so as to avoid sending duplicate information | 11:55 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Read error: Connection timed out] | 11:57 | |
| jrayhawk | also there's no particular reason not to | 11:58 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 11:59 | |
| jrayhawk | The default advice of "use http for everything" is an artifact of raising a generation of programmers on cripply fractally-bad web technologies | 12:00 |
| jrayhawk | like programmers raised on the web have never thought about how a synchronous textual s | 12:02 |
| jrayhawk | unmultiplexed non-streaming and non-dynamically-negotiable object retrieval system could possibly be suboptimal | 12:03 |
| -!- nickjohnson [sid789@gateway/web/irccloud.com/x-lrazrwoaopzrrwqi] has joined ##hplusroadmap | 12:07 | |
| nickjohnson | Hello. I felt my ears burning from this direction. | 12:08 |
| jrayhawk | it's just the soup they swim in | 12:08 |
| nickjohnson | (I'm the entirety of Arachnid Labs) | 12:10 |
| nmz787_i | yo | 12:12 |
| nmz787_i | i just found your recent dds repo on github | 12:12 |
| nickjohnson | *nods* | 12:12 |
| nmz787_i | have half an email written to you, which basically said as much | 12:12 |
| nmz787_i | was wondering about the Loki project as I don't see it on tindie | 12:12 |
| nickjohnson | I wrote it because I'm currently designing a project based on that and an AVR | 12:12 |
| nmz787_i | but also found this http://midnightdesignsolutions.com/dds60/index.html#Overview | 12:13 |
| nickjohnson | The Loki got shelved after Cypress released their PSoC pioneer board for less than its BoM cost in materials | 12:13 |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Ping timeout: 256 seconds] | 12:13 | |
| nmz787_i | and was thinking since it doesn't have an ADC, I might use an LPC-Link v2 | 12:13 |
| nickjohnson | (Also after I realised just how complicated launching a new microcontroller ecosystem is) | 12:13 |
| nmz787_i | (which has a triple core ARM and 80 MSPS ADC) | 12:13 |
| -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap | 12:14 | |
| nmz787_i | basically was looking for something a little higher freq than this, which is a nice complete-package solution http://www.analog.com/en/digital-to-analog-converters/direct-digital-synthesis-dds/ad5933/products/product.html | 12:14 |
| nmz787_i | for generation and sensing | 12:14 |
| nmz787_i | but it looks like they have no integrated one-chip solutions that are faster | 12:15 |
| nickjohnson | Yeah, that's a seriously cool chip | 12:15 |
| nickjohnson | The board I'm working on should be able to do everything it can, however | 12:16 |
| nickjohnson | By doing peak amplitude and phase sensing instead | 12:16 |
| nmz787_i | I am a bit confused as to what this means "This self-contained functional module generates a good-quality RF signal from 1-60 MHz" re the DDS60 project | 12:16 |
| nickjohnson | But my board only has a 16MHz oscillator, not 80 | 12:16 |
| nmz787_i | like, is that 1MHz or 1Hz | 12:16 |
| nmz787_i | the oscillator is not 80, 80 is the ADC sample rate | 12:16 |
| nmz787_i | http://www.embeddedartists.com/products/lpcxpresso/lpclink2.php | 12:17 |
| nickjohnson | Well, the 9851 goes up to 150MHz | 12:17 |
| nmz787_i | http://www.nxp.com/products/microcontrollers/cortex_m4/lpc4300/LPC4370FET100.html | 12:17 |
| nickjohnson | Right. But 16msps is the upper limit on my board because of the 16mhz clock | 12:18 |
| nmz787_i | but honestly I am not sure what is required | 12:18 |
| nmz787_i | ah | 12:18 |
| nmz787_i | I see | 12:18 |
| nickjohnson | Required for what exactly? | 12:18 |
| nmz787_i | "Alternating current measurements are sometimes used in clinical hematology instruments, due to the special nature of cell membranes. At low frequencies (below 500 kHz), alternating and direct current measurements behave essentially the same way. At intermediate frequencies (500 kHz - 6 MHz), the plasma membrane of cells can become polarized, leading to a decreased capacitance of the measurement systems. However, at high | 12:18 |
| nmz787_i | frequencies (6-20 MHz), the cell membrane loses its polarization, and the electrical pulses provide information about the cell cytoplasm" | 12:18 |
| nmz787_i | .wik coulter counter | 12:18 |
| yoleaux | "A Coulter counter is an apparatus for counting and sizing particles suspended in electrolytes. It is used for cells, bacteria, prokaryotic cells and virus particles." — http://en.wikipedia.org/wiki/Coulter_counter | 12:18 |
| nickjohnson | That's what you want to do? | 12:19 |
| nmz787_i | and in general impedance spectroscopy (I personally am interested in nucleic acid detection and quantification) | 12:19 |
| nmz787_i | another person here is brewing beer | 12:19 |
| nmz787_i | they were looking for some solution to do cell counts | 12:20 |
| nickjohnson | ah | 12:20 |
| nmz787_i | and I happen to also need to do similar for my own lab-on-a-chip stuff | 12:20 |
| nmz787_i | aren't AVR adcs limited to 80ksps with decent bitwidth? | 12:21 |
| nmz787_i | I thought 250ksps was the limit but it lost a few bits of range | 12:21 |
| nickjohnson | So, the AD9834 takes up to 75MHz clock, and therefore does the frequency range you care about | 12:21 |
| nickjohnson | While still being fairly affordable and straightforward | 12:21 |
| nickjohnson | One thing I don't like about the DDS-60 you linked to is that they use AC coupling | 12:22 |
| nickjohnson | So it's not good for frequencies down to DC | 12:22 |
| nickjohnson | Though you probably don't care about that :) | 12:22 |
| nickjohnson | And yes, the AVR ADCs are slow. I'm working around that by building peak detector and phase discriminator circuits into the board | 12:23 |
| nickjohnson | So you can detect amplitude and phase without sampling the whole waveform | 12:23 |
| nmz787_i | ah | 12:24 |
| nmz787_i | hmm, yeah the different articles that mention the dds60 change between saying 1-60Mhz or 0-60MHz | 12:25 |
| nickjohnson | You can avoid needing a DC cap with a differential amp approach | 12:27 |
| nickjohnson | one sec | 12:27 |
| nickjohnson | What exactly do you need to measure about the return waveform? Is phase and amplitude enough? | 12:27 |
| nickjohnson | https://drive.google.com/file/d/0B_iKMAYU3t1LdDlaVkdRYWdwQW8/view?usp=sharing | 12:28 |
| nickjohnson | The structure on page 1 with 3 opamps is a combined differential -> single ended amp to avoid AC coupling, and a filter | 12:28 |
| nmz787_i | umm | 12:29 |
| nickjohnson | It's less horrendous than it looks ;) | 12:29 |
| nmz787_i | i think the freq response curve | 12:29 |
| nickjohnson | Right | 12:29 |
| nickjohnson | But that's just amplitude and possibly phase at each step of a frequency sweep | 12:29 |
| nickjohnson | RE DDSes, the other option is to pick one that has a built in PLL, so you don't need to feed it a high frequency clock | 12:30 |
| nmz787_i | then you would apply matching algorithms to different curves you obtained, or some other alogorithm to know how much one curve changed from another... and deduce some info (# of particles, length of polymer maybe) | 12:30 |
| nickjohnson | *nods* | 12:30 |
| nickjohnson | Sounds like you should be able to do it with peak and maybe phase measurements, then | 12:30 |
| nickjohnson | You can still use a high speed ADC if you want :) | 12:30 |
| nmz787_i | $20 for that dev board isn't too bad | 12:31 |
| nmz787_i | though it has a 0 to 0.8V input | 12:31 |
| nmz787_i | so shifters would be needed | 12:31 |
| nickjohnson | Which one are you talking about? | 12:31 |
| nickjohnson | The DDS-60 daughterboard? | 12:31 |
| @ParahSailin_ | ups uses golf carts with trailers now? | 12:32 |
| nmz787_i | no the ADC input of the LPC-linkv2 | 12:32 |
| nmz787_i | (which uses an NXP LPC4370 processor chip) | 12:32 |
| nickjohnson | Oh, I see | 12:33 |
| nickjohnson | Well, there are lots of ways to get a fast ADC | 12:33 |
| nmz787_i | compared to making phase/peak detectors | 12:33 |
| nmz787_i | which I haven't thought about | 12:33 |
