631600+ entries in 0.58s
kolinko: let's say
that I wanted
to build a business
that relies on options... a web wallet
that maintains
the users' bitcoin balance stable in relation
to usd (e.g. I put $100 worth of btc into
the wallet, and after a year I'm almost guaranteed
to be able
to withdraw $100, regardless of bitcoin price)
The20YearIRCloud: mircea_popescu: re : keynesians -
they still seem
to be popular in
the US. Alot worship at
the
temple
ThickAsThieves: so at first i was frustrated at
this 1bTCXE
trading challenge because basically what it is is
they gave everyone a fake account with half BTC / half USD credits and people are set wild
to
trade
the assets. However,
there is no
tie
to
the real market so it's mostly a psychological/economical experiment. In
this case, is
there
theory on how such a market would play out?
mircea_popescu: but really,
there doesn't seem
to be
that much demand for options once
the darkpool and bitbet respectively
take
the slack.
mircea_popescu: it'll just work from
there relying on
the wot and
that.
mircea_popescu: imo
the avenue
to
this standardisation is, once moloko delivers
the registrar of deeds, and if indeed we want
to do a lot of option
trade,
ThickAsThieves: people will
try anything if youre wearing
the right clothes
kolinko: what I'm really
trying
to figure out is whether people would be willing
to
try such a system
mircea_popescu: nah, it can actually be answered in
theory
too. just, it gets esoteric quickly.
kolinko: as for people being bothered into doing
this - it's a valid question, but one
that can only be answered by
trying it in practice
mircea_popescu: a system
that either works as intended or not at all is
the holy grail of unbreakability.
mircea_popescu: