--- Log opened Wed Aug 16 00:00:51 2017
00:00 < fltrz> in this way the whole database is replicated, in parts by subject. those who want the complete db no longer need to download it, they just need to say hello to some students from each faculty
00:01 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has quit [Remote host closed the connection]
00:02 < fltrz> without partitioning it, much fewer people would get the humongous db, and those who want it but can't afford (in money/time/hassle) the connection wont at all
00:03 < fltrz> also, tracking offline copying behaviour is going to be a bit harder
00:04 -!- duper [~super@pitbull.blackcatz.org] has joined ##hplusroadmap
00:14 < fltrz> hmm, perhaps provable geocaching or smth
00:32 -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-hweobyjufygbkmzw] has quit [Quit: Connection closed for inactivity]
00:35 < fltrz> how much would it cost as a ROM mask?
00:36 < fltrz> design a ROM mask with encrypted contents, distribute, release key?
00:51 -!- augur [~augur@198-27-215-123.static.sonic.net] has joined ##hplusroadmap
01:02 -!- poppingtonic [~brian@unaffiliated/poppingtonic] has joined ##hplusroadmap
01:08 -!- poppingtonic [~brian@unaffiliated/poppingtonic] has quit [Ping timeout: 260 seconds]
01:28 -!- augur [~augur@198-27-215-123.static.sonic.net] has quit [Remote host closed the connection]
01:30 -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-ndlqgsmqlzezhvya] has joined ##hplusroadmap
01:32 -!- poppingtonic [~brian@unaffiliated/poppingtonic] has joined ##hplusroadmap
01:51 < fltrz> theres absolutely no good reason why 10.1016/j.jpainsymman.2016.11.014 should be 10MB, this is just poor publishing standards
01:52 -!- CRM114 [~urchin@unaffiliated/urchin] has joined ##hplusroadmap
01:52 < fltrz> I actually believe we can create a tool to detect the problematic cases, and using pdf editing libraries (inkscape libs or whatever) identify the offending element
01:53 < fltrz> I guess in this case the author/publisher converted a vector graphic to high resolution raster graphic
01:56 < fltrz> another example I checked was just 10 pages but 30MB: it did have one page with high resolution scanned BW photos from a printed paper article, but at too high a resolution (the dithering of the toner was resolvable and even caused moire patterns...)
01:58 < fltrz> compared to 60kB articles this is 500x ... I believe the database can be made about 200 times smaller with appropriate strategy and automation.
01:58 < fltrz> just like theres donate buttons on sci-hub, there could be a "this article seems messed up, show me the original" button
01:59 < fltrz> if sci-hub keeps track of how often a user clicks the original button, problematic conversions can be identified
02:00 < fltrz> a small team could then inspect the article and classify them, so we have statistics on what went wrong and why
02:01 < fltrz> every instance of too large filesize is always someone being conservative (the author in choosing to high a resolution for an embedded image, or the digitization people who did not dare downsample a scan)
02:02 < fltrz> I don't mind the scan being high res, its easy to downsample in software compared to having to scan again.. its the publisher fault really
02:03 < fltrz> if sci-hub becomes the new publisher it might as well one-up the original publishers at their own job
02:04 < fltrz> @2017-04-07 the article count was 64 million
02:04 < fltrz> libgen scimag
02:06 < fltrz> apparently sci-hub is not the storage of the papers but 1) the portal for users to access the papers 2) the mechanism by which access credentials are shared, when a new article is encountered however it is stored in libgen
02:11 < fltrz> may 30 2016: "Around 53 million papers, ~18 TB" so that would make the current estimate for 64M papers 22 TB
02:12 < fltrz> divide by 100 (I think divide by 200 would certainly be possible) => 220GB
02:12 < fltrz> this starts to get in the direction of flash storage
02:14 < fltrz> if they want to stop it, they have to stop it right now, with a relatively small team I believe these papers can be compacted tremendously and then the ghost will be out of the bottle
02:14 < fltrz> it probably already is
02:18 < fltrz> if only Aaron could see this day
02:27 < fltrz> for rasterized vector art like the diagrams in the paper 10.1016/j.jpainsymman.2016.11.014 for example, the pdf elements could be detected by size, and then the color space histogram would detect that only a few colors are used
02:28 < fltrz> hence either raster -> vector conversion or (jpg or whatever) raster -> (GIF or whatever) raster could be easily automated
02:29 < fltrz> instead of the captcha's on sci-hub, it could present an ABX test, where you see three images, the original left, the conversion right, and in the middle unknown to the user either original or conversion, and the user must classify as middle==left or middle==right
02:31 < fltrz> the first say 20 users would determine ground truth, after which it can serve as a real captcha
02:31 < fltrz> if users fail to classify, its probably a good conversion
02:38 -!- y0ur1 [~y0ur1@gateway/tor-sasl/y0ur1] has quit [Read error: Connection reset by peer]
02:40 -!- justan0theruser [~justanoth@unaffiliated/justanotheruser] has quit [Quit: WeeChat 1.7.1]
03:06 -!- jnplx [~jnplx@h57.89.213.151.dynamic.ip.windstream.net] has quit [Ping timeout: 260 seconds]
03:35 < kanzure> perhaps it would make more sense to do something like: for each pdf that a user downloads, they also download 9 other files, appended at the end.
