2014-10-22.log

--- Log opened Wed Oct 22 00:00:23 2014
justanotherusernmz787: okay, did you read shamirs paper?00:01
justanotheruserwould you care to tell me whats wrong with it?00:01
nmz787never heard of it00:06
justanotherusernmz787:  "Quantitative Analysis of the Full Bitcoin Transaction Graph"00:07
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-gnsjxpktnhemnbfo] has joined ##hplusroadmap00:13
nmz787sorry, busy looking at 'Use of domestic detergents in the California mastitis test for high somatic cell counts in milk'00:14
nmz787paperbot: http://veterinaryrecord.bmj.com/content/163/19/566.long00:14
paperbothttp://libgen.org/scimag/get.php?doi=10.1136%2Fvr.163.19.56600:14
justanotheruserinteresting...00:15
nmz787paperbot: http://veterinaryrecord.bmj.com/content/163/19/566.full.pdf+html00:17
paperbothttp://libgen.org/scimag/get.php?doi=10.1136%2Fvr.163.19.56600:17
nmz787justanotheruser: sorry i'm only just above a newb when it comes to bitcoin00:22
justanotherusernmz787: but as you said, that shouldn't matter00:24
-!- ebowden [~ebowden@147.69.54.84] has quit [Remote host closed the connection]00:41
-!- ebowden [~ebowden@147.69.54.84] has joined ##hplusroadmap00:41
-!- andytoshi [~andytoshi@unaffiliated/andytoshi] has quit [Ping timeout: 272 seconds]01:12
archelsrelevant to last night's discussion http://cameramaker.se/microlenses.htm01:45
archels.title01:45
yoleauxThe Camera Maker - Making Microlenses01:45
-!- nsh_ [~nsh@wikipedia/nsh] has joined ##hplusroadmap01:45
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap02:14
-!- snuffeluffegus [~snuff@5.150.254.180] has joined ##hplusroadmap02:40
-!- snuffeluffegus [~snuff@5.150.254.180] has quit [Remote host closed the connection]02:40
-!- snuffeluffegus [~snuff@5.150.254.180] has joined ##hplusroadmap02:40
jrayhawk_coooool02:51
-!- Urchin [~urchin@unaffiliated/urchin] has joined ##hplusroadmap02:58
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 255 seconds]03:38
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap03:46
-!- Vutral__ [~ss@176.10.107.235] has joined ##hplusroadmap03:53
-!- Vutral_ [~ss@176.10.107.228] has quit [Ping timeout: 260 seconds]03:53
-!- snuffeluffegus [~snuff@5.150.254.180] has quit [Quit: May the force be with you. Always.]04:06
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 245 seconds]04:23
-!- HEx1 [~HEx@hexwab.plus.com] has joined ##hplusroadmap04:30
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap04:32
-!- Beatzebub [~beatzebub@S0106b81619e8ecee.gv.shawcable.net] has quit [Quit: No calling card for the unsung bard]04:40
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 260 seconds]05:02
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap05:06
-!- ebowden [~ebowden@147.69.54.84] has quit [Remote host closed the connection]06:00
JayDuggerGood morning.06:41
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Ping timeout: 260 seconds]06:51
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-gnsjxpktnhemnbfo] has quit [Quit: Connection closed for inactivity]07:07
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-mizesietfccgspgb] has joined ##hplusroadmap07:31
-!- d4de^^ [~d4de@197.160.156.98] has quit [Ping timeout: 260 seconds]07:34
-!- d4de^^ [~d4de@197.160.62.123] has joined ##hplusroadmap07:36
-!- pete4242 [~smuxi@boole.london.hackspace.org.uk] has quit [Remote host closed the connection]07:36
-!- d3vz3r0 [~d3vz3r0@jsr.6502.ws] has quit [Quit: Bye 1.0.1]07:44
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Ex-Chat]07:59
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap08:03
-!- lichen [~lichen@c-50-139-11-6.hsd1.or.comcast.net] has joined ##hplusroadmap08:10
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap08:45
-!- h0rsep0wer [~h0rsep0we@unaffiliated/h0rsep0wer] has joined ##hplusroadmap08:48
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 260 seconds]08:56
kanzureoof09:00
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 250 seconds]09:02
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap09:03
kanzurestalk: Rusnak <stick@gk2.sk>09:04
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has joined ##hplusroadmap09:08
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 244 seconds]09:14
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap09:24
fennthis was cute, literal AR sandbox http://www.youtube.com/watch?v=9IRm-IamCqo09:27
kanzurehttp://www.blockstream.com/wp-content/uploads/2014/10/sidechains.pdf09:28
fenn502 bad gateway09:29
kanzurehttp://diyhpl.us/~bryan/papers2/bitcoin/sidechains.pdf09:30
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap09:33
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Quit: Reconnecting]09:36
-!- justanot1eruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap09:37
kanzurehttp://0bin.net/paste/PpyKcrm5Y1wb5AUJ#kwDsCN6TwWDn61j8QKCn7ifRzSlqwmBJxHUyvsUiHRJ09:40
-!- justanot1eruser is now known as justanotheruser09:44
-!- nmz787_i [nmccorkx@nat/intel/x-qlkiiquehdksupsr] has joined ##hplusroadmap10:00
fenna client side encrypted pastebin?10:00
kanzurewhere have you been10:01
kanzurewhat was megaupload?10:01
justanotherusermega you mean?10:01
kanzurewhatever? heh10:01
* justanotheruser pedants10:01
fenni mean i get the concept, but it just seems totally lame to fail completely if there isn't actually any encryption. if there is encryption then where is the decryption key?10:02
kanzureurl10:02
fennthat's fucking retarded10:03
kanzurehahah10:03
justanotheruseris it?10:03
fennyes10:03
justanotheruserwhy?10:03
kanzurenot even https too10:03
justanotheruseryeah, lack of https is stupid10:03
justanotheruserbut I don't see the problem with the url having the key10:03
fennbecause a) the host gets the key, thus negating the point of client-side encryption, b) random packet sniffers get the key c) its in your browser history, d) and on and on10:04
justanotheruserfenn: well most of those are http problems afaix10:05
justanotheruser*afaix10:05
fennand i cant read the faq because i dont have javascript? srsly10:05
justanotheruser*afaik10:05
kanzurewell you should double check what 0bin is doing, i haven't really checked10:05
kanzurei just always assumed the thing in the url was the key10:05
kanzurebecause i can't figure out why else you would do it like this10:05
justanotheruserkanzure: either its in the URL or it isn't encrypted10:05
kanzureright, except i haven't read 0bin's site ever10:05
justanotheruserkanzure: and it gives you bad key if you change the url :p10:05
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 260 seconds]10:05
kanzurejust viewed a few pastes once in a while10:05
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap10:07
fenn"The browser never sends the hash to the server"10:07
fennthey are talking about the # anchor hash10:08
fennThe goal of 0bin is <strong>not</strong> to protect the user and their data10:08
fennInstead, it aims to protect the host from being sued for the10:08
fenncontent users pasted on the pastebin.10:09
fennmeh10:09
kanzureheh10:12
justanotherusersmart10:13
justanotheruserI wonder if them saying the encryption prevents them from being sued would hurt them in court10:14
chris_99oh i wrote something to scrape pastebin for CCs just to see if there were any10:14
kanzurenothing can prevent you from being sued10:15
justanotheruserlol10:18
fennpresumably they could just implement "client side rot13" with a button you click to get the cleartext10:18
justanotheruserliterally you can sue anyone for anything. A lot of the time it will be thrown out though.10:18
fennthere's certainly enough (stupid) legal precedent to claim that rot13 is a form of encryption10:19
justanotheruserkanzure: could you help me with a datastructures problem? I want to make a search engine that finds bitcoin scripts. If someone searches "OP_ADD, OP_CHECKSIG", I want to return results that have both of those opcodes. The scripts are just byte arrays, so I'm wondering what datastructure or database setup is good for searching 10s of millions of these byte arrays. I have a hacky solution right now I'm thinking of involving ...10:19
justanotheruser... a single table where each column is a boolean representing whether that opcode is present.10:19
justanotherusertl;dr: how to preprocess list of byte arrays so I can search and find byte arrays containing a list of bytes?10:20
fennsort them as if they were integers?10:21
kanzurethis sounds like some sort of set search thingy. xapian?10:21
chris_99so you wan't to search for multiple strings, within a string?10:21
