--- Log opened Tue Oct 21 00:00:22 2014 | ||
-!- peteros [~asakharov@24.60.79.55] has left ##hplusroadmap [] | 00:06 | |
-!- Beatzebub [~beatzebub@S0106b81619e8ecee.gv.shawcable.net] has quit [Quit: No calling card for the unsung bard] | 00:28 | |
-!- pete4242 [~smuxi@boole.london.hackspace.org.uk] has joined ##hplusroadmap | 01:19 | |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 245 seconds] | 01:26 | |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has joined ##hplusroadmap | 01:51 | |
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap | 01:58 | |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 246 seconds] | 02:05 | |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has joined ##hplusroadmap | 02:07 | |
-!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 02:12 | |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 246 seconds] | 02:14 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 245 seconds] | 02:14 | |
-!- Viper168_ is now known as Viper168 | 02:25 | |
-!- |b| [~|d|@ip68-107-37-158.sd.sd.cox.net] has joined ##hplusroadmap | 03:19 | |
-!- JingoFett [~|d|@ip68-107-37-158.sd.sd.cox.net] has quit [Ping timeout: 245 seconds] | 03:20 | |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has joined ##hplusroadmap | 03:41 | |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 272 seconds] | 03:46 | |
-!- Vutral_ [~ss@31.7.56.132] has quit [Ping timeout: 258 seconds] | 03:52 | |
-!- Vutral_ [~ss@176.10.107.228] has joined ##hplusroadmap | 03:53 | |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 03:54 | |
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 260 seconds] | 04:02 | |
-!- Vutral [~ss@vutral.net] has joined ##hplusroadmap | 04:04 | |
-!- Vutral [~ss@vutral.net] has quit [Changing host] | 04:04 | |
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 04:04 | |
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Excess Flood] | 04:13 | |
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 04:17 | |
-!- TC [~talinck@66-161-138-110.ubr1.dyn.lebanon-oh.fuse.net] has quit [Remote host closed the connection] | 04:21 | |
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Excess Flood] | 04:23 | |
-!- lichen [~lichen@c-50-139-11-6.hsd1.or.comcast.net] has quit [Quit: Lost terminal] | 04:25 | |
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 04:25 | |
-!- Netsplit *.net <-> *.split quits: Merovoth | 04:37 | |
-!- gnusha [~gnusha@131.252.130.248] has joined ##hplusroadmap | 04:37 | |
-!- Topic for ##hplusroadmap: biohacking, nootropics, transhumanism, open hardware | sponsored by george church and the NRA, banned by the Federal Death Administration (4 times) | this channel is LOGGED: http://gnusha.org/logs | http://diyhpl.us/wiki | not intentionally unrepeatable | 04:37 | |
-!- Topic set by kanzure [~kanzure@131.252.130.248] [Fri Jun 6 17:48:33 2014] | 04:37 | |
[Users ##hplusroadmap] | 04:37 | |
[ altersid ] [ Daeken ] [ JayDugger ] [ sivoais ] | 04:37 | |
[ andytoshi ] [ dingo ] [ jrayhawk_ ] [ smeaaagle ] | 04:37 | |
[ archels ] [ docl_ ] [ juri_ ] [ strages ] | 04:37 | |
[ audy ] [ DonnchaC_ ] [ justanotheruser] [ strangewarp ] | 04:37 | |
[ augur_ ] [ dpk ] [ juul ] [ streety ] | 04:37 | |
[ balrog ] [ drazak ] [ kanzure ] [ superkuh ] | 04:37 | |
[ bbrittain ] [ drethelin ] [ kjskjskjs ] [ superobserver] | 04:37 | |
[ bkero ] [ drewbot ] [ night ] [ ThomasEgi ] | 04:37 | |
[ blueskin ] [ dvorkbjel ] [ nmz787 ] [ thundara ] | 04:37 | |
[ Burn_ ] [ ebowden ] [ nsh ] [ tigger ] | 04:37 | |
[ catern ] [ ElixirVitae] [ paperbot ] [ Twey ] | 04:37 | |
[ CheckDavid] [ faceface ] [ ParahSailin_ ] [ upgrayeddd ] | 04:37 | |
[ chris_99 ] [ fenn ] [ pasky ] [ Viper168 ] | 04:37 | |
[ comma8 ] [ gnusha ] [ pete4242 ] [ Vutral ] | 04:37 | |
[ cpopell2 ] [ HashNuke ] [ rak[1] ] [ Vutral_ ] | 04:37 | |
[ crescendo ] [ heath ] [ rigel ] [ yoleaux ] | 04:37 | |
[ d3vz3r0 ] [ hypron ] [ saurik ] [ yorick ] | 04:37 | |
[ d4de^^ ] [ ivan` ] [ sheena ] [ |b| ] | 04:37 | |
-!- Irssi: ##hplusroadmap: Total of 72 nicks [0 ops, 0 halfops, 0 voices, 72 normal] | 04:37 | |
-!- Channel ##hplusroadmap created Thu Feb 25 23:40:30 2010 | 04:37 | |
-!- Irssi: Join to ##hplusroadmap was synced in 8 secs | 04:37 | |
-!- kenju254 [~kenju254@static-41-242-0-196.ips.angani.co] has joined ##hplusroadmap | 04:37 | |
-!- helleshin [~talinck@66-161-138-110.ubr1.dyn.lebanon-oh.fuse.net] has joined ##hplusroadmap | 05:10 | |
-!- ebowden [~ebowden@147.69.54.84] has quit [Remote host closed the connection] | 05:28 | |
-!- juri_ [~juri@vpn166.sdf.org] has quit [Ping timeout: 265 seconds] | 06:00 | |
-!- hypron [~hypron@p8120-ipngn100105yosemiya.okinawa.ocn.ne.jp] has quit [Quit: hypron] | 06:20 | |
-!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has quit [Read error: Connection reset by peer] | 07:44 | |
-!- ElixirVitae [~Shehrazad@88.252.235.148] has joined ##hplusroadmap | 07:44 | |
-!- ElixirVitae [~Shehrazad@88.252.235.148] has quit [Changing host] | 07:44 | |
-!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has joined ##hplusroadmap | 07:44 | |
kanzure | .to fenn "The scarcity of consumer goods, in fact, helped promote of ubiquitous fix-it (*remont*) shops for small appliances wherever I walked in Moscow; try to find their counterpart in the capitals of the throwaway West." | 07:50 |
---|---|---|
yoleaux | kanzure: I'll pass your message to fenn. | 07:50 |
kanzure | http://communicationnation.blogspot.co.uk/2010/02/design-philosophy-of-ak-47.html | 07:51 |
-!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has quit [Read error: Connection reset by peer] | 07:52 | |
-!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has joined ##hplusroadmap | 07:52 | |
JayDugger | "Dr. Catsby Food Bowl for Whisker Relief?" | 08:05 |
JayDugger | 1) use a small plate, 2) use a saucer, 3) use a paper towel, 4) feed smaller portions | 08:06 |
kanzure | i think his point was that he wasn't aware the whiskers were a problem | 08:07 |
JayDugger | I think he's imagining the whiskers are a problem, or at worst, his sample size of 1 cat led him to overestimate the problem's frequency. | 08:09 |
kanzure | i wonder if whiskers are really a problem there | 08:14 |
JayDugger | Not in my experience, but I am not a cat. | 08:17 |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has joined ##hplusroadmap | 08:54 | |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 246 seconds] | 08:58 | |
-!- juri_ [~juri@vpn166.sdf.org] has joined ##hplusroadmap | 09:00 | |
-!- drewbot [~cinch@ec2-54-161-245-196.compute-1.amazonaws.com] has quit [Remote host closed the connection] | 09:14 | |
-!- drewbot [~cinch@ec2-54-80-140-243.compute-1.amazonaws.com] has joined ##hplusroadmap | 09:14 | |
-!- pete4242 [~smuxi@boole.london.hackspace.org.uk] has quit [Remote host closed the connection] | 09:24 | |
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 240 seconds] | 09:29 | |
kanzure | "Looking through their filed patents, it looks like their main innovation is a HMD device that uses a high speed digital zone plate that is used to focus/diverge light to form images at different depths. They have a paired high-speed mask LCD that blocks off parts of the images based on a z-buffer and finally a conventional imaging device (LCD/LCos/etc) that displays the left and right eye images. The zone plate and mask run 12X times the ... | 09:34 |
kanzure | ... frame rate of the imaging device, which runs at 30 or 60Hz. The result is that the users sees 12 frames (rather portions of a single frame) focuses at different depth levels for every image displayed bucketed by the z-buffer values in the image." | 09:34 |
kanzure | from https://news.ycombinator.com/item?id=8486801 | 09:34 |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has joined ##hplusroadmap | 09:35 | |
superkuh | Cool. | 09:35 |
-!- nmz787_i [~nmccorkx@134.134.137.71] has joined ##hplusroadmap | 09:37 | |
-!- nmz787_i1 [~nmccorkx@134.134.137.73] has joined ##hplusroadmap | 09:41 | |
-!- nmz787_i [~nmccorkx@134.134.137.71] has quit [Ping timeout: 255 seconds] | 09:42 | |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 246 seconds] | 09:44 | |
nmz787_i1 | superkuh: mfg got back to me, but only to ask what uF and voltage (replied 2Uf 2kV)... they also said they didn't MKP52 right now | 09:50 |
superkuh | Ah well. Like suggested, there's always microwave oven caps and a beefy power supply to deal with the bleeders. | 09:51 |
nmz787_i1 | superkuh: they also said it's some holiday there now, and they won't reply again til next monday | 09:51 |
nmz787_i1 | but the FMD52 are the AC ones anyhow | 09:52 |
nmz787_i1 | so that is what they seem to still be making | 09:52 |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 245 seconds] | 10:15 | |
fenn | great, i will just move to moscow and be a 3d printer repairman | 10:20 |
yoleaux | 14:50Z <kanzure> fenn: "The scarcity of consumer goods, in fact, helped promote of ubiquitous fix-it (*remont*) shops for small appliances wherever I walked in Moscow; try to find their counterpart in the capitals of the throwaway West." | 10:20 |
kanzure | soviet scientists are still super cheap to hire, right? | 10:21 |
fenn | in soviet russia | 10:21 |
fenn | which no longer exists | 10:22 |
fenn | um, actually the scientists were the highest "paid" of all | 10:22 |
kanzure | but their scientists didn't just do or anything | 10:22 |
kanzure | *die or anything | 10:22 |
kanzure | they're still around, just doing .. different stuff. | 10:22 |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 10:23 | |
fenn | take a look at the energia website http://www.buran-energia.com/boutique-shop/?language=en | 10:23 |
fenn | yes those are the scientists who designed and launched an orbiting laser cannon, selling t-shirts and coffee mugs | 10:23 |
kanzure | perfect | 10:24 |
fenn | (aforementioned laser cannon) http://www.buran-energia.com/polious/polious-desc.php | 10:24 |
kanzure | world's so fucking broken | 10:24 |
kanzure | oh look they accept paypal | 10:24 |
fenn | buran was a terrible idea, buy only because the space shuttle was a terrible design | 10:25 |
fenn | they actually had better designs but were ordered to make an imitation shuttle instead | 10:26 |
fenn | oh look you can buy pieces of it | 10:27 |
-!- [nsh] [~nsh@wikipedia/nsh] has joined ##hplusroadmap | 10:32 | |
-!- [nsh] [~nsh@wikipedia/nsh] has quit [Remote host closed the connection] | 10:46 | |
superkuh | I wonder how Magic Leap's device would handle the chromatic aberration of a zone plate lens. In VR HMD with a static set of lenses they can render the image to correct for it. But they'd need to apply different corrections to the display image for each optical depth. | 10:48 |
