--- Day changed Tue Jun 10 2008
00:02 < fenn> its not even up to date.. kurzweil already demonstrated his 'pocket sized reading machine for the blind'
00:04 < kanzure> it's very focused on Kurzweil
00:04 < fenn> kurzweil is amazingly accurate
00:04 < kanzure> I wonder why Kurzweil is so popular
00:04 < kanzure> really? he completely ignores so much though
00:06 < fenn> clarke has some really weird predictions
00:06 < kanzure> '
00:06 < kanzure> Immediately forget any preconceptions you may have about Salman Rushdie and the controversy that has swirled around his million-dollar head. You should instead know that he is one of the best contemporary writers of fables and parables, from any culture. Haroun and the Sea of Stories is a delightful tale about a storyteller who loses his skill and a struggle against mysterious forces attempting to block the seas of inspiration fro
00:06 < fenn> Largest living creature filmed: a 75-metre octopus in the Mariana Trench. [Clarke99]
00:07 < kanzure> :)
00:07 < fenn> compared to something you might actually expect a futurist to say, like "2012: billion digital video cameras posting online realtime, personal privacy is history"
00:09 < kanzure> I don't know about the 'idle predictions' thing though
00:09 < kanzure> "the best way to predict the future is to create it"
00:09 < kanzure> I'm not feeling good recently
00:09 < kanzure> have not had enough time to express all that I really should be
00:09 < kanzure> for example, it may appear that I do not know about GPL licensing
00:09 < kanzure> but that's really just my laziness showing
00:10 < kanzure> too much time in the lab probably :) I get more work done juist sitting alone in five minutes.
00:13  * kanzure needs to go read up on the latest in filesharing
00:13 < fenn> look at distributed hash table on wikipedia
00:14 < kanzure> I mean actual networks :)
00:15 < kanzure> hurray for lists
00:15 < kanzure> http://www.zeropaid.com/bbs/showthread.php?t=38811
00:15 < fenn> coral cache isnt an actual network?
00:16 < kanzure> http://www.coralcdn.org/
00:16 < kanzure> ?
00:16 < fenn> nevermind
00:17 < kanzure> ?
00:18 < fenn> i thought you were talking about the implementation, not just what's available and popular at the moment
00:22 < kanzure> I'm either shutting down or completely confused now.
00:22 < kanzure> re: the lab work, I think I'm going to possibly insert the complaint that since they don't know how to do the manufacturing, it's all bunk
00:22 < kanzure> unless they want to propose the crazy idea of doing the chemistry from scratch
00:22 < kanzure> which is a very big task that I am not too sure about ... wouldn't somebody have done this by now ? heh'
00:23 < fenn> reinventing biochem? why bother
00:23 < fenn> it already works
00:24 < kanzure> I'm not too certain about the protein engineering stuff
00:24 < kanzure> and again
00:24 < kanzure> you mention that the writozyme (or just general easy DNA synthesizer) proliferation might help that out
00:24 < kanzure> but I'm not quite sure how, other than letting lots of people tinker and try new stuff out :)
00:24 < fenn> yep, pretty much
00:24 < kanzure> I suppose it has to be ultimately tractable
00:24 < kanzure> protein folding, I mean.
00:26 < fenn> i think a system of legos could be built, even using standard biotech tools
00:26 < kanzure> http://heybryan.org/mediawiki/index.php/Wittig_cohort#2008-06-09 <-- meeting with Zack, the guy who shows up and likes amorphous computing stuff
00:26 < kanzure> he seems to be fairly knowledgable about the general space of stuff that this involves
00:26 < kanzure> and isn't exactly a biochemist-specialist, so he looks like he's from the software side
00:26 < fenn> but if you have to do some nobel-prize-winning effort every time you want to build something with those lego's, its no good
00:27 < kanzure> system of legos? so, a new biochemistry though?
00:27 < kanzure> I mean, that sounds nasteh
00:27 < fenn> no
00:27 < fenn> i mean sticking protein blobs together
00:27 < kanzure> oh
00:27 < kanzure> puzzle pieces?
