--- Log opened Mon Jun 03 00:00:45 2013
00:29 -!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has joined ##hplusroadmap
00:34 -!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has quit [Ping timeout: 260 seconds]
01:36 -!- superkuh [~superkuh@unaffiliated/superkuh] has joined ##hplusroadmap
01:54 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap
01:57 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds]
03:02 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap
03:05 -!- Adifex [~Adifex@97-122-125-109.hlrn.qwest.net] has joined ##hplusroadmap
03:10 -!- quantumkat [~kat@ip70-171-6-179.ga.at.cox.net] has joined ##hplusroadmap
03:13 -!- cogitokat [~kat@ip70-171-6-179.ga.at.cox.net] has quit [Remote host closed the connection]
03:13 -!- devrandom [~devrandom@50-0-206-254.dsl.dynamic.sonic.net] has quit [Remote host closed the connection]
03:13 -!- makoLime [~mako@103-9-42-133.flip.co.nz] has quit [Ping timeout: 276 seconds]
03:13 -!- makoLime [~mako@103-9-42-133.flip.co.nz] has joined ##hplusroadmap
03:18 -!- devrandom [~devrandom@50-0-206-254.dsl.dynamic.sonic.net] has joined ##hplusroadmap
03:41 -!- EnLilaSko [EnLilaSko@unaffiliated/enlilasko] has joined ##hplusroadmap
03:47 -!- FooQuuxman [~test@c-98-215-254-55.hsd1.il.comcast.net] has joined ##hplusroadmap
03:49 -!- makoLime [~mako@103-9-42-133.flip.co.nz] has quit [Ping timeout: 248 seconds]
04:22 -!- ThomasEgi [~thomas@185.5.8.81] has joined ##hplusroadmap
04:22 -!- ThomasEgi [~thomas@185.5.8.81] has quit [Changing host]
04:22 -!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap
04:31 -!- yorick [~yorick@oftn/member/yorick] has joined ##hplusroadmap
04:58 -!- Adifex [~Adifex@97-122-125-109.hlrn.qwest.net] has quit [Quit: Adifex]
05:51 -!- ielo [~ielo@184.88.56.168] has joined ##hplusroadmap
05:58 -!- ielo [~ielo@184.88.56.168] has quit [Ping timeout: 264 seconds]
06:06 -!- yorick [~yorick@oftn/member/yorick] has quit [Remote host closed the connection]
06:34 -!- ielo [~ielo@184.88.56.168] has joined ##hplusroadmap
06:44 -!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Remote host closed the connection]
06:56 -!- ielo [~ielo@184.88.56.168] has quit [Ping timeout: 260 seconds]
07:54 -!- FooQuuxman [~test@c-98-215-254-55.hsd1.il.comcast.net] has quit [Ping timeout: 260 seconds]
07:57 -!- kajetan [122@unaffiliated/kmo] has quit [Quit: Leaving]
08:02 -!- Urchin [~user@unaffiliated/urchin] has joined ##hplusroadmap
08:08 -!- AshleyWaffle [~quassel@unaffiliated/anastasiawyatt] has joined ##hplusroadmap
08:27 < kanzure> "injecting malware into iOS devices via malicious chargers" http://www.blackhat.com/us-13/briefings.html#Lau
08:30 -!- FooQuuxman [~test@c-98-215-254-55.hsd1.il.comcast.net] has joined ##hplusroadmap
08:31 < kanzure> "Methuselah Foundation could establish a grant specifically for the purpose of empirically exploring the effectiveness of characterization of a particular individual's (ie Steve Coles) cancer tumor genome used for the purpose of developing and administering targeted treatment based on said genetic characterization. If list members are interested in doing this, then Methuselah would match donations 1 for 1 (ie .50 per dollar) up to $5,000 (to ...
08:31 < kanzure> ... achieve the 10,000 target). The match would not happen until funds sufficient to achieve the target had been donated. There would need to be a physician or researcher willing to document/write a paper/transactional letter for submission to the foundation and to a journal (such as Rejuvenation Research) or published on the GRG list. There would additionally need to be submitted a formal grant request specifying and stipulating that the use ...
08:32 < kanzure> ... of the grant funds would be applied to the intended purpose. This request would be short - appx one page."
08:32 < kanzure> bizarre way to go about fundraising for such a small amount of money.
08:33 < kanzure> this seems like missing the point:
08:33 < kanzure> https://www.google.com/webmasters/tools/disavow-links-main
08:33 < kanzure> "If you believe your site's ranking is being harmed by low-quality links you do not control, you can ask Google not to take them into account when assessing your site. You should still make every effort to clean up unnatural links pointing to your site. Simply disavowing them isn't enough"
08:33 < kanzure> https://support.google.com/webmasters/bin/answer.py?hl=en&answer=2648487
08:33 < kanzure> "Google works very hard to make sure that actions on third-party sites do not negatively affect a website. In some circumstances, incoming links can affect Google’s opinion of a page or site. For example, you or a search engine optimizer (SEO) you’ve hired may have built bad links to your site via paid links or other link schemes that violate our quality guidelines. First and foremost, we recommend that you remove as many spammy or ...