superkuh | And chromatic aberration is a lot worse with zone plates than refractive lenses. | 10:49 |
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 260 seconds] | 11:10 | |
fenn | doesnt it also need 12X the power to run, or conversely has 1/12 the contrast ratio of a system without a focus modulator layer | 11:11 |
fenn | i can't imagine needing more than 3 layers anyway | 11:12 |
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 11:16 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 244 seconds] | 11:17 | |
fenn | i'd rather have a plenoptic light field display | 11:21 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 11:24 | |
-!- sheena [~home@d154-20-132-7.bchsia.telus.net] has joined ##hplusroadmap | 11:24 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 244 seconds] | 11:29 | |
fenn | "Magic Leap and Weta Workshop are collaborating on a truly next-generation Dr. Grordbort’s first person shooter on a world-changing new platform in an effort to defend Earth from robotic overthrow" | 11:31 |
fenn | wow, that deserves $500m fer sure | 11:31 |
kanzure | maybe if you're snarky enough they will see the errors of your ways and grant you unlimited funding | 11:32 |
fenn | doesn't it seem like a lot of money? | 11:32 |
fenn | even assuming the thing does what it claims | 11:33 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 11:35 | |
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 260 seconds] | 11:36 | |
-!- _0bitcount [~big-byte@81.61.34.185.dyn.user.ono.com] has joined ##hplusroadmap | 11:38 | |
fenn | .title https://www.youtube.com/watch?v=deI1IzbveEQ | 11:41 |
yoleaux | NVIDIA's Light-field Glasses Prototype demo @ Siggraph 2013 - YouTube | 11:41 |
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 11:45 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 265 seconds] | 11:52 | |
chris_99 | interesting video | 11:56 |
yoleaux | 05:34Z <nmz787> chris_99: http://www.reddit.com/r/opencv/ has some good links, been reading through SIFT and a few others on here http://aishack.in/category/computer-vision/ | 11:56 |
chris_99 | ah cheers nmz787 | 11:56 |
kanzure | .g site:youtube.com siggraph 2014 | 11:57 |
yoleaux | http://www.youtube.com/user/ACMSIGGRAPH | 11:57 |
kanzure | uh, i meant "will see the errors of their ways", not your ways :) | 11:57 |
fenn | i'm not a fan of steampunk i guess | 11:58 |
-!- sheena [~home@d154-20-132-7.bchsia.telus.net] has quit [Ping timeout: 260 seconds] | 12:39 | |
fenn | "Think of it: A bunch of geniuses with the bodies of jocks each crafting superweapons in his basement to toss against his immediate neighbors (because they were bred for strength and smarts, but not cooperation) or spending too much time indoors reading and/or lifting weights to build a functioning society (because the Chief Eugenicist went a little overboard with the introversion)" | 12:44 |
fenn | http://i3.kym-cdn.com/photos/images/original/000/019/928/Steroids.jpg | 12:45 |
kanzure | .wik face of boe | 12:57 |
yoleaux | "The Face of Boe is a fictional character in the British science fiction television series Doctor Who. Not portrayed on-screen by an actor, the Face of Boe is a wholly mechanical effect, resembling a gigantic human-like head with a weathered face and in place of hair, numerous tendrils which terminate in round pod-like structures." — http://en.wikipedia.org/wiki/Face_of_boe | 12:57 |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Ex-Chat] | 12:59 | |
Viper168 | wait | 13:03 |
Viper168 | doctor who isn't real? | 13:03 |
fenn | yes virginia168 there is a santa claus | 13:05 |
fenn | your little friends are wrong. they have been effected by the skepticism of a skeptical age. they think that nothing can be which is not comprehensible by their little minds. | 13:06 |
kanzure | 12:26 < hearn> http://eprint.iacr.org/2014/838 - not exactly bitcoin related but still pretty amazing | 13:07 |
kanzure | 12:26 < hearn> 100x speedup for homomorphic encryption, when a GPU is in use | 13:07 |
kanzure | 12:26 < hearn> they used it to build a Bayesian spam filter that operates over encrypted messages | 13:07 |
fenn | .rot13 viagra | 13:08 |
yoleaux | ivnten | 13:08 |
fenn | mark as spam all messages containing 'ivnten' | 13:08 |
kanzure | i should throw together a bad blockchain idea generator before someone scoops me on that | 13:14 |
kanzure | {random_word} on ze blockchain | 13:14 |
fenn | or you could just do a blockchain idea generator and see what it generates | 13:15 |
fenn | .g ideonomy | 13:15 |
yoleaux | http://ideonomy.mit.edu/ | 13:15 |
kanzure | "Technical Consultant" | 13:15 |
kanzure | 404s :) | 13:15 |
archels | http://spectrum.ieee.org/green-tech/wind/norway-wants-to-be-europes-battery | 13:16 |
superkuh | "When the power-reversal scheme interrupts the circuit, the converters use subsea electrodes at either shore to feed the return currents across the strait through the water. " | 13:20 |
fenn | i didn't get that either | 13:21 |
fenn | is it just bleeding off inductance? | 13:21 |
fenn | the ocean is just a snubber resistor | 13:22 |
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 260 seconds] | 13:24 | |
kanzure | .title https://www.mitre.org/publications/technical-papers/presentation-extreme-privilege-escalation-on-windows-8uefi-systems | 13:33 |
yoleaux | Presentation: Extreme Privilege Escalation On Windows 8/UEFI Systems | The MITRE Corporation | 13:33 |
kanzure | oh. link was okay. | 13:33 |
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 13:34 | |
kanzure | http://counterparty.io/news/counterparty-community-update-oct-2014/ | 13:37 |
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Ping timeout: 260 seconds] | 13:39 | |
nmz787_i1 | "SCs normally reverse power flow by reversing a line’s current, whereas classic HVDC converters must flip the line’s voltage polarity." | 13:41 |
nmz787_i1 | I don't understand that... doesn't changing the polarity mean a change in current direction? | 13:42 |
nmz787_i1 | 'Sending current through seawater can corrode subsea infrastructure such as natural gas pipelines, but here, the dose makes the poison. “They can accept even 2,000 amps for up to 2 hours. So for a short pulse, it’s no problem,”' | 13:43 |
* nmz787_i1 waits for fish to die | 13:43 | |
fenn | free fish for everyone | 13:43 |
nmz787_i1 | I guess fish would be more resistive than sea water | 13:44 |
fenn | they'd still die | 13:44 |
kanzure | hm this presentation is much more elaborate than i expected | 13:49 |
nmz787_i1 | some guys from sec dept here did a talk on uefi at defcon | 13:50 |
nmz787_i1 | err | 13:51 |
nmz787_i1 | that is from defcon | 13:51 |
fenn | mitre presente at defcon? | 13:51 |
kanzure | (public) bitlicense responses https://www.coinprices.io/articles/industry-response-to-bitlicense-guidelines | 13:51 |
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap | 13:51 | |
nmz787_i1 | oh, i guess theirs was on secureboot from windows 8 | 13:52 |
nmz787_i1 | https://www.defcon.org/html/defcon-22/dc-22-speakers.html#Bulygin | 13:52 |
nmz787_i1 | https://media.blackhat.com/us-13/us-13-Bulygin-A-Tale-of-One-Software-Bypass-of-Windows-8-Secure-Boot-Slides.pdf | 13:53 |
fenn | secure boot is fine as long as there's a guaranteed way around it | 13:54 |
fenn | until one day they remove the off switch | 13:55 |
fenn | .g tivoisation | 13:55 |
yoleaux | http://en.wikipedia.org/wiki/Tivoization | 13:55 |
kanzure | fenn: someone should scheme up a joint eff/nra thing for homebrew cmos fabrication because <insert nsa paranoia stuff here> and <insert ic fab paranoia here> | 13:56 |
fenn | but guns dont have cmos backdoors (yet | 13:57 |
kanzure | on the bright side, they already might | 13:57 |
kanzure | .g drm gun backdoor nra | 13:57 |
yoleaux | http://www.factcheck.org/2013/12/no-back-door-gun-control/ | 13:57 |
kanzure | .g "drm" gun backdoor nra | 13:57 |
yoleaux | http://whyevolutionistrue.wordpress.com/2014/05/09/u-s-gun-lobby-opposes-safety-equipped-guns/ | 13:57 |
nmz787_i1 | fenn: i'm pretty sure an admin can turn off secureboot completely at will, the point is for non-admin to not be able to do that | 13:58 |
nmz787_i1 | i was thinking recently that most likely a bullet could kill a bacteria | 13:58 |
nmz787_i1 | maybe a few at a time | 13:58 |
kanzure | "But not to the right-wing gun lobby. No, they fear that this is the first step on the road to the government controlling all the guns, for, supposedly, smart-gun technology could enable the U.S. Gubbmint to stop ALL the guns from firing, eliminating our God-given right to have a lethal weapon. And so we have the weird situation in which the gun lobby not only opposes new kinds of guns, but also threatens those gun dealers who sell them." | 13:58 |
kanzure | perfect | 13:58 |
jrayhawk_ | Yeah, being able to disable it is part of the requirements, though the certification process doesn't test for it AFAIK | 13:58 |
fenn | sort of a continuation of the missile defence/feminist superweapon thought from yesterday | 13:59 |
nmz787_i1 | i see folks running ubuntu when I walk around here... so hopefully that means everything is OK as far as free-computing | 13:59 |
fenn | a hardware switch to disable secureboot would work for me | 14:00 |
fenn | but that would add $0.05 to the cost | 14:00 |
nmz787_i1 | probably more realistically, since they'd need to add a new pinout from the chip package too or something | 14:01 |
fenn | and that would add $0.05 to the cost | 14:02 |
nmz787_i1 | hah | 14:02 |
nmz787_i1 | that'd be like weeks of work probably | 14:03 |
nmz787_i1 | and reviews and more reviews of the process | 14:03 |
fenn | not my fault they didnt think of it first | 14:03 |
archels | there are plenty of bull-goose gun loonies, but you have to admit, placing a fingerprint scanner on a gun or coupling it with a smartwatch is the dumbest thing ever | 14:03 |
nmz787_i1 | probably some security concern that keeps it out | 14:03 |
nmz787_i1 | (unless we're both wrong and that already exists) | 14:04 |
nmz787_i1 | seems like a wrist strap like treadmills have might work OK | 14:04 |
fenn | 'might work OK' is not what you want to hear about a piece of survival equipment | 14:04 |
nmz787_i1 | someone steals the gun from your hand, the pin attached to the string attached to your wrist pulls out, gun won't fire | 14:05 |
nmz787_i1 | well, 'OK' in that it's annoying to need a wrist strap | 14:05 |
nmz787_i1 | doesn't work for concealed carry | 14:05 |