chris_99aho-corasick if so10:21
kanzurethey shouldn't really be considered strins10:21
kanzure*strings10:21
kanzurethis dataset is much more restricted than normal stringy datasets10:22
kanzureafk10:22
chris_99aho-corasick works on binary10:22
chris_99too10:22
chris_99AV guys use it10:22
fennis OP_ADD a bit vector?10:24
fennor is it an integer10:24
justanotheruserkanzure: restricted only by validity10:24
justanotheruserbut I'll check out xapian10:25
kanzurefenn: integer10:25
justanotheruserfenn: 0x9310:25
kanzurefenn: https://en.bitcoin.it/wiki/Script10:25
justanotheruserkanzure: in what sense is it an integer?10:25
fennok nevermind then10:26
justanotheruserwhen the byte has math done on it it is an integer "When used as numbers, byte vectors are interpreted as little-endian variable-length integers with the most significant bit determining the sign of the integer. Thus 0x81 represents -1. 0x80 is another representation of zero (so called negative 0). Positive 0 is represented by a null-length vector. Byte vectors are interpreted as Booleans where False is represented by any ...10:26
justanotheruser... representation of zero, and True is represented by any representation of non-zero."10:26
kanzure0x93 is an integer10:27
fenni was thinking you had a bit array for each script, where each opcode is represented by one bit. then you bitwise AND those fields with your search query (also a bit array)10:27
kanzure.py 0x9310:28
yoleaux14710:28
justanotheruserkanzure: it is all about interpretation10:28
justanotheruserthey're signed so its actually -1310:28
justanotheruserbut that is only when a byte is on the stack and is being interpreted by something that cares about booleans and ints iirc10:29
fenn"a single table where each column is a boolean" is basically a bit field (array of bit vectors)10:31
justanotheruserfenn: its indexed too10:31
justanotheruserwell...10:32
justanotheruserthere are 10 million entries10:32
justanotheruserI wonder how bad it would be to read 2 mb10:32
justanotheruserprobably worse than the space that it would take up on disk10:32
fenni dont know enough about how databases search things like that to say whether it's efficient or not (my gues is they aren't designed for that type of search)10:32
justanotheruserprobably not... but it may still be faster since its O(log n)10:33
fennthere are on the order of 255 opcodes, so each entry needs 255 bits or 319MB for 10 million entries10:34
fennwhy do you want to search for the presence of any particular combination of opcodes?10:35
justanotheruserfenn: because it could yeild interesting scripts10:36
-!- _0bitcount [~big-byte@81.61.34.185.dyn.user.ono.com] has joined ##hplusroadmap10:36
kanzureyeah, there's a bunch of interesting stuff floating around in there10:36
kanzureit's sort of like searching for malware too10:36
kanzureand then various automatic code analysis benefits10:36
fennit may be worth your time to do cluster analysis and various types of dimensionality reduction10:37
fennare you familiar with basic machine learning algorithms?10:38
justanotheruserya10:38
justanotheruserjust took a class on it10:38
kjskjskjsjustanotheruser: the right data structure for this problem depends a lot on the distribution of the data10:38
-!- ElixirVitae [~Shehrazad@unaffiliated/shehrazad] has quit [Ping timeout: 245 seconds]10:38
justanotheruserI wonder if it would be faster to do it with numpy or find a C++ class10:38
kjskjskjsif almost all scripts contain OP_ADD and almost all scripts contain OP_CHECKSIG then you probably need to precompute the OP_ADD ∧ OP_CHECKSIG result10:39
kjskjskjsfenn's suggested approach involves iterating over all the scripts.  ideally taht's what you'd like to avoid10:40
justanotheruserkanzure: you think xapian would be a good fit for this?10:40
fennif you're interested in outliers it would seem a statistical analysis approach would be better suited than searching one query at a time10:40
justanotheruserI want exact results btw, not fuzzy results ofcourse10:40
kanzurejustanotheruser: not particularly, it's just something on the top of my mind10:40
kjskjskjsthings like xapian and lucene can solve this problem, but maybe not efficiently10:41
kjskjskjs10 million is not very big though10:41
kjskjskjsif you only have 10 million scripts to search then just iterate over them10:41
justanotheruserkjskjskjs: yeah, probably 99% of scripts are of the form <somebytes> <somebytes> OP_DUP OP_HASH160 <somebytes> OP_EQUALVERIFY OP_CHECKSIG10:41
kjskjskjsdon't even bother with fancy bit-field data structures10:41
justanotheruserand another .9% is probably p2sh10:42
kanzurei think these scripts can even fit in memory10:42
kjskjskjshmm, in that case if you had a list of scripts including OP_ADD then you could reduce the number of things to iterate over by 100×10:42
kjskjskjsfor that query10:42
kanzureyou can also do a histogram-style, where you don't record duplicate scripts10:42
kjskjskjsyeah10:42
kanzureyou know, the simplest implementation that i can think of would be sqlalchemy + sqlite or something, and then just iterate from there10:42
justanotheruserkjskjskjs: yeah, I will probably just end up doing that hack for now10:42
kanzureit wont have fancy search features beyond sql querying, but that doesn't sound problematic10:42
justanotheruserIt likely wont always be true that 99% of scripts are of that form10:43
kanzurethere's even a bitcoin sqlalchemy module somewhere that has common bitcoin structures  that you can readily use with sqlalchemy10:43
kanzure99% of all scripts are just pay to sighash or whatever10:43
kanzureand storing a billion of those is stupid10:43
nmz787_iwon't any-sql add overhead, for something like this you seem to just want a sliding search window over all the data10:43
justanotheruserkanzure: yeah, I probably should just store their pubkey and pubkey hash10:43
kanzurewell he wants querying or something10:43
kanzureand implementing your own query engine is dumb10:44
kjskjskjsnmz787_i: it will add overhead but probably not enough to make a difference for only 10 million10:44
kjskjskjsno, implementing your own query engine is not a big deal10:44
nmz787_iso store the results in the db, not the raw data10:44
kanzuresqlite can easily fit in memory with sqlite://:memory: in sqlalchemy10:44
nmz787_isince the raw data has no structure other than as a vector10:44
kanzureraw data does have a structure10:44
kanzure(in this case)10:44
fennthere is a sequence of opcodes10:45
kanzureand params10:45
nmz787_ii heard 'script' which seems like a sequence of 0s and 1s10:45
fennyeah well this irc channel is a sequence of 0s and 1s10:45
nmz787_ithe channel itself, yes, but the viewer programs are separate streams10:46
justanotheruserkanzure: you sure it's that dumb to make my own query engine?10:47
kanzureyou should make up some example queries that you want to ru nfirst10:47
kanzurethen evaluate that against what each implementation gives you10:47
kanzurethen you can decide to make your own or not10:47
kanzurebut defaulting to making your own... pfft.10:47
fennthis whole conversation seems like premature optimization anyway10:48
justanotheruserkanzure: good call.10:48
justanotheruserit will be a pain to test all of these though since I need to copy all this blockchain data10:48
kanzurehere are some things that might be useful https://github.com/monetizeio/sqlalchemy-bitcoin10:48
justanotheruserI think I might just mentally test it10:48
justanotherusersee if my algorithm would be efficient10:49
kanzurehttps://github.com/monetizeio/sqlalchemy-bitcoin/blob/master/sa_bitcoin/core.py10:49
kanzurehttps://github.com/monetizeio/sqlalchemy-bitcoin/blob/master/sa_bitcoin/ledger.py10:49
kanzuremaaku: i see your name here10:49
justanotheruserneat, I can put scripts in the DB automatically10:50
justanotheruserstill need to handle them specially though, which is why it seems doing it myself would be easier10:50
kanzurepetertodd/python-bitcoinlib already parses script pretty well, i highly recommend plugging his CScript implementation in with sqlalchemy-bitcoin stuff10:51
justanotheruserif I'm going to query the for p2pk, then I need to index pubkeys and pubkey hashes10:51
justanotheruserkanzure: yeah, already was planning on using that10:51
justanotheruserwould be a pain reimplementing all that for each opcode10:51
kanzure10:47 <+deltab> "Experience historically accurate terrain collision rendering." — https://www.youtube.com/watch?v=5pfgMS3nDn410:52
justanotheruserok, thx for the help10:52
* justanotheruser vanishes10:52
kanzurewizard evidence10:53
