293400+ entries in 0.189s
Framedragger: how do you ensure
that it's still bob you're
talking
to, or do you not care (on some
technical or literal level)
Framedragger: right, so your idea of pfs is perfect unlinkability, so
to speak. of course
this makes sense
mircea_popescu: sure, but merely "chaning"
the key is not good enough. if it signs it...
they're
there.
Framedragger: butbut, yes of course, hence
the whole idea of otr actually changing session key constantly, for plausible deniability and all
that (and won't argue
that it's not broken)
mircea_popescu: there's one definition of pfs, and
that is, later keys being compromised does not show anything about earlier keys.
Framedragger: hm, see i put more weight on
the "can't encrypt if permanent key is compromised" part in pfs
mircea_popescu: anyway,
to understand each other : if M are messages signed by K[eys],
then if on
the basis of M2 you can prove
that indeed K1 was used for M1
then your system is not pfs, and K aren't ephemeral.
☟︎ Framedragger: right,
the example, however, only probed a general scheme, and no protocol particulars. but fair enough
Framedragger: however if you want
to be sure
that it was bob who constructed
this session
that you're now participating in - your old friend bob will have
to show
that he owns
the private key
Framedragger: all i meant was
that you e.g. have an otr fingerprint - which represents a permanent keypair. but your otr client establishes ephemeral key per each session
Framedragger: ephermeral keys are usually used for encryption; i don't
think
the functions of permanent vs. ephermeral keys are mutually exclusive at all
Framedragger: asciilifeform: won't argue as it's obvious
that it'd be fruitless. i can certainly see
the viewpoint. "what he pushes == snakeoil" is a strong accusation
tho. however, i'm not prepared
to provide defence
till i'm old, weary, and wise (or at least
till i'm old)
Framedragger: concepts not contradictory, you use one as session key,
the other
to link sessions in
terms of authorship / who participates in said sessions. unless you meant something more
technical/particular mircea_popescu
mircea_popescu: not entirely clear what
the benefit of ~signing~ said ephemeral keys is however
☟︎ Framedragger: asciilifeform: btw
the prekey idea is sound, with or without ad hominem
Framedragger: "this one weird
trick" kind of
thing, not
too smart, but possibly just smart enough.
Framedragger: in other news, am casually looking into perfect forward secrecy in *asynchronous* communications.
this *may*
turn out
to be relevant for gossipd enthusiasts and connoisseurs in
the long run,
too.
the most simple way appears
to be for nodes
to generate a bunch of "pre-keys" (halves of ephemeral key exchange), and store/cache
them somewhere (obvs signed by
their permanent identity key), a la
https://whispersystems.org/blog/asynchronous-sec Framedragger: mircea_popescu: apologies -
too much presumption on my part,
then!
thestringpuller: Framedragger: I was being hyperbolic. My point was more
that
the mass adoption crowd is
too stupid
to regulate
themselves, and as such would like
to punish all of Bitcoin.
Framedragger: but
to show
this / give examples of some differences.. eh fuck
that,
too much effort
Framedragger: the mapping between
the sets is not necessarily bijective, i.e. one-to-one
Framedragger: i would maintain
that
the socialist set and
the doesnt-deserve-bitcoin set are not necessarily mutually exclusive; at least
this cannot be known a priori. mp et al. would argue
that it can; so be it
thestringpuller: I refuse
to live in a world where my hard work is plundered by mr. millenial who was sucking on his momma's
tit until 22.
thestringpuller: They are actually one in
the same. Socialist by any other name. For fuck's sake.
thestringpuller: Framedragger:
The mass adoption crowd is much like
the Bernie Sanders crowd.
Framedragger: thestringpuller: "do people deserve bitcoin?" is
this still a non-rhetorical question? :)
Framedragger: vc: cool, and i remember you saying
this
the first
time otherwise wouldn't have done it, just wanted
to let you know
that
these particular scans won't (*completely*)
trash
the reputation of
the currently assigned ipv4 :)
thestringpuller: "Without any regulation you can't stop a bad actor before
they start.
They only need
to scam people once
to make a lot of money.