03:35 < kanzure> and the 9 others are random
03:39 < kanzure> .tw https://twitter.com/petertoddbtc/status/897633067673821185
03:39 < yoleaux> @jermhu You put your mining nodes in space. Point is, the *profit* is being returned as data, rather than beamed power. (@petertoddbtc, in reply to tw:897632716644122625)
03:42 -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-ndlqgsmqlzezhvya] has quit [Quit: Connection closed for inactivity]
04:21 -!- y0ur1 [~y0ur1@gateway/tor-sasl/y0ur1] has joined ##hplusroadmap
04:23 -!- cluckj [~cluckj@static-98-114-202-142.phlapa.ftas.verizon.net] has quit [Ping timeout: 240 seconds]
04:24 -!- cluckj [~cluckj@static-98-114-202-142.phlapa.ftas.verizon.net] has joined ##hplusroadmap
04:27 -!- y0ur1 [~y0ur1@gateway/tor-sasl/y0ur1] has quit [Quit: y0ur1]
04:33 -!- y0ur1 [~y0ur1@gateway/tor-sasl/y0ur1] has joined ##hplusroadmap
04:37 -!- jtimon [~quassel@143.29.134.37.dynamic.jazztel.es] has joined ##hplusroadmap
05:08 -!- Gurkenglas_ [~Gurkengla@dslb-178-000-218-087.178.000.pools.vodafone-ip.de] has joined ##hplusroadmap
05:15 -!- streety [streety@2a01:7e00::f03c:91ff:feae:ded6] has left ##hplusroadmap []
05:37 -!- y0ur1 [~y0ur1@gateway/tor-sasl/y0ur1] has quit [Quit: y0ur1]
05:37 -!- jaboja [~jaboja@jaboja.pl] has joined ##hplusroadmap
05:40 < kanzure> don't most pdfs have font definitions inside of them? those should be removed.
05:46 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
06:00 -!- Gurkenglas_ [~Gurkengla@dslb-178-000-218-087.178.000.pools.vodafone-ip.de] has quit [Ping timeout: 240 seconds]
06:09 < TMA> kanzure: that would render some texts illegible -- those that have their letters scrambled so that an a looks like an n you would get something like "gung jbhyq eraqre fbzr grkgf vyyrtvoyr -- gubfr gung unir gurve yrggref fpenzoyrq fb gung na n ybbxf yvxr na a lbh jbhyq trg fbzrguvat yvxr" then
06:14 < pasky> but these are nowadays relatively rare, at least if you aren't intentionally obfuscating
06:22 -!- g0d355__ [~lmao@104.131.75.159] has joined ##hplusroadmap
06:37 -!- jtimon [~quassel@143.29.134.37.dynamic.jazztel.es] has quit [Ping timeout: 255 seconds]
06:52 -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-zofqhmtlhaizlwyr] has joined ##hplusroadmap
06:54 -!- Gurkenglas_ [~Gurkengla@dslb-178-000-218-087.178.000.pools.vodafone-ip.de] has joined ##hplusroadmap
07:07 -!- jaboja [~jaboja@jaboja.pl] has quit [Ping timeout: 255 seconds]
07:12 -!- jcluck [~cluckj@static-98-114-202-142.phlapa.ftas.verizon.net] has joined ##hplusroadmap
07:16 -!- cluckj [~cluckj@static-98-114-202-142.phlapa.ftas.verizon.net] has quit [Ping timeout: 248 seconds]
07:22 < kanzure> what?
07:24 -!- jcluck is now known as cluckj
07:26 < kanzure> "The glEnd() of Zelda" https://www.cs.cmu.edu/~tom7/zelda/zelda.pdf
07:36 < kanzure> .title https://www.youtube.com/watch?v=xDxjbXAqTPg
07:36 < yoleaux> Automatic 3Dification of Nintendo games: The glEnd() of Zelda - YouTube
07:44 -!- fltrz [d5d38b31@gateway/web/freenode/ip.213.211.139.49] has quit [Ping timeout: 260 seconds]
07:53 -!- jaboja [~jaboja@jaboja.pl] has joined ##hplusroadmap
08:06 < kanzure> maybe you need recursive timelock: each party re-encrypts and transfers with a new timelock. first distributors can unlock the earliest.
08:21 < kanzure> i think the ideal is to target elimination of leaking originator identity bits (and also not screw over any of the users that participated in the distribution).
08:29 -!- jaboja [~jaboja@jaboja.pl] has quit [Ping timeout: 255 seconds]
08:35 -!- y0ur1 [~y0ur1@gateway/tor-sasl/y0ur1] has joined ##hplusroadmap
08:37 -!- emeraldgreen [~user@92.222.68.248] has joined ##hplusroadmap
08:38 -!- preview [~quassel@2407:7000:842d:4000::3] has joined ##hplusroadmap
08:42 -!- poppingtonic [~brian@unaffiliated/poppingtonic] has quit [Ping timeout: 255 seconds]
08:56 < kanzure> 07:53 <kanzure> hey can you use intermediate states in the timelock cracking to unlock other less-secret secrets?
08:56 < kanzure> 08:36 <petertodd> sure! intermediate states IIRC are handled no different than other states
09:19 -!- preview [~quassel@2407:7000:842d:4000::3] has quit [Remote host closed the connection]
09:38 -!- g0d355__ [~lmao@104.131.75.159] has quit [Ping timeout: 248 seconds]
09:44 -!- y0ur1 [~y0ur1@gateway/tor-sasl/y0ur1] has quit [Remote host closed the connection]
09:44 < kanzure> with softer timelocks you could have progressive unlocking of increasingly larger chunks
09:45 < kanzure> the first few might have only a handful of documents
09:45 < kanzure> and if there's a way to prove the content in the encrypted blob, that would be nice (commit to a merkle root, then prove that the items are all member of the tree, and then let some users choose some random items from the collection, and prove that they are part of the blob? i dunno.)
09:47 -!- y0ur1 [~y0ur1@gateway/tor-sasl/y0ur1] has joined ##hplusroadmap
09:48 < kanzure> y0ur1: greetings.