fennpoof10:55
fenni still dont get the focus on querying and databases if the whole point of the exercise is data science/exploration10:56
kanzurehe wants to be able to do things like "find me every script that has this op code followed by this other one"10:56
kanzureor "find me all scripts that are invalid"10:56
fennuh..10:56
kanzure... i assume.10:57
fennisnt that basically the halting problem10:57
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 245 seconds]10:57
fennhm no loops10:58
kanzurehehe10:59
fennlisp has no loops but it's trivial to construct one10:59
kanzurethere are also limits on how long a script can execute or how many ops10:59
kanzureask in #bitcoin-wizards or #bitcoin-dev11:00
maakukanzure: I wrote it, although right now it is pre-blockstream abandonware11:00
fennthere's too much to learn bitcoin before i can ask a decent question11:00
maakuhappy to entertain questions about it or pull request11:00
fennabout bitcoin*11:00
kanzuremaaku: abandonware because it is borked or abandonware because limited time11:00
archelspaperbot: http://link.springer.com/chapter/10.1007/978-1-4615-0039-1_15?no-access=true11:01
maakulimited time + no longer doing the project which necessitated its existence in the first place11:01
archels?no-access=true lol11:01
paperbothttp://diyhpl.us/~bryan/papers2/paperbot/430b0ce28ae82bd792044c3bfb8d300c.txt11:01
kanzuremaaku: cool. i'll probably use it eventually, nice to have sqlalchemy stuff available.11:01
kanzurefenn: yes, that's a big problem, but i dunno how to solve it... "sorry, but yes there are multiple components."11:02
maakunothing wrong with it although there may be a few undocumented warts. it certainly wasn't up to my "release-ready" standards when I stopped working on it11:02
archelspaperbot: http://link.springer.com/content/pdf/10.1007%2F978-1-4615-0039-1_15.pdf11:02
paperbotXMLSyntaxError: None (file "/srv/ikiwiki/paperbot/modules/papers.py", line 524, in parse_html)11:02
-!- snuffeluffegus [~snuff@5.150.254.180] has joined ##hplusroadmap11:02
archelswell, frack11:02
fennhttp://www.frackingfacts.com.au/blog/wp-content/uploads/2011/06/noFrack.jpg11:04
-!- drewbot [~cinch@ec2-54-80-140-243.compute-1.amazonaws.com] has quit [Remote host closed the connection]11:06
-!- drewbot [~cinch@ec2-54-166-61-180.compute-1.amazonaws.com] has joined ##hplusroadmap11:07
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Ping timeout: 260 seconds]11:11
kanzurefenn: if it makes you feel better there's basically only ten people that understand bitcoin11:13
fennand half of them are fictional?11:17
kanzurei mean, if nobody reads papers anymore, how many people do you think also read code and papers11:18
kanzurethis isn't like linux where you have 10,000 people that already understand kernels floating around your society11:18
fenni still dont see anything wrong with my suggestion above about how to do queries... it seems like there's a fair amount of research on indexing bitmaps https://en.wikipedia.org/wiki/Bitmap_index11:19
kanzureliterally everyone had to learn by reading the source code, because that was all that was produced11:19
kanzureoh, i don't think i said there was anything wrong with that idea?11:19
fennkjskjskjs didnt like it because "you'd have to iterate over every entry" but you have to do that anyway in order to parse the data in the first place11:19
kanzurei just can't imagine the exact implementation11:20
kanzurei would have to do work or something11:20
fenni probably misunderstood the problem he was trying to solve in the first place11:20
kanzurewell, i assume the problem is something like:11:22
kanzure"here is a bunch of possibly cool data, what are typical analysis things i should be doing with it and how"11:23
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap11:23
fennok so where does database queries enter into that11:23
fennk-means would be a good place to start11:24
fennmarkov modeling maybe11:24
fennthere are a zillion ways to slice large data sets11:24
kanzurealso there's some static analysis (erm, w.r.t code/software) aspects to this11:25
kanzurequeries over code, was what i was thinking..11:25
fennbayesian filters11:26
kanzurehm?11:26
fennas a way of sorting through large quantities of irrelevant data11:26
kanzure(brb)11:26
fennhttps://en.wikipedia.org/wiki/Bayesian_spam_filtering11:27
kanzureapplied to source code?11:27
fennsure why not11:27
kanzurewhat would that look like?11:27
kanzure(brb4realz)11:27
fennin this case you'd be prejudiced against things that look like "<somebytes> <somebytes> OP_DUP OP_HASH160 <somebytes> OP_EQUALVERIFY OP_CHECKSIG"11:30
fenn(not that i have any idea what that means)11:30
fennyou can use bayes filtering to sort things into multiple bins too, not just spam/nospam11:32
chris_99dumb question, why do different transactions need to do different things, via scripts11:32
fennbecause they're the basis for contracts11:33
chris_99oh i didn't know it could do that kind of stuff just looking at the wiki on that thanks11:34
fennsay you want to sell a car on a different continent, how do both parties ensure that they dont get screwed? the traditional solution is to send both money and keys to an escrow service11:34
chris_99i never noticed anything to do this kind of stuff in the btc client i used, but it's been a long time11:35
fennwith smart contracts, the keys don't work unless the money has been exchanged, and the money doesnt work unless the keys have been received11:35
chris_99nice11:35
fenni havent been following bitcoin much, but it seems like a set of possible enhancements rather than a 1.0 feature11:36
fennthere is also stuff like multiple signatures required to transfer money, so if you get hacked your bank wont send money, or if your bank gets hacked they wont send money (word "bank" being used loosely)11:37
chris_99clever11:39
kanzuresome of those things are in the bitcoin client and others are not11:41
kanzurepay-to-script-hash for example works (the addresses start with '3' or 'm' on testnet or something)11:41
kanzureand it just looks like you're paying a funny looking address11:41
-!- eridu [~eridu@gateway/tor-sasl/eridu] has joined ##hplusroadmap11:44
chris_99ah11:46
-!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##hplusroadmap11:47
fennso do miners actually run the scripts to see if they're valid?11:48
kanzureyes11:48
kanzureevery bitcoin node runs the scripts11:48
kanzurepetertodd even implemented a python version in python-bitcoinlib called VerifyScript11:48
fennnot every node? just the ones that are verifying that block, right?11:49
kanzureeven the blocks are verified by nodes11:49
kanzurenow, there are some bitcoin software implementation things that do not do any verification- these are usually called "spv" for some reason11:49
-!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 265 seconds]11:49
kanzure(however, bitcoind.git does not implement that)11:49
kanzure(actually it might; that may have been in a recent pull request that i saw? it's a config option maybe. and it's definitely not enabled by default.)11:49
fennguh. true/false: every node verifies every block11:50
kanzurewell what the fuck is a node11:50
kanzureit's sort of like asking "everything that uses the bittorrent protocol does everything the official implementation does?"11:51
fennthis is only part of why i'm confused11:51
kanzureanyone can write some software that does not verify every block in the blockchain11:52
kanzurehowever, this is dangerous because you can be mitm'ed much more easily11:52
-!- h0rsep0wer [~h0rsep0we@unaffiliated/h0rsep0wer] has quit [Quit: Leaving]11:52
kanzureeveryone that deploys bitcoind by default with no config changes will be veryifying every transaction that they see from the p2p network, including every transaction in every block11:53
-!- _0bitcount [~big-byte@81.61.34.185.dyn.user.ono.com] has quit [Quit: Leaving]11:53
fennand that actually scales?11:54
kanzurehttps://en.bitcoin.it/wiki/Scalability11:54
fennwhy cant they just use multiple blockchains11:55
kanzureif your balance is on one blockchain (like the dogecoin blockchain) and i want to receive on the other, how do you communicate that to your users11:55
kanzureor to your payees/payers or w/e11:56
kanzurethere are other reasons why multiple blockchains are a bad idea, related to security and mining hashrate11:58
kanzure(and then less strong reasons like "well nobody understands the first one, and you want to start a second secured only by your own lack of understanding?")11:58
fenngreat11:59
