| --- Log opened Fri Mar 11 00:00:56 2016 | ||
| -!- urchin [~urchin@unaffiliated/urchin] has joined ##hplusroadmap | 01:49 | |
| -!- urchin is now known as Urchin | 01:50 | |
| -!- TOB1 [~ObitO@41.101.165.99] has joined ##hplusroadmap | 02:05 | |
| -!- T0BI [~ObitO@41.101.163.58] has quit [Ping timeout: 276 seconds] | 02:06 | |
| -!- Gurkenglas [Gurkenglas@dslb-188-106-113-006.188.106.pools.vodafone-ip.de] has joined ##hplusroadmap | 02:21 | |
| -!- Gurkenglas [Gurkenglas@dslb-188-106-113-006.188.106.pools.vodafone-ip.de] has quit [Ping timeout: 248 seconds] | 02:58 | |
| -!- zadock [~outsider@cthulhu.tuiasi.ro] has joined ##hplusroadmap | 03:22 | |
| -!- zadock [~outsider@cthulhu.tuiasi.ro] has quit [Quit: Leaving] | 03:39 | |
| -!- zadock [~outsider@cthulhu.tuiasi.ro] has joined ##hplusroadmap | 03:39 | |
| -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-emxrajejnxrpgsml] has quit [Quit: Connection closed for inactivity] | 03:40 | |
| -!- zadock [~outsider@cthulhu.tuiasi.ro] has quit [Remote host closed the connection] | 03:40 | |
| -!- zadock [~outsider@cthulhu.tuiasi.ro] has joined ##hplusroadmap | 03:45 | |
| -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-aoevjiquizydrdqt] has joined ##hplusroadmap | 03:46 | |
| -!- jtimon [~quassel@35.31.134.37.dynamic.jazztel.es] has joined ##hplusroadmap | 03:50 | |
| -!- AmbulatoryCortex [~Ambulator@173-31-155-69.client.mchsi.com] has joined ##hplusroadmap | 03:55 | |
| -!- jaboja [~jaboja@ejw72.neoplus.adsl.tpnet.pl] has joined ##hplusroadmap | 04:12 | |
| -!- sandeepkr__ [~sandeep@111.235.64.4] has quit [Ping timeout: 260 seconds] | 04:19 | |
| -!- zadock [~outsider@cthulhu.tuiasi.ro] has quit [Quit: Leaving] | 04:19 | |
| -!- sandeepkr__ [~sandeep@111.235.64.4] has joined ##hplusroadmap | 04:22 | |
| -!- eudoxia [~eudoxia@r167-57-37-8.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 04:59 | |
| -!- yorick__ [~yorick@oftn/oswg-member/yorick] has joined ##hplusroadmap | 04:59 | |
| -!- yorick [~yorick@oftn/oswg-member/yorick] has quit [Ping timeout: 240 seconds] | 05:01 | |
| -!- eudoxia [~eudoxia@r167-57-37-8.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving] | 05:37 | |
| -!- Aurelius_Work [~cpopell@209.48.69.2] has quit [Ping timeout: 240 seconds] | 05:57 | |
| -!- Aurelius_Work [~cpopell@209.48.69.2] has joined ##hplusroadmap | 06:00 | |
| -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-aoevjiquizydrdqt] has quit [Quit: Connection closed for inactivity] | 06:20 | |
| kanzure | http://shitmyreviewerssay.tumblr.com/ | 06:22 |
|---|---|---|
| kanzure | .tw https://twitter.com/AcademiaObscura/status/708223626877788160 | 06:25 |
| yoleaux | New rule: if your footnotes cover more than half the page, I'm skipping that page. https://t.co/2dIxvPJM71 (@AcademiaObscura) | 06:25 |
| kanzure | https://twitter.com/FakeElsevier | 06:31 |
| fenn | ‘Hypothesized’ is such an ugly word | 06:31 |
| kanzure | FakeElsevier needs to meet FakeEliezer | 06:34 |
| fenn | "Not only does this strike me as the worst kind of postmodern legerdemain, but if true the statement would transform ethics into a hopelessly muddled enterprise." | 06:35 |
| fenn | i think these are all psychology paper reviewers | 06:36 |
| cluckj | what a bizarre statement you quoted | 06:57 |
| -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-kkimnnyujoknfayb] has joined ##hplusroadmap | 07:02 | |
| -!- AmbulatoryC0rtex [~Ambulator@173-31-155-69.client.mchsi.com] has joined ##hplusroadmap | 07:16 | |
