2009-01-24.log

--- Day changed Sat Jan 24 2009
genehackerwell I would do it if it worked00:02
genehackerjeez, I don't know what excuse I can have for shaving my head00:05
kanzure3cancer00:05
genehackeryeah then I have to have to learn all about cancer so I can keep my stories straight00:06
kanzurehm, I need to hunt down Osaka, Kitamura and Mizoguchi [5,31]00:06
genehackeryes I would00:08
genehackerI want to get a brain implant as soon as they're safe00:09
genehackerI want to get a brain implant as soon as they're safe00:09
genehackeroops00:09
genehackerhttp://en.wikipedia.org/wiki/Transcranial_direct_current_stimulation00:19
genehackerhmmmm.... induces noticeable psychological changes, I might want to read up on what exactly those were00:19
genehackerhmmm... 10-30 minutes of stimulation is good for five hours00:30
kanzurehave we talked about RoBlock in here before?00:32
genehackerno00:35
genehackerwait yes00:35
genehackerI think so00:35
genehackerwhere's the right orbita?00:37
kanzureWhat?00:37
genehackerIn the00:38
genehackerfirst experiment, different electrode positions were tested to find the00:38
genehackeroptimal positions for DC stimulation. In the subsequent experiments,00:38
genehackerthe optimal electrode arrangement (motor cortex—forehead above00:38
genehackerthe contralateral orbita),00:38
kanzurerawr, this is why I was working on a 3D model viewer of the brain last year00:39
kanzureI wanted to tag regions with the many different names that they are given just for this purpose..00:39
kanzure"show me teh cerebellums!"00:39
genehackerwell guess what?00:39
genehackeryou get to learn all about regions of the brain in pyschology, or at least the names of them00:40
kanzureno you don't, I've been through psych00:41
genehackeroh00:41
genehackerthat's what one my friends said00:41
kanzureI learned more about the brain going through ontology lists than I have ever in all of my years in psych classes00:41
kanzureblah, don't listen to them00:41
kanzurehold on a sec.00:41
kanzureaww crap00:41
fenngenehacker: sponge+salt water electrodes00:41
kanzurewell, if I was on my laptop,00:41
genehackerthat is what it is fenn00:41
kanzureI have this huge section of my bookmarks for structures of the brain00:41
kanzureit lists >200 regions00:42
kanzure(only about ~12 are ever mentioned in psych classes..)00:42
genehackerheck that's practically what the commercial unit is00:42
genehackerI just don't want to shave my head00:42
fenni dont think you need to shave anything00:43
genehackerthe electrodes are 35 cm sq00:44
fenn3.5cm radius is not terribly large00:44
kanzure3http://www.sisostds.org/ simulation interoperability standards organization00:45
kanzure3what is a base object model (BOM) and how is it not a Bill of Materials (BOM)?00:45
kanzure3http://www.simventions.com/boms/00:46
fenngod that paper would be so much less annoying if they called things "interface" instead of "port"00:46
fennbut then i guess they wouldn't have anything because it'd be so obvious00:47
fennis this part of the basis for design repo silliness?00:48
fenni havent read the phrase 'mechanical flow' anywhere else (no gene, it's not anything like optic flow)00:48
kanzurewhat?00:48
kanzureoh00:48
kanzurethis is just me googling around for legos+interoperability00:49
kanzurebut it turns out that that 'port ontology' paper was cited by a paper by Tolga, the guy that I sort of replace at ADL.00:49
kanzure(via the "citations" Google link..)00:49
kanzureerm, "cited by" link.00:49
genehackerhmmmm... this paper on tDCS had a sample size of 1900:50
kanzureew, the port-ontology paper just turns into "tag it with lego-circular-shape"00:52
kanzurepage 208 (pg 3), where the Kitamura and Mizoguchi paragraph is located, is what I found the more interesting00:55