| nmz787_i | I would assume using the input-capture interrups on the AVR somehow? | 12:33 |
| nickjohnson | A peak detector is just a diode followed by a capacitor | 12:33 |
| nickjohnson | You pull it low, then let it charge up to the peak value of the input signal | 12:34 |
| nmz787_i | oh, but that doesn't give you time | 12:35 |
| nmz787_i | i guess that would be your phase detector that gives you that | 12:35 |
| nickjohnson | yup | 12:39 |
| nickjohnson | For that, you have to threshold the signals, then XOR them | 12:39 |
| nmz787_i | so would your project be able to produce a graph easily? | 12:39 |
| nickjohnson | Which gives you a square wave with duty cycle proportional to the phase difference | 12:39 |
| nmz787_i | of the freq response? | 12:39 |
| nickjohnson | Yup, it can produce a bode plot | 12:39 |
| nickjohnson | Only up to about 2MHz, though | 12:40 |
| nickjohnson | You'd need a faster DDS to get higher than that | 12:40 |
| nmz787_i | is there any kind of FFT bode plot production out there, where you only need one/few measurements, rather than needing to sweep and take a measurement at each step? | 12:40 |
| nmz787_i | to reduce the complete-plot acquisition time | 12:40 |
| nickjohnson | Not really - in theory it could be any curve at all. You can reduce the number of steps you make and interpolate, is all | 12:41 |
| nmz787_i | (i am pretty much making that idea up) | 12:41 |
| nickjohnson | And that's what the combined chip you linked to does, too | 12:42 |
| nmz787_i | any idea how long a bode plot takes to acquire? would the non-combined chip (like you're using) need closer to microseconds or milliseconds to hop from one freq to the next? | 12:43 |
| nmz787_i | (obv it depends on hop/step increment size) | 12:43 |
| kragen | glad to see this interaction is turning out productive :) | 12:45 |
| nickjohnson | Depends on how wide a sweep you want and how many steps per octave | 12:48 |
| nickjohnson | The DDS chip can change frequencies instantly | 12:49 |
| nickjohnson | The peak and phase detectors depend on what values were chosen for the rectifying cap and resistor. If it responds slowly (milliseconds), it will give a stable value for slow signals. If it responds quickly (microseconds), you can take more samples per second | 12:50 |
| -!- cpopell [~cpopell@c-76-26-144-132.hsd1.dc.comcast.net] has joined ##hplusroadmap | 12:50 | |
| nickjohnson | For instance, in mine I'm using a 10k resistor and 0.01u cap, which has a ~1600hz cutoff, so it'd give stable values for sweeps above about 10KHz (conservatively), and settle in a few milliseconds. | 12:51 |
| nickjohnson | Here's a simulation of it settling on a 100khz input signal, for instance: http://imgur.com/z6mABLx | 12:53 |
| nickjohnson | Looks like it takes about 3ms to settle to within 99% of its true value | 12:53 |
| nickjohnson | My reasoning for picking that value was that for lower than about 10khz, you can probably just sample the waveform directly :) | 12:54 |
| kragen | I was just watching a ten-million-frame-per-second video of a water drop getting diverted by a laser pulse vaporizing one side of it | 12:57 |
| kragen | taken with a 20 000 frame per second camera | 12:57 |
| kragen | this is very similar to the peak-and-phase-detection approach you're talking about | 12:58 |
| kragen | I mean in a sense it's more similar to heterodyning downconversion with a mixer | 12:59 |
| nickjohnson | kragen: Been reading XKCD's What-if? | 13:00 |
| nickjohnson | Incidentally, I have a friend who's about to launch a Kickstarter for a high speed LED based flash that I worked on. 1 million lumens for 500 nanoseconds :) | 13:00 |
| nickjohnson | And yeah, I believe downconversion is another possible approach. But I don't understand the frequency domain well enough to say :) | 13:01 |
| kragen | yup | 13:01 |
| kragen | oh, you just multiply (somehow) by a sinewave that's close to the frequency of the phenomenon you'r etrying to measure | 13:01 |
| kragen | and the product contains sum and difference components | 13:02 |
| nickjohnson | Here's one of AD's fancier DDSes. 500MHz, with built in PLL so you can feed it a slower clock, and dual outputs: http://www.analog.com/static/imported-files/data_sheets/AD9958.pdf | 13:02 |
| kragen | the difference components are close to 0Hz, which makes it easy to measure their amplitude and phase | 13:02 |
| kragen | I think you can do a lot better than 500 nanoseconds with a fairly normal LED if you're willing to accept a much smaller number of lumens | 13:03 |
| nickjohnson | kragen: Right. I can say the words, but I don't fundamentally grok how it works. :) | 13:03 |
| nickjohnson | Yes, the limitation on this was the rise times given parasitic capacitance and inductance, and not wanting to burn out the LED bond wires with the surge | 13:03 |
| nickjohnson | 500ns is short enough to stop a bullet :) | 13:03 |
| kragen | :) | 13:03 |
| nickjohnson | And an order of magnitude faster than a xenon flash | 13:04 |
| kragen | actually you can stop a bullet pretty well with a xenon flash too | 13:04 |
| -!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 240 seconds] | 13:04 | |
| kragen | I mean Edgerton was famous for that | 13:04 |
| nickjohnson | Only if you do it from a long way away or don't mind a lot of blur | 13:04 |
| kragen | yeah, true | 13:04 |
| nickjohnson | Xenon flashes have a long decay time, so you get a lot of blur in the photo | 13:04 |
| kragen | as for heterodyning: well, you can rotate and scale by a vector by multiplying by [[a b] [-b a]], right? | 13:04 |
| nickjohnson | The LED flash has a nice square waveform | 13:04 |
| nickjohnson | It was fun working with it - 180A, 2.8KW, for half a microsecond. | 13:05 |
| nickjohnson | 20 times their rated working current :P | 13:05 |
| kragen | like say [[cos θ, sin θ], [-sin θ, cos θ]] | 13:05 |
| nickjohnson | kragen: With you so far, sure. Though somewhat rusty. | 13:05 |
| heath | so much broken software on the internets | 13:06 |
| nickjohnson | nmz787_i: So, the question is, what frequency range do you most care about? | 13:06 |
| kragen | you can sort of draw it out on paper and see how if you sort of move the axes to [cos θ, sin θ] and [-sin θ, cos θ], along with the point you're trying to rotate, that it's correct | 13:06 |
| kragen | anyway if you do that twice you end for two different angles you can get the rotation by the sum of the angles | 13:06 |
| kragen | which gives you the cos and sin of the sum | 13:07 |
| kragen | which you probably learned when you were like 13 involves sums and differences of the angles | 13:07 |
| nickjohnson | I think I need Vi Hart to explain this to me ;) | 13:07 |
| kragen | yeah | 13:07 |
| kragen | I wish I were Vi Hart | 13:07 |
| kragen | also that IRC had graph paper and markers | 13:08 |
| nickjohnson | *nods* | 13:08 |
| kragen | anyway it falls out of the math pretty easily | 13:08 |
| kragen | you can also go back in the other direction | 13:08 |
| nickjohnson | Yeah. The trouble is, I can follow the math (I don't) but still not fundamentally 'get' it | 13:08 |
| kragen | two nearby frequencies produce "beating" | 13:08 |
| kragen | summing sinewaves at two nearby frequencies I mean | 13:09 |
| nickjohnson | Like, you can modulate some amount of data with a signal around 1MHz. How can you mix it down to near-DC and still retain the modulated information? | 13:09 |
| kragen | which is to say, a third frequency (in between the two) amplitude-modulated (which is to say multiplied) by a sinusoidal modulation of the difference of the two | 13:09 |
| nickjohnson | Yup | 13:09 |
| nickjohnson | hmm | 13:09 |
| nickjohnson | Okay, light begins to dawn, perhaps | 13:09 |
| kragen | that is, if you start with the sum and difference signals, and you add them together, you get something that is evidently a product of two sinewaves of the different frequencies | 13:10 |
| nickjohnson | Isn't that effectively downmixing, then? You're producing a low frequency with the sum of two high frequencies? | 13:10 |
