--- Log opened Tue Mar 26 00:00:38 2013 | ||
klafka_ | man i can't believe i've optimized shit all up | 00:29 |
---|---|---|
-!- SolG [~Sol@c-174-57-58-11.hsd1.pa.comcast.net] has joined ##hplusroadmap | 00:31 | |
-!- _Sol_ [~Sol@c-174-57-58-11.hsd1.pa.comcast.net] has quit [Ping timeout: 256 seconds] | 00:34 | |
-!- superkuh [~superkuh@unaffiliated/superkuh] has joined ##hplusroadmap | 00:35 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 245 seconds] | 00:46 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 00:59 | |
-!- ThomasEgi [~thomas@185.5.8.81] has joined ##hplusroadmap | 01:07 | |
-!- ThomasEgi [~thomas@185.5.8.81] has quit [Changing host] | 01:07 | |
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap | 01:07 | |
-!- ParahSailin [~eg@unaffiliated/parahsailin] has quit [Read error: Connection reset by peer] | 01:21 | |
-!- klafka_ [~klafka@c-67-174-253-229.hsd1.ca.comcast.net] has quit [Remote host closed the connection] | 01:31 | |
-!- cpopell [~cpopell@pool-96-231-37-73.washdc.fios.verizon.net] has quit [Ping timeout: 256 seconds] | 01:35 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 240 seconds] | 01:47 | |
-!- undersco2 [~quassel@192.210.211.75] has joined ##hplusroadmap | 01:59 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 02:00 | |
-!- Netsplit *.net <-> *.split quits: JayDugger, undersco2_ | 02:06 | |
superkuh | paperbot: http://www.sciencemag.org/content/318/5856/1574.full.pdf | 02:06 |
paperbot | error: HTTP 504 http://diyhpl.us/~bryan/papers2/paperbot/6611c402eb354ee63d5a192e354d8ee6.pdf | 02:07 |
-!- Netsplit over, joins: JayDugger | 02:14 | |
-!- mako [~mako@103-9-42-31.flip.co.nz] has quit [Ping timeout: 264 seconds] | 02:20 | |
-!- BioGuy [~BioGuy@184-76-124-69.war.clearwire-wmx.net] has joined ##hplusroadmap | 02:32 | |
-!- SolG is now known as _sol_ | 02:35 | |
superkuh | paperbot: http://www.nature.com/nature/journal/v493/n7433/full/nature11775.html | 02:36 |
paperbot | http://diyhpl.us/~bryan/papers2/paperbot/LTP%20requires%20a%20reserve%20pool%20of%20glutamate%20receptors%20independent%20of%20subunit%20type.pdf | 02:36 |
-!- abetusk_ is now known as abetusk | 02:38 | |
-!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has joined ##hplusroadmap | 02:42 | |
-!- BioGuy [~BioGuy@184-76-124-69.war.clearwire-wmx.net] has quit [Ping timeout: 245 seconds] | 02:42 | |
-!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has quit [Ping timeout: 272 seconds] | 02:47 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 272 seconds] | 02:48 | |
-!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 02:52 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 240 seconds] | 02:53 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 03:00 | |
-!- EnLilaSko [EnLilaSko@unaffiliated/enlilasko] has joined ##hplusroadmap | 03:22 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 246 seconds] | 03:48 | |
-!- d00hann [~root@84.120.228.103.dyn.user.ono.com] has joined ##hplusroadmap | 03:49 | |
d00hann | que canal mas silencioso | 03:52 |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 03:59 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 04:01 | |
d00hann | hello | 04:17 |
archels | .head http://www.sciencemag.org/content/318/5856/1574.full.pdf | 04:18 |
yoleaux | 200, text/html;charset=UTF-8 | 04:18 |
-!- d00hann [~root@84.120.228.103.dyn.user.ono.com] has quit [Quit: Leaving] | 04:24 | |
archels | Content-Disposition: inline; filename="Science-2007-De Pontieu-1574-7.pdf" | 04:25 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 240 seconds] | 04:48 | |
-!- yorick [~yorick@vredebest.xs4all.nl] has joined ##hplusroadmap | 04:51 | |
-!- yorick [~yorick@vredebest.xs4all.nl] has quit [Changing host] | 04:51 | |
-!- yorick [~yorick@oftn/member/yorick] has joined ##hplusroadmap | 04:51 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 05:02 | |
chris_99 | nmz787, be thee about per chance? | 05:25 |
-!- ikeyjoe [~ikeyjoe@cpc1-hari11-2-0-cust217.hari.cable.virginmedia.com] has joined ##hplusroadmap | 05:40 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 248 seconds] | 05:49 | |
-!- ikeyjoe [~ikeyjoe@cpc1-hari11-2-0-cust217.hari.cable.virginmedia.com] has left ##hplusroadmap [] | 05:51 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 06:02 | |
EnLilaSko | paperbot: http://www.medical-hypotheses.com/article/S0306-9877%2808%2900641-5/abstract | 06:19 |
paperbot | no translator available, raw dump: http://diyhpl.us/~bryan/papers2/paperbot/2cd5f7306a88bc8fa0b85ada68431f6a.txt | 06:19 |
-!- eudoxia [~eudoxia@r186-52-24-97.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 06:24 | |
eudoxia | bina technologies? don't tell me that's rothblatt's wife/partner/transdimensional fellow being | 06:25 |
kanzure | rothblatt has been many things but i'm not sure transdimensional is one of them | 06:26 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 272 seconds] | 06:50 | |
-!- eudoxia [~eudoxia@r186-52-24-97.dialup.adsl.anteldata.net.uy] has left ##hplusroadmap ["leaving"] | 06:59 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 07:02 | |
-!- chidori is now known as chido | 07:11 | |
-!- etheros [~etheros@2601:0:8980:65:21f:f3ff:fe5c:f708] has joined ##hplusroadmap | 07:16 | |
kanzure | man, the "design by committee" vibe is very strange. | 07:25 |
kanzure | "first we need to assemble a committee to define what open source hardware packaging means".... what. | 07:27 |
kanzure | because if none of you know it, maybe if you get together you'll suddenly have a clue? that's never worked. | 07:28 |
-!- eudoxia [~eudoxia@r186-52-24-97.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 07:30 | |
eudoxia | http://opensourceecology.org/wiki/Induction_Furnace | 07:31 |
eudoxia | >Qwiki Reference is no longer available . We've decided to take down that product and focus on Qwiki for iPhone. | 07:31 |
eudoxia | this sort of shit is going to ruin them | 07:31 |
eudoxia | 'where did all our dev files go?' | 07:31 |
kanzure | they are already dead | 07:32 |
eudoxia | you mean OSE as a whole? | 07:32 |
kanzure | i was just channeling kenshiro for a moment there | 07:39 |
eudoxia | ah ok | 07:39 |
kanzure | but yes, this sort of problem is critical | 07:39 |
eudoxia | that's probably some kind of anime reference | 07:39 |
eudoxia | why use google docs for spreadsheets? doesn't mediawiki have tables? can't they write a little js spreadsheet editor to make it easier? | 07:40 |
eudoxia | for example: http://handsontable.com/ | 07:41 |
eudoxia | just make it output HTML <tables> or Medawiki's {| ... |} | 07:42 |
JayDugger | Google Docs spreadsheet works much as Excel does. | 07:43 |
JayDugger | Good luck fighting years of habit. | 07:43 |
JayDugger | Pardon the cynical comment, but I just watched a senior co-worker butcher a software upgrade at work. | 07:44 |
JayDugger | Too busy to back-up the system first; too impatient to read the installer's instructions, its changelog, or its error messages. | 07:46 |
JayDugger | Then he wondered why it didn't go well. | 07:46 |
JayDugger | The kind of person who'd blame his wife for not trying harder and finishing pregnancy ahead of schedule. | 07:47 |
eudoxia | :) | 07:47 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 240 seconds] | 07:50 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 08:04 | |
-!- etheros [~etheros@2601:0:8980:65:21f:f3ff:fe5c:f708] has quit [Ping timeout: 245 seconds] | 08:10 | |
kanzure | storing tables in mediawiki is not efficient | 08:12 |
-!- paperbot [~paperbot@131.252.130.248] has quit [Remote host closed the connection] | 08:15 | |
-!- paperbot [~paperbot@131.252.130.248] has joined ##hplusroadmap | 08:16 | |
kanzure | paperbot: http://www.sciencemag.org/content/339/6121/823.abstract | 08:16 |
kanzure | paperbot: http://www.sciencemag.org/content/339/6121/819.abstract | 08:16 |
paperbot | http://diyhpl.us/~bryan/papers2/paperbot/RNA-Guided%20Human%20Genome%20Engineering%20via%20Cas9.pdf | 08:17 |
paperbot | http://diyhpl.us/~bryan/papers2/paperbot/Multiplex%20Genome%20Engineering%20Using%20CRISPR_Cas%20Systems.pdf | 08:17 |
kanzure | paperbot: http://rna.berkeley.edu/files/jinek_etal_elife_2013.pdf | 08:17 |
paperbot | no translator available, raw dump: http://diyhpl.us/~bryan/papers2/paperbot/d7833ae4c612df9102a65bfa694c67a4.pdf | 08:17 |
kanzure | paperbot: http://www.sciencemag.org/content/337/6096/816.short | 08:18 |
kanzure | paperbot: http://www.nature.com/nature/journal/v471/n7340/abs/nature09886.html | 08:18 |
paperbot | http://diyhpl.us/~bryan/papers2/paperbot/A%20Programmable%20Dual-RNAGuided%20DNA%20Endonuclease%20in%20Adaptive%20Bacterial%20Immunity.pdf | 08:18 |
paperbot | http://diyhpl.us/~bryan/papers2/paperbot/CRISPR%20RNA%20maturation%20by%20trans-encoded%20small%20RNA%20and%20host%20factor%20RNase%20III.pdf | 08:18 |
-!- ParahSailin [~eg@unaffiliated/parahsailin] has joined ##hplusroadmap | 08:23 | |
-!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has joined ##hplusroadmap | 08:34 | |
-!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has quit [Remote host closed the connection] | 08:37 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 245 seconds] | 08:51 | |
-!- klafka [~klafka@c-67-174-253-229.hsd1.ca.comcast.net] has joined ##hplusroadmap | 08:55 | |
-!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 240 seconds] | 08:55 | |
-!- klafka [~klafka@c-67-174-253-229.hsd1.ca.comcast.net] has quit [Remote host closed the connection] | 09:01 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 09:04 | |
-!- eudoxia_ [~eudoxia@r186-52-34-251.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 09:09 | |
-!- eudoxia [~eudoxia@r186-52-24-97.dialup.adsl.anteldata.net.uy] has quit [Ping timeout: 245 seconds] | 09:13 | |
-!- JayDugger [~duggerjw@pool-173-74-81-239.dllstx.fios.verizon.net] has quit [Quit: Leaving.] | 09:28 | |
-!- eudoxia_ [~eudoxia@r186-52-34-251.dialup.adsl.anteldata.net.uy] has quit [Quit: leaving] | 09:38 | |
-!- randallagordon [~randall@174-25-124-45.ptld.qwest.net] has quit [Quit: Changing server] | 09:44 | |
-!- randallagordon [~randall@174-25-124-45.ptld.qwest.net] has joined ##hplusroadmap | 09:45 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 264 seconds] | 09:52 | |
-!- superkuh [~superkuh@unaffiliated/superkuh] has quit [Remote host closed the connection] | 10:04 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 10:05 | |
-!- klafka [~klafka@204-16-157-18-static.ipnetworksinc.net] has joined ##hplusroadmap | 10:20 | |
-!- superkuh [~superkuh@unaffiliated/superkuh] has joined ##hplusroadmap | 10:25 | |
-!- OldCoder_ [~OldCoder_@c-69-181-139-128.hsd1.ca.comcast.net] has quit [Remote host closed the connection] | 10:50 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 256 seconds] | 10:52 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 11:05 | |
-!- lichen_ is now known as lichen | 11:12 | |
-!- archbox_ [~archie@unaffiliated/archbox] has quit [Quit: bye] | 11:30 | |
-!- augur [~augur@208.58.5.87] has quit [Remote host closed the connection] | 11:41 | |
-!- augur [~augur@208.58.5.87] has joined ##hplusroadmap | 11:42 | |
-!- augur [~augur@208.58.5.87] has quit [Ping timeout: 240 seconds] | 11:46 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 256 seconds] | 11:53 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 12:06 | |
-!- augur [~augur@129-2-129-35.wireless.umd.edu] has joined ##hplusroadmap | 12:17 | |
-!- mako [~mako@103-9-42-31.flip.co.nz] has joined ##hplusroadmap | 12:24 | |
-!- mako is now known as sylph_mako | 12:31 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 264 seconds] | 12:53 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 13:07 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 13:07 | |
-!- superkuh [~superkuh@unaffiliated/superkuh] has quit [Quit: when you clone partitions make sure you change the uuid.] | 13:14 | |
kanzure | i am not sure how i am supposed to reply to malcolm | 13:14 |
kanzure | On Tue, Mar 26, 2013 at 2:44 PM, malcolm stanley <a.malcolm.stanley@gmail.com> wrote: | 13:14 |
kanzure | > Bryan, if I organize a Google Hangout for it, will you put together a short | 13:14 |
kanzure | > presentation on ISO 10303 and do a Q&A for us on what it is, what it says, | 13:14 |
kanzure | > and why it matters? | 13:14 |
kanzure | you can go Hangout by yourself | 13:15 |
kanzure | isn't it their *jobs* to know about that long list of standards i sent them | 13:15 |
kanzure | if they think the existing software is buggy, they should file bug reports, or propose something that doesn't have whatever problems they are imagining. | 13:15 |
kanzure | i don't think it should be my job to teach them about all the things they have already claimed to have a responsibility to know about, unless they want to pay me. | 13:16 |
-!- superkuh [~superkuh@unaffiliated/superkuh] has joined ##hplusroadmap | 13:16 | |
kanzure | (also i don't see why they are fixating on ISO. i guess that's a start.) | 13:18 |
kanzure | the fact that he claims to be interested in hardware packaging, but knows nothing about these existing solutions (including existing package managers), indicates to me that he is not very serious about his role and that i would be wasting my time talking with him for any extended period. | 13:20 |
kanzure | ... right? | 13:22 |
ParahSailin | im probably missing some context, whats iso 10303 | 13:26 |
kanzure | cad file format standards, covers a lot of shit | 13:27 |