08:33 < kanzure> ... low-quality links from the web as possible. If you’ve done as much work as you can to remove spammy or low-quality links from the web, and are unable to make further progress on getting the links taken down, you can disavow the remaining links. In other words, you can ask Google not to take certain links into account when assessing your site."
08:33 < kanzure> "i pinky promise that i didn't cause those links to appear"
08:35 < archels> what is "characterization of a genome" and how do they plan to base treatment on this?
08:36 < kanzure> they are inserting his cancerous cells into a rat and then watching the rat grow tumors/test treatments on the rat.
08:36 < kanzure> *rats.
08:37 < archels> where does the genome come in
08:37 < kanzure> sequencing
08:37 < archels> yes, but how does this fit into personalized treatment?
08:37 < kanzure> there are many ways that cancer can happen
08:37 < kanzure> when cancer happens you can look at the genome to see wtf happened to your cell
08:38 < kanzure> this then informs your decisions about which treatments to attempt
08:41 < archels> that's the theory, yes, but are we really at a stage yet where this is practicable? I mean, let's say it turns out there's a deletion of a few thousand bp on chromosome 5. And let's say that we get lucky, and some functions of those genes are known--say that a bunch of chaperones are coded for. Then what?
08:42 < archels> it's just a pileup of luck at every stage
08:42 < kanzure> sometimes you might luck out and your strain of cancer will have surface membrane receptors with really easy mutations. and those can be killed selectively.
08:42 < kanzure> yes but it's better than being dead
08:50 -!- quantumkat [~kat@ip70-171-6-179.ga.at.cox.net] has quit [Remote host closed the connection]
08:50 < archels> aye
08:51 -!- hehelleshin [~talinck@69-61-156-24.ubr1.dyn.lebanon-oh.fuse.net] has quit [Ping timeout: 245 seconds]
08:52 < archels> but even if you find a slightly mutated receptor, you'd still need to find a suitable ligand
08:52 < kanzure> most receptors degrade into a number of specific motifs
08:55 < kanzure> wish i had a ref on this one
08:55 < kanzure> there's probably an 'annual review' paper somewhere that summarizes cancers that we can molecularly target
09:33 < Urchin> http://spacecraft.ssl.umd.edu/old_site/academics/akins_laws.html
09:34 < kanzure> .title
09:34 < yoleaux> Akin's Laws of Spacecraft Design
09:35 -!- phillyj [~chatzilla@static-96-227-241-38.phlapa.fios.verizon.net] has joined ##hplusroadmap
09:41 -!- ielo [~ielo@184.88.56.168] has joined ##hplusroadmap
09:41 < Urchin> link paranoia, good
09:55 -!- FooQuuxman [~test@c-98-215-254-55.hsd1.il.comcast.net] has quit [Read error: Connection reset by peer]
09:57 -!- FooQuuxman [~test@c-98-215-254-55.hsd1.il.comcast.net] has joined ##hplusroadmap
09:58 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 260 seconds]
09:59 -!- klafka [~klafka@204-16-157-18-static.ipnetworksinc.net] has joined ##hplusroadmap
10:01 < kanzure> angular.js has a bunch of open pull requests (94).. https://github.com/angular/angular.js/pulls
10:09 -!- AshleyWaffle [~quassel@unaffiliated/anastasiawyatt] has quit [Read error: Connection reset by peer]
10:20 -!- helleshin [~talinck@69-61-156-24.ubr1.dyn.lebanon-oh.fuse.net] has joined ##hplusroadmap
10:20 -!- undersco2 [~quassel@192.210.211.75] has quit [Ping timeout: 248 seconds]
10:41 < archels> paperbot: http://www.jove.com/video/3691/dissection-culture-mouse-dopaminergic-striatal-explants-three
10:41 < paperbot> no translator available, raw dump: http://diyhpl.us/~bryan/papers2/paperbot/25eb4cabc45a966e09b3e5ecb9f4b205.txt
10:41 -!- underscor [~quassel@192.210.211.75] has joined ##hplusroadmap
10:41 -!- underscor is now known as Guest50861
10:42 -!- kmo [~kmo@apn-37-109-222-50.dynamic.lte.plus.pl] has joined ##hplusroadmap
10:42 -!- kmo [~kmo@apn-37-109-222-50.dynamic.lte.plus.pl] has quit [Changing host]
10:42 -!- kmo [~kmo@unaffiliated/kmo] has joined ##hplusroadmap
11:03 -!- FooQuuxman [~test@c-98-215-254-55.hsd1.il.comcast.net] has quit [Ping timeout: 256 seconds]
11:13 -!- eudoxia [~eudoxia@r190-134-62-35.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap
11:13 < eudoxia> so i found this link some years ago http://www.choo-institute.com/
11:13 < eudoxia> and i'm still wondering exactly what it is
11:14 < Urchin> first time I heard about it
11:16 < eudoxia> >Our project work impacts the nine billion people across the world currently excluded from the Diamond Age economy.