00:27 < fenn> right
00:27 < kanzure> or lego
00:27 < kanzure> right
00:27 < kanzure> heh'
00:27 < fenn> not artificial amino acids or whatever
00:28 < kanzure> I see
00:28 < kanzure> that's structure, not necessarily function
00:28 < fenn> you can have functional subunits though
00:28 < fenn> turn them into engineering blackboxes
00:28 < fenn> with levers and buttons
00:28 < kanzure> okay
00:29 < kanzure> I guess I just have to get over the fact that coming up with new blackboxes isn't going to be the same as writing a new function
00:29 < kanzure> (in terms of code)
00:29 < kanzure> unless we solve some other problems re: protein folding of course
00:30 < fenn> if my humble opinion is worth anything, i dont think compartmentalization is going to go much anywhere
00:30 < kanzure> how are you going to get evolution? 
00:31 < kanzure> the idea is to get "genotype and phenotype in the same place"
00:31 < kanzure> one way that this might happen is Mike's, uh, "RNA polymerase + biotin + attach it to the strand of RNA" - which isn't a compartment, but it's functionally equivalent
00:31 < fenn> yes, perhaps you'll demonstrate a "DNA computer" but it'll be slow and require external 2d patterning
00:31 < kanzure> attaching the two components together.
00:32 < fenn> eh, i'm talking past you i think. sorry
00:33 < kanzure> it might be, I'm falling asleep
00:33 < kanzure> maybe you mean the problem with compartments and how you're just creating lots of problems
00:33 < kanzure> like communication barrier stuff
00:33 < kanzure> I've been listening to these guys boast about emulsions
00:33 < kanzure> and it sounds like painfully, retardingly intense work
00:33 < kanzure> for minimal returns that are barely related to the ideals of amorphous fabrication and their other dreamgoals
00:34 < fenn> did my comment about wifi networks and noise threshold make sense?
00:35 < kanzure> no, unless you meant that only a certain number of the packets (signal) get to the more distant wifi nodes and thus those get to be counted as noise
00:35 < fenn> yes, that's what i meant
00:35 < kanzure> I'm guessing it's dependent on 802.11b
00:35 < kanzure> so, ok
00:35 < fenn> but instead of a bunch of houses with transmitters on the roof, you have a microarray with transcriptional switch genes
00:35 < kanzure> right
00:36 < kanzure> this scenario has shown up a number of times
00:36 < kanzure> it's even what I recommended a few months ago
00:36 < fenn> so, you can re-use the same toehold sequence if they're far enough apart
00:36 < kanzure> http://heybryan.org/mediawiki/index.php/In_vitro_liquidwars
00:36 < fenn> but you lose the whole amorphous part
00:36 < kanzure> aha!
00:36 < kanzure> you see
00:36 < kanzure> they don't agree with me on that
00:36 < kanzure> they tell me that it's ok for it to be attached
00:36 < kanzure> that "everything has to do symmetry breaking"
00:37 < kanzure> but I keep pointing out that it's not amorphous if it's attached
00:37 < fenn> well, its certainly easier than building a chip fab
00:37 < fenn> but not amorphous
00:37 < kanzure> the idea is that you have *both* (1) you don't know their location, and (2) NEITHER DO THEY.
00:37 < kanzure> except relative concentration gradients and so on
00:37 < kanzure> and whatever established relative axises between each other etc.
00:37 < fenn> it doesnt matter if you dont know the location, the point is you have to pattern this microarray thingy to do anything
00:38 < kanzure> right, I don't like the prospects of losing the biological advantage
00:39 < fenn> heh you could make a DNA origami array :)
00:40 < kanzure> ruthemund's work involves self-assembling structures based off of a scaffolding system, so that's close, I guess
00:40 < kanzure> I wonder if building stuff from DNA like that is worth it?
00:40 < kanzure> http://dna.caltech.edu/
00:40  * fenn pats self on the back
00:40 < kanzure> eh?
00:40 < fenn> 100% DNA computer
00:40 < kanzure> well, are you trying to lead back to the DNA FPGA idea?