kanzure | implemented by solidworks/opencascade/autocad/pro-engineer/catia/etc. | 13:27 |
ParahSailin | whats he asking you to do? | 13:28 |
kanzure | most of those implementations are primarily focused on mechanical part representation (the CAD aspects), and i don't think they do anything like list out dependencies for building hardware. but there are existing solutions for managing dependencies in projects that seem completely absent from his discussions/emails. | 13:28 |
-!- augur [~augur@129-2-129-35.wireless.umd.edu] has quit [Remote host closed the connection] | 13:29 | |
kanzure | he is asking me to prepare a presentation to bring him up to speed on open source hardware packaging, file formats, cad standards, etc., to do his job for him | 13:29 |
kanzure | prepare and give the presentation, i should say | 13:29 |
-!- sylph_mako [~mako@103-9-42-31.flip.co.nz] has quit [Ping timeout: 245 seconds] | 13:31 | |
nmz787 | dependencies in hardware can be so varied | 13:34 |
kanzure | if i tell him these things, then he will say "that's nice, i will take these to the committee for consideration" and what's the point?? am i missing something here. how does he think this isn't a waste of time? | 13:34 |
nmz787 | you could take a CAD and make it on a wood lathe, makerbot, resin DLP printer | 13:34 |
kanzure | that's a different type of dependency | 13:35 |
-!- augur [~augur@129-2-129-35.wireless.umd.edu] has joined ##hplusroadmap | 13:35 | |
kanzure | cad-compile-time dependencies are the primary type that most people consider (especially considering the >30000 objects on thingiverse) | 13:35 |
nmz787 | what dependency are you talking about then? | 13:36 |
kanzure | tangiblebit and skdb both had considerations for different types of dependencies (like "run time" and "physical build time") but neither had any traversal algorithm (which is trivial, but it just wasn't written) | 13:36 |
kanzure | cad-compile-time dependencies | 13:36 |
nmz787 | like simply the CAD compiler (SCAD, solidworks, sketchup, rhino, etc) | 13:36 |
nmz787 | ? | 13:36 |
kanzure | for instance, elmom's scad library of parts is an okay example of a dependency if you include any of his parts in your project | 13:36 |
kanzure | have you ever looked at the dependencies of a package when you force apt-get or aptitude to do something? | 13:37 |
nmz787 | yeah, it used to be no fun like 5 years ago | 13:37 |
kanzure | what was no fun? | 13:37 |
nmz787 | but it's been that long since things have really acted up on me | 13:37 |
nmz787 | hunting for thee right version | 13:37 |
kanzure | package managers do that for you | 13:38 |
nmz787 | maybe it wasn't maintained in apt anymore | 13:38 |
kanzure | that's the whole point | 13:38 |
chris_99 | nmz787, you know that linear sensor you used for your spectrometer do you know the highest nm it can pick up. I'm wondering if i can measure the temp. of a flame using it | 13:38 |
nmz787 | then i'd have to go find the right version, download, compile, move on in the apt-get install and hope it didn't find another dependency that wasn't maintained anymore | 13:38 |
-!- superkuh [~superkuh@unaffiliated/superkuh] has quit [Quit: sorry for the reboot cycle spam.] | 13:38 | |
kanzure | what were you recompiling? | 13:38 |
kanzure | i don't understand. | 13:39 |
nmz787 | chris_99: prob not higher than 900 or 1100nm | 13:39 |
chris_99 | ooh | 13:39 |
nmz787 | probably openCV | 13:39 |
chris_99 | that's probably be enough | 13:39 |
nmz787 | or multi-pointer X | 13:39 |
nmz787 | yeah, xorg i think it was | 13:39 |
kanzure | i'm pretty sure opencv has a debian maintainer | 13:39 |
nmz787 | it was a fork at the time | 13:39 |
kanzure | xorg also has lots of package maintainers | 13:39 |
nmz787 | mpx wasn't mainstreamed | 13:39 |
kanzure | i don't know why you are bringing it up | 13:39 |
nmz787 | my experience with dependencies | 13:40 |
kanzure | i was trying to illustrate a software tool that resolves dependencies without human intervention | 13:40 |
nmz787 | findind them manually | 13:40 |
nmz787 | chris_99: that's most silicon based light sensors, now if they have an IR filter that would kill the sensitivity | 13:41 |
nmz787 | chris_99: try removing the IR filter from a common webcam | 13:41 |
chris_99 | mm, do you know how you would actually measure the temp from the lines from the grating, i'm a bit confused how that'd be done | 13:42 |
nmz787 | you mean like flame spectroscopy? | 13:44 |
nmz787 | chris_99: you could just try hacking this $12 item http://www.aliexpress.com/item/Professional-hand-held-non-contact-DT-380-Digital-LCD-infrared-IR-thermometer/629051292.html | 13:44 |
chris_99 | alas that won't handle the temp. i plan on looking at 3500C | 13:45 |
chris_99 | and yep flame spectroscopy i guess | 13:45 |
nmz787 | chris_99: that might simply require adding a resistor | 13:45 |
nmz787 | or an aperature | 13:45 |
chris_99 | mm that's an idea, the aperture could work maybe | 13:46 |
chris_99 | i've got one of those myself, will see how that effects it | 13:46 |
chris_99 | my bet is even with an aperture it'll overwhelm the sensor | 13:46 |
nmz787 | how? | 13:50 |
nmz787 | an aperature reduces photons coming in | 13:50 |
chris_99 | hmm i guess so, but why can't i seem to find any IR thermometers that do above 1500C though | 13:51 |
-!- OldCoder_ [~OldCoder_@209.237.22.146] has joined ##hplusroadmap | 13:51 | |
nmz787 | if that was the case, than the aperature wasn't big enough, and probably the sensitivity of the sensor amp needs changed | 13:51 |
nmz787 | whihc would likely just be adding or removing a resistor | 13:51 |
nmz787 | (either in=line with the sensor, or on the amp gain selector, if there's a gain selector) | 13:52 |
kanzure | "I would be happy to speak with you about existing standards and on-going progress in open source hardware file formats/packaging, either over email, irc, voip, phone, or in person. I am curious as to what outcomes you are expecting from that sort of conversation. In particular, it strikes me as odd that you would ask why these things matter-- you run into them constantly if you do any hardware development (it's seriously hard to find a ... | 13:52 |
kanzure | ... commercial CAD tool that doesn't support STEP), and in the software world there's a tremendous amount of standardization on packaging formats, which has led to hundreds of thousands of standard 'components'. If these things are not evident to you, I would be interested in knowing what you are doing here..." | 13:52 |
kanzure | perhaps too harsh? | 13:53 |
nmz787 | maybe the end | 13:53 |
nmz787 | yes | 13:53 |
kanzure | but i mean it :( | 13:53 |
nmz787 | well | 13:53 |
nmz787 | you can still mean it, but say it better | 13:53 |
kanzure | "yo dawg, what the fuck?" | 13:54 |
nmz787 | heh | 13:54 |
nmz787 | better in some ways, lol, worse in others | 13:54 |
nmz787 | meh | 13:54 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 248 seconds] | 13:54 | |
nmz787 | it's pretty good | 13:54 |
nmz787 | it | 13:54 |
nmz787 | it's straightforward | 13:54 |
nmz787 | which I tend to like | 13:54 |
nmz787 | maybe change what you are doing here... | 13:55 |
nmz787 | I would be interested to know your status and understanding. Understanding STEP and other standards is pre-requisite to advancing our cause... | 13:56 |
nmz787 | or something | 13:56 |
kanzure | no, that would sound elitist | 13:57 |
nmz787 | now /that/ i don't understand | 13:57 |
kanzure | and frankly, STEP isn't really that important, the format sucks, and there's like 100 different 500 page documents defining AP101-AP4xx, and they are all based on a custom language (EXPRESS) that compiles the standards into C++ that compiles into god-knows-what. | 13:57 |
nmz787 | really it's true though, how can you evaluate a field if you don't dig in? | 13:58 |
klafka | that is true | 13:58 |
nmz787 | klafka: what? | 13:58 |
klafka | how can you evaluate a field if you don't dig in | 13:58 |
nmz787 | it seems that these folks want to dig in, via you bryan, via a braindump | 13:59 |
nmz787 | so honestly give them what you can | 13:59 |
nmz787 | what you don't need to prep for | 13:59 |
nmz787 | too much prep at least | 13:59 |
nmz787 | it's still their responsibility if they have some grander vision | 13:59 |
kanzure | brain dumping doesn't guarantee me that they will take my information and make something useful out of it | 13:59 |
nmz787 | no | 13:59 |
nmz787 | kids paying $50k/year to college doesn't mean they understand shit either | 14:00 |
nmz787 | but they still pay | 14:00 |
nmz787 | you're not paid | 14:00 |
nmz787 | so you have to assume/hope they'll use it | 14:00 |
kanzure | so far they have demonstrated an unwillingness to review solutions that others (including myself) have proposed, or file bug reports (not just on my own work, but everything else too), etc., so why would i think this time is any different? | 14:00 |
nmz787 | that's one reason having a slide deck or a recorded google hangout will be valuable to the greater community if these folks ignore you | 14:01 |
nmz787 | so other's can take your facts/knowledge and make use | 14:01 |
nmz787 | kanzure: maybe that should be your second or third email paragraph | 14:01 |
nmz787 | not harsh, not sugarcoated | 14:01 |
nmz787 | very realistic for a busy person | 14:02 |
kanzure | i don't think so, it feels wrong to suggest that they are acting in bad faith | 14:02 |
kanzure | even though i fully think they are :) | 14:02 |
nmz787 | saying 'i've been burned' is OK though | 14:02 |
nmz787 | that's why i said give minimal prep time | 14:03 |
nmz787 | 'i've been burned, but I still have faith to give more' | 14:03 |
kanzure | i don't have faith. | 14:03 |
nmz787 | you'll anyway cross-post the hell out of a chat session/video/slide deck | 14:04 |
kanzure | even if this "open source hardware" committee of theirs comes up with something, does that mean they are going to go look around for someone to write all the software.. and then what.. we end up with some shitty software? | 14:04 |
nmz787 | so it won't be totally wasted if they don't listen | 14:04 |
nmz787 | or they come back to you | 14:04 |
kanzure | and if it ends up me writing their shitty software, ugh | 14:04 |
nmz787 | and you say you need $ for more commitment | 14:04 |
kanzure | committees have this tendency to come up with really really terrible solutions ("XML FOR EVERYTHING") | 14:04 |
nmz787 | lol | 14:05 |
nmz787 | i can't even understand why the youtube API uses XML in something called an Atom feed | 14:05 |
kanzure | ("WE DON'T HAVE MONEY BECAUSE WE ARE OPEN SOURCE") | 14:05 |
kanzure | (which is a lie) | 14:05 |
kanzure | atom is based on xml | 14:05 |
kanzure | pretty sure. | 14:05 |
kanzure | atom and rss were two competing syndication standards, i think. | 14:05 |
kanzure | "atom is an IETF standard while rss is not" aha | 14:06 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 14:07 | |
@fenn | give quantumG my apologies, holy crap how high can it go: http://blockchain.info/charts/market-price | 14:08 |
kanzure | 2010-02-10.log:23:08 < fenn> ah well at least they sell bitcoins via paypal | 14:09 |
kanzure | 2010-02-10.log:23:12 < fenn> currently $0.0038/bitcoin | 14:09 |
ParahSailin | are silk road prices lower than street prices? | 14:10 |
@fenn | let me just point out that nobody else hoarded any bitcoins either | 14:10 |
kanzure | did quantumg hoard any? | 14:10 |
@fenn | dunno | 14:10 |
kanzure | i kept thinking ybit did, but he says he didn't. | 14:10 |
kanzure | wasn't he going through his crypto-currency streak around that time? | 14:11 |
@fenn | ParahSailin: generally silk road is 2-10 times street price, depending on where you live and what's being sold | 14:11 |
ParahSailin | hm, so im not seeing much rational utility of the currency yet | 14:12 |
@fenn | some people value "not having to talk to some shady drug dealer and maybe get busted" even if this isn't an unbiased depiction of reality | 14:13 |
kanzure | transfers seem to be very much easier than using other systems, if that counts for anything. | 14:14 |
@fenn | also i think you can't buy porn with paypal(?) | 14:14 |
@fenn | sometimes i think alien time travelers inserted puritanical morals into the timeline in order to spur human technological development | 14:15 |
kanzure | the bitcoin laundering schemes are sorta hilariously wrong (they don't actually hide originating sources) | 14:16 |
@fenn | you have to actually swap coins, no? | 14:16 |
ParahSailin | well you have to have some sort of basic understanding of cryptography and math to understand why bitcoin is not in fact anonymous | 14:17 |
kanzure | well there's some services where they just distribute your bitcoins into a pool of a few thousand wallets and then they try to obfuscate the path the money followed | 14:17 |
ParahSailin | can't blame the majority for being completely clueless | 14:17 |
kanzure | by "obfuscate" i mean "attempt to hide by increasing the path length" | 14:17 |
kanzure | software development is a really unique way to launder money, though. | 14:17 |
@fenn | maybe i'm clueless but can't you just swap coins and not tell anyone? | 14:17 |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 264 seconds] | 14:17 | |
kanzure | transactions appear in the blocks http://blockexplorer.com/ | 14:18 |
@fenn | then don't do the transaction that way | 14:18 |
kanzure | oh you mean sell a wallet's identity? | 14:18 |