nmz787_i1 | well at least | 14:05 |
fenn | the impetus for this is to prevent people from stealing the gun and using it for crimes or accidents | 14:05 |
nmz787_i1 | i guess if it was a gold chain it could be jewelry | 14:05 |
fenn | presumably the rightful owner is smart enough to not shoot himself in the face on accident | 14:06 |
archels | it also creates a false sense of security | 14:06 |
fenn | this is what i consider a "smart gun" http://tracking-point.com/ | 14:06 |
fenn | fire control computer system based on actual input data about what it's pointing at | 14:07 |
nmz787_i1 | some sensor tech seems like it could work | 14:07 |
archels | neat | 14:09 |
nmz787_i1 | fenn: that is very neat | 14:09 |
nmz787_i1 | hah | 14:10 |
nmz787_i1 | wow, like 8X the price of a 'dumb' AR | 14:10 |
fenn | huh apparently there's an interferometer measuring the barrel flex | 14:10 |
nmz787_i1 | it has WiFi | 14:11 |
nmz787_i1 | wtf is that for? | 14:11 |
nmz787_i1 | oh, must be the glasses | 14:11 |
fenn | saving recorded video and link to the HUD | 14:11 |
nmz787_i1 | 'You can shoot from completely protected positions behind trees or around corners when it’s necessary to remain unexposed to the target.' | 14:12 |
nmz787_i1 | 'Collaborative shooting and mentoring' | 14:12 |
nmz787_i1 | someday it might only be fair for opposing fighters to share each others video links | 14:13 |
nmz787_i1 | link in the old split-screen game days | 14:13 |
fenn | green on blue | 14:14 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 14:30 | |
-!- _0bitcount [~big-byte@81.61.34.185.dyn.user.ono.com] has quit [Quit: Leaving] | 14:45 | |
drethelin | nmz that's what the honor system is for | 14:47 |
drethelin | if your friend is screen watching you have to yell at them | 14:47 |
-!- HEx2 [~HEx@hexwab.plus.com] has joined ##hplusroadmap | 14:48 | |
-!- HEx2 is now known as HEx1 | 14:49 | |
superkuh | paperbot: http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=1903723 | 14:49 |
paperbot | http://libgen.org/scimag/get.php?doi=10.1117%2F12.2061547 | 14:49 |
justanotheruser | paperbot http://hbr.org/2009/12/the-innovators-dna/ar/1 | 14:54 |
justanotheruser | paperbot: http://hbr.org/2009/12/the-innovators-dna/ar/1 | 14:54 |
justanotheruser | pls | 14:54 |
paperbot | http://diyhpl.us/~bryan/papers2/paperbot/7d9ae6739de3f843d9cf0bbcc3e636f4.txt | 14:54 |
paperbot | http://diyhpl.us/~bryan/papers2/paperbot/c04a23a5121e337dac1160e17906c005.txt | 14:55 |
fenn | justanotheruser: DNA is not a metaphor, it actually exists. | 14:59 |
nmz787_i1 | paperbot: http://hbr.org/product/the-innovator-s-dna/an/R0912E-PDF-ENG | 14:59 |
paperbot | http://diyhpl.us/~bryan/papers2/paperbot/a86e5f75f3951086e624028f1614ab3e.txt | 15:00 |
fenn | i doubt paperbot has access to "harvard business review" anyway | 15:00 |
kanzure | in fact, paperbot is an avid reader of harvard business review and other fine periodicals | 15:00 |
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap | 15:01 | |
nmz787_i1 | superkuh: does the libgen link work? | 15:01 |
-!- |b| [~|d|@ip68-107-37-158.sd.sd.cox.net] has quit [Read error: Connection reset by peer] | 15:03 | |
fenn | justanotheruser: that paper would have been much more interesting if they did statistical analysis on the genomes of the 25 innovative entrepreneurs instead | 15:03 |
-!- HEx1 [~HEx@hexwab.plus.com] has quit [Ping timeout: 255 seconds] | 15:03 | |
fenn | tho 25 or even 3000 is probably not a big enough sample to draw meaningful conclusions | 15:04 |
-!- |c| [~|d|@ip68-107-37-158.sd.sd.cox.net] has joined ##hplusroadmap | 15:04 | |
nmz787_i1 | that paper mentions http://www.mcnpvised.com/visedtraining/penelope/penelope0.pdf https://www.oecd-nea.org/science/pubs/2009/nea6416-penelope.pdf (which seems to have become https://www.oecd-nea.org/science/docs/2011/nsc-doc2011-5.pdf) and this one looks fun too http://www.slac.stanford.edu/grp/arb/tn/arbvol5/AARD459.pdf | 15:04 |
sheena | nmz787: ever heard of using OCR on a video? | 15:06 |
yoleaux | 05:15Z <kanzure> sheena: http://imgur.com/a/8DfqE | 15:06 |
yoleaux | 05:33Z <nmz787> sheena: why not just use a small plate? | 15:07 |
kanzure | yes you can ocr each frame | 15:07 |
kanzure | there's a way to dump frames from a video | 15:07 |
kanzure | each frame would be an image | 15:07 |
sheena | does software already exist for this? | 15:07 |
superkuh | nmz787, no. Nor entering the DOI manually. | 15:08 |
sheena | i imagine you would be best served by finding 4 identical frames in a row (or 40, or whatever threshold), and picking one of them to OCR, then repeating | 15:08 |
sheena | imagining a video of a book, with pages turning | 15:08 |
fenn | it's hard to get a good image from a moving page | 15:08 |
kanzure | ffmpeg -i video.mp4 -r 1 -f image2 image-%3d.jpg | 15:08 |
kanzure | this will generate image-00x.png files | 15:08 |
fenn | unless you just mean someone turning pages by hand with a delay between page turns | 15:09 |
nmz787_i1 | sheena: I was doing all of that via python recenty (though not combining the two ideas, video and ocr) | 15:09 |
sheena | nmz787_i1: do you have codes and things you would share? | 15:10 |
sheena | fenn: by hand with a delay, yes | 15:10 |
nmz787_i1 | sheena: I think I was using pytesseract for OCR (more details in the logs of this channel) | 15:10 |
sheena | i turn on camera and set up tripod, and just sit and flip the book slowly | 15:10 |
nmz787_i1 | sheena: yeah not right now though, when I get home | 15:10 |
sheena | but faster than scanning by a long shot | 15:10 |
fenn | that's how they do it at google | 15:11 |
sheena | ive looked at tesseract in general before | 15:11 |
sheena | fenn: video you mean? or tesseract? | 15:11 |
nmz787_i1 | i remember one of the image formats (TIFF vs PNG I think) worked 100% better than the other | 15:11 |
fenn | there's a special spring loaded book holder to keep the pages flat and perpendicular to the two overhead cameras | 15:11 |
nmz787_i1 | which was unexpected | 15:11 |
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 260 seconds] | 15:11 | |
fenn | i dont know if there's a button or what | 15:12 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 260 seconds] | 15:12 | |
fenn | occasionally you'll see fingers in google books | 15:12 |
nmz787_i1 | sheena: here's some code for getting the video camera frames (and then displaying to screen, but you'd need to save as [TIFF I think] then push to tesseract) http://stackoverflow.com/a/26457671/253127 | 15:12 |
sheena | fenn: cooll | 15:13 |
nmz787_i1 | sheena: what OS are you on? | 15:13 |
sheena | lubuntu 14.04 | 15:13 |
nmz787_i1 | cool, should be easy to install opencv (libopencv I think) and wxPython if you wanted a GUI | 15:14 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 15:14 | |
sheena | i wonder if i need OCR | 15:14 |
fenn | opencv is probably overkill for this task | 15:14 |
nmz787_i1 | if you want the text to be searchable | 15:14 |
kanzure | opencv does ocr? | 15:14 |
nmz787_i1 | no | 15:14 |
nmz787_i1 | but it gets camera frames quite easily | 15:14 |
sheena | if i can just pull the page image frames as png/tiff, and pdf into an ebook.... but no search and no text-to-audio option | 15:14 |
nmz787_i1 | and can de-skew if needed also quite easily | 15:14 |
fenn | sheena you may want to look into djvu | 15:15 |
kanzure | yeah what's the level of ocr needed here, sheena? | 15:15 |
kanzure | like 1 million pages of text vs 10 ? | 15:15 |
sheena | books of 100 pages | 15:15 |
fenn | also are there pictures? | 15:16 |
sheena | some ya | 15:16 |
sheena | so just exporting video to image/pdf might be better, then could ocr from there if needed in some of the cases? | 15:16 |
fenn | i have no idea | 15:16 |
fenn | i could never get open source OCR software to work right | 15:16 |
sheena | im (of course for personal use only) lookig to make ebook copies of some books that aren't available in ebook versions.. | 15:17 |
sheena | some of the books are self published or done by small publishers.. | 15:17 |
sheena | and.. i want to be able to read them digitally, for my own personal entertainment | 15:17 |
sheena | heh | 15:17 |
kanzure | fenn, nmz787_i1 claims that pytesseract is much better than it was | 15:17 |
nmz787_i1 | fenn: I was able to get tesseract to work nicely | 15:17 |
nmz787_i1 | fenn: the key was changing the input file format (which seemed quite strange to me) | 15:18 |
nmz787_i1 | sheena: opencv can save uncompressed TIFF files | 15:18 |
nmz787_i1 | so dumping video to tiff would be super easy | 15:18 |
nmz787_i1 | prob like 10 lines of pythoin | 15:18 |
nmz787_i1 | figuring out when to take a pic would be more work (i.e. only when the frame is not moving (the page isn't turning)) | 15:19 |
sheena | how to do the "dont save every frame, just the different/good ones" part? | 15:19 |
sheena | yeah. that. | 15:19 |
nmz787_i1 | should be a matter of diff-ing subsequent frames till there is no/minimal diff | 15:19 |
sheena | diffing? | 15:20 |
nmz787_i1 | in new python opencv bindings images are just numpy arrays | 15:20 |
fenn | subtracting the pixel values | 15:20 |
kanzure | well text in a video usually lasts for multiple frames | 15:20 |
nmz787_i1 | so you can do array subtraction (diff-ing) | 15:20 |
kanzure | so if you ocr one frame and then another, you usually get different results if the frames were different | 15:20 |
kanzure | and then you hvae to diff text n' stuff i think | 15:20 |
sheena | what about the stuff that detects motion? | 15:21 |
sheena | like, lots of cameras for wildlife only record/display the parts where there is motion | 15:21 |
nmz787_i1 | sheena: that's what subtracting images would do | 15:21 |
fenn | kanzure: actually that's a good way to automatically detect OCR errors | 15:21 |
fenn | since you know the underlying text hasn't changed | 15:21 |
nmz787_i1 | sheena: I say get opencv installed and fire up python and start playing | 15:21 |
sheena | so i'm doing; if frame 2 - frame 1 == 0, frames 1 and 2 are identical? | 15:22 |
nmz787_i1 | sheena: lots of sample code in the opencv repo/source https://github.com/Itseez/opencv/tree/master/samples/python2 | 15:22 |
sheena | (concept, not specific example) | 15:22 |
nmz787_i1 | sheena: exactly | 15:22 |
sheena | ok | 15:22 |
nmz787_i1 | and with the images being numpy arrays, that operation is pretty much just that easy | 15:22 |
fenn | there's a package in ubuntu called 'motion' which is designed for detecting movement in webcam security cameras | 15:22 |