| -!- Aurelius_Work [~cpopell@209.48.69.2] has quit [Ping timeout: 252 seconds] | 07:16 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-kbbafmbvbalprtiv] has joined ##hplusroadmap | 07:18 | |
| -!- xrr [~quassel@c21-76.uvn.zone.eu] has quit [Remote host closed the connection] | 07:19 | |
| -!- AmbulatoryCortex [~Ambulator@173-31-155-69.client.mchsi.com] has quit [Ping timeout: 244 seconds] | 07:19 | |
| -!- xrr [~quassel@c21-76.uvn.zone.eu] has joined ##hplusroadmap | 07:19 | |
| -!- Gurkenglas [Gurkenglas@dslb-188-106-113-006.188.106.pools.vodafone-ip.de] has joined ##hplusroadmap | 07:20 | |
| -!- xrr [~quassel@c21-76.uvn.zone.eu] has quit [Remote host closed the connection] | 07:24 | |
| -!- xrr [~quassel@c21-76.uvn.zone.eu] has joined ##hplusroadmap | 07:25 | |
| -!- Aurelius_Work [~cpopell@209.48.69.2] has joined ##hplusroadmap | 07:48 | |
| -!- xrr [~quassel@c21-76.uvn.zone.eu] has quit [Remote host closed the connection] | 07:52 | |
| -!- xrr [~quassel@c21-76.uvn.zone.eu] has joined ##hplusroadmap | 07:53 | |
| -!- AmbulatoryCortex [~Ambulator@173-31-155-69.client.mchsi.com] has joined ##hplusroadmap | 07:54 | |
| -!- AmbulatoryC0rtex [~Ambulator@173-31-155-69.client.mchsi.com] has quit [Ping timeout: 250 seconds] | 07:57 | |
| -!- jaboja64 [~jaboja@emf122.neoplus.adsl.tpnet.pl] has joined ##hplusroadmap | 08:07 | |
| -!- jaboja [~jaboja@ejw72.neoplus.adsl.tpnet.pl] has quit [Ping timeout: 240 seconds] | 08:10 | |
| -!- xrr [~quassel@c21-76.uvn.zone.eu] has quit [Remote host closed the connection] | 08:44 | |
| -!- xrr [~quassel@c21-76.uvn.zone.eu] has joined ##hplusroadmap | 08:44 | |
| kanzure | singinst forum thingy https://agentfoundations.org/ | 08:53 |
| kanzure | maaku: your best bet for ai collaboration on freeonde is pasky | 08:57 |
| kanzure | *freenode | 08:57 |
| maaku | i have a couple of gpu workstations otherwise sitting idle 90% of the time. i would love to have an open-source Watson-like clone running on libgen data | 09:01 |
| pasky | maaku: I run live.ailao.eu + movies.ailao.eu; it doesn't run on gpu, but i'm currently using a gpu grid heavily (with some backlog right now actually) to solve https://github.com/brmson/dataset-sts (and I plan to hook that up to YodaQA pretty soon, a student is working on that for one of the tasks already) | 09:03 |
| pasky | a guy on germany is working on applying this to papers instead of webpages+enwiki, he doesn't want to show me anything much before he submits his masters thesis but claims it's working pretty well :) | 09:03 |
| pasky | *in | 09:04 |
| pasky | brb | 09:04 |
| maaku | interesting! | 09:04 |
| kanzure | maaku: yeah i have been wondering about queries to run over this huge dataset... there's probably a bunch of interesting things in here that nobody has looked at. | 09:04 |
| kanzure | a pretty large citation graph could be extracted | 09:06 |
| kanzure | which is usually proprietary data | 09:06 |
| kanzure | unfortunately i don't think this data set contains supplemental docs...... much less other attached media like external images, videos, software. | 09:09 |
| kanzure | are supplemental docs assigned doi numbers? | 09:10 |
| kanzure | maaku: how about #machinelearning on freenode? | 09:16 |
| kanzure | Douhet: does stuff happen in there? | 09:17 |
| * pasky op there | 09:18 | |
| pasky | it's getting better than it used to be :) wide spectrum of questions, from newbie to highly technical, and sometimes someone even replies | 09:19 |
| kanzure | pasky: maaku is looking for more agi people to hang with. i have told him that he might have to settle for just you :). | 09:20 |
