log☇︎
2000+ entries in 0.444s
assbot: Request successful, get your OTP: http://w.b-a.link/otp/905653adafb2fba6
assbot: Request successful, get your OTP: http://w.b-a.link/otp/0d66340b173959c4
assbot: Request successful, get your OTP: http://w.b-a.link/otp/ad75876b5345660f
assbot: Request successful, get your OTP: http://w.b-a.link/otp/e6bf926d05455a98
assbot: Request successful, get your OTP: http://w.b-a.link/otp/85ef912e10a7ae08
assbot: Request successful, get your OTP: http://w.b-a.link/otp/0aa13d1a7b98679a
asciilifeform: but this is still idiocy, because otp is the ~definitional~ strong cipher. it is provably ludicrous to ask for a cipher in any sense whatsoever stronger than otp.
asciilifeform: mircea_popescu: the 'otp failed' part is idiocy
asciilifeform: (lulzier when we get ~different~ arithmetic!11 as in the otp mega-thread)
asciilifeform: anyway mircea_popescu , m. blaze's 'turtle' is, afaik, the only ~claimed~ provably-hard block cipher (non-otp) that i was able to track down.
punkman: where is this znc otp plugin?
jurov: kek. but it's true somebody here sponsored otp for znc plugin
phf: somebody here i believe sponsored otp for znc plugin
assbot: Request successful, get your OTP: http://w.b-a.link/otp/d854434c72d2880f
ascii_butugychag: whereas your otp in the field is the one you have.
ascii_butugychag: running out of otp is rather like running out of ammo,
mircea_popescu: (meanwhile i asked for the why and wherefore of the "debiasing" of the plaintext. "no you idiot, it's so you burn less otp. what the fuck's wrong with you!")
mircea_popescu: the recent debias-otp-plaintext thing being a splendid if amusing example in this exact line.
assbot: Logged on 08-02-2016 02:23:29; phf: for extra lulz paper includes proof of otp perfect secrecy taken from a coursera cryptography course
phf: for extra lulz paper includes proof of otp perfect secrecy taken from a coursera cryptography course ☟︎
phf: ;;later tell maqp on a cursory inspection i couldn't figure out how the protocol decides between otp and cev, how those are identified on the wire, etc. is that up to the user?
phf: ;;later tell maqp if i were you i'd split tfc.pdf into separate papers. HWRNG, data diode communication, tfc otp, tfc cev and "rationale", the last one to include all the superfluous NSA shoutouts
punkman: "why use information theoretically secure ciphers" << not really plural there, there is only otp
maqp: basically it's like OTP but with forward secret cascading encryption
maqp: It's also a lot easier with NaCl than with OTP/CEV (there's a separate command for adding PSKs)
punkman: maqp, is that a carter-wegman MAC in your otp version?
assbot: Request successful, get your OTP: http://w.b-a.link/otp/32540d5779b7cecb
maqp: thanks. I wanted to recommend you guys take a look at the TFC-NaCl that's fresh out of oven and has better design compared to OTP/CEV versions
mircea_popescu: you're the guy with the open source otp / airgapped thing are you ?
punkman: the otp implementation is possibly decent though
assbot: Logged on 07-02-2016 09:30:48; punkman: https://github.com/maqp/tfc-otp
assbot: GitHub - maqp/tfc-otp: Tinfoil Chat (OTP) ... ( http://bit.ly/1S8th8K )
punkman: https://github.com/maqp/tfc-otp ☟︎
mircea_popescu: that's before the otp.
punkman: /me goes look for otp-compatible MAC scheme
asciilifeform: funnily enough, last year there was some derp who shat into mircea_popescu's comment section with 'otp doesn't work because rng might burp out N zeros and then what'
punkman: you can't get the bitmap penguin after otp
asciilifeform: i realize that it cries against every mathematical intuition, but with correctly functioning otp, it makes not a whit of difference what your plaintext is.
mircea_popescu: looky : otp bias, message bias, same fucking story.
asciilifeform: or you could just skip the otp alltogether
mircea_popescu: asciilifeform otp bias cheifly doesn't matter here, as the same otp is delibverately used for both messages.
mircea_popescu: thus nth position in otp xor nth position in ciphered a must yield n and in ciphered b must yield 65535-n
punkman: what's the point of public otp
mircea_popescu: http://trilema.com/otp.php for the bored
punkman: mp must guess how many of the 100 ciphertexts are made from the string ""mircea_popescu: long, deeply biased plaintexts are dangerous for otp.""
punkman: other variant: ascii makes 100 otps, makes 100 plaintexts, X of which are the string "mircea_popescu: long, deeply biased plaintexts are dangerous for otp.", then passes 100 ciphertexts to mp. mp must guess X withing some range.
mircea_popescu: long, deeply biased plaintexts are dangerous for otp.
