2008-09-28.log

--- Day changed Sun Sep 28 2008
kanzureSteve's brother really, really hates ESR apparently.02:25
kanzureSomething about terminfo being broken.02:26
kanzurebkero: Know anything about Urszula Chajewska?03:23
bkeroSorry04:11
ybitconcening skdb, instead of manually entering in information for a material, anyone want to share thoughts on using an "intelligent" (bot|program) to do classify and identify properties for the user04:16
kanzureybit: *cough*04:20
ybitguess a realistic approach would just be to buckle down and enter in the info04:20
kanzureybit: you know I'm downloading matweb for a reason, right?04:20
kanzuremeep04:20
kanzurehttp://designworld.matweb.com/search/datasheet_print.aspx?matid=4474404:20
kanzureunless you mean automated material testing equipment04:20
ybitwhat about fruit?04:21
ybitplants04:21
kanzure?04:21
fennwhat about meat on a stick?04:21
fennor popsicles?04:21
ybit:P04:22
ybitSeed the database with a number of functional parts, components, tools, etc.04:22
ybit    * Basic: lemon battery04:22
fennfuck lemon battery04:22
kanzureseeding, right04:22
genehuh?04:22
ybitfrom the wiki04:22
genewhy?04:22
kanzureso that you don't have to manually enter all this bullshit04:22
kanzurere: meat on a stick04:23
kanzureor really just re: bio in general,04:23
kanzurenote that I've been working off and on that synthetic biology circuit creator app04:23
kanzureso somehow biobricks fit into the scheme of things somehow04:23
kanzureyet it's not obvious to me that it's appropriate to place 'biobrick' next to a 'gear' in the same dir for a db. meh.04:24
fennyou dont like the debian "throw twenty thousand packages in the same dir" strategy? :)04:24
kanzureno04:25
kanzurejust my typical turtle objections04:25
kanzurethe molecule is not the gene04:25
kanzureerm04:25
kanzurethe protein is not the gene04:25
fenna biobrick is not a gear, unless of course it is04:25
kanzurebiobricks could /encode/ gears04:26
kanzuredoes that still make it a gear?04:26
kanzurejust wondering.04:26
kanzureI could go either way on this04:26
fennthe gene is source code for the protein04:26
fennthere are a number of ways we could have "source code" for a gear as well04:26
fennso how do you classify a gear generator program04:27
* fenn grumbles about e-prime04:27
kanzureare we going to have stupid N part generators?04:27
kanzurecompiler theorem04:27
fennN part?04:27
kanzureany problem can be broken into smaller parts04:28
genewhat .stl files?04:28
kanzuresay you have these parts: gear, train, rocket04:28
kanzureso are you going to infinitely splice this? 04:28
kanzureinto smaller and smaller turtle shells ?04:28
kanzureso we have a turtle shell generator?04:28
kanzurethat sounds unreasonable, frankly04:28
fennno, that leads to running a simulation of the universe, which is computationally impossible04:28
genehuh what are you guys talking about?04:28
kanzureso we don't need a gear generator04:28
ybitsorry, was distracted, just got a txt from my mum. supposedly, some previous u.s. president is related to me somehow. whatever. how is the data being formated into uniform yaml files?04:28
kanzurebut maybe we need a gear-configurator, yes04:28
kanzurewhich data, ybit?04:29
genejust store em as stl files04:29
ybitall of it04:29
geneall of it04:29
fenna gear generator program is one layer of turtles up in the abstraction hierarchy04:29
geneon magnetic tape or something04:29
fennstl is practically useless for manufacturing gears04:29
genethen convert em in to g code04:29
kanzuregene: stl is a step backwards here04:29
kanzureplease don't04:29
kanzureno04:29
geneI know it's an arbitrary file formate04:30
fenngene just shut up04:30
kanzureybit: you check the repo yet?04:30
ybitautogenix?04:30
kanzureno no04:30
kanzurethis is recent04:30
ybitah04:30
kanzureybit: http://heybryan.org/~bbishop/docs/repo/ try one of the smaller files04:30
fennok there's this abstraction hierarchy, going from "computer, build me a thing" to machines running machine code04:30
ybitnot yet04:30
kanzureybit: go look at a file in there that won't take forever to download04:31
kanzureand that's the format that I've found out that the lab is working with04:31
kanzurewe need to improve this a bit04:31
kanzurepreferably in yaml of course04:31
fenngenes are source code that eventually gets translated into DNA and then RNA and then protein04:31
fennbut you can't store DNA in your database04:31
fennbecause it's on a hard disk04:31
geneyou can but it's tricky04:32
genethere are dna databases out there04:32
genereal meatspace dna04:32
fennyay04:32
geneon bacterial plasmids04:32
fenni'm talking about SKDB04:32
fennskdb is supposed to be easily distributed via protocols like git and http04:33
fenntherefore it cant contain a freezer full of eppendorf tubes04:33
geneyup04:33
fennor a warehouse full of gear manufacturing equipment04:33
fennso, what was the problem?04:34
fenni look at some arbitrary chunk of metal and can confidently say "this ia a 10DP 52 tooth gear"04:34
fennor some protein is firefly luciferase04:35
fennif you dont like it, learn to speak e-prime04:35
genehow do you compile it?04:37