| kanzure | fenn is living with a handful of agi people but i doubt they are actively playing around with novel software experiments | 09:22 |
| pasky | heh :) i'd be careful to call myself agi, though it's my long-term goal; i think #machinelearning is more practically oriented though, otoh if he wants to discuss neural turing machines and whatnot or get appreciation of the realities of machine learning, it may be okay | 09:22 |
| kanzure | no i think he is looking for people that have been working on recurrent networks and other strange feedback architectures | 09:23 |
| kanzure | e.g. people who have said "huh we should apply backpropagation to this" and then did it, and then maaku wants to eat their brains | 09:23 |
| pasky | okay, i have been playing with that a lot lately | 09:23 |
| pasky | but my brain is mine! | 09:23 |
| fenn | pasky it needs some work: http://fennetic.net/irc/yodaqa_fail.png | 09:40 |
| fenn | prioritizing wikipedia intro sections is a good idea | 09:42 |
| fenn | .wik dog | 09:42 |
| yoleaux | "" — https://en.wikipedia.org/wiki/Dog | 09:42 |
| fenn | hm | 09:42 |
| fenn | .wik dog | 09:42 |
| yoleaux | "" — https://en.wikipedia.org/wiki/Dog | 09:42 |
| fenn | well saxo gets it | 09:42 |
| maaku | kanzure: Houshalter could be helpful if he stopped worrying about safety | 09:44 |
| kanzure | fenn has been unsuccessful in his attempts to get people to stop worrying about unsafety | 09:44 |
| maaku | but yes I'm looking for people to discuss strange recurrent network architectures with | 09:45 |
| fenn | is "strange" a technical term? | 09:45 |
| kanzure | you should drop by fenn's house in the bay area, he has interesting people you will like to meet | 09:45 |
| maaku | I admit to having been wrong about the capabilities of deep learning networks, and the generality of a deep learning + reinforcement approach DeepMind is taking | 09:45 |
| kanzure | fenn: strange probably means "not regularly discussed or attempted" | 09:45 |
| maaku | what kanzure said | 09:45 |
| fenn | i advise against visiting, this place is terrible | 09:46 |
| kanzure | fenn: maaku just wants some programmers to throw around ideas with and then see what sticks and what does interesting things | 09:46 |
| kanzure | you know.... like gangbangers. | 09:46 |
| fenn | uh, uhm... etymology | 09:46 |
| maaku | pretty much | 09:47 |
| kanzure | fenn: er, wu tang clan of agi | 09:47 |
| maaku | kanzure: the citation graph is useful in a pagerank sense for figuring out what weight should be given to which papers | 09:47 |
| maaku | it also can be used for supervised training -- a generative model should be able to create the content of the child paper from its citation parents | 09:47 |
| maaku | although obviously this is a little bit more complex because it is a one-to-many relationship. but over the corpus as a whole an artificial scientist with access to papers before date T, and data for all future experiments, should be able to recreate future works | 09:48 |
| kanzure | it's a cyclic graph, because sometimes you cite papers not yet published | 09:48 |
| kanzure | because people are manipulative donkeys | 09:48 |
| maaku | true | 09:48 |
| kanzure | there was once someone who was rendering pdfs on his server for google scholar crawl bot | 09:49 |
| kanzure | and he generated a giant tree of papers that cited each other | 09:49 |
| kanzure | to become the most cited paper on google scholar | 09:49 |
| kanzure | it would be interesting to do the same but also citing 50 million papers | 09:49 |