@fenn | make two new wallets with 0 bitcoins, add money, trade wallets | 14:18 |
kanzure | but then you have 2 people who each have information about where the bitcoins went | 14:19 |
ParahSailin | thats exactly what you're supposed to do-- recipient is supposed to create new wallet for money to be sent to | 14:20 |
kanzure | ok, if the recipient creates the new wallet, then that's fine, as long as that new wallet never interacts with the recipients' other wallets. ever. | 14:20 |
ParahSailin | kanzure: yes, i know | 14:21 |
kanzure | but this seems particularly hard to make useful, say you have 1000 wallets with bitcoins in each and you are trying to make use of your money.. you're back to the same problem basically. | 14:21 |
ParahSailin | bitcoin is not anonymous-- it can't be, to be zero trust irrevocable transactions | 14:21 |
ParahSailin | zero trust irrevocable transactions is a valuable property, but orthogonal to anonymity | 14:22 |
ParahSailin | if you want anonymity, you gotta trust a middle man to try to launder | 14:22 |
@fenn | hm you'd expect google to at least recognize "BTC" as a currency type | 14:24 |
-!- plur [~nonentity@121-73-87-49.cable.telstraclear.net] has quit [Remote host closed the connection] | 14:25 | |
ParahSailin | i havent been paying attention to whatever laundering services are out there, but it's conceivable that you could do a safe laundering if you deposited currency with them to maintain a positive balance in your account, and they would forward on a non-equal amount to what you deposited to the end set of destination wallets | 14:25 |
ParahSailin | it's entirely possible that that's how the existing ones work | 14:25 |
kanzure | ParahSailin: i think another way to make anonymity happen is to have a unit of work/value that you can pay for in tainted currency while anonymous, and then take that "created valuable" and sell it somewhere for clean money. the trick is that the item has to be something that neither the original creator (the person getting the tainted money) or the end customer (the person paying you non-tainted money) will be able to recognize after the ... | 14:27 |
ParahSailin | i dont think there's been any major bitcoin depository "institution" that hasn't gotten compromised in some way in the last year | 14:27 |
kanzure | ... fact... e.g. painting logos for an oil company wont do it, the worker will know. | 14:27 |
ParahSailin | anonymity relies on trust, regardless of the scheme | 14:28 |
kanzure | generic software could work, especially if it's privately-deployed software that is unlikely to reach lots of consumers. | 14:28 |
-!- plur [~nonentity@121-73-87-49.cable.telstraclear.net] has joined ##hplusroadmap | 14:28 | |
kanzure | or you can go the other way around, and make it completely open source work that you are paying for with tainted coinage | 14:29 |
kanzure | what are they going to do? trace bad money and they find an open source project.. and what.. repossess it? take it down? at that point, everyone has had a chance to access it and copy it / fork it. | 14:29 |
@fenn | the idea with money laundering is to have lots of legitimate users (or at least the perception of such) | 14:30 |
@fenn | in order to get other types of money/services in and out | 14:30 |
kanzure | security through numbers? really, fenn? | 14:30 |
kanzure | disappointedi n you | 14:30 |
kanzure | *in you | 14:30 |
kanzure | *through quantity | 14:31 |
@fenn | if you can just arrest all 5 people using pokecoins you lose | 14:31 |
@fenn | think about tor; if there were only 5 exit nodes it would be easy to just block them all | 14:32 |
kanzure | man, we should get a cypherpunk in here. let's see some real paranoia. | 14:32 |
ParahSailin | the only purpose of bitcoin laundering services is to allow you to buy from people you don't trust, without revealing identity | 14:32 |
kanzure | eugen leitl doesn't count (he's insufficiently paranoid, even though he's a cypherpunk member) | 14:32 |
ParahSailin | if delivery of service would not reveal identity to the seller, it's unncessary | 14:33 |
@fenn | given the size of the cypherpunks list i'd be surprised if there weren't one here already | 14:34 |
kanzure | eleitl. | 14:34 |
@fenn | he's not here :P | 14:34 |
ParahSailin | you could run paperbot off bitcoin, and the delivery mechanism would be a random web based drop box | 14:34 |
kanzure | ok how does this sound http://diyhpl.us/~bryan/irc/oshw-standards-email.txt | 14:34 |
kanzure | ParahSailin: part of the point of paperbot is community reading | 14:35 |
@fenn | but botsnacks would be so much more useful if you could buy processing time with them | 14:35 |
@fenn | think of all the captchas | 14:35 |
ParahSailin | tor hidden service could work | 14:35 |
kanzure | so far i haven't seen an academic publisher that implements a captcha system | 14:36 |
@fenn | "which are hard to escape the reaches from on most computers | 14:36 |
kanzure | also, deathbycaptcha.com is sufficiently cheap for me to fund all future paperbot accesses that require captcha solving | 14:36 |
kanzure | fenn: at least computers worth using | 14:36 |
@fenn | you'll use" should just be "which you will find on most computers" | 14:36 |
kanzure | heh servers | 14:37 |
kanzure | okay | 14:37 |
kanzure | re deathbycaptcha see | 14:37 |
kanzure | https://github.com/kanzure/python-deathbycaptcha | 14:37 |
@fenn | what is a Hangout anyway | 14:38 |
kanzure | i dunno if thingdoc is worth mentioning; i still hate .scad, but at least there's a documentation tool. | 14:38 |
kanzure | Google Hangout is this in-browser video conferencing thing.. | 14:39 |
@fenn | is it just group video chat? | 14:39 |
kanzure | yes | 14:39 |
kanzure | also you can add funny hats to people | 14:39 |
@fenn | why does that need a name | 14:39 |
brownies | and moustaches | 14:39 |
kanzure | brownies: can you wordsmith-check that link? http://diyhpl.us/~bryan/irc/oshw-standards-email.txt | 14:40 |
kanzure | i sort of have to be political around these people, because the CEO of sparkfun ends up reading my email | 14:40 |
kanzure | and other bigwigs | 14:40 |
@fenn | industrial titans of our era | 14:40 |
kanzure | hardly. | 14:41 |
ParahSailin | who is malcolm stanley | 14:41 |
kanzure | but i would like them to think positively of me if at all possible, while at the same time also telling them that i am disappointed in them. if that's possible. | 14:41 |
brownies | heh | 14:41 |
brownies | kanzure: well, i think he is asking more "how it matters" not "does it matter" | 14:41 |
brownies | it's possible that he doesn't know if or why it matters, but that seems... at least just from this piece of text... less likely. | 14:42 |
kanzure | that's an *absurd* question for these people to ask. they are supposed to be the "open source hardware association". how the fuck could they not know... | 14:42 |
@fenn | does he know that ISO-10303 == .step files? | 14:43 |
kanzure | dunno | 14:43 |
@fenn | that would be a good place to start | 14:43 |
kanzure | btw there's a lot of stuff in 10303 other than just .step files | 14:43 |
@fenn | eh i guess | 14:43 |
kanzure | i shouldn't really be pimping ISO shit here anyway | 14:43 |
@fenn | i was the one that started it :\ | 14:44 |
kanzure | charlie stix runs a company called costvision and i think he's written a bunch of software for iso, so i am just deferring that shit to him now | 14:44 |
kanzure | *bunch of software for iso 10303 things | 14:44 |
kanzure | plus he has read opencascade source code and seems to be vaguely interested in open source things | 14:44 |
kanzure | ParahSailin: seems to be this person http://www.linkedin.com/in/amstanley | 14:45 |
nmz787 | kanzure: email looks ggood to me | 14:46 |
kanzure | On Tue, Mar 26, 2013 at 9:20 AM, Matt Maier <blueback09@gmail.com> wrote: | 14:47 |
kanzure | > I suspect that once we figure out what we need it will be easy to discover | 14:47 |
kanzure | > several pre-existing software tools that either already do exactly that, or | 14:47 |
kanzure | > can easily be forked to do exactly that. | 14:47 |
kanzure | > The problem is that we don't know what we need. | 14:47 |
kanzure | wtf | 14:47 |
kanzure | > overhead. I doubt even OSE is doing any project big enough to justify that | 14:47 |
kanzure | > much complexity. A giant, complicated format like STEP could obviously | 14:47 |
kanzure | with you so far | 14:47 |
kanzure | > describe simple projects, but it would also create a huge barrier to entry | 14:48 |
kanzure | > The best way to find out is to | 14:48 |
kanzure | > compare it to a bunch of other people's impressions so we can come up with a | 14:48 |
kanzure | > generally applicable solution | 14:48 |
kanzure | .... do people seriously read the shit they type? | 14:48 |
nmz787 | what's wrong with comparing? | 14:48 |
@fenn | it sounds like they are seriously underestimating the scale of the problem | 14:48 |
nmz787 | comparing *exists* with *expected*/*desired* | 14:48 |
kanzure | comparing impressions != technology development | 14:48 |
kanzure | > At any rate, the discussion has to happen before any implementation makes | 14:49 |
kanzure | > sense. | 14:49 |
kanzure | i think that's a blatant lie.. thingdoc works really well for .scad. and there was little discussion about it. it was an obvious implementation based on previous things like docstrings, javadoc, and other comments-to-documentation generators. | 14:49 |
kanzure | i suppose he will just handwave that by saying thingdoc is simple and therefore irrelevant | 14:50 |
brownies | wtf is his point? | 14:51 |
nmz787 | so which of these formats simply stores voxels that == atoms? | 14:51 |
nmz787 | that seems like the space-wasting but easy way out | 14:51 |
kanzure | pdb | 14:51 |
nmz787 | lol | 14:51 |
kanzure | it's not that helpful. | 14:51 |
nmz787 | new standard CAD format | 14:52 |
nmz787 | :P | 14:52 |
kanzure | no | 14:52 |
kanzure | sending stl or voxels is like sending me a jpeg of your compiled binary, instead of sending me your C source code. | 14:52 |
nmz787 | no more NURBS, just atoms | 14:52 |
nmz787 | true | 14:52 |
nmz787 | unless voxel files could be compressed well | 14:53 |
kanzure | email ahoy. now to sit around and wait for the hate mail to roll in. | 14:53 |
nmz787 | set your expectations higher, call it 'love mail' | 14:54 |
@fenn | mail of intense yet undescribable feelings | 14:54 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 245 seconds] | 14:54 | |
nmz787 | you've already been nominated for a nobel prize this week | 14:54 |
kanzure | my space ship is powered by the heat from burning hopes, dreams and nobel prizes | 14:55 |
nmz787 | so i was thinking of getting my truck painted | 14:55 |
kanzure | god damn it i got charlie's name wrong | 14:55 |
kanzure | none of you caught that? | 14:55 |
nmz787 | then i thought maybe i would just drive it to burning man with a case or three of spray paint and clear coat | 14:56 |
brownies | a nobel prize? | 14:56 |
nmz787 | fenn: would that fly there? aside from dust coating the truck? | 14:56 |
@fenn | nmz787: you could paint it like this http://i3.asn.im/Pikachu-airplane-_tsi5.jpg | 14:56 |
brownies | nmz787: just spray paint it | 14:56 |
nmz787 | brownies: well i want a reason to go to burning man anyway | 14:56 |
brownies | fenn++ | 14:56 |
nmz787 | brownies: and I'm not too creative, it would end up simply being black | 14:57 |
nmz787 | s/creative/artistic/ | 14:57 |
kanzure | oh look, someone cared enough to merge my changes https://github.com/jeanphix/Ghost.py/pull/73 | 14:57 |
@fenn | you're painting it because you want it to look good for burning man? 0_o | 14:57 |
nmz787 | no | 14:57 |
nmz787 | i just want to keep the rust off for a few more years | 14:58 |
nmz787 | 85 born and going strongish | 14:58 |
nmz787 | that pokemon scheme isn't bad | 14:58 |
@fenn | http://rpad.tv/wp-content/uploads/2010/03/Pikachu-school-bus-2.jpg | 14:59 |
nmz787 | getting it painted at burning man would likely mean it would look somewhat artistic, rather than like I was parked outside a paint factory that exploded | 14:59 |
-!- mako [~mako@wireless-nat-7.auckland.ac.nz] has joined ##hplusroadmap | 15:00 | |
@fenn | wtf there's an inflatable version of that airplane http://s1108.photobucket.com/user/Pinkproposal/media/plane.png.html | 15:02 |
brownies | nmz787: turn it into its own cheesy modern-art installation | 15:03 |
kanzure | fenn: there is a cheap (~$300) linux 2.6.32 quadrotor on the market. i am pondering ways to make it automatically replace its own battery, possibly through a docking station, so that it could fly continuously. | 15:03 |
brownies | nmz787: drive your truck there, leave out a bunch of paint, and let people express themselves | 15:03 |
brownies | nmz787: bam, free paint job | 15:03 |
@fenn | sounds like a job for lego | 15:03 |
nmz787 | brownies: pretty much my thinking | 15:03 |
@fenn | kanzure: the magnet idea sounded pretty reasonable | 15:03 |
brownies | you might look a damn dirty hippie driving it around afterwards | 15:04 |
nmz787 | fenn: hmm, i wonder if I should use legos to rig up my projector to my microscop | 15:04 |
brownies | kanzure: link to quadrotor? | 15:04 |
kanzure | fenn: what about the forces due to running tight curves in air? wouldn't that make the battery not stay attached.. | 15:04 |
@fenn | the forces aren't very high until it crashes | 15:04 |
kanzure | brownies: http://diyhpl.us/~bryan/irc/nodecopter.txt | 15:05 |