| kragen | no, other way around | 13:10 |
| kragen | you're producing two high frequencies by multiplying a low frequency by a high frequency | 13:10 |
| nickjohnson | Sorry, I mean in the case of two nearby frequencies producing a beat | 13:11 |
| kragen | yeah | 13:11 |
| kragen | the low frequency isn't actually in the resulting "beating" signal | 13:11 |
| kragen | unless you rectify it or something | 13:11 |
| nickjohnson | Yeah, it's amplitude modulation of the higher frequency signal | 13:11 |
| kragen | I mean it isn't an additive component of the Fourier analysis of the signal | 13:12 |
| nickjohnson | Self-evidently true because you put in two sine waves, so all you can have is two sine waves | 13:12 |
| nickjohnson | right | 13:12 |
| nickjohnson | So how do you get from there to downmixing? | 13:12 |
| kragen | you multiply frequencies that are close together instead of far apart :) | 13:13 |
| nmz787_i | i thought when you multiply, you get the additive and the subtractive | 13:13 |
| nickjohnson | In the example with beats, you're summing, not multiplying, though | 13:13 |
| nmz787_i | then if you were down or up converting, you select the one you want | 13:13 |
| kragen | it's the other direction | 13:13 |
| nmz787_i | with just a filter | 13:13 |
| nickjohnson | ...wasn't there some sort of tool that let you experiment with chaining signals and filters etc? At a higher level than ltspice? | 13:14 |
| kragen | brb | 13:14 |
| nickjohnson | kragen: I don't know what the rest of that sentence fragment is | 13:14 |
| superkuh | GNU Radio? | 13:15 |
| nickjohnson | That might be it, yeah | 13:15 |
| nmz787_i | superkuh: I think he means to model them | 13:15 |
| superkuh | Specifically GNU Radio Companion. | 13:15 |
| superkuh | Oh. | 13:15 |
| nickjohnson | No, I just meant to experiment with them | 13:16 |
| nickjohnson | I figure maybe being able to actually see waveforms will help me grok this better | 13:16 |
| superkuh | Definitely GNU Radio Companion then. | 13:16 |
| nickjohnson | Just like KSP helped me grok orbital mechanics :P | 13:16 |
| superkuh | Get a $10 rtlsdr dongle and install GR; best way to learn DSP. | 13:16 |
| kragen | sorry, some dude I met at PyConAr was phoning me up to invite me to come stay at his house on the othe side of the country | 13:16 |
| nickjohnson | I need to do that anyway so I can do pre-compliance CE checks on my hardware | 13:16 |
| kragen | which was awesome but really reduced my bandwidth for DSP chat | 13:16 |
| nickjohnson | heh | 13:17 |
| kragen | so what I was saying is that if you start with a sum and difference signal that are similar in frequency, and you add them together, you get a signal that is very evidently a sine wave carrier being AM-modulated by a much lower frequency, which is to say multiplied by it | 13:17 |
| kragen | which perhaps makes it somewhat more intuitive that if you start by multiplying together two sine waves and then Fourier-analyze the result, the Fourier spectrum will contain sum and difference frequencies | 13:18 |
| -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap | 13:18 | |
| nickjohnson | Can we back up for a minute and go over exactly what multiplying two sine waves entails? It's not simply pointwise multiplication, is it? | 13:18 |
| kragen | it's maybe not obvious how the phase of the sum and difference frequencies relates to the phase of the signal being multiplied | 13:18 |
| kragen | yes, it is | 13:19 |
| nickjohnson | okay then | 13:19 |
| kragen | in practice, in analog hardware, it's hard to do actual multiplication | 13:19 |
| fenn | huh i always thought a "phase discriminator" was just some star trek technobabble | 13:19 |
| nickjohnson | Yeah, that requires variable gain | 13:19 |
| nickjohnson | Which is a real PITA | 13:19 |
| kragen | so you settle for feeding them through some component that is significantly nonlinear | 13:19 |
| kragen | such as a diode | 13:19 |
| nickjohnson | fenn: It'd be more commonly called a "phase detector", but "discriminator" is legit too :) | 13:19 |
| nickjohnson | Hah, right, take the logs and sum them | 13:20 |
| kragen | yeah, you can build a fairly precise analog multiplier out of diodes | 13:20 |
| kragen | but you don't need to | 13:20 |
| kragen | you just need something that has significant nonlinearity, but not *so much* nonlinearity that the higher-order terms swamp the second-order terms | 13:21 |
| kragen | I mean the first-order (linear) terms are just x and y | 13:21 |
| nickjohnson | AM modulation == multiplication does help clear things up. It's a lot easier to visualise a LF signal multiplied with an HF one than two similar frequencies, intuitively | 13:21 |
| kragen | the second-order terms are x², xy, and y² | 13:21 |
| kragen | if both x and y are high in frequency, which is to say periodic on a very short timescale, their squares will also be periodic on a very short timescale | 13:21 |
| kragen | and so won't contribute significantly to the LF or IF signal that you're trying to measure | 13:22 |
| kragen | I haven't actually built any kind of heterodyne, so maybe I shouldn't be trying to explain this | 13:22 |
| nickjohnson | You're doing pretty well so far | 13:23 |
| kragen | but it seems fairly intuitive to me. I hope I'm not making any major mistakes. | 13:23 |
| kragen | anyway so you use roughly any kind of device with two input terminals and an output that is a significantly nonlinear, but not too nonlinear, combination of the inputs | 13:23 |
| nmz787_i | i first learned about them in a humanities/english/liberal-arts class, which was on 'the history of radio' | 13:24 |
| kragen | and then you filter out the high-frequency components from it, and you're left with the signal with the difference of the frequencies | 13:24 |
| nickjohnson | One thing that is currently surprising me is that you can use an LPF and get a low frequency component out that didn't exist in the fourier transform of the input signal | 13:25 |
| nickjohnson | As in the case of an AM modulated signal | 13:25 |
| kragen | well, you can't | 13:25 |
| kragen | if you low-pass filter the signal from an AM antenna, you won't hear anything | 13:26 |
| kragen | you have to rectify it first | 13:26 |
| nickjohnson | Didn't we establish something like that? The sum of, say 10KHz and 9KHz gives a 1KHz beat frequency; couldn't you use an LPF and actually get that frequency out? | 13:27 |
| kragen | nope! | 13:27 |
| nickjohnson | ...and in this case by rectify you mean? | 13:27 |
| kragen | traditionally you put the signal through a rusty razor blade or galena crystal or something | 13:27 |
| nickjohnson | I mean, I've seen certain cases of an LPF used as a definition of "rectify" | 13:27 |
| kragen | but those have mostly been supplanted with silicon diodes :) | 13:27 |
| nickjohnson | Which is just multiplying the signal with the tuning frequency again | 13:28 |
| nmz787_i | bbl | 13:28 |
| nmz787_i | thanks for stopping by nickjohnson | 13:28 |
| nickjohnson | nmz787_i: my pleasure | 13:28 |
| kragen | is it really? | 13:28 |
| nmz787_i | will be back in prob 20 or 30 mins | 13:28 |
| nickjohnson | I thought it was | 13:28 |
| kragen | I don't think it is | 13:28 |
| nickjohnson | Happy to recommend a DDS architecture later if you want one | 13:28 |
| kragen | it's more like multiplying the signal by its own signum | 13:28 |
| nickjohnson | kragen: Okay, so what operation is rectifying? | 13:28 |
| kragen | taking the absolute value | 13:28 |
| nickjohnson | Literally abs(signal)? | 13:28 |
| kragen | yes | 13:28 |
| kragen | like what a bridge rectifier does | 13:29 |
| kragen | I mean you can use a single diode too obviously | 13:29 |
| nickjohnson | It seems unintuitive that the absolute value of a signal would have different frequency components to the input signal | 13:29 |
| nickjohnson | To me, at least | 13:29 |
| kragen | it is! but it's also true! | 13:29 |
| kragen | absolute value is a very highly nonlinear operation | 13:29 |