| maaku | I'd be interested in a YODA like system not so much for the natural language query interface, but how the internal data structures of the knowledge graph could be used for both making and running an artificial scientist | 09:49 |
| kanzure | you have very high expectations | 09:50 |
| kanzure | i see what you are saying but there's a massive amount of hand-holding that you will have to do | 09:50 |
| fenn | maaku: jake cannell is worth talking to about neural networks, but i don't know where he hangs out online | 09:50 |
| maaku | kanzure: hah you mean of the software or the people? | 09:50 |
| kanzure | i have never encountered an agi project where programmers don't have to do massive amounts of setup work to hand-hold the damn thing | 09:51 |
| maaku | the whole enterprise is a moonshot, you have to aim high. people aren't generally so reliable though | 09:51 |
| kanzure | integration work, writing new parsers, templates, etc. | 09:51 |
| kanzure | also, if your data set is not already tagged, then you are even worse off because you have to do semantic data labeling to even begin to train anything- which requires even more manual labor | 09:52 |
| maaku | kanzure: I'm exclusively interested in unsupervised training for that reason... | 09:53 |
| maaku | but as mentioned above, pagerank applied to citations can give you some measure of automated supervised training for a given data set | 09:53 |
| kanzure | if you can find me an example of unsupervised learning where there was minimal or zero integration work to get interesting results, like no manual human programmer intervention to do setup work, then you might be able to convince me of... something.. | 09:54 |
| maaku | kanzure: that's the goal, not where we are | 09:55 |
| kanzure | most of my machine learning work has been extremely similar to all of my web scraping work: tons and tons of manual intervention and ugh | 09:55 |
| c0rw1n | huh, if you can generate the next AI papers with a trained AI, does that count as recursive self-improvement | 09:55 |
| maaku | c0rw1n: yes | 09:55 |
| c0rw1n | yay | 09:55 |
| kanzure | this is going to end up with one of those runaway effects like "great now we have a lot of worthless papers being generated, and human labor has to sort out this mess" | 09:56 |
| fenn | heh i had the same problem, formatting data correctly just to get it to parse takes more work than running the machine learning algorithm | 09:56 |
| kanzure | fenn: to be fair, that might just be my extreme cynicism showing through.... | 09:56 |
| fenn | also PDF is a terrible way to transmit data | 09:57 |
| maaku | fenn: I'd previously been of the opinion that it would be better to render papers as images and have the machine consume it the same way people do ... | 09:58 |
| maaku | shoves parsing into a computer vision problem | 09:58 |
| pasky | fenn: yeah it's the second-ranked answer, it does need more & better features | 09:58 |
| pasky | right now we are trying to find some good (and sufficiently simple) commercial applications of these systems, as well as add more deep learning to the mix | 09:59 |
| kanzure | i am not so convinced about making it computer vision only; i agree that it is a computer vision problem but, having the other data available is quite useful. | 09:59 |
| -!- Filosofem [~Jawmare@unaffiliated/jawmare] has joined ##hplusroadmap | 09:59 | |
| fenn | what went wrong there? why did it parse "socializing these dogs is a must" to mean "dogs is a must"? | 09:59 |
| maaku | pasky: this is a commercial effort? | 10:00 |
| maaku | kanzure: no reason to not do both. | 10:00 |