fenncompile what?04:37
genethe gear, the luciferase04:38
kanzure:)04:38
fennwell, they are different "targets"04:38
fennone targets a gear hobbing machine and associated robots04:38
geneindeed they are04:38
fennthe other targets some dna sequencing machinery and expression vectors04:38
genea gear hobbing machine can only make gears04:39
fennnot true04:39
genewhat else can it make?04:39
fenndepends on the machine04:39
kanzureybit: anyway, what were you asking? or did I answer it already or something04:39
genea mill is more universal04:40
fennmost machine tools are just three sliding parts and a spinning part04:40
geneyup04:40
fennsometimes they have two spinning parts, these can be used as gear hobs04:40
geneunless it's one of those fancy noncartesian setups04:40
fennbut there are other problems, like how do you mount the gear, and do you have encoders on both spindles, and what tooling to use, etc04:40
genejigs04:41
genelots and lots of jigs04:41
kanzureanybody remember where I put baez.zip ?04:41
fenntry locate04:41
genejigs are pretty much the best method for automated construction04:42
fennwhat is a "jig"04:42
genea part that holds another part in place during assembly04:42
genesay a part that holds a gear and part that connects with that to align the gears onto the rod04:43
fennwhy is it i'm picturing big blocks of metal with bolts all over them instead of a robot arm04:43
geneit's sort of like that04:43
geneexcept on the end of the arm04:43
genestratasys 3d printers are used to make jigs04:44
fenngood for them04:44
* fenn is so sick of hearing about 3d printers it's not funny anymore04:44
geneMan I could really make you sick04:44
gene** printers are fairly universal04:45
fennyou could synthesize some ebola and send it to me in the mail04:45
geneNO04:45
geneI don't synthesize ebola04:45
fennmeh04:45
kanzureguess it's not universal then04:45
genebut, you could make parts to make a machine to make ebola04:46
fennif i did that, could i say the printer made ebola?04:46
fennwhat if i just pressed the green button, who did it?04:46
genesort of04:46
fennis anyone here a philosopher? please step into the chamber to your left04:47
geneyou pressed the button you caused it04:47
genebeen wanting to make a machine to click ok on EULAs for me04:47
fennwhat if we have a cat, in a box, with a gamma ray source04:47
genethe cat dies04:48
fennthere's a fifty-fifty probability the light will blink, based on whether our detector senses a gamma ray or not04:48
genenope04:48
fennif i see a blinking light i'll press the button04:48
fennat what time did the two trains pass?04:48
genegreen04:48
kanzuregod I'm confused.04:49
genewhy is a mouse when it spins?04:49
genethis topic isn't very productive04:51
geneit must be ended04:51
fennchannel rule proposal: no philosophy04:51
kanzurewhere will the turtles live?04:51
kanzuream I missing messages or something?04:52
fennno04:52
kanzurebecause those previous messages I'm seeing make very little sense04:52
genehttp://en.wikipedia.org/wiki/Jig_(tool)04:52
genejig04:52
fenndo not adjust your computer monitor04:52
fennirssi is currently experiencing some mental difficulties04:53
kanzurefenn: I was going to stop you after your first message to gene because I didn't see what you were going for04:53
geneSo you're worried about degradation with time?04:54
fennis it clear now? i can eloquate04:54
gene"a single Fanuc two-armed robot assembles smaller robots using a three-dimensional vision sensor and six force sensors that correct random positional errors."04:55
genethe answer is vision sensors 04:55
fenna three dimensional vision sensor.. i want one of those04:55
geneunfortunately04:55
geneyou have one already04:55
fennnuh-uh04:55
geneunless you wear an eyepatch04:56
kanzurestereo?04:56
geneyeah04:56
geneeyes04:56
genehttp://www.molecularassembler.com/KSRM/3.7.htm04:56
genesource04:56
geneYes I know you guys hate freitas04:57
kanzurenot true04:57
fenni dont hate freitas, but i dont blindly believe everything i read in it04:57
fennor extrapolate to absurd degree04:57
genefreitas has sticky fingers04:57
geneback to the topic04:58
genehow to replicate a camera04:58
geneor 3d vision sensor04:58
fennthat's the topic?04:58
genehow do you replicate a camera04:58
kanzuregod, I haven't been this confused since the great freenode disconnect of 200604:59
fennme personally, i havent replicated any cameras, but if i were to choose the simplest path it would probably involve a woman at some point04:59
geneheh04:59
fennand a lot of extraneous production04:59
geneI mean in a mechanical replicator04:59
fenni suppose the process could be simplified with a tank of guppies or something04:59
genehmmm...05:00
geneguppies do have the computational power of a PS205:00
fennis that so? who did you parrot that from?05:00
fennwhat is the computational power of a cup or warm tea?05:00
genenot much05:01
fennwhat is "computational power" anyway?05:01
geneSome dude trying to extrapolate to the singularity05:01
geneso how do you make a ccd05:01
fennyou take a slice of ultrapure silicon...05:02
ybitybit, you answered my question, essentially, it's a work in progress05:02
