log☇︎
5134 entries in 0.896s
decimation: https://mvideos.stanford.edu/graduate#/SeminarDetail/Spring/2015/EE/380/9469 < this Nadia Heniger chick claims that rsa keys can be poorly generated by bad rngs (around 50:00 or so)
decimation: it's in the comments of that article that replaced yours on hacker news https://blog.hboeck.de/archives/872-About-the-supposed-factoring-of-a-4096-bit-RSA-key.html
mircea_popescu: http://trilema.com/2015/full-disclosure-4096-rsa-key-in-the-strongset-factored/#comment-114229 << lol check it out, ~they~ are going to ~lend me~ credence. i've been visited by backwards world wtf is with today.
justJanne: Nah, I was just reading on the RSA factorization article where someone claimed this channel would still DDoS everyone who joins. Sadly I can’t see any of that (was hoping to see if it actually is true xD)
decimation: https://blog.hboeck.de/archives/872-About-the-supposed-factoring-of-a-4096-bit-RSA-key.html#comments < heh the hboeck guy admits he doesn't really know of which he speaks
mircea_popescu: asciilifeform http://trilema.com/2015/full-disclosure-4096-rsa-key-in-the-strongset-factored/#comment-114226 "Because Phuctor shows me an Internal Server Error when I try to test a key."
joshbuddy: hiya, just hanging out after reading the hilarious rsa key blog post by .. umm you
alphonse23_: I came because of the 4096-rsa article
assbot: Logged on 17-05-2015 16:10:13; mircea_popescu: achtung BingoBoingo cazalla : phuctor broke a rsa.
mircea_popescu: asciilifeform http://blog.hboeck.de/archives/872-No,-nobody-has-factored-a-4096-bit-RSA-key.html if you care.
mircea_popescu: actually, a fully-explicit "how to extract your rsa moduli as numbers, and how to test things" write-up may be a good use of someone's time.
JPT: I decided to read along a little since it got to me that some rsa keys were broken ;)
davout: asciilifeform: it doesn't parse plain rsa keys?
asciilifeform: this is a point, if you convert your ssl or whatever rsa key to pgp format,
mircea_popescu: http://trilema.com/2015/full-disclosure-4096-rsa-key-in-the-strongset-factored/#comment-114212 << apparently the ddos guy is still lurking.
mircea_popescu: http://trilema.com/2015/full-disclosure-4096-rsa-key-in-the-strongset-factored/#comment-114210 << the problems of people.
davout: http://www.reddit.com/r/Bitcoin/comments/36a1m2/full_disclosure_4096_rsa_key_in_the_strongset/
jurov: http://www.reddit.com/r/technology/comments/36a3mu/today_a_4096_bit_rsa_key_was_factored_more_coming/
mats: http://mercemolist.net/2014/10/07/colofon-navaja-negra-la-clave-rsa-que-al-final-se-rompio/
deedbot-: [Trilema] Full disclosure : 4096 RSA key in the strongset factored. - http://trilema.com/2015/full-disclosure-4096-rsa-key-in-the-strongset-factored/
mircea_popescu: http://trilema.com/2015/full-disclosure-4096-rsa-key-in-the-strongset-factored/
asciilifeform: jurov: most peculiar. it is a valid rsa key.
assbot: Successfully updated the rating for asciilifeform from 4 to 5 with note: 4 First man to factor a 4096 RSA key in the history of computing.
mircea_popescu: !rate asciilifeform 5 4 First man to factor a 4096 RSA key in the history of computing.
mircea_popescu: ;;rate asciilifeform 4 First man to factor a 4096 RSA key in the history of computing.
asciilifeform: mxtm: some 2 schmucks (of which i've identified 1, but the other will follow in a week or so) had common factor in their rsa private keys.
mircea_popescu: achtung BingoBoingo cazalla : phuctor broke a rsa. ☟︎
ascii_field: this plus rsa is good enough to sig-only-revoke.
mircea_popescu: there are some things that don't need fixing. what needs fixing re gpg is to use rsa throughout, rather than the current lulzatron.