00:40 < fenn> you heard it here first
00:41 < fenn> sorta.. but instead of using semiconductors you use transcriptional switches
00:41 < fenn> and that way the complexity is already programmed in
00:41 < fenn> the problem of course is that you can't reprogram it
00:41 < kanzure> yeah ..
00:41 < kanzure> I'm still not quite convinced
00:41 < kanzure> DNA is not the best substrate for much
00:41 < fenn> cant work around defects in tile crystal growth
00:41 < kanzure> I mean, it degrades in all sorts of weird shit
00:42 < kanzure> there are ways to do "checks" on tile growth
00:42 < fenn> DNA doesnt normally degrade
00:42 < kanzure> that Ruthemund has apparently implemented
00:42 < kanzure> eh, I guess I just need to give up on the idea of semiconductors being so solid state
00:42 < kanzure> but I guess I could argue something about DNA having lots of repair enzymes to help fix it, does that count?
00:42 < fenn> DNA is not a semiconductor (yet)
00:42 < kanzure> you think that could change ? how ?
00:42 < fenn> well, i read it was superconducting somewhere.. that sorta blew my mind
00:43 < kanzure> hrm
00:44 -!- ybit [n=h@unaffiliated/ybit] has quit [Read error: 110 (Connection timed out)]
00:44 < fenn> at 1 kelvin
00:44 < kanzure> at 1 kelvin
00:44 < kanzure> heh'
00:44 < fenn> yeah :(
00:45 < kanzure> I'm sure we could throw in some salts or phase change stuff
00:46 < fenn> http://www.physics.ucla.edu/research/biophysics/news/pdf/outlook.pdf
00:47 < fenn> first time i've seen a retraction of prior unlikely claims in the new scientist
01:05 -!- fenn_ [n=pz@adsl-75-60-173-55.dsl.bltnin.sbcglobal.net] has joined #hplusroadmap
01:05 -!- Topic for #hplusroadmap: Intro: http://www.youtube.com/watch?v=ZXKbzbeipmI http://diybio.org/ http://openwetware.org/ | diy bio toolkit: http://biohack.sf.net/ | Automated societal knowledge (put it to work): http://heybryan.org/exp.html | Channel wiki: http://heybryan.org/mediawiki/ | F/OSS perspectives on Kurzweil: http://heybryan.org/fernhout/
01:05 -!- Topic set by kanzure [] [Sun Jun  8 13:00:15 2008]
01:05 [Users #hplusroadmap]
01:05 [ fenn ] [ kanzure] [ parodyoflanguage] [ pupnik ] [ wrldpc] 
01:05 [ fenn_] [ nsh    ] [ procto          ] [ Splicer] 
01:05 -!- Irssi: #hplusroadmap: Total of 9 nicks [0 ops, 0 halfops, 0 voices, 9 normal]
01:05 -!- Channel #hplusroadmap created Sat Mar 22 15:44:12 2008
01:05 < kanzure> fenn_: you still need a surface with your dna assemblers
01:06 -!- Irssi: Join to #hplusroadmap was synced in 37 secs
01:09 < kanzure> why do the Wachowski brothers spell 'woah' as 'whoa' ?
01:11 -!- kanzure [n=bryan@cpe-70-113-54-112.austin.res.rr.com] has quit ["Leaving."]
