110000+ entries in 0.064s
ckang: it increases
the surface area
ckang: i just
thought it was interesting for
them
to code protection in for such
things
ckang: but
that could be because
theres not large adoption of it
ckang: there hasnt really been anything major in
terms of flaws I could find.
ckang: and not have super
terrible performance
ckang: its much lighter such
that it can even be run on ARM routers
ckang: yea but its nice
to see in a working product
that i can use now
ckang: its faster
than watch's 1s updates
ckang: showing how quickly it will close one
tunnel and open up another
mircea_popescu: the only important question in computing is what i end up using anyway.
to
that standard, what difference could it possibly make.
ckang: so assuming
the applications
timeout is set appropriately, its like what mosh is
to ssh (wireguard
to openvpn)
ckang: which
takes a bit of
time on openvpn
ckang: you can hop between various
towers or APs seamlessly without re-authenticating
mircea_popescu: asciilifeform you mean gossipd ? how can it possibly displace something
that doesn't even exist!
ckang: for mobile applications
this stuff makes a lot more sense
ckang: im hoping he finds
the resources
to get an iOS client done
ckang: it is 'better' in same sense as ethereum 'better
than' paypal <- so its slightly better
than
the worst? ;)
mircea_popescu: there's a process we go
through here, first
the engineers
throw a fit,
then i pick up
the pieces.
a111: Logged on 2018-04-11 04:19 mircea_popescu: i guess. on a superficial look it's certainly better
than whatever
tls bs.
ckang: im not sure how
to interpret what you are saying without knowing your opinions on
those
topics
mircea_popescu: asciilifeform i know, but prepare a list of q's for when/if
the guy shows up.
ckang: can you summarize
that down
to: 'it sucks' 'indifferent' 'has a future' ;)
a111: Logged on 2018-04-11 04:11 mircea_popescu: ckang you can read up on all
the crypto functions, be
they
trapdoors or whatever, if you are interested. it's not illegible arcana.
a111: Logged on 2018-04-11 04:12 ckang:
this is what got me interested, can push a
ton more data and with less latency on
the same hardware vs. openvpn
ckang: sure
thing yea, didn't
think
to check if
they had a channel before
mircea_popescu: ckang feel free
to idle
there, smooth
things over if need be.
ckang: mullvad does offer it and its pretty good
through
them, I lose about 2Mbit off my
top end and and only gain 2ms when pinging 'google.com' (hard
to do
this
test since geography and routing)
mircea_popescu: i guess. on a superficial look it's certainly better
than whatever
tls bs.
☟︎ ckang: the project has a lot of potential, its just not well known so providers arent selling it
to
the general masses
mircea_popescu: ckang you know
the ancient story of how openbsd got saved from death ? it was eerily similar.
ckang: yea,
thats awesome, hopefully he
takes you up on
that
mircea_popescu: "<mircea_popescu> (on #wireguard) zx2c4 (the owner, j. donenfeld) : if you're willing
to set
two hours apart on any day of your choosing
to answer wireguard questions on #trilema, i'm willing
to donate 1 btc
to your project. let me know, i'm usually on freenode (this nick).
thanks & gl." << asciilifeform spyked whoever else might care.
☟︎ ckang: theres mesh capability but ive not delved into
that much yet
ckang: this is what got me interested, can push a
ton more data and with less latency on
the same hardware vs. openvpn
☟︎ mircea_popescu: ckang you can read up on all
the crypto functions, be
they
trapdoors or whatever, if you are interested. it's not illegible arcana.
☟︎ ckang: that stuff
to a layman like myself goes right over my head ;)
ckang: ive pushed 2TB
though a
tunnel before i rebooted
the router for updates
ckang: ive been using it for some
time and its been super solid, i just didnt know how strong it was from
the cryptography side of
things
ckang: sure was just curious since it was crypto related and you guys seem
to know your stuff
there
ckang: its pretty interesting
though if you need such a
thing
mircea_popescu: the fundamental problem with formal verification is
that it's not currently implemented seriously (which is
to say -- completely, on small codebases). it's just machines poking at
things generally, in an untenable
theoretical model.