ybit"""05:02
ybit[23:19] <kanzure> and that's the format that I've found out that the lab is working with05:02
ybit[23:19] <kanzure> we need to improve this a bit05:02
ybit[23:19] <kanzure> preferably in yaml of course05:02
kanzureybit: work in progress as in, I'm too lazy to fix it *right now*. I should though.05:02
kanzureif I was more awesome I would.05:02
fennspray some photo-mask solution on it, photo-sensitize the mask, spray some etchant on it, repeat with various patterns of etchant and dopants05:03
geneby dope do you mean coat with thin film of metal?05:04
kanzureare we talking typical semiconductor manufacturing?05:04
kanzureno05:04
geneok05:04
fennno, dopants are usually salts that diffuse into the silicon05:04
geneso shoot ions at it05:04
genebtw05:04
kanzurethis is all so random05:05
geneI have always wondered05:05
* fenn blames gene05:05
kanzureI have a strong urge to go read ruby docs now.05:05
genehow do they keep the wafers aligned during manufacturing05:05
fennwith piezo stages05:05
fennand alignment marks05:05
* fenn pulls explanation out of ass05:05
genethat's bad05:05
ybitmanufacturing of CCDs is interesting05:06
* ybit listens05:06
ybitand i'm spent05:06
genepiezoes are hard to replicate05:06
fennyou know i'm just making all this up.. if you really want to know, go read a book05:06
genethey require exotic elements05:06
genethat actually sounds about right05:06
ybitkidding of course05:06
geneeven worse05:07
genethey move the whole damn mask around05:07
geneover the whole chip05:08
genehttp://en.wikipedia.org/wiki/Stepper05:09
kanzureSteve's thinking we're going to be needing an ontology of ways for parts to be in surface contact, in order to get the subcomponents in designs working correctly, for simulation and such05:09
ybitkanzure, kind of on-topic with fenn and gene, i remember you asking me what i would be doing in my research project, it's essentially ccd image reductions with iraf, nothing special. see http://iraf.net/irafdocs/#  for more details05:11
fennkanzure: you mean like, helical=nut/bolt, planar=chip/heatsink ?05:12
kanzuresort of. also glues between things, too05:12
fenna lot of that is what i call 'interfaces'05:12
kanzurei.e., given any two parts that are next to each other, WTF is their interface and how does that influence the simulation?05:12
genemaybe you should think of it in terms of force05:13
kanzurethe big example of what I'm thinking of is the car design scenario05:13
kanzuretwo sheets of metals in various shapes05:13
kanzurehow do they connect?05:13
geneKanzure, you really need to learn statics05:13
kanzurethis greatly influences CFD/aerodynamics05:13
fennit's more generic than that, you have to worry about electrical and chemical interfaces too05:13
kanzureyep05:13
geneor solids05:13
kanzureoh boy.05:13
kanzurewell, I know that there are chemical simulations05:13
kanzurebut this is highly domain dependent05:13
kanzureright?05:13
genethey have methods for describing things such as ball joints, rollers, bearings and such05:14
geneso you can calculate the forces on them05:14
fennthere are a lot of unpredictable ways that interfaces can interact05:15
kanzureSteve thinks 'geometrical computations' are the only showstoppes here05:15
kanzurehrm05:15
fennthere are also predictable ways, which i think we should focus on05:15
kanzuregive me something other than geometry that is a Big Deal05:15
genesand05:15
fennwhy is geometry a Big Deal?05:15
genesand can clog up parts05:15
kanzurefenn: two sheets of metal example05:16
kanzurehow do you specify two metals and their interface?05:16
kanzureone has more stress than another; etc. there's shape involved there. 05:16
kanzureyou don't just have it magically determine the most optimal shape or something.05:16
kanzurethere's all sorts of variables on shape.05:16
kanzureuhm05:16
genewell what are you trying to make05:16
kanzureI'm not sure if that's a good characterization o the problem05:16
kanzureSKDB05:16
fennthere's all sorts of ISO fit classes, is that helpful?05:16
genewhat needs to be made from metal?05:17
kanzurefenn: I'm trying to track down a link.05:17
fennthere should be a standard process for flexible chip fab05:19
kanzureno, I mean,05:20
fennso you can run your chip through, and they align the mems mirror array according to your cad spec05:20
kanzurecould you help me find a ref to ISO fit classes05:20
fennoh, yes, sorry05:20
genephotosensitize the chips with lasers05:20
* kanzure kicks google05:20
geneor even better a electron beam05:20
fenniso = information scarcity organization :)05:20
kanzurehttp://www.gizmology.net/nutsbolts.htm maybe05:20
kanzureheh05:20
geneyeah, Kanzure you need to learn solids05:21
genethat is solids05:21
geneyou have to do Finite element mesh to model solids correctly though05:22
kanzureFEM isn't a big deal05:22
kanzureunless I'm supposed to do it on paper05:22
genegood05:22
fennyeah the nuts/bolts thing is representative, they also have other classes for pegs in holes and so on05:22
fennall rather arbitrary if you ask me05:22
kanzureclasses 1 through 405:23
kanzureyep05:23
kanzurearbitrary.05:23