09:53 -!- maaku [~mark@173.234.25.100] has quit [Quit: Lost terminal]
09:56 -!- maaku [~mark@173.234.25.100] has joined ##hplusroadmap
10:03 -!- jtimon [~quassel@143.29.134.37.dynamic.jazztel.es] has joined ##hplusroadmap
10:32 < nmz787> are PDFs already compressed very well? could you avoid introspection and just compress with 7z or something better (if exists)>
10:32 < nmz787> ?
10:33 < kanzure> some of the objects inside of the pdf files are compressed using flate/deflate
10:34 < kanzure> ya better compression could be considered for the raw objects.
10:34 < kanzure> open up a pdf file in a text editor and you'll see a bunch of "deflate" stuff.
10:40 -!- preview [~quassel@2407:7000:842d:4000::3] has joined ##hplusroadmap
10:51 -!- emeraldgreen [~user@92.222.68.248] has quit [Quit: Leaving.]
10:54 < Taek> kanzure: can you explain the problem you highlighted me about a little better?
10:54 < Taek> seems like you want to be able to spread data around without people knowing whether it's legal or not?
10:55 < Taek> If the original uploader has the plaintext, all they need to do is encrypted 100 different ways and then upload each copy somewhere new
10:55 < kanzure> Taek: want to increase redundancy of data distribution, and then release the key using timelock encryption
10:55 < kanzure> mega.co.nz failed because everything was immediately decryptable and mega.co.nz was obviously the host
10:56 < kanzure> it needs to be more of a "push" model, not mega's "pull" model.
10:56 < Taek> If you give people the power to increase the redundancy, you also give them the power to deduplicate and cheat. If your model assumes that they are not being paid, then this is not an issue
10:56 < Taek> it's an issue for Sia because cheating hosts would make a lot of money by cheating
10:56 < Taek> but otherwise it's pretty easy to take some ciphertext, make it redundant, and then spread those redundant copies around
10:57 < kanzure> deduplication would kinda suck, but it's not entirely a problem in the timelock puzzle scenario (it's only a problem if the originator releases the decryption key once they measure a certain amount of redundancy)
10:57 < kanzure> (and the adversary could simply cause lots of fake duplication to appear. and then they capture the originator in the act.)
10:58 < kanzure> the goal is that by the time that the key is released (or cracked), the identifying bits of knowledge regarding the originators or where the data originally came from, is just gone from memory.
10:59 < Taek> The best thing I know for that is Tor
11:07 -!- poppingtonic [~brian@unaffiliated/poppingtonic] has joined ##hplusroadmap
11:13 -!- poppingtonic [~brian@unaffiliated/poppingtonic] has quit [Remote host closed the connection]
11:20 -!- jtimon [~quassel@143.29.134.37.dynamic.jazztel.es] has quit [Ping timeout: 240 seconds]
11:26 -!- esmerelda [~mabel@unaffiliated/jacco] has quit [Ping timeout: 240 seconds]
11:28 -!- fltrz [d5d38b31@gateway/web/freenode/ip.213.211.139.49] has joined ##hplusroadmap
11:29 -!- poppingtonic [~brian@unaffiliated/poppingtonic] has joined ##hplusroadmap
11:54 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has joined ##hplusroadmap
11:58 -!- emeraldgreen [~user@92.222.68.248] has joined ##hplusroadmap
11:59 -!- emeraldgreen [~user@92.222.68.248] has quit [Client Quit]
12:00 -!- emeraldgreen [~user@92.222.68.248] has joined ##hplusroadmap
12:10 -!- poppingtonic [~brian@unaffiliated/poppingtonic] has quit [Ping timeout: 240 seconds]
12:27 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has quit [Ping timeout: 246 seconds]
12:51 -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-zofqhmtlhaizlwyr] has quit [Quit: Connection closed for inactivity]
13:12 -!- jaboja [~jaboja@jaboja.pl] has joined ##hplusroadmap
13:29 -!- jaboja [~jaboja@jaboja.pl] has quit [Ping timeout: 260 seconds]
13:47 -!- Gurkenglas_ [~Gurkengla@dslb-178-000-218-087.178.000.pools.vodafone-ip.de] has quit [Ping timeout: 248 seconds]
13:54 -!- jaboja [~jaboja@jaboja.pl] has joined ##hplusroadmap
14:38 -!- sachy [~sachy@nat.brmlab.cz] has quit [Quit: Leaving.]
14:46 < fenn> .title http://youtu.be/7f6WvIkZ9wU
14:46 < yoleaux> Sam Altman Live Q&A : Open AI, Future Technology, Basic Income - YouTube
14:46 < fenn> seems like he's got his head on straight
14:46 < fenn> 10 years before AGI seems like too long of an estimate to me
14:47 < fenn> wish they had put a transcript in the description
14:55 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has quit [Remote host closed the connection]
14:56 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has joined ##hplusroadmap
15:00 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has quit [Ping timeout: 264 seconds]
15:04 < fltrz> fenn: about timeframe I have no prediction but about if the basic income will happen: it depends on what exactly this basic income is.
15:05 -!- emeraldgreen [~user@92.222.68.248] has quit [Ping timeout: 255 seconds]
15:05 < fltrz> the most popular conception will never happen (people mistakenly think they want it, but they want something even better and this better thing will happen)
15:14 < fenn> what is "the most popular conception"
15:14 < fenn> marshall brain style foamcrete prison complexes?