kanzurein their paper, they separate objects that "augment the value of feedstock" versus the 'conduits' which try to maintain the value as much as possible (conduits being the "glue" between an output port and an input port somewhere) - 'stasis'00:56
kanzurebut all of these 'ontologies' suck immensely and are the wrong approach00:56
kanzure(another aspect of their paper was "a functional way", so as to separate POV of different actors entering information..)00:56
kanzurehm01:01
kanzuremy rebel spies inform me that this would totally be the time to make suggestions for rebranding and restructuring of the Singularity Institute01:01
genehackerhttp://clinicaltrials.gov/ct2/show/NCT0075749701:03
genehackerThis study will test whether transcranial direct current stimulation (TDCS) can be used safely in children with schizophrenia and if it can improve memory and attention span or auditory hallucinations in these children, at least temporarily. TDCS has temporarily improved memory and attention span in healthy adults and a similar method called TMS has relieved auditory hallucinations in adults...01:04
genehacker...with schizophrenia.01:04
genehacker;)01:04
genehackerimproved attention span with electricity to the head01:04
kanzurefor "improved attention span", please see-01:05
kanzurehttp://heybryan.org/mediawiki/index.php/Sustained_attention01:05
fennimproved auditory hallucinations01:08
fenndoes it come with bluetooth?01:08
genehackerspeaking of that01:13
genehackerfenn, ever watched ghost in the shell?01:13
fenncertainly01:13
genehackerwell you know how they all talked without talking?01:13
genehackerwe have the technology to do that01:13
fennthe electrodes you put on your neck?01:14
kanzuresubvocalization cheap trickery? :/01:14
genehackerno, a throat microphone01:14
fennthroat microphone doesn't count01:14
genehackerhttp://www.clearercom.com/01:14
genehackerit's close enough01:14
fennbah01:15
genehackersaw one at an outdoor store the other day designed for use by hunters01:15
genehackerof course it was in the bargain bin...01:16
fenn" An electrode implanted in the patient's brain made it possible for the patient to produce vowels by thinking them, using a speech synthesizer"01:16
fennthat's a bit more invasive than i'm thinking01:16
genehackeroh yeah I remember that one01:16
kanzuresufficiently invasive to grab my attention01:16
fennthis was a throat cuff with an array of like 16 electrodes over the trachea01:16
fennit was mentioned in this channel a while ago01:17
kanzurewas it the one about the browser?01:17
fennoh wait maybe they were using it to drive a wheelchair or something01:17
genehackerthough we really need something like in GITS people talking on the phone can be quite annoying01:17
fenni'd rather text-based communication became more prevalet01:18
kanzureI shouldn't be thinking at this hour of the night01:20
kanzurebut should I implement my command line interoperability stuff?01:20
kanzureit'll work, but I'm not convinced that it is analogous to the skdb stuff any mroe01:20
kanzure*more01:20
genehackerhey kanzure, they got parkour training by the tower at 10 in the morning01:21
kanzureparkour?01:21
kanzureI'm already booked01:21
kanzuremy comp sci teacher from 2005 is paying me a visit :)01:21
genehackerhttp://www.youtube.com/watch?v=jquXcwooV6A01:21
genehackerah fun01:22
genehackerwhen I took compsci, my teacher got fired for mysterious reasons01:22
fennreminds me of "mystic arts of the ninja"01:22
genehackerhe was a really cool guy too, gave out copies of certain softwarez01:23
genehackerended up learning about how to play flashgames and write html code the rest of the year instead of CNC machining or other fun stuff01:24
