log☇︎
152000+ entries in 0.095s
asciilifeform: BingoBoingo: sorta like comp market in 1995 was superficially similar to walking into one from '85, except for the detail where all of the 'different comps you can CHOOSE from' run microshit on x86
mircea_popescu: asciilifeform oh, the little wolverine huh. she's one spirited hussy.
asciilifeform: BingoBoingo: i run into these continuously
mircea_popescu: scratch the surface of "diversity" and "World of choice" juyst a little, admire the self-same oroborus maggot.
BingoBoingo: So, in new lulz... Telestra is soliciting business by posing as independent datacenters
asciilifeform tried, and failed, to distinguish linked site from the usual inmate-fat-powered spamulag
jhvh1: mircea_popescu: (ticker [--bid|--ask|--last|--high|--low|--avg|--vol] [--currency XXX] [--market <market>|all]) -- Return pretty-printed ticker. Default market is Bitfinex. If one of the result options is given, returns only that numeric result (useful for nesting in calculations). If '--currency XXX' option is given, returns ticker for that three-letter currency code. It is up to you to make sure the code is a valid (1 more message)
asciilifeform: can also sum how much they 'dun make' at nasdaq; and while we're at it, how much fish they don't have at the proverbial sov fish counter
mircea_popescu: should i have said " how much dough the etsy chicks don't make" ?
asciilifeform: and 0 to do with american chixen knitting
asciilifeform: i thought 'etsy' et al were long ago colonized by folx farming out grunt labour to pakistani etc
mircea_popescu: and if anyone was wondering how much dough the etsy chicks make, and which ones, and out of what exactly, http://www.craftcount.com/category.php?cat=3&subcat=4
mircea_popescu: the proper equivalent of "he has either x of q or y or p but not both" is equivalent to "i only use p, and he has an x but IT IS NEVER THE X HE NEEDS haha! take that bitch! murphy works for me!"
asciilifeform: but this is nonsense. he has x because he is able to - sometimes - infer it from a partial knowledge of q, because relation between x and q stream was 1:1. with multiple generators, it is now many to many.
mircea_popescu: now it should be apparent : in the first case you let attacker have X of Q. in the second case, you MUST let him have X of Q and Y of P.
mircea_popescu: here's what you move to : "I have a Q and a P". here's the correct statement : "Of all the possibles that may exist, Q and P, I have both".
mircea_popescu: here's how you state the story : "I have a Q". here is the correct statement : "Of all the possible Q that could ever exist, I have Q, the one and sole and only."
asciilifeform: if 'enemy is a telepath and can access anything', you don't have crypto, you have eating pistol or surrender
mircea_popescu: nope. in fact, thinking about enemy in this specified sense is inviting error.
asciilifeform: it does tho
mircea_popescu: lobbes how the fuck is bringing watermelons supposed to be discriminatory ? he wasn't gonna let the white guys have any ?
asciilifeform: mircea_popescu: in your gedankenattack, to what items does enemy have access, and how did he get it ?
a111: Logged on 2017-10-11 16:53 asciilifeform: proposed item is that if using 1 stream, knowledge of a segment of plaintext gives you something valuable to work with; if using 2+ -- nothing
mircea_popescu: asciilifeform nevermind that waste of time/effort. think : when you say http://btcbase.org/log/2017-10-11#1724006 you are actually imposing further conditions on the attacker than the null case. ☝︎
shinohai: lobbes: They were prolly pissed because he didn't bring a bucket of KFC too
asciilifeform: which then gets used as otp per mircea_popescu's original algo.
mircea_popescu: what shall i do with this guy.
asciilifeform: the streams exist solely to be xored together into a pad
mircea_popescu: asciilifeform point there was that enemy will have access to some plaintext of each stream.
asciilifeform: 'every time i add two numbers i get diff answer; arithmetic IS BROKEN'
BingoBoingo: mircea_popescu: ABout to find out
a111: Logged on 2017-10-11 14:25 phf: but at the same time it's handy strawman: ~web of trust~ is broken, because our take on it is a piece of shit. i thought that's some pantsuit pattern discussed elsewhere
mircea_popescu: tl;dr : sad story of non-tmsr wot implemented by heathens. 60k or so ratings totally exchanged, item collapsed into uselessness. why ? because not had and therefore not observed the http://trilema.com/2014/what-the-wot-is-for-how-it-works-and-how-to-use-it/ bible.