| kanzure | i think nmz787 proposed using opencv for doing image segmentation of papers, to isolate graphs and figures. maybe that was fenn.. | 10:00 |
| fenn | oh there is a tool that does this exact thing already | 10:01 |
| c0rw1n | the vision part would just return characters and then shove them into word2vec ... | 10:01 |
| kanzure | (spitting out pdf objects is not enough, since figures often come with surrounding explanatory text and titles) | 10:01 |
| c0rw1n | and diagrams would be classified to be parsed otherwise ... ah, the position of the diagrams in the text is also valuable | 10:02 |
| maaku | er, no, I meant something more like either DeepMind or OpenCog's perceptual system. | 10:02 |
| -!- Jawmare [~Jawmare@unaffiliated/jawmare] has quit [Ping timeout: 252 seconds] | 10:02 | |
| maaku | not hand-crafted rules like OpenCV | 10:03 |
| maaku | http://goertzel.org/DeSTIN_OpenCog_paper.pdf | 10:03 |
| fenn | ugh why do people keep using these awful top level domains | 10:04 |
| fenn | i feel embarrassed even linking to it | 10:04 |
| kanzure | i hope this is x.ai | 10:04 |
| fenn | tabula.technology | 10:04 |
| fenn | can you tell it's a link? | 10:05 |
| kanzure | yes but only because i recently weeped over a list of gTLDs | 10:06 |
| fenn | there's both .car and .cars | 10:06 |
| -!- sandeepkr__ [~sandeep@111.235.64.4] has quit [Ping timeout: 244 seconds] | 10:10 | |
| -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-kkimnnyujoknfayb] has quit [Quit: Connection closed for inactivity] | 10:10 | |
| pasky | fenn: i'd wager a parser error which decide it's of the form <adjphrase> dogs -is_a- must; i'm a bit lazy to look at the log files now :) | 10:13 |
| pasky | next step for me is throw a lot of this away and replace it with some deep nets that'll also select the answer-bearing tokens, but i might need a larger dataset of questions for that, and noise is a problem if you don't want humans annotating hundreds of thousands of sentences | 10:14 |
| fenn | would it help to have anonymous visitors pick the most correctest answer from a list? | 10:15 |
| -!- kanzure [~kanzure@unaffiliated/kanzure] has quit [Ping timeout: 276 seconds] | 10:15 | |
| pasky | kanzure: I think my colleague could do this easily if you can get a gold standard (some pdfs where the table regions are selected); he's working on having deep convolutional NNs segment webpages (eshops in particular for now - identifying product name, image and pricetag) based on their screenshots + info about charbigrams at that place in the webpage | 10:16 |
| -!- kanzure [~kanzure@bryan.fairlystable.org] has joined ##hplusroadmap | 10:16 | |
| pasky | and it works pretty nice | 10:16 |
| pasky | oh | 10:16 |
| pasky | kanzure: I think my colleague could do this easily if you can get a gold standard (some pdfs where the table regions are selected); he's working on having deep convolutional NNs segment webpages (eshops in particular for now - identifying product name, image and pricetag) based on their screenshots + info about charbigrams at that place in the webpage | 10:16 |
| pasky | fenn: it would! a lot of them do, thankfully | 10:16 |
| kanzure | why use machine learning for segmenting webpages? why not just analyze DOM and stylesheets. | 10:16 |
| pasky | kanzure: because every eshop does it differently | 10:17 |
| pasky | and looks differently | 10:17 |
| pasky | but by looking at the page, humans have no trouble with this | 10:17 |
| pasky | maaku: i'm doing my phd by this, but i also need some money for living and we are trying to set up a startup around this nn stuff we are doing | 10:18 |