ckang: ah
this, is what i was looking for
ckang: the pdf goes a lot deeper but
there some info on
the site
mircea_popescu: (which goes back
to a long held asciilifeform notion, of "mining is a bug" -- certainly, but looky here : mining is also
the direct result of "i want a shunt for
the bruteforce, so i can say
to people, "x is cheaper
therefore y won't happen")
ckang: hah, yea
thats hard
to comprehend
mircea_popescu: this is roughly speaking 376440772360506502753317342245835
times
the age of our present universe (the big bang having
taken place 13.8 or so billion years ago).
mircea_popescu: ckang for an ad-hoc illustration : admitting
that you own sunway
taihulight (the chinese supercomputer discussed in
http://trilema.com/2017/resplenduminous/ ), which does something like 9.3 * 10^16 flops ; and admitting you
take 1 flop
to generate a key (it's more like 150-200 irl, but w/e) and 0
time
to check for its correctness,
then you could expect a correct guess about once every 51948826585749897379957793229925273575140
mircea_popescu: but, for
the expert minds
tuned in : ckang 's question does not, as we currently stand, have a published canonical answer i can link him
to. if you write it, i will link it next
time someone asks.
mircea_popescu: there's also some ecdsa involved, but
that's cryptographically less valuable.
mircea_popescu: the proper formula is : address = ripemd160(sha256(secret)).
to go from an address
to its corresponding private key (which is what "bruteforce" requires in
this context) you'd have
to reverse a ripemd160 and a sha256 op.
ckang: i didnd even know base58 was a
thing
ckang: oh wow, ive never seen it broken down like
that
ckang: that just show
the magnitude of something
mircea_popescu: ckang it's one of
the
things people do for
their own satisfaction, but your question isn't without marrow. let's see here...
ckang: just
thought it would be a fun read
ckang: yea
thats why i was just wondering if anyone has done a paper about it ;)
a111: Logged on 2018-04-03 18:39 mircea_popescu: lobbes
the only important consideration here is
that design is not a haphazard activity driven by occurence and circumstance.
that's implementation. design is a deductive activity, it proceeds from first principles and does not break faith.
a111: Logged on 2018-04-11 01:20 lobbes: speed of implementation really (I got other
tmsr irons in fire). I figured I could get a
tickerbot up and running quicker just using a gribble instance >>
http://btcbase.org/log/2018-04-10#1795728 mircea_popescu: in other words : it will never happen,
the world ends first.
mircea_popescu: ckang
the
timescale involved in bruteforcing a bitcoin address exceeds
the computable capacity of an alternate universe in which every single atom extant would be part of an ideal processor working at it.
ckang: its just on what
timescale
ckang: hmm yea,
true, didnt
think about it from
that angle
mircea_popescu: ckang
the idea isn't you bruteforce it,
the idea is
that in between world A, where 100 people living have 1 btc each, and world A', where 99 people living and 1 person dead have 1 bitcoin each, is
that
the apparent value of
the bitcoin will be 101% in A` vs A.
ckang: has
there been any studies done on
that?
ckang: it would
take
too long
to brute force it
though
mircea_popescu: but whether
this properly means you have
taken it or you haven't
taken it is very much an open question of metaphysics, "what is
the meaning of
taking".
mircea_popescu: technically bitcoin you can't
take
to grave either,
the passive result of dead keys is a slight increase in
the value of circulating bitcoin.
ckang: ive mostly just lived under
the "cant
take it
to
the grave or regret it when dead" idea
mircea_popescu: so you never heard of "Do not store up for yourselves
treasures on earth, where moths and vermin destroy, and where
thieves break in and steal." and all
that ?
ckang: i just know
that
they are christian
ckang: im not well read in
theology by any means
mircea_popescu: really ? it's
the cornerstone of "protestant" ideology.
ckang: heh not familiar with
that one
mircea_popescu: quite a lulzy implementation of
the whole "do not build yourself perishable
treasures" meme.
ckang: anything material has
to be
thrown out, sold, etc...
mircea_popescu: the fantasy
the empire is living is
that "nobody can be poor enough
to starve".
the only meaning of which is, "nobody can have any incentive
to hold anything besides bitcoin".
ckang: yea
thats
true, you can literally
take it
to your grave