Results 1 ... 118 found in all logged channels for 'bernstein f:asc'
(asciilifeform) asciilifeform: http://logs.nosuchlabs.com/log/asciilifeform/2022-07-07#1110799 << asciilifeform walked around for many yrs looking at heathen (nominally) 'p2p' items, errywhere found some moxyesque absurdities/atrocities (e.g. acceptance of heavy memory/cpu load from strangers ; sslism; bernsteinism; and, more often than not, sham-'p2p' where some nodes
(asciilifeform) asciilifeform: in which capacity it continues to perform a++ (w/ e.g. bernstein's shilling) despite the continued nonexistence of the 'philosopher's stone', err, qc, per se.
(asciilifeform) asciilifeform: for brief time d. bernstein was a kind of exception -- but swamp quickly swallowed him, and for many yrs now he has not been distinguishable from other swamp creatures.
(trilema) asciilifeform: http://logs.ossasepia.com/log/trilema/2019-10-22#1947549 << asciilifeform is partial to serpent , but strictly because actually had a chance to do some analysis of it ; and dislikes sha for the obv. reason, and chacha ditto for the reason of originating from post-brainrot bernstein. but theoretically all of these snake oils are equally snake oils, and difficult to argue against the standard 'drink t
(trilema) asciilifeform expected to find that linked item is re ~mass~ gcd, in the sense of 'bernsteinization' used in phuctor. but apparently djb has nothing to add re ~that~
(trilema) asciilifeform: ( in selected cases, e.g. where bernstein prefers to smoke crack rather than showing up , can discern. but their song is sung , not much to add to'em )
(trilema) asciilifeform: at some point i'ma do the bernsteinizer in straight asm. ( but this is ~9000th on the list of things-to-do atm )
(trilema) asciilifeform: the 'uncut' gmp is not afaik used anywhere in our mechanisms, with the exception of phuctor, where a (patched) variant is used for bernsteinization
(trilema) asciilifeform: phunphakt : bernstein (aka djb) pushes urandomism. actively.
(trilema) asciilifeform: i dun remember any such thing, never thought to do it with bernstein et al
(trilema) asciilifeform: ( i.e. why bernstein not replied ? only odin knows )
(trilema) asciilifeform: in other lulz, the 'bernstein as nsa stooge' concept slowly percolates from the republic, https://sporaw.livejournal.com/538323.html
(trilema) asciilifeform: thing is, any idjit (turns out ) can bernsteinize
(trilema) asciilifeform: mircea_popescu: sorta hilarious, imho, took usg 5 whole yrs to finally implement bernstein ( supposing thing even worx, i haven't found the giving shit to actually try it )
(trilema) asciilifeform: i had this knob on the implementation conveyor regardless, because of a certain peculiarity of bernstein's algo which i noticed, where if there is an even multiple of any given factor present in the whole, it will not get found
(trilema) asciilifeform: as for 'uwb', they were content to issue ban edicts, and then for good measure to bernsteinize all of the remaining 'experts' they could get to.
(trilema) asciilifeform: funnily related story, back when asciilifeform was a uni student, cellpnojeism was beginning to spread among plebelands, and asciilifeform noticed that folx were paying telco fiddycents/minute to talk over 1-2km ! so asciilifeform went, like naive idjit, to engineering profs, with 'p2p telephony' proposal thing, reaction was the exact sort of through-the-teeth laughter of today's bernstein re phuctor .
(trilema) asciilifeform: ( these, as alert reader prolly realizes, signal immediately, they do not require bernsteinization to find )
(trilema) asciilifeform: ( keys with dupe mods get processed correctly, they are addressable as distinct keys but mods are references to a mods table, and dupes get marked as dupes. bernsteintron works on deduped mods table. )
(trilema) asciilifeform: bernsteinization dunwork if you dun dedupe.