| pasky | but so far we have more product ideas for the segmentation task than for the question answering / information extraction from text task :( | 10:19 |
| fenn | maybe you're too smart | 10:19 |
| fenn | i have a theory that it's impossible to think like someone dumber than you | 10:20 |
| maaku | Thankfully mechanical Turk is a thing. | 10:21 |
| pasky | possibly, i have trouble thinking of what people or companies want or how to easily find out | 10:21 |
| fenn | so you should focus on making your QA bot answer things you actually want to know, rather than "anything at all" which seems to be targeted at a hypothetical average internet person | 10:21 |
| fenn | keep a list of questions | 10:22 |
| maaku | Tell me about it. I'm in the midst of a Solve All The Problems startup. | 10:23 |
| pasky | fenn: one idea we had was to have it hooked up to stackoverflow and possibly even paste the code from top-rated question | 10:23 |
| fenn | eh sounds dangerous | 10:23 |
| fenn | presumably you'd be using this on a command line and automatically executing it | 10:23 |
| pasky | that does sound dangerous :) i thought more of uttering "for cycle in javascript" and it'd paste it to where my cursor is in the code | 10:24 |
| fenn | "Your own gTLD demonstrates confidence and vision and may accelerate your brand and its value. An internet address at the Top Level is far better than registration at the 'low rent' Second Level." | 10:33 |
| fenn | for the low low price of $185,000 you too can participate in the next generation of creativity and inspiration | 10:35 |
| kanzure | apparently last weekend i was hanging out with an ICANN board member, https://aso.icann.org/aso-icann/ac-appointments-to-icann-board/previous-aso-appointments-to-the-icann-board-2/pindar-wong/ | 10:39 |
| fenn | i just have to wonder what the fuck were they thinking? | 10:40 |
| cluckj | get. money. | 10:40 |
| kanzure | they have formed a sovereign political organization to manage the interwebs as of.. uh.. last year or something. | 10:41 |
| maaku | Can I be a sovereign too ? | 10:43 |
| fenn | sure, if you buy this plot of land in south sudan | 10:43 |
| kanzure | i feel like this is an unreasonably dangerous requirement | 10:46 |
| maaku | Guess it's back to the old fashioned way | 10:54 |
| fenn | put on your napoleon hat | 10:55 |
| kanzure | https://www.xkcd.com/simplewriter/ | 10:59 |
| maaku | I could use one of those hats | 11:00 |
| -!- Houshalter [~Houshalte@oh-71-50-63-56.dhcp.embarqhsd.net] has joined ##hplusroadmap | 11:02 | |
| maaku | pasky: have you considered an artificial patent examiner? | 11:16 |
| maaku | Could be business there. | 11:16 |
| kanzure | all patents should be re-purposed as fire kindling | 11:22 |
| c0rw1n | ^ yes | 11:23 |
| c0rw1n | good idea : make a patent-overlap-finder and public dump the graph of precisely how much they all invalidate each other | 11:26 |
| maaku | .title http://blog.samaltman.com/hard-tech-is-back | 11:27 |
| yoleaux | Hard Tech is Back - Sam Altman | 11:27 |
| kanzure | c0rw1n: https://groups.google.com/d/msg/openmanufacturing/vS4ju1VqXb0/jD_TZ8U47b4J | 11:28 |
| maaku | well that's the pitch to patent offices -- look our software found all this prior art that invalidates all these patents, pay us a huge subscription fee and you can use this data moving forward | 11:36 |
| maaku | patents need to die i a fire, but this at least makes the system better and creates useful tech for wha twe're doing here | 11:37 |
| kanzure | they are a mostly political thing. nothing about the patent system is a good tech decision at all. | 11:42 |