11:17 < eudoxia> sort of makes me think it's some sort of future history arg
11:36 < kanzure> http://octopart.com/blog/archives/2013/6/api-update%253A-access-datasheets-from-the-octopart-api
11:37 < kanzure> "I'm happy to report that you can now access datasheets from Octopart API V3. In addition to datasheets, we have added two new features which I'm very excited about: Show/Hide Directives and Slice Directives. Show/Hide Directives are url arguments that you can use to control the size of API response objects by only requesting the fields that you need. For example, you can use the directives to only return MPNs and manufacturer brand names for ...
11:37 < kanzure> ... each part matched in a call to the parts/match endpoint. Slice Directives are url arguments that you can use to perform slice operations on lists. For example, you can choose to only return the first five offers matched in a parts/match request. Slice Directives are very useful for more find grained control over API response sizes where you want to limit the number of objects in a list."
11:37 < kanzure> http://octopart.com/api/docs/v3/overview#datasheets
11:38 < kanzure> i wonder if it just returns pdf
11:43 < kanzure> http://blogs.scientificamerican.com/oscillator/2013/06/03/glowing-futures/ biocurious things
11:43 < kanzure> ugh "Amirav-Drory says that the Glowing Plant project is “so obviously beautiful, and non-dangerous and legal and ethical,”"
11:43 < kanzure> he actually said that about his own project? jeeze.
11:43 < eudoxia> shameless
11:44 < kanzure> ooh i like this author
11:44 < kanzure> "So is it the same or is it different? While aggressively selling themselves as something new and something inherently moral and democratic, Genome Compiler is unfortunately acting out a very familiar story. They are a for-profit corporation trying to wedge themselves into a crowded market of DNA-based services that has many viable open source alternatives, exploiting the ideas, work, and the enthusiasm of idealistic people excited about the ...
11:44 < kanzure> ... potential of synthetic biology, overpromising and overhyping the potential of the technology in the short term in order to market their derivative, proprietary software tools and themselves as leaders in the future of the field."
11:44 < kanzure> hey look! someone saw through genome compiler's bullshit for once. cool.
11:44 < kanzure> "In an email exchange with the Marketing Manager of Genome Compiler Corp. about the patent situation, the company stated that “[We] initially thought of doing some kind of defensive patent, but we decided not to in the end. Genome Compiler was built with the thought of democratizing creation and we would be thrilled if someone wanted to use or alter the design. The design is already up for anyone to see on Genome Compiler in the cloud. We ...
11:44 < kanzure> ... were not concerned with other patents since we were using a different method.”"
11:44 < kanzure> "the design" ?
11:45 -!- eudoxia [~eudoxia@r190-134-62-35.dialup.adsl.anteldata.net.uy] has left ##hplusroadmap ["leaving"]
11:45 < kanzure> hahah
11:45 < kanzure> "Genome Compiler Corp. has gotten into patent trouble before, not over uses and applications of their software, but with the actual software itself. The DNA synthesis company DNA 2.0 has a patent on their version of “drag-and-drop” synthetic biology design software called Gene Designer, and has sued Genome Compiler for infringement."
11:45 < kanzure> that's excellent.
11:45 < kanzure> "Fortunately, many other options for this kind of software exist, including some well developed open source options that allow you to edit, build on, and contribute to the future development of the software itself rather than just releasing your designs to a corporation’s “cloud” so that they might one day decide to do a Kickstarter about your idea too. These include programs like Clotho or TinkerCell to name just two of many, as well ...
11:45 < kanzure> ... as a host of much simpler DNA viewing tools that allow users to design small-scale plasmids, not to mention the open and completely free databases of genetic sequences that anyone can access like the National Center for Biotechnology Information or the iGEM Parts Registry."
11:46 < kanzure> "In general I don’t agree with biotechnology patents and I support the efforts of the BioBricks foundation to explore different arrangements and licensing schemes. It is difficult to tell what the role of Genome Compiler will be in the long run of the debate about intellectual property in biotechnology, and it’s difficult to tell what role patents will end up playing in the Glowing Plant story. The language of the kicktarter campaign and ...