kanzure"html code"01:24
kanzureorly01:24
fennit's about as much of a programming language as g-code01:25
genehackeryeah01:25
genehackerthough I forgot most of it01:25
genehackerhmmm... what's cognitive safety?01:25
genehackerI'll I can find is weird how to manage people bs01:26
genehackerhttp://en.wikipedia.org/wiki/10-20_system_(EEG)01:27
genehackerhow's that for a brain model kanzure?01:28
genehackerstandardized eeg placement locations, also used in placement of TMS01:28
genehackerhttp://www3.interscience.wiley.com/cgi-bin/fulltext/112702866/PDFSTART01:35
genehackeraccording to standard attention measuring tests, tDCS resulted in a 40-60% improvement above baseline01:36
genehackern = 1801:38
-!- genehacker [n=chatzill@wireless-128-62-174-19.public.utexas.edu] has quit [Read error: 110 (Connection timed out)]02:38
kanzure3http://ejohn.org/blog/ocr-and-neural-nets-in-javascript/   11:40
kanzure3Heh. GreaseMonkey userscript (javascript) OCR for solving Megaupload captchas.11:40
kanzure3http://www.youtube.com/watch?v=ZN60Ob5x7f0 Autodesk Labs Project Draw11:43
kanzure3http://draw.labs.autodesk.com/ADDraw/draw.html11:43
kanzure32D AJAX diagramming thingy.11:43
kanzure3I wonder if they are interested in doing 2D CAD nonsense in the browser?11:43
-!- PeerInfinity [i=PeerInfi@brndmb02bbp-ac09-53-249.dial.mts.net] has joined #hplusroadmap12:28
-!- PeerInfinity [i=PeerInfi@brndmb02bbp-ac09-53-249.dial.mts.net] has quit []12:34
kanzure3http://www.youtube.com/watch?v=JgBgmw-2U8c NASA ISS tour13:36
-!- genehacker [n=chatzill@wireless-128-62-174-19.public.utexas.edu] has joined #hplusroadmap13:56
-!- samrose [n=samrose@99.147.180.206] has joined #hplusroadmap16:15
kanzureHi Sam.16:22
kanzureI think your proposal to only keep the drupal-enabled site as a 'showcase' fell on deaf ears ('cept me :-))16:22
-!- samrose [n=samrose@99.147.180.206] has quit [Remote closed the connection]16:26
kanzurehm, MIT used to have a project called "LEGO Constructopedia"16:44
kanzure3http://heybryan.org/books/Manufacturing/Designing%20effective%20step-by-step%20assembly%20instructions.pdf  designing effective step-by-step assembly instructions16:53
kanzure3in that paper, they use a "planner" to figure out a good sequence for assembling some given object from multiple parts16:53
kanzure3I'm pretty sure s/planner/wizard/ (in terms of what we've called it before) is sufficient16:53
kanzurehahah!17:01
kanzuredesignosaur17:01
kanzure3computer-aided tool selection  <-- another buzzword17:27
-!- genehacker [n=chatzill@wireless-128-62-174-19.public.utexas.edu] has quit [Read error: 110 (Connection timed out)]17:43
kanzureWhy do USB male/female adapters not look like IEEE 1394 adapters?17:51
kanzure3http://www.usb.org/developers/docs/  pg 76 of the USB 3.0 spec has the diagrams.18:02
kanzureanyway, for interop purposes, just include a "USB" part in the assembly - screw tagging. USB doesn't look like 1394 because it's all based off of typical part mating..18:05
-!- ppk [n=pk@c-98-217-103-22.hsd1.ma.comcast.net] has joined #hplusroadmap18:12
ppkbryan have you seen the scribd diybio section?18:12
kanzureNot yet, no. How much is there?18:13
ppknot much18:13
ppkI don't know much about it either18:13
ppkbeyond that it's supposed to be a more accessible alternative to your git repository18:13
ppkdiybio has 70 gigs of scientific papers to distribute18:14
ppkwe should be able to parse them all somehow18:14
kanzureI sent a hard drive up to them.18:14
kanzureActually, I sent it up to the fablab people. Go ask the Boston Fab Lab for my hard drive and get a copy.18:15
ppkyeah I was in somerville on wednesday18:15
kanzurehm?18:15
kanzurewere they handing out copies?18:15
ppkno18:15