(trilema) asciilifeform: meanwhile, in a galaxy far away, 'where did bernstein go?' 'to write crappy stego apps' >> http://elligator.cr.yp.to ( https://archive.li/KwRAu )
(trilema) asciilifeform: ^ from 2018 ! ( asciilifeform very recently fixed the corner case in bernstein which permitted this kind of thing )
(trilema) asciilifeform: ( the way bernsteinism works, a given run does not reduce the work of subsequent run in any way )
(trilema) asciilifeform: + entirely separate c proggy that actually does the bernsteinization.
(trilema) asciilifeform: iirc i originally introduced phuctor as 'catalogue of rsa keys which are inexpensively breakable'. for said formulation it does not matter precisely ~how~, if tomorrow i conceive of a wholly novel inexpensive break, i will apply it to phuctor with the others ( as i applied classical methods, gcd, bernsteinistic gcd, fermat, ( in the worx...) lenstra , etc . )
(trilema) asciilifeform: and find better bernsteins lol
(trilema) asciilifeform: this sounds great until it is time to 'unfind' a factor, as in http://btcbase.org/log/2018-04-25#1805504 or the prev case where the thing turned up a corner case in bernstein and found 300+ or wat was it spurious factors
(trilema) asciilifeform: i've no particular objection to snake oil from king cobra vs from japanese viper; but as i observed earlier, the sudden popularity of bernsteinism has never been explained to my satisfaction.
(trilema) asciilifeform: mircea_popescu: blake2 is bernstein's hash ( consists of a slightly modified chacha, his symmetric algo )
(trilema) asciilifeform: how did you settle on the use of bernsteinian cryptoprimitives ( e.g. chacha ) ?
(trilema) asciilifeform: http://btcbase.org/log/2018-04-11#1796064 << naturally fast -- uses post-betrayal bernsteinian cryptosnakeoil, instead of old-fashioned nsa-inhouse snakeoil
(trilema) asciilifeform: hell, bernstein et al ~with~ theorems still went to ~same place.
(trilema) asciilifeform: i thought bernstein were a bloke
(trilema) asciilifeform: http://btcbase.org/log/2017-11-16#1739455 << not so surprising, considering that bernstein himself is a quantumist
(trilema) asciilifeform: apeloyee: did you read same bernstein as was linked ? pretty clearly pins 'constructed primes'
(trilema) asciilifeform: in other lullies, bernstein exposes the item the original 'authors' sat on in http://btcbase.org/log/2017-10-17#1725945 >>> https://blog.cr.yp.to/20171105-infineon.html
(trilema) asciilifeform: i still think that it makes sense to do this only after every other bolt is as tight as physically possible -- bernsteinian karatsuba, unrolled comba, etc
(trilema) asciilifeform: phf: the bernsteinian 'werker' is also c proggy. ( phuctor , in all 4 versions to date, consisted of 2 pieces, running wholly separately )
(trilema) asciilifeform: phf: i dun think i ever posted the modern ( bernsteinian ) one.
(trilema) asciilifeform: but hypothetically it may even be possible to ffaize bernstein's tree. or even to do it in such a way that doesn't wipe out the cpu winning from it. and even possibly to prove that it works and doesn't leak bits and doesn't let composites through once in a while.
(trilema) asciilifeform: i used bernstein's tree in phuctor, where it made actual sense
(trilema) asciilifeform: and then bernsteinian karatsuba, possibly, and whatever else i can think of.
(trilema) asciilifeform: http://btcbase.org/log/2017-10-07#1722400 << bernstein's gcd method is neither here nor there, i certainly don't need anything of the kind in ffa, and quite likely it fundamentally does not ffaize
(trilema) asciilifeform: incidentally ~95% of the work ffa does in modexp, now, is multiplication. which means that there is further 20-25% speedup waiting to be had when i get bernsteinian optimization for karatsuba ( haven't yet figured it out, he buried it deep in a paper , as if he were an alchemist, quite cryptically ) and another 10-20% optimization if we move to unrolled comba ( see august thread. )
(trilema) asciilifeform: he's an older, smaller-calibre bernstein
(trilema) asciilifeform: ( bernsteintronic phuctorator dun care how many, nor ever uses the individual factors per se for anything )
(trilema) asciilifeform: he ain't named bernstein by any chance??