11:46 < kanzure> ... of the Genome Compiler software is language of the open source movement, emphasizing sharing and “democratization” over profit. However, while Genome Compiler cites the PLoS ONE paper from the Stony Brook team, they fail to mention the fact that this group has started a company, BioGlow Inc., to produce and sell ornamental glowing plants, and that this company holds a broad patent on bioluminescent plants (there are a number of other ...
11:46 < kanzure> ... patents on transgenic bioluminescent plants as well). Does the power of the story of openness on the part of Genome Compiler trump the previous work of the BioGlow group? What or who is being democratized, who is being protected or infringed on here? Do crowdfunding “rewards” infringe on patents the same way that a product for sale would?"
11:47 < kanzure> "The image of the glowing plant from the 1986 Science paper that the team is using as the logo of their campaign combines many of the tensions present in the design and marketing of the project. First of all, the image is a direct copy of the work of past researchers, copyrighted by Science, and used with ambiguous citation in promotional materials, proposed Kickstarter rewards like t-shirts, advertising and branding on their website and ...
11:47 < kanzure> ... their software. The use of this copyrighted image to raise funds is a symptom of a larger issue with how intellectual property is being dealt with in this story (more on that later), but more importantly the image provides the team both credibility in terms of feasibility of the project as well as in terms of “wow”-factor. The image of the glowing plant shows that this is technology that has been around for a long time–nearly 30 ...
11:47 < kanzure> ... years!–but is also misleadingly reminiscent of the science fiction plants from the movie “Avatar,” which are frequently cited as inspiration."
11:53 -!- phillyj [~chatzilla@static-96-227-241-38.phlapa.fios.verizon.net] has quit [Quit: ChatZilla 0.9.90 [Firefox 21.0/20130511120803]]
12:07 -!- lichen_ [~lichen@c-24-21-206-64.hsd1.or.comcast.net] has joined ##hplusroadmap
12:10 -!- lichen [~lichen@c-24-21-206-64.hsd1.or.comcast.net] has quit [Ping timeout: 248 seconds]
12:12 -!- eudoxia [~eudoxia@r190-134-62-35.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap
12:13 < eudoxia> http://www.google.com/patents/US20090056802
12:13 < eudoxia> >Practical method and means for mechanosynthesis and assembly of precise nanostructures and materials including diamond, programmable systems for performing same; devices and systems produced thereby, and applications thereof
12:13 < eudoxia> nope, not the freitas patent
12:13 < eudoxia> >Inventors	Eli Michael Rabani
12:13 < eudoxia> who is this guy? this looks like it was filed pre the freitas patent
12:14 -!- lichen_ [~lichen@c-24-21-206-64.hsd1.or.comcast.net] has quit [Ping timeout: 248 seconds]
12:20 < kanzure> there were a bunch of people on usenet newsgroups back in the day talking about nanotech things, it could be any number of those people.
12:20 < kanzure> i think it was sci.nano or sci.nanotech
12:26 -!- lichen [~lichen@c-24-21-206-64.hsd1.or.comcast.net] has joined ##hplusroadmap
12:42 -!- eudoxia [~eudoxia@r190-134-62-35.dialup.adsl.anteldata.net.uy] has quit [Quit: leaving]
12:45 < kanzure> octopart responded to my hatemail and added an example to their documentation: http://octopart.com/api/docs/v3/overview#datasheets
12:46 < kanzure> huh they also support CORS.
12:46 < kanzure> or erm.. cross-domain requests.
12:49 -!- Guest50861 is now known as underscor
12:53 -!- ielo [~ielo@184.88.56.168] has quit [Ping timeout: 256 seconds]
13:11 -!- makoLime [~mako@103-9-42-133.flip.co.nz] has joined ##hplusroadmap
13:19 < klafka> http://izbicki.me/public/papers/icml2013-algebraic-classifiers.pdf btw one of the craziest ml papers in awhile
13:50 -!- randallagordon [~randall@75-164-226-12.ptld.qwest.net] has quit [Ping timeout: 248 seconds]
13:50 -!- randallagordon [~randall@75-164-222-73.ptld.qwest.net] has joined ##hplusroadmap
14:02 -!- makoLime [~mako@103-9-42-133.flip.co.nz] has quit [Ping timeout: 246 seconds]
14:29 -!- kirka [~Kirka@109.205.253.139] has joined ##hplusroadmap
14:29 < kirka> Hi~
14:35 < kanzure> hello
14:35 < klafka> 	hey
14:47 < kirka> I have read some http://www.n-a-n-o.com/nano/cavstuf/cavstuf-0.2.2.html sources, and found that some files are written by K.E.Drexler. Didn't know he was also a CL programmer.