mircea_popescu: of relevance to republican interests -- https://www.blackhatworld.com/seo/bhw-trade-volume.978942/ ☟︎
mircea_popescu: and in other (lengthy) lulz : i cursed new girl with the old mpoe-pr curse, if you haven't guessed by now. the tardstalk this time is "black hat world", dedicated to "seo" and shit.
asciilifeform: ground rule of crypto is that enemy has access to ciphertext and the occasional bit of plaintext ( guessed. ) if he has access to inside of your box we ain't doing cryptology, it is called perimeter defense
mircea_popescu: what is so SPECIAL about him that you get to move but he has to stay put ?
mircea_popescu: you rpopose to move to situation two, where there's two streams AND YOU ALSO PROPOSE to not permit enemy to know 500bytes of each stream. why not ?
mircea_popescu: look. situation 1, there's one stream, and enemy knows 1kb.
asciilifeform: see the otp thread
asciilifeform: exactly that
asciilifeform: knowing xor tells you 0 about what went into the xor.
mircea_popescu: if you get to double your heads, so does eve : now she has knowledge of a segment half size in each stream. tada.
mircea_popescu: but that nothing is entirely predicated of assumptions on how exactly "knowledge of segment" translates into the new world.
asciilifeform: proposed item is that if using 1 stream, knowledge of a segment of plaintext gives you something valuable to work with; if using 2+ -- nothing ☟︎
mircea_popescu: the problem now is at the next link in the chain, the fs.
mircea_popescu: the problem is not the seeds per se, that fg solved.
asciilifeform: ftr i dun particularly like this style of mechanism. it is suggested solely as continuation of the theme from linked thread, where 'make best of a cpu-poor situation with existing algos'
asciilifeform: proposed mechanism is applicable to 'stacking' any but the most screamingly rubbish prng. elementarily if the seeds are independent ( and they are, having been birthed from FG, ideally even from separate physical units ) the value of the streams at whatever index are also uncorrelated; and the xor will have period no shorter than of the longest period prng.
mircea_popescu: leaving us with this situation where "a different hash function" is not even a meaningful mathematical construction.
asciilifeform: after 3 generations of dried dog shit, quite conceivably the last actual crytoanalyst ate nagant 30yrs ago.
mircea_popescu: yet nevertheless, the "g is a different hash function from f" "how do we know ?" "because george named his g and florence named hers f" is nonsense.
asciilifeform: i dun buy the 'no one has proven rsa to be hard so it dun matter how to implement it, let's use wet noodles and dried shit' argument. ☟︎
mircea_popescu: and so in certain context, such as this one, the correct assumption...
mircea_popescu: asciilifeform i don't have the proof it is, but that's irrelevant. you don't have the proof IT ISNT.
mircea_popescu: that was the point of all those galois fields discussions recently.
mircea_popescu: asciilifeform you don't know it's ACTUALLY different ; the expression sounds different, but there's a clear modulo space / elliptic curve relation.
asciilifeform: if mircea_popescu has the proof that 'all the same' quite a few folx would be interested to read.
mircea_popescu: discrete curves, diophantine equations, it's all the fucking same.
mircea_popescu: and this isn't a joke : the "distinction" between rsa and ecc, whereby "ecc is faster" or "has longer effective key" is bs. ECC is exactly RSA in polar coordinates, if either fails mathematically both do.
mircea_popescu: to recapitulate for the noobs : the converse of "we have no strong security models for hash functions" does ALSO imply that we CAN NOT DISTINGUISH THEM.
asciilifeform: let's take however the variant where they all use same type of hash (e.g. keccak) but with independently FG-generated seeds. you still remove the i+i==H(i) known-relation problem.
mircea_popescu: that's nice, but tell you what, 1880 airplanes all failed if one did. and one did. any one.
asciilifeform: who said i had 1 much less n to offer with this recipe. i learn from the masters!11
mircea_popescu: i can't fucking find one. you want me to use n
mircea_popescu: cuz this exists now "?
mircea_popescu: well how. if hash function is preimage weak, they all will be.
asciilifeform: each one in particular runs as mircea_popescu's old algo does. the otps, however, are xored to form the otp that gets used in the end.
asciilifeform: they ain't hashed together, no
asciilifeform: you can trivially prove that, for crypto break (vs e.g. side chans) enemy must break either a) all of the rsagrams b) all of the hashes
mircea_popescu: i don't get it. so instead of one item with the above three problems, i now have three, and they're hashed together. so ?