fenni guess i had this fantasy that there were all these smart crypto people out there who actually understood all this11:59
kanzurethere's a small group that is growing at a very slow rate11:59
kanzureand all of their knowledge comes from "reading the bitcoin source code" basically12:00
fennso all this microtransaction $0.0001 per gif viewed crap is basically impossible with current bitcoin protocol12:01
kanzuretx fee might have to be pretty high to get the transaction picked up12:01
nmz787_iback to real news: http://www.independent.co.uk/life-style/health-and-families/health-news/british-doctors-on-brink-of-cure-for-paralysis-9807010.html12:02
kanzureunfortunately it is hard to determine what is real or not in the news12:02
fennspeaking of news, apparently someone shot up the canadian parliament12:02
kanzure12:03 < nsh> "There are trade-offs between scalability and decentralisation. For example, a larger block size would allow the network to support a higher transaction rate, at the cost of placing more work on validators — a centralisation risk."  -- is it worth elaborating on how this leads to centralization?12:04
nmz787_ii saw that too12:04
kanzure12:03 < sipa> if fewer people can validate, validation becomes more centralized.12:04
kanzure12:03 < sipa> any question? :012:04
fenni just think its crazy to have everyone validating every transaction12:06
kanzurego tell #bitcoin-wizards12:07
kanzurethey are fun to talk with too12:07
kanzureand nsh is there so he has your back i guess12:07
kanzure12:07 < nsh> so the degree of centralization depends on the tail of resource-availability or usability of the cadre of potential validating participants12:07
fennwait, what time is it now?12:08
kanzurei don't know man12:08
kanzurefuck time?12:08
fenndid nsh just say that?12:08
kanzureyes12:08
nshhad you asked already?12:09
nshdidn't see12:09
nmz787_iwhy is it crazy to have everyone validate everything?12:09
kanzureand approximately how crazy on a log scale12:11
fennbecause it's an O(N^2) algorithm12:18
kanzurealso, log context http://gnusha.org/bitcoin-wizards/2014-10-22.log12:19
nmz787_ifenn: but like, so what, you either fear getting ripped off, or spend some watts making sure that doesn't happen. this is why we invented computers.12:24
maakufenn: ideally full nodes should not be running scripts. that is crazy. they should validating execution traces of scripts (which is cheaper!)12:27
fennnot if you have to download a gigabyte of data and you're in subsaharan africa with a cellphone12:27
maakubut that' snot how bitcoin works12:27
maakufenn: i don' tthink your scaling numbers are right though12:28
fennmaaku: it seems like the sheer number of transactions would still not scale even with no scripts12:28
maakuscript validation is O(n) with script size12:28
fenntransaction validation is O((n/2)^2) with n transactions12:29
kanzureyou're both in wizards this is silly12:29
maakuoh well that's a complaint against bitcoin. having a distributed consensus where N people validate M transactions scales as O(N*M)12:29
maakuso don't use bitcoin. that's the cost12:29
kanzurethere are many bitcoin software thingies that do *not* validate all transactions.12:29
nmz787_iwhat maaku said12:29
fenner, yes you were right O(N*M)12:29
maakukanzure: only becaues miner *do* validate12:30
nmz787_iif bitcoin sucks, find a patch12:30
fennguh i dont even know what bitcoin is12:30
maakunmz787_i: not quite, this isn't something you can fix12:30
maakuwell, not without SNARK magic that doesn't exist yet12:30
nmz787_iwell then in that case, deal with the overhead of not getting ripped off in transactions12:30
fennnmz787_i: currently, as i understand it, "validate everything" is equivalent to "only validate my transaction", and you should be able to only validate a subset of transactions12:33
maakubut yeah, the true cost of a bitcoin transaction is something like $20-$4012:33
fennwhat!12:33
maakubut because of subsidies most people don't know that12:33
kanzurethe "true" cost also has to include things like forever storage costs i think12:33
fennthe reptilians from sirius B must be pumping some serious intergalactic cash into this12:33
kanzuresirius-m hasn't been committing in a while i don't know what you're talking about12:34
maaku25btc every 10 minutes, and it's hard to fit more than a thousand or so transactions in a block. do the math on that12:34
maakuwell ok, $10. price has declined12:34
maakukanzure: true12:34
kanzureit's also hard to calculate what the real value of a non-centralized ledger is supposed to be12:35
fennit's hard to calculate any value12:35
kanzureso nobody knows if $10-$40 is too much or too low or what12:35
fenna glass of water is priceless if you're dying of thirst12:35
maakueh it's not the centralization that's important, it is the trustlessness. the promise of smart contracts that can't be revoked by fiat12:36
fennwhat exactly is the promise of smart contracts?12:36
maakufenn: making contracts enforced by the universe instead of courts and guns12:37
kanzurejust to be snarky.... but courts and guns are part of the universe.12:38
fennso the promise is we can get rid of governments and still have law?12:38
fennor some kinds of law at least12:38
maakuwhich i'm sorry is so abstract, but it promises huge efficiencies (eliminating double-digit percentages of global GDP that is wasted), plus anti-corruption by being able to restrict what is done with money, etc.12:39
fennso you can buy porn on ebay?12:39
maakuand democratizing contracts to the poorest people12:39
fenn(what restrictions on what is done with money are you upset with currently?)12:40
maakuright now if you are poor and put in the position of needing to exchange with someone rich, the power dynamic ensures that many times you're screwed with no recourse12:40
maakua smart contract, on the other hand, doesn't care who you are. it executes according to its script irregardless12:40
fennthe being screwed is mostly due to lack of choice and market monopolies12:41
fennany corruption involved is usually supported by law in the first place12:41
maakufenn: I'm getting at why there are no choices, and why being a monopolist is a position of power12:41
kanzuremaaku: i really appreciate your thermodyamics-based explanation of proof-of-work and consensus sets, and i am going to be stealing it for the rest of forever.12:44
maaku:)12:47
fennif verizon locks me into a "smart contract" with lousy terms, how is that any better than the current situation?12:49
maakufenn: the important point is that with a smart contract they can't change the terms on you later, or pay off someone to interpret the terms in their favor, etc.12:51
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap12:51
maakuthere is still a lot of work to be done to make sure that users have tools to inform them of the implications of a smart contract before they sign it12:51
fennso the benefits are just that it's specified in a formal language with unabiguous interpretation?12:51
maakuright, that's it really12:52
maaku-but- i will claim that the implications of that simple difference are huge12:52
maakubut of course if I didn't feel that way I wouldn't be a full time bitcoin developer :P12:53
maakukanzure: so on that theme and in line with the topic of this channel, one way of interpreting bitcoin is similar to a nanny AGI12:53
maakuthe nanny AGI is omnipotent, but only uses that power to enforce various constraints12:54
maakubitcoin effectively does the same: it is a construction that makes the universe enforce particular constraints about transaction validity and revokation12:55
fenni am trying to find something jmc (of lisp fame) wrote about formalized legal systems (basically a robot judge)12:55
maakudue to the connection between proof-of-work and the entropy law12:56
fennit was somewhere around here http://www-formal.stanford.edu/jmc/elephant/elephant.html12:59
* justanotheruser appears out of nowhere13:00
fennanyway this sounded sort of relevant "Programs that engage in commercial transactions assume obligations on behalf of their owners in exchange for obligations assumed by other entities. It may be part of the specifications of an Elephant 2000 programs that these obligations are exchanged as intended, and this too can be expressed by a logical sentence."13:00
maakunick szabo has done a lot of work in this area. you can start on his stuff here : http://szabo.best.vwh.net/smart_contracts_idea.html13:01
maakufurther links at the bottom13:01
