2008-12-26.log

--- Day changed Fri Dec 26 2008
* kanzure_ sleeps.00:26
geneSleep is for slackers00:38
kanzure_Sleep is for the weak00:46
bkerokanzure_: That's generally what happens when you get news coverage.00:47
bkeroEspecially AP and slashdot.00:47
kanzure_I've got Slashdot coverage before, but not AP.00:47
kanzure_AP >>>> Slashdot in terms of distribution.00:47
kanzure_I like one of the Slashdot comments:00:48
kanzure_"Bah, nothing good will come from garages!"00:48
kanzure_"I was conceived in a garage, you insensitive clod!"00:48
geneheh it made fark too, got a scary tag00:49
bkeroBah00:52
bkeroFundies scared of a little engineering00:52
geneand idiots00:52
bkeroI just hope that the gov't doesn't start cracking down at at-home science.00:52
bkeroYou already need a physicians license to order some medical equipment. :(00:53
kanzure_Deadly.00:55
bkeroMy ass is deadly.  Motorcycles are deadly.  People are wimps.00:55
kanzure_http://www.reddit.com/r/science/comments/7lom8/amateurs_are_trying_genetic_engineering_at_home/00:57
genewhere I live it's illegal to sell glassware, but that doesn't stop fry's from selling it00:59
bkeroglassware?  You're kidding.  that's just stupid.01:02
geneI live in texas01:02
genego figure01:02
bkeroWhy stop from selling glass?01:02
geneoh I forgot to say scientific glassware01:02
geneyou know in maryland it's harder to get a flask than it is to get a gun01:03
bkeroYea I just go to the campus surplus store here01:03
bkeroWash them out and stick 'em in an autoclave and they're pretty good.01:03
geneTHERE IS SUCH A THING AS A CAMPUS SURPLUS STORE?01:03
kanzure_the 1989 act, 71st Leg., ch. 678, Sec. 1,01:04
kanzure_eff. Sept. 1, 1989. Sec. 481.002.53.H.01:04
kanzure_that's where the texas-flask bullshit comes from.01:04
genethe strange thing is that maryland has a bunch of biotech companies01:04
bkeroHere there is :)01:05
bkeroGlassware for $0.50 each01:05
genepsst fry's still sells beakers, heck I even seen them carry glass tubes01:05
bkeroBeakers, graduated cylinders, all sorts of things.01:05
willPow3rfry's electronics?01:05
geneyes fry's electronics01:06
genedon't tell the coppers01:06
willPow3rmaybe i *am* the coppers01:10
geneno you aren't the coppers, the coppers were a 1980s heavy metal band01:11
willPow3rwhere are they now?01:12
genewho knows01:13
* kanzure_ likes 80s heavy metal / hair metal.01:13
* bkero has the hair, and the band.01:16
bkeroAs noted by his picture on http://www.osuosl.org/info/people01:16
geneI for one like heavy metal binding proteins01:16
-!- ferrouswheel [n=jp@121-73-144-159.cable.telstraclear.net] has quit [Read error: 54 (Connection reset by peer)]01:35
-!- willPow3r [n=will@cpe-66-75-6-181.san.res.rr.com] has quit [Read error: 131 (Connection reset by peer)]01:41
-!- jm|earth [n=jm@p57B9CB9F.dip.t-dialin.net] has joined #hplusroadmap01:59
xp_prg2my flash is coming along quite nicely :>02:00
* bkero flashes xp_prg2 02:07
xp_prg2wahoo :>02:07
* bkero flashes xp_prg2 to xp_prg302:08
bkeroCRITICAL: Checksum error.02:08
-!- jm [n=jm@p57B9CF3B.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)]02:17
-!- xp_prg2 [n=xp_prg3@c-24-130-128-182.hsd1.ca.comcast.net] has quit ["Leaving"]02:39
-!- gene [n=chatzill@pool-71-164-238-185.dllstx.fios.verizon.net] has quit [Read error: 113 (No route to host)]02:54
-!- willPow3r [n=will@cpe-66-75-6-181.san.res.rr.com] has joined #hplusroadmap03:42
UtopiahGHMLhow well do classical compression algorithms handle DNA? achieving significant compression or not really?03:47
willPow3rhttp://shop1.frys.com/product/2648172?site=sr:SEARCH:MAIN_RSLT_PG04:10
willPow3rim not sure i trust this "wild goose" brand04:10
willPow3ri wonder if making meth would be a good science fair project04:12
willPow3rcarson-dellosa, the company that makes wild goose chemistry equipment, is also a christian education company04:13
wrldpcbizarre04:25
willPow3rthey must be banking on muslim fundamentalists04:37