nmz787_i1 | you might need to check how many non-zero values there are relative to total pixel count | 15:23 |
nmz787_i1 | so you can get % difference | 15:23 |
nmz787_i1 | sheena: motion is also pretty good and easy to get setup | 15:23 |
sheena | https://www.youtube.com/watch?v=lzRGp3nc8Tw relevant?\ | 15:23 |
kanzure | .title | 15:23 |
yoleaux | Motion detection in OpenCV - YouTube | 15:23 |
nmz787_i1 | idk if there are python bindings to play/interact with it though | 15:23 |
fenn | Motion is a program that monitors the video signal from one or more cameras and is able to detect if a significant part of the picture has changed. | 15:23 |
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 15:24 | |
sheena | sudo apt-get motion ? | 15:24 |
nmz787_i1 | sheena: googling motion detection example opencv provides lots of decent looking examples | 15:25 |
nmz787_i1 | sheena: something like that | 15:25 |
nmz787_i1 | (re apt-get) | 15:25 |
fenn | sudo apt-get install motion | 15:26 |
sheena | oh jeesus | 15:26 |
sheena | its saving jpgs of me!! lol | 15:26 |
fenn | i've never actually used it either | 15:26 |
nmz787_i1 | i've used it for security webcams from a server years ago, and also from a rasp-pi recently | 15:27 |
sheena | this is awesome but im not sure its best for this application lol | 15:27 |
sheena | cause i need it to be looking at a recorded video, not a camera.. off to read some man files | 15:27 |
nmz787_i1 | sheena: you can probably bug jrayhawk_ about how to setup a loopback fom a video to a v4l device | 15:27 |
sheena | might have a function already for that | 15:28 |
sheena | but it may also be backwards to what i want..... | 15:28 |
fenn | is there actually an option to only record images with no motion? (i haven't found it yet if there is) | 15:28 |
nmz787_i1 | fenn: as opposed to what? | 15:29 |
nmz787_i1 | oh, 'no motion' | 15:29 |
nmz787_i1 | hmm, yeah I guess that's the inverse! | 15:29 |
sheena | motion_video_pipe string | 15:29 |
sheena | Values: Max 4095 characters / Default: Not defined | 15:29 |
sheena | The video4linux video loopback input device for motion images. | 15:29 |
sheena | If a particular pipe is to be used then use the device filename | 15:29 |
sheena | of this pipe, if a dash '-' is given motion will use | 15:29 |
sheena | /proc/video/vloopback/vloopbacks to locate a free pipe. Default: | 15:29 |
sheena | not set | 15:29 |
sheena | oops :( | 15:29 |
sheena | sorry | 15:29 |
sheena | but that looks like the "use afile instead of my webcam" thing | 15:30 |
nmz787_i1 | yeah, you said you wanted to use a video file | 15:30 |
sheena | on_event_end string Values: Max 4095 characters / Default: Not defined Command to be executed when an event ends after a period of no motion. The period of no motion is defined by option gap. You can use Conversion Specifiers and spaces as part of the command. | 15:31 |
-!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has quit [Read error: Connection reset by peer] | 15:31 | |
sheena | motion video pipe looks like it will let me put a file as input? | 15:31 |
-!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has joined ##hplusroadmap | 15:32 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 250 seconds] | 15:32 | |
fenn | okay that's not what i would have called it | 15:32 |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 15:34 | |
nmz787_i1 | sheena: this is basic code that would just save images http://paste.pound-python.org/show/fRtaQrlb2rhuiibc2cAQ/ | 15:34 |
nmz787_i1 | 8 lines | 15:34 |
nmz787_i1 | you can replace the VideoCapture with a call to open a video file instead | 15:34 |
fenn | don't save as jpg | 15:34 |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 240 seconds] | 15:34 | |
nmz787_i1 | yeah and that | 15:35 |
nmz787_i1 | but you should be able to just change the file extension and it will use the correct imge encoder | 15:35 |
sheena | i need to pip instlal cv2? | 15:35 |
nmz787_i1 | probably use apt-get | 15:35 |
nmz787_i1 | sudo apt-cache search opencv | 15:35 |
nmz787_i1 | there should be one main package that installs the rest | 15:36 |
sheena | python-opencv? | 15:36 |
nmz787_i1 | my guess is something like libopencv | 15:38 |
nmz787_i1 | or libcv | 15:38 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 15:39 | |
kanzure | it's possible that python-opencv will install opencv things in addition to the python bindings or python wrapper | 15:39 |
fenn | python-opencv is what you want | 15:39 |
sheena | call to open a video.. how do i find the 'man' for cv2 inside of python? | 15:40 |
nmz787_i1 | sheena: http://docs.opencv.org/modules/highgui/doc/reading_and_writing_images_and_video.html#videocapture-videocapture | 15:41 |
nmz787_i1 | sheena: idk how to use python help, sorry | 15:41 |
sheena | modules. thanks | 15:41 |
kanzure | help() | 15:41 |
kanzure | is how you use help in python | 15:41 |
nmz787_i1 | kanzure: so like help(cv2) ? | 15:41 |
kanzure | yes | 15:41 |
nmz787_i1 | huh | 15:41 |
kanzure | it just reads cv2.__doc__ for you | 15:41 |
fenn | it's not really a manual, just a list of what's in it (unless someone has specifically provided documentation that way) | 15:42 |
sheena | it does not like the syntax of that paste nmz787_i1 sent | 15:43 |
fenn | the if block shouldn't have that many levels of indentation | 15:46 |
fenn | how do i get my cursor back on screen #6 (i lost it) | 15:48 |
fenn | like there should be a white box where i'm entering text, but there is none | 15:49 |
fenn | i must have entered some terminal escape sequence on accident | 15:50 |
kanzure | you might have typed ctrl-s? | 15:50 |
fenn | gah now i have a split screen | 15:51 |
nmz787_i1 | sheena: must be something with the copy-paste, spaces vs tabs or something like that | 15:51 |
sheena | okie | 15:52 |
nmz787_i1 | sheena: ahh, fenn is right, everything after the ret, new_frame is indented one level too much | 15:52 |
nmz787_i1 | whoops! | 15:52 |
fenn | ESC Pn p Cursor Visibility (97801) | 15:53 |
fenn | i have no idea what this means | 15:54 |
dingo | hehe you got questions on terminal control sequences, just let me know | 15:55 |
sheena | the code doesnt end. should it? | 15:55 |
sheena | like | 15:55 |
dingo | its \x1b[25l | 15:55 |
sheena | it does the jpg output stuff, butthen just... hangs? | 15:55 |
dingo | i know that off the top of my head | 15:55 |
dingo | give me something easier | 15:55 |
fenn | too late, i killed the window | 15:55 |
fenn | sheena: it was written as an infinite loop, that's all it does | 15:55 |
fenn | while True: | 15:55 |
sheena | ok | 15:56 |
sheena | so i can just kill it when its done? | 15:56 |
sheena | thats how it should work? | 15:56 |
fenn | yes | 15:56 |
nmz787_i1 | ctrl-d or ctrl-c should end it | 15:56 |
sheena | i assume ther is no easy way to have it tell me when its done? can i just add break to the last line? | 15:56 |
dingo | even if you don't use python, you can use blessed to find the raw sequences for your TERM: http://github.com/jquast/blessed | 15:56 |
nmz787_i1 | sheena: there is a way to tell if the file ended | 15:57 |
dingo | python -c 'import blessed; print repr(blessed.Terminal().cursor_hidden)' or something like that | 15:57 |
kanzure | if you know when it's done then you don't need to use "while True". | 15:57 |
kanzure | alternatively, you can break inside of the while loop when you know it's done (some api call, i assume) | 15:57 |
sheena | ok no worries | 15:58 |
nmz787_i1 | sheena: the ret value will tell you if it's done | 15:58 |
kanzure | ah | 15:58 |
nmz787_i1 | 'If no frames has been grabbed (camera has been disconnected, or there are no more frames in video file), the methods return false and the functions return NULL pointer.' | 15:58 |
sheena | so now i jut need to make a video of pages turning to test this properly | 15:58 |
sheena | iv'e tested it with a random AVI and it seems to work | 15:58 |
nmz787_i1 | sheena: you'll need to do the subtraction too | 15:59 |
sheena | nmz787_i1: what is the ret value? (in simple explanation to satisfy my curiosity) | 15:59 |
sheena | oh, it's just getting each frame? | 15:59 |
sheena | this video has only 111 frames? that seems unlikely.. | 15:59 |
nmz787_i1 | sheena: ret should just be True or False, or True or None... something that indicates success of the read() | 16:00 |
nmz787_i1 | sheena: looking at this now http://wiki.scipy.org/Tentative_NumPy_Tutorial#head-c5f4ceae0ab4b1313de41aba9104d0d7648e35cc | 16:00 |
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Ping timeout: 260 seconds] | 16:00 | |
fenn | i'm assuming for each frame it's reinitializing the camera, and that takes time | 16:01 |
fenn | dingo: how do i type in a terminal control sequence at all? escape key doesn't seem to send an escape sequence | 16:02 |
fenn | this is what too many layers of screen and virtual terminals does to you http://fennetic.net/irc/input_out.mp3 | 16:08 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Quit: Reconnecting] | 16:08 | |
kanzure | makes sense to me | 16:09 |
nmz787_i1 | sheena: this may work to give you a percent motion estimate http://paste.pound-python.org/show/mv8YD0DDgrrvrHp5sq7v/ | 16:10 |
nmz787_i1 | sheena: oops, I meant to put a break after the last print | 16:10 |
nmz787_i1 | sheena: also that last print statement has a tab for indent which will break things, you need to change it to 4 spaces | 16:11 |
nmz787_i1 | sheena: fixed http://paste.pound-python.org/show/s6TWwHGfR3V0WXTaUpNm/ | 16:11 |
nmz787_i1 | sheena: err, whoops, I didn't save the last_frame ! | 16:11 |
sheena | do i have numpy? | 16:11 |
nmz787_i1 | if you have cv2 then yes | 16:12 |
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap | 16:12 | |
fenn | how are tabs still a thing | 16:13 |
kanzure | they are input rotation | 16:13 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 16:13 | |
nmz787_i1 | sheena: try this http://paste.pound-python.org/show/7fLTADIWJIxgHoUJCIdV/ | 16:13 |
sheena | and does it not need frame and last_frame defined? | 16:13 |
nmz787_i1 | sheena: sorry for overloading with wrong code | 16:13 |
nmz787_i1 | sheena: ugh, once again there was a bug... the not should not be there | 16:14 |
nmz787_i1 | http://paste.pound-python.org/show/qmXmTX7ntWsrMPiyzOUh/ | 16:14 |
nmz787_i1 | oh god | 16:14 |
nmz787_i1 | i think i need to stop! | 16:14 |
fenn | what's last frame = | 16:14 |
nmz787_i1 | http://paste.pound-python.org/show/f7pRougJBEVBVwWG9rNx/ | 16:15 |
nmz787_i1 | there | 16:15 |
* nmz787_i1 back to work! | 16:15 | |
fenn | yes sir | 16:15 |
kanzure | .title https://news.ycombinator.com/item?id=8489954 | 16:16 |