kanzurethis doesn't help much with the issues05:23
fennno05:23
kanzuremaybe I'm missing something. are there general principles for dealing with the different iso fit classes?05:24
fennit just makes it easy to write "H7" on a shop drawing05:24
fennGD&T is more general and "pure" but it's harder to read and write05:25
kanzurewrite?05:25
kanzurewhat?05:25
fennwith a pencil05:26
fennhere read this http://www.efunda.com/designstandards/gdt/introduction.cfm05:26
fenntat's kinda long but at least you'll know wtf i'm talking about05:28
kanzurehttp://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6V1P-4MNRMXW-2&_user=108429&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_version=1&_urlVersion=0&_userid=108429&md5=f138e2c93b47ea0fa213557c78658776 Flexural behaviour of bonded-bolted butt joints due to bolt looseness05:30
genewhat does the link concern05:31
kanzurehttp://heybryan.org/~bbishop/docs/manufacturing/Flexural%20behaviour%20of%20bonded-bolted%20butt%20joints%20due%20to%20bolt%20looseness.pdf05:31
genebolt looseness?05:32
geneoh shoot05:32
geneI think I found a problem05:32
genewith space replicators05:32
genemetal parts cold weld in a vacuum05:33
genenever mind not a problem05:33
genefor what I was thinking about05:34
fennmetal parts cold weld just about anywhere05:34
genebut especially in vacuums05:35
genehttp://en.wikipedia.org/wiki/Cold_welding05:35
geneyou just oxide coat your metal parts05:36
geneto avoid it05:36
geneProduct design by an advanced "expert system" software package or by humans remotely or interactively, using a computer design system that stores data on models, computes optimal designs for different options, displays results for approval, and allows efficient process iteration.05:38
geneProduction planning, an optimized plan for the manufacturing processes generated by a computer on the basis of product-design outputs, scheduling, and line-balance algorithms, and varying conditions of ore-feedstock deliveries, available robot resources, product mix and priorities. Planning includes routing, timing, work stations, and operating steps and conditions.05:38
geneguess someone already thought of SKDB05:39
genehttp://www.molecularassembler.com/KSRM/3.7.htm05:39
kanzureeverybody's *thought of* it05:39
kanzurehttp://adciv.org/05:39
kanzurehttp://peoplescapitalism.org/05:39
kanzurehttp://p2pfoundation.net/Category:Design05:39
kanzurehttp://oscomak.net/05:40
geneyeah05:40
geneI hope we actually implement it05:40
fenni wish everybody had already thought of it, then i wouldnt have to explain every damn time05:42
geneand i wish I had a particle accelerator in my dorm room05:44
fennhah - i've got one right next to me05:45
fennit's a Mazellan05:45
genewhat's a Mazellan05:46
fennHansol Mazellan 400A05:46
genenah05:47
fenni've got another one, of a different design05:47
genehow many KV05:47
genehow dense is the beam?05:48
fenna couple kV.. no fancy model name/number05:48
fennmade by Ionizair05:48
fennbeam width is about 10cm05:48
genetoo bad05:48
geneDang i though the beams on those things were smaller05:49
genelike the size of a the pixels or something05:49
fennwell, it's not circular or gaussian, it's planar05:49
fennabout 2cm in the other dimension05:49
geneis that really how big the beam is on tvs and such?05:50
fennno, this isn't a TV05:50
genethey both use beams of electrons to light up phosphors05:51
fennno, this uses beams of ion to accelerate a working fluid gas05:51
genehuh?05:51
geneplasma05:51
geneof course05:51
fennyou can barely see the plasma actually, has to be totally dark05:52
fennthen you will see it at the tips of the field concentrators05:52
geneOk I am confused05:53
fenni dont like sticking my face up that close though, too much ozone05:53
geneyour monitor makes fucking OZONE?05:54
fennit's not a monitor05:54
genewhat the heck is it?05:54
fennan air purifier, silly05:54
geneheh05:55
genethat's sort of a particle accelerator05:55
kanzure" Uh, well, if an implementation detail constraint is logically  implied by known design constraints, then coming up with an implementation that has that detail is part of what the optimizer's job is"05:55
fennactually i guess it's planar gaussian beam profile05:55
genein other words 05:56
genejust run it through a genetic algorithm05:56
fennGA has to start with something that works05:56
geneyup05:56
fennbefore GA you have to use monte carlo or something that covers the whole search spce05:57
geneyou can start with something that doesn't work05:57
genebut that takes too long05:57
fennif it's stuck in a local minimum that doesn't work, probably wont ever get out of it05:57
geneever read Prey by Michael Crichton?05:58
fennno05:58
genedang05:58
geneyou just start accepting stuff with lower fitness levels05:58
kanzurefenn: Steve is suggesting compartmentalized optimizers. So each turtle layer gets an optimizer.05:59
kanzurewith each optimizer trying to figure out the lower layer's results.05:59
fennyes that simplifies things a lot05:59
genespeaking of GAs check this out: http://www.framsticks.com/05:59
fennblack-boxing05:59
genecould you document all this stuff about turtles?06:00