-!- jm|earth [n=jm@p57B9CB9F.dip.t-dialin.net] has quit [Read error: 60 (Operation timed out)]05:54
-!- wrldpc [n=ben@ool-43503bd5.dyn.optonline.net] has quit []07:25
-!- wrldpc [n=ben@ool-43503bd5.dyn.optonline.net] has joined #hplusroadmap07:48
-!- wrldpc_ [n=ben@ool-43503bd5.dyn.optonline.net] has joined #hplusroadmap07:58
-!- wrldpc [n=ben@ool-43503bd5.dyn.optonline.net] has quit [Read error: 104 (Connection reset by peer)]07:58
-!- wrldpc_ [n=ben@ool-43503bd5.dyn.optonline.net] has quit [Read error: 104 (Connection reset by peer)]08:14
-!- wrldpc [n=ben@ool-43503bd5.dyn.optonline.net] has joined #hplusroadmap08:15
-!- elias` [n=me@unaffiliated/elias/x-342423] has quit [Read error: 145 (Connection timed out)]10:33
-!- elias` [n=me@unaffiliated/elias/x-342423] has joined #hplusroadmap10:42
-!- elias` [n=me@unaffiliated/elias/x-342423] has quit [Read error: 145 (Connection timed out)]11:25
kanzure_http://science.slashdot.org/comments.pl?sid=1073317&cid=26233119  what's a reflow oven?11:32
-!- elias` [n=me@unaffiliated/elias/x-342423] has joined #hplusroadmap11:34
-!- percent [n=percent@66.158.193.46] has joined #hplusroadmap12:21
-!- percent is now known as jihaaad12:21
jihaaaddrazak: 12:21
jihaaadpaging drazak 12:21
jihaaadDRAZAK12:22
jihaaadDRAZAK12:22
jihaaadI HEARD YOU WERE BUILDING A CHEAP PCR MACHINE12:23
jihaaadI WANTED TO HELP LOL12:23
-!- elias` [n=me@unaffiliated/elias/x-342423] has quit [Read error: 145 (Connection timed out)]12:26
jihaaadcuz i know 'bout heaters n shut12:27
-!- elias` [n=me@unaffiliated/elias/x-342423] has joined #hplusroadmap12:35
drazakjihaaad: uh, I'm using peltier devices and fans12:52
jihaaadwhat is yo black ass using to measure temp12:52
drazaksome thermistors12:52
drazak4 of them, in 4 different tubes each with the solution, to use as temperature controlls but not final electrolysis12:53
jihaaadWhy not something like a thermocouple?12:53
jihaaadCan you show me what you have so far?12:53
jihaaadOr is it just an idea in your head?12:53
drazaknowhere12:53
drazakI haven't had time to build it yet12:53
jihaaadYou have plans?12:54
drazakI haven't had time to draw them up12:54
drazakbut it seems pretty simple12:54
jihaaadDon't you think we could have a few more tubes?12:55
drazakthere'll be more than 412:55
jihaaadAlso, is aluminum our ideal material, since it heats and cools so quickly?12:55
drazakjust 4 of them are for temp control12:55
jihaaad(For the block)12:55
drazakuhm12:55
drazakno12:56
drazakyou want something with a little more temp control than aluminium12:56
jihaaadhey if you don't want input/help then you can quit being a faggot12:56
jihaaadand say so12:56
drazakI'm not being a faggot12:56
jihaaadYou're always being a faggot.12:56
-!- jihaaad changed the topic of #hplusroadmap to: fags12:56
jihaaadFag.12:56
jihaaadAnyway12:57
jihaaadWhat do you suggest, if not aluminum? 12:57
drazakwater, use the peltier device to get water hot, and then have a rack for 50um cuvettes, and a peltier device right on top to prevent condensation, and then jut hook it up to a microcontroller12:58
jihaaadSo you're proposing a rack sitting above a little reservoir? 12:58
jihaaadOr therein?12:59
drazakyeah12:59
jihaaaddraw me a pictar12:59
drazakno, I'm busy12:59
jihaaaduse mspain12:59
drazaklol, mspaint12:59
drazakI'm on linux12:59
jihaaadYOU ARE NOT TOO BUSY FOR SCIENCE12:59
jihaaadFUCKING FAGS12:59
drazakyes, I am12:59
drazaksorry!12:59
jihaaadwhat are you doing that is more important than science12:59
drazakstuff that I get paid for13:00
drazak!13:00
jihaaadScience comes first. Sex follows closely thereafter. 13:00
-!- samrose [n=samrose@c-24-11-214-181.hsd1.mi.comcast.net] has joined #hplusroadmap13:02
-!- gene [n=chatzill@pool-71-164-238-185.dllstx.fios.verizon.net] has joined #hplusroadmap13:09
kanzure_jihaaad: Wrong. You're supposed to combine the two, you see.13:29
drazak:P13:33
drazakyep13:34
drazakmy ex and I used to joke about having sex with molecular interactions on the ceiling13:34
kanzure_wait, what?13:35