| kragen | if you want to visualize the waveforms resulting from multiplying two similar-frequency signals intuitively, then think about multiplying a sinewave by itself | 13:30 |
| kragen | it's always positive | 13:30 |
| kragen | in fact it's another sinewave of twice the frequency with a DC offset | 13:30 |
| nickjohnson | Right | 13:30 |
| nickjohnson | http://www.wolframalpha.com/input/?i=sin%5E2+x | 13:30 |
| kragen | and if you multiply it by its negative, you get the negative result | 13:30 |
| kragen | right | 13:30 |
| nickjohnson | (Wolfram might almost be as good a tool as marker and graph paper for this conversation) | 13:30 |
| kragen | so if you have two sinewave signals that are shifting in and out of phase, you shift gradually back and forth between that positive DC offset and that negative DC offset | 13:31 |
| nickjohnson | Like http://www.wolframalpha.com/input/?i=sin%28x%29+*+sin%281.1x%29 | 13:31 |
| kragen | I mean, that DC offset is basically the phase difference between them | 13:31 |
| nickjohnson | yup | 13:31 |
| kragen | right | 13:31 |
| kragen | so you can kind of see that that's composed of a sum frequency and a difference frequency | 13:32 |
| kragen | which should hopefully be intuitive now for the special cases where the frequencies are very close together and where they're very far apart | 13:32 |
| nickjohnson | The difference frequency being the very slow change in the DC offset (sorta) | 13:33 |
| kragen | xactly | 13:33 |
| kragen | and hopefully you can also see now that a phase shift in one of the input signals produces a phase shift in the output signal | 13:33 |
| nickjohnson | The sum frequency being... what frequency exactly? | 13:33 |
| kragen | well, roughly twice the original frequency | 13:33 |
| nickjohnson | Yes, that seems evident enough by the nature of the multiply operation | 13:33 |
| kragen | like how sin² x has a sine wave in it that oscillates at twice the frequency of the original | 13:33 |
| nickjohnson | Yes, okay, visually obviously twice the original: http://www.wolframalpha.com/input/?i=sin%28x%29+*+sin%281.1x%29%2C+sin%28x%29 | 13:34 |
| kragen | (which should also be evident from thinking about rotating vectors) | 13:34 |
| kragen | I remember being astonished when I first plotted sin² x with Quattro Pro on MS-DOS | 13:34 |
| kragen | "is that really a sine wave? how?" | 13:35 |
| nickjohnson | heh | 13:35 |
| kragen | anyway, so if you can repeat some phenomenon frequently while you're multiplying it by a sine wave and observing the result, you can downconvert the frequent phenomenon into a much less frequent phenomenon | 13:35 |
| nickjohnson | And yeah, evidence that abs is very nonlinear: http://www.wolframalpha.com/input/?i=abs%28sin%28x%29+*+sin%281.1x%29%29 | 13:36 |
| kragen | which is how SDR frontends work | 13:36 |
| nickjohnson | But if it happens less frequently, what happens to the information that was encoded in the occurrences of that phenomenon? | 13:37 |
| kragen | well, you're sort of hoping that all the repetitions are identical and so you can extract them stroboscopically | 13:38 |
| kragen | well, sort of stroboscopically | 13:38 |
| nickjohnson | I can see how you could decode phase shift keying, for instance. Your difference signal would phase shift at the same times, by the same phase quantities | 13:38 |
| kragen | stroboscopic "downconversion" works by multiplying your phenomenon by a periodic impulse rather than a sine wave, in order to "alias" it to a lower frequency | 13:39 |
| kragen | yes | 13:39 |
| nickjohnson | And as long as your difference signal is fast enough that you get a decent chunk of sine wave between each shift, you can still decode it | 13:39 |
| nickjohnson | Right, I remember seeing a bit on that. | 13:39 |
| nickjohnson | Likewise frequency shift keying would manifest as a proportional difference in your output frequencies | 13:39 |
| nickjohnson | (Proportional, or the same? The same, I guess?) | 13:39 |
| kragen | the same | 13:41 |
| kragen | so if your frequency shifts from 20MHz to 20.01MHz, your difference signal could shift from 1Hz to 10,001Hz | 13:42 |
| nickjohnson | right | 13:42 |
| kragen | A lot of times people do two-stage downconversion so they can filter out extraneous signals at an intermediate-frequency stage | 13:42 |
| kragen | which I should understand but don't | 13:43 |
| kragen | you can do all this in the spatial domain, too | 13:43 |
| kragen | moiré can provide you a kind of magnification of a periodic pattenr | 13:43 |
| nickjohnson | I was thinking about that, oddly | 13:44 |
| nickjohnson | It comes up all the time at train stations with those perforated screens | 13:44 |
| kragen | yup! | 13:44 |
| -!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 272 seconds] | 13:44 | |
| nmz787_i | there's also fooplot.com | 13:44 |
| nmz787_i | which has a github repo if you ever wanted to extend it | 13:44 |
| kragen | you can think of that as being a result of the nonlinear opacity operation combining two spatially periodic signals | 13:45 |
| nickjohnson | Neat | 13:45 |
| nickjohnson | kragen: Which are the same signal but, er, spatially frequency shifted by perspective | 13:45 |
| kragen | and one of the terms is the product of the two, so it has the difference frequency in it | 13:45 |
| kragen | right | 13:45 |
| kanzure | someone posted this silly stuff about google to hacker news https://www.mail-archive.com/kragen-tol@canonical.org/msg00268.html | 13:46 |
| -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap | 13:46 | |
| kragen | kanzure: what a bozo | 13:46 |
| kanzure | oh shit | 13:46 |
| kragen | haha | 13:46 |
| kragen | If you could get a layer of cells to be very periodic you could use this as a microscope | 13:47 |
| kanzure | huh now it's gone from the front page | 13:47 |
| kragen | it was probably controversial | 13:47 |
| kragen | I have the vague impression that what nmz787_i is doing is sort of like trying to get a linear sequence of cells to be very periodic | 13:47 |
| nmz787_i | umm, well there are holographic imagers/microscopes | 13:49 |
| nmz787_i | but I'm not trying to get 'ducks in a row'/cells-in-a-periodic-row | 13:49 |
| kragen | that's interesting! they're using the wave nature of light itself to provide the "multiplication by a sine wave"? | 13:49 |
| kragen | it was apparently controversial when it got posted four years ago and got voted up to 351 points: https://news.ycombinator.com/item?id=2933619 | 13:50 |
| nickjohnson | kragen: Reading your post, I'm sure you must have heard how reviled the "real names" policy was internally, too | 13:50 |
| nickjohnson | It's fixed now :P | 13:50 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 272 seconds] | 13:51 | |
| fenn | kragen doesn't seem so silly now | 13:51 |
| kragen | nickjohnson: naturally, yes :) | 13:51 |
| kragen | fenn: I always seem silly! | 13:51 |
| nickjohnson | It was always a colossally stupid idea | 13:51 |
| fenn | nickjohnson: did you used to hang out at hackerdojo? | 13:52 |
| nickjohnson | The fact it persisted so long is arguably an example of Google's slow reorganization from bottom-up engineering driven to top-down management driven | 13:52 |
| kragen | I used to hang out at SuperHappyDevHouse. Does htat count? | 13:52 |
| nickjohnson | fenn: Nope | 13:52 |
| * nickjohnson goes back to designing this IBM punched card reader | 13:54 | |
| nmz787_i | kragen: so you posted that? did you used to work there and quit or you were just making a general statement? | 13:56 |
| kragen | No, I never worked there. A lot of my friends work there or used to work there. | 13:56 |
| kanzure | i was just confused to see him appearing in two places at the same time | 13:57 |
| kanzure | async is weird like that | 13:57 |
| kragen | Instead, I moved to South America so I could watch teenagers sending each other naked pictures of themselves on Facebook and get laughed at by people who don't understand why that's important | 13:57 |
| kragen | I left Silicon Valley in 2006 and moved into a Volkswagen bus | 13:58 |