yoleaux | Introducing Consul Template | Hacker News | 16:16 |
sheena | ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all() | 16:16 |
kanzure | confd recently implemented long-polling, which is way better than my implementation in pyconfd... but it's still only go templates. sooo. i might have to steal that feature and put it into pyconfd. | 16:17 |
kanzure | sheena: it's right, you know. basically use "if all(whatever):" instead of "if watever:". or you can use any() if you mean any() instead of all(). | 16:17 |
nmz787_i1 | sheena: what line? | 16:17 |
nmz787_i1 | is it the if statement? | 16:18 |
sheena | yeah | 16:18 |
fenn | if last_frame is not None: maybe | 16:18 |
nmz787_i1 | change to if last_frame is not None | 16:18 |
sheena | nod | 16:18 |
sheena | working | 16:18 |
sheena | i think | 16:18 |
sheena | doing things, anyway | 16:18 |
sheena | so now i need a test video, right? one that doesnt have constant motion | 16:19 |
nmz787_i1 | sheena: you can also add a live video display easily http://docs.opencv.org/trunk/doc/py_tutorials/py_gui/py_image_display/py_image_display.html#display-an-image | 16:19 |
nmz787_i1 | sheena: webcam should work with you just moving your head in front of it | 16:20 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 272 seconds] | 16:20 | |
fenn | my recommendation is to use as bright a light as possible, preferably a halogen or fluorescent | 16:20 |
nmz787_i1 | sheena: if you want to use that video display, change waitKey(0) to something like waitKey(30) | 16:20 |
nmz787_i1 | since 0 blocks until the user presses a key | 16:21 |
nmz787_i1 | while 30 would display the image for 30 milliseconds before continuing processing code | 16:21 |
fenn | nmz787_i1: how would you detect/quantify motion blur in an arbitrary image? | 16:24 |
nmz787_i1 | like only having a single image? | 16:24 |
sheena | thanks. i suddenly checked the time and have to go :( back later!! | 16:25 |
nmz787_i1 | probably do edge detection from several directions and compare the signal strength | 16:25 |
nmz787_i1 | sheena: hope you get something working! | 16:26 |
fenn | hm. "blur detection is actually a very active research field" | 16:26 |
fenn | paperbot: http://ieeexplore.ieee.org:80/xpl/articleDetails.jsp?reload=true&arnumber=1038902 | 16:27 |
fenn | it's thinking | 16:27 |
kanzure | deep thought | 16:27 |
fenn | paperbot: http://dx.doi.org/10.1109/ICIP.2002.1038902 | 16:27 |
kanzure | heh paperbot should rate papers based on pretentious vocabulary or something | 16:28 |
paperbot | http://diyhpl.us/~bryan/papers2/paperbot/2a79a7e5e63276fd7ce0c34dc9708cca.txt | 16:28 |
fenn | unfortunately 99% of them would have losing scores | 16:28 |
nmz787_i1 | fenn: pg 5 shows the edge and their derivatives http://www.cse.unr.edu/~bebis/CS791E/Notes/EdgeDetection.pdf | 16:28 |
nmz787_i1 | this also has some stuff that might be interesting http://bit.kuas.edu.tw/~jihmsp/2010/vol1/JIH-MSP-2010-01-003.pdf | 16:29 |
nmz787_i1 | they only have one graph that I like (the first) | 16:29 |
paperbot | http://libgen.org/scimag/get.php?doi=10.1109%2FICIP.2002.1038902 | 16:31 |
fenn | maybe i shouldn't be using as my test case "videos of sand dunes filmed in dark conditions" | 16:31 |
fenn | though it is a valid use case | 16:33 |
fenn | i want to make 3d environment maps from waving a cellphone around, but it's hard to figure out which images are usable | 16:34 |
fenn | also high resolution "scanning" of architecture and the like | 16:35 |
nmz787_i1 | fenn: that's been a project i've thought of a lot | 16:35 |
nmz787_i1 | especially back when I was commanded to help build my campus in google sketchup | 16:35 |
fenn | if you just pan around at constant velocity all your images are equally blurred | 16:35 |
nmz787_i1 | my idea was using people who biked around campus to collect data | 16:36 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 16:36 | |
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 245 seconds] | 16:38 | |
fenn | yeah this has actually been done, see i think http://www.youtube.com/watch?v=0Ef-ofGocK4&list=PL39C03377F937DCCC&index=7 | 16:38 |
fenn | gah stupid playlists | 16:39 |
fenn | should be Museum 01 Entrance Map Creation | 16:39 |
fenn | .title | 16:40 |
yoleaux | PTAMM: Museum 01 Entrance Map Creation - YouTube | 16:40 |
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 260 seconds] | 16:40 | |
fenn | wouldn't take long to do a low resolution map | 16:40 |
nmz787_i1 | hmm | 16:41 |
kanzure | somewhat-elaborate terraform example https://github.com/18F/fec-infrastructure/blob/master/bootstrap/bootstrap.tf | 16:41 |
nmz787_i1 | that doesn't seem to be in 3D space | 16:41 |
fenn | the building facade is mostly 2d (2.5d) | 16:42 |
-!- CharlieNobody [~CharlieNo@97-85-246-37.static.stls.mo.charter.com] has joined ##hplusroadmap | 16:43 | |
fenn | welcome, welcome, welcome | 16:44 |
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 16:44 | |
kanzure | man i'm glad i forgot about greeter bots | 16:45 |
fenn | the worst are the ones that spew a bunch of rules at you the instant you say something | 16:45 |
kanzure | hmm http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using-network-security.html | 16:47 |
fenn | why am i reading a paper about edge detection | 16:48 |
kanzure | fenn: http://ourincrediblejourney.tumblr.com/ | 16:49 |
fenn | kanzure: i don't care about that either? | 16:50 |
kanzure | it wasn't meant as news | 16:51 |
fenn | is it some kind of demotivational propaganda? | 16:51 |
kanzure | yes.. sort of. | 16:52 |
kanzure | mockery | 16:52 |
fenn | so the lesson here is ... don't use online startup services for anything important | 16:53 |
kanzure | basically. | 16:53 |
kanzure | and "throw users under a bus whenever you please" | 16:53 |
fenn | well i already learned that when microsoft bought out hotmail | 16:53 |
kanzure | how original | 16:53 |
fenn | it was actually original at the time | 16:53 |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has joined ##hplusroadmap | 16:55 | |
fenn | It also is about to acquire CompuServe, which earlier this week launched a Web-only service, dubbed "C." | 16:55 |
fenn | "The Microsoft acquisition of Hotmail is another great validation that Web-based email is here to stay," said Scott Chasin, USA.Net's chief technology officer. "Things will get very exciting in the coming year, as everyone is looking for a dance partner." | 16:56 |
fenn | shortly thereafter they deleted any accounts if you hadn't logged in for more than 1 month (!) | 16:56 |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 244 seconds] | 16:59 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 244 seconds] | 16:59 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 17:02 | |
fenn | A company that can afford to pay millions for some new staff but not for what those staff built. This repeated pattern only encourages more people to create flashy services that have no hope of being sustainable businesses in their own right, but may survive long enough, with VC funding, to attract the attention of a large company eager for new ideas and staff. | 17:05 |
kanzure | "do whatever it takes to grow 5% every week" | 17:05 |
fenn | is that really a thing? | 17:08 |
-!- maaku [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has joined ##hplusroadmap | 17:08 | |
kanzure | http://www.paulgraham.com/growth.html | 17:08 |
kanzure | yes it is a thing | 17:08 |
kanzure | hello maaku | 17:08 |
fenn | does it exist in reality? | 17:08 |
kanzure | they certainly claim it does | 17:08 |
-!- maaku is now known as Guest46292 | 17:08 | |
kanzure | maaku: i have stolen your words and claimed them as my own, be amused https://github.com/kanzure/bitcoin-incentives/blob/master/bitcoin-incentives.tex | 17:08 |
kanzure | fenn: companies like airbnb had to have had some non-zero growth rate to go from zero to $5B/year | 17:10 |
-!- Guest46292 is now known as maaku | 17:11 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 17:13 | |
-!- nmz787_i1 [~nmccorkx@134.134.137.73] has quit [Ping timeout: 240 seconds] | 17:14 | |
fenn | i wonder what percentage of acqui-hires actually stick around for more than a year | 17:15 |
kanzure | depends on the contract | 17:16 |
fenn | well it seems like you usually get stock options after a year | 17:17 |
fenn | but then there's no incentive to stay | 17:17 |
-!- Netsplit *.net <-> *.split quits: bbrittain, rak[1], Vutral, chris_99, helleshin, Burn_, drewbot, DonnchaC_ | 17:20 | |
-!- yashgaroth [~ffffff@2606:6000:cb85:6a00:106:6dc5:c4e3:6f74] has joined ##hplusroadmap | 17:20 | |
kanzure | 17:14 <@gwern> in any event, I'm not a chemist but more of a statistician/data-scientist, so I'm not sure what I would do with a spectrometer | 17:20 |
fenn | make sure your nootropics are what you think they are | 17:21 |
kanzure | nah let's wing it | 17:24 |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has joined ##hplusroadmap | 17:25 | |
-!- Netsplit over, joins: Vutral, helleshin, DonnchaC_ | 17:25 | |
fenn | ugh reading about "acqui-hire strategy" makes me sad | 17:26 |
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Max SendQ exceeded] | 17:26 | |
-!- chris_99 [~chris_99@static.35.151.76.144.clients.your-server.de] has joined ##hplusroadmap | 17:26 | |
-!- Netsplit over, joins: drewbot, Burn_, bbrittain, rak[1] | 17:26 | |
kanzure | hah | 17:26 |
kanzure | from their perspective though, acquisitions help you not die | 17:26 |
fenn | individuals who would choose a startup will likely never even consider applying at a large corporation. And even if they do apply, up to 90% will be rejected because they are found not to be a “corporate fit” somewhere during the recruiting process. | 17:27 |
fenn | ^quote | 17:27 |
fenn | "Eventually many will acclimate – just like anyone who moves into a new environment, over time most will eventually acclimate and even lose their dislike for the corporate world." | 17:28 |
kanzure | have you never read pg | 17:28 |
kanzure | how can this be | 17:28 |
fenn | of course i have, i read all of his essays once | 17:28 |
kanzure | another lifetime ago perhaps | 17:28 |
kanzure | which iteration is this cycle? | 17:28 |
fenn | i just think it's incredible that the corporate HR/hiring people acknowledge that their process is so broken it weeds out 90% of the people they are trying to recruit in the first place | 17:29 |
kanzure | oh, most of the people in the hr department don't really acknowledge that | 17:29 |
-!- chris_99 [~chris_99@static.35.151.76.144.clients.your-server.de] has quit [Remote host closed the connection] | 17:29 | |