kanzure_you mean, written out on the ceiling?13:35
kanzure_Hrm. I have a challenge to set forth- can anybody find some ridiculously complicated step-by-step instructions? I'm wondering how to classify what's unapproachable to step-recipes.13:44
kanzure_For instance, control structures are a bit different in step-by-step recipes. Timers and so on have to be represented as "set the timer" and such, and then "link up the timer to such-and-such activity" or something.13:44
kanzure_so in that case, time-based instructions can be encoded in individual linear steps13:45
kanzure_http://www.twine.com/item/11yngqm3s-34z/space-com-simulation-shows-bacteria-could-live-on-mars Mars Environment Simulation Chamber (MESC)13:48
kanzure_http://www.opensourcesensing.org/  wtf13:59
kanzure_Joseph Jackson just linked to that on om13:59
jihaaadmy ex draws molecules on me13:59
jihaaadbut not very often13:59
jihaaadit's usually math13:59
jihaaadusually trig for whatever reason13:59
jihaaadbitch is insane13:59
kanzure_jihaaad: What's the most complicated set of instructions you've had to ever execute?14:05
jihaaadin what?14:06
kanzure_anything but programming14:06
jihaaadC? ASM? Cooking?14:06
jihaaadoh14:06
kanzure_anything that could be expressed as a recipe, though14:06
jihaaadThat's an incredibly abstract question14:06
kanzure_Yes :(14:06
jihaaad"Go to college, make good grades"?14:06
kanzure_Heh.14:06
jihaaadwhy?14:06
kanzure_I'm still working on recipe representation14:07
kanzure_if a linear model of recipes is ok, then I'll go do that14:07
jihaaadI can tell you, the most complex things I've done in C are various rainbow-table related things14:07
kanzure_but PSL provides for some more abstract concepts14:07
jihaaadand14:07
jihaaadin ASM14:07
jihaaadJust shellcode.14:07
jihaaadAnd in cooking14:07
jihaaadeggs14:07
jihaaadthose fucking things14:07
jihaaadare impossible14:07
jihaaadyou should buy a PSL14:07
jihaaadremarkable guns14:07
kanzure_http://www.protocol-online.org/  <-- most of these are "step one, step two" things, where each step is pretty easy14:07
kanzure_but thereeee's a few that I'm either forgetting or vageuly remember that are ridiculously complicated14:08
kanzure_and can't be reduced to "step one, step two" sort of things.14:08
jihaaadwhy are you doing this14:08
kanzure_"Harvest log phase growing cells (trypsinization is OK), wash twice in PBS, pellet by centrifugation (2000 rpm, 5 min in table top centrifuge, or 3 min in microfuge), resuspend carefully and completely in Lysis Buffer (see solutions), incubate at RT 5 min with frequent vortexing, and freeze in ETOH/dry ice bath. Repeat freeze/thaw cycle twice (3 times total) and store at -70°C until ready to quantitate proteins. Try to suspend 105 cells/l5 ml LB. The number of cells need not be exact, but you need a minimum of 5x105 cells in 75 ml to have accurate results, and if you have too many cells in a small volume, they will not resuspend or lyse well."14:09
kanzure_to systematize it.14:09
jihaaadmore complex action requires a longer recipe14:09
jihaaadIn order to tell them how to do something like that14:09
kanzure_I'm also thinking of stuff like "How to assemble your new chair"14:09
jihaaadI'd take at least 3 pages14:10
kanzure_assembling a chair has easy instructions though14:10
jihaaadWhy, though?14:10
kanzure_I guess I should be asking about "nonlinear instructions/recipes". but usually those can be broken down into linear recipes, right? All parallel programs can be simulated on a linear processor, is that correct?14:10
kanzure_something like that.14:10
kanzure_why => because it has to be done? :)14:11
jihaaadWhy must it be done?14:12
jihaaadHonestly, it seems like pointless semantic BS to me14:12
kanzure_semantic is about "meaning". I don't care about meaning.14:12
jihaaadApparently so14:12
kanzure_I'm trying to figure out a computational representation of recipes so that we can generate instructions that are both machine readable and human  readable. Now, I don't care what they actually mean, but as long as there's a computational representation of the steps, that's the important part.14:13