14:48 < kirka> I mean Common Lisp
14:51 < kirka> https://www.google.com/search?q=antibiotic+timeline&um=1&ie=UTF-8&hl=ru&tbm=isch&source=og&sa=N
14:52 < kirka> Looks like development of new antibiotic drugs has slowed down in last 40 years
14:52 < kirka> I wonder about the cause
14:53 < kirka> Maybe it's because FDA has made testing of new drug a ~100 million dollar deal
15:12 -!- eudoxia [~eudoxia@r190-135-50-209.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap
15:12 < eudoxia> hello kirka
15:12 < ParahSailin> maybe more likely that we've run out of easy to find antibiotic-producing strains
15:12 < kirka> hi, eudoxia
15:14 < eudoxia> i didn't know drexler was a coder... at all
15:14 < eudoxia> i would assume he just gave theoretical input
15:15 < kirka> ;;; Historical Note: an initial version of this file was made by K. E. Drexler.
15:15 < kirka> ;;;                  subsequent modifications by KR.
15:15 < eudoxia> oh, lol
15:15 < kirka> eudoxia Interesting patent I see in logs, found by you
15:16 < kirka> google finds something relevant on "Ely Rabani"
15:17 < eudoxia> i briefly googled him but didn't see links to anything but more patent indexing sites
15:17 < eudoxia> i honestly thought the only MNT-related patent was from freitas, but this was a patent filed almost on the same date and almost on the same subject
15:17 < eudoxia> i think some of the figures are from nanosytems
15:19 < kirka> https://groups.google.com/groups/profile?enc_user=B9u4ahYAAADOqoJgV8IIafbxARdVDngf4s_GnoCDq-EqGGyHnuiveQ
15:20 < kirka> Just some random guy briefly related to sci.nanotech
15:21 < eudoxia> huh, kanz was right after all
15:22 < eudoxia> god damn google groups can't search shit
15:22 < kirka> https://groups.google.com/groups/search?q=ely+rabani+nano
15:23 < kirka> >For teaching me to use the atomic force microscope, I thank Anca Segall, Bob Moision,
15:23 < kirka> and Ely Rabani; with their encouragement and help I saw the first exciting images of DNA lattices.
15:23 < kirka> http://cba.mit.edu/events/03.11.ASE/docs/Winfree.pdf
15:26 -!- helleshin [~talinck@69-61-156-24.ubr1.dyn.lebanon-oh.fuse.net] has quit [Ping timeout: 252 seconds]
15:27 < kirka> ParahSailin Isn't it possible to find new antibacterial drugs with drug design, as it is practiced by big pharma?
15:27 < ParahSailin> i think thats pretty unusual in practice
15:28 < ParahSailin> i think cipro is a fully synthetic one not resembling a natural antibiotic
15:29 < kirka> Interesting, thanks.
15:30 < ParahSailin> also one of the least desirable ones to feed humans
15:30 < ParahSailin> linezolid also looks fully synthetic
15:31 < ParahSailin> sulfa drugs also fully synthetic
15:36 -!- eudoxia [~eudoxia@r190-135-50-209.dialup.adsl.anteldata.net.uy] has quit [Quit: Saliendo]
15:44 -!- eudoxia [~eudoxia@r190-135-50-209.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap
16:13 -!- eudoxia [~eudoxia@r190-135-50-209.dialup.adsl.anteldata.net.uy] has quit [Quit: Saliendo]
16:18 -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving]
16:28 -!- FooQuuxman [~test@c-98-215-254-55.hsd1.il.comcast.net] has joined ##hplusroadmap
16:31 -!- EnLilaSko [EnLilaSko@unaffiliated/enlilasko] has quit [Quit: - nbs-irc 2.39 - www.nbs-irc.net -]
16:38 < kanzure> i really dislike all the pandering to hawkins/numenta, like here: https://news.ycombinator.com/item?id=5814382
16:39 < kanzure> it's basically a bunch of people with no knowledge of neuroscience trying to claim that his neural networks are realistic, except they aren't.
16:41 < kanzure> the typical numenta fanboy is best characterized as that guy who read the news clip about the blue brain project receiving a few billion euros and was like "So what? It doesn't sound like they have read *On Intelligence* so I don't think this will go anywhere." as if really liking a book makes it useful/true.
16:48 -!- Charlie [~quassel@74.63.212.44] has quit [Ping timeout: 245 seconds]
16:50 -!- Charlie [~quassel@74.63.212.44] has joined ##hplusroadmap
16:51 < kirka> I have never heard about comparsion of Numenta's approach to other ML algorithms in standard benchmarks. Numenta looks overhyped.
16:51 < kirka> I haven't seen formal definition of their algo either.
16:56 -!- klafka [~klafka@204-16-157-18-static.ipnetworksinc.net] has quit [Remote host closed the connection]
16:57 -!- lichen [~lichen@c-24-21-206-64.hsd1.or.comcast.net] has quit [Quit: Lost terminal]
17:00 < kirka> kanzure What do you think about Mrkram and his billion euros?