| kanzure | wow so hipster. ugh. | 13:58 |
| kragen | haha | 13:58 |
| kanzure | (i'm saying that as someone with family that has chosen to live in a volkswagen bus) | 13:58 |
| kragen | yeah, it was just me and Beatrice at the time | 13:58 |
| kragen | it was sure a crash course in auto mechanics | 13:59 |
| nmz787_i | by family do you mean Beatrice, or that like, your parents also decided to move into a VW? | 14:00 |
| nmz787_i | or do you now also have kids? | 14:00 |
| kanzure | i was the one who said family | 14:00 |
| kanzure | he was the one that said beatrice | 14:00 |
| fenn | kanzure you're living in a bus now? :P | 14:01 |
| fenn | so hipster | 14:01 |
| nmz787_i | the party bus | 14:01 |
| fenn | the bitcoin bus/boat | 14:01 |
| kanzure | "with family" does not mean i am "with family" | 14:01 |
| kanzure | it means that i have the unfortunate association of involvement in some family that happens to do the thing | 14:02 |
| fenn | maybe you have a tapeworm | 14:02 |
| nmz787_i | 1.5m long one | 14:02 |
| fenn | or a family of tapeworms | 14:02 |
| nmz787_i | in your ear | 14:02 |
| fenn | in your eye | 14:02 |
| nmz787_i | in your 'thinkin box' | 14:02 |
| fenn | ok i'll stop | 14:02 |
| kanzure | http://media.liveauctiongroup.net/i/5557/8524972_1.jpg?v=8CBEB089E9CD270 | 14:03 |
| fenn | i like the idea of living in a microbus, but they seem poorly engineered for the purpose | 14:03 |
| kragen | hahah | 14:04 |
| fenn | low insulation mainly | 14:04 |
| kragen | fenn: they're brilliantly engineered for living in, at least in warmish places. they just suck at driving | 14:04 |
| kragen | they're small enough that a small propane heater heats them adequately most places | 14:04 |
| fenn | and how do you not die from CO poisoning? | 14:04 |
| nmz787_i | I've seen some conversions to upgrade them to a newer TDI diesel engine | 14:05 |
| fenn | kanzure that bus is in great condition | 14:05 |
| kanzure | not theirs | 14:05 |
| kanzure | just some random pic | 14:05 |
| kanzure | you'd think that driving around in a vw bus since 1995 would yield at least one pic on the web but nope | 14:05 |
| kragen | mine was a Vanagon | 14:06 |
| fenn | there's one down the street just like that with "ich nein hasse kugelshrieber" written on it | 14:06 |
| fenn | what's the thing on top? | 14:07 |
| kragen | a tree | 14:07 |
| kragen | in http://media.liveauctiongroup.net/i/5557/8524972_1.jpg?v=8CBEB089E9CD270? | 14:07 |
| -!- Beatzebub [~beatzebub@S0106b81619e8ecee.gv.shawcable.net] has joined ##hplusroadmap | 14:07 | |
| nmz787_i | 'i hate no pen' is what google translate shows | 14:07 |
| nmz787_i | so he must be a politician | 14:08 |
| kanzure | mandatory diversity pic https://xa.yimg.com/kq/groups/9200033/homepage/name/788016.jpg?type=hr | 14:08 |
| fenn | i'm probably remembering it wrong, but it didn't make sense anyway | 14:08 |
| nmz787_i | 'We are sorry, you can not display images hosted by Yahoo! Groups on non Yahoo! Groups pages' | 14:08 |
| kanzure | (hint: they're the tall white ones) | 14:08 |
| kragen | https://secure.flickr.com/photos/blmurch/tags/magic-bus has photos of ours, with my ex-mother-in-law in them, and photos of the cracked cylinder heads | 14:08 |
| nickjohnson | Anyway, bed for me | 14:09 |
| nickjohnson | Thanks for the conversations :) | 14:09 |
| fenn | https://secure.flickr.com/photos/blmurch/607214287/ <- a bus | 14:09 |
| fenn | oh this is beatrice | 14:10 |
| kragen | it certainly is | 14:10 |
| kragen | were you asking what the thing on top of that bus is? | 14:10 |
| fenn | i was asking about https://en.wikipedia.org/wiki/File:Vanagon_Santana_VW.png | 14:11 |
| fenn | i thought they all had the thing | 14:11 |
| fenn | i guess it's just a cargo pod | 14:11 |
| kragen | it's a pop top | 14:11 |
| kragen | that one is an aftermarket pop top | 14:11 |
| kragen | with a sort of pantograph parallel linkage to raise it vertically | 14:11 |
| fenn | is it just for ventilation? | 14:12 |
| fenn | hm no that is a cargo pod. there are pop tops on the same page though | 14:13 |
| kragen | no, you pop it up when you park | 14:14 |
| kragen | it expands the volume of the bus by almost a meter vertically, which means you can stand up inside in front of the stove and sink, and provides a bunk that can comfortably accommodate two more people | 14:14 |
| kragen | you can also put cargo in it, but only while you're parked | 14:14 |
| kanzure | i need a recommendation for a tool to convert from irssi-style irc log format to html-with-timestamp-anchors format. | 14:15 |
| nmz787_i | kanzure: wouldn't that uglify grepping the logs? | 14:16 |
| fenn | how about just appending anchor tags to the txt file | 14:16 |
| nmz787_i | kanzure: could it do the conversion on-the-fly as they're requested? | 14:16 |
| kanzure | fenn: no | 14:16 |
| * fenn hates html logs | 14:16 | |
| kanzure | nmz787_i: no thanks | 14:16 |
| kanzure | fenn: i would never replace plaintext logs with only-html logs | 14:17 |
| fenn | i mean adding </a><a id="123456"> 12:34:56 < joebob> hay guyz | 14:18 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 14:18 | |
| kanzure | yes i know what you mean | 14:18 |
| kanzure | you were already very specific, i don't know how you could expect me to not have understood that | 14:18 |
| fenn | um, i managed to interpret it a few different ways | 14:19 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Max SendQ exceeded] | 14:21 | |
| nmz787_i | well fenn's example shows the anchor tags interspersed in the text, appending would be at the end of the file | 14:21 |
| fenn | time.mktime((year, month, day, hour, min, 0, 0, 0, -1)) | 14:22 |
| fenn | i don't think there's an off the shelf tool to do this | 14:24 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 14:24 | |
| kanzure | he meant appending after every write | 14:24 |
| fenn | i don't know what i meant, but you probably have an existing body of logs you wish to convert, and also an ongoing logging process that needs to be modified | 14:25 |
| nmz787_i | .tell nickjohnson do you know what the difference between say the DDS60 (or your project) and these ~$5 modules on ebay (search AD9850 ) | 14:29 |
| yoleaux | nmz787_i: I'll pass your message to nickjohnson. | 14:29 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Max SendQ exceeded] | 14:37 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 14:39 | |
| -!- juri_ [~juri@vpn166.sdf.org] has quit [Ping timeout: 264 seconds] | 15:01 | |
| -!- juri_ [~juri@vpn166.sdf.org] has joined ##hplusroadmap | 15:01 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 15:06 | |
| -!- delinquentme [~dingo@74.61.157.78] has quit [Read error: Connection reset by peer] | 15:08 | |
| -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 15:13 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 265 seconds] | 15:15 | |
| chris_99 | anyone seen these before http://ibidi.com/xtproducts/en/ibidi-Labware/Customer-Specific-Slides/Custom-Specific-Flow-Slides-and-Channels | 15:26 |
| chris_99 | i was wondering if you could easily use something like the branching channel one | 15:26 |
| chris_99 | to mix dye with cells | 15:26 |
| jrayhawk | irclog2html.py anchors timestamps, if that's close enough | 15:26 |
| jrayhawk | does CSS coloration, though, which is a bit lame | 15:27 |
| jrayhawk | coloration should be done in HTML for compatibility with lynx/links/w3m | 15:28 |
| -!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has joined ##hplusroadmap | 15:28 | |
| -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap | 15:30 | |
| jrayhawk | oh, you specified timestamps, nevermind | 15:32 |
| jrayhawk | i mentally replaced that with "anchors specific lines" | 15:32 |
| kanzure | anchoring specific lines is fine | 15:33 |
| kanzure | probably? | 15:33 |
| jrayhawk | yeah, doing that or both seems optimal, but i don't think i have ever seen it done | 15:33 |
| -!- sivoais [~zaki@unaffiliated/sivoais] has quit [Ping timeout: 264 seconds] | 15:38 | |
| chris_99 | you guys seen this - http://www.microfluidic-chipshop.eu/Download/Lab-on-a-Chip%20Catalogue_032014.pdf | 15:42 |