@fenn | and even then, the magnets are self-healing | 15:05 |
kanzure | oops, refresh nodecopter.txt for the better version | 15:06 |
@fenn | you'd want separate electrical spring contacts and magnetic mechanical contacts | 15:06 |
kanzure | one of these papers/videos was saying 2G's on some of the curves | 15:06 |
@fenn | 2G is nothing | 15:06 |
kanzure | neodymium magnets will hold? | 15:07 |
@fenn | well it depends on the magnets doesn't it? how long is a string? | 15:07 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 15:08 | |
@fenn | you just need a good ratio of (force to remove battery)/(expected operational forces) | 15:08 |
@fenn | if it's too high you won't be able to get the battery off | 15:09 |
@fenn | do they run on cell phone batteries or what? | 15:09 |
kanzure | looks like lipo batteries the size of a cell phone | 15:10 |
kanzure | http://ardrone2.parrot.com/usa/ is the one that the linux/javascript community has been cross-compiling to | 15:10 |
-!- qu-bit [~shroedngr@unaffiliated/barriers] has quit [Remote host closed the connection] | 15:12 | |
brownies | kanzure: nice link collection, thanks | 15:12 |
kanzure | 1000mAH or 1500mAH battery that provides 12 minutes of 5m/sec (18 kmph) | 15:12 |
@fenn | yeah that's about the same as a cell phone battery | 15:14 |
kanzure | it looks like a really appealing toy, except i absolutely know i would be frustrated with short flights, so even if i wrote out some code, it would involve waiting to test alterations, unless i buy 8-10 batteries and swap them out, ugh. | 15:16 |
-!- qu-bit [~shroedngr@unaffiliated/barriers] has joined ##hplusroadmap | 15:16 | |
@fenn | some of those "technical specifications" are things you would not be advertising if your customers were well informed | 15:16 |
kanzure | hm? | 15:16 |
@fenn | like "emergency stop controlled by software" | 15:16 |
kanzure | it's in their ipad app.. it's an emergency stop button you can press. | 15:17 |
@fenn | no that's talking about the "fully reprogrammable motor controller" | 15:17 |
@fenn | so your friends can hack into your drone and make it explode in mid-air | 15:17 |
@fenn | what the hell were they thinking | 15:18 |
kanzure | your friends can hack into them anyway, port 23 is open and gives you a shell. | 15:18 |
@fenn | yes but they shouldn't be able to reprogram the motor controller | 15:18 |
kanzure | https://github.com/substack/virus-copter/blob/master/virus.js | 15:18 |
kanzure | i don't think you can reprogram the motor controller with the default software. | 15:18 |
@fenn | seems easy enough to immunize, just broadcast "INFECTED" regardless | 15:19 |
kanzure | haha | 15:19 |
kanzure | his point was that the default security is weak | 15:20 |
@fenn | um. but port 23 is open and gives you a shell | 15:21 |
kanzure | don't you think that's dangerous? | 15:21 |
@fenn | yes, but at least there's a point to it | 15:22 |
@fenn | whereas i can't see any reason for letting people reprogram motor controllers aside from the glee of watching your opponent go down in flames | 15:22 |
kanzure | given the option, i would take the open port because it means i wouldn't have to flash it | 15:22 |
kanzure | or, at least, i wouldn't have to flash it so soon | 15:22 |
@fenn | werent we supposed to have chip turbines by now | 15:25 |
@fenn | methanol or butane just seems so much better suited for this kind of thing | 15:25 |
kanzure | the timeline oracle hasn't reported any discrepancies. | 15:25 |
@fenn | i mean 12 minutes is pretty weak | 15:26 |
kanzure | yes. | 15:26 |
@fenn | why are people using batteries at all | 15:26 |
kanzure | no clue | 15:26 |
kanzure | i'm very conflicted on it. on the one hand.. flying robot with linux and javascript. on the other hand, low battery life and i would have to build that docking station for auto swapping. :/ | 15:26 |
kanzure | constant flying robot is very appealing to my nerd senses. especially with what.. it has a 2 mile max range before it has to turn around (assuming no wind resistance issues)? | 15:26 |
nmz787 | fenn: wouldn't that just be removing a jumper on the data line of the controller before flying it? | 15:27 |
nmz787 | kanzure: pretty cool, how much $ total with batteries? | 15:27 |
nmz787 | and legos | 15:27 |
kanzure | $300 with one battery, extra batteries look $40-$50/each or something.. so <$1000 for an always-on system. | 15:27 |
nmz787 | and the weight limit is already there with those batteries? | 15:28 |
ParahSailin | make it an actual airplane if you want it to go farther | 15:28 |
nmz787 | already maxed* | 15:28 |
kanzure | someone said 100g payload limit, dunno if that's with the plastic shit or not | 15:28 |
@fenn | make it a blimp if you want longer airtime :P | 15:28 |
nmz787 | ahh so adding more batteries isn't much an option | 15:28 |
ParahSailin | you get a factor of 15 extra range just going to airplane | 15:28 |
kanzure | but no positional stability | 15:28 |
nmz787 | VTOL? | 15:29 |
kanzure | loops? | 15:29 |
ParahSailin | ulrich flyer? | 15:29 |
nmz787 | inflatable on-demand blimp? | 15:29 |
nmz787 | get to your location fast and with precision once there, then cruise slowly home | 15:30 |
nmz787 | though there's the wind factor | 15:30 |
@fenn | kytoon is the answer to wind | 15:30 |
nmz787 | do they make non-electric copters? | 15:31 |
@fenn | yes | 15:31 |
nmz787 | do they have better range? | 15:31 |
@fenn | yes | 15:31 |
nmz787 | kanzure: have you looked into that? | 15:31 |
nmz787 | no batteries | 15:32 |
@fenn | they're usually nitromethane or "diesel" which is more like ether mixed with lamp oil | 15:32 |
nmz787 | wonder if you could use veg oil | 15:32 |
nmz787 | with some ether | 15:32 |
@fenn | it would gunk up | 15:32 |
-!- augur [~augur@129-2-129-35.wireless.umd.edu] has quit [Remote host closed the connection] | 15:32 | |
nmz787 | not light enough? | 15:32 |
@fenn | crosslinking caused by peroxidation at high temperatures | 15:33 |
nmz787 | he /is/ in TX | 15:33 |
@fenn | wut? | 15:33 |
ParahSailin | internal combustion engines dont really scale well to that small | 15:33 |
kanzure | i expect that they are larger | 15:33 |
kanzure | yes | 15:33 |
nmz787 | yeah but they're already popular in RC cars | 15:33 |
@fenn | they are very noisy and not as efficient as larger engines | 15:33 |
kanzure | quadrotors are already pretty noisy | 15:34 |
ParahSailin | you can only do two-stroke below a certain size | 15:34 |
@fenn | so the range benefit is not as good as you'd expect | 15:34 |
nmz787 | and kanzure wanted range which fenn said increases | 15:34 |
kanzure | no i wanted flight time | 15:34 |
kanzure | or i want auto-refueling (which is much easier with liquid fuels i think) | 15:34 |
nmz787 | i assume that's the same within a copter | 15:34 |
ParahSailin | kanzure: what is the application? | 15:35 |
-!- strangewarp [~C@c-67-176-51-26.hsd1.co.comcast.net] has quit [Read error: Connection reset by peer] | 15:35 | |
nmz787 | i guess hovering would take less fuel than forward motion | 15:35 |
-!- strangewarp [~C@c-67-176-51-26.hsd1.co.comcast.net] has joined ##hplusroadmap | 15:35 | |
kanzure | oh, you know. just have a flying robot around the house or to go out and do my evil bidding. | 15:35 |
nmz787 | so it's not a great assumption | 15:35 |
nmz787 | 'in the house fumes would then be an issue | 15:35 |
ParahSailin | ah, then quadrotor and an inductive charger station | 15:35 |
kanzure | so it would perch on the charging station for an hour, then go fly for 10 minutes, and then perch again? | 15:35 |
kanzure | i think the battery idea was better :P | 15:35 |
nmz787 | you'd have to have the gasser for outdoor ops only | 15:36 |
nmz787 | 'get me a taco' | 15:36 |
kanzure | http://tacocopter.com/ | 15:36 |
nmz787 | that would totally fucking work | 15:36 |
@fenn | nmz787: actually slight forward motion tends to work better because there's this "donut effect" if you stay still | 15:36 |
nmz787 | you could have it deliver in your neighborhood for convenience fees too | 15:36 |
ParahSailin | for deliveries an airplane is much more efficient than quadrotor | 15:36 |
-!- safitan [~safitan@75-105-12-23.cust.wildblue.net] has joined ##hplusroadmap | 15:36 | |
-!- ArmilusDajjal [~safitan@75-105-12-23.cust.wildblue.net] has quit [Ping timeout: 256 seconds] | 15:36 | |
nmz787 | maybe, if it was cheap enough in fuel and maintenance costs | 15:36 |
@fenn | kanzure: i talked to her about tacocopter and she seemed amused but thought FCC was too big of a hurdle to seriously go for it | 15:37 |
kanzure | fixed-wing is definitely better for payloads, sure | 15:37 |
nmz787 | ParahSailin: how would you land on a porch? | 15:37 |
kanzure | fenn: who is her? | 15:37 |
@fenn | star simpson | 15:37 |
kanzure | nmz787: sky drops | 15:37 |
ParahSailin | parachute drop | 15:37 |
@fenn | so tacocopter.com is just a joke as far as i can tell :( | 15:37 |
nmz787 | that seems, unreliable... | 15:37 |
kanzure | yes it's a joke | 15:37 |
nmz787 | 'my taco is on the porch below' | 15:37 |
ParahSailin | if payments are in btc, who cares about fcc, faa etc | 15:37 |
@fenn | i don't really understand why it's a joke | 15:37 |
kanzure | because people are cruel | 15:37 |
@fenn | but she's an engineer that could actually do it | 15:38 |
kanzure | oh is she? | 15:38 |
kanzure | are you sure | 15:38 |
* fenn shrugs | 15:38 | |
@fenn | it's not that hard, "a simple matter of programming" | 15:38 |
nmz787 | throw a pico projector on it too maybe, to have a facial 'presence' | 15:39 |
kanzure | why? | 15:39 |
nmz787 | people like that | 15:39 |
nmz787 | trust, etc | 15:39 |
@fenn | haven't you played halo | 15:39 |
nmz787 | same reason people want androids | 15:39 |
kanzure | i want androids because i want to enslave the human race | 15:39 |
kanzure | what reason do you have? | 15:39 |
nmz787 | or to display ads | 15:39 |
-!- Juul [~Juul@50-0-83-116.dsl.static.sonic.net] has joined ##hplusroadmap | 15:39 | |
nmz787 | ad-droids | 15:40 |
nmz787 | adroids | 15:40 |
kanzure | Juul: do you live with max ogden? | 15:40 |
nmz787 | no i didn't mistype | 15:40 |
@fenn | nmz787: i'm going to send killer androids back in time if you do that | 15:40 |
Juul | kanzure, no. his place is awesome though | 15:40 |
nmz787 | fuck tacos | 15:40 |
kanzure | "his place" ? | 15:40 |
nmz787 | just hover around apartment buildings beaming ads in | 15:40 |
@fenn | i'm gonna deliver an atomic taco to your mouf | 15:41 |
nmz787 | then i'd vote for legislation for micro ground to air missiles allowed for home use | 15:41 |
nmz787 | screw halo | 15:42 |
@fenn | i, uh, i'd vote for that too | 15:42 |
Juul | he has the first floor on the corner of an old warehouse-type building with lotsa big windows. one big open space with a few smaller rooms in the back. | 15:42 |
kanzure | in sf? | 15:42 |
nmz787 | oakland | 15:42 |
kanzure | hm. for whatever reason i got the wires crossed and thought he was hanging out at your place. | 15:43 |
Juul | nah, he hangs out at sudo room | 15:43 |
brownies | is that the hackerspa e? | 15:43 |
Juul | one of them yeah | 15:44 |
-!- superkuh [~superkuh@unaffiliated/superkuh] has joined ##hplusroadmap | 15:44 | |
-!- augur [~augur@129-2-129-35.wireless.umd.edu] has joined ##hplusroadmap | 15:47 | |
-!- augur [~augur@129-2-129-35.wireless.umd.edu] has quit [Remote host closed the connection] | 15:47 | |
nmz787 | is he the eastbay biohackerspace? | 15:48 |
@fenn | typical nitromethane heli flight time is 7-15 minutes, "gassers" (?) up to 30 minutes, and there exist rc helis with 2 hours flight time but apparently it's a specialized thing | 15:49 |
brownies | why not just put a solar panel on top of the 'copter? | 15:50 |
@fenn | uff. this 25 pound payload helicopter sells for the low low price of $10300 | 15:50 |
brownies | or is that the dumbest idea in the entire world? | 15:50 |
brownies | fenn: it's an RC 'copter that can carry 25 pounds? that seems pricey. that's only like 10 large burritos. | 15:51 |
brownies | fenn: link, though? | 15:51 |
kanzure | east bay diybio is not at sudo room | 15:51 |
@fenn | brownies: solar power actually works okay for glider planes | 15:51 |
@fenn | brownies: http://www.bergenrc.com/IndTurbine.php | 15:51 |
kanzure | solar panel is not going to work for something that guzzles as much energy as a quadrotor | 15:51 |
brownies | it's 5 feet long. that's... nontrivial. | 15:52 |
-!- safitan is now known as mutagen | 15:52 | |
brownies | "lead time is 3-4 weeks" ... so they build it on-demand. | 15:52 |
brownies | i wonder who buys such things? | 15:52 |
@fenn | hollywood film makers | 15:53 |
brownies | not a very large market. | 15:53 |
brownies | if it was 10x cheaper, i wonder who else could buy it? | 15:53 |
ThomasEgi | police, photographers... | 15:53 |
ThomasEgi | i know a couple of people who live off selling airial pictures of stuff like hotels, events, and locations | 15:54 |
@fenn | they have a lot of other helicopters, i just liked the red MSR fuel bottles | 15:55 |
kanzure | "but we are all aware of the fact that Github cannot handle the complexity of hardware projects. " | 15:55 |
brownies | ThomasEgi: are they in the US? | 15:55 |
ThomasEgi | no. germany | 15:55 |
brownies | ThomasEgi: how regulated is commercial UAV usage in germany? | 15:55 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 264 seconds] | 15:55 | |
kanzure | ughhh | 15:55 |
kanzure | On Tue, Mar 26, 2013 at 5:17 PM, Matt Maier <blueback09@gmail.com> wrote: | 15:55 |