ascii_field: one cannot claim the same degree of 'hardness' for this as for the actual rsa, correct
asciilifeform: link rsa'd
copypaste: gpg: encrypted with RSA key, ID 16B8E32E
assbot: Logged on 12-05-2015 22:03:27; mircea_popescu: so ascii_field from the above, am i correct in deducing that there have been at most 47728 - 31262-4584 = 11882 (out of 47728, or ~1/4) keys without any RSA subkeys in them ?
mircea_popescu: so ascii_field from the above, am i correct in deducing that there have been at most 47728 - 31262-4584 = 11882 (out of 47728, or ~1/4) keys without any RSA subkeys in them ? ☟︎
ascii_field: http://log.bitcoin-assets.com/?date=07-05-2015#1124071 << the 258 figure does not include the non-rsa submissions ☝︎
asciilifeform: because the paucity of rsa moduli susprised even me
mircea_popescu: which, non rsa ?
asciilifeform: ;;later tell mircea_popescu there are quite a few non-rsa keys. what were folks thinking, l0l
BingoBoingo: For shame, 768 bit RSA key from 2002 http://pgp.mit.edu/pks/lookup?op=get&search=0xA8F0C01E3330EE4B
mircea_popescu: "rsa. use for text. rsa/aes. use for anything that's not text."
mircea_popescu: anyway, optimal thing would be user specified, and full rsa option.
jurov: and? so the optimal thing would be to parse the email and encrypt some parts by rsa and some by aes or what?
jurov: in this case it's not moore's law but bandwidth. you can't have megabytes/second with rsa, or if you do, battery dies
mircea_popescu: funny how moore law is always there to justify doing something stupid, never to justify doing things like "rsa all the way - we can afford it now"
jurov: iirc pgp/rsa is computationally expensive, so it's only used to establish rsa keys
pete_dushenski: i dun follow that part at all. the pgp/rsa makes sense but where aes comes in is anyone's guess.
mircea_popescu: do they do the "rsa encryption of an aes symmetric key" thing ?
pete_dushenski: "For maximum security, our BlackBerry PGP encryption service utilizes AES 256 bit PGP encryption with a 4096 bit RSA key format." << also uses session keys, but these are presumably generated on the phone, so...
mircea_popescu is really curious what gcd over rsa keys actually looks like in terms of resource consumption.
assbot: Generate RSA keys in browser. Store public and private ke... - justpaste.it ... ( http://bit.ly/1HLv90n )
decimation: it would be handy to have a 'diceware' version for rsa keys
assbot: Logged on 04-04-2015 03:39:35; decimation: but I would also like to pack enough bits on an 8.5x11 sheet of paper to convey a 4096 bit rsa privkey
decimation: but I would also like to pack enough bits on an 8.5x11 sheet of paper to convey a 4096 bit rsa privkey
mircea_popescu: how does it even work, you hash the "public key" and then multiply it with something ? sorta halfass rsa ?
decimation: asciilifeform: it strikes me that a handy product would be a little ada microcontroller with an implementation of rsa
assbot: Logged on 13-10-2014 19:14:19; asciilifeform: ulbricht << unless he's learned to grind rsa in his head, how is this to happen? his jailers will otherwise have the key at their leisure.
mircea_popescu: asciilifeform rsa per se is not even really involved in all this o.O
asciilifeform: mircea_popescu: i think he was speaking of rsa per se
nubbins`: ascii_field so what's the path from rsa key -> seed node
ascii_field: you can't ddos an rsa key
ascii_field: ultimately the only proper box is one where the only magical constant is an rsa key, and same key is also prominently burned into the chassis
ascii_field: 'Provides a working implementation of a backdoor embedded into the RSA modulus of a Certification Authority public-key certificate and the code for a minimalistic client and server communicating over a TLS channel:'
mike_c: "RSA received $10 million in a deal that set the NSA formula as the preferred, or default, method for number generation in the BSafe software"
mike_c: mats: http://www.reuters.com/article/2013/12/20/us-usa-security-rsa-idUSBRE9BJ1C220131220
mike_c: what? i thought it was common knowledge rsa sold their soul.
mats: and re: BSafe, i'm not convinced RSA could be sold on destroying their reputation for packing peanuts.
mats: RSA upheld their end by backdooring the BSafe product
asciilifeform: oh and mats, forgot to ask: why have you forgotten about the massive (iirc, 10-20 mil) usg bribe to rsa?