15:15 < fenn> i really have no idea what people are saying about basic income in general
15:22 -!- emeraldgreen [~user@188.227.115.178] has joined ##hplusroadmap
15:39 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has joined ##hplusroadmap
15:51 < fltrz> I think most popular conception of basic income is: everybody gets an unconditional income of X dollars/euros per month/day, off course you can get a job and earn more on top, or you can chillax and keep the economy running
15:52 < fltrz> problem 1: what is the 'just' amount of X? what happens when currencies change value? it is not pegged to a concrete reference
15:54 < fltrz> problem 2: prices of food and housing will simply rise (people will have more money to bid on housing and food or alternatively if supermarkets and housing companies/owners know everybody has more money, they will just ask more for food and rent
15:55 < fltrz> note that with 'follow the money' problem 2 suggest a solution to both problem 1 and 2:
15:55 < fltrz> oh I forgot to mention problem 3: where do we source all this basic income from?
15:56 < fltrz> solution: everybody gets the average rent. number of people N stimes average A = total T spent on rent
15:57 < fltrz> so if you rent an average place its basically free since you get average rent, if you want to live in more than average place you take a job
15:57 < fltrz> if you satisfied with less than average place, you have money left over, in a sense "making space" then is a job
15:58 < cluckj> why not cut all social and public services to fund it, and make everyone pay for what they use out of their basic income?
15:58 < cluckj> </s>
15:59 < fltrz> BUT it is a reappropriation of the world, we are effectively saying the world belongs to everybody, so total revenue of rent is divided equally. so it pays for itself, everybody gets average rent
16:00 < fltrz> cluckj: I would see average rent as substituting social services, but not necessarily all public services
16:00 < kanzure> alright this is gettig off-topic
16:00 < cluckj> kanzure, you're welcome :P
16:00 < kanzure> your society sucks and you suck for being born into it.
16:00 < fltrz> the system I describe is simultaneously free market (i.e. highest rent bidder gets to rent a place) and communist (the world belongs to everybody, so we are equal owners and receive equal share of worldrent)
16:01 < kanzure> this is not the specialty of hplusroadmap--- unless you propose forcing everyone to upload their brains and forcing them to eternal parallel hells of endless labor.
16:02 < fltrz> I am definitely against  1) forcing people to upload brains or 2) forcing people to work
16:02 < kanzure> i think the most political we get is something like "everyone should have a complete copy of all relevant technology, and if they want to go fuck off and try a new political experiment somewhere else in the galaxy, they are perfectly welcome to".
16:03 < kanzure> fltrz: there was once a user in here who alleged that i am part of a conspiracy to force everyone to upload their brains.
16:03 < fltrz> I view politics as technology
16:03 < kanzure> it's a waste of time.
16:03 < fltrz> why did he come to such a conclusion?
16:04 < kanzure> people are extremely bad at politics. it's off-topic. banhammerworthy.
16:04 < cluckj> technology is politics made concrete *flees*
16:05 < kanzure> at most you can say biology is technology. "politics" is just noise.
16:05 < fltrz> yeah, bandwagonry self-declared as politics does not make it politics
16:05 < kanzure> i don't care. get out.
16:06 < fltrz> dont worry, the subject is out ;)
16:06 < kanzure> here is an overview of this channel: http://diyhpl.us/wiki/hplusroadmap/
16:06 < cluckj> oh no, don't drag me into this
16:07 < cluckj> the no politics/philosophy is about keeping bounding the channel focus on topic
16:07 < cluckj> -keeping
16:10 < kanzure> also, i don't think we should be responsible for figuring out basic income. that's not our job.
16:10 < fltrz> kanzure: I'm keeping a text document with possible approaches for compacting the scimag terrabase, regarding the whole issues vs couple of pages: the sql dump contain first and last page entries according to DOI, if the actual pdf page count is substantially larger those files can be highlighted
16:11 < kanzure> you got lotta storage available? you should go see what the seeder situation is for those torrents.
16:11 < fltrz> I don't have lots of storage (a single 1TB drive, partially used) nor lots of connectivity
16:12 < fltrz> but I got the tsv (as in csv) file of the 65M papers as of april
16:15 < fltrz> Im making a Lorentz curve for the filesize and a Lorent curve for the (filesize/pagecount)
16:18 < fltrz> ideally this all results in me writing a command line application that can take a pdf, and returns any "issue" with a suggestion code
16:20 < fltrz> then if scimag people would run it on their db, we could see which kind of issues would result in most space saving
16:22 < fltrz> some documents are at least 1000x larger than some other docs
16:24 < kanzure> oh you want bloat stats
16:24 < fltrz> yep
16:25 < fltrz> I just got to write some code to go through the 65M line tsv file
16:26 < kanzure> i'm worried that some pdf files might be using float coordinates for something that is imperceptibly close to rational integer divisions of the page or something, for alignment.  and adding extra bytes for floats in every single file... geeze.
16:26 < fenn> sorry for bringing up basic income, that wasn't really the maint point of the sam altman interview
16:26 < fenn> main*
16:28 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has quit [Remote host closed the connection]
16:29 < fenn> there's also the option of just dropping the top 1st percentile of files, and seeing how much data that trims off
16:33 < fltrz> fenn: exactly thats what youll see with the Lorentz curve https://en.wikipedia.org/wiki/Lorenz_curve
16:33 < fenn> er, are we talking about basic income again? :P
16:33 < fltrz> horizontal axis would be % of database by paper count
16:34 < fltrz> vertical would be total size for that bottom percentage %
16:36 < fltrz> the second lorentz curve would be horizontal: bottom percentage sorted by bytes/page, to maximize total pagecount instead of papercount (without actually splitting up papers)
16:36 -!- jtimon [~quassel@143.29.134.37.dynamic.jazztel.es] has joined ##hplusroadmap
16:36 < fltrz> i.e. a file would be allowed to be large if it also represented many pages
16:37 < fenn> thanks for doing this analysis
16:37 < fltrz> and a relatively small file could be rejected if it was just one page that took the size of an average 10 page document
16:37 < fltrz> fenn: it's not done yet ;)
16:38 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has joined ##hplusroadmap
16:38 < fenn> yes but it's actually useful to know the result
16:38 < kanzure> i think you should run your bloat stat tool on the arxiv data set.