jihaaadbut14:13
jihaaadcurrent machine-readable instructions ARE human-readable14:14
jihaaadiff human is not a complete retard14:14
kanzure_right, so you can convert stuff like "FOLD on line A to point B" to something more verbose and bullshitty that a human would pur about (whatever)14:15
jihaaador you can represent it in c14:15
kanzure_http://www.wgbh.org/cainan/article?item_id=1991546 <-- a supposedly hard recipe,14:15
jihaaadstop reinventing the wheel14:15
kanzure_but ass I read it, it's just a linear sequence of steps14:15
jihaaadit ain't gonna get no rounder14:15
kanzure_I agree that it should directly use an instruction set architecture that everyone is familiar with14:16
kanzure_preferably RISC14:16
jihaaadno please god no14:16
kanzure_better than CISC.14:16
jihaaadi bet you want it to be big-endian too14:16
jihaaadfucking freak14:16
jihaaadx86 4 lyfe14:17
jihaaadrepresent14:17
jihaaadComp sci folks need to quit making pointless abstractions14:18
jihaaadThey do not serve to clarify, only to distort and confuse14:18
jihaaadPeople do not and will not ever think like computers.14:18
jihaaadComputers do not and will not think like people.14:18
kanzure_It's pretty obvious to me that recipes are a list of instructions.14:19
kanzure_Sorry.14:19
jihaaadWe add complexity without adding features14:19
jihaaadwell, yeah14:19
jihaaadbut14:19
jihaaadWhat's the point of translating recipes into an invented machine code?14:20
jihaaadOne which no existing machine can currently read?14:20
jihaaadYou have a perfectly good C compiler14:20
jihaaadand yet you want to draft up some zany RISC processor which nobody will know how to use14:20
kanzure_Actually, machines do read these instructions, like: gcode, whatever format various printers accept (some old stuff accepted straight-up ASCII), etc.14:20
jihaaadOkay14:21
jihaaadExplain to me14:21
jihaaadin two sentences14:21
jihaaadwhat you are trying to do14:21
jihaaadand then, explain, in one sentence, why14:21
kanzure_overall, or what I'm asking of you? heh'14:22
jihaaadOveral.14:22
jihaaad*Overall14:22
kanzure_I'm coming up with a recipe representation format that rivels PSL ( http://www.mel.nist.gov/psl/ ) for the representation of processes, recipes, instructions, in a specific data format. 14:23
kanzure_The recipes can't just be "Natural Language" crap, because that's just asking for another Wikipedia.14:23
kanzure_ooh: Clinical Laboratory Procedure Markup Language, CLP-ML14:24
jihaaadSo, you're creating a language to represent algorithms.14:24
jihaaad...14:24
jihaaadreal inventive14:24
jihaaadfucking why14:24
jihaaadC14:24
jihaaadthey had it right in the 70's14:25
kanzure_it just needs to be a data format, not executable code. 14:25
jihaaadC works anyway14:25
jihaaadwhy not xml14:25
kanzure_the code is simple: go to each referenced package and call the English serialization function given the parameters (like where to fold or whatever) to make more complete instructions for that specific action14:25
kanzure_well, what schema in xml? etc.14:25
kanzure_psl provides one, but nobody knows how to use psl really14:26
kanzure_and the psl guys are dead14:26
jihaaadlearn to use PSL14:26
jihaaadThis is even worse than reinventing the wheel14:26
kanzure_so that's why I'm thinking of just using functionCAD, which is really just this graph-based approach of "A -> B"14:26
jihaaadThey reinvented the wheel so you're reinventing their wheel14:26
jihaaadMore abstraction != good14:26
jihaaadOkay, let's say we have a manufacturing thing, right? A robot.14:27
jihaaadIn theory, we could feed it your language and it would build a firearm14:27
jihaaador something14:27
jihaaadyes?14:27
kanzure_we'd feed it parameters or whatever, yeah14:27
kanzure_like gcode :)14:27
jihaaadso14:27
jihaaadYour hypothetical language would be more abstract14:27