ppkbut we did talk about finding a way to easily distribute the papers18:15
ppkin a way that won't get people in trouble18:15
kanzuretorrents.18:15
ppkbut is also very easy18:15
kanzuredebtorrents, etc.18:15
ppkand keeps the papers parsable18:16
kanzureAlec has been mentioning that he is cooking something up.18:16
kanzurewell, part of the problem is that most PDFs suck anyway18:16
kanzureand are just image scans18:16
ppkyeah18:16
kanzurethe Science Commons folks have prodded me in the past for some thoughts on image segmentation algorithms18:16
ppkI agree18:16
kanzureto separate graphs, figures, etc., from the text18:16
ppkyeah18:16
ppkwell I'm checking out scribd now18:17
kanzureimagine a simple script: ./image-segment my.pdf and out pops fig1.png, fig2.png, fig3.png18:17
ppkit has mac's attention for some reason18:17
ppkhttp://www.scribd.com/group/66598-diybio18:17
ppkand ywah that would be an awesome script18:17
kanzureI only see a handful of papers there .. 618:18
kanzurewhereas I mailed 120,000 to Boston :p18:18
ppkright18:18
kanzureget to work! :)18:18
ppkthis scribd doesn't seem much better than torrents18:18
kanzureseems worse to me, IMHO18:18
ppkpapers aren't parsable here either18:18
kanzurebad experience loading the pages too18:18
ppkpersonally I would just make a private tracker18:18
kanzurenothing wrong with torrents18:18
kanzureright18:19
kanzureI'd host it, but my upload speed is shit18:19
kanzureI mean, a tracker doesn't need to have a high upload speed18:19
kanzurebut it helps18:19
ppkespecially when you're distributing papers18:19
kanzuresuperseed mode I guess18:19
ppkseedboxes aren't super expensive18:20
ppkand I'm willing to seed a copy of my own18:20
kanzurethat's right, there's a few sites now that let you pay a bit for 'em18:20
kanzureyou should go bring an external hard drive with you and go track down whoever currently has a copy of the hdd that I sent up there :)18:20
kanzureit will take maybe 8 hours to copy over18:20
ppkmac cowell has it18:20
kanzureoh goodie18:20
kanzuredidn't know.18:21
ppkwell when he presented this problem the first thing I said was use a private tracker18:21
ppkbut it needs to be easily searchable18:21
ppkso I need to find a way to automate 'uploading' the torrents18:21
ppknobody will do 120k by hand18:22
kanzurewhat?18:22
kanzureoh, so you don't want a super-massive 70 GB torrent?18:22
ppkunless you wanted to throw all the papers in one torrent18:22
kanzurewell that'd be a good start18:22
ppkright, I think that would be cumbersome18:22
ppkbut, it's easy18:22
kanzureanyway, it was my understanding that there are some shell utilities to do torrent packaging18:22
kanzureand if so, that just means one would have to split the directory index into sections of, say, 10k papers18:23
kanzurewhich again is a task fit for a shell script18:23
ppkthat would be more manageable18:23
ppkperhaps people would be willing to download the entire torrent if they were granulated small enough18:24
kanzureppk, I suspect that people would be interested in downloading it regardless18:24
kanzurethere's an amazing super awesome nerd factor in knowing that you have downloaded the entirety of Nature18:24
ppkyeah18:24
ppkwell, here's hoping18:24
ppkI didn't know that Mac got the papers from you18:25
kanzure3Port-Compatibility and Connectability Based Assembly Design18:25
kanzure3J. Comput. Inf. Sci. Eng.  -- September 2004 --  Volume 4,  Issue 3, 197 (9 pages) 18:25
kanzure3DOI:10.1115/1.177965918:25
kanzure3is anyone able to get this? 18:25
kanzureyeah, I didn't know he got them from me either18:25
kanzureI was expecting it to circulate around up there, glad it got to the diybio folks.18:25
ppkare you familiar with what.cd?18:26
