275100+ entries in 0.175s
mircea_popescu: yes, everybody did.
that's
the problem. when derp mcherperson wants commit access
to vomit her versiuon of invidious gossip
therein as "comments" and "sensitivity", PEOPLE FORCE IT
mircea_popescu: she's well equipped for
the
things she does, invidious gossip, raising children and burning
the cabbage.
there's really no call for cabbage os and burned "literature".
mircea_popescu: nobody asked you
to force
the poor
thing
to art ; much like nobody asked you
to force
the minimal brain of
the huswife into producing political discourse and directing ships on
the sea.
mircea_popescu: see, but
that's not it. same FUNCTION of organ.
that's
the issue.
the anal ring is made
to keep farts in, and ~arguably~
to pleasure a beloved penis now and again.
this is very much
the same function.
mircea_popescu: one
time one person inquired at party "but what's with all
the monster cocks"
mircea_popescu: well no, feminist you see. had no use for
those e=mc2 shapes
that predilect
the edge of light over roundnesses much more important ...
to "us".
mircea_popescu: as per aristotle,
there is no such
thing as "no design"
a111: Logged on 2016-08-03 13:57 mircea_popescu can't
think of any valid excuse anyone involved in os design could possibly have not
to already be in wot.
mircea_popescu occasionally lulz at girlies earnestly clucking around
to speed up key production.
mircea_popescu can't
think of any valid excuse anyone involved in os design could possibly have not
to already be in wot.
☟︎ mircea_popescu: i prefer
to discern between
the
two by
their reaction
to events.
the fact
that
they're not all here means
they're all
there.
mircea_popescu: maybe (there is a valid argument here,
that at
the
time
they were making
this shit it was for
toys not btc nodes). or maybe
they're just usefully idiotarian.
mircea_popescu: (incidentally
those "accidental"-deliberate fucktards : fixed whitening scheme ? really ? what
the FUCK! if
there's anything in
the entire fucking os
the user must be able
to customize, it's NOT
the fucking glyph
table / window size for
the love of christmas fucks!)
mircea_popescu: it just doesn't actually speak
to
the foregoing. i suppose
the correct rsa implementation comes with a kernel patch.
mircea_popescu: the concept of
test is, definitionally, "on user end".
mircea_popescu: all
tests must always and forever be in
the hand of hte user.
mircea_popescu: but i do dispute
that for
this reason it
then follows
there also can't be put any.
mircea_popescu: anywya, i don't dispute
that "accidentally"-deliberately nobody put any effort into rng quality assurance ; key quality assurance ; etc.
mircea_popescu: passing it better
than
the real
thing is failing, innit.
mircea_popescu: (also by definition whitening doesn't
trivially pass
the
tests, just superficially seems
to.)
mircea_popescu: otherwise you're in
the position of
the camper who doesn't pack a burner because,
technically speaking, he doesn't know
there will be oxygen where he goes, not having been
there before.
mircea_popescu: asciilifeform yes, but
this is not a reason
to not have
the
tool.
a111: Logged on 2016-08-03 06:15 mircea_popescu: and speaking of gpg deplorable state asciilifeform can you
think of any possible reason
the damned
thing doesn't come a) bundled with ent and b) with ready implemented
tests of local entropy while c) key generation is a subset of entropy
testing in all cases ?
shinohai: Well after getting his fudge packed for so long, I guess it finally made it's way
to his brain.
mircea_popescu: and speaking of gpg deplorable state asciilifeform can you
think of any possible reason
the damned
thing doesn't come a) bundled with ent and b) with ready implemented
tests of local entropy while c) key generation is a subset of entropy
testing in all cases ?