geneso that I can start understanding it06:00
fennturtle just means abstraction layer06:01
fenn"it's turtles all the way down"06:01
geneabstraction layer means06:01
kanzurefenn: he doesn't do programming06:01
fennoh ffs06:01
geneyes I don't 06:01
* kanzure knows gene is thinking "fuck"06:01
kanzureyep06:01
kanzure:)06:01
geneindeed I am06:01
fennthis is sorta hard over text-based medium06:02
kanzurehttp://www.whatisthought.com/bayes2.ps06:02
geneabstraction layer you mean like concepts?06:03
kanzureuh.06:03
kanzureforget everything you know about english06:03
fennenglish is an abstraction layer06:03
geneoh so what you call the parts06:04
fenni point in the direction of a collection of hair, bones, whiskers, claws, and teeth06:04
kanzurefenn: best to have him do hard experience first before poisoning his mind06:04
fennwe call it a "cat"06:04
geneI am starting to see what you mean06:04
fenni point at a collection of subatomic particles, we call it "iron"06:04
geneI was thinking that SKDB should imitate what the brain does06:04
geneit looks like you already are06:04
fennbrain doesn't do abstraction, it does categorization06:05
geneI can't read .ps files06:05
fennsome paper about bayesian tree algorithms06:06
fennnot relevant06:06
kanzuregene: you might as well ignore it, I'm not sure of its relevance really06:06
geneok then06:06
genebut what I was thinking is that you make a program that is brain like06:07
fennwe dont know enough about how the brain works to do that06:07
genehave it read stuff online to "learn" how things work06:07
genebut we know the basics06:07
fennand if we did, it would be a bad idea anyway, because then your program would get tired and lie to you and generally have all the problems that humans do06:07
genewhy would it lie?06:08
fenn(and skdb is not yet another agi project)06:08
geneI'm talking about something with the intelligence of a cow here06:08
geneand maybe not even that06:08
kanzurewhat the fuck is intelligence06:08
fennyes let's please not go down the philosophy pit06:08
fenngene have you ever played video games?06:09
geneyes06:09
fennvideo games are full of simplified abstractions that are presented more or less directly to the player06:09
fennpress a button, door opens06:09
geneok06:09
fennthere's no fine-grained simulation of joined pieces of wood going on06:10
genethat's sort of what I am hitting at with the brain thing06:10
genefire can be made by making wood hot06:10
kanzuregene: there are neural networks that can be modularly involved with the automated design work06:10
kanzurewe're calling it "rule rewriting"06:10
kanzurewhich is just connectivity rewiring in design graphs06:10
kanzurethis is _not_ a one-to-one mapping06:10
kanzurein conceptspace.06:11
geneok06:11
fennrule rewriting?06:11
geneI am just wondering about having something to put stuff into SKDB automatically06:11
kanzurehttp://www.whatisthought.com/eric.html has some papers including "Where Genetic Algorithms Excel"06:11
fenn(i didnt download your zip file)06:11
kanzuregraph rewriting rules, sorry06:11
fennisnt a graph just a collection of connections and nodes?06:12
kanzuregiven such and such in the graph, try such-and-such06:12
kanzureyes06:12
fennso what is there to change?06:12
kanzurethese are possible mutation operators06:12
kanzurewell06:12
kanzureimagine what you'd write to go through all possible designs06:12
kanzure(much like the schedule optimizer)06:12
kanzureinstead of going through them one by one you would be making some rather large jumps06:12
fennall possible designs = an infinite set06:12
kanzurecorrect!06:12
fenn-_-06:12
kanzureso you have an upper limit, like 10 million, but the first 10 million might just be going down the left hand side of the tree06:13
geneonly some will be autocatalytic06:13
kanzurefenn: so obviously we don't want to be left-hand-travelers, right?06:13
kanzurefenn: so that's why we use some graph rewriting rules. 06:13
fenni dunno, maybe the left hand path is imbued with certain mystical powers06:13
kanzureheh :)06:13
kanzuretheoretically, there might be some rewriting rules that tend to work for certain problem domains06:13
fennwhat basis is there for thinking that?06:14
kanzurenone whatsoever, I just want to see really06:14
kanzureisn't that why we're doing turtle optimizers though?06:14
kanzuresuperturtle06:14
-!- superautomation is now known as superturtle06:14
fennwe're doing turtles because it reduces the amount of computation necessary06:14
kanzureor because we know no other way.06:15
kanzureheh06:15
fennsure, nobody has found the theory of everything06:15
superturtleright06:15
kanzureerm, wrong face06:16
fennand if we knew it, it would just be equivalent to running another instance of the universe06:16
genewe might have found the theory of everything06:16
fennstuff it06:16
kanzurehahah06:16
genebut we don't know if it is right06:16
geneand using it for automated design like this would be very impractical06:17
fenngah, gene you should change your nick to train-derailment06:17
geneok06:17
fennso, graph recombination is good for GA's for some reason, that's why we have sex and gene splicing and so on06:18