kanzure | > I can't build software, but my personal expertise does cover things like | 15:55 |
@fenn | ThomasEgi: but you don't need 25 pound payload for just a camera | 15:55 |
kanzure | > building consensus around technical subjects. So that's what I'll try to do. | 15:55 |
kanzure | > That requires collaborating with people other than myself; preferably as | 15:55 |
kanzure | > many stakeholders as possible. Thus, a committee with outreach access to as | 15:55 |
kanzure | > much of the relevant community as possible. | 15:55 |
kanzure | "stakeholders" | 15:56 |
kanzure | software is the only reasonable way to solve these problems | 15:56 |
kanzure | you can't just wish open source hardware into existence | 15:56 |
brownies | kanzure: "I can't build software, so I'm just going to talk about building software instead." | 15:56 |
kanzure | hahah | 15:56 |
kanzure | brownies: thank you for making my day better. | 15:56 |
@fenn | in summary, "i won't be of any help at all, but i'll get in the way and slow things down as much as possible. i'm actually a plant by the anti-open hardware consortium" | 15:56 |
brownies | hahaha | 15:56 |
kanzure | i love you guys | 15:57 |
brownies | seriously, who sees an effort starting up and is like "I KNOW! i'll ASSEMBLE A GIANT COMMITTEE!" | 15:57 |
kanzure | dude you should see this email | 15:57 |
@fenn | i'm surprised there aren't more autopilot/assisted pilot coaxial helicopters, since they're much harder to learn how to fly than a quadrotor | 15:59 |
@fenn | maybe coaxial is th wrong word | 15:59 |
kanzure | https://groups.google.com/group/openmanufacturing/msg/bf3342e661f6c6a2 | 15:59 |
brownies | i think coax is correct? | 15:59 |
@fenn | anyway, conventional single-rotor helicopter | 16:00 |
kanzure | http://www.evilmadscientist.com/2011/improving-open-source-hardware-visual-diffs/ | 16:00 |
@fenn | github does visual diffs btw | 16:01 |
kanzure | that's what that link is talking about. | 16:01 |
@fenn | well, image diffs | 16:01 |
kanzure | it's true that it's a pain in the ass to do visual diffs of cad files on github, but who cares. make a better git repository viewer if you care that much. | 16:02 |
kanzure | someone did do that with cubegithero or whatever, but his solution is proprietary and annoying | 16:02 |
@fenn | i automatically disregard anything-hero | 16:02 |
brownies | i honestly don't see anything about git that constraints it to software-only applications | 16:02 |
brownies | i use it all the time for documents, marketing materials, and all kinds of other stuff | 16:02 |
kanzure | well, people want things like auto merging | 16:03 |
@fenn | and i want a magical flying pony cat | 16:03 |
kanzure | except auto-merging 3d point clouds and curved manifolds => disaster | 16:03 |
brownies | right | 16:03 |
@fenn | there's a reason it's called "git" (aside from being easy to type) | 16:03 |
@fenn | and the reason is that merging doesn't work | 16:04 |
@fenn | in any domain | 16:04 |
brownies | merging branches in works fine if only 1 person has fiddled with every given file | 16:04 |
kanzure | i think it would be a huge improvement just to get people to put their fucking files in a shared repository with some sort of naming convention ("pulley.git", not "make-magazine-super-awesome-mechanical-motion-thing-for-your-thingiprinter.git") | 16:04 |
-!- Juul [~Juul@50-0-83-116.dsl.static.sonic.net] has quit [Ping timeout: 264 seconds] | 16:04 | |
brownies | and, as a corollary, it tends to fall apart hilariously if that is not the case. | 16:04 |
-!- cpopell [~cpopell@pool-96-231-37-73.washdc.fios.verizon.net] has joined ##hplusroadmap | 16:04 | |
kanzure | i think it is a mistake for these people to assume that they can just "let the software work itself out" | 16:05 |
brownies | kanzure: yes. generally, some standards and conventions and such? | 16:05 |
kanzure | no, putting files into a folder is pre-standards | 16:05 |
kanzure | it's just getting people to actually do that step itself -_- | 16:05 |
brownies | naming things intelligently usually requires conventions though =/ | 16:05 |
kanzure | ok fine, i'll settle with "here are all of the required files" | 16:05 |
kanzure | and "we promise we're not hiding shit and making your job annoying" | 16:05 |
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving] | 16:06 | |
@fenn | how is it that scientific papers are all more or less the same format? | 16:06 |
kanzure | same publishers | 16:06 |
@fenn | i mean there was never any committee that sat down and published a standard | 16:06 |
kanzure | they all have the same latex templates | 16:06 |
@fenn | but aside from that, there's a general consensus as to how the content should be written | 16:07 |
-!- klafka [~klafka@204-16-157-18-static.ipnetworksinc.net] has quit [Remote host closed the connection] | 16:07 | |
@fenn | title, abstract, authors, background, methods, results, pontification, blather, references | 16:07 |
kanzure | sometimes authors appear at the end with pictures | 16:07 |
kanzure | and sometimes the references appear on each page, instead of at the end | 16:07 |
@fenn | hey what if we just omit the whole names part and just share pictures of the lab group | 16:08 |
@fenn | you can really get to know them, after all a picture's worth a thousand words | 16:08 |
kanzure | feeling alright? | 16:08 |
@fenn | i am more than a name! | 16:08 |
kanzure | aname is going to be hurt that you said that | 16:08 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 16:09 | |
@fenn | anyway it's a rare paper without a title, abstract, or bibliography | 16:09 |
kanzure | maybe next to my picture in my next paper i will write "He has coauthored over 200 emails." | 16:09 |
-!- Juul [~Juul@50-0-83-116.dsl.static.sonic.net] has joined ##hplusroadmap | 16:12 | |
brownies | "published over a thousand puns in major discussion channels" | 16:12 |
kanzure | *cat puns | 16:12 |
brownies | good catch. | 16:13 |
kanzure | or your small "bibliography" can be the type of bullshit that passes for profiles on VC sites | 16:13 |
kanzure | "He spends his weekends biking in the hills contemplating the value his iphone app brings to monks." | 16:13 |
@fenn | your bioblography | 16:15 |
kanzure | oh. biography. bibliography. whatever. | 16:16 |
abetusk | can someone explain what the open hardware meetings are about? What are they hashing out exactly? | 16:17 |
kanzure | they claim they want to be hashing out open source hardware packaging formats, for putting things into reusable containers that you can use in other projects. | 16:18 |
kanzure | but they don't seem to know anything about any of the previously proposed solutions | 16:18 |
kanzure | or most of the existing cad formats (although they seem to be vaguely aware of some electronics software tools??) | 16:18 |
brownies | how about JSON? that solves every problem, right? | 16:19 |
-!- Juul [~Juul@50-0-83-116.dsl.static.sonic.net] has quit [Ping timeout: 246 seconds] | 16:19 | |
kanzure | well, we chose yaml a few years ago and fenn wrote a specification | 16:19 |
kanzure | and we wrote a bunch of python first | 16:20 |
kanzure | and then someone reviewed it and decided to make his own version, but then abandoned that | 16:20 |
kanzure | the problem is that if i tell this matt guy that this is fundamentally a software problem, he wont believe me because he will think i have a vested interest in making people think it's about "this thing that i just happen to know" | 16:21 |
abetusk | but does software even have this? I'm still confused as to what that means. What is a reusable component? A widget you can plug into KiCAD to act as a voltage regulator? An adapter to put onto your widgets case? | 16:21 |
kanzure | abetusk: do you use a package manager on your computer? | 16:21 |
abetusk | Yes, git | 16:21 |
kanzure | git isn't /exactly/ a package manager | 16:22 |
* fenn smells the pungent aroma of slack | 16:22 | |
kanzure | do you run linux? | 16:22 |
abetusk | ah, as in dpkg | 16:22 |
abetusk | I'm running ubuntu | 16:22 |
kanzure | i think part of the problem is that apt-get sounds impossible/magical to just about everyone | 16:22 |
@fenn | because debian is in fact impossible/magical but somehow it exists, and nobody knows why or how | 16:23 |
* kanzure pokes debian | 16:23 | |
brownies | what's so magical about apt-get? | 16:23 |
brownies | there are packages; it gets them. | 16:23 |
kanzure | most people don't know what a package is | 16:24 |
brownies | ah, well, then, i suppose that would make it more mysterious. | 16:24 |
kanzure | and even if they do have a general concept of "installing things", that doesn't completely tell you about the ecosystem of package maintainers | 16:24 |
@fenn | yeah if you install an "app" it just downloads a tarball for each app | 16:24 |
@fenn | there's no such thing as "app dependency" | 16:25 |
kanzure | and then you have to explain the scale/scope of this | 16:25 |
@fenn | well, there is, but it's awful and never works | 16:25 |
kanzure | CPAN is fucking huge | 16:25 |
brownies | i've never had any big problems with aptitude | 16:25 |
abetusk | ok, so 'dpkg for hardware'. I still don't know what that looks like | 16:25 |
kanzure | cpan is 114,000 packages | 16:25 |
kanzure | abetusk: http://gnusha.org/skdb/ welcome to the channel | 16:25 |
brownies | shit does tend to be outdated, but considering the magnitude of the problem, i think they've done quite well. and shit has never broken or conflicted, which is really something. | 16:26 |
@fenn | kanzure: btw the packages directory has bit-rotted and i'm not sure what to do about it | 16:26 |
kanzure | they are in git repos on http://diyhpl.us/cgit | 16:26 |
kanzure | it's not worth it, i would prioritize getting someone to review the system and offer some better ideas at this point | 16:26 |
kanzure | appamatto: wow wtf. always always always use a contract. what the hell man. | 16:27 |
@fenn | how does versioning work in cpan? does each package specify an exact version it depends on? | 16:28 |
brownies | can we back up a bit and add some context? | 16:29 |
-!- yorick [~yorick@oftn/member/yorick] has quit [Remote host closed the connection] | 16:29 | |
brownies | what is the goal of a "packaging system for hardware schematics"? what are the use cases? | 16:29 |
kanzure | use case is sudo apt-get install nodecopter and out pops a nodecopter | 16:30 |
brownies | from... where? | 16:30 |
@fenn | from the 16 other manufacturing robots that also got installed as dependencies | 16:30 |
kanzure | materials and parts in your inventory, online ordering when necessary or wherever you damn well please, or from using machines in your inventory to make other machines | 16:30 |
brownies | "install" does not seem semantically valid, since in theory you'd want to magically build a second nodecopter later | 16:30 |
kanzure | correct | 16:30 |
brownies | so then let's use a better word... build? clone? | 16:31 |
kanzure | build doesn't work because there's software building too | 16:31 |
brownies | this feels very OOP-like tbh | 16:31 |
kanzure | anyway, before you can get to that point, you need hardware to exist in consumable packages | 16:31 |
brownies | you have a package repository, those are your classes | 16:31 |
brownies | then you grab a class and "instantiate" a nodecopter | 16:31 |
kanzure | http://thingiverse.com/ is just a bunch of stl files with no metadata | 16:32 |
brownies | you can instantiate other nodecopters later, and fiddle with each one without affecting the others, if you so like | 16:32 |
@fenn | except in this case the objects aren't metaphors, they're objects | 16:32 |
brownies | yes, sure. | 16:32 |
brownies | just makes the vocabulary even more applicable | 16:32 |
kanzure | brownies: http://gnusha.org/skdb/package_spec.html | 16:32 |
brownies | especially when you consider that we have a robust vocabulary for all sorts of related things like dependencies, parents/children, mix-ins... | 16:32 |
appamatto | kanzure, heh, why? | 16:33 |
@fenn | i think the idea of a 'class' first was applied to CAD stuff, like a square is part of the class of rectangles, and you can build houses and so forth | 16:33 |
appamatto | It seems like contracts increase the likelihood of someone being sued | 16:33 |
kanzure | appamatto: because that's the minimum level of respect for business that you absolutely must demonstrate to your customers. | 16:33 |
appamatto | kanzure, I'm talking about people I hire | 16:34 |
appamatto | Not my customers | 16:34 |
@fenn | why would you hire someone without a contract | 16:34 |
brownies | kanzure: (still reading your link) one of the problems here is that it is all well and good to have a pile of source files like you would for software, but you need more for hardware | 16:34 |
@fenn | yes you also need resources | 16:35 |
brownies | kanzure: specifically, software is "self-assembling" ... each source file includes/requires other stuff, there's an entry point, etc. but if you just do that with hardware, you just get a pile of parts. | 16:35 |
kanzure | brownies: i am aware of how broken this is. i am not aware of how to fix this. | 16:35 |
brownies | so you need some sort of dependency graph or assembly instructions. | 16:35 |
kanzure | yes.. that's in there. | 16:35 |
kanzure | or ... well. we have years of logs of talking about that in here. | 16:35 |
@fenn | software is not self-assmebling, it requires a computer to do so | 16:35 |