☟︎ BingoBoingo: "The results mean
that, on Nov. 8,
the state's gubernatorial race will be between Koster, a former Republican
turned Democrat, and Greitens, a former Democrat
turned Republican."
mod6: <+mircea_popescu> more practical would be
to force
the code
to make a pile of say 64 byte keys and sieve
them. <<
this would be interesting.
mircea_popescu: mod6 ok but none of
this resulted in "current gpg shitfest is acceptable"
a111: Logged on 2015-05-20 15:03 mircea_popescu:
http://log.bitcoin-assets.com/?date=20-05-2015#1139680 << speaking of
this, am I
the only one nonplussed by all
this "we use <<best practices>> fixed exponent" bs ? it's an unavoidalbe magic number , okay, but it's
tyhe sort
that should eminently be a knob for
the user. a proper gpg would have e user-settable at
the key generation phase (with 65536+1 as a default, sure)
mircea_popescu: yes,
that's right : landed at chatham, burned
the ships, FAILED
TO BURN
THE FUCKING DOCKYARDS.
a111: Logged on 2016-08-03 02:58 mircea_popescu: incidentally,
the curious amateur historian may be well served by a review of
the 2nd anglo-dutch war. some underlined parts :
the anglos were deeply overextended politically -
their crummy country consisting of one single
town and a bunch of retarded peasants ; while
the republic had many more merchants, with lots more money. in practice
this meant
that
the english could pay for one ship where
the dutch could afford seven ;
mircea_popescu: actually
to formalize
that : a 4096 bits key means a p
that is 257
to 259 bytes long ; and a q
that is 258
to 260 bytes long. end of fucking story.
☟︎ mod6: <+asciilifeform> could've sworn
this was in
the logz << i recall a few discussions, ya.
a111: Logged on 2015-05-24 14:45 Apocalyptic: "there's no guarantee p and q have
the same bitsize is
there ?" // I
think
there is, a couple of lines above it generates
them both with nbits/2 bits, so I would say yes, unless
there is a bug in "generate_secret_prime", because
this function specifically sets
the
two high bits
to 1
mircea_popescu: /* make sure
that nbits is even so
that we generate p, q of equal size */ << what ?
mircea_popescu: more practical would be
to force
the code
to make a pile of say 64 byte keys and sieve
them.
mircea_popescu: yes but no. for one
thign you don't do lattice search iirc ? for
the other,
the keys we're working on are
too large.
mircea_popescu: one obviously hopes
that gpg did. but
then again
the brutal
teacher
that is experience shows it's wiser
to suspect everyone merely hoped someone else did it.
mircea_popescu: asciilifeform btw re
the fermat discussion, i wonder if anyone ever did a proper review of rsa code for lattice and fermat-closeness weakness in p,q generation.
nosuchlabswww: Not
trying
to spam. Check out qntra dot net. Fourth article down is about rsa and brought me here. Good read and new article.
fabio__: ok I
think I understand your position a bit better now
thanks
mircea_popescu: well
the ecc justification is "shorter keys", other
than a bunch of "new! better!" crapola.
mircea_popescu: always and everywhere in engineering,
this is
the case.
fabio__: so alternatives
that are more complex are regarded as lesser
fabio__: your said "fabio__ rsa has
the advantage
that it's
the simpler solution."
mircea_popescu: but "the new better replacement for copper pipe" is not worth contemplating
today like it wasn't worth contemplating in 516 ad.
mircea_popescu: you mean, "alternatives for rsa are not worth contemplating" ? sure,
they are, much like anything's worth contemplating. it's educative if nothing else.
fabio__: so in your mind, new comers are not worth considering RSA unless
they are equivalent in complexity or simpler. and until RSA is broken no need
to migrate? <-- is
this fair?
mircea_popescu: if someone breaks rsa you have all sorts of other problems
to contend with.
mircea_popescu: fabio__ rsa is not breakable in
the "shit someone broke openssl" sense you seem
to be
thinking of.
fabio__: so in your mind, new comers are not worth considering RSA unless
they are equivalent in complexity or simpler. and until RSA is broken no need
to migrate?
mircea_popescu: last
the
topic was
took up it came out
that no, it's absolutely not.
mircea_popescu: not
that we're even sure "complexity class" is meaningfully defined for
this purpose
fabio__: you did, but I'm fighting
through
the snark
to ask for more details ;)