17:02 < kirka> Thay have an ambition
17:04 < kirka> It's hard to believe that they will reach their set of goals.
17:09 < Urchin> who?
17:09 < kirka> Human Brain project
17:11 -!- makoLime [~mako@wireless-nat-7.auckland.ac.nz] has joined ##hplusroadmap
17:11 < Urchin> oh, I'm not following a lot of such developments lately
17:17 < kanzure> i believe there were formal papers about numenta's work that have been published.
17:17 < kanzure> i think markram is really hypnotic. i wish he would release more software.
17:18 < kanzure> their goals were really well picked and clearly defined, i don't think there's any chance of them missing them ("do some research projects")
17:18 < kanzure> the news hyped up their project to make it seem like "we will emulate a human brain by 2015", but the reality is that their documentation says nothing like that.
17:18 < kanzure> http://diyhpl.us/~bryan/papers2/neuro/HBP_flagship.pdf
17:18 < kirka> Ah, that's it.
17:19  * kirka is reading
17:22 < fenn> kanzure: you shouldn't discount a project just because it has fanboys
17:23 < fenn> apparently "on intelligence" is responsible for andrew ng getting interested in statistical machine learning
17:23 < kanzure> i discount numenta's book for other reasons
17:23 < kanzure> not because of fanboys. the fanboys are a separate issue.
17:24 < kirka> Maybe there is some evaluation of their approch on standard problems? E.g. classification, regression.
17:25 < kanzure> kirka: well, i'm talking about the book itself
17:25 < kanzure> http://en.wikipedia.org/wiki/On_Intelligence
17:25 -!- klafka [~klafka@ip-64-134-145-116.public.wayport.net] has joined ##hplusroadmap
17:26 < fenn> numenta isn't in academia so they haven't released the sort of standard papers you'd expect of a new algorithm
17:26 < kirka> I have heard about but haven't rad ir. I have read some docs from their site, but they don't present complete algorithm there.
17:26 < fenn> (that's not an excuse though)
17:26 -!- klafka [~klafka@ip-64-134-145-116.public.wayport.net] has quit [Remote host closed the connection]
17:27 < kanzure> iirc there was some source code they released for a tool or a thing.
17:27 < kanzure> http://en.wikipedia.org/wiki/Hierarchical_Temporal_Memory
17:27 -!- eudoxia [~eudoxia@r190-135-50-209.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap
17:27 < kanzure> i'm also not sure why his book gets much more attention than any of the other work that attempts to make a biologically-accurate representation
17:28 < kirka> That was it https://www.groksolutions.com/htm-overview/education/HTM_CorticalLearningAlgorithms.pdf
17:28 -!- AshleyWaffle [~quassel@63-228-107-211.tukw.qwest.net] has joined ##hplusroadmap
17:28 -!- AshleyWaffle [~quassel@63-228-107-211.tukw.qwest.net] has quit [Changing host]
17:28 -!- AshleyWaffle [~quassel@unaffiliated/anastasiawyatt] has joined ##hplusroadmap
17:28 < fenn> because you're a packer, not a mapper. mappers love elegant abstractions
17:29 < fenn> packers get distracted by details
17:29 < fenn> it wasn't an attempt to "make a biologically-accurate representation", it was trying to explain all the details in a coherent whole framework
17:30 < kanzure> what is elegant about it in particular? the fact that it talks about cortical columns?
17:30 < kanzure> oh oops you said abstractions
17:30 < fenn> the fact that there's a single algorithm used everywhere in the cortex
17:31 < kanzure> why did i think you said elegant?
17:31 < fenn> i said elegant
17:31 < eudoxia> he said both
17:31 < kanzure> oh good
17:32 < kanzure> wow these wikipedia articles are junk
17:33 < kanzure> although the list of his predictions of future discoveries is nice of them to bother with.
17:34 < kanzure> "For example, for the purposes of his framework, the nerve impulses can be taken to form a temporal sequence (but phase encoding could be a possible implementation of such a sequence; these details are immaterial for the framework)."
17:35 < fenn> btw if anyone would like to read the book: http://fennetic.net/irc/on_intelligence.pdf
17:35 < eudoxia> thanks fenn
17:36 < kanzure> so what's elegant about it, again? i mean, i've seen so many artificial neural network models that i'm struggling to remember the exact differences between his feed forward model and everyone else's.
17:36 < fenn> it explains firing patterns
17:37 < kirka> Maybe someone should take their NuPIC, take some state of the art temporal sequence prediction algorithms and benchmark them? That'd solve the question once and for all.
17:37 < fenn> afaik other neural networks fire more or less randomly
17:38 < fenn> kirka: not necessarily. until recently, SVM did better than neural networks on benchmarks. then computational power went up and neural networks did better on the new larger data sets
17:38 < kanzure> what part of it explains firing patterns?