brownies | is that the point of "ontology"? | 16:35 |
brownies | because having an ontology does not satsify that, i think | 16:35 |
kanzure | you don't "get a pile of parts" if you are actually using connected cnc machines and arms | 16:36 |
kanzure | also, you can just tell your human to do things | 16:36 |
kanzure | but, this is all down-the-road things that cannot be designed into the system from day one (believe me, i've checked) | 16:36 |
brownies | i see. ok. | 16:36 |
kanzure | the minimum start to this is packages. | 16:36 |
brownies | so for v1 it would suffice to just have a "pile of parts" (if you will) | 16:36 |
@fenn | BOM | 16:36 |
kanzure | holy hell a pile of parts would be way better than the current situation | 16:37 |
kanzure | a pile of models and a BOM would be grand. | 16:37 |
brownies | BOM? | 16:37 |
kanzure | also plus dependencies, somehow. | 16:37 |
@fenn | bill of materials | 16:37 |
kanzure | like, there must be dependencies from day one, otherwise this will go nowhere. | 16:37 |
@fenn | it's like a shopping list for resources needed to construct the project | 16:37 |
brownies | kanzure: the other thing is, if you contemplate the way RubyGems came about (ignoring for the moment that it sucks in lots of exciting ways) you had a core group of people who declared a spec and then produced a bunch of packages matching that spec | 16:37 |
kanzure | dependencies are how projects can leverage other projects and become better | 16:37 |
kanzure | yes, we produced skdb packages | 16:38 |
brownies | ah. and then what happened? | 16:38 |
kanzure | for screws, legos, and other items | 16:38 |
kanzure | nobody cared | 16:38 |
@fenn | eh, the packages weren't very useful | 16:38 |
@fenn | also it wasn't sufficiently automated to make it easy to use | 16:38 |
kanzure | there was even this really ridiculous 3d lego editor | 16:38 |
appamatto | fenn, mostly because I don't want to get my friends to sign contracts | 16:38 |
appamatto | Yeah, if the project were such that they could cause serious damages somehow then I'd want a contract | 16:39 |
@fenn | also there were unresolved theoretical problems in engineering and manufacturing that nobody had ever considered, and problems with closed data (alloy specs etc) | 16:39 |
kanzure | s/closed/proprietary | 16:40 |
kanzure | not closed in the math sense | 16:40 |
@fenn | also we had some problems defining the scope of the problem | 16:40 |
@fenn | like, should the system be able to simulate two lego bricks being mashed together? should it simulate functioning machines? | 16:41 |
brownies | why would it simulate such things? | 16:41 |
kanzure | testing compatibility | 16:41 |
brownies | just have packages for lego bricks, and let the builder sort it out | 16:41 |
@fenn | so you can determine if the project can be constructed with the available resources | 16:41 |
kanzure | not all building blocks are legos, but some are interface/port-compatible | 16:41 |
brownies | you can declare packages (with fixed versions) that are compatible with your package, but that is not ever going to be a complete list | 16:41 |
brownies | anyone can examine your package and make something compatbile, of course. | 16:42 |
brownies | if there is a central package repo, then it can report the union of such sets. | 16:42 |
kanzure | oh it's so easy is it | 16:42 |
brownies | i think guaranteeing/testing compatibility is a rather high bar for v1, though, don't you think? | 16:42 |
@fenn | yes | 16:42 |
kanzure | (the point was not having to manually currate all of the possible combinations) | 16:42 |
brownies | kanzure: yes, leave me with my whiskey and an hour, i'll solve the problems of the hardware world. ;) | 16:42 |
kanzure | go ahead | 16:42 |
kanzure | you may proceed | 16:42 |
brownies | anyway i was just asking why SKDB didn't take off, since it seems like you covered all these bases? | 16:43 |
kanzure | also there's a bit of a bootstrapping problem with packages-with-dependencies | 16:43 |
@fenn | there's no bootstrapping problem | 16:43 |
kanzure | a pile of dependencies and... what are you going to do with it? oh great another part you don't have. | 16:43 |
@fenn | you have a computer and a human capable of reading | 16:43 |
kanzure | and then you have to get to ordering from the right places | 16:43 |
kanzure | well anyway; openscad looks interesting because people seem to sometimes use previous parts from other scad scripts, which is exactly like a dependency. | 16:44 |
kanzure | (just.. only a dependency for rendering-time or complete-the-final-shape-of-the-object-before-manufacturing-it time) | 16:44 |
brownies | ordering stuff also seems like very much a v2 feature | 16:45 |
kanzure | so you have a list of parts you don't have. how is that system useful? | 16:45 |
@fenn | brownies: part of it was that i lost interest just before people started talking about it (the first time around, on openmanufacturing) | 16:45 |
@fenn | motivation is such a fickle thing | 16:46 |
kanzure | brownies: https://github.com/kanzure/skdb (go easy on me. obviously a lot of these tools should be separate projects.) | 16:46 |
abetusk | so, can you walk me through how to use skdb to build a widget? Say I have KiCAD files for the electronics, maybe even gerber files, a BOM, 3D design files for the case. When I 'skdb-get widget', what happens? All those files are downloaded? | 16:46 |
kanzure | if you already have those files, why would you download them again? | 16:47 |
abetusk | There's meta information for each type of file to tell me what type of file it is, how to use it, where to get it fabbed? | 16:47 |
abetusk | Sorry, say I wish to create a 'widget' skdb package that I would then want a friend to install | 16:47 |
@fenn | assuming you have a circuit board etcher, pick and place robot, and automated soldering station, and all the parts loaded and ready to go, the finished product just pops out | 16:48 |
@fenn | when pieces of this chain are missing, things get interesting | 16:48 |
@fenn | unfortunately we aren't at that point yet, because nobody has written the software to handle the in-between steps | 16:49 |
@fenn | so instead of a robot arm that knows the layout of your workspace and machinery, there would be a dialog that pops up saying "okay now put the circuit board in machine #1" and you click okay | 16:50 |
abetusk | so you're talking about an integration between all available tools? So skdb now knows how to talk to LinuxCNC, OpenPNP, etc. etc? | 16:51 |
@fenn | oh, brownies a big reason nobody cares is that it's easier to just pay someone in china to mail it to you | 16:51 |
@fenn | abetusk: that's the idea | 16:53 |
@fenn | i got stuck on how to define a milling machine | 16:53 |
abetusk | ok, but this is a 'down the road' kind of integration, right? | 16:53 |
@fenn | sort of | 16:53 |
@fenn | even if you have all the cad files, it's a lot of work to figure out how to set up the machine to hold the raw material, what CAM settings to use | 16:54 |
@fenn | alleviating that burden is a large part of the goal for skdb | 16:54 |
abetusk | so what does skdb offer as of today? How much help does skdb provide to let me download a widget and create it? | 16:55 |
@fenn | it lets you download the files for the widget | 16:55 |
@fenn | that's all | 16:55 |
@fenn | and sub-widgets | 16:55 |
abetusk | What does it offer over downloading a zip file with all of the files? What does it offer of cloning a git repo with all of the files? | 16:55 |
kanzure | the package format is a git repo | 16:56 |
@fenn | it tells you what the files are | 16:56 |
kanzure | the whole point is to use standard things | 16:56 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 248 seconds] | 16:56 | |
@fenn | do you know how many file formats use .brd as the suffix | 16:56 |
abetusk | yep. And are you running grbl or LinuxCNC? because one has features the other doesn't support | 16:57 |
abetusk | etc. etc. | 16:57 |
brownies | that's crazy | 16:57 |
brownies | i mean, it's a nice vision, but starting with that in v1 is thoroughly insane | 16:57 |
-!- mako [~mako@wireless-nat-7.auckland.ac.nz] has quit [Ping timeout: 245 seconds] | 16:57 | |
kanzure | i think packaging is a good start. | 16:58 |
brownies | it would suffice to literally have a list of parts and a pile of schematics, with some declared dependencies and compatabilities | 16:58 |
@fenn | brownies: but 99% of open source hardware projects don't even have that, and the specification was supposed to force things to comply | 16:58 |
brownies | they don't have that? well then a spec isn't going to force them to get those things | 16:59 |
@fenn | this is where yaml turned out to not be such a good choice, because there's not very good validation tools | 16:59 |
brownies | i thought they would at least have that =/ but anyway it would suffice for v1 to simply have a way to organize the sorts of things that most people have | 17:00 |
kanzure | yeah the other issue is that to get good data you basically have to redo a lot of open source hardware projects on your own | 17:00 |
kanzure | (with possibly the exception of reprap, maybe.) | 17:00 |
@fenn | there are some well documented projects, but even then most of it is implicit information | 17:01 |
@fenn | "of course you need a soldering iron and gcc, duh!" | 17:01 |
@fenn | which is easy if you understand what's going on, but not easy if you don't know what gcc is or what it's for | 17:02 |
kanzure | and then there's those evil BOMs that only list certain things but not all things | 17:02 |
abetusk | most hardware projects involve human intervention at some point | 17:02 |
kanzure | yeah but you shouldn't have to become an expert just to assemble some parts | 17:02 |
kanzure | there's no way you're going to build your space colony at that rate | 17:02 |
abetusk | It sounds like most instructables pages have more documentation than most hardware projects | 17:03 |
kanzure | instructables are crap :( | 17:03 |
@fenn | i haven't even been able to get instructables pages to load for 2 years now | 17:03 |
kanzure | it's just huge blocks of text with some pics | 17:03 |
abetusk | some of them, but they tend to have step by step instructions and BOMs | 17:03 |
kanzure | and then you have to figure out NPL parsing of their documentation to figure out some representation of whatever hardware they are talking about... | 17:03 |
abetusk | If human intervention is key, then how can you get around it? | 17:04 |
abetusk | You either need human readable instructions for humans or automated machines that understand machine languages | 17:04 |
@fenn | nobody ever said human intervention is key | 17:04 |
kanzure | reading millions of pages of documentation is just not tractable | 17:04 |
abetusk | Then this is a completely moot discussion until we get the infrastructure for completely autonomous construction | 17:04 |
kanzure | (for a person) | 17:05 |
abetusk | I understand | 17:05 |
abetusk | and I agree | 17:05 |
kanzure | no, you can have humans put things together | 17:05 |
kanzure | and that's what they are doing now | 17:05 |
abetusk | By providing them instructions? | 17:05 |
kanzure | how is it moot to make that process even slightly less hilariously dmb | 17:05 |
@fenn | the trick here is to treat humans as just a class of agent | 17:05 |
kanzure | *dumb | 17:05 |
abetusk | trying to go all the way and try and get a full fledged system that is as simple as 'apt-get widget' and have it constructed right then and there is shooting way too high now | 17:06 |
@fenn | often when you say "and now the human puts the bread into the toaster" your audience starts assuming magical powers of intelligence and consciousness on the part of the human | 17:06 |
kanzure | i said packaging :| | 17:06 |
-!- mako [~mako@wireless-nat-7.auckland.ac.nz] has joined ##hplusroadmap | 17:07 | |
brownies | kanzure: looking at your github repo, the most salient feature is that it's just too much damn stuff | 17:07 |
@fenn | meanwhile g-code can't even represent "object 1, location 1", much less "bread, toaster" | 17:07 |
-!- EnLilaSko [EnLilaSko@unaffiliated/enlilasko] has quit [Quit: - nbs-irc 2.39 - www.nbs-irc.net -] | 17:07 | |
kanzure | maybe it would be useful to have a sort of low quality hardware packaging format where all user bugs are sent directly to the upstream projects | 17:09 |
abetusk | Look, I'm sorry for being dense, but my initial question is "what are the osh people doing with their documentation jam". I don't understand what problem it is they're trying to solve that couldn't be done with github. Or is that the answer? Standardize hardware packages to github in some manner so that people have a BOM, a supplier and the necessary design files? | 17:09 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 17:09 | |
@fenn | abetusk: that's exactly what the skdb package spec says | 17:09 |
kanzure | abetusk: skdb packages are git repos that include those things (although not necessarily supplier info, that wasn't written down in the spec) | 17:09 |
kanzure | abetusk: the "ohw documentation jam" thing is just them being pigheaded about looking at previously existing solutions | 17:10 |
kanzure | and apparently they feel that the solution "can't" be software or something | 17:10 |
kanzure | see https://groups.google.com/group/openmanufacturing/msg/bf3342e661f6c6a2 | 17:10 |
abetusk | yeah, I'm on the list, I've been watching the thread | 17:11 |