kanzure | but the people who use the phrase "strategic partnerships" think about that, sure | 17:29 |
fenn | "strategic Talent Management solutions" | 17:30 |
kanzure | github is like the only startup doing fully remote | 17:30 |
-!- justanot1eruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 17:30 | |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Read error: Connection reset by peer] | 17:30 | |
fenn | they have a batcave or whatever tho | 17:31 |
fenn | i was supposed to build some arcade consoles or something for them | 17:31 |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 17:31 | |
kanzure | oh that's unfortunate | 17:31 |
kanzure | it's pretty weird how nobody has been able to pull off fully remote | 17:32 |
fenn | it was an interesting idea but my contact flaked out on me | 17:32 |
kanzure | they all get pressured by vc to get an office or something | 17:32 |
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 17:32 | |
fenn | it's rare for people to know how to a) collaborate effectively online, and b) keep track of who's doing what (if anything) | 17:32 |
kanzure | if you're not within 30 miles you don't exist | 17:33 |
fenn | strange how it's more acceptable to telework from nearby | 17:33 |
fenn | as if the extra millisecond of latency matters | 17:33 |
fenn | it's nice to have an office available to go to if you need it (to get away from family or distraction) | 17:35 |
fenn | that's hardly worth building a custom capital-hemorrhaging extravaganza for tho | 17:36 |
* kanzure wonders why paperbot hasn't been into the office lately | 17:36 | |
fenn | him and gradstudentbot are having a three-way with gnusha | 17:36 |
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-kehaimpesyazzqza] has quit [Quit: Connection closed for inactivity] | 17:37 | |
fenn | .botsnack | 17:37 |
yoleaux | :D | 17:37 |
-!- Beatzebub [~beatzebub@S0106b81619e8ecee.gv.shawcable.net] has joined ##hplusroadmap | 17:38 | |
fenn | i read paul graham, then studied lisp and fought a lot with foreign function interfaces and SLIME and decided maybe there was a slight mismatch between paul graham and reality | 17:40 |
kanzure | when was this | 17:41 |
kanzure | not sure if growth.html was new to you or not | 17:41 |
kanzure | http://googlescholar.blogspot.com/2014/10/caselaw-is-set-free-what-next.html | 17:42 |
fenn | august 2006 apparently | 17:42 |
kanzure | he has written many things since then heh | 17:43 |
kanzure | nearly all of his experience with fast-growing startups was after that | 17:43 |
kanzure | viaweb counts a bit but then there's the other 500 companies he managed | 17:43 |
kanzure | https://www.youtube.com/watch?v=j2ijUU4CwzA | 17:47 |
kanzure | http://d2bx26666ysa1s.cloudfront.net/wp-content/uploads/2014/10/BitGo-BitLicense-Comment-2014-10-21.pdf | 17:47 |
fenn | i think he was just starting what would become YC and doing angel investing | 17:47 |
kanzure | right.. 2006 was even pre reddit? | 17:47 |
fenn | what's reddit? | 17:48 |
fenn | 2006 sounds like a long time ago but it was only 3 years before i moved to the bay area.. | 17:48 |
kanzure | in 2006 i was still stuck in high school | 17:49 |
fenn | somehow i managed to read all of his essays and think that "a startup" was just a small business in the software/technology space | 17:49 |
kanzure | i'm sure that his emphasis has become stronger in the past 8 years. | 17:50 |
fenn | i had no idea there was this huge game around it, attracting capital, "series A financing" etc | 17:50 |
kanzure | cheat book http://diyhpl.us/~bryan/venture-deals.pdf | 17:50 |
fenn | is this like "business school for dummies" | 17:51 |
kanzure | this is like "heuristics that you will eventually learn after doing everything wrong when fundraising" | 17:51 |
fenn | "only a small percentage of companies raise venture capital" | 17:54 |
-!- justanot1eruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 244 seconds] | 17:55 | |
kanzure | that's true, the vast majority of businesses do not raise venture capital | 17:55 |
drethelin | our company was started with a credit card, dad likes to say | 17:56 |
kanzure | that was possibly not the best choice | 17:56 |
kanzure | was the credit card in his name? | 17:56 |
fenn | is it that hard to get a small business loan? | 17:56 |
kanzure | uh, yes? | 17:56 |
kanzure | you have to put up your 4th home as collateral or something | 17:56 |
kanzure | and then the interest rate is absurd | 17:57 |
fenn | doesnt it depend on the amount? i mean most credit cards dont go over $15k | 17:57 |
kanzure | hahah | 17:57 |
fenn | hard to find a worse rate than credit card interest | 17:57 |
kanzure | $50k credit is the normal limit i see around these parts | 17:58 |
drethelin | I think it varies a lot | 17:58 |
drethelin | plus it depends on what you can ask for | 17:58 |
kanzure | basically 1/3rd total annual cash salary | 17:58 |
drethelin | my current credit limit is like 2500 for some reason | 17:59 |
drethelin | on the card I get from my back | 17:59 |
drethelin | bank | 17:59 |
fenn | i have no category to file this book under :( | 17:59 |
drethelin | according to my room-mate he just had to go in and ask to get it raised | 17:59 |
dingo | yeah you have to ask for an increase | 17:59 |
dingo | you'll get one | 17:59 |
dingo | tell them you're buying furniture or some shit | 17:59 |
kanzure | mom spent a few years unable to find a $100k loan for her business | 18:00 |
dingo | i decided to stop playing the fucking game a while ago, recently bought a truck, and even though i didn't get a loan they still have to run your credit report for some new terrorist-tracking reason, and i got my credit score, its actually really fucking good, wtf | 18:00 |
kanzure | because they all wanted evidence of her personal salary (and she doesn't pay herself anything) | 18:00 |
dingo | i carry no CC's for years now | 18:00 |
kanzure | electricity companies sometimes report your bill payments to credit agencies | 18:00 |
fenn | terrorists need trucks | 18:00 |
fenn | and credit | 18:01 |
fenn | those bombs don't pay for themselves, son! | 18:01 |
kanzure | first they come for the terrorists, and then nobody was left to complete this joke | 18:01 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 18:02 | |
kanzure | pedophiles are constitutionally protected | 18:02 |
drethelin | kanzure how does she get money out of the company? | 18:02 |
drethelin | if she doens't pay herself anything? | 18:02 |
dingo | louis ck has a dark joke about pedophiles | 18:02 |
kanzure | she doesn't. she's not the smartest bear. | 18:02 |
kanzure | "i'm growing my company! 0.3% per year is growth, right?" (i don't know what the actual number is.) | 18:03 |
kanzure | (but it sounds low to me) | 18:03 |
fenn | you're supposed to write yourself an IOU for when you do make money in the future, and then it's counted as income for that year on your taxes (instead of one big lump which presumably has a higher tax rate) | 18:03 |
kanzure | 0.3% is less than what you can get on the stock market or something, geeze | 18:04 |
kanzure | might as well not bother with the company at all | 18:04 |
kanzure | that's even less than inflation | 18:05 |
fenn | but furniture appreciates over time, right? | 18:05 |
kanzure | i wonder if she realizes how awful that would be | 18:05 |
kanzure | she switched to cabinets | 18:05 |
kanzure | and sex furniture | 18:05 |
kanzure | which i really hope doesn't appreciate over time | 18:05 |
fenn | depends on who owned it | 18:05 |
kanzure | hey good point | 18:06 |
kanzure | that's looking on the bright side | 18:06 |
fenn | tell her to switch business models to renting furniture out to rock stars and selling it as a collector's item | 18:06 |
kanzure | she's not so good at the switching part, but i'll mention it | 18:06 |
fenn | i wasn't serious | 18:06 |
fenn | how can you not make money on cabinets | 18:07 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 265 seconds] | 18:07 | |
fenn | all that home renovation crap is ridiculously overpriced | 18:07 |
fenn | a new kitchen costs more than a new car | 18:08 |
kanzure | well, there's a bunch of ways that it can go wrong | 18:09 |
kanzure | i agree that these customers have big budgets | 18:09 |
kanzure | but it also turns out they are nag machines and you have to sue them frequently | 18:09 |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Ex-Chat] | 18:09 | |
fenn | has she heard of the 80/20 rule | 18:09 |
kanzure | i'm sure. | 18:09 |
fenn | so dump 20% of the customers; 80% of the problems go away | 18:10 |
kanzure | right. | 18:11 |
fenn | "we're overwhelmed with projects right now, but so and so is available and does quality work blah blah blah" | 18:11 |
kanzure | i could rant for hours about things that i think are going wrong, but i don't really want to | 18:11 |
kanzure | it's not even an industry that i would want to be involved in | 18:11 |
fenn | it's not even an industry | 18:11 |
drethelin | why does capitalism do such a bad job | 18:11 |
kanzure | and there has been no indication that she takes any of my advice ever, so there's also that | 18:12 |
kanzure | drethelin: because saving is required for capitalism to work | 18:12 |
fenn | saving? | 18:12 |
kanzure | hey that sounds almost insightful, go me! | 18:12 |
kanzure | yeah. | 18:12 |
fenn | explain | 18:12 |
kanzure | hoarding money. | 18:13 |
kanzure | not losing it on the market in an attempt to flee inflation | 18:13 |
kanzure | money counts as a capital asset | 18:13 |
kanzure | and it's perhaps the most common | 18:13 |
fenn | so are you blaming it on monetary policy? | 18:13 |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 260 seconds] | 18:13 | |
kanzure | i'm suggesting that money would be a very useful capital asset to conduct capitalism with | 18:13 |
drethelin | kanzure how is that relevant to over-priced slow industries | 18:14 |
kanzure | if more money was saved overall, high-capital-cost industries would become cheaper because of deflation | 18:14 |
kanzure | but, i was just commenting on your capitalism line, not necessarily about this industry i guess | 18:15 |
fenn | but if everyone saves doesnt that reduce the growth rate and cause a recession death spiral | 18:15 |
drethelin | that's more the part I'm curious about | 18:15 |
drethelin | than capitalism in general | 18:15 |
drethelin | why are there so many fields which seem obviously broken | 18:15 |
fenn | (i think this is what's happening in japan) | 18:15 |
drethelin | is it all just like | 18:15 |