kanzuresure.06:18
kanzureoh06:18
kanzureit's not just recombination though06:18
kanzurefrom previous designs.06:18
geneit's imperfect recombination06:19
fennif i 'recombine' one element, is that any different from a random mutation?06:19
kanzuremight not be entirely random06:19
kanzurebut no, not much different06:19
fennwhat's a non-random mutation?06:19
genedo you know about meiosis fenn?06:19
fennyes06:19
genegood06:20
kanzuresuppose a subvariable has a linear improvement curve on the output of a simulation06:20
kanzureoh god, what did I just write06:20
kanzureI mean to say that it's no longer a PDE06:20
kanzureand a mutation in a certain direction offers some amount of improvement06:20
kanzureI don't know how this integrates 06:20
kanzurelet me shut up06:20
fennok so you think the algorithm is smart enough to extrapolate the curve and set the variable to maximum06:20
geneI believe that is sort of what GA's do with regards to combinationt06:20
kanzureuser/designer is the guy changing from single-mutation to GA06:20
geneDIRECTED MUTATION 06:20
geneare you fucking crazy?06:21
fennuser/designer is doing the mutation?06:21
fennrawr not enough context06:22
fenni assume this has something to do with ADL06:22
fennor DEL or whatever you call it these days06:22
kanzureno, user/designer is saying "I hate this you bitch, give me something better"06:22
geneheh why not use lamarkian evolution06:22
kanzurein other words06:22
kanzurehe's denying/affirming the rewrite rules06:22
kanzure"show me things less with rule #34"06:22
geneheh06:22
fennso the user is requesting the computer perform a random mutation?06:22
geneoh 06:22
kanzureon some set of designs with graphs06:23
genethat's easy to do06:23
kanzuresometimes, yes, fenn06:23
kanzurebut the user isn't specifying which mutation to make, really06:23
geneGAs can be quite pesky06:23
fenni just dont get what these rules are doing in the first place06:23
genethey cheat06:23
genethey take advantage of physics system glitches06:23
fennwe all do06:23
genewell I need sleep06:24
fennhe's gone! :)06:24
kanzureWTF is he06:25
kanzurerewriting the connectivity of the graph.06:26
kanzureokay, so in the HTML example06:26
fenni wish i could just download information into other peoples' brains sometimes..06:26
kanzurewhat I was going to implement.06:26
UtopiahGHMLfenn: learn pedagogy06:26
kanzureor don't06:26
kanzurejust eat the newbies for breakfast06:26
fenni'm too nice06:26
kanzureme too :(06:26
kanzureHTML example => graph was the HTML elements as nodes06:27
fennok06:27
kanzurerewrite would transform one recognized "chunk" of HTML into another chunk06:27
kanzurepredefinededly.06:27
UtopiahGHMLyou can learn manipulation instead but I think it's pretty close anyway.06:27
fenntransform one chunk into another chunk- please elaborate on that06:28
kanzurehttp://www.cs.mdc.ac.uk/research/PhDArea/saeed/paper1.pdf 06:28
kanzurefenn: <b>HELLO</b   dfieowqiq> <-- Wrong. That type of element closure could be recognized and rewritten as something else.06:28
kanzurehuh06:28
fennUnknown host www.cs.mdc.ac.uk06:28
kanzureerror correction is actually a good example06:28
kanzurebut obviously we're not interested in correcting errors06:28
kanzureerm06:28
kanzurewait06:28
kanzureerrors have unrecognizable forms usually06:28
kanzurebad example. shit.06:29
fennhtml has a syntax which defines the order characters are allowed to be arranged06:29
fennwithin that syntax you can make arbitrary rearrangements06:29
kanzures/characters/nodes/ and I'd say you have the idea06:30
kanzureerm06:30
kanzureI mean s/characters/elements/06:30
kanzurewith the understanding that elements == nodes06:30
UtopiahGHMLyou might to look at XHTML then XML then SGML if you want to understand HTML structure06:30
fennsyntax defines something like <tag element = "foo"> text </tag>06:31
kanzureBeen dong that for years, UtopiahGHML ;-)06:31
kanzurecorrect fenn.06:31
fennsyntax says nothing about what "text" should be06:31
kanzureoh, ignore text right now06:31
kanzurejust imagine that autogenerated latin BS ther06:31
kanzurethere06:31
fennlorem ipsum dolor sit amet06:31
kanzureyes, that crap06:32
kanzurein HTML, you have parent nodes and subnodes06:32
kanzure<HTML><head><title><hi I am a child of title></hi I am a child of title> ......06:32
fennand there is a grammar for those as well06:33
kanzurereword06:33
kanzureare you trying to affirm me?06:33
fennbody tag can't be child of head tag06:33
kanzuresure06:33
kanzurethat's a good constraint to mention.06:33
UtopiahGHMLmight want to look at DOM and libexpat06:34
fennok, where's the "rule" you're wanting to rewrite06:34
kanzureuh oh06:34
kanzureokay okay06:34
kanzurethat's just because of bullshit terminology06:34
kanzure*you're* making a rule *for* rewriting the graph06:34
fennis a rule just an algorithm for satisfying constraints?06:35
kanzurerule: you see a,b,c you turn it into a->b, b->c, so if I give you a randomly generated graph you look for matches for that rule to apply.06:36
kanzureerm, stupid syntax there. commas would be nonconnections, -> would be a connection. 06:36