maakufenn: right so there are engineering reasons why a higher level language like elephant is probably not a good choice for the protocol level, but the idea is similar13:02
fennsure, i was thinking of it like C and assembly13:03
maakuright13:03
fennelephant gets compiled to bitcoin_script or whatever its called13:03
maakuright, so another language in that area is E by Mark Miller : http://www.erights.org/13:04
-!- HEx1 [~HEx@hexwab.plus.com] has quit [Ping timeout: 255 seconds]13:05
kanzuremaaku: instead of nanny agi i have been saying anti-replay oracle13:06
fennit wouldnt (and shouldnt) be a general intelligence anyway13:08
fennor you're back where we started with judges and corruptability13:08
fennfind me one person in the world who knows what's legal or not legal13:09
maakufenn: side issue. a fully general AI can have a strictly unchanging goal set.13:13
maakukanzure: eh, nanny AGI is a bit broader than that. e.g. the source of magic in HPMoR is (I speculate) something akin to a nanny AGI enforcing programmable conditions13:14
-!- paperlooker [~tiktaalik@50-79-188-182-static.hfc.comcastbusiness.net] has joined ##hplusroadmap13:15
maaku(with Merlin having basically programmed in the operating system that contains charms, etc. as programs)13:15
paperlookerpaperbot: http://onlinelibrary.wiley.com/doi/10.1111/pbr.12190/pdf13:15
paperlookerpaperboooot do your thing13:15
-!- HEx1 [~HEx@hexwab.plus.com] has joined ##hplusroadmap13:15
maakumaybe nanny AGI isn't the right word, i'd be curious if you can suggest something better13:15
paperlookerpaperbot: http://onlinelibrary.wiley.com/doi/10.1111/pbr.12190/full13:15
paperlookerT_T13:16
paperlookerkanzure: ?13:16
paperbothttp://diyhpl.us/~bryan/papers2/paperbot/64a723ca996d95a9fef285cf027d695.txt13:16
kanzuremaaku: i saw a lamport paper recently that suggested that every node hsould simulate a "virtual leader" that does the correct thing, instead of an actual leader.13:18
paperbothttp://diyhpl.us/~bryan/papers2/paperbot/8e2dc82d65c69cd716d80b3198a9a340.txt13:19
paperlookerpaperbot: http://onlinelibrary.wiley.com/enhanced/doi/10.1111/pbr.12190/13:19
kanzureunfortunately paperbot is not omnipotent13:19
* paperlooker is sad about that13:19
paperlookerI worship paperbot13:20
nmz787_ijust watched this https://www.youtube.com/watch?v=sKOlfR5OcB4#t=220213:20
nmz787_i.title13:20
yoleauxAlfred Leitner - Liquid Helium II the Superfluid - YouTube13:20
paperbothttp://diyhpl.us/~bryan/papers2/paperbot/cc2832407347eda545b7b9c1e9cc598f.txt13:21
nmz787_ipaperbot: http://onlinelibrary.wiley.com/doi/10.1111/pbr.12190/pdf13:21
kanzureyou should check the .txt file to see if it is an access issue or a parser issue13:22
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has quit [Ping timeout: 265 seconds]13:23
paperbothttp://diyhpl.us/~bryan/papers2/paperbot/79ab9636eb86d0418073ba5810cee394.txt13:23
nmz787_ipaperbot: http://onlinelibrary.wiley.com/store/10.1111/pbr.12190/asset/pbr12190.pdf13:25
paperbothttp://diyhpl.us/~bryan/papers2/paperbot/36b05a58c8a64e2ade7f33aff3c882b6.txt13:25
-!- justanotheruser [~Justan@unaffiliated/justanotheruser] has joined ##hplusroadmap13:31
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Ping timeout: 260 seconds]13:32
paperlookerkanzure: FYI, it did work. the page loads dynamically with javascript so you have to find a .pdf within the text/html spew13:34
* paperlooker rolls eyes13:34
kanzurehm?13:40
kanzurepaperbot spews a .txt file when it can't access a .pdf13:40
kanzure.title http://dangerousprototypes.com/?p=8395013:42
yoleauxUpdate on CERN’s investment in KiCAD | Dangerous Prototypes13:42
paperlookeryea, but the link to the pdf I wanted was within13:43
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap13:43
kanzurepaperbot doesn't get *links* to pdfs, it gets the actual pdfs13:44
kanzureor at least attempts to13:44
nmz787_iif that pdf link worked for you, you didn't need paperbot13:44
kanzureyeah13:44
justanotheruserlol13:45
fennshould i read "true names"?13:45
kanzurelink13:46
fennhttps://en.wikipedia.org/wiki/True_Names http://fennetic.net/irc/true_names.pdf13:47
superkuhIf you're looking for cyberpunk fiction I just finished Raphael Carter's "The Fortunate Fall" and it was fairly good. It played up the "if you die in the matrix you die in real life" bit but overall the setting (censorship dystopia) and prose were worth it. Rare in a cyberpunk story.13:47
paperlookernah, I did. the "*.pdf" one it returned was a html spew but there was a link to the appropriate pdf contained within13:48
kanzurefenn: yes i think we have been lacking on vernor vinge in general for a while now13:49
kanzureso someone should pick up this slack13:49
drethelinthe last thing he wrote was a lame sequel to a fire upon the deep right13:49
fennsequel to a deepness in the sky i think13:49
kanzurewas the rainbow thing him or was that stross?13:50
fennrainbows end was him (AR meets evil AI meets biotech)13:50
paperlooker-_-13:50
fennor at least "AI with questionable motives"13:50
paperlookerrainbows end was a poor book13:50
superkuhUnrelated: I feel like half the existing Chinese optical crystal websites are just fronts for a single company. The html tables and english wording are exactly the same. They just slap a different header and company name above it.13:50
fennit wasnt so much a book as a technology proposal13:51
kanzurei was not impressed with a deepness in the sky when i was 12, but i think the interweb is telling me a fire upon the deep is the proper one to read13:51
kanzuresuperkuh: i have been collecting some china spam for a while now. eventually i will package this up into a .mbox or .zip but it confirms your observations.13:51
fenni was not impressed with a fire upon the deep when i was .. well, not much younger than i am now13:51
kanzurethe emails are all from companies like "Quanghing Xian Technology Corporation Company Ltd. LLC, Shanghai Development Zone, Dalian, Liaoning, China"13:52
fenni liked the tines world13:53
kanzure"Shanghai Yancui Import And Export Co., Ltd"13:54
kanzure"SuZhou Transtone Auto Parts Co.,Ltd""13:54
superkuh"Shanghai Daheng Optics and Fine Mechanics Co.,LTD."13:55
fennoh no wonder i'm getting them mixed up. "Vinge's novel, A Deepness in the Sky (1999), is a prequel to A Fire Upon the Deep set 20,000 years earlier and featuring Pham Nuwen. Vinge's The Children of the Sky, "a near-term sequel to A Fire Upon the Deep", set ten years later"13:55
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Ping timeout: 260 seconds]13:56
kanzurea fire upon the deep is the one that wei dai really likes13:56
kanzureand happens to be the one that wasn't on my dad's bookshelf so that's how i ended up with the crap ones13:57
fenni guess i just wasnt that impressed with his presentation of the blight13:58
fenncompared to, say, 3340 in ventus13:58
fennthe internet: a vasty deep of broken links and spontaneous time warps14:03
* fenn summons the demon spirit of ted nelson14:04
kanzureit's too bad that science fiction hasn't really caught up to stuff.14:04
kanzurei don't have time to write their scifi, that's their job14:04
superkuharxiv has the best scifi.14:05
kanzure.g site:arxiv.org "Greg Egan"14:06
drethelinI liked rainbows end14:06
yoleauxhttp://arxiv.org/abs/gr-qc/020801014:06
kanzure.title14:06
yoleaux[gr-qc/0208010] Asymptotics of 10j symbols14:06
drethelinlike it had no good cfharacters14:06
drethelinbut I still enjoyed it14:06
fennit was intended to be played as an alternate reality game on the san diego campus14:07
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has joined ##hplusroadmap14:08
superkuhThe Benford brothers "Gregory (scifi author) and James" are have a bit on there.14:08
superkuhBut the funniest I've run across is, http://arxiv.org/abs/1204.052614:09
kanzurei ran into gregory benford once over at some conference when he was pimping for genescient i think14:09
fennyeah herbal tinctures to recompile your DNA, uh huh14:10
fennthey seem to have buried that product/research14:11
nmz787_ichemo recompiles yer dna, sort of14:12
-!- nmz787_i [nmccorkx@nat/intel/x-qlkiiquehdksupsr] has quit [Quit: Leaving.]14:14
fennsuperkuh: is that an april fools joke? it's not funny and it sounds perfectly reasonable except for the "telescopes such as myself" bit14:15
superkuhKeep reading.14:15
superkuhAnd yes, a great april fools joke.14:16
-!- eridu [~eridu@gateway/tor-sasl/eridu] has quit [Remote host closed the connection]14:27
-!- eridu [~eridu@gateway/tor-sasl/eridu] has joined ##hplusroadmap14:28