kanzureno, but a friend of mine used to be a big member18:26
kanzuresomething about interviews for entry :)18:26
ppkit's the best/most popular music torrent site18:27
kanzuredidn't it die?18:27
ppkI'd like to do something similar for diybio18:27
ppkalbeit less elitist18:27
ppkvery much alive18:27
ppkhave an invite only community - that way you can minimize leechers and people trying to stop the distribution of papers18:28
kanzure3http://heybryan.org/books/Manufacturing/Port-compatibility%20and%20connectability%20based%20assembly%20design.pdf18:30
* kanzure3 had to ssh in as a professor to get that..18:30
ppklol18:31
ppkhave you heard of tom knight's ftp?18:31
kanzureNo?18:31
kanzuregimme gimme gimme18:31
ppkhe has an ftp with a user/pass he unwittingly gives to people18:32
ppkwink wink nudge nudge18:32
ppkI forget the user/pass, I'm sure Mac has it though18:32
kanzurewhat's the server?18:32
ppksomeone mentioned it when we were talking about distributing the papers18:33
ppkdunno18:33
ppkbut supposedly it has a lot of papers on it18:33
kanzurehopefully not as much as me :-)18:33
kanzureI might have to go beat my own record18:33
ppkheh18:33
ppkwell18:33
kanzureI was attempting to do a total rip of ScienceDirect last year,18:33
ppkmy terabyte can't come soon enough18:33
kanzurebut I didn't bother when I realized how much I hate their site18:33
ppkmmm18:34
ppkdinner, bbl18:34
kanzure3http://www.part-solutions.com/index_main.asp18:40
fenngee i wish i could make money selling part-solutions18:45
kanzure3the paper that I last linked to has a description of their service on page 218:45
fenna catalog of things that almost work?18:45
kanzure3they seem to provide a 'configuration model' that complements some CAD data18:45
kanzure3but I'm unable to determine if it's just physical/mechanical connections, or if they are magically doing something more18:46
kanzure3'configuration model' is a terrible name. The paper says that their configuration model consists of 'connection points'- what does that have to do with configuration?18:47
kanzure3guess it's sorta related in a hack-eyed way..18:47
fennthey're wrong about lego, most lego instructions use action diagrams18:57
kanzurepg 6 gets interesting18:57
kanzurebetween 2 and 6 I must have been on Mars18:57
fennwow this paper is quite old19:02
kanzure2004?19:02
fenn... ok nevermind19:03
fennthey were running stuff on a 500MHz SGI workstation19:03
kanzurethat's about my 1998 box (the server)19:04
fennwell, that's a very cool paper, i hope something Free comes along eventually19:06
kanzurehrm. so it turns out that they don't really go into detail about 'working principle' (applied forces, etc.)19:06
kanzure'permissible forces, torques, etc.'19:07
fennno it's all geometry19:07
fenn"permissible forces" would require arduous FEA at each step or proposed step19:08
kanzurewouldn't that be specified by the designer though?19:08
fennuh, i dunno?19:08
kanzurethe previous two sections before that were talking about labels19:08
kanzurewhich would be manually typed in, presumably19:08
kanzure(the paper then says how much labels suck (yay))19:08
fennforward to OM?19:10
kanzureshould I include the link to the PDF?19:11
fennya19:13
fennthe website doesnt mean much to me19:13
-!- genehacker [n=chatzill@wireless-128-62-174-19.public.utexas.edu] has joined #hplusroadmap19:14
kanzurefenn: heh. "therefore function is insufficient for evaluating compatibility"19:37
kanzureunfortunately, Rob Stone, one of the guys running VOICED from another university,19:37
kanzureand Wood, the professor in the manufacturing/design lab (parent lab to ADL), 19:38
kanzurewrote "Development of a Functional Basis for Design"19:38
kanzure*do'h*19:38
fennwell, he's right19:38
genehackerWood as in Kristen L Wood?19:44