kanzure_what I'm trying to find is the "thingy" that says "ok, give the robot here this gcode file. now stir for 20 minutes"14:27
jihaaadthe problem is14:28
jihaaadimagine a human doing that14:28
jihaaadand a robot doing it14:28
kanzure_doing gcode? Yeah,  it's a bad example14:28
jihaaadWhich is going to be more precise?14:28
kanzure_a better example would be something like printing out vs. hand-copying a book14:28
jihaaadThe human might stir for 20ish minutes14:28
jihaaadexactly14:28
jihaaadso14:28
kanzure_or the origami folding machine versus a professional origami folder14:28
jihaaadhuman language sucks for manufacturing/science in general14:28
jihaaadthis is why we have math14:28
kanzure_and yet people use it.14:28
jihaaadWe need to specify exactly, human language leaves a lot of wiggle room for interpretation14:29
jihaaad"Well, he said stir, but fuck that, I'm going to vortex"14:29
jihaaadyou keep reinventing wheels, and as a result, you'll never build the car14:30
jihaaadand if you do14:30
jihaaadnobody will know how to drive it14:30
jihaaador power it14:30
jihaaador which way to sit therein14:31
kanzure_you're not really helping to answer my question14:31
jihaaadWhich was?14:31
kanzure_whether there's something that I'm going to miss with linear recipes/instructions14:31
jihaaadBranches.14:31
kanzure_hrm14:31
jihaaadYou'll have to branch out and execute each branch14:31
jihaaadOOE14:31
jihaaadis what it's called when processors do it14:32
jihaaadout of order execution14:32
jihaaadthen you're looking at branch prediction, branch table buffers, etc...14:32
kanzure_No, I think linear instructions do branching. "If the chicken is black, then go to section 4."14:32
jihaaadOf course, but you'll be writing an insane amount of code14:32
jihaaadjust write an interpreter!14:33
kanzure_Saadawi G, Harrison JH. A lightweight XML editor for clinical14:35
kanzure_laboratory procedure manuals. Arch Pathol Lab Med 2004;128:14:35
kanzure_1104.14:35
jihaaadUNREASONABLY COMPLEX MARKUP LANGUAGES AHOY, CAPTAIN14:37
jihaaad"sir C is too easy to use and makes too much sense"14:38
jihaaadFIRE THE ABSTRACTION TORPEDOES 14:38
jihaaadyes sir14:38
kanzure_yeah, because all chair assembly instruction manuals should just be a giant .c file.14:39
jihaaad"sir it worked nobody has any fucking clue what the hell is going on and the language is actually not useable"14:39
jihaaadBAHAHAHAHAHAHA14:39
jihaaador14:39
jihaaadphrase them in simple G code14:39
jihaaadfor a machine14:39
jihaaadfor a person14:39
jihaaadTell him how to do it14:39
jihaaadaloud. or in writing.14:39
jihaaadmaybe some pictures.14:39
-!- jihaaad [n=percent@66.158.193.46] has quit [" "]14:51
-!- ferrouswheel [n=jp@121-73-144-159.cable.telstraclear.net] has joined #hplusroadmap15:05
-!- samrose [n=samrose@c-24-11-214-181.hsd1.mi.comcast.net] has quit [Read error: 60 (Operation timed out)]15:13
kanzure_if only overloading was more dynamic. "We see that there are 20 different solutions to this function, and 30 to this one, but only 5 combinations that make it so that the two functions match up specifically."15:14
-!- elias` [n=me@unaffiliated/elias/x-342423] has quit [Read error: 145 (Connection timed out)]15:35
-!- elias` [n=me@unaffiliated/elias/x-342423] has joined #hplusroadmap15:44
genelol FIRING AWAY CAPTAIN!16:09
genedrazak I am perplexed16:11
geneas to how that is possible16:12
drazakgene: what is?16:44
-!- elias` [n=me@unaffiliated/elias/x-342423] has quit [Read error: 145 (Connection timed out)]16:54
-!- wrldpc [n=ben@ool-43503bd5.dyn.optonline.net] has quit []16:59
-!- elias` [n=me@unaffiliated/elias/x-342423] has joined #hplusroadmap17:02
kanzure_part of the problem goes back to kernel time sharing algorithms and the like, the traditional monolithic-vs.-microkernel debate. but ultimately userspace should be full of linear stuff, so.17:42
kanzure_so maybe I'll just write a few wrappers for linked/ordered lists and call it quits.17:42
kanzure_http://news.slashdot.org/comments.pl?sid=1073673&cid=26235239   Bruce Perens did 4chan?18:11