assbot: 6 results for 'rsa pill' : http://s.b-a.link/?q=rsa+pill
asciilifeform: !s rsa pill
asciilifeform: http://www.cs.gmu.edu/~zduric/cs803/Simmons.pdf << unrelated but interesting from historical standpoint re: usg and practical use of rsa
asciilifeform: if real-time satellite is costly and unreliable, a 'poor man's' version will contain two rsa keys. one will sign every frame and globe coordinate tuple -until package is opened-, the other - after
asciilifeform: regarding your intentions to compute rsa funktion
asciilifeform: to borrow mircea_popescu's terminology, buying a dedicated, e.g., rsa silicon, communicates information to enemy which he has no business learning
asciilifeform: that, say, cranks 8192-bit rsa.
mircea_popescu: more like cracking rsa : for the techology they had at the time, absolute security.
BingoBoingo: IN the very worst case and RSA fails at least private languages aren't too bad compared to DES, provided the grammar and semantics are unique
BingoBoingo: 70's and 90's RSA is the strong stuff
BingoBoingo: danielpbarron: Even with GPG not everything is RSA. Some people are stupid and use version 2-ish though it lacks backwards compatibility or they use 80's fail crypto which breaks easily.
mircea_popescu: danielpbarron in the rsa scheme
adlai: although not all gpg keys are rsa
BingoBoingo: danielpbarron: Could even the good doktor mark such a large keyspace as RSA as his numebers?
asciilifeform: gpg: Signature made Wed Feb 25 20:47:07 2015 EST using RSA key ID 01ABFFC7
BingoBoingo: It is amusing. Still when it comes time to replace GPG it will look more like https://github.com/tedu/reop than any Moxenslit iPerson thing (except... with RSA support and done in ADA/Common Lisp)
BingoBoingo: gmaxwell: I'm just inclined towards giant RSA keys and presenting a hard problem. It's just I'd like the option. That and ECC is already keeping the money safe...
mircea_popescu: asciilifeform if nubbins` weren't such a noob, he'd have laser etched a mechanical rsa by now
decimation: asciilifeform: it would be amusing if someone were to make an electro-mechanical RSA machine
asciilifeform: i'll elaborate re: above: 'directorate xxx was unable to break rsa despite five quadrillion bezzlars blown in five years' is only mildly interesting because directorate yyy might have a functioning voodoo doll from whose arsehole keys drip on demand. etc.
asciilifeform: the 15 lines also contain rsa?
asciilifeform: to log readers wondering about the purpose of the disposable rsa key: forward secrecy
asciilifeform: decimation: assume 4096-bit rsa. and udp mtu (576.)
asciilifeform: in response to a valid, non-replayed 'hello' (from known pubkey, naturally), machine sends back a single-use rsa pubkey (generated for the occasion) to the requester
decimation: also, today I was thinking about your rsa-signed-udp-packets thingy
phillipsjk: DSA is assumed to be weak because 768bit RSA primes have been publicly factored.
phillipsjk now has 2048 bit RSA
decimation: decimation: i wonder if the device keys have even been 'phuctored' ever. << "CMs MUST store and maintain the CM Device Certificate RSA private/public key pairs. The CM MUST store the CM Device Certificate private keys in a manner that deters unauthorized disclosure and modification. Also, CMs SHOULD prevent debugger tools from reading the CM Device Certificate private key in production devices by restricting or blocking physical
decimation: ah "The CM MUST have two factory installed CM Device Certificates (and their associated private keys). The CM MUST have a CM Device Certificate installed that is issued from the new PKI. The CM MUST have a CM Device Certificate installed that is issued from the legacy PKI. The CM MUST have the same RSA public key in the CM Device Certificate as the RSA public key in the BPKM Attributes depending upon which CM Device Certificate is
decimation: apparently docsis 3.1 uses 2048-bit rsa for the device, 4096 for the root ca
asciilifeform: if uses civilized rsa at all, it is with ludicrously shrunken key lengths
jurov: <davout> it occurs to me it would be really nice to have a way to easily dice-generate an rsa keypair << needs just some 1585 dice throws to get two 2048bit seeds (from them prime search is sequential)