UtopiahGHMLif you want to define rule of a well formed XML document (including XML) you might want to look at DTD06:36
kanzureuhm, it could go either way so I guess I didn't need to send that less message06:36
fennyes UtopiahGHML we understand, we're not talking about html, we're talking about engineering design graphs06:36
kanzureUtopiahGHML is right in a sort of different way06:36
kanzureDOM is relevant, sure06:36
kanzureBNF files are useful in a certain way too, but we're talking about looking at an example/sample file06:37
fennDTD is just a constraint grammar i think06:37
kanzureright06:37
kanzuresomething like that.06:37
UtopiahGHMLfenn: well to me it looks like you are talking about the underlying technical structure to support them and maybe using a lot of energy for things that doo already exist so Im just throwing suggestions in06:37
kanzurethe DOM could aide in some of this06:37
kanzureargh06:38
kanzurehow do I word how this is not DOM?06:38
kanzureI give you a graph. You see a graph. I give you a rule, "when you see this, you change it into this". You implement rule, painfully checking the whole graph. NP-hard. You cry. You finish. You didn't really check the DOM except maybe to make sure the rule wasn't retarded.06:38
UtopiahGHMLif it's an HTML document you can manipulate it with DOM (but it should be well-formed)06:38
fennDOM is totally unrelated06:39
UtopiahGHMLDOM manipulate nodes of a tree, HTML document are trees.06:39
kanzureby retarded I mean "breaking the well-form-edness constraints"06:39
kanzureget it?06:40
kanzurefenn?06:40
fennwhat is the engineering purpose of the rule? what's the desired outcome?06:41
fennis it to prevent square pegs in round holes?06:41
kanzureoh, yes06:42
kanzurebut also06:42
kanzureconsider you have a set of designs, and you have a list of the rules used to make each of them (and thus all of them)06:42
fennbecause i'd say that should be automatic, and why would the user be modifying a rule that prevented something obviously stupid?06:42
kanzureand you figure you don't like the results of a particular rule application06:42
kanzureso you just eliminate all results using that rule06:42
UtopiahGHMLbecause the user doesn't understand a software or model like the author indented it to be understood...06:43
kanzurethen user preferences might develop for certain rules. would be worth keeping track of, no? then you can switch between exploration v. exploitation (SA v. GA) with the 'knob' mechanism I was mentioning. this would be for selecting certain designs06:43
fennthere is no author06:43
kanzurekind of for going through the many, many results06:43
UtopiahGHMLoh06:43
UtopiahGHMLno author :)06:43
kanzurefiltering over possibility space, I guess06:43
fennunless you mean the guy who made the parts you're currently manipulating06:44
kanzurejust one? poor guy :)06:44
fennyeah :(06:44
kanzureleft hand path06:44
fenntoo many cooks spoils the tag-soup06:44
UtopiahGHMLif you invite a user to use a system, you are the author of the system06:44
UtopiahGHMLif your system provide affordance to your model06:44
UtopiahGHMLyou are the author of the underlying model06:44
UtopiahGHMLso there is an author06:44
fenni guess06:45
UtopiahGHML(you could even consider a genetic algorythm an author in some sort of way)06:45
fennthat's what i mean by no author06:45
UtopiahGHMLto me the only option to "no author" is evolution thus evolutionary algo06:45
fennthere's also brute force06:45
fennor some other "fuzzy" learning algorithm06:46
UtopiahGHMLbrute force is evo algo with no return and selection only by final sucess06:46
UtopiahGHML1 step mutation/selection06:46
fennnot necessarily06:46
kanzurethat's what I say SA is06:47
fennbrute force can be systematic06:47
kanzurehad a paper ref06:47
kanzureSAGA_hybrids.pdf06:47
UtopiahGHMLsystematic or random you just build a space of solutions and try them with selection being only work,do not work06:47
fennimagine you're trying to cover a landscape in marbles, you can roll marbles into depressions in the landscape (GA) or glue them randomly, or glue a grid of marbles on  the landscape06:48
fennwhat does SA stand for?06:49
kanzuresimulated annealing06:49
fennhrm.06:49
fennrequires continuous changes to be good06:50
fennotherwise it might jump out of the hole06:50
fennor is that the point06:51
kanzureneed a little mutation to get evolution out of the gutters06:51
fennwhere's my magic quantum computer that can just give me the optimum answer06:52
fennhey, you know, in reality we want to support standards, which typically involve discrete choices ie M5 screw M6 screw etc06:52
fennso that reduces the search space drastically06:53
fennso i still dont know what "rules" are06:54
kanzure"if you find this in the graph, do this"06:54
kanzurehttp://heybryan.org/~bbishop/docs/rules/ check one of these files06:54
fennthat's just a substitution system06:54
kanzureyes06:54
fennthat's as bad as cellular automata06:55
kanzure(the files have a "left hand side" and a "right hand side")06:55
kanzureeh?06:55
fenni.e. unpredictable output06:55
kanzurerules are constrained by those standards though06:55
kanzurethat you just mentioned.06:55