16:38 < fenn> we can also sample the distribution to see what makes a file larger than others
16:38 < kanzure> (or the nature data set that we don't talk about)
16:39 < fenn> shh
16:39 < fltrz> I hope it will turn out useful, but if so I'm not sure the analysis will reach the people from scimag
16:39 < kanzure> elbakyan is on my buddy list and she replies from time to time.
16:39 < fltrz> kanzure: I don't need the actual files for this analysis (assuming the columns in the index file are accurate)
16:39 < kanzure> also if you put your analysis in a pdf and slap it up on arxiv, then people are bound to look at it eventually.
16:40 < kanzure> (also you should minimize the pdf file size just for laughs.)
16:40 < fltrz> kanzure: haha lol
16:40 < fltrz> demoscene pdf?
16:40 < fenn> append 10TB of zeroes
16:41 < kanzure> "arxiv refuses 2 page scihub analysis paper".
16:41 < fltrz> fenn: about sampling to see the cause/issue, thats exactly what I have in mind, I will randomly sample to top x percent and then fetch these random samples and see for myself
16:42 < kanzure> "also, if you load it as mimetype application/png, it's a picture of a cat."
16:43 < fltrz> yesterday there was also a (slightly older) paper posted here that had stats of percentage of scihub coverage for closed and open access papers separately, I'll check their methods out
16:44 < fltrz> to see if I can selectively sample the open access papers from the top of the distribution, so that people can replicate results if I claim things like "27% of issues is due to unnecessarily rasterized vector graphics" etc
16:51 -!- emeraldgreen [~user@188.227.115.178] has quit [Quit: Leaving.]
16:52 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
16:52 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has quit [Client Quit]
16:52 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
16:53 < fltrz> Just as computers ship with an OS by default, I can imagine a future where consumer grade computers ship with humanities library by default
16:57 -!- jaboja [~jaboja@jaboja.pl] has quit [Ping timeout: 240 seconds]
16:59 < fenn> i'd be satisfied with just wikipedia tbh
17:08 -!- Qfwfq [Qfwfq@gateway/shell/suchznc/x-ngzxhrzkzsvvfznj] has quit [Ping timeout: 255 seconds]
17:33 -!- jnplx [~jnplx@h57.89.213.151.dynamic.ip.windstream.net] has joined ##hplusroadmap
17:35 < adlai> fltrz: "making space" isn't a job, it's an album -- and not just any, but one of the best :P
17:39 -!- Gurkenglas_ [~Gurkengla@dslb-178-000-218-087.178.000.pools.vodafone-ip.de] has joined ##hplusroadmap
17:41 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has quit [Remote host closed the connection]
17:44 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has joined ##hplusroadmap
17:47 < fltrz> adlai: listening now
17:52 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has quit [Quit: Malvolio]
17:55 -!- CRM114 [~urchin@unaffiliated/urchin] has quit [Ping timeout: 258 seconds]
17:55 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has quit [Remote host closed the connection]
18:00 -!- augur [~augur@208.66.31.98] has joined ##hplusroadmap
18:03 -!- Qfwfq [Qfwfq@gateway/shell/suchznc/x-vchfbskswadwetlc] has joined ##hplusroadmap
18:03 -!- Qfwfq [Qfwfq@gateway/shell/suchznc/x-vchfbskswadwetlc] has quit [Changing host]
18:03 -!- Qfwfq [Qfwfq@unaffiliated/washirving] has joined ##hplusroadmap
18:03 -!- Qfwfq [Qfwfq@unaffiliated/washirving] has quit [Changing host]
18:03 -!- Qfwfq [Qfwfq@gateway/shell/suchznc/x-vchfbskswadwetlc] has joined ##hplusroadmap
18:05 -!- augur [~augur@208.66.31.98] has quit [Ping timeout: 264 seconds]
18:07 -!- mrdata- [~mrdata@unaffiliated/mrdata] has joined ##hplusroadmap
18:07 < kanzure> blort
18:09 < kanzure> .tw https://twitter.com/robinhanson/status/897892479646781440
18:09 < yoleaux> I critique accuracy of em depictions in @doctorow's Walkaway. Some stuff is right, but a whole lot looks very wrong. http://www.overcomingbias.com/2017/08/ems-in-walkaway.html (@robinhanson)
18:11 < superkuh> Walkaway was just a vehicle for Doctorow to distance himself from his prior (positive) views on meritocracy and reputation systems.
18:11 < superkuh> As such things are no longer politically correct.
18:12 < superkuh> The em stuff was just sort of tacked on at the end.
18:12 < mrdata-> meritocracy was never politically correct as such
18:13 < superkuh> It was in tech circles until the last couple years.
18:13 < mrdata-> what undermined it?
18:13 < superkuh> Lets not get into that.
18:13 -!- streety [streety@2a01:7e00::f03c:91ff:feae:ded6] has joined ##hplusroadmap
18:16 < fltrz> adlai: that was nice music (at least the 6 tracks I found as a youtube playlist)
18:17 < kanzure> gosh what's with the politics around here lately, am i gonna have to permaban everyone
18:18 < cluckj> we're in a heated political moment outside of this channel, it's hard to avoid
18:19 < kanzure> you're not trying hard enough.
18:20 -!- jaboja [~jaboja@jaboja.pl] has joined ##hplusroadmap
18:20 < kanzure> i am disappointed in all of you
18:20 < cluckj> lol
18:20 < kanzure> btw the origin of the word 'meritocracy' also sucks (it was some sort of warning about degrees/reputation) (whereas tech circles see meritocracy as something like "evaluate the idea and determine whether it is a good idea")
18:21 < cluckj> I...don't try at all
18:21 < fltrz> whats an "em" ?