kanzurehttp://techcrunch.com/2014/10/22/sosventures-takes-on-y-combinator-with-a-pure-biotech-accelerator/14:29
kanzurehaha "yes we're exactly like ycombinator with none of the lessons about software and growth"14:30
kanzure"Also of note, IndieBio is acquiring Berkeley BioLabs and bringing in Ryan Bethencourt"14:31
kanzureah that's news..14:31
fennWarning14:34
fennSome browsers may have trouble displaying this image at full resolution: This image has a large number of pixels and may either not load properly or cause your browser to freeze.14:34
fennhttps://en.wikipedia.org/wiki/File:VISTA%E2%80%99s_infrared_view_of_the_Orion_Nebula.jpg14:34
nmz787www.minimetalmaker.com/technology14:50
nmz787.title14:50
yoleauxTechnology of the Mini Metal Maker | http://minimetalmaker.com14:50
nmz787blah14:50
nmz787justanotheruser: re: last night (for me at least) talk on bitcoin paper... I said I was relatively a newb to an expert, then you replied nmz787: but as you said, that shouldn't matter14:53
justanotheruserindeed14:53
nmz787justanotheruser: but originally I was saying that a non-expert wouldn't be able to feasibilt evaluate the work of an expert when it comes to details14:53
justanotheruserI thought you said 02:53 < nmz787> that seems like an issue with the reader, that their fact/non-fact processing isn't robust enough to check or make note of thinks they don't understand, for someone else to check14:54
nmz787yes14:54
nmz787so it is a personal issue that i lack knowledge to evaluate the paper14:54
justanotherusernmz787: however you would likely be able to read it and understand the message14:55
nmz787it's not the paper's fault... 'it's math's fault'14:55
justanotheruserwat?14:55
nmz787there's a reason math is a bachelors and masters and phd level degree14:55
nmz787it's non-trivial14:56
nmz787especially when you're talking about security and looking for sidechannels14:56
justanotheruseryeah?14:57
justanotheruserMy original question was just how often such papers slipped by14:57
kanzurea lot of papers are crap14:58
kanzurethe vast majority of papers are terrible14:58
nmz787how many math nerds are there relative to non-math nerds, or people who don't like math at all14:58
nmz787(very few math nerds relative to population at large)14:59
fennover nine thousandth14:59
kanzurefenn: ever feel like the world is just a giant gish gallop attack against you15:00
fenn.wik gish gallop15:01
yoleaux"Duane Tolbert Gish (February 17, 1921 – March 5, 2013) was an American biochemist and a prominent member of the creationist movement. A Young Earth creationist, Gish was a former vice-president of the Institute for Creation Research (ICR) and the author of numerous publications about creation science. Gish was called "creationism's T. H." — http://en.wikipedia.org/wiki/Gish_gallop15:01
fennsorry i'm all out of tabs15:01
kanzurehttp://rationalwiki.org/wiki/Gish_Gallop15:01
kanzure"The Gish Gallop is the debating technique of drowning the opponent in such a torrent of small arguments that their opponent cannot possibly answer or address each one in real time. More often than not, these myriad arguments are full of half-truths, lies, and straw-man arguments — the only condition is that there be many of them, not that they be particularly compelling on their own. They may be escape hatches or "gotcha" arguments that ...15:02
kanzure... are specifically designed to be brief, but take a long time to unravel. Thus, galloping is frequently used in timed debates (especially by creationists) to overwhelm one's opponent."15:02
nmz787duh15:02
nmz787complex answers take time to explain to people not in the know15:02
kanzurethere's really nothing that is particularly complex15:03
-!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Ping timeout: 272 seconds]15:03
fenni disagree, there are lots of things that are humongously complex and nobody understands them15:03
fennthere are also plenty of things that are quite simple and elegant but for unknown reasons have been massacred over and over again until nobody knows wtf they are talking about15:04
fennlike quantum mechanics for example15:04
maakui understand quantum mechanics15:04
kanzurehe wasn't using quantum mechanics as an example of complex15:05
fennplease dont be offended if i dont take your word for it15:05
maakuthe problem is most physics professors don't understand quantum mechanics, because there's this culture that developed in which it was considered smart, even wise to label it as 'mysterious' and stop thinking15:05
fennthings that are complex: the brain, the economy, the immune system15:05
maaku(physics degree here, partly left physics for comp sci because i was disgusted with this phenomenon)15:06
fenni'm hoping garrett lisi's theory turns out to be true (or at least 'valid')15:06
kanzurelook on the bright side, you are still capable of experiencing disgust15:06
fennthen at least physics will be more than just a pile of observations15:07
maakuheh15:07
nmz787kanzure: also modelling that disgust with comp sci!15:08
nmz787we have a winner15:08
justanotheruserkanzure: have you intentionally been recruiting bitcoin devs into here?15:08
nmz787they should be called something better, like trustbit15:09
kanzurejustanotheruser: nope15:09
nmz787or safebit15:09
nmz787bithappy15:09
kanzurejustanotheruser: they just don't completely hate me15:09
fennthey're taking over, we never even talk about transhumanism anymore15:09
nmz787but not funbut15:09
kanzurei have benefits like email and bookmarks15:09
nmz787funbit15:10
kanzureand paperbot15:10
nmz787we should just store paperbot in the blockchain15:10
justanotheruseryeah. If only there was a channel with paperbot but no kanzure /s15:10
maakui was into transhumanism before bitcoin :)15:10
kanzurewe tried that but it turns out i'm the only one doing anything in here15:10
nmz787and have the propagtion delay serve as a delay loop15:10
nmz787to get computation cycles15:10
nmz787or something15:10
fennwe should just implement paperbot on a negative time delay switches, then he can fetch papers from the future15:11
nmz787psh15:11
nmz787i am doing micro/nanoscopy and fab shit15:11
kanzureyes but you're increasingly proprietary-only and that will eventually get inconvenient for me15:12
kanzureand anyone else in here.15:12
nmz787devs take time to release shit for OS15:12
fennwhats the point then15:12
fennopen source is not a dumping grounds15:12
kanzurecorrect.15:13
-!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 265 seconds]15:13
nmz787by shit i meant stuff15:13
fennby stuff you meant "finished" software15:13
nmz787did 'satoshi' release half-implemented code or papers or whatever?15:14
kanzureyes15:14
kanzurethat's why bitcoin is so broken15:14
nmz787well who did he have to compete with anyway15:14
nmz787ah15:14
kanzureall the other p2p currency ideas at the time15:14
kanzurethat's why 2 days after his release i shitted all over it15:14
kanzure"your software is fucking shit and i hate you"15:14
fennit's great they let people use it when they're done, but a) they lose the advantages of open source development, and b) there's no community built around it15:14
kanzurean open source strategy is actually pretty nuanced, it's not just "throw it over the wall"15:15
fennif it werent for marc fawzi i would probably be a millionaire now15:15
* fenn shrugs15:15
kanzurei should sue him15:15
kanzurehah15:15
kanzuremental distress15:15
kanzureand .. stuff.15:15
fennhire a palestinian to blow up his house :P15:15
fennnaw its just that he completely turned me off of virtual currencies15:15
kanzureit was pretty endless15:16
justanotheruserI feel like every time I argue with PoS people it becomes some petty subargument about how miners having the risk of DoS is a much smaller risk than someone reverse engineering a program to attack it15:16
kanzurehis emails were super long15:16
kanzures/endless/relentless15:16
kanzuremaaku: funny thing is that satoshi directly references marc fawzi in a few of his emails15:16
nmz787joules are the new jewels15:17
nmz787i'm telling you15:17
kanzureyou were around for those emails?15:17
kanzurevery early 2009?15:17
nmz787me?15:18
kanzureyes15:19
nmz787i existed15:19
nmz787I knew about bitcoin via silkroad media mention15:19
kanzureat the time, marc fawzi was spamming open manufacturing with his p2p joule currency15:19
nmz787huh15:19
kanzurewhich is the stuff satoshi referenced in his p2presearch emails15:19
nmz787idk15:19
kanzureanyway i was banning him a few times etc15:19
kanzuretypical "kanzure is an evil psychopath from the future" slurs were thrown around...15:20