17:39 < fenn> i dont remember, it's been several years since i read the book
17:39 < fenn> maybe i should read it again
17:39 < kanzure> it's been almost 10 years for me. i think when i first read it i was so excited that i wrote in pen over the pages.
17:39 < fenn> i should note that i never could make any sense out of neuroscience until i read this book
17:40 < fenn> maybe now i will have a different opinion, having learned more about neuroscience since then
17:41 < kanzure> the book is definitely a good introduction to neural networks of all sorts
17:41 < kirka> fenn That's true. Also that fact that it's temporally-oriented makes benchmarking difficult.
17:41 < kanzure> very basic models like perceptrons are useful introductory material (although i think he skips this anyway, who cares)
17:41 < eudoxia> perceptron is like the name of some 1950's atompunk big brother supercomputer
17:42 < kanzure> nah
17:42 < kanzure> oh actually brownies was complaining to me the other day that he wanted to brush up on sparse coding or something
17:42 < fenn> the literature about perceptrons is horrendously obtuse and offputting for a newbie
17:42 < kanzure> so i dug out some old notes from an intro class
17:42 < kanzure> http://web.archive.org/web/20081205173159/http://heybryan.org/school/buildingbrains/all.html
17:43 < kirka> I have talked to guys that made hardwired (analog circuit) perceptrons in 1950s. They were tuning weights manually with a hundred of potentiometers.
17:44 < kanzure> except the class didn't go into actual neurophysiology-informed neuroscience
17:44 < fenn> computer science really should start out with shannon style probabilistic information theory
17:45 < fenn> you can get orders of magnitude more bandwidth down a wire if you're willing to screw up a bit here and there
17:47 < fenn> anyway i think understanding how i.e. huffman coding works would be a good thing to know
17:49 < fenn> i'm impressed they were teaching the stuff at the above link in your high school
17:49 < kanzure> man i want a giant perceptron machine with potentiometer weights. damn.
17:49 < kanzure> sadly that was a filler class i took in college
17:49 < fenn> my CS class was like, "what is an integer"
17:50 < fenn> "what does CMOS stand for" (as if it mattered)
17:51 < kanzure> my high school comp sci class had me writing java in 2004-2005
17:51 < kanzure> http://web.archive.org/web/20081202064047/http://heybryan.org/school/gift1.java
17:52 < fenn> Person[] people = new Person[totalNumberOfUselessnesses];
17:52 < kanzure> i was jaded and bored
17:52 < fenn> java will do that
17:53 < fenn> this is probably 5 lines of bash
17:53 < kanzure> i think i literally said that exact line, out loud
17:55 -!- helleshin [~talinck@69-61-156-24.ubr1.dyn.lebanon-oh.fuse.net] has joined ##hplusroadmap
18:01 -!- hehelleshin [~talinck@69-61-156-24.ubr1.dyn.lebanon-oh.fuse.net] has joined ##hplusroadmap
18:03 < kanzure> "Markus Götz and Peter Hartmann are busy working on the QNetworkAccessManager stack. Their work includes improving the parsing speed of HTTP headers, making sure to start HTTP connections after the first iteration of the mainloop instead of the third."
18:03 < kanzure> "In one of my tests wget is still twice as fast as the Qt stack to download the same set of files. And wget is using one connection at a time, no pipelining... and Qt is attempting to have up to 6 connections in parallel. This means there is still some work to do in reducing latency and improving scheduling of requests. I'm pretty confident that Markus and Peter will work on this!"
18:04 -!- helleshin [~talinck@69-61-156-24.ubr1.dyn.lebanon-oh.fuse.net] has quit [Ping timeout: 256 seconds]
18:08 < kanzure> "Inbound links don't immediately get detected as negative. That takes a while." https://news.ycombinator.com/item?id=5813704 welp.. that seems strange. i guess google takes a while to compute negative pagerank.
18:10 < fenn> or they're just messing with the spammers
18:10 < fenn> if you could immediately measure google's response to a particular link, it would be much easier to optimize your strategy
18:10 -!- ParahSailin_ [~ropoctl@99-25-200-252.lightspeed.hstntx.sbcglobal.net] has joined ##hplusroadmap
18:10 -!- ParahSailin_ [~ropoctl@99-25-200-252.lightspeed.hstntx.sbcglobal.net] has quit [Changing host]
18:10 -!- ParahSailin_ [~ropoctl@unaffiliated/parahsailin] has joined ##hplusroadmap
18:10 < kanzure> you can measure, though. based on inbound traffic. from google.