| -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has quit [Read error: Connection reset by peer] | 15:48 | |
| -!- strangewarp [~strangewa@c-76-25-206-3.hsd1.co.comcast.net] has joined ##hplusroadmap | 15:57 | |
| nmz787_i | chris_99: I used ibidi stuff for that cell-flow monitoring i mentioned earlier | 16:00 |
| chris_99 | ah i think i was on that site before | 16:01 |
| chris_99 | i'll register on their site then to get the prices | 16:01 |
| chris_99 | eeeek | 16:04 |
| chris_99 | they're not cheap | 16:04 |
| chris_99 | e.g. http://www.thistlescientific.co.uk/acatalog/u-slide-iii-3in1.html | 16:04 |
| nmz787_i | chris_99: so you'd probably want something with two inputs, an in-line backflow prevention valve for the cell-line (maybe), and some mixing posts... and either electrodes or optically clear so you can hook up to a microscope and camera or a spectrometer/photometer | 16:15 |
| nmz787_i | I am not sure if PDMS swells with alcohol, but I would assume if it does it's minor | 16:16 |
| chris_99 | yeah i'm just looking at ones with reaction chambers | 16:20 |
| chris_99 | *mixing | 16:20 |
| chris_99 | regarding the pumps, what kind of pump did you use | 16:20 |
| -!- sivoais [~zaki@unaffiliated/sivoais] has joined ##hplusroadmap | 16:26 | |
| chris_99 | from what i can tell the two main types are peristalic and syringe | 16:27 |
| fenn | gravity works for some things too | 16:27 |
| chris_99 | mmm | 16:29 |
| fenn | "When you started to publish the Whole Earth Catalogue in 1968, you said that you wanted to create a database so that “anyone on Earth can pick up a telephone and find out the complete information on anything.” Is that the idea of the internet, before the internet? | 16:29 |
| fenn | Brand: Right, I had forgotten about that quote. Isn’t it nice that I didn’t have to go through the work of collecting that information, it just happened organically. Some people say to me that I should revive the catalogue and my answer is: The internet is better than any catalogue or encyclopedia could ever be." | 16:30 |
| -!- sivoais_ [~zaki@199.19.225.239] has joined ##hplusroadmap | 16:31 | |
| -!- nmz787_i [nmccorkx@nat/intel/x-esjyfkctowipchgy] has quit [Ping timeout: 250 seconds] | 16:33 | |
| fenn | kanzure/faceface_ there's your pre-internet wikipedia | 16:34 |
| kanzure | if you say so | 16:34 |
| fenn | not really, it was heavily biased | 16:35 |
| kanzure | you said web yesterday not pre-internet | 16:35 |
| kanzure | stop changing the rules on me | 16:35 |
| fenn | whatever | 16:35 |
| fenn | i would hate to have to use a telephone to browse wikipedia | 16:36 |
| kanzure | you guys act like wikipedia is some magical thing | 16:36 |
| fenn | it is magical | 16:36 |
| kanzure | it's just a set of files that get edited | 16:36 |
| kanzure | (and then some overbearing politics, but let's not mention that) | 16:37 |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Ex-Chat] | 16:44 | |
| kragen | the overbearing politics are probably what make it magical | 16:45 |
| kragen | I mean they're what determine what edits survive | 16:47 |
| kragen | and what edits get made in the first place | 16:48 |
| kanzure | yes the ability to edit pages is not novel or unique to wikipedia or any wiki for that matter | 16:48 |
| fenn | uh what internet are you on | 16:49 |
| kragen | Wikipedia without the overbearing politics is just the c2.com Wiki | 16:49 |
| fenn | "the one with RFC-123456 that lets you edit pages, duh" | 16:49 |
| fenn | webdav? | 16:50 |
| fenn | anyway c2.com sucks | 16:51 |
| fenn | you get 5 pages down and realize you've read the same thing multiple times but nobody bothered reading that far before re-posting | 16:51 |
| kragen | because of the lack of overbearing politics, imo | 16:51 |
| fenn | can you think of something with the ability to edit pages pre-wiki? | 16:53 |
| kragen | pre-wiki is pretty early | 16:54 |
| kragen | like 1994 I think | 16:54 |
| fenn | hm i had never heard of wikis until around 2000 i guess | 16:55 |
| kragen | 1995 | 16:55 |
| kragen | 2000 was the first time I installed wiki | 16:56 |
| kragen | it seems like there must have been web pages editable through the web before 1995 but I can't remember any | 16:57 |
| fenn | heh "Writeups in E1 were limited to 512 bytes in size" so that's why they're called "nodes" | 16:57 |
| kragen | even Block Stackers Everything was a couple years later | 16:57 |
| kragen | and that wasn't, I think, editable. just writable. there were I think some things like that on Philip Greenspun's site | 16:57 |
| kragen | comment sections | 16:58 |
| fenn | E1 was started march 1998 | 16:58 |
| kragen | three years later, then | 16:58 |
| kragen | in 1996 it was still unusual to use dynamically generated web pages | 16:58 |
| kragen | I mean CGI dates from before that obviously | 16:58 |
| kragen | but people didn't run their web site off it | 16:59 |
| kragen | AuctionWeb might have been contemporary. and ViaWeb. | 16:59 |
| fenn | there were things like FAQs and HOWTOs but they were usually something like "email this person" and they would incorporate your changes if they liked them | 16:59 |
| kragen | yeah, there was a thing called faq-o-matic at some point | 16:59 |
| fenn | i kinda miss HOWTOs | 17:00 |
| kragen | eBay was founded in September 1995 | 17:00 |
| kragen | already post-Wiki | 17:00 |
| fenn | how can cunningham have started so early, the www barely existed then | 17:01 |
| kragen | he's a very smart guy | 17:01 |
| kragen | I think he's currently working on a thing he calls the Simplest Federated Wiki | 17:01 |
| fenn | SFW | 17:01 |
| kragen | which is like a decentralized Wiki that supports varying content types | 17:01 |
| kragen | including bytebeat! | 17:02 |
| nmz787 | i saw him once at a local bar | 17:02 |
| kragen | I met him in 2000 because he was interested in Comet-enabling Wiki, in particular for federation | 17:02 |
| nmz787 | well, the bar was hosting a gizmo meetup | 17:02 |
| kragen | Craigslist didn't go web-based until 1996 | 17:02 |
| kragen | Viaweb was founded July 1995 | 17:03 |
| kragen | which I think is barely post-Wiki | 17:03 |
| nmz787 | i didn't know craigslist existed before like 2005 or something | 17:03 |
| kragen | yeah, 1995 | 17:03 |
| kragen | but it was a mailing list at the time | 17:03 |
| nmz787 | i thought it started briefly before I first learned of it | 17:03 |
| nmz787 | :P | 17:04 |
| kragen | people think that about everything | 17:04 |
| -!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 240 seconds] | 17:06 | |
| kragen | I can't tell how old FAQ-O-Matic is | 17:07 |
| kragen | it might predate wiki but I don't think it does | 17:07 |
| kragen | there was a thing called HyperNews | 17:07 |
| kragen | an attempt to do Usenet on the Web | 17:08 |
| kragen | still at hypernews.cern.ch apparently | 17:08 |
| kragen | presumably you couldn't edit things you'd already posted but at least you could respond to them | 17:08 |
| fenn | that doesn't count | 17:10 |
| bbrittain | what's a diff except a response? | 17:10 |
| fenn | it would still be up to the reader to apply a patch | 17:10 |
| fenn | or 9 million patches | 17:11 |
| bbrittain | don't assume you have a lazy reader :P | 17:11 |
| fenn | sure you could implement a newsgroup patch protocol with client-side patching, but it didn't happen as far as i know | 17:13 |
| nmz787 | how would that work when someone responds to another "YOU'RE WRONG" | 17:15 |
| fenn | it also breaks if you don't have a complete history | 17:18 |
| -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap | 17:26 | |
| -!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has quit [Remote host closed the connection] | 17:32 | |
| -!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has joined ##hplusroadmap | 17:36 | |
| -!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has quit [Remote host closed the connection] | 17:36 | |
| -!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has joined ##hplusroadmap | 17:36 | |