kanzureactually06:55
kanzureit might just be an equivalency06:55
kanzurethey are standards for a reason06:58
fennmost standards are proven designs06:58
fennbut there is some arbitrariness, like what diameter is the screw06:58
kanzureok, so some rules are bendable06:58
* kanzure recalls scenes from the matrix06:58
fenni mean why 5mm why not 5.1mm06:59
kanzureyes, I know06:59
fennbut the standard only defines 5mm screws06:59
kanzuresay you had to have 5.1 mm . that's a breakable rule methinks.06:59
kanzurewell that's probably a stupid standard anyway06:59
kanzuresize, jeesh06:59
fennhowever you can make a 5.1mm screw some of the rules for defining a 5mm screw (just change the diameter value)07:00
kanzurelets do a more important thing in design space for standardization, I dunno, something that actually helps07:00
kanzure5 mm screws help of course, when buying parts07:00
kanzurethat's probably what it is.07:00
fennvery much so07:00
fennused to be you had to call up the local blacksmith to make each screw since no two were alike07:00
kanzureuh, how long ago?07:00
fennnot that long ago :(07:01
fenn1910?07:01
kanzureshouldn't have asked.07:01
fennoptical comparator revolutionized screw making because it introduced some GD&T-like principles07:01
fennso a screw doesn't have to be exactly 5mm, as long as it falls within a certain range of shapes (optical comparator is because helical shapes are hard to measure with calipers)07:02
kanzureright07:16
kanzurefenn gets it yet?07:22
ybitfigured p.36 of http://web.mit.edu/endy/www/scraps/talks/03.06/EndyMarch2006.pdf might help gene07:26
ybitmaybe i will remember later07:26
ybitand the 31m mark in http://video.google.com/videoplay?docid=-6950604815683841321&ei=li7fSO-PKZycrALHy9igCw&q=synthetic+biology+drew+endy 07:28
kanzureeffn07:31
kanzuredon't get me wrong, substitution is a complementary method to everything else (like bruteforce)07:31
kanzurebut to some extent it's somewhat equivalent07:31
kanzureI'd want to hash out the code for bruteforce first to see/show etc.07:31
kanzurehm, weird typo on 'fenn'07:32
kanzurewhy'd I type?07:32
UtopiahGHML[tab] don't type07:33
kanzureright07:33
UtopiahGHMLactually07:33
UtopiahGHMLcomment on http://seedea.free.fr/wiki/pmwiki.php?n=Oimp.Ubiquitousvocabulary please07:34
kanzurewhere'd my log analysis scripts go?07:35
kanzureI had 100+ MB of analyzed logs from my many years of chatting07:35
kanzuretechnically the information is still in my mysqldb :)07:35
* ybit takes a few capsules of melatonin and calls it a night07:36
kanzuredictionary based tab completion, eh? 07:36
ybitgoodnight07:36
UtopiahGHMLnight07:36
UtopiahGHMLkanzure: 100MB of raw data or of resulting analysis?!07:36
kanzureraw data07:36
UtopiahGHMLthat's different ;)07:36
kanzure/me was working on auto grammar expansion stuff a while back (conceptually) .. would be nice to have computer assisted grammar / sentence construction for conveying sufficient meaning/detail07:37
kanzurethe analysis was like 300 MB or something07:37
kanzurelots of stupid crazy overhead07:37
kanzuretimestamp on each and every word07:37
kanzure(not connectivity though, why? mostly because I took the code from another guy)07:37
UtopiahGHMLwell the point if to use [tab] all anytime anywhere instead of typing07:38
UtopiahGHMLs/if/is/07:38
kanzurehave you seen the japanese text input interfaces for cell phones?07:38
kanzureI hear it's kind of like tab completion except more gui07:38
UtopiahGHMLyep, chinese actually but I guess it's the same07:38
UtopiahGHMLIve spend 6 months in Shanghai07:39
UtopiahGHMLdidn't think about it actually, thanks :P07:39
UtopiahGHML(duh)07:39
UtopiahGHMLthey basically phonetically type and tab cycle amonst the different possibilities based on the word frequencies07:40
UtopiahGHMLsound very similar indeed07:41
kanzuresentence construction help next please.07:41
UtopiahGHMLthat's why I love to share ideas, you always have people looking at it with a different perspective07:41
kanzureor people sufficiently annoyed with the problems.07:41
UtopiahGHMLthey basically phonetically-type and tab-cycle amonst the different possibilities (possibilities ordererd by the frequency of usage of each word)07:43
kanzuretheoretically that's how standards should be crystalizing07:45
kanzurebut anyway.07:45
kanzuresleep. g'night07:45
UtopiahGHML++07:45
superturtleuh15:56
superturtlecan anyone explain why precisely my laptop is off?15:56
UtopiahGHMLsounds like a job for the X-files guys15:59
superturtleI suspect the fans.16:07
superturtle48585. This delay has set it back.16:19
kanzureyep, no fan movement16:58
kanzurehttp://forums.somethingawful.com/showthread.php?threadid=296064117:51
kanzuresomethingawful feels like the missing link in my internet ancestry17:51
kanzureinfopop brings me back.17:51
kanzure*takes me back17:51
kanzurewhich one is it?17:51
bkeroI'm a goon.17:52
kanzurepatrick sucks18:04
kanzureon openmanufacturing18:04
superturtleblah20:48
superturtlewould anyone be interested in accounts to the file server?22:21
bkerofiles?22:45
superturtle300 GB collection of stuff.22:49

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