kanzurehttps://groups.google.com/forum/#!msg/openmanufacturing/o_3Tp9mPaA4/Y_HuEb36rHUJ15:20
fennto be fair it was completely off topic15:20
kanzureOn Sun, 8 Feb 2009, Bryan Bishop wrote:15:20
kanzure> On Sun, Feb 8, 2009 at 9:15 PM, marc fawzi wrote:15:20
kanzure>> The two biggest dogmas present in current social theories are:15:20
kanzure>15:20
kanzure> Marc, could you please move this on to another mailing list now, maybe15:20
kanzure> the postscarcity mailing list, or even better, Michel's p2presearch15:20
kanzure> list? Since we've discussed these topics, and your P2P model, forever15:21
nmz787my openamnufacturing archive dates to 3/200915:21
kanzurehahaha. go me!15:21
kanzureah, so that may have been after15:23
kanzurebut just barely15:23
justanotheruserkanzure: I see you in my inbox15:38
justanotheruserand I am now curious who this dsmurrel guy is15:38
kanzurejust some spammer15:39
kanzurejustanotheruser: http://www.reddit.com/r/Bitcoin/comments/2k070h/enabling_blockchain_innovations_with_pegged/clgokbf15:39
justanotherusernevermind, based on the context I realized he was opencryptoreview15:39
justanotheruserhis website is a good idea, he's just to spammy IMO15:40
kanzureright, independent of whether it's good/bad, he's too spammy15:40
kanzurealso, there was once a time- if you can believe it- where it was unusual to /not/ see me in your inbox15:40
justanotheruserkanzure: what, you used to be a regular in the mailing list?15:41
kanzurenot that particular mailing list15:41
justanotheruserwell why would it be my inbox :P15:41
kanzurethat's just how it worked, my email would find ways of getting to people15:41
* justanotheruser crawls all transhumanism mailing lists for bryan15:41
justanotheruserlol15:42
justanotheruserkanzure: how do you host diyhpl.us15:44
jrayhawk_i throw servers at him15:45
justanotheruserI am tired of paying so much for ec215:45
justanotheruseryou probably use s3 if you use aws though15:45
kanzurehmm someone posted this https://github.com/kanzure/bitcoin-incentives/issues/415:45
kanzurejustanotheruser: i use aws, but i don't actually host anything on s3 except for the consulting stuff i do15:46
kanzurei mean, i use aws for commercial stuff. not personal stuff.15:46
kanzurei mean.. for commercial stuff for clients and customers.15:46
justanotheruser"Miners have incentives to reorganize the blockchain to change who gets transaction fees. They can do this without eroding trust by enforcing the same ordering of transactions."15:48
justanotheruserthe anti-incentive of lost rewards are very high15:48
kanzuretake it to -wizards, i don't want to proxy conversations between both channels15:49
kanzure(i just quoted from that page actually)15:49
jrayhawk_dis- is the operative prefix15:49
kanzurein irc anyone can make up whatever words they please, from henceforth i shall deem these malincentives15:50
kanzure(kidding)15:50
fennincensives15:55
kanzure"agitates into action or inaction"15:55
kanzurefenn did you look at the versum paper15:56
kanzurehttp://diyhpl.us/~bryan/papers2/bitcoin/VerSum:%20Verifiable%20computations%20over%20large%20public%20logs.pdf15:57
justanotheruserjrayhawk_: I anti-agree16:01
jrayhawk_oh god16:01
justanotheruserlol16:02
justanotheruserjrayhawk_: what is the operative prefix of "scales of economy"16:02
jrayhawk_while we're taking the english language out back and shooting it, i am going to go with unscales of economy16:05
justanotheruserreally?16:05
-!- ThomasEgi [~thomas@panda3d/ThomasEgi] has quit [Remote host closed the connection]16:05
kanzurewhat about just "economic inefficiency" there?16:05
justanotheruserkanzure: doesn't cover that scales matters16:06
justanotheruser*matter16:06
jrayhawk_"unscalability" is usually the term16:06
justanotheruseryeah, that makes sense16:07
kanzureso re: https://github.com/kanzure/bitcoin-incentives/issues/416:07
kanzurewait, no, wrong channel16:07
jrayhawk_hahaha16:07
justanotheruser19:07 < kanzure> justanotheruser: what do you think, is it safe for me to include that text verbatim and give attribution to that user?16:08
justanotheruserI'd include some...16:08
justanotheruserjk16:08
-!- Viper168 [~Viper@unaffiliated/viper168] has joined ##hplusroadmap16:10
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has quit [Ping timeout: 245 seconds]16:11
fennnmz787: how does metal clay toughness and tensile strength compare to the equivalent bulk metal?16:38
kanzurehttps://soundcloud.com/ilyadeep/ilya-gerus-alireza-dp16:41
fennmitsubishi precious metal clay looks to be around 10-20ksi (about the same as epoxies) but that isn't an engineering alloy16:41
fennkanzure your music taste is improving16:48
kanzurehttps://soundcloud.com/ilyadeep/ilya-gerus-vague-hallucinations-original-mix16:49
fenntoo bad i can't just paste the url in mocp16:50
kanzureyoutube-dl or cclive should be able to get it16:51
fennyeah16:51
fenni'm just wondering if the computer world will ever not be fractured and broken16:52
kanzurewould probably have to be something about a set of principles for expectations and estimations that everyone would have to design around, so that they can avoid making unreasonable expectations of systems or implementations16:53
fennan expectation of making the data available would be a good start16:53
fennyoutube-dl is one hell of a compatibility layer16:54
fennit has 7 pages of options16:54
fennit's designed to actively circuvent limiting measures16:55
kanzure904 open issues https://github.com/rg3/youtube-dl/issues16:55
kanzurehell i should just put paperbot in there16:56
justanotheruserkanzure: they might appreciate it. you should ask16:57
kanzurei'd want to wait until i finish up unit tests for paperbot v2 first16:58
kanzureand then switch over to actively using paperbot v216:58
kanzurewhich has just been sitting there. all alone.16:58
justanotheruserkanzure: wait, you didn't give paperbot the ability to feel loneliness did you?16:59
nmz787fenn: not sure, but you can probably find more info by looking into MIM (metal injection molding)16:59
kanzurehonestly i forget16:59
fennyoutube-dl  --list-extractors | wc -l17:00
fenn45117:00
nmz787fenn: in both cases it seems like organic binders, two stages (a wax and a polymer), and metal particles of varying mixtures17:00
nmz787fenn: and then again in both cases, kiln curing accompanied by some shrinkage17:00
fenn30% shrinkage is a lot but what's more important is the uniformity of shrinkage17:01
nmz787the few youtube vids on MIM seemed very nice17:01
kanzurehah what17:01
nmz787like a scale factor would be fine17:01
nmz787g2g now though17:01
justanotheruserpaperbot:  http://eprint.iacr.org/2014/85217:09
justanotherusercan we ban paperlooker :P17:09
justanotheruserpls don't make me failtab17:09
kanzure.title17:14
yoleauxCryptology ePrint Archive: Report 2014/85217:14
kanzurehmm i don't know if paperbot supports this17:14
kanzurezotero seems to https://github.com/zotero/translators/blob/master/ePrint%20IACR.js17:14
justanotherusersomeone posted in -wizards17:15
-!- lichen [~lichen@c-50-139-11-6.hsd1.or.comcast.net] has quit [Quit: Lost terminal]17:22
-!- |b| [~|d|@ip68-107-37-158.sd.sd.cox.net] has joined ##hplusroadmap18:05
-!- JingoFett [~|d|@ip68-107-37-158.sd.sd.cox.net] has quit [Ping timeout: 256 seconds]18:07
-!- andytoshi [~andytoshi@wpsoftware.net] has joined ##hplusroadmap18:08
-!- andytoshi [~andytoshi@wpsoftware.net] has quit [Changing host]18:09
-!- andytoshi [~andytoshi@unaffiliated/andytoshi] has joined ##hplusroadmap18:09
-!- Beatzebub [~beatzebub@S0106b81619e8ecee.gv.shawcable.net] has joined ##hplusroadmap18:20
-!- CheckDavid [uid14990@gateway/web/irccloud.com/x-mizesietfccgspgb] has quit [Quit: Connection closed for inactivity]18:47
-!- nsh_ [~nsh@wikipedia/nsh] has quit [Ping timeout: 258 seconds]18:47
kanzurei need more monitors18:47
kanzureand keyboards18:47
-!- Vutral__ [~ss@176.10.107.235] has quit [Ping timeout: 246 seconds]19:05
-!- maaku [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has quit [Remote host closed the connection]19:09
-!- maaku [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has joined ##hplusroadmap19:12
-!- maaku is now known as Guest5607219:12
fennmore ... brains ...19:15
fennbraaains19:15
-!- rak[1] [~rak@opensource.cse.ohio-state.edu] has quit [Remote host closed the connection]19:16
-!- eridu [~eridu@gateway/tor-sasl/eridu] has quit [Ping timeout: 246 seconds]19:18
kanzurethe problem is too disjoint and stuff.19:19
kanzurealso, lulls are annoying19:21
fenni do it for the lulls19:22