asciilifeform: also eliminates the problem where there is a known relation between adjacent pieces of ciphertext where plaintext is known
a111: Logged on 2016-02-10 19:29 asciilifeform: (incidentally, the problem of an inner cipher introducing known-plaintextisms is solved routinely by splitting the payload into xor-able halves, using rng, and enciphering each ~half~ with different cipher, rather than box-in-box composition)
asciilifeform: it buys you same thing http://btcbase.org/log/2016-02-10#1402315 buys you -- strength of the strongest one ☝︎
mircea_popescu: and srsly don't later tell me when im right here wth.
mircea_popescu: other than more complexity, what's this buy me ?
jhvh1: asciilifeform: The operation succeeded.
a111: Logged on 2017-10-06 23:13 mircea_popescu: basically the scheme is, you rsa a random bitfield, then you expand that into as much otp as you want by doing recursively Fi = hash(bitfield + Fi-1). there's a limit on i, obviously, which can be set to 1.
asciilifeform: !~later tell mircea_popescu i had a thought re your http://btcbase.org/log/2017-10-06#1722036 algo : instead of one stream generator, have N independent ones (each with own seed, and not necessarily all same hash fn) running in parallel, xor together. every rsagram changes a seed in ONE of these at a time. strength is bought by sending rsagrams moar often ☝︎
mircea_popescu: lmao "atenCOIN is rebranding to blablaBITCOIN". oook ?
mircea_popescu: it's a good thing the human rights crowd informs us "everyone wants to be and is an individual", because from experience i'd never have arrived at the idea.
mircea_popescu: quite exactly, it's never futurism as the 1920s artistic blaseness. it's always "i will r entrepreneurize" or "i will marry well" or such, translated in the new paints.
mircea_popescu: phf just about same here. i particularily enjoy situations where the "group expert" in "new trends" brings bitcoin to the group. particularily if 20something female.
jhvh1: shinohai: The operation succeeded.
shinohai: !~later tell jurov http://wotpaste.cascadianhacker.com/pastes/zOfhc/?raw=true
phf: asciilifeform: well to do middle aged ru dentist types talked to me of bitcoin mining with a "please explain what the hell is going on". it's very much the opposite of the beobachter take, still as lulzy.
asciilifeform: phf: it was part of the fiction where 'no we didn't surrender to americans, your rape is a traditional ancient greek, dontchaknow, rape style, nuffin to do with usa'
shinohai: Since event is being held in Russia, inb4 Russian hackers fucked with the team
phf: (ipoteka is a ru word by way of greece for mortgage, i've not heard it before when growing up, but now at a certain age it's a main conversation subject/as well as advertisement, so i've been cracking up every time i hear it. i've been forcing it as a catch all term for "orc financial matters" to the confused looks of my ru friends)
asciilifeform: asciilifeform's relatives are ~all elderly emigre type and read volkischer beobachter and so naturally 100% of mention bitcoin is 'hey didja hear, that bitcoin thing finally HACKED!11!! soon valueless, president of imf proclaimed!' etc snoar
shinohai: lmao the U.S. futbol team was defeated 2-1 by *Trinidad and Tobago* and will not advance to the World Cup.
phf: depending on the person either the context of "you're a banker, what do bitcoin investment" or else futurism "soon there won't be ИПОТЕКА everything will be in bitcions, have you heard?"
phf: every single 'civilian' conversation in ru i had with friends and friends of friends at some point had bitcoin brought up, and obviously not coming from me. in fact not even as part of "talking technology" track
asciilifeform: my current understanding is that the ethertardium etc folx today target moneyed orcs, rather than homo-mmmicus .
mircea_popescu: local orcs have their own homegrown equivalents of graham-style "entrepreneurship".
mircea_popescu: idiocy is not really cross-cultural in this strict formal sense.
asciilifeform: i saw a little reportage, it was interestingly the ~exact same spamcarnival as seen prev in usa & elsewhere. oddly ~no homegrown orc scamola .
mircea_popescu: all this "conference" bs is much a subleaf of that, for the loser/womenz to try hand at.
asciilifeform: the ukrs did in fact recently put on some public spectacle where they hosted 'up and coming' shitcoinists, ethertards, et al
mircea_popescu: i guess they don't know what cyprus is huh.