| maaku | anyone building a business needs to be willing to hold their nose with respect to revenue opportunities. you're getting someone else to give you money, so you need to work on their terms | 11:43 |
| c0rw1n | ( you can also blackmail all the patent trolls for more money than has ever been invented , thereby solving all your funding problems forever ) | 11:44 |
| maaku | c0rw1n: meh, they're setup to survive that | 11:51 |
| c0rw1n | really not worth a try? :( | 11:52 |
| maaku | see newegg vs everybody | 11:53 |
| -!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:5196:be:65fb:8473] has joined ##hplusroadmap | 11:56 | |
| -!- ArturShaik [~ArturShai@195.114.248.236] has quit [Ping timeout: 240 seconds] | 12:17 | |
| -!- kanzure [~kanzure@bryan.fairlystable.org] has quit [Changing host] | 12:18 | |
| -!- kanzure [~kanzure@unaffiliated/kanzure] has joined ##hplusroadmap | 12:18 | |
| pasky | maaku: i have considered it but it is a hard problem that has very little alternative applications, so it carries a big opportunity cost; but i guess my main reason is that it's soul-eating stuff that doesn't sound like fun at all to me | 12:21 |
| -!- jdqx [uid146399@gateway/web/irccloud.com/x-hrscuzyzbghasifj] has joined ##hplusroadmap | 12:23 | |
| Aurelius_Work | anyone else here play factorio? | 12:24 |
| c0rw1n | o/ | 12:24 |
| Aurelius_Work | c0rw1n : my buddy has decided to try to make an adaptive forecaster | 12:26 |
| fenn | heh i just started | 12:26 |
| Aurelius_Work | Factorio : The Great Nerdsniper | 12:26 |
| c0rw1n | hehe yes ^_^ | 12:26 |
| fenn | .title http://youtu.be/9yDZM0diiYc | 12:29 |
| yoleaux | Factorio - Trailer 2014 - YouTube | 12:29 |
| -!- nmz787_i [~ntmccork@134.134.137.73] has joined ##hplusroadmap | 12:34 | |
| nmz787_i | sup | 12:36 |
| fenn | swarmsim x factorio = exponentially growing bug colony | 12:36 |
| fenn | i'd like to see some competitive professional-level factorio play | 12:36 |
| kanzure | nmz787_i: thinking about decapping a 21 inc bitcoin mining asic | 12:36 |
| nmz787_i | ah, cool | 12:38 |
| nmz787_i | I still haven't thrown the CPU I decapped recently under a microscope or ion beam | 12:38 |
| -!- jtimon [~quassel@35.31.134.37.dynamic.jazztel.es] has quit [Ping timeout: 264 seconds] | 12:55 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-kbbafmbvbalprtiv] has quit [Quit: Connection closed for inactivity] | 13:21 | |
| -!- psyreal [~psyreal@104.200.154.94] has joined ##hplusroadmap | 13:26 | |
| -!- yorick [~yorick@ip51cd0513.speed.planet.nl] has joined ##hplusroadmap | 13:31 | |
| -!- yorick [~yorick@ip51cd0513.speed.planet.nl] has quit [Changing host] | 13:31 | |
| -!- yorick [~yorick@oftn/oswg-member/yorick] has joined ##hplusroadmap | 13:31 | |
| -!- yorick__ [~yorick@oftn/oswg-member/yorick] has quit [Ping timeout: 244 seconds] | 13:32 | |
| -!- yorick [~yorick@oftn/oswg-member/yorick] has quit [Ping timeout: 244 seconds] | 13:35 | |
| -!- yorick [~yorick@ip51cd0513.speed.planet.nl] has joined ##hplusroadmap | 13:36 | |
| -!- yorick [~yorick@ip51cd0513.speed.planet.nl] has quit [Changing host] | 13:36 | |
| -!- yorick [~yorick@oftn/oswg-member/yorick] has joined ##hplusroadmap | 13:36 | |
| kanzure | factorio looks like an extremely industrious dude | 13:54 |
| pasky | oh hey i know the directors of the factorio company! tomas was my sensei when i was learning go | 13:54 |
| kanzure | he's going to put the martian out of business | 13:54 |
| -!- irseeyou [~irseeyou@c-67-168-101-20.hsd1.wa.comcast.net] has joined ##hplusroadmap | 14:12 | |