-!- ybit [n=heath@unaffiliated/ybit] has joined #hplusroadmap18:52
kanzure_lo ybit.19:02
genedrazak the molecular interactions part19:04
ybitoi kanzure_19:05
ybitgene [nod]19:05
geneUgh I don't want to restart my computer to install this software19:06
ybitwhat software?19:06
ybitmessing with your kernel?19:06
genedriver software for a clearance USB microscope19:06
kanzure_you're doing it wrong.19:06
kanzure_use modprobe.19:06
genewhat's modprobe?19:07
kanzure_adds a driver to kernelspace without rebooting19:07
ybitman modprobe, no? :)19:07
genewell this driver was probably hacked from a webcam19:08
kanzure_So, Paul was complaining earlier today about safety and diybio, and was saying "ok, so we should just put everything into remote cells for tinkering". I wonder how he lives with his wife?19:08
geneHAHAHAHA19:08
kanzure_I'm serious though.19:08
kanzure_an honest issue.19:08
geneKanzure let me tell you a little something about e.coli19:09
genelaboratory strains of e.coli have gone through so many generations, that they are now adapted to only living in a lab19:09
genethey can19:09
gene't live in your gut19:10
genebtw this driver isn't even workin19:10
geneit thinks the webcam in my laptop is the microscoe19:10
-!- gene [n=chatzill@pool-71-164-238-185.dllstx.fios.verizon.net] has quit [Read error: 104 (Connection reset by peer)]19:15
kanzure_I think the point of modprobe was lost on him.19:15
ybitmy feeds are getting out of hand19:19
kanzure_I've stopped reading.19:19
ybitthe curse of holidays19:19
ybitcome on, you got to read some19:19
ybitwikinews at the least19:20
kanzure_No, as it turns out, I get the same crap through email.19:20
ybit(to stay in touch with the world)19:20
ybitheh, there is a lot of redundancy19:20
kanzure_stay in touch with .. what?19:20
ybitpossible disasters? there could be a natural disaster occuring in a city you are about to visit19:21
ybitwould help to know that19:21
ybitsomeone could beat you to what you're working on, would be nice to know that19:22
ybitagree, disagree?19:26
kanzure_I'm not sure any more. News reading hasn't really led to anything spectacularly productive except obsessive compulsive information hoarding, meanwhile the fundamental underlying issues are still at stake (i.e., all of these damn iGEM projects in sloppy submission styles)19:27
ybiti've noticed this19:34
-!- gene [n=chatzill@pool-71-164-238-185.dllstx.fios.verizon.net] has joined #hplusroadmap19:42
genewoo the microscope works19:42
kanzure_Neat, Bruce Perens replies to email.20:07
genewho's bruce perens?20:22
geneand what would you do with a 130x cheapass usb microscope?20:22
kanzure_Make lots of microbe videos.20:26
-!- elias` [n=me@unaffiliated/elias/x-342423] has quit [Read error: 145 (Connection timed out)]20:26
kanzure_Bruce Perens is the guy who did OSI and at one point was a big head guy behind debian. If you saw "Revolution OS", he was one of the guys talking after ESR, RMS, and Linus.20:27
geneyeah big microbes20:27
kanzure_erm, I'm sorry, not microbes, planarians and the like20:27
geneYeah I know what you mean20:27
genehey you know how I can tell the software this microscope uses is from a webcam?20:28
genethe software has various "effects"20:29
genenot very useful for a microscope if you ask me20:30
kanzure_"Relational database theory describes the manipulation of hyper-rectangles, but we fake it very badly with indexes we actually have algorithms for.  Did you ever wonder why no one has built a massively distributed SQL database despite the obvious value?  It is not because it is theoretically impossible, but because it is only possible if someone discovers a general algorithm for indexing hyper-rectangles -- faking it is not distributable. "20:31
willPow3rtl;dr20:32
kanzure_I thought that Google does distributed SQL sort of though20:32
kanzure_i.e., caching and the like.20:32
kanzure_What's the deal?20:32
genereminds me of my compsci roommate's homework20:32