18:11 -!- FooQuuxman [~test@c-98-215-254-55.hsd1.il.comcast.net] has quit [Ping timeout: 246 seconds]
18:12 -!- Urchin [~user@unaffiliated/urchin] has quit [Quit: ERC Version 5.2 (IRC client for Emacs)]
18:27 < ParahSail1n> platinum wire too expensive for electrophoresis? man, you're never gonna do science if $1/cm is a dealbreaker
18:30 -!- yashgaroth [~ffffff@cpe-66-27-118-94.san.res.rr.com] has joined ##hplusroadmap
18:53 -!- lichen [~lichen@c-24-21-206-64.hsd1.or.comcast.net] has joined ##hplusroadmap
18:57 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap
19:13 -!- FooQuuxman [~test@c-98-215-254-55.hsd1.il.comcast.net] has joined ##hplusroadmap
19:31 -!- AlonzoTG [~atg@pool-108-28-74-221.washdc.fios.verizon.net] has quit [Ping timeout: 276 seconds]
19:37  * kirka switches into sleep mode
19:38 -!- eudoxia [~eudoxia@r190-135-50-209.dialup.adsl.anteldata.net.uy] has quit [Quit: Saliendo]
19:38 -!- kirka [~Kirka@109.205.253.139] has left ##hplusroadmap []
19:47 -!- AlonzoTG [~atg@pool-108-28-74-221.washdc.fios.verizon.net] has joined ##hplusroadmap
19:59 -!- Charlie [~quassel@74.63.212.44] has quit [Ping timeout: 256 seconds]
20:02 -!- FooQuuxman [~test@c-98-215-254-55.hsd1.il.comcast.net] has quit [Ping timeout: 256 seconds]
20:04 -!- abetusk [~abetusk@cpe-24-58-232-122.twcny.res.rr.com] has joined ##hplusroadmap
20:16 -!- AlonzoTG [~atg@pool-108-28-74-221.washdc.fios.verizon.net] has quit [Ping timeout: 246 seconds]
20:17 < nmz787> what is this weird pay-wall http://www.jce.divched.org/video/electrolysis-sodium-hydroxide-solution
20:23 -!- makoLime [~mako@wireless-nat-7.auckland.ac.nz] has quit [Ping timeout: 276 seconds]
20:32 -!- AlonzoTG [~atg@pool-108-28-74-221.washdc.fios.verizon.net] has joined ##hplusroadmap
20:57 -!- makoLime [~mako@wireless-nat-7.auckland.ac.nz] has joined ##hplusroadmap
21:05 -!- AlonzoTG [~atg@pool-108-28-74-221.washdc.fios.verizon.net] has quit [Ping timeout: 264 seconds]
21:11 -!- lichen [~lichen@c-24-21-206-64.hsd1.or.comcast.net] has quit [Quit: Lost terminal]
21:21 -!- AlonzoTG [~atg@pool-108-28-74-221.washdc.fios.verizon.net] has joined ##hplusroadmap
21:26 < brownies> kanzure: sparse coding? what?
21:26 < kanzure> ensemble coding
21:26 < kanzure> sorry
21:27 -!- ParahSailin_ [~ropoctl@unaffiliated/parahsailin] has quit [Quit: ParahSailin_]
21:34 -!- makoLime [~mako@wireless-nat-7.auckland.ac.nz] has quit [Ping timeout: 264 seconds]
21:42 -!- Charlie [~quassel@74.63.212.44] has joined ##hplusroadmap
21:43 -!- HEx1 [~HEx@hexwab.plus.com] has quit [Ping timeout: 256 seconds]
22:12 -!- superkuh [~superkuh@unaffiliated/superkuh] has quit [Remote host closed the connection]
22:13 -!- kmo [~kmo@unaffiliated/kmo] has quit [Quit: Leaving]
22:26 -!- lichen [~lichen@c-24-21-206-64.hsd1.or.comcast.net] has joined ##hplusroadmap
23:02 -!- klafka [~klafka@c-67-174-253-229.hsd1.ca.comcast.net] has joined ##hplusroadmap
23:02 -!- makoLime [~mako@103-9-42-133.flip.co.nz] has joined ##hplusroadmap
23:11 -!- AshleyWaffle is now known as Ashley420SwagOba
23:11 -!- Ashley420SwagOba is now known as Ashley420Swag69
23:23 -!- Ashley420Swag69 is now known as AshleyWaffle
23:34 -!- yashgaroth [~ffffff@cpe-66-27-118-94.san.res.rr.com] has quit [Quit: Leaving]
23:41 -!- klafka [~klafka@c-67-174-253-229.hsd1.ca.comcast.net] has quit [Ping timeout: 256 seconds]
23:45 -!- AshleyWaffle [~quassel@unaffiliated/anastasiawyatt] has quit [Read error: Connection reset by peer]
23:46 -!- klafka [~klafka@c-24-6-18-31.hsd1.ca.comcast.net] has joined ##hplusroadmap
--- Log closed Tue Jun 04 00:00:46 2013