18:21 < kanzure> it's a poor attempt at an abbreviation, he means "human brain emulations".
18:21 < fltrz> "Doctorow also presents the rather silly theory that World War I was caused by elites turning on each other because they ran out of “new worlds to plunder.” " < thats basically what we were taught in school
18:21 < kanzure> and he should just say "human brain emulation" instead.
18:22 < kanzure> alright i regret linking to a robin hanson thing. i should have thought better of it.
18:22 < cluckj> fltrz, lots of silly things get taught in school, especially in history classes
18:23 < kanzure> ttp://diyhpl.us/~bryan/papers2/neuro/Feasibility%20of%20whole%20brain%20emulation%20-%20Anders%20Sandberg.pdf
18:23 < kanzure> http://diyhpl.us/~bryan/papers2/neuro/Whole-brain%20imaging%20with%20single-cell%20resolution%20using%20chemical%20cocktails%20and%20computational%20analysis%20-%20Susaki%20-%202014.pdf
18:23 < kanzure> http://diyhpl.us/~bryan/papers2/brain-emulation-roadmap-report.pdf
18:23 < kanzure> http://diyhpl.us/~bryan/papers2/ai/nengo/BioSpaun:%20A%20large-scale%20behaving%20brain%20model%20with%20complex%20neurons%20-%20nengo%20-%202016.pdf
18:23 < kanzure> http://diyhpl.us/~bryan/papers2/neuro/Connectomics-based%20analysis%20of%20information%20flow%20in%20the%20Drosophila%20brain%20-%202015.pdf
18:23 < kanzure> ttp://diyhpl.us/~bryan/papers2/neuro/Reconstruction%20and%20simulation%20of%20neocortical%20microcircuitry.pdf
18:23 < kanzure> er, i mean, http://diyhpl.us/~bryan/papers2/neuro/Reconstruction%20and%20simulation%20of%20neocortical%20microcircuitry.pdf
18:23 < fltrz> I see, thats a poor attempt at abbreviation indeed
18:24 -!- justan0theruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
18:25 < kanzure> 09:53 < archels> "-is Dave going to be in the meeting? -yeah, why? -his hardware is so old, this meeting is going to take up so much of my objective time..."
18:25 < kanzure> 09:53 < kanzure> i think everything is going to be slow (for everything) and the costs of computation so high that sleep will be expensive
18:25 < kanzure> 09:54 < kanzure> if you can just double your speed or whatever, you would have done it already. cpu power doesn't just magically fall out of the sky- and when it does, software isn't guaranteed to run much faster on more hardware anyway.
18:25 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has quit [Ping timeout: 264 seconds]
18:28 < kanzure> 10:55 < archels_> not reading the book isn't really an option though
18:28 < kanzure> 10:57 < kanzure> it's definitely an option. if hanson is more imaginative than you about human brain emulation, then you already have serious problems that a book isn't going to fix.
18:28 < kanzure> 11:02 < kanzure> seems that his timeline is a little fucked up; he says 2 years total until there's enough simulated time to have created malicious+effective ai, but meanwhile says that the doubling rate would exhaust the supply of computational capacity without near constant expansion of equipment..... and i'm pretty sure near constant expansion of equipment cannot be brought online in under ...
18:28 < kanzure> ...2 years.
18:28 < kanzure> 11:03 < kanzure> e.g. 2 years doesn't make sense for all his breathless talk about 10,000x simulation rates and 1/1000000th simulation slowdown because where is all that equipment coming from so quickly?
18:29 < kanzure> 04:24 < TMA> "90% of everything is crap" principle. An uploaded da Vinci might be valuable asset, his uploaded illiterate contemporary less so (maybe few of them might satisfy someone's curiosity)
18:29 < kanzure> 04:41 < kanzure> so, i think that's actually broken in a number of ways. ordinary folks getting uploaded is valuable because it's cheaper than healthcare and food costs in the long run, so at minimum there's that.
18:29 < kanzure> 04:42 < kanzure> regarding your insistence on slavery, i don't think that's necessary either. it's the same as like "you're free to go out to sea with no infrastructure, no funding, and no taxation, sure, but have fun". additionally, not-quite-human emulations will probably be valuable anyway and maybe even pay for all the human-specific junk.
18:29 -!- Gurkenglas_ [~Gurkengla@dslb-178-000-218-087.178.000.pools.vodafone-ip.de] has quit [Ping timeout: 264 seconds]
19:01 -!- jaboja [~jaboja@jaboja.pl] has quit [Ping timeout: 260 seconds]
19:03 < kanzure> 010b938802952aa888ea340d5bb1690b93bbe2bdc71e6fe9914c7b9f9e1b4f03
19:11 -!- yashgaroth [~yashgarot@2606:6000:cd4d:3300:f5e0:f867:a11d:8d52] has joined ##hplusroadmap
19:14 -!- Darius [~quassel@66-215-89-229.dhcp.psdn.ca.charter.com] has joined ##hplusroadmap
19:24 -!- mrdata- [~mrdata@unaffiliated/mrdata] has quit [Read error: Connection reset by peer]
19:27 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has joined ##hplusroadmap
19:52 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
19:54 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has quit [Client Quit]
19:55 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
19:56 -!- justan0theruser [~justanoth@unaffiliated/justanotheruser] has quit [Ping timeout: 264 seconds]
20:00 < fenn> down and out in the magic kingdom was hardly a positive view of reputation systems
20:03 < kanzure> i don't think reputation works that well.