mircea_popescu: asciilifeform : Let message A consist of individual bytes counting down from FFFFFFFF ; let message B consist of individual bytes counting up from 00000000. Let the enemy xor one of these two against a random, unbiased OTP of the same length and supply the enciphered result. Take that result, and count the instances where byte n is larger than byte n+1. Take that result, and count the instances where byte n is larger t
asciilifeform: if there is one otp key, and it gets used two or more times, with mircea_popescu controlling the input and knowing anything whatsoever about the output, he learns the key trivially.
asciilifeform: one otp bit, one xor.
punkman: you pick one of 1000 plaintexts, generate one otp
asciilifeform: this is not how otp is used.
asciilifeform: otp doesn't get reused.
punkman: mp makes 2 plaintexts, ascii generates 1000 otps, for each otp: picks one of the 2 plaintexts and xors with otp. mp must guess guess correctly 501?, 600? more?
mircea_popescu: and if the plaintext is long enough, this is equivalent to a requirement of minimal bias in the otp pad.
mircea_popescu: how biased the otp needs to be is part of the crc spec, for instance "every 8th bit may be a 1" etc.
mircea_popescu: let me put it this way : stuff like CRC, or ECC etc, exists fundamentally out of "we guarantee you can recover the plaintext after it has been otp'd with a pad which is AT LEAST this biased"
mircea_popescu: if you're making 1 mb of 01111110 and 1mb of 10000001 and then otp them against a random pad
asciilifeform: mircea_popescu: think about it, with otp, there is no reason for you to actually intercept the ciphertext
mircea_popescu: they aren't all equally probable if i can rely on your otp being random.
asciilifeform: an actual otp conveys no information whatsoever via the ciphertext.
mircea_popescu: if i know you will be encrypting one of shakespeare's plays, otp won't save you.
mircea_popescu: there is another way to die using otp, and that way is to use a lengthy biased message the enemy knows most of.
asciilifeform: ergo the linked thread, where i posit that an ideal otp is actually a physical object which brings the bits somehow into existence one at a time
mircea_popescu: asciilifeform technically speaking, the s-box cipher crapolade is an ellaborate exercise in reusingselect parts of otp
punkman: you can't do frequency analysis on otp
asciilifeform: there are several possible ways to die when otp
mircea_popescu: punkman compressing the plaintext, not the otp.
punkman: isn't compressing your otp akin to whitening?
assbot: Logged on 03-02-2016 01:53:21; asciilifeform: actually for many years i have thought about the ideal electric otp.
asciilifeform: http://log.bitcoin-assets.com/?date=03-02-2016#1394833 << obligatory otp megathread ☝︎
punkman: biased otp???
mircea_popescu: to be plainer : otp works better with biased pad of unknown bias than with unbiased pad of known lack of bias.
punkman: this sounds like you want to do frequency analysis on otp, but perhaps I'm just thick
mircea_popescu: the correct way to apply otp to something like human readable text is to weigh it.
BingoBoingo: <punkman> gotta have something to remember how much of the otp has been used << burn the used pages of your cipherbook
punkman: gotta have something to remember how much of the otp has been used
mircea_popescu: but yes, otp on top of prng is asking for trouble.
BingoBoingo: otp implementation is as good as your random
punkman: is there a decent otp implementation?
mircea_popescu: ben_vulpes otp key as long as message.
ben_vulpes: what does c-s buy one over the otp in that case?
mircea_popescu: in the EP? general scheme of true cryptography, otp occupies a peculiar spot, equivalent to rsa's use of multiplication, where otp uses "multiplication modulo 1" or "multiplication in the binary group" for a º function
ben_vulpes: and the need to share the key does not impose the same operational considerations as otp?
asciilifeform: ben_vulpes: otp is a particular very specific thing
ben_vulpes: cramer shoup + shared key does not reduce to...otp?
asciilifeform: iirc mircea_popescu wanted a non-otp that demonstrably doesn't suck
mircea_popescu: i already have otp.
asciilifeform: mircea_popescu: neato! turns out otp has its very own lilienfeld
asciilifeform: 'peak crypto!1111' was otp, vernam, ww1.
asciilifeform: otp is ww1 state of the art.
mircea_popescu: asciilifeform> otp. << if the key is 64kb, technically otp would work fine for message up to 64kb.
asciilifeform: in any ciphersystem ~other~ than otp, the ciphertext carries, theoretically, ~some~ information re: the key.
asciilifeform: (the otp proof is kindergarten-level - the ciphertext tells you nothing at all - as in 0 bits - about the key or the plaintext)
asciilifeform: otp.
assbot: Request successful, get your OTP: http://w.b-a.link/otp/813ea42d8e6d9d8a
asciilifeform: the ideal otp machine:
asciilifeform: actually for many years i have thought about the ideal electric otp. ☟︎☟︎
mircea_popescu: dun need computers for otp.
mircea_popescu: anyway - it's not that i don't like paper otp, or otp generally. it's that if that's the best you can do, you should have been a clockmaker