kanzureyes, I believe so.19:45
kanzurethe mechanical guy19:45
genehackerdo you know what research he does in Solid Freeform Fabrication?19:46
kanzureHe doesn't any more.19:47
kanzureHe now helps run the parent lab to ADL19:47
genehackerok19:47
genehackerI want to know what this university does in the way of SFF research19:48
kanzurehey isn't there a site that lets you search government grants to researchers?19:50
kanzureand wasn't there also a way to narrow down the search by the institution and keywords?19:50
genehackernevermind19:51
genehackerfound the universities site on it19:51
genehackerinteresting, someone here is trying to do nanolithography freeform fabrication19:54
genehackerI wonder how that works out19:54
kanzurelink?19:54
genehackerhttp://utwired.engr.utexas.edu/lff/research/#SURF19:54
kanzureTM 319:54
genehacker???19:55
kanzure<--- has the dual functionality of also being a giant associative pokemon lookup table19:55
kanzuresurf was TM 3, "technical machine 3"19:56
genehackerfound a cool paper, can't find the paper though19:57
genehackerD.Y. Fozdar, W. Zhang, M. Palard, C.W. Patrick, S.C. Chen, “Flash Imprint Lithography Using a Mask Aligner:  a Method for Printing Nanostructures in Photosensitive Hydrogels,” Nanotechnology, Vol. 19, pp. 215303 (1-13), 2008.19:57
kanzureooh19:58
genehackerhttp://www.me.utexas.edu/~scchen/publications.html19:58
genehackerhere we go19:58
kanzurethis other paper that I'm reading, ref 8 from the previous one,19:58
kanzurehas a full assembly planner implementation19:59
kanzurepage 385 (17 of the PDF)- 'rotate pallet/trunion .. insert PART-23 .. orient and install DETECTOR-0"19:59
kanzure3http://heybryan.org/books/Manufacturing/A%20prototype%20of%20feature-based%20design%20for%20assembly.pdf19:59
kanzurehuh, also a task planner for reconfiguring a flexible manufacturing floor .. wtf? this must be bullshit20:00
genehackerhttp://www.me.utexas.edu/~scchen/pdf/2007JVST.pdf20:01
kanzureUV lamp heh20:04
genehackerwell I'll be20:04
genehackerone of the papers mentioned in a paper about membraneless filtration was written by that guy20:11
kanzurefenn: so I went through an entire conversation with you without knowing that we were talking about two different papers20:35
kanzurehuh..20:35
fennyeah i was sort of confused what that german website had to do with anything20:36
fenni guess i skipped a couple pages while reading the backlog20:38
kanzureanyway, just for completeness' sake, /books/Manufacturing/ always has the last-modified option so that you can see the latest papers dumped in there.20:44
-!- ybit [n=heath@unaffiliated/ybit] has quit [Read error: 104 (Connection reset by peer)]20:56
kanzuremy inbox has been talking to itself today about a few cases of parents taking their kids to Costa Rica for a stem cell therapy treatment for autism21:09
kanzurethe youtube video shows little girls getting injections through the wrist21:10
kanzure3http://www.youtube.com/watch?v=d0ZanIBoGHU&feature=related21:11
kanzure3http://www.youtube.com/watch?v=FlSaXCzLW6w&feature=related21:11
-!- Phreedom [n=freedom@ip-194-50-167-184.mir.dn.ua] has quit [SendQ exceeded]21:48
kanzure3http://www.secretlifeofmachines.com/22:34
kanzure3http://www.timhunkin.com/22:38
kanzure3http://www.timhunkin.com/44_secretlifeofmachines2.htm  huh, he encourages people to grab his tv show via torrenting22:40
-!- PeerInfinity [i=PeerInfi@brndmb02bbp-ac09-53-43.dial.mts.net] has joined #hplusroadmap22:48
genehackercheck this out23:31
genehackerhttp://micromanufacturing.com/showthread.php?t=39923:31
genehacker0.5 mm borescope23:33
-!- ppk [n=pk@c-98-217-103-22.hsd1.ma.comcast.net] has quit []23:40

Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!