20:04 < fenn> what's the alternative
20:04 -!- jtimon [~quassel@143.29.134.37.dynamic.jazztel.es] has quit [Ping timeout: 264 seconds]
20:05 -!- UnknownRogue [~UnknownRo@2a04:9dc0:c1:112:301:200:0:1a4d] has joined ##hplusroadmap
20:06 < fenn> what channel are all these crossposts coming from?
20:09 < kanzure> hplusroadmap
20:10 -!- preview [~quassel@2407:7000:842d:4000::3] has quit [Remote host closed the connection]
20:17 -!- Solgriffin [~Sol@c-69-141-24-242.hsd1.nj.comcast.net] has joined ##hplusroadmap
20:24 -!- UnknownRogueX [~UnknownRo@2a04:9dc0:c1:113:404:200:0:18c2] has joined ##hplusroadmap
20:25 -!- UnknownRogue [~UnknownRo@2a04:9dc0:c1:112:301:200:0:1a4d] has quit [Ping timeout: 264 seconds]
20:35 -!- UnknownRogueX [~UnknownRo@2a04:9dc0:c1:113:404:200:0:18c2] has quit [Quit: Leaving]
20:43 < kanzure> hmph
20:45 < kanzure> "Polylox barcoding reveals haematopoietic stem cell fates realized in vivo" http://www.nature.com/nature/journal/vaop/ncurrent/full/nature23653.html https://twitter.com/bloodgenes/status/897882658457481216
20:47 < kanzure> http://www.nature.com/news/china-launches-brain-imaging-factory-1.22456 https://twitter.com/minikerri/status/897452648793686016
20:48 < kanzure> "In-chamber reel-to-reel system for random-access volume electron microscopy" https://www.cambridge.org/core/services/aop-cambridge-core/content/view/14669427EDB473E88C99E19813976AB6/S1431927617001453a.pdf https://twitter.com/brainmaps/status/896809653824364546
20:50 < kanzure> basal forebrain and basal nucleus projection mapping using retrograde (non)cholinergic cell labeling https://academic.oup.com/cercor/article/25/1/118/367269/Neurons-in-the-Basal-Forebrain-Project-to-the#.WYxhw5vC9Os.twitter https://twitter.com/MargolisLab/status/895643643830251520
20:52 < kanzure> sperm/eggs/artificial gametes stuff https://www.technologyreview.com/s/608452/a-new-way-to-reproduce/ https://twitter.com/amy_harmon/status/895261251261145089 https://twitter.com/antonioregalado/status/894573452434300930
21:05 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has joined ##hplusroadmap
21:07 -!- abetusk [~abe@68.175.143.22] has quit [Ping timeout: 240 seconds]
21:08 -!- abetusk [~abe@68.175.143.22] has joined ##hplusroadmap
21:08 < fenn> "Mitinori Saitou and Katsuhiko Hayashi, who last November reported they had turned mouse tail cells into iPS cells and then into eggs."
21:08 -!- abetusk is now known as Guest25737
21:09 < fenn> ... and then they made mice, right? right? because that's what sensible researchers do
21:10 < fenn> yes they did
21:11 < fenn> http://www.sciencemag.org/news/2016/10/mouse-egg-cells-made-entirely-lab-give-rise-healthy-offspring
21:13 < fenn> .title http://dx.doi.org/10.1038/nature20104
21:13 < yoleaux> Reconstitution in vitro of the entire cycle of the mouse female germ line : Nature : Nature Research
21:18 < mrdata> did they use an artificial womb as well?
21:18 < fenn> no
21:18 < fenn> it seems like a logical step in the pipeline
21:19 < mrdata> that would also solve a bunch of problems surrounding bringing extinct species back
21:20 < fenn> we do have existing animals that are close enough
21:20 < fenn> ostriches for dinosaurs, elephants for mammoths, etcf
21:21 < mrdata> sure; that's kind of ish
21:21 < fenn> artificial wombs are even more so
21:21 < mrdata> admittedly
21:21 < fenn> ok so maybe an ostrich is the wrong temperature
21:22 < mrdata> idk
21:22 < mrdata> is it?
21:22 < mrdata> many dinosaurs were also warm blooded
21:22 < fenn> birds are warm blooded, dinosaurs probably not
21:22 < fenn> how do we know that?
21:23 < mrdata> i dont recall well, but i do remember this was confirmed
21:23 < mrdata> checking
21:23 < mrdata> https://www.livescience.com/51162-dinosaurs-warm-blooded-growth-rates.html
21:23 < fenn> there's also the problem of getting dinosaur DNA
21:24 < fenn> .wik mesotherm
21:24 < yoleaux> "A mesotherm (from Greek μέσος mesos "intermediate" and thermē "heat") is a type of animal with a thermoregulatory strategy intermediate to cold-blooded ectotherms and warm-blooded endotherms." — https://en.wikipedia.org/wiki/Mesotherm
21:27 < fenn> sounds like it's not a simple yes/no answer
21:27 < mrdata> mhm
21:28 < fenn> a near term use for the whole iPS -> gamete thing would be for cloned embryo organs
21:28 < fenn> there's significant demand for that right now
21:29 < fenn> ethical wankers aside
21:29 < fenn> ethics is all fine and dandy when it's not you that's dying
21:30 < fenn> i sort of wonder how big the embryo has to get before you can transplant
21:33 < fenn> .wik gastruloid
21:33 < yoleaux> "Gastruloids are three dimensional aggregates of mouse embryonic stem cells (ESCs) that display many features of early development such as symmetry-breaking, axial organisation, germ layer specification, polarised gene expression and axial elongation." — https://en.wikipedia.org/wiki/Gastruloid
21:40 < mrdata> can you coax those to grow you a new liver or kidney?