-!- rak[1] [~rak@opensource.cse.ohio-state.edu] has joined ##hplusroadmap19:22
fennlook, fresh brains19:23
kanzurethose are lolz or lulz not lulls19:24
kanzurei'm not sure how to organize all this bitcoin information19:25
kanzureeven in https://github.com/kanzure/bitcoin-incentives/blob/master/bitcoin-incentives.tex19:25
kanzurea list without numbered or named entries is a bad idea19:26
kanzureit's a very foreign method of software development19:27
fennso give them numbers19:29
fennwhat does \begin{enumerate} do19:29
kanzureoh right, numbers.19:29
kanzureit does numbering things.19:29
kanzurebut still.19:29
fenndo the numbers change if you add or remove stuff?19:30
kanzurewouldn't it be better to have some method of cranking these out automatically instead of it just being "whatever bryan happens to remember or come across or intuit"?19:30
kanzureyes :(19:30
kanzure\item{name} works i think?19:30
fennyou're essentially doing observations on a thought experiment19:31
fennyou can also observe actual in-the-wild behavior and infer incentives from that19:32
fennsometimes things you think will be incentives, aren't, and for no good reason19:32
fennlike, people are stupid19:32
fennor they just dont know any better19:33
kanzurethese should all be explicit though19:33
kanzureand the code should make them as explicit as possible19:33
kanzureand then optionally give you some stupid parameters for tweaking things if possible19:34
fennwhat about "you get mad propz for running a node"19:34
kanzuredo you?19:34
fennhave you interviewed miners about what they do and why19:35
kanzurei have talked with miners regularly for quite a while now19:35
fennaccording to traditional economic theory linux never should have existed19:35
kanzurebut not interviewed, not really19:35
fennso i think it's unreasonable to expect you can discover all existing incentives just by thinking real hard19:36
kanzurelukejr operates eligius and he hangs out in wizards and often mentions what sort of custom tweaks he has made re: incentives for his pool19:36
kanzureeligius is an unfair example because it's probably the best maintained pool (probably the only one with a bitcoin core developer wizard person hanging around?)19:36
fennwhy is 40 TBC == 0.04194304 BTC19:46
kanzureTBC?19:48
kanzureis this an altcoin?19:48
fenni dont know what TBC is but apparently 0.1 TBC == 0.00004096 BTC19:48
kanzureoh this might be testnet bitcoin? yeah they had to reset testnet a few times becaues people started trading testnet bitcoin for real money19:48
kanzureso now they are on testnet319:48
kanzureanyway there's also a mode called regtest, which is like testnet except infinitely easier to test against because it doesn't require testnet's blockchain (starts at block 0 or 1?)19:49
fennits being used as payout for eligius mining reward19:49
fenndoesnt make sense to me why that would use testnet19:50
fenni also found some wackjob numbering system called "tonal bitcoin"19:50
kanzureah, then it's probably not testnet. you best go ask in #bitcoin i suppose (but there's sharks in those waters).19:51
-!- |b| is now known as JingoFett19:51
* kanzure considers http://people.xiph.org/~greg/simple_verifyable_execution.txt19:52
fenni'm basically treating all this like reading a foreign language19:52
fenn"hey i know that word"19:52
kanzure"that is not a word i would ever have any interest in knowing"19:52
fenna hash is like an exploded tater tot right20:01
fenndid you actually understand that last link20:02
-!- yashgaroth [~ffffff@2606:6000:cb85:6a00:58e4:dbb7:29ec:b236] has joined ##hplusroadmap20:03
fennit made sense up until "Now lets define an adaptation key for two wires."20:04
kanzurewell, i was reading that link while also watching some tv, replying to an email, selecting fine music, and double checking some thoughts about sidechains.pdf -- so essentially, no not yet.20:04
kanzureand also apparently contributing to https://github.com/TheBlueMatt/bitcoinninja/commits/master20:06
kanzurealthough i don't remember doing that20:06
kanzurei never really understood wikipedia's explanation of zero-knowledge proofs, it was always something about some stupid tunnels and caves.20:07
fenni never really understood proofs20:07
fennit always just seemed like a let down, or some kind of cheat20:07
kanzurethese sorts of proofs seem very different from the traditional theorem kind20:08
kanzurethey actually seem a little bit more accessible to me.20:08
fenni havent read it yet but i expect a zero knowledge proof is just a machine that you know will work even though you dont know all the inputs20:09
kanzurepaperbot: http://link.springer.com/chapter/10.1007/978-3-540-72540-4_4#page-120:09
paperbothttp://diyhpl.us/~bryan/papers2/paperbot/f8a545c66f53888e6690517d9481af92.txt20:10
kanzurepaperbot: http://link.springer.com/content/pdf/10.1007%2F978-3-540-72540-4_4.pdf20:12
paperbotXMLSyntaxError: None (file "/srv/ikiwiki/paperbot/modules/papers.py", line 524, in parse_html)20:12
kanzure<iframe src="https://library.rit.edu/depts/assets/ezproxy/ip-check.php" class="iframe" frameborder="0" allowtransparency="true">This library resource requires additional authentication for use from off-campus or because of specific licensing agreements.</iframe>20:13
fenni guess its more like "prove you are actually my friend from high school and not an impostor from an intelligence agency who studied his dossier"20:15
fennthe more unlikely facts he knows, the less likely it is that he is an impostor20:17
-!- eridu [~eridu@gateway/tor-sasl/eridu] has joined ##hplusroadmap20:24
* fenn sleeps20:25
kanzurefun stuff: http://nozdr.ru/data/media/biblioteka/kolxo3/Cs_Computer_science/CsLn_Lecture%20notes/20:38
kanzure.title http://www.eroids.com/20:45
yoleauxSteroid Source Reviews. Check your supplier!20:45
kanzurealright let's do a group buy20:49
kanzurewho's in?20:49
-!- kumavis_ [~kumavis@107-219-148-42.lightspeed.sntcca.sbcglobal.net] has joined ##hplusroadmap20:58
-!- eridu [~eridu@gateway/tor-sasl/eridu] has quit [Quit: Leaving]21:03
-!- yash [~ffffff@cpe-76-167-105-53.san.res.rr.com] has joined ##hplusroadmap21:11
-!- yashgaroth [~ffffff@2606:6000:cb85:6a00:58e4:dbb7:29ec:b236] has quit [Ping timeout: 265 seconds]21:12
-!- Guest56072 [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has quit [Remote host closed the connection]21:22
-!- maaku [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has joined ##hplusroadmap21:24
-!- maaku is now known as Guest682321:24
-!- Guest6823 [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has quit [Remote host closed the connection]21:25
kanzure"Proudly processing 2 billion nops per second since 2003."22:02
-!- paperlooker [~tiktaalik@50-79-188-182-static.hfc.comcastbusiness.net] has quit [Ping timeout: 265 seconds]22:07
-!- yash [~ffffff@cpe-76-167-105-53.san.res.rr.com] has quit [Quit: Leaving]22:14
-!- maaku_ [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has joined ##hplusroadmap22:18
-!- bbrittain_ [~bbrittain@172.245.212.12] has joined ##hplusroadmap23:08
-!- DonnchaC1 [~DonnchaC@windmill.donncha.is] has joined ##hplusroadmap23:09
-!- bbrittain [~bbrittain@172.245.212.12] has quit [Ping timeout: 272 seconds]23:10
-!- bbrittain_ is now known as bbrittain23:10
-!- Burnin8 [~Burn@pool-71-191-174-26.washdc.fios.verizon.net] has joined ##hplusroadmap23:10
-!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 260 seconds]23:10
-!- DonnchaC_ [~DonnchaC@windmill.donncha.is] has quit [Ping timeout: 260 seconds]23:10
-!- hehelleshin [~talinck@66.161.138.110] has joined ##hplusroadmap23:12
-!- Burn_ [~Burn@pool-71-191-174-26.washdc.fios.verizon.net] has quit [Ping timeout: 272 seconds]23:13
-!- helleshin [~talinck@66-161-138-110.ubr1.dyn.lebanon-oh.fuse.net] has quit [Ping timeout: 260 seconds]23:13
-!- maaku_ [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has quit [Read error: Connection reset by peer]23:15
-!- maaku [~quassel@50-0-37-37.dsl.static.fusionbroadband.com] has joined ##hplusroadmap23:15
-!- maaku is now known as Guest3391323:15
-!- sheena [~home@S0106c8be196316d1.ok.shawcable.net] has joined ##hplusroadmap23:16
-!- Vutral [~ss@mirbsd/special/Vutral] has joined ##hplusroadmap23:19
-!- Guest33913 is now known as maaku23:21
-!- Vutral_ [~ss@31.7.56.132] has joined ##hplusroadmap23:26
-!- ebowden [~ebowden@147.69.54.84] has joined ##hplusroadmap23:49
-!- TC [~talinck@66-161-138-110.ubr1.dyn.lebanon-oh.fuse.net] has joined ##hplusroadmap23:56
--- Log closed Thu Oct 23 00:00:25 2014

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