195400+ entries in 0.131s
phf: probably better
than injecting for >1 whitespace in a row...
phf: never
the
the less, i put a hack
that fixes
the issue without using <pre>
phf: asciilifeform: browser doesn't render repeat whitespace, unless you do <pre> but
then, i believe, it doesn't automatically wrap lines
a111: Logged on 2017-04-27 22:46 phf: it is not in fact phf who posted
this nonsense but an adversary, a sybil attack is in progress!
a111: Logged on 2017-04-27 22:39 asciilifeform:
this is
trickier
than it appears, because you can't ever have a branch condition itself on a secret bit.
a111: Logged on 2017-04-27 21:26 asciilifeform: after working as exterminator long enough, all rats begin
to look quite alike
a111: Logged on 2017-04-27 19:11 mircea_popescu: pretty much all
that grows on
the fertile soil of
the red skin's plain is potatoe.
phf: it is not in fact phf who posted
this nonsense but an adversary, a sybil attack is in progress!
☟︎ phf: well,
the whole
thing can just be shift by exponent
a111: Logged on 2017-04-27 16:45 mircea_popescu:
these
together mounted on say quadcopter suddenly are rather dangerous. for one
thing, no way
to intercept its comms anymore.
mircea_popescu: (no idea why not,
they certainly
thought 8 bits were a byte for
the same reason)
mircea_popescu: apparently
there's a lot of
them. /me is looking at cliki.net
mircea_popescu: could have stopped at
the "properly specified multi implementation" part
mircea_popescu: lol is
that ffa implemented in slutty what's hjer name ?
phf: i decided
to learn ada
today
phf: unfortunately it seems
to work as expected
phf: well, it's
the correct result, nothing unexpected
phf: asciilifeform: i did a
thing
☟︎ Framedragger: asciilifeform: did, with and without numlock. and
the
thing you're
trying
to catch in code (e.keyCode) is 0 in both cases. however, e.which is correct (and is more 'canonical' here anyway). but i'm on laptop currently; but
that shouldn't make a difference
Framedragger: asciilifeform: e.which more browser-agnostic
than e.keyCode; in my case (firefox), e.keyCode is 0, expects e.charCode for $reasons. however, e.which will also work, and 'should' work everywhere except on very old IE. am not js expert
tho, so, disclaimer
Framedragger considers opening js console
to check exceptions, but goes for
tea
to not ruin evening
Framedragger: ah, i remember
the microwriter article. cool. hm yeah dun seem
to work, on first attempt
mircea_popescu: <andytoshi> how parallelizable is shor's algo? if i want
to break 100 discrete logs is
this much faster
to do in batch
than
to do
them separately?"
mircea_popescu: in other "Riastradh> Design of RNGs for crypto is on-topic,
though physical phenomona
that lead
to
them is at
the edge of
the
topic of
this channel -- not because it's irrelevant but because it lies outside
the area of expertise in mathematical cryptography one expects
to find here." : "* andytoshi (~apoelstra@unaffiliated/andytoshi) has joined ##crypto
mircea_popescu: phf Framedragger
there's actually a specific law forbidding your approach from prevailing.
mircea_popescu: apparently
the whole academic
thing isn't very high with
the contemporaneous crop.
phf: Framedragger: we're
trying
to say
the same
thing, but i don't
think it's quite
there yet
a111: Logged on 2017-04-27 21:22 Framedragger: isn't it a kind of probability of possible messages / possible states, given a message / given some system. can you really infer anything about "how it came about" (infer *any* info, in
terms of probability or w/e)?
mircea_popescu: but yes, i agree
there's a huge difference between "spit out string hunter2 half
tyhe
time" and entropy eh
mircea_popescu: basically what happens is
they hang out in
these "professional" venues being "polite" and as a result
they misunderstand something and it becomes a
thing for
them and
then
they end up in arguments with outsiders over it.
mircea_popescu: they're like mere hairs away from coming up with an ideal compressor,
these folks.
mircea_popescu: and check out
the "min-entropy" "best strategy"
thing on slide 10.
mircea_popescu: basically
this
tribe
thinks
that what shannon entropy is, is when P
takes value "hunter2" in 50% of
the cases and a random in
the remainder of cases and
therefore
this is "no good for crypto because i can guess what your password will be".
mircea_popescu: oh shit.
turns out what
the dude means is actually rather
trivial, and also a rather ridiculous
tribal misunderstanding of what shannon said etc. read
the slides 8 and 9 of
this dude's presentation.
Framedragger: isn't it a kind of probability of possible messages / possible states, given a message / given some system. can you really infer anything about "how it came about" (infer *any* info, in
terms of probability or w/e)?
☟︎ mircea_popescu: phf if it can categorize
the events it necessarily can say ~something~ about ["the past","how
they came about"], neh ?
phf: afaiu shannon's entropy being a probability is descriptive, rather
than prescriptive. so it can categorize a sequence of events, but it can't really say anything about how
those sequence of events come about. so i'm not entirely sure how it even applies
to engineering problem of event generation..