kanzure | they just have a lot of attention because of their oreilly connections i think, but that doesn't mean they understand how to solve this problem... or that they are right about it being non-software.. | 17:11 |
@fenn | i have no idea what matt maier is trying to say there, "approach software problems as customers" | 17:12 |
-!- mako [~mako@wireless-nat-7.auckland.ac.nz] has quit [Quit: Leaving] | 17:12 | |
@fenn | like, we should be buying software? | 17:12 |
@fenn | we should be selling software? | 17:12 |
-!- mako [~mako@wireless-nat-7.auckland.ac.nz] has joined ##hplusroadmap | 17:12 | |
abetusk | ok, so this is what I was trying to get at with an skdb example. So if I were to use skdb to package my widget, I would have some standard files/directories to put in BOMs, design files and the like? When someone downloads it, they would know to go to the 'design' directory, say, to find the openscad files for the case that they can then 3d print? | 17:12 |
@fenn | -_- | 17:12 |
-!- cpopell [~cpopell@pool-96-231-37-73.washdc.fios.verizon.net] has quit [Ping timeout: 252 seconds] | 17:13 | |
@fenn | abetusk: yeah, more or less | 17:13 |
@fenn | instead of directories we have a metadata file | 17:13 |
abetusk | Is there an example tutorial on how to package a widget and how to get and interpret the downloaded widget? | 17:13 |
kanzure | actually yes | 17:14 |
kanzure | i wonder where that tutorial went | 17:14 |
@fenn | i don't remember where the tutorial was, but i think the spec itself is pretty self-explanatory: https://github.com/kanzure/skdb/blob/master/doc/package_spec.yaml | 17:15 |
@fenn | hmm | 17:16 |
kanzure | brownies: long list of "mvp" issues. | 17:17 |
abetusk | so is there a lego or bolt example skdb package I can get? | 17:19 |
kanzure | http://diyhpl.us/cgit some of these are sample packages | 17:20 |
kanzure | like cathal garvey's dremelfuge | 17:20 |
abetusk | thanks | 17:21 |
-!- eudoxia [~eudoxia@r186-54-239-26.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 17:22 | |
eudoxia | >13:36 < kanzure> tangiblebit and skdb both had considerations for different types of dependencies (like "run time" and "physical build time") but neither had any traversal algorithm (which is trivial, but it just wasn't written) | 17:23 |
eudoxia | kanzure, was this supposed to be in dep.py? | 17:23 |
@fenn | hrm. the template thing in lego/metadata.yaml is pretty confusing if it's the first thing you see | 17:23 |
kanzure | fenn: a lot of this should just be removed, or even just replaced with a separate tool/library that doesp recisely one thing very well | 17:29 |
kanzure | eudoxia: tangiblebit is in tangiblebit.git | 17:29 |
@fenn | yeah. abetusk this is the canonical example package metadata; i'd just ignore the "template" part: http://diyhpl.us/cgit/screw/tree/metadata.yaml | 17:29 |
eudoxia | oh i didn't know about tangiblebit | 17:30 |
eudoxia | oh wait that's that riseup group | 17:30 |
@fenn | where the hell did screw.step go | 17:30 |
@fenn | gah | 17:30 |
@fenn | kanzure: do you remember why the interface description ("template") is even in the metdata file? | 17:32 |
kanzure | wtf "The (open and/or free) tools available at the moment, including the ones on Github, are not up to the task of tracking all of the interrelated dimensions of a hardware project well enough to allow an "actively open" level of activity. When we remix the digital elements of a hardware project we are not messing around with the project itself; we are merely adjusting a description of the project." | 17:32 |
@fenn | something about not downloading the whole package in order to determine whether to download it or not | 17:32 |
kanzure | fenn: some yaml reason | 17:32 |
kanzure | fenn: or that | 17:32 |
kanzure | whatever. it's dumb. it's all dumb. | 17:32 |
kanzure | "By way of a supporting example, I'd like to direct your attention to the LifeTrac Fabrication instructions over at Open Source Ecology. I know those instructions are pretty good because I wrote them. But, for the same reason, I also know that they are dead." | 17:33 |
kanzure | duh because it's on a fucking wiki and we warned you against this | 17:33 |
eudoxia | fenn: well if you're building some package and you need a part that fits into an interface with, say, diameter x length y, you'd want to know the properties on the matching interface on that package | 17:33 |
@fenn | kanzure: he holds that view because they have no automation, they do everything by hand | 17:34 |
-!- eudoxia [~eudoxia@r186-54-239-26.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving] | 17:34 | |
kanzure | yeah, this guy just seems like an idiot | 17:34 |
@fenn | there is no "project" beyond instructions for a human | 17:34 |
kanzure | "To sum up: hardware projects are more complicated because more variables have to be accounted for if the instructions are to be correct. It is currently possible to capture that complexity and to store it on Github, but it is not possible to interact with it in a way that allows for the project to be "truly" open. " | 17:34 |
kanzure | i bet he thinks doc tools are evil magic | 17:34 |
-!- eudoxia [~eudoxia@r186-54-239-26.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 17:34 | |
kanzure | "what! you generate your documentation from thoughtful comments and docstrings?" | 17:34 |
@fenn | eudoxia: the diameters and lengths are in a different file though | 17:35 |
kanzure | i don't even know where to begin delineating a list of reasons for why he is doing it wrong | 17:35 |
@fenn | ugh, has he ever tried to use doxygen | 17:35 |
@fenn | sadface | 17:36 |
kanzure | doxygen has its warts but.. uh. better than not doxygen. | 17:36 |
kanzure | actually, this would be a good moment for you to reply to them and re-iterate your OSE complaints | 17:36 |
@fenn | but it's not documentation | 17:36 |
kanzure | especially since it will be heard by some people who claim they are interested in helping | 17:36 |
kanzure | in fact, you can probably just reuse a previous OSE hatemail letter | 17:36 |
kanzure | probably word for word | 17:36 |
kanzure | just send it every 4 years you'll be fine | 17:36 |
@fenn | mrr | 17:37 |
@fenn | [Folder "manufacturing" opened with 7,305 messages - 4,514 new] | 17:37 |
kanzure | well 1000 of those are probably paul fernhout | 17:38 |
kanzure | also you should consider his email in the context of his last one | 17:39 |
kanzure | https://groups.google.com/group/openmanufacturing/msg/bf3342e661f6c6a2 | 17:40 |
kanzure | https://groups.google.com/group/openmanufacturing/browse_thread/thread/8d82fda1bb088c5a | 17:40 |
@fenn | okay so why not translate .openproj to yaml or whatever | 17:45 |
kanzure | i haven't seen the openproj format | 17:45 |
@fenn | it's a binary database format | 17:45 |
kanzure | why | 17:45 |
@fenn | personally i don't give a shit about openproj, but it seems like a pretty wimpy excuse | 17:45 |
kanzure | http://sourceforge.net/projects/openproj/ | 17:46 |
kanzure | oh this looks disgusting. a project management gui? | 17:46 |
@fenn | would he prefer that its native format were arbitrarily-ordered XML? how about some other text format? | 17:46 |
kanzure | what information is in here? TODO? | 17:46 |
@fenn | i mean diffs are nice and all, but usually you look at the commit message first | 17:47 |
kanzure | why the fuck would you make this "No one can read or edit the actual OpenProj file without the software." | 17:47 |
kanzure | this guy is insane | 17:47 |
@fenn | he didn't make openproj, and "insane" is a bit of an overreaction | 17:47 |
kanzure | if your goal is to make gantt charts it seems like an ok tool | 17:48 |
@fenn | the analogy is with binary cad files and needing a cad program to read/write them | 17:48 |
@fenn | but since he's not an engineer he only knows gantt charts i guess | 17:48 |
@fenn | anyway i don't think a project hosting site needs to be able to read and write every file format imaginable | 17:50 |
@fenn | but it's not too much to ask someone to download the required software to edit the source file if they want to contribute to the project | 17:51 |
@fenn | would you rather they be able to commit code changes without having compiled the code? no way, and cad is no different | 17:52 |
kanzure | fine; but why is the gui mandatory | 17:52 |
@fenn | because whoever wrote openproj hates you | 17:53 |
@fenn | i'm sure there's some excellent essay by the project gutenberg guy about why ascii text will never die | 17:55 |
kanzure | you mean the textfiles.com guy? | 17:55 |
@fenn | no, that seems to be about the bbs scene in the early 90's | 17:56 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 248 seconds] | 17:56 | |
@fenn | http://hart.pglaf.org/ | 17:58 |
@fenn | he has a specific interpretation of what an "ebook" is that aligns with my ideas about sharing hardware designs | 17:59 |
@fenn | maybe it's just OCD, i dunno | 17:59 |
kanzure | gah this stuff just hurts to read http://lists.oshwa.org/pipermail/discuss/2013-March/000470.html | 18:00 |
@fenn | today i've been reading about extinct alternative phoneticization systems for japanese and chinese | 18:00 |
@fenn | it took a government mandate for everyone to stop inventing their own personal lettering system and use the same letters | 18:01 |
kanzure | neat | 18:02 |
@fenn | a visiting german thought there were 3 alphabets, but actually there were 2 major contenders and thousands of others | 18:02 |
kanzure | off with their heads! | 18:03 |
kanzure | what did the government mandate happens if someone uses their own lettering system? | 18:03 |
@fenn | they never really explained that | 18:05 |
kanzure | "if you don't eat your green beans, ISO will deploy drones to shoot you in the butt" | 18:06 |
@fenn | midori doesn't do so well with huge pages of japanese text.. oh the irony | 18:07 |
@fenn | apparently they just ordered all the schools to teach it and everyone else died out :( | 18:09 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 18:10 | |
kanzure | excellent strategy | 18:10 |
ThomasEgi | i wish they'd finally issue such an order for the imperial unit crap | 18:18 |
ThomasEgi | ordering metric. and waiting for everyone else to die out | 18:18 |
@fenn | "you only get the immortality treatment if you swear to use metric" | 18:19 |
eudoxia | "sorry, i can't operate on that tumor unless you tell me its mass in grams" | 18:19 |
ThomasEgi | i'd be perfectly fine with that | 18:20 |
-!- qu-bit [~shroedngr@unaffiliated/barriers] has quit [Read error: Connection reset by peer] | 18:38 | |
-!- qu-bit [~shroedngr@unaffiliated/barriers] has joined ##hplusroadmap | 18:41 | |
-!- eudoxia [~eudoxia@r186-54-239-26.dialup.adsl.anteldata.net.uy] has quit [Ping timeout: 246 seconds] | 18:47 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 248 seconds] | 18:57 | |
-!- eudoxia [~eudoxia@r186-52-163-212.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 19:00 | |
-!- mako [~mako@wireless-nat-7.auckland.ac.nz] has quit [Ping timeout: 257 seconds] | 19:00 | |
kanzure | On Tue, Mar 26, 2013 at 8:56 PM, malcolm stanley <a.malcolm.stanley@gmail.com> wrote: | 19:00 |
kanzure | > There are millions of people like me who 'could' become competent and | 19:00 |
kanzure | > valuable members of this movement, but only if we are careful to consider | 19:00 |
kanzure | > their needs and assume, first and foremost, that they know nothing, and | 19:00 |
kanzure | > bring a willingness to explain and educate as part of our core activity. | 19:00 |
kanzure | but... we were talking about making a standard for open source hardware packaging.. and somehow i am supposed to feel bad that you don't know what you're talking about? | 19:01 |
eudoxia | why can't people just sit down and code | 19:01 |
brownies | eh? where are the MVP issues? | 19:03 |
brownies | kanzure: first we must educate the world and learn to live in peace with one another | 19:03 |
brownies | only then we can hope to create a reasonable package format for hardware | 19:04 |
eudoxia | hahahhaha | 19:04 |
kanzure | brownies: MVP issues with skdb? you were recognizing the issues earlier, except in more words. | 19:05 |
kanzure | it occurs to me that maybe he sent that email just to genuinely ask the question, and not because he felt like he is going to define some standards or whatever. | 19:08 |
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Remote host closed the connection] | 19:10 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 19:11 | |
-!- eudoxia [~eudoxia@r186-52-163-212.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving] | 19:13 | |
kanzure | fenn: are you going to write that counter ose email? | 19:31 |
-!- yashgaroth [~ffffff@cpe-66-27-118-94.san.res.rr.com] has joined ##hplusroadmap | 19:31 | |
-!- appamatto [~appamatto@copyfree/advocate/appamatto] has quit [Remote host closed the connection] | 19:32 | |
kanzure | > According to our current community standards, the LifeTrac Fabrication instructions | 19:44 |
kanzure | > would generally not be regarded as an "open" part of the tractor's open source hardware | 19:44 |
kanzure | > release because you have chosen to not publish the original design files. So as a "supporting | 19:44 |
kanzure | > example" to argue that releasing the design files alone isn't enough to be considered | 19:44 |
kanzure | > "truly open," it doesn't provide much support. | 19:44 |
kanzure | haha i like this guy | 19:44 |