21:41 < fenn> i just learned of them
21:42 -!- fltrz [d5d38b31@gateway/web/freenode/ip.213.211.139.49] has quit [K-Lined]
21:42 -!- y0ur1 [~y0ur1@gateway/tor-sasl/y0ur1] has quit [K-Lined]
21:42 -!- sandeepkr [~sandeepkr@ec2-52-29-251-54.eu-central-1.compute.amazonaws.com] has quit [K-Lined]
21:42 -!- BobaMa [bobama@kapsi.fi] has quit [K-Lined]
21:42 -!- Iriez [wario@distribution.xbins.org] has quit [K-Lined]
21:42 -!- gnusha [~gnusha@unaffiliated/kanzure/bot/gnusha] has quit [K-Lined]
21:52 -!- gnusha [~gnusha@unaffiliated/kanzure/bot/gnusha] has joined ##hplusroadmap
21:52 -!- Topic for ##hplusroadmap: biohacking, nootropics, transhumanism, open hardware | sponsored by lobsters everywhere, banned by the Federal Death Administration (5 times) | this channel is LOGGED: http://gnusha.org/logs | http://diyhpl.us/wiki | "ray kurzweil is a pessimist" - george church
21:52 -!- Topic set by kanzure [~kanzure@unaffiliated/kanzure] [Wed May 20 12:46:25 2015]
21:52 [Users ##hplusroadmap]
21:52 [ abetusk_      ] [ c0rw1n     ] [ FourFire       ] [ juul       ] [ redlegion] [ wrldpcmbl ] 
21:52 [ adlai         ] [ catern     ] [ gnusha         ] [ kuldeep    ] [ Reventlov] [ y0no      ] 
21:52 [ AgenttiX      ] [ cluckj     ] [ Guest60956     ] [ kwgwk      ] [ sandeepkr] [ y0ur1     ] 
21:52 [ anonnumberanon] [ danfox     ] [ Guest84243     ] [ marciogm   ] [ strages  ] [ yashgaroth] 
21:52 [ axlshear      ] [ Darius_    ] [ helleshin      ] [ mf1008     ] [ streety  ] [ yoleaux   ] 
21:52 [ balrog        ] [ Dumuzi     ] [ hylleddin      ] [ nanotube   ] [ superkuh ] 
21:52 [ berndj        ] [ dustinm    ] [ indiebio       ] [ nickjohnson] [ Taek     ] 
21:52 [ bkero         ] [ EnabrinTain] [ Iriez          ] [ nsh        ] [ thxffo   ] 
21:52 [ BobaMa        ] [ ensign     ] [ Joshchamp      ] [ rancyd     ] [ TMA      ] 
21:52 [ Burninate     ] [ fenn       ] [ justanotheruser] [ rancyd_    ] [ vikraman ] 
21:52 -!- Irssi: ##hplusroadmap: Total of 55 nicks [0 ops, 0 halfops, 0 voices, 55 normal]
21:52 -!- Channel ##hplusroadmap created Thu Feb 25 23:40:30 2010
21:52 -!- Irssi: Join to ##hplusroadmap was synced in 14 secs
21:52 -!- entity8421[m] [entity8421@gateway/shell/matrix.org/x-ciyyxtiszwwnhsal] has joined ##hplusroadmap
21:52 -!- Cory [~Cory@unaffiliated/cory] has joined ##hplusroadmap
21:53 -!- midnightmagic [~midnightm@unaffiliated/midnightmagic] has joined ##hplusroadmap
21:53 -!- Malvolio [~Malvolio@unaffiliated/malvolio] has joined ##hplusroadmap
21:53 -!- augur [~augur@noisebridge130.static.monkeybrains.net] has joined ##hplusroadmap
21:54 -!- ExeciN [ExeciN@bnc.stormbit.net] has joined ##hplusroadmap
21:57 -!- Darius_ [~quassel@66-215-89-229.dhcp.psdn.ca.charter.com] has quit [Remote host closed the connection]
21:59 -!- Rmesil8O4b[m] [rmesil8o4b@gateway/shell/matrix.org/x-oqzgcluhwdilxsgy] has joined ##hplusroadmap
21:59 < fenn> it's such a new process that nobody has figured out how to grow organs from gastruloids yet, although they are hopeful/expectant that it will be possible
22:08 -!- mrdata [~mrdata@unaffiliated/mrdata] has joined ##hplusroadmap
22:08 -!- SolG [~Sol@c-69-141-24-242.hsd1.nj.comcast.net] has joined ##hplusroadmap
22:09 -!- SolG is now known as Solgriffin
22:23 -!- hehelleshin [~talinck@cpe-174-97-113-184.cinci.res.rr.com] has joined ##hplusroadmap
22:25 -!- sivoais [~zaki@unaffiliated/sivoais] has joined ##hplusroadmap
22:26 -!- helleshin [~talinck@cpe-174-97-113-184.cinci.res.rr.com] has quit [Ping timeout: 246 seconds]
22:41 -!- docl [~docl@159.203.115.16] has joined ##hplusroadmap
22:56 -!- andytoshi [~apoelstra@wpsoftware.net] has joined ##hplusroadmap
22:56 -!- andytoshi [~apoelstra@wpsoftware.net] has quit [Changing host]
22:56 -!- andytoshi [~apoelstra@unaffiliated/andytoshi] has joined ##hplusroadmap
23:02 -!- fltrz [d5d38b31@gateway/web/freenode/ip.213.211.139.49] has joined ##hplusroadmap
23:12 -!- yashgaroth [~yashgarot@2606:6000:cd4d:3300:f5e0:f867:a11d:8d52] has quit [Quit: Leaving]
--- Log closed Thu Aug 17 00:00:51 2017