01:13 -!- ybit [n=h@unaffiliated/ybit] has joined #hplusroadmap
01:16 -!- ybit [n=h@unaffiliated/ybit] has quit [Remote closed the connection]
01:18 -!- ybit [n=h@unaffiliated/ybit] has joined #hplusroadmap
01:18 -!- ybit [n=h@unaffiliated/ybit] has quit [Remote closed the connection]
01:18 -!- fenn [n=pz@adsl-75-60-172-87.dsl.bltnin.sbcglobal.net] has quit [Read error: 110 (Connection timed out)]
01:19 -!- parodyoflanguage [i=pseudony@mmds-216-19-34-118.twm.az.commspeed.net] has quit [Read error: 110 (Connection timed out)]
01:19 -!- parodyoflanguage [i=pseudony@mmds-216-19-34-118.twm.az.commspeed.net] has joined #hplusroadmap
01:28 -!- You're now known as fenn
01:45 -!- ybit [n=h@unaffiliated/ybit] has joined #hplusroadmap
02:01 < ybit> internet connection issues all day while i was at work it seems
02:01 < ybit> so what'd i miss? :)
02:07 -!- parodyoflanguage [i=pseudony@mmds-216-19-34-118.twm.az.commspeed.net] has quit [Read error: 113 (No route to host)]
02:33 -!- ybit [n=h@unaffiliated/ybit] has quit [Remote closed the connection]
02:36 -!- ybit [n=h@unaffiliated/ybit] has joined #hplusroadmap
03:23 -!- ybit [n=h@unaffiliated/ybit] has quit [Connection timed out]
03:43 -!- Phreedom [n=freedom@www.online.dn.ua] has joined #hplusroadmap
05:22 < wrldpc> fnord
06:12 -!- nsh [n=nsh@wikipedia/nsh] has quit [Read error: 110 (Connection timed out)]
06:54 -!- nsh [n=nsh@eduroam-80.uta.fi] has joined #hplusroadmap
07:42 -!- Splicer [n=p@h71n3c1o261.bredband.skanova.com] has quit []
07:51 -!- Vedestin [n=Vedestin@d122-109-35-58.sbr3.nsw.optusnet.com.au] has joined #hplusroadmap
08:48 -!- Vedestin [n=Vedestin@d122-109-35-58.sbr3.nsw.optusnet.com.au] has quit ["User pushed the X - because it's Xtra, baby"]
08:50 -!- Vedestin [n=Vedestin@d122-109-35-58.sbr3.nsw.optusnet.com.au] has joined #hplusroadmap
10:09 -!- Vedestin [n=Vedestin@d122-109-35-58.sbr3.nsw.optusnet.com.au] has quit ["User pushed the X - because it's Xtra, baby"]
10:50 -!- kanzure [i=bryan@dhcp-146-6-213-183.icmb.utexas.edu] has joined #hplusroadmap
10:59 -!- faceface [n=chatzill@bioinformatics.org] has joined #hplusroadmap
10:59 < faceface> anyone seen a guy called don in here?
11:00 < faceface> don / mindreck 
11:09 < kanzure> Nope.
13:15 -!- pupnik_ [n=pupnik@dslb-088-068-117-029.pools.arcor-ip.net] has joined #hplusroadmap
13:26 -!- pupnik [n=pupnik@unaffiliated/pupnik] has quit [Read error: 110 (Connection timed out)]
13:49 -!- cis-action [n=cis-acti@c-66-31-201-16.hsd1.ma.comcast.net] has joined #hplusroadmap
13:56 -!- kanzure [i=bryan@dhcp-146-6-213-183.icmb.utexas.edu] has quit [Connection timed out]
17:02 < cis-action> ping
17:03 -!- cis-action [n=cis-acti@c-66-31-201-16.hsd1.ma.comcast.net] has quit []
17:03 -!- cis-action [n=cis-acti@c-66-31-201-16.hsd1.ma.comcast.net] has joined #hplusroadmap
18:14  * procto is at the cambridge semantic web gathering
18:20 -!- Splicer [n=p@h176n2c1o261.bredband.skanova.com] has joined #hplusroadmap
19:44 -!- kanzure [n=bryan@cpe-70-113-54-112.austin.res.rr.com] has joined #hplusroadmap
19:45 < kanzure> Hey all.
19:45  * kanzure got his radiation training done today.
19:45 < kanzure> The whole "don't be a terrorist" thing.
20:16 -!- cis-action [n=cis-acti@c-66-31-201-16.hsd1.ma.comcast.net] has quit []
20:35 < kanzure> fenn: So I put some thought into it, and I think I can get their 'amorphous fabrication' goals via ribosome+polymerase+switches all hopelessly tied together. However, it requires the same writozyme/polymerase ideas. The good news is that ribosomes have been very well characterized.