kanzure | "Case in point: the actual hardware designs for the LifeTrac *are* on github, even though a pricey chunk of software is required to modify them." | 19:45 |
kanzure | where? | 19:45 |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 19:55 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Ping timeout: 252 seconds] | 19:58 | |
-!- mako [~mako@wireless-nat-7.auckland.ac.nz] has joined ##hplusroadmap | 19:59 | |
kanzure | hah what https://github.com/Amakaruk/OSELifeTrac | 20:06 |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has joined ##hplusroadmap | 20:11 | |
-!- lupfantomo [~lupfantom@24-159-24-229.dhcp.jcsn.tn.charter.com] has quit [Quit: Leaving] | 20:45 | |
-!- augur [~augur@208.58.5.87] has joined ##hplusroadmap | 21:06 | |
-!- augur [~augur@208.58.5.87] has quit [Remote host closed the connection] | 21:07 | |
-!- augur [~augur@208.58.5.87] has joined ##hplusroadmap | 21:07 | |
kanzure | heh nice to see bram cohen laying it down https://news.ycombinator.com/item?id=5443618 | 21:10 |
kanzure | "Can anyone with a more recent exposure to p2p stuff comment on whether this is indeed an innovation or is it just a PR spin on a patent application?" | 21:10 |
kanzure | "This patent does not claim anything novel. This idea of "clubs of peers" or groups has been around for a while. This well known paper from Torino, Italy, written a few years ago: "overlays are maintained by peers organized in clusters that represent sets of collaborating peers"," | 21:11 |
kanzure | and bram says.. "The traditional overlay approach involves a bunch of full trees, and uses multiple trees as a way around dealing with leaf nodes being unutilized. My approach uses multiple groupings, which do not overlay, screams within them, and does something completely different for the last hop. They're completely different architectures. I have trouble taking seriously any paper which says that it makes heavy use of multiple description ... | 21:11 |
kanzure | ... coding. If you have congestion control, skips should be an extreme and bad event." | 21:11 |
-!- augur [~augur@208.58.5.87] has quit [Read error: Connection reset by peer] | 21:21 | |
-!- augur [~augur@208.58.5.87] has joined ##hplusroadmap | 21:21 | |
-!- OldCoder_ [~OldCoder_@209.237.22.146] has quit [Remote host closed the connection] | 21:34 | |
-!- lilnicky [46c20834@gateway/web/freenode/ip.70.194.8.52] has joined ##hplusroadmap | 21:45 | |
-!- lilnicky [46c20834@gateway/web/freenode/ip.70.194.8.52] has quit [Client Quit] | 21:46 | |
kanzure | wow... wtf | 21:51 |
kanzure | On Tue, Mar 26, 2013 at 11:50 PM, Matt Maier <blueback09@gmail.com> wrote: | 21:51 |
kanzure | > Then we disagree about another point of "openness." Unlike software, certain | 21:51 |
kanzure | > electronics, or 3D printable parts, the LifeTrac was not completely designed | 21:51 |
kanzure | > in a computer. It wasn't even completely planned out. The design evolved | 21:51 |
kanzure | > after metal touched metal and those instructions merely document a stable | 21:51 |
kanzure | > phase. They were already out of date when they were finalized, but they do | 21:51 |
kanzure | > describe a complete version of the machine. | 21:51 |
kanzure | this is getting weird | 21:52 |
kanzure | On Tue, Mar 26, 2013 at 11:50 PM, Matt Maier <blueback09@gmail.com> wrote: | 21:52 |
kanzure | > As far a I'm concerned, that qualifies a "open" because that's all anyone | 21:52 |
kanzure | > needs. Unlike the places where open source is most popular, structures and | 21:52 |
kanzure | > mechanisms do not always depend on digital files. Blueprints and sketches | 21:52 |
kanzure | > and cardboard prototypes aren't any good to anyone else. What everyone else | 21:52 |
kanzure | > needs is an after-the-fact description of how the project actually works, | 21:52 |
kanzure | > not a before-the-fact description of how it might work. Even if the | 21:52 |
kanzure | > "original" files exist in any meaningful way they are only useful to satisfy | 21:52 |
kanzure | > historical curiosity. | 21:52 |
kanzure | "It wasn't even completely planned out." | 21:53 |
kanzure | and this, kids, is what we call technical debt. | 22:01 |
kanzure | http://blog.cloudflare.com/the-ddos-that-knocked-spamhaus-offline-and-ho | 22:17 |
kanzure | "The basic technique of a DNS reflection attack is to send a request for a large DNS zone file with the source IP address spoofed to be the intended victim to a large number of open DNS resolvers. The resolvers then respond to the request, sending the large DNS zone answer to the intended victim. The attackers' requests themselves are only a fraction of the size of the responses, meaning the attacker can effectively amplify their attack to ... | 22:17 |
kanzure | ... many times the size of the bandwidth resources they themselves control." | 22:17 |
kanzure | "We recorded over 30,000 unique DNS resolvers involved in the attack. This translates to each open DNS resolver sending an average of 2.5Mbps, which is small enough to fly under the radar of most DNS resolvers." | 22:18 |
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Read error: Connection timed out] | 22:25 | |
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap | 22:27 | |
kanzure | > far out of the computer that it requires welding you are beyond the ability | 22:37 |
kanzure | > of "source files" to be anything other than helpful. The records of a | 22:37 |
kanzure | > machine always have to be updated with all of the little things that you | 22:37 |
kanzure | > ACTUALLY did when you built it. If you were to give someone the original | 22:37 |
kanzure | > files, the ones that haven't been updated based on reality, they would not | 22:37 |
kanzure | uh.. why does welding nto get to be recorded? i think he is lying. | 22:37 |
-!- OldCoder_ [~OldCoder_@c-69-181-139-128.hsd1.ca.comcast.net] has joined ##hplusroadmap | 23:02 | |
-!- sbean [~sbean@67.51.1.15] has joined ##hplusroadmap | 23:03 | |
kanzure | sbean: hello | 23:03 |
sbean | kanzure: hello | 23:03 |
kanzure | sbean: what brings you these ways? | 23:06 |
sbean | wanted to check out paperbot | 23:06 |
sbean | reading the source to see how to use it :) | 23:10 |
sbean | http://www.sciencedirect.com/science/article/pii/S1477513107003312 | 23:11 |
kanzure | sbean: paperbot does not "try harder" unless you explicitly tell paperbot to fetch something. if it fails using the zotero translation-server, then it wont try again unless you asked it to (because we don't want error messages flooding the channel on every link). | 23:13 |
kanzure | paperbot: http://www.sciencedirect.com/science/article/pii/S1477513107003312 | 23:13 |
paperbot | no translator available, raw dump: http://diyhpl.us/~bryan/papers2/paperbot/Outcome%20of%20surgical%20management%20of%20concealed%20penis.txt | 23:14 |
sbean | got it | 23:14 |
kanzure | unfortunately sciencedirect seems to be broken at the moment | 23:14 |
kanzure | paperbot needs to be rewritten, if you have any python chops then it would be awesome if you could just delete the current bullshit and write a plugin system. | 23:14 |
kanzure | paperbot: http://www.sciencedirect.com/science?_ob=MiamiImageURL&_cid=273056&_user=1694017&_pii=S1477513107003312&_check=y&_origin=article&_zone=toolbar&_coverDate=2007--31&view=c&originContentFamily=serial&wchp=dGLzVlk-zSkzV&pid=1-s2.0-S1477513107003312-main.pdf&_valck=1&md5=59c8e027d4280a8948a6fd3be464a3e9&ie=/sdarticle.pdf | 23:14 |
paperbot | no translator available, raw dump: http://diyhpl.us/~bryan/papers2/paperbot/b49fabc978a97b632983dbb59c6c5850.txt | 23:14 |
kanzure | damn. no access. | 23:14 |
sbean | paperbot: http://ieeexplore.ieee.org/xpl/articleDetails.jsp?tp=&arnumber=4773330&contentType=Standards&queryText%3Dpassword | 23:16 |
paperbot | http://diyhpl.us/~bryan/papers2/paperbot/da724a6706fccb9e9614774e795be47c.txt | 23:16 |
kanzure | ieeexplore access is splotchy at best. | 23:16 |
sbean | whats a good site to try? | 23:17 |
kanzure | in the future i will have paperbot try different ezproxy logins to attempt to grab a paper instead of just giving up. | 23:17 |
kanzure | well paperbot never has 100% access to any closed publication because of how subscriptions tend to work. | 23:17 |
kanzure | nature tends to work alright. | 23:17 |
kanzure | paperbot: http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=5509165 | 23:18 |
paperbot | http://diyhpl.us/~bryan/papers2/paperbot/820efc6e1b24bacba0d9e978ecac95f8.pdf | 23:18 |
sbean | very cool :) | 23:18 |
kanzure | paperbot: http://www.nature.com/nprot/journal/v7/n6/abs/nprot.2012.048.html | 23:18 |
paperbot | HTTP 401 unauthorized http://www.nature.com/nprot/journal/v7/n6/pdf/nprot.2012.048.pdf | 23:18 |
kanzure | oops | 23:18 |
sbean | :) | 23:18 |
kanzure | paperbot: http://www.nature.com/nprot/journal/v7/n12/abs/nprot.2012.140.html | 23:18 |
paperbot | HTTP 401 unauthorized http://www.nature.com/nprot/journal/v7/n12/pdf/nprot.2012.140.pdf | 23:18 |
kanzure | paperbot: http://apl.aip.org/resource/1/applab/v100/i18/p181112_s1 | 23:19 |
paperbot | error: HTTP 500 http://diyhpl.us/~bryan/papers2/paperbot/Tesla%20coil%20discharges%20guided%20by%20femtosecond%20laser%20filaments%20in%20air.pdf | 23:19 |
kanzure | there we go. | 23:19 |
kanzure | are you in utah? | 23:20 |
sbean | yep\ | 23:20 |
-!- yashgaroth [~ffffff@cpe-66-27-118-94.san.res.rr.com] has quit [Quit: Leaving] | 23:22 | |
-!- ParahSail1n [~pwang@unaffiliated/parahsailin] has quit [Ping timeout: 256 seconds] | 23:22 | |
sbean | doing some coding for a torrent site that has journals, ran across paperbot | 23:22 |
-!- ParahSail1n [~pwang@unaffiliated/parahsailin] has joined ##hplusroadmap | 23:22 | |
kanzure | sbean: which torrent site? | 23:23 |
sbean | a private tracker | 23:23 |
kanzure | sbean: you might also be interested in http://groups.google.com/group/science-liberation-front | 23:23 |
kanzure | btw if you can get me access to someone who is willing to seed 10 TB of data, lemme know. | 23:23 |
sbean | thanks | 23:23 |
sbean | lol | 23:24 |
sbean | 10tb of journals? | 23:24 |
kanzure | don't ask me that! what's wrong with you. | 23:24 |
sbean | sorry :) | 23:24 |
sbean | pdfs? | 23:25 |
-!- cerillio [~androirc@217.72.221.154] has joined ##hplusroadmap | 23:26 | |
sbean | I'll just pm you | 23:26 |
cerillio | Hey bryan | 23:26 |
kanzure | hello cerillio | 23:27 |
cerillio | Aka Nils | 23:27 |
kanzure | sbean is being introduced to paperbot and http://groups.google.com/group/science-liberation-front | 23:27 |
cerillio | But that was taken | 23:27 |
sbean | kanzure pm | 23:29 |
kanzure | ok. | 23:37 |
sbean | this tracker is run by staff from some of the biggest private trackers out there | 23:37 |
kanzure | On Wed, Mar 27, 2013 at 12:52 AM, Matt Maier <blueback09@gmail.com> wrote: | 23:37 |
kanzure | > As far as I know OSE didn't design the LifeTrac in CAD; they just used CAD | 23:37 |
kanzure | > afterwards to create a digital blueprint. | 23:37 |
kanzure | tsk tsk | 23:38 |
kanzure | sbean: neat, glad to hear that | 23:38 |
sbean | started as magazines/newspapers but users want more journals | 23:38 |
kanzure | are you paid by the tracker? | 23:38 |
sbean | no | 23:39 |
kanzure | volunteer? | 23:39 |
sbean | everyone is | 23:39 |
sbean | the reward is even being able to access the site | 23:40 |
-!- mako [~mako@wireless-nat-7.auckland.ac.nz] has quit [Ping timeout: 264 seconds] | 23:40 | |
sbean | just like what.cd, passthepopcorn, etc | 23:40 |
kanzure | haha whatever | 23:40 |
sbean | no, really | 23:40 |
kanzure | didn't what.cd ask for blood samples from your maternal grandmother or something | 23:40 |
sbean | yep | 23:41 |
sbean | so people value their access :) | 23:41 |
kanzure | why should i be tempted? | 23:41 |
sbean | I had to do a 4 hr interview to get on | 23:41 |
kanzure | what do you have that i do not? | 23:41 |
sbean | well right now, not much as far as journals go | 23:41 |
sbean | what is zotero? | 23:42 |
kanzure | it was originally a firefox plugin for extracting bibliographies and pdfs from academic publisher's sites. | 23:42 |
kanzure | paperbot uses zotero in headless mode to extract metadata from a publisher's site, including a link to the pdf (although not always; it's hit-or-miss, so there's some other shitty paperbot code for the other situations) | 23:43 |
kanzure | http://zotero.org/ | 23:43 |
kanzure | https://github.com/zotero/translators | 23:43 |
kanzure | https://github.com/zotero/translation-server | 23:43 |
sbean | interesting | 23:44 |
kanzure | https://github.com/kanzure/paperbot | 23:44 |
kanzure | also i guess you should also look at https://github.com/kanzure/pdfparanoia if you are interested in maintaining private tracker anonymity | 23:44 |
sbean | seen that | 23:46 |
-!- cerillio [~androirc@217.72.221.154] has quit [Ping timeout: 264 seconds] | 23:48 | |
-!- cerillio [~androirc@tmo-107-187.customers.d1-online.com] has joined ##hplusroadmap | 23:48 | |
cerillio | Yawn | 23:52 |
sbean | thanks for the info, I'll come back when I have 10tb free :) | 23:52 |
cerillio | I really need to hack my sleep cycle | 23:52 |
kanzure | get rid of the kids, that's step 1 | 23:53 |
-!- sbean [~sbean@67.51.1.15] has left ##hplusroadmap [] | 23:53 | |
kanzure | who was that guy. | 23:53 |
cerillio | Lol | 23:54 |
cerillio | Never, i wait until they grow up. That's when most of the other folks get theirs and I'll be only 45 then | 23:55 |
--- Log closed Wed Mar 27 00:00:39 2013 |
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!