| -!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 258 seconds] | 17:38 | |
| -!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has quit [Ping timeout: 240 seconds] | 17:40 | |
| -!- sivoais_ [~zaki@199.19.225.239] has quit [Quit: leaving] | 17:41 | |
| -!- ebowden [~ebowden@CPE-121-223-157-226.lns2.bat.bigpond.net.au] has joined ##hplusroadmap | 17:53 | |
| -!- Viper168_ is now known as Viper168 | 17:54 | |
| -!- Netsplit *.net <-> *.split quits: HEx1, maaku, ThomasEgi, kenju254, audy, Vutral, rk[1] | 18:00 | |
| -!- lichen [~lichen@c-50-139-11-6.hsd1.or.comcast.net] has quit [Quit: Lost terminal] | 18:02 | |
| -!- Netsplit over, joins: Vutral | 18:12 | |
| -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap | 18:18 | |
| -!- delinquentme [~dingo@74.61.157.78] has quit [Remote host closed the connection] | 18:18 | |
| -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap | 18:19 | |
| nmz787 | this is tempting to buy http://www.thermal.com/what_is_thermal_desktop.html | 18:27 |
| -!- eridu [~eridu@gateway/tor-sasl/eridu] has joined ##hplusroadmap | 18:27 | |
| nmz787 | .wik http://en.wikipedia.org/wiki/Vanadium(IV)_oxide#Infrared_radiation_management | 18:29 |
| yoleaux | "Vanadium(IV) oxide is the inorganic compound with the formula VO2. It is a dark blue solid. Vanadium(IV) oxide is amphoteric, dissolving in non-oxidising acids to give the vanadyl ion, [VO]2+ and in alkali to give the [V4O9]2− ion, or at high pH [VO4]4−." — http://en.wikipedia.org/wiki/Vanadium(IV)_oxide | 18:29 |
| nmz787 | "The thermochromic phase transition between the transparent semiconductive and reflective conductive phase, occurring at 68 °C, can happen in times as short as 100 femtoseconds" | 18:29 |
| nmz787 | "Vanadium dioxide can act as extremely fast optical shutters, optical modulators, infrared modulators for missile guidance systems, cameras, data storage, and other applications." | 18:30 |
| fenn | finally a usb thermal camera | 18:30 |
| fenn | $200 is not bad | 18:30 |
| nmz787 | no, especially compared to the other's avail now | 18:31 |
| nmz787 | FLIR has one for iPhone that is only 60x80 pixels | 18:31 |
| fenn | i don't care about pixels really | 18:31 |
| nmz787 | that one is 206x156 | 18:31 |
| nmz787 | and I think the FLIR one was also like 3 microns | 18:31 |
| nmz787 | http://www.flir.com/uploadedFiles/CVS_Americas/Cores_and_Components_NEW/Products/Uncooled_Cores/Lepton/FLIR-Lepton-DataBrief.pdf | 18:32 |
| nmz787 | oh, no it's 8-14 too | 18:32 |
| nmz787 | oh, same frame rate (9hz) | 18:32 |
| -!- kenju254 [~kenju254@static-41-242-0-196.ips.angani.co] has joined ##hplusroadmap | 18:32 | |
| -!- rk[1] [~rak@opensource.cse.ohio-state.edu] has joined ##hplusroadmap | 18:32 | |
| -!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap | 18:32 | |
| -!- HEx1 [~HEx@hexwab.plus.com] has joined ##hplusroadmap | 18:33 | |
| -!- maaku [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has joined ##hplusroadmap | 18:33 | |
| -!- audy [~audy@107.170.175.57] has joined ##hplusroadmap | 18:33 | |
| nmz787 | their 'shop' has a place for a coupon code, but I can't find any online | 18:33 |
| kanzure | .title http://www.youtube.com/watch?v=WPE7mU3myfk | 18:34 |
| yoleaux | Cellular Mechanisms of Transcranial Direct Current Stimulation (tDCS) - YouTube | 18:34 |
| fenn | what is the mechanism? | 18:34 |
| fenn | don't make me watch an hour of video to find out "more research is needed" | 18:35 |
| kanzure | i wasn't planning on actually watching it right now | 18:36 |
| kanzure | ybit's more the "watch a video for an hour" type | 18:36 |
| fenn | heath: i'll be waiting for your summary :P | 18:36 |
| * nmz787 put it on the TV | 18:38 | |
| kanzure | tdcs user videos? https://www.youtube.com/playlist?list=PLHl2ybrb92mWwoJWoMPK0j7mCnrEeQexI | 18:38 |
| -!- Netsplit *.net <-> *.split quits: HEx1, audy, maaku | 18:40 | |
| nmz787 | wait a sec, I just talked to a friend recently who told me he had an electrode put in his brain to give him more neurotransmitter because the side-effects from the meds he had been taking were getting too bad | 18:41 |
| kanzure | .title http://www.ncbi.nlm.nih.gov/pubmed/22949089 | 18:41 |
| nmz787 | and now he doesn't have to take any meds or at least less | 18:41 |
| yoleaux | Transcranial current brain stimulation (tCS): models and technologies. - PubMed - NCBI | 18:41 |
| kanzure | "Neuroanatomical, Neurophysiological and Neuropsychological Terminology" https://faculty.washington.edu/chudler/neuroroot.html | 18:41 |
| -!- eridu [~eridu@gateway/tor-sasl/eridu] has quit [Ping timeout: 250 seconds] | 18:42 | |
| fenn | nmz787: that's probably deep brain stimulation, which is considered to be a different thing | 18:42 |
| kanzure | but still a cool thing | 18:42 |
| fenn | yeah but definitely not "DIY" | 18:42 |
| nmz787 | i was super surprised that he said that | 18:42 |
| kanzure | "average cost for subthalamic nucleus deep brain stimulation implantation with microelectrode recording per patient is $26,764.79 for unilateral" | 18:43 |
| bbrittain | I wanna hook a battery up to my head and be smarter | 18:44 |
| bbrittain | hell, I wann hook up 20 battries | 18:44 |
| bbrittain | let's hope it scales linearly | 18:44 |
| kanzure | ultrasound. | 18:45 |
| nmz787 | or exponentially | 18:45 |
| nmz787 | really? pressure waves? | 18:45 |
| kanzure | hmm i wonder how deep those deep brain stimulation electrodes need to go | 18:45 |
| kanzure | because ultrasound will be way cheaper than $26k | 18:45 |
| kanzure | although harder to position around your head on a regular basis | 18:45 |
| kanzure | compared to having something static/fixed. hrm. | 18:45 |
| nmz787 | that sounds less effective/hackable | 18:45 |
| nmz787 | less hackability | 18:45 |
| kanzure | deep brain stimulation is the one that is less hackable | 18:46 |
| nmz787 | tcds? | 18:46 |
| kanzure | drill -> brain surgery -> close -> go back to brain surgery to fix anything | 18:46 |
| kanzure | you said electrode in his brain | 18:46 |
| kanzure | that's not tcds | 18:46 |
| -!- HEx1 [~HEx@hexwab.plus.com] has joined ##hplusroadmap | 18:46 | |
| -!- maaku [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has joined ##hplusroadmap | 18:46 | |
| -!- audy [~audy@107.170.175.57] has joined ##hplusroadmap | 18:46 | |
| nmz787 | isn't chemistry and electronics based on... electrons, not pressure differentials? | 18:46 |
| nmz787 | oh, yeah that friend | 18:46 |
| kanzure | ultrasound has been shown to cause action potentials | 18:47 |
| * nmz787 is watching tcds on tv | 18:47 | |
| kanzure | http://diyhpl.us/~bryan/papers2/neuro/ultrasound/ | 18:47 |
| nmz787 | bbrittain: your song can only use words from https://faculty.washington.edu/chudler/neuroroot.html | 18:47 |
| nmz787 | or 2 degrees of separation from them | 18:48 |
| nmz787 | or maybe just 1 degree | 18:48 |
| kanzure | i should really use sulcus more often | 18:48 |
| fenn | so one mechanism they propose is "bimodal polarization" of neurons, where one end becomes positively charged by some number of mV and the other end is negatively charged; since pyramidal neurons always have the soma (the "important end") at the bottom, the soma is positively charged and will fire more actively | 18:54 |
| fenn | this charge happens because V=IR and the R is across the neuron | 18:55 |
| kanzure | "at the bottom" | 18:55 |
| kanzure | hm? | 18:55 |
| kanzure | what is my frame of reference | 18:55 |
| fenn | in this reference frame "top" is the skull | 18:55 |
| kanzure | huh. okay. | 18:56 |
| fenn | of course the cortex has wrinkles | 18:56 |
| kanzure | sulcri! | 18:56 |
| kanzure | someone's not paying attention | 18:56 |
| nmz787 | 'tcds changes the sensitivity to tms' | 18:56 |
| fenn | "this is a first-order approximation to start the thought process" | 18:56 |
| nmz787 | what is tms? | 18:56 |
| kanzure | transcranial magnetic stimulation | 18:56 |
| fenn | "0.3mV doesn't sound like very much when action potential is 100mV" | 18:58 |