20:41  * kanzure feels a stroke of evil genius coming along
20:42 < kanzure> I just found a paper that explains how to couple mRNA, ribosome, and the protein output together. 
20:42 < kanzure> i.e., chemically/physically
20:46 < Splicer> are you talking about a bacterial amorphous computer project? 
20:49 < kanzure> "coupled transcription-translation"
20:49 < kanzure> Yes.
20:49 < kanzure> But it's not bacterial.
20:50 < kanzure> In the lab we're doing transcriptional switches that we hope will one day ("a very long time from now") can lead up to amorphous computation+fabrication (an intersection of the two). The writozyme idea and some others kind a jump the gun and say "screw a long time, let's do it now."
20:52 < Splicer> how would a transcriptional switch work?
20:53 < kanzure> Splicer: Imagine a DNA fragment. A polymerase would bind, during a run through a PCR protocol, and copy a portion which would then go off to promote the transcription of another part of the switch, which would then ... :)
20:54 < kanzure> Splicer: http://heybryan.org/mediawiki/index.php/Wittig_cohort <-- notes from the cohort.
20:55 < Splicer> checking
21:08 < kanzure> uh, well, technically we don't necessarily want polymerase bound to the ribosome display techniques
21:08 < kanzure> simply a transcriptase
21:08 < kanzure> DNA -> mRNA converter
21:09 < kanzure> so, RNA polymerase.
21:19 < Splicer> but how is the switching done?... I mean the polymerase follows the singlestranded DNA so either that has to contain the blueprint or the polymerase has to be modified? 
21:20 < kanzure> The polymerase copies a part of the DNA molecule which then goes off to signal a "neighboring" (one that matches) strand.
21:21 < Splicer> you mean you steer what gene to read?
21:22 < kanzure> There are inhibition regions and promoter regions. So what if the promoter region is blocked? Suddenly the strand doesn't transcribe. :)
21:22 < kanzure> (the small signaler portion, at least)
21:24 < Splicer> the promroter/inhibitor regions are gene wide?
21:26 < kanzure> What?
21:26 < Splicer> mmm... i have to read up on this... i thought they regulated the whole gene, so the mRNA is a separate piece ending in the STOP.
21:27 < Splicer> wait.. you are building a single logic gate?
21:28 < Splicer> not a program
21:28 < kanzure> It's just an oscillator.
21:28 < kanzure> but that's basically a flip flop, if you can get some hooks into it and so on
21:28 < kanzure> and from flip flops you can build gates
21:28 < Splicer> ah.. cool
21:28 < kanzure> well
21:28 < kanzure> no
21:29 < Splicer> fuck... i have a lot to learn
21:29 < kanzure> http://en.wikipedia.org/wiki/NOT_gate :)
21:29 < Splicer> shut up
21:30 < Splicer> hehehe
21:30 < kanzure> not sure how to implement NOT on transcriptional switches really
21:30 < kanzure> in fact, I'm still working through the entire scheme myself
21:30 < kanzure> to see what we actually have
21:31 < kanzure> it's kind of weird that the other students that I am working with seem to have it nailed down, but when I actually try to ask them about any of the computational aspects or the meaning of any of the certain aspects, they just kind of go walk off :-p
21:31 < Splicer> wans't there an iGEM team who built an oscilator... it blinked 10 times i think
21:31 < kanzure> Maybe.
21:32 < Splicer> maybe cause you are a high profile guy
21:32 < Splicer> ( http://openwetware.org/wiki/IGEM:IMPERIAL/2006/project/Oscillator )
21:32 < kanzure> I think it's because I know the grad student is a comp sci guy, I'd like to think I know how these guys work. We get along pretty well. But then they shush me when they think he's mad (just getting excited about a certain aspect of a system, really), etc.
21:34 < Splicer> you do get excited
21:36 < kanzure> No, not me, him. 
21:36 < Splicer> you meant mad mad?
21:37 < Splicer> like nuts?
22:10 < kanzure> No.
22:11 < kanzure> I don't know how to characterize it. I'll get back to you on that.