genehe had to turn a sentence into an nth dimensional vector20:33
geneso you can compare sentences with each other20:34
-!- elias` [n=me@unaffiliated/elias/x-342423] has joined #hplusroadmap20:35
kanzure_gene: what was the comparison operation?20:37
kanzure_i.e., how do you compare two given words?20:37
geneI forgot you represent each word as a vector and it's incidence is the magnitude of the vector20:39
geneI don't know20:39
geneI'll have to ask my roommate when I can get back20:39
kanzure_You could do a hack with WordNet, but I'm wondering what the assignment was asking specifically for.20:39
geneI don't remember what it was about my roommate didn't understand it20:40
kanzure_multidimensional vector stuff is common in web search algorithms, for instance. but anyway.20:41
geneyeah it was something like that20:43
-!- willPow3r [n=will@cpe-66-75-6-181.san.res.rr.com] has quit [Read error: 54 (Connection reset by peer)]20:44
drazakgene: ?21:31
kanzure_fenn: what about specific materials in your Measurements class? i.e., "20 g of CAS ID #338024" ? I guess we can come up with a class that encapsulates sourcing of things, which has a member variable for the measurements/units21:59
-!- willPow3r [n=will@cpe-66-75-6-181.san.res.rr.com] has joined #hplusroadmap22:27
proctogene: he was either turning it into a markov chain or a a document-term matrix22:27
proctokanzure_: google does something totally different22:28
proctokanzure_: BigTable is a column based db22:28
proctorelational DBs are incredibly powerful... for some thirgs.22:28
proctothings*22:28
proctothey are NOT good for things like documents22:28
proctowhich is what google indexes and what most of human-produced content is like22:29
proctowhich is why I use couchdb for my personal projects22:29
proctoa lot of people use RDBs because they have become ubiquitous, and are proctically synonymous with what people understand DBs to be22:31
proctokanzure_: the assignment sounds like it has to do with latent semantic analysis22:31
* kanzure_ has some done some wacky cross-table queries with indices pointintg back to indices in weird criss-cross directions22:41
kanzure_It's not pretty.22:41
kanzure_hrm. Somebody from Australia just phoned me to tell me to keep my grades up. I wonder how much that costed him.22:41
geneagain?22:57
geneI don't get it, I don't want to know22:57
genekanzure do you know any good PCB layout tools a23:12
drazakgene: they all suck23:13
drazak:S23:13
geneyeah but which one sucks least23:14
-!- ybit [n=heath@unaffiliated/ybit] has quit [Read error: 110 (Connection timed out)]23:20
kanzure_http://www.brpreiss.com/books/opus7/html/page84.html  what's the meaning of the (object) on line 1? Inheritance?23:26
kanzure_class blah (object):23:26
genedrazak do you prefer freeforming23:27
drazakgene: paper and a mech drawing table23:29
* drazak nods23:29
genewhat's a mech drawing table?23:30
kanzure_probably what you were drawing on in the mech drawing class.23:30
drazakgene: a table with a thing that adjusts to different angles23:31
geneoh23:31
geneyou seem very old fashioned, how do I use the save function on paper?23:31
drazakscanner23:32
drazak:P23:32
genewhat OS does paper run?23:32
kanzure_drazak: http://brlcad.org/23:32
genealibre.com23:32
genehttp:/alibre.com23:32
geneit ain't open source but it's free23:33
genebah BRL cad's a primitive based constructor isn't it?23:33
genethat's what second life uses23:34
kanzure_what do you mean 'uses'23:34
drazakeaglecad23:34
drazakI just find them restraining23:34
kanzure_what, you can't import custom data?23:34
geneis eaglecad free?23:34
kanzure_import custom data to second life, I mean23:34
genethat's why I hate it23:35
geneoops I meant to say that's why I hate it vehemently23:35
geneWHAT? I HAVE TO PAY TO USE THIS TEXTURE?23:36
geneeagle cad link ?23:38
geneone of my favorite apps for primitive type modelling and mesh based modeling is metasequioa23:40
genemight've spelled it wrong23:41

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