Framedragger: (gun resume and finish ipv4 open ssh port scans
tonight with vc's node (vc: it's randomized ip range scan and only 30 kpackets/s, before you ask), and
then deploy ~10 vps nodes for ssh key extraction, feeding port-22-open-list from
the former into
the latter.)
a111: Logged on 2016-06-12 18:35 mircea_popescu: asciilifeform what sense does
that make anyway ?
Framedragger: mircea_popescu: yeah afk shit is calming down, so getting back
to variations packet madness :D
vc: yes,
that would be aftables
Framedragger: with botched mac addr so shouldnt even reach internet, i
think
theyre getting dropped at some local router
vc: are you doing
tests right now?
vc: there is an outbound cap of 100Mbit/s on
the interfaces
vc: Framedragger: I have definitely seen waaaaaaay over 10k packets / s on
these interfaces so
that's not
the problem
Framedragger: note, stricly-localhost
test gives me much more
than
that. but
test with bogus mac address (so it doesn't hit
the actual internet) gives only 10k pps, which is real low. some hardware firewall doing excessive ratelimiting?
Framedragger: vc: just fyi,
there appears
to be a ~10k packets / second limit somewhere upstream, are you aware of anything of
the kind? i'm just running some self-tests (using a program which has its own
TCP stack, i.e. no use of kernel networking / sockets). same
tests produce at least 10
times as much elsewhere. cpu not
the bottleneck. just wondering what it could be
Framedragger: hehe yeah, ssh keys, welcome
to idiotland i guess
jurov: or just say you want
to donate
them
to foundation and we'll work it out
jurov: you can have
them sent
to mpex account,
too
vc: and if
the 140 slots fill up I will purchase
the other 144GB memory which is
the last possible upgrade for
the server
vc: I'm going
to have
to put like a
tutorial on
the checkout page
jurov: yes, your shares are
tied
to your gpg fingerprint, so pls make a signed request
to onetime/automatic delivery
toi certain coinbr account
vc: Framedragger:
things have been great, users have been fine save for
the billions of
them
that messed up
their SSH key
shinohai: Bad news is all
the good blow vendors are on
tor
mod6: anyway,
thanks, we'll
talk 'round month end
then
to get it figured out.
mod6: i suppose
that can be arranged.
mod6: or, if you'd rather i signed a statement saying such, i can do
that when
they get distributed.
jurov: sorry, i can't.
they are meant
to be used strictly for hookers and blow.
mod6: so just
take your coinbr fee and forward
the rest
to
the btcf addy if
that's ok?
mod6: but i'd like
to send
the proceeds
to
The Bitcoin Foundation
mod6: i haven't, yet. but i wrote an article. so i presume
they will give me a few.
thestringpuller: usually its
the week after
the statement on
trilema confirming share count
mod6: Cool jurov, when
they are distrbuted (month end?), just hit me up.
Framedragger: by which i mean, any crazy customer support / admin stories
thus far? :)
jurov: ;;later
tell BingoBoingo yes i'll gladly help anyone
to dispose of
their shares ;)
BingoBoingo: But yes. Not
the question we deserve, but
the one we need.
thestringpuller: well his hot granddaughter is out fucking and having a good
time now
that her brainwashing is gone
mircea_popescu: in other news, "whosoever withdraws from stupidity
thereby becomes its avant-garde."
mircea_popescu: prolly get a million an' a half or some shit for
them,
too.
mircea_popescu: bring a friend, sit in for
the filming of any sex scene during
the season.
mircea_popescu: so here's what i'd do if i were actually running any of
the studios :
shinohai: Takes "choking your chicken"
to a new level.
mircea_popescu: (the cameras were installed in
the first place by old woman wanting
to know wtf her hens are so fucking noisy at night)
BingoBoingo: Literally
this involves simply deleting a file called "xml-rpc.php" One keystroke for great justice.
mircea_popescu: not like i'm doing anything ;
there's
the antispam
thing
that's documented in an article ; also
the xml-rpc hardening idem documented.
that's about it. haven't
touched any of it in years.
BingoBoingo: mircea_popescu: But at what cognitive cost? Certainly a greater one
than simply removing
the entirety of
the xml-rpc and
turds
therein.
mircea_popescu: BingoBoingo
trilema works ok with
trackbacks, but i guess it's one of
the few.