416400+ entries in 0.281s
shinohai: I hope
that somewhere in
the AM data, it is revealed Josh Duggar snorted coke off of a
tranny's cock.
punkman: shadowsocks makes your
tunnels look like
traffic
that's not blocked
punkman: mircea_popescu: is
that
trolling ? << chinese problems
mircea_popescu: "A
totalitarian society which succeeded in perpetuating itself would probably set up a schizophrenic system of
thought, in which
the laws of common sense held good in everyday life and in certain exact sciences, but could be disregarded by
the politician,
the historian, and
the sociologist."
mircea_popescu: you know,
they speak casually of how
the bitches
they know're gonna introduce
the new ones. except... all
this is a work of feverish imagination.
mircea_popescu: "The reappearance of HELEN'S BABIES, in its day one of
the most popular books in
the world–within
the British Empire alone it was pirated by
twenty different publishing firms,
the author receiving a
total profit of £40 from a sale of some hundreds of
thousands or millions of copies–will ring a bell in any literate person over
thirty-five."
mircea_popescu: cuz i might want
to be able
to distinguish orwell from orlov
mircea_popescu: because it's where it's headed anyway, and why
the fuck wait.
mircea_popescu is
tempted
to deed a list of key,values of names of various people and start referring
to people as "Primitive Man AFMrikO2o3Ka2pcpZ" instead of "Obama" or "Jean Jacques Rousseau" or whatever.
phf: g
the wrong RSA key." message format spec explains "First 2 bytes of
the Message Digest inside
the RSA-encrypted integer,
to help us figure out if we used
the right RSA key
to check
the signature."
phf: so
to continue
this archaeological dig, GPG 2.6 clarifies
the usage of 2 octets. reads
the header, reads
the rsa ciphertext, decrypts rsa. rsa contains a digest of some fields from header and
the body of message. so first
thing he does next is check
the first 2-octets of digest againts
the 2-octets in header. if
the
two don't match program bails with "Error: RSA-decrypted block is corrupted.
This may be caused either by corrupted data or by usin
mod6: asciilifeform: ... if you regenerate
the 'orchestra' patches with new vdiff, i will sign'em. otherwise might have
to wait a while << hey, no problem. I'll start in on
this sometime
tomorrow probably.
thestringpuller: dunno if asciilifeform is
talking about human female, or some random puppy he has.
phf: i actually forgot
that you were working on gnupg source for cardano
phf: oh? i missed
that, i
think
that's before i started
to read closely
phf: surprisingly hard
to get a copy
phf: yes,
trying
to see what he was doing with signatures
there
phf: incidentally after bitcoind gnupg source is pleasure
to read
phf: (so it's not compare.
they run a digest on payload while reading (or separately if detached),
then
they run digest on own reconstruction of header,
then
they finalize
the digest and do sig verification of provided signature against
the digest
that
they calculated)
mircea_popescu: ;;google "pentru ca esti o vita spastica si
te uraste soarta"
mircea_popescu: and if youy for some incomprehensible reason MUST use a hybrid scheme, use
the following : 1. generate random 4096 hash ; 2. cut your message up into N chunks of size up
to 2048 ; 3. xor
the chunks with 1; encrypt each chunk via rsa
mircea_popescu: yes. but none of
these actually requires a hybrid scheme
mircea_popescu: (and i maintain
that in all context hybrid encryption is retarde dwithout exception. either fuck or get off
the woman.)
mircea_popescu: i wasn't at any point contemplating "rsa encryption = rsa run once over
the message herp"
mircea_popescu: make
the damned
thing match
throughout. bapg = 4096 rsa, 4096 hash, etc.
mircea_popescu: actually. is
there a 4096 bit hash function of any merit ?
phf: logic looks like "read a bunch of
things", "hash data", "put relevant bits into an in memory copy of
the header", "hash in memory copy", "compare"
phf: i'm mostly curious if my guess about
the purpose of
the field was correct, but no insight from gpg
phf: asciilifeform: i'm failing
to see
the clusterfuck part
mircea_popescu: phf yeah it's not actually as bad as it seemed on
the basis of
the snippet. still crud.
phf: so for
the curious it doesn't look like 1.4.19 uses
the
two-byte hash value anyway. it reads it from disk, it populates it during sign and it can write it
to disk, but no actual logic done using it
assbot: Logged on 22-08-2015 01:11:30; mircea_popescu: never, inb
the entire history of reichs, was
this
the case. not once.
mircea_popescu: it "had" it. suppose
there's madonna burried in your back yard
mircea_popescu: i dunno dood. but once i figure it out i'll make
the computer do what you mean.
mircea_popescu: never, inb
the entire history of reichs, was
this
the case. not once.
☟︎ mircea_popescu: your notion
that hitler's ahead of
the curve is spurious.