(trilema) asciilifeform: ( the comp sci papers that actually mattered, plus a handful of 1990s bernsteins, are already on my hdd... )
(trilema) asciilifeform: massive pile of moving parts, aes, various post-conversion bernsteinisms, null ciphers, 'this is faster on 32-bit cpu so we're using it', let's-give-enemy-raw-bytes-from-prng, and other jokes.
(trilema) asciilifeform: but in very other olds, apparently in an obscure article in '09 bernstein shows how to eliminate one of the middle-term additions of karatsuba .
(trilema) asciilifeform: and nao bernstein, henninger ( this is what, 3rd paper since she was attached to him ) 'unhappened and rehappened' it
(trilema) asciilifeform: phf: also at a certain point 'they don't know about tmsr' evaporates and turns into 'they studiously ignore, like bernstein'
(trilema) asciilifeform: bernstein, who would not even ANSWER message re phuctor
(trilema) asciilifeform: possibly the only example of something that comes close, that i know of , is bernstein's gcd
(trilema) asciilifeform: Framedragger: i wrote to, e.g., bernstein, more than once.
(trilema) asciilifeform: not the lispm folx, not bernstein, whoever.
(trilema) asciilifeform: fwiw i wrote to him after i deployed phuctor-cum-bernsteintron. 0 answer.
(trilema) asciilifeform: or how about bernstein vs. unitedstates.
(trilema) asciilifeform: mircea_popescu: bernstein ?! has multi-MB www of unique and valuable work
(trilema) asciilifeform: meanwhile, daniel bernstein ( of the smooth-parts-of-integers algo used in phuctor, and 1,001 other useful mathematical works ) is headed for gasenwagen: https://eindhoven.cr.yp.to/false-statements-by-henry-de-valence.txt
(trilema) asciilifeform: 'Remember that most of these """journalists""" grew up either during Watergate or in its shadow, and so it's hardly surprising that they see the role of the press as some kind of unelected fourth branch of government whose job is to "take down" the President, if all else fails. Bernstein and Woodward are their gods.'
(trilema) asciilifeform: relatedly vintage bernstein, http://archive.is/xZZmb >> ''Some people argued that randomness generation should be centralized--- whether in the OS or hypervisor or CPU---and fixed at that central location if it doesn't work properly. Other people argued that each library and application should defend itself against failures of the centralized mechanism. There actually seem to be three positions...''
(trilema) asciilifeform: mircea_popescu: quite obviously i do not know bernstein as a human, but can only conjecture, that he felt like a proper rube , ripe for fleecing in orc land.
(trilema) asciilifeform: http://btcbase.org/log/2016-12-31#1594814 << bernstein's algo does not use a running global product. see his 'finding the smooth parts of integers'.
(trilema) asciilifeform: so it is special case, it is marked by bernstein but must stay green until another factor is found.
(trilema) asciilifeform: it does not. but by bernstein's algo -- does.
(trilema) asciilifeform: worst case is that we (or somebody ELSE) found an undocumented boojum concerning bernstein's algo.
(trilema) asciilifeform: so for next couple of hours, until the second half of this bernsteinization, there will be 100s of mods marked 'phuctored' for whom the only marked factor -- is themselves. which is Wrong
(trilema) asciilifeform: (and bernsteinization requires access to ~all~ moduli, as i think is obvious, and not simply 'most recent ones')
(trilema) asciilifeform: understand, the only reason why the thing works at all, is that this one small part of it, the bernsteinization, can be made ~entirely~ independent from the db locking idiocy
(trilema) asciilifeform: and not the bernsteining.
(trilema) asciilifeform: http://btcbase.org/log/2016-12-30#1593516 << recall, i wrote to bernstein himself.
(trilema) asciilifeform: phf: bernstein's algo operates on ~all known moduli simultaneously~
(trilema) asciilifeform: phf: nope. the only thing that happens to db as a result of bernsteinization is N queries 'do we already know this factor'
(trilema) asciilifeform: trinque: i need random-access in O(1) to them for bernsteining
(trilema) asciilifeform: oh and then, factors are found, largely the same set every time (how bernsteinization works) and each one is queried to the db
(trilema) asciilifeform: also did i mention that the entire db get shat out every time we bernstein ?