drethelin | regulatory capture | 18:15 |
kanzure | there's lots of regulatory capture, but then lots of bad businesses too in general | 18:15 |
drethelin | eg you can hire a random guy to remodel your house for way cheaper than actual contractors | 18:15 |
kanzure | fenn: specifically you're worried about "everyone saving" or "nobody spending" ? | 18:16 |
kanzure | drethelin: there's also some weird knowledge arbitration going on, especially in obscure biotech niche industries. | 18:17 |
kanzure | and then you have a limited number of customers anyway, with limited funds, so you don't have much room to grow anyway | 18:17 |
kanzure | bridgeport operates on $10M/year revenue, that's it | 18:17 |
kanzure | that's super super tiny for having a majority stake of the entire global cnc market | 18:18 |
fenn | bridgeport is not a large player | 18:18 |
kanzure | that might explain something | 18:18 |
fenn | they're mostly used by universities and vocational schools | 18:18 |
kanzure | right, hm | 18:19 |
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 260 seconds] | 18:20 | |
fenn | i guess it actually matters what people spend their money on | 18:21 |
fenn | if everyone spends all their money on ice cream and kitchen remodeling, that doesn't really make the economy go | 18:21 |
kanzure | oops, i should not have called that hoarding http://archive.mises.org/5763/saving-versus-hoarding/ | 18:21 |
fenn | but if they spend it on a new truck or a computer that helps them do other stuff.. | 18:22 |
kanzure | a kitchen remodeling can be a capital expenditure because commercial kitchen efficiency, i dunno | 18:23 |
fenn | i'm talking about HGTV "granite countertops and accent backsplash" stuff | 18:24 |
kanzure | right, she doesn't do commercial kitchens anyway | 18:24 |
fenn | which drives up local property values but doesn't really contribute anything to the efficiency of making stuff | 18:25 |
kanzure | anyway if inflation is happening then why would you want to work for worthless dollars? | 18:25 |
kanzure | weird how http://en.wikipedia.org/wiki/Capitalism only mentions saving twice, "Neoclassical economics explain capitalism as made up of individuals, enterprises, markets and government. According to their theories, individuals engage in a capitalist economy as consumers, laborers, and investors. As laborers, individuals may decide which jobs to prepare for, and in which markets to look for work. As investors they decide how much of their ... | 18:26 |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has joined ##hplusroadmap | 18:26 | |
kanzure | ... income to save and how to invest their savings. These savings, which become investments, provide much of the money that businesses need to grow." | 18:26 |
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap | 18:26 | |
fenn | because of loss aversion, i guess. she's afraid of "throwing away everything i've worked for over the past N years" | 18:26 |
kanzure | sunk costs too | 18:27 |
fenn | same thing | 18:27 |
kanzure | and lots of psychological battery like "of course i should be doing this, to think otherwise would be silly" | 18:27 |
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap | 18:27 | |
fenn | i dont think "saving" is a very good description of how money is used as capital | 18:28 |
kanzure | *could be used | 18:28 |
fenn | an agile investor will maintain a diversity of asset types in order to anticipate different situations. money is good for when you need to just buy something quickly | 18:28 |
drethelin | http://www.theonion.com/articles/responsible-man-sets-aside-small-portion-of-every,37200/ | 18:29 |
fenn | there's a whole range of different commodities with varying degrees of liquidity | 18:29 |
fenn | but there's also productive capital, which can also be a commodity | 18:29 |
fenn | for example your mom could sell her 8 foot table sander since she's not using it, but the faster she has to sell the less likely she is to get a good price for it | 18:30 |
kanzure | here's a weird thing for consideration, http://nakamotoinstitute.org/mempool/working-and-saving-are-revolutionary-acts/ | 18:30 |
fenn | "revolutionary" eh | 18:30 |
drethelin | ugh | 18:31 |
drethelin | that's a new trend I could do without | 18:31 |
drethelin | turns out actual revolutions are terrible but people still want to identify as revolutionary | 18:31 |
drethelin | and thus revolutionary goes the way of epic | 18:32 |
kanzure | programmers are a little bit stranger in the capital game because our labor has some really fucked up results | 18:32 |
fenn | i hope you're not generalizing all revolutions based the crap going on in the arab world | 18:32 |
kanzure | e.g. laboring for a few hours can put a million people out of work or something | 18:32 |
kanzure | erm i didn't mean to reference the other people aspect really | 18:33 |
kanzure | just the capital efficiency aspects | 18:33 |
fenn | yes all this theory was invented before automation | 18:33 |
kanzure | see also http://www.forbes.com/sites/venkateshrao/2011/12/05/the-rise-of-developeronomics/ | 18:33 |
drethelin | fenn: nope | 18:34 |
drethelin | most revolutions are bloody shitfests | 18:34 |
drethelin | that don't accomplish what they claim or want to | 18:34 |
drethelin | kanzure: there's also like, musicians | 18:34 |
drethelin | who can now produce stuff that can be copied infinitely | 18:34 |
fenn | but the french revolution brought us the metric system :P | 18:34 |
kanzure | drethelin: what about it | 18:34 |
drethelin | the metric system is kind of shitty | 18:34 |
drethelin | a meter is not a natural kind | 18:34 |
drethelin | a foot is | 18:34 |
drethelin | feet and inches are easy to intuit | 18:35 |
kanzure | that article should have elaborated more on "Companies and individuals cannot develop new technologies without having savings or capital set aside to engage in or finance speculative research." | 18:35 |
fenn | /unfriend drethelin | 18:35 |
kanzure | without any money in savings, people can't take risks really | 18:35 |
drethelin | base 10 is better for science but there's a reason traditional measuring systems are the way they are | 18:35 |
kanzure | (accepting a salary from a company isn't really a risk :p) | 18:35 |
fenn | holy shit are you seriously advocating pounds and fortnights | 18:35 |
-!- snuffeluffegus [~snuff@5.150.254.180] has joined ##hplusroadmap | 18:35 | |
kanzure | his family is made up of biologists, go easy on him | 18:36 |
fenn | i understand how dozenalism is desirable, but the fact is we use base 10 | 18:36 |
drethelin | it's more that I wish the metric system wasn't based on weirdness | 18:37 |
fenn | what weirdness | 18:37 |
drethelin | we can use a base ten system that takes into account intuitional measurements | 18:37 |
kanzure | you mean estimation? | 18:37 |
fenn | a person is 2 meters tall. there's yardsticks waltzing around everywhere | 18:37 |
drethelin | most people aren't 2 meters tall | 18:38 |
kanzure | i am | 18:38 |
fenn | most feet aren't one foot long | 18:38 |
kanzure | mine are | 18:38 |
fenn | i really don't get why the US didnt convert in the 1970s after the auto industry switched over to metric | 18:39 |
fenn | (the US auto industry) | 18:39 |
fenn | now there are two kinds of everything floating around | 18:39 |
kanzure | drethelin: also, many of those weirdo industries are actually very poor | 18:40 |
kanzure | even if you do $1M/year in revenue your company can still be poor | 18:41 |
kanzure | especially when everything is tied up in inventory or supply contracts etc etc | 18:41 |
kanzure | i suspect that a good accountant can make them seem less poor and arrange liquidity deals with banks, but it would be a struggle | 18:42 |
drethelin | mm | 18:43 |
drethelin | lots of stuff you can do with good accounting | 18:43 |
fenn | ouch | 18:43 |
fenn | "In capitalism, every human is either a capitalist, somebody else’s capital, or economically worthless. Today, this abstract point specifically translates to: people who can invest in developers, developers, and everybody else." | 18:43 |
drethelin | like maintain a profitable float | 18:43 |
kanzure | in the context of capitalism i believe developers are more like wizards or magicians | 18:44 |
fenn | yes i've always thought that in general | 18:44 |
kanzure | they do not bode well for your modeling heh | 18:44 |
fenn | like the one reason i tolerate the harry potter franchise is that it may actually represent the culture kids today will grow into | 18:45 |
kanzure | hm. | 18:45 |
drethelin | what do you mean tolerate it | 18:47 |
drethelin | like allow it to exist | 18:48 |
fenn | didn't you know, i run the internet censorship bureau | 18:48 |
kanzure | tolerate as in, he doesn't disappear in a huff of smoke whenever it's mentioned | 18:48 |
fenn | that would be a very wizardly thing to do | 18:51 |
kanzure | yes i have been known to carry a conversation once in a while | 18:51 |
kanzure | i've heard it can be quite magical | 18:51 |
fenn | i mean disappearing in a huff of smoke | 18:52 |
fenn | all the cool wizards are doign it | 18:52 |
kanzure | robot pirate wizard, of course | 18:53 |
kanzure | anywho that author has some other curious things on http://ribbonfarm.com/ | 18:53 |
drethelin | one of my favorite farms | 18:54 |
fenn | you linked the gervais principle thing but i never got to it | 18:54 |
fenn | i got hung up on the name | 18:55 |
fenn | i think he's wrong to lump "chemists, mechanical engineers," in with other kinds of commodity labor | 18:56 |
justanotheruser | https://gist.github.com/dsueiro/4997152 | 18:56 |
justanotheruser | Quote #1: "On May 13th 2012 we downloaded the full public record of this system, which consisted of about 180,000 HTML files." | 18:57 |
justanotheruser | Quote #2: "Nodes broadcast transactions to this network, which records them in publicly available web pages, called block chains, after validating them with a proof-of-work system." | 18:57 |
justanotheruser | Quote #3: "The entire activity in the Bitcoin network is publicly available through the internet and is recorded in the form of a block chain, starting at block 0 (created back on the 3rd of January 2009). Each block reports on as little as a single transaction to as much as over a thousand transactions, and provides hyperlinks to other blocks and to other activities of each address." | 18:57 |
kanzure | i thought it was well known that nobody should bother reading shamir's papers about bitcoin | 18:57 |
kanzure | well, i suppose it's necessary to still write up criticism anyway | 18:59 |