| -!- jtimon [~quassel@35.31.134.37.dynamic.jazztel.es] has joined ##hplusroadmap | 15:06 | |
| -!- sandeepkr__ [~sandeep@111.235.64.4] has joined ##hplusroadmap | 15:34 | |
| -!- SDr [~SDr@unaffiliated/sdr] has joined ##hplusroadmap | 15:48 | |
| juri_ | hio. | 15:50 |
| juri_ | been very busy. | 15:50 |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-ndsdzrahyktvheqi] has joined ##hplusroadmap | 15:57 | |
| -!- sandeepkr__ [~sandeep@111.235.64.4] has quit [Ping timeout: 248 seconds] | 15:58 | |
| -!- jaboja64 [~jaboja@emf122.neoplus.adsl.tpnet.pl] has quit [Remote host closed the connection] | 16:02 | |
| -!- sandeepkr__ [~sandeep@111.235.64.4] has joined ##hplusroadmap | 16:05 | |
| -!- irseeyou [~irseeyou@c-67-168-101-20.hsd1.wa.comcast.net] has quit [Quit: Leaving.] | 16:06 | |
| maaku | Factorio is The Last Starfighter for hplus projects | 16:16 |
| -!- psyreal [~psyreal@104.200.154.94] has quit [] | 16:38 | |
| -!- sandeepkr__ [~sandeep@111.235.64.4] has quit [Ping timeout: 260 seconds] | 16:47 | |
| -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-xuxwmnpeoqykhznp] has joined ##hplusroadmap | 17:18 | |
| -!- |node [uid125132@gateway/web/irccloud.com/x-ljnxzeenvcycbnts] has joined ##hplusroadmap | 17:21 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-ndsdzrahyktvheqi] has quit [Quit: Connection closed for inactivity] | 17:41 | |
| kanzure | http://www.theatlantic.com/science/archive/2016/02/the-unexplored-marvels-locked-away-in-our-natural-history-museums/459306/?single_page=true | 17:58 |
| -!- |node [uid125132@gateway/web/irccloud.com/x-ljnxzeenvcycbnts] has quit [] | 18:01 | |
| -!- Gurkenglas [Gurkenglas@dslb-188-106-113-006.188.106.pools.vodafone-ip.de] has quit [Ping timeout: 248 seconds] | 18:05 | |
| -!- nmz787_i [~ntmccork@134.134.137.73] has quit [Ping timeout: 244 seconds] | 18:36 | |
| -!- yashgaroth [~yashgarot@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has joined ##hplusroadmap | 18:36 | |
| -!- drewbot [~cinch@ec2-54-205-107-25.compute-1.amazonaws.com] has quit [Remote host closed the connection] | 19:05 | |
| -!- drewbot [~cinch@ec2-54-80-238-96.compute-1.amazonaws.com] has joined ##hplusroadmap | 19:05 | |
| -!- ArturShaik [~ArturShai@195.114.248.236] has joined ##hplusroadmap | 19:50 | |
| -!- AmbulatoryCortex [~Ambulator@173-31-155-69.client.mchsi.com] has quit [Quit: Leaving] | 19:53 | |
| -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-xuxwmnpeoqykhznp] has quit [Quit: Connection closed for inactivity] | 20:00 | |
| -!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:5196:be:65fb:8473] has quit [Read error: Connection reset by peer] | 22:06 | |
| -!- xrr [~quassel@c21-76.uvn.zone.eu] has quit [Remote host closed the connection] | 23:03 | |
| -!- xrr [~quassel@c21-76.uvn.zone.eu] has joined ##hplusroadmap | 23:04 | |
| -!- yashgaroth [~yashgarot@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has quit [Quit: Leaving] | 23:16 | |
| -!- xrr [~quassel@c21-76.uvn.zone.eu] has quit [Remote host closed the connection] | 23:17 | |
| -!- xrr [~quassel@c21-76.uvn.zone.eu] has joined ##hplusroadmap | 23:18 | |
| -!- irseeyou [~irseeyou@c-67-168-101-20.hsd1.wa.comcast.net] has joined ##hplusroadmap | 23:33 | |
| JayDugger | Good morning. | 23:40 |
| -!- irseeyou [~irseeyou@c-67-168-101-20.hsd1.wa.comcast.net] has left ##hplusroadmap [] | 23:48 | |
| -!- Houshalter [~Houshalte@oh-71-50-63-56.dhcp.embarqhsd.net] has quit [Quit: Quit] | 23:55 | |
| --- Log closed Sat Mar 12 00:00:57 2016 | ||
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!