(trilema) asciilifeform: the one obvious optimization i was considering was to avoid all dupe checks on key submit and simply deduplicate prior to each bernsteining. but this has serious cost in ui consistency, no more could submitters expect to see a result that is guaranteed to make sense after they submit.
(trilema) asciilifeform: mircea_popescu: incidentally, when i wrote to bernstein, there was 0 answer.
(trilema) asciilifeform: in other non-news, https://mjos.fi << possibly interesting d00d of the dan bernstein type
(trilema) asciilifeform: nope, that's not how bernstein's algo worx
(trilema) asciilifeform: in other definitely not-news, https://archive.is/Cbzko << lulzy vintage dan bernstein.
(trilema) asciilifeform: i'll say that this is the first not-wholly-plagiarized paper on subj since bernstein's.
(trilema) asciilifeform: iirc bernstein's 'salsa' thing works on this method.
(trilema) asciilifeform: bernstein's also is, what was once called 'embarrassingly parallel'
(trilema) asciilifeform: jurov: understand, there has to be a db, because ALL mods get bernsteined against ALL mods AND the 8ball.
(trilema) asciilifeform: the remainder-tree (see bernstein's paper)
(trilema) asciilifeform: and when it is, we'll simply bernstein in 2 processed, one just with 8ball.
(trilema) asciilifeform: iirc she is bernstein's coauthor in pretty much everything, also.
(trilema) asciilifeform: eh henninger's proggy from that site (straight implementation of bernstein) works.
(trilema) asciilifeform: what the hell happened to the last time folks collected ssh pubkeys and bernsteined'em
(trilema) asciilifeform: and i think it's herr doktor bernstein
(trilema) asciilifeform: because bernsteinatronic.
(trilema) asciilifeform: in other nyooz, i am readying the bernsteinatronic phuctorer for blastoff.
(trilema) asciilifeform: see mega-p4p3r 'How to find smooth parts of integers' (Bernstein) .
(trilema) asciilifeform: ^ bernstein's replacement for sendmail.
(trilema) asciilifeform: (bernstein's method)
(trilema) asciilifeform: sbp: and ftr i'm fond of bernstein but not of ecc.
(trilema) asciilifeform: 'A number of IETF standards groups are currently in the process of applying the second-system effect to redesigning their crypto protocols. A major feature of these changes includes the dropping of traditional encryption algorithms and mechanisms like RSA, DH, ECDH/ECDSA, SHA-2, and AES, for a completely different set of mechanisms, including Curve25519 (designed by Dan Bernstein et al), EdDSA (Bernstein and colleagues), Poly
(trilema) asciilifeform: 1305 (Bernstein again) and ChaCha20 (by, you guessed it, Bernstein).'
(trilema) asciilifeform: now who wants to try bernstein!
(trilema) asciilifeform: though it won't surprise me if bernstein dies of mystery 'accident'
(trilema) asciilifeform: but somebody ought write to bernstein.
(trilema) ascii_butugychag: punkman: the one with bernstein's proposed enhancements - possibly.
(trilema) ascii_butugychag: http://log.bitcoin-assets.com/?date=04-02-2016#1396374 << sorta what bernstein tried to do. fits-in-head ciphers.
(trilema) ascii_butugychag: i like bernstein but the adoption of his blockciphers by the enemy casts darkness on them.
(trilema) ascii_field: 'Dan Bernstein speaks of interesting crypto and boring crypto. Interesting crypto is crypto that supports plenty of academic papers. Boring crypto is crypto that simply works, solidly resists attacks, [and] never needs any upgrades." Dan asks, in his typically flippant way, 'What will happen if the crypto users convince some crypto researchers to actually create boring crypto? No more
(trilema) asciilifeform: he's talking about carl bernstein
(trilema) asciilifeform: unrelated: neat find, from today's trip to library, for all crypto aficionados: 'Post-Quantum Cryptography.' Bernstein, Buchmann & Dahmen. (2009)
Download hourly DB snapshot | Get Source Code