| kanzure | http://i.imgur.com/5iHr8fz.jpg | 18:59 |
| fenn | what do you have that tagged as? | 19:02 |
| kanzure | bishop | 19:02 |
| fenn | you should eat some raw garlic, just a little | 19:06 |
| kanzure | hm? | 19:08 |
| kanzure | will that stop my eye from falling out? | 19:08 |
| fenn | potentially | 19:09 |
| kanzure | anyway i'm like 98% sue that show was written based on me | 19:10 |
| fenn | was it from "fringe"? | 19:10 |
| kanzure | you didn't even do a reverse image search :( | 19:10 |
| kanzure | what's the point of having gigantic image indexes if nobody uses them | 19:10 |
| delinquentme | are there legal implications for journalists who fuck up being journalists? | 19:11 |
| fenn | tineye kinda sucks and google image search only works with javascript | 19:11 |
| nmz787 | how can you find an IRC user, like what room they are in? | 19:11 |
| fenn | "Placement of the dot around a glyph may indicate digital value, or cypher coding, in addition to the alphabetic value that is represented." | 19:12 |
| fenn | nmz787: freenode turned that off (mode invisible by default) | 19:15 |
| nmz787 | that video was kinda lame | 19:17 |
| nmz787 | he mostly talked about cells in a petri dish and electrical concepts | 19:17 |
| nmz787 | I guess it was a room full of biologists | 19:17 |
| nmz787 | but like the top and bottom voltage offset stuff is stuff the electron/ion microscope user group was talking about last month | 19:18 |
| nmz787 | except they were interested in a specimen, not a cell | 19:18 |
| fenn | imaging artifacts? those are much higher voltages | 19:18 |
| nmz787 | to get around charging and stuff sometimes, or to deflect certain electrons from getting to the detector... to get better signal from the other types of electrons | 19:19 |
| fenn | i don't buy this mechanism as the entire explanation because tDCS has an aftereffect that lasts for hours to months | 19:22 |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-vphbxwbezadqurkd] has quit [Quit: Connection closed for inactivity] | 19:27 | |
| kanzure | "Compute Engine VMs boot within 30 seconds [7] which is considered to be 4-10x faster than the competition." | 19:30 |
| kanzure | sigh | 19:30 |
| catern | 30 seconds isn't very fast for full boot time | 19:30 |
| catern | i could put together a VM host that boots faster than that with just tools from around the house | 19:31 |
| catern | clone a copy on write disk image, start qemu, done | 19:31 |
| fenn | systemd fanboys claim 900ms boot time | 19:31 |
| catern | fenn: i was assuming most of the time was spent in allocation | 19:32 |
| catern | yeah, the OS itself shouldn't take more than 2 seconds | 19:32 |
| catern | in a virtualized environment | 19:32 |
| catern | (okay, exaggeration, maybe on first boot you want to do some things, so let's say 5 seconds) | 19:33 |
| catern | full system containers can be allocated and booted and networked in 100ms or so | 19:33 |
| catern | (something like that) | 19:33 |
| nmz787 | i was looking at tablets the other day, trying to find a cheap ($200 ish) device with a 1080 screen.... and learned that windows RT 8.1 doesn't allow apps to run unless they're from the MS store | 19:34 |
| nmz787 | unless it's jailbroken or something | 19:34 |
| nmz787 | fenn: you may take interest in https://cloud.sagemath.com | 19:40 |
| fenn | i never developed the enthusiasm for hieroglyphics that my contemporaries have | 19:42 |
| fenn | that page is hilariously badly rendered in dillo btw | 19:43 |
| kanzure | where's your tapeworm? | 19:43 |
| fenn | suprachiasmatic nucleus | 19:44 |
| fenn | "clock gene" what a remarkably clearly named gene | 19:46 |
| fenn | "Drugs that activate dopamine receptors speed up ones perception of time, while dopamine antagonists cause one to feel that time is passing slowly" | 19:57 |
| fenn | does that mean "time flies when you're having fun" | 19:58 |
| kanzure | hmph | 19:59 |
| -!- yashgaroth [~ffffff@2606:6000:cb85:6a00:991c:987c:d2fe:b825] has joined ##hplusroadmap | 20:00 | |
| fenn | "In his book "Awakenings", Oliver Sacks discusses how patients with Parkinson's disease experience deficits in their awareness of time and tempo. For example, Mr E, when asked to clap his hands steadily and regularly did so for the first few claps before clapping faster and irregularly; culminating in an apparent freezing of motion. When he finished, Mr E asked if his observers were glad he did | 20:05 |
| fenn | it correctly to which they replied "no". Mr E was offended by this because to him, his claps were regular and steady." | 20:05 |
| fenn | your sense of time offends me, sir | 20:05 |
| -!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 255 seconds] | 20:12 | |
| -!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Remote host closed the connection] | 20:18 | |
| -!- delinquentme [~dingo@74.61.157.78] has joined ##hplusroadmap | 20:21 | |
| kanzure | .title https://www.youtube.com/watch?v=K1Ix6dRNNQs | 20:36 |
| yoleaux | steve lovesey-soaring high-alfonso muchacho remix - YouTube | 20:36 |
| kanzure | haha they let star trek on youtube now? https://www.youtube.com/watch?v=hQwoT5y7R4Y | 20:48 |
| kanzure | "Computer production requires optical and electronic devices far beyond the capability of an individual to produce. The crystals of silicon used to produce most computer chips require sophisticated ovens and production facilities that few people could build. The tools used to manufacture wafers from silicon crystals depends on generations of development that could not be built in even a sophisticated machine shop. Even something as seemingly ... | 20:53 |
| kanzure | ... simple as a pocket calculator was beyond the capability of the technology that put a man on the moon in the 1960s. | 20:53 |
| kanzure | i hate this person already | 20:53 |
| kanzure | and he calls himself "the librarian"? man librarians suck | 20:54 |
| kanzure | jokes on him because he's missing out on the cool name "dread pirate archivist" | 20:55 |
| kanzure | "dread pirate just minding my own business" | 20:55 |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Read error: Connection timed out] | 21:17 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 21:27 | |
| -!- juri_ [~juri@vpn166.sdf.org] has quit [Ping timeout: 250 seconds] | 21:37 | |
| -!- JayDugger [~jwdugger@pool-173-57-55-138.dllstx.fios.verizon.net] has joined ##hplusroadmap | 21:39 | |
| JayDugger | Good evening, everyone. | 21:41 |
| -!- gene_hacker [~chatzilla@c-50-137-46-240.hsd1.or.comcast.net] has joined ##hplusroadmap | 21:51 | |
| -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 244 seconds] | 21:52 | |
| -!- gene_hacker [~chatzilla@c-50-137-46-240.hsd1.or.comcast.net] has quit [Ping timeout: 250 seconds] | 21:59 | |
| -!- gene_hacker [~chatzilla@c-50-137-46-240.hsd1.or.comcast.net] has joined ##hplusroadmap | 21:59 | |
| -!- Jaakko9113 [~Jaakko@host86-155-144-71.range86-155.btcentralplus.com] has joined ##hplusroadmap | 22:04 | |
| -!- Vutral [ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 22:08 | |
| -!- yashgaroth [~ffffff@2606:6000:cb85:6a00:991c:987c:d2fe:b825] has quit [Quit: Leaving] | 22:23 | |
| -!- gene_hacker [~chatzilla@c-50-137-46-240.hsd1.or.comcast.net] has quit [Ping timeout: 272 seconds] | 22:39 | |
| -!- gene_hacker [~chatzilla@c-50-137-46-240.hsd1.or.comcast.net] has joined ##hplusroadmap | 22:40 | |
| -!- juri_ [~juri@vpn166.sdf.org] has joined ##hplusroadmap | 22:55 | |
| -!- gene_hacker [~chatzilla@c-50-137-46-240.hsd1.or.comcast.net] has quit [Ping timeout: 272 seconds] | 23:03 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 256 seconds] | 23:16 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 23:23 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Max SendQ exceeded] | 23:29 | |
| -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 23:30 | |
| -!- delinquentme [~dingo@74.61.157.78] has quit [Ping timeout: 240 seconds] | 23:57 | |
| --- Log closed Sat Nov 22 00:00:21 2014 | ||
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!