22:41 < procto> kanzure: re the cybernetic view of models: http://pespmc1.vub.ac.be/MODEL.html
22:41 < kanzure> Principia Cybernetica gets *close* to some pretty awesome insights.
22:42 < procto> more stuff here: http://pespmc1.vub.ac.be/EPISTEM.html
22:42 < procto> and on the sidebar on the right
22:42 < procto> yes, certainly
22:42 < procto> I'm trying to remember where I saw some other really interesting stuff on cybernetic modelling but it's elluding me
22:43 < procto> soon as I remember a keyword other than "cybernetics" and "world models" i'll google it
22:57 < procto> I'm not sure I'm totally with the cybernetic epistemology, though
22:59 < kanzure> How so?
23:06 < procto> you could say i'm epistemologically agnostic, but not because I don't want to commit, but rather because I don't feel I know enough as yet to properly defend a position to myself
23:06 < procto> at the same time, I rather like the cybernetic framewolk
23:06 < procto> framework*
23:07 < procto> much od my conception of epistemology I derive from decision theoretic semantics, which is a sort of meta-semantic theory. that is, it subsumes within it referential and possible world models
23:07 < procto> and utilizes decision theoretic and game theoretic techniques
23:08 < procto> claiming that this models the way humans decide on consensus meaning
23:08 < procto> i.e. a meaning of an utterance is ascribed on the basis of maximizing utility
23:09 < procto> and in a conversation, some sort of game theoretic approach may be used to decide on the equilibrium point, which is the then decided upon meaning
23:10 < kanzure> who cares if it models 'consensus' ?
23:10 < procto> some prefer a cooperative approach. i.e. in your head, you are modeling the decision problems of the person your are speaking to, and thus an utterance is an answer to a hypothetical question one might ask to increase the utility of that model person in your head
23:11 < procto> well, by "consensus" I mean a situation where in the process of an interaction, clashes are minimized
23:11 < procto> myself, I prefer an adversairal approach
23:11 < procto> adversarial*
23:12 < procto> if we are talking together about how much we love mashed potatoes in standard informal english, we have little to no misunderstanding
23:13 < procto> and you can model the way in which we would both ascribe the word "potate" to the same-ish category of tuber in the decision theoretic semantics framework
23:13 < procto> potato*
23:14 < fenn> mashed potatoes are awesome
23:14 < fenn> i'm glad you guys can udnerstand
23:15  * Splicer wonders if nsh speaks finnish
23:17 < procto> kanzure: although the epistemology derived from this is quite similar to the cybernetic one, the difference is that the cybernetic framework indicates that the internal representation is created through interaction with the environment
23:19 < procto> kanzure: whereas I currently prefer a framework where the internal representation is created by internal evaluation of the environment in terms of relative value
23:20 < procto> the difference is subtle and i have not plumbed its depths. it could be a non-existant one, since my feeling of distinction could be wrong
23:22 < procto> perhaps to illustrate what I percieve to be the difference I can use an example
23:22 < procto> take Searle's Chinese Room (I am assuming familiarity)
23:22 < procto> searle's argument is that it lacks semantics, that is, meaning
23:24 < procto> hmmm
23:24 < procto> actually
23:24 < procto> scratch all that
23:24 < procto> (I am expressing ideas and such that are relatively nebulous in my head, so revision once they are regurgitated is a welcome side effect)
23:25 < procto> it seems that i was anthropomorphizing systems
23:25 < procto> i.e. robbing non-value laden systems such as searle's chinese room of the ability to practice an epistemological process
23:27 < procto> that is, "if it doesn't have a mind it can't know things"
23:27 < procto> but cybernetics says nothing of minds
23:28 < procto> (my initial thurst was, rephrased, that minds are born when a cybernetic system becomes value-laden)
23:29 < procto> definition of value-laden: able to ascribe utility to "stuff"
23:29 < procto> (this utility can be ordinal or cardinal, not picking a system)
23:31  * procto takes a deep breath
23:31 < procto> was my train of thought followable? (it's ok if there was no desire to follow it :>)
23:45 < procto> sleep times now