kanzure | very unfortunate waste of time | 19:00 |
kanzure | dingo: https://github.com/schematics/schemalchemy | 19:05 |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Quit: Leaving] | 19:29 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 19:35 | |
dingo | this guy is a bit scant on details, i've read the tests and .py file twice over and i'm a bit perplex | 19:39 |
dingo | i think i need to toy with sqlalchemy a bit to figure out.. why this | 19:39 |
kanzure | here is your 10 second sqlalchemy intro: http://blog.miguelgrinberg.com/post/the-flask-mega-tutorial-part-iv-database | 19:41 |
kanzure | eww it's using sqlalchemy-migrate. ignore all those parts. | 19:41 |
justanotheruser | Isn't it strange that one of the creators of RSA got it so wrong? | 19:44 |
kanzure | nah, academia has its share of conservatism as much as anywhere else | 19:44 |
justanotheruser | conservatism? | 19:44 |
kanzure | hm how do i explain this | 19:46 |
fenn | groupthink? | 19:47 |
dingo | https://en.wikipedia.org/wiki/RC4#History | 19:47 |
dingo | its not rsa but its interesting, the origins | 19:47 |
dingo | jon oberheide is the one who brought it to your linux kernel | 19:48 |
dingo | at the time everybody was like, "is this legal?" | 19:49 |
dingo | and ppl were like "idk ..." | 19:49 |
kanzure | it's also something like... protecting territory or claiming territory by getting papers out faster than everyone else. | 19:49 |
kanzure | so the conservative thing to do is to publish absolutely anything as fast as possible | 19:49 |
kanzure | you'll even notice his dismissals in his email responses to the criticisms | 19:49 |
kanzure | "Yes, well, your criticism is flawed because clearly I published and you didn't." | 19:50 |
drethelin | hah | 19:50 |
kanzure | their responses here are very peculiar: https://bitcointalk.org/index.php?topic=118797.msg1280496#msg1280496 | 19:52 |
justanotheruser | by god that is the most well written email I think I've ever seen | 19:57 |
justanotheruser | "2. We noted that knowledge of multiple private keys is required in this case, and while it is always possible that different owners will share their private keys, this is not likely to happen very often." | 19:59 |
justanotheruser | oh wow, they keep digging themselves deeper | 19:59 |
kanzure | deflection yo | 19:59 |
justanotheruser | just bad assumptions | 20:01 |
justanotheruser | you don't need to share private keys to share a tx | 20:02 |
kanzure | there's also some sort of time-based advantage to publishing early and defending your territory | 20:02 |
kanzure | (citations) | 20:02 |
justanotheruser | what, you gain some legitemacy from citations? | 20:02 |
fenn | citations are the currency of the academic world | 20:03 |
kanzure | you absolutely gain legitimacy from citations | 20:03 |
fenn | it even directly quantitatively impacts your hireability, salary etc in the form of impact factor | 20:03 |
fenn | right now there's this huge explosion of bullshit journals in china just to get people published, otherwise their career will die | 20:05 |
fenn | mostly for chinese academics i think | 20:05 |
kanzure | it's everywhere. most of the u.s. researchers don't need to resort to that. | 20:06 |
kanzure | this guy has been tracking various spammy journals: | 20:06 |
kanzure | http://scholarlyoa.com/ | 20:06 |
fenn | well there's a difference between spammy and scammy | 20:06 |
kanzure | it's both | 20:07 |
fenn | oh wait i'm thinking conference scams | 20:08 |
kanzure | it's very interesting how effective the "call for paper" scams are | 20:08 |
kanzure | to the extent that they can even collect publication fees from the authors | 20:08 |
kanzure | ooh http://scholarlyoa.com/2014/05/08/scholarly-publishing-phishing-attempts-noted/ | 20:08 |
fenn | well they're just doing what the big time criminals have been doing for decades | 20:09 |
kanzure | those are not very good phishing attempts. | 20:09 |
kanzure | as far as i know, elsevier emails don't look like that | 20:10 |
fenn | i was talking about the absurdity of publication fees | 20:10 |
kanzure | i meant the link. | 20:11 |
-!- snuffeluffegus [~snuff@5.150.254.180] has quit [Quit: May the force be with you. Always.] | 20:38 | |
-!- CharlieNobody [~CharlieNo@97-85-246-37.static.stls.mo.charter.com] has quit [Quit: Leaving] | 20:56 | |
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 245 seconds] | 20:57 | |
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Remote host closed the connection] | 20:59 | |
-!- yash [~ffffff@cpe-76-167-105-53.san.res.rr.com] has joined ##hplusroadmap | 21:23 | |
-!- snuffeluffegus [~snuff@5.150.254.180] has joined ##hplusroadmap | 21:24 | |
-!- yashgaroth [~ffffff@2606:6000:cb85:6a00:106:6dc5:c4e3:6f74] has quit [Ping timeout: 265 seconds] | 21:24 | |
-!- Beatzebub is now known as SodelTheBeefcake | 21:26 | |
-!- SodelTheBeefcake is now known as Beatzebub | 21:31 | |
fenn | this is the first "security" thing i've seen that actually makes sense http://www.langsec.org | 21:52 |
kanzure | ""Shotgun parsers", Meredith L. Patterson, Sergey Bratus, November 2012-February 2013" | 21:53 |
kanzure | hm it is all maradydd | 21:53 |
fenn | in simple terms http://www.langsec.org/occupy/ | 21:53 |
kanzure | "Hard-to-parse protocols require complex parsers. Complex, buggy parsers become weird machines for exploits to run on. Help stop weird machines today: Make your protocol context-free or regular." | 21:54 |
kanzure | fair enough | 21:54 |
nmz787 | kanzure: freecad vs blender? | 21:59 |
kanzure | depends on what you're doing | 22:00 |
nmz787 | can you expound on that? | 22:00 |
kanzure | animated videos, use blender | 22:01 |
fenn | blender is not so good at dimensioning | 22:01 |
kanzure | complex scene rendering, with beautiful reflections and impossible light physics.... blender. | 22:01 |
* nmz787 ponders fenn's comment | 22:01 | |
nmz787 | rendering seems like an onion in that is seems like layering of effects | 22:02 |
nmz787 | dimensioning sounds more 'different' than display | 22:02 |
nmz787 | if that makes any sense | 22:02 |
nmz787 | i was reading a bit about that CGAL vs opencascade | 22:02 |
kanzure | http://diyhpl.us/wiki/cadfaq/ | 22:03 |
kanzure | for a gui for mesh tweaking blender is fine. especially for aesthetic-style mesh tweaking. | 22:04 |
nmz787 | kanzure: it seemed like you were concerned with NURBS support and also surface to surface meshing | 22:04 |
nmz787 | or finding the knit-points? | 22:04 |
nmz787 | to connect two surfaces that are getting added | 22:05 |
nmz787 | something like that | 22:05 |
nmz787 | ? | 22:05 |
kanzure | cgal is not a nurbs kernel | 22:05 |
kanzure | i only mentioned surface-surface because verbnurbs does not have a surface-surface intersection algorithm implemented at the moment | 22:06 |
kanzure | here are my notes on opencascade http://diyhpl.us/wiki/cad/opencascade/ | 22:07 |
kanzure | fenn will be able to handle your cad questions /me sleeps | 22:08 |
fenn | sorry /me also sleeps | 22:08 |
nmz787 | i should too | 22:09 |
fenn | i am years out of date wrt open source cad anyway | 22:09 |
kanzure | nothing has happened, cept verbnurbs | 22:10 |
kanzure | and python-brlcad | 22:10 |
kanzure | and cadquery | 22:10 |
fenn | you're going to be up all night remembering cad programs | 22:10 |
kanzure | you have no idea | 22:10 |
nmz787 | so it seems you could write python scripts for blender, render them, but never save the triangles, only the scripts | 22:12 |
nmz787 | then if you needed to massage triangles, you can save your 100 gigs of triangles or whatevefr | 22:13 |
nmz787 | like, isn't NURBS just some symbolic math? so like the sage package might be able to handle? | 22:13 |
-!- snuffeluffegus [~snuff@5.150.254.180] has quit [Quit: May the force be with you. Always.] | 22:13 | |
-!- yash [~ffffff@cpe-76-167-105-53.san.res.rr.com] has quit [Quit: Leaving] | 22:13 | |
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap | 22:17 | |
fenn | yes you could save the scripts only | 22:17 |
fenn | (until the blender api changes and your scripts suddenly dont work) | 22:18 |
fenn | sage might be able to render nurbs data, but trying to do anything useful in it would be equivalent to writing a geometry kernel in some other programming language | 22:19 |
nmz787 | i thought nurbs was just a curved line | 22:22 |
nmz787 | wiki says it came from olden days when making a ship's bow, a strip of wood that was fixed at the end points then tugged on with varying directions and amounts of force... then some dude found out you could model that with vectors along a line | 22:24 |
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 256 seconds] | 22:31 | |
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap | 22:41 | |
nmz787 | has anyone seen a plastic powder/bead FDM print head? something like the HDPE milk bottle recycling to filament would work (with a hopper and an archimedes screw)... except going to a print head diameter, rather than a filament diamter. | 22:52 |
-!- pete4242 [~smuxi@boole.london.hackspace.org.uk] has joined ##hplusroadmap | 22:57 | |
justanotheruser | kanzure: strange. Is there anything that can be done to make scientific papers more robust? | 23:06 |
justanotheruser | The only scientific papers that I understand enough to say why they're certainly wrong are bitcoin related for the most part. Is this common in other papers? | 23:07 |
justanotheruser | I should say understand the topic well enough | 23:07 |
-!- ebowden [~ebowden@147.69.54.84] has joined ##hplusroadmap | 23:13 | |
nmz787 | justanotheruser: how the hell can you evaluate some organic chemistry paper if you don't understand it? I don't see how that's related to the content being robust or not | 23:41 |
justanotheruser | nmz787: I mean understand it on a greater level than the author does | 23:42 |
justanotheruser | Most people who read the document we were talking about would probably take it as is | 23:42 |
justanotheruser | you likely aren't going to do research on every single statement | 23:42 |
-!- |c| is now known as JingoFett | 23:47 | |
nmz787 | that seems like an issue with the reader, that their fact/non-fact processing isn't robust enough to check or make note of thinks they don't understand, for someone else to check | 23:54 |
nmz787 | paperbot: http://dx.doi.org/10.1039/C4LC00576G | 23:55 |
paperbot | http://diyhpl.us/~bryan/papers2/paperbot/c46c0940f219b98d8a17b0dbb2d45dba.txt | 23:56 |
--- Log closed Wed Oct 22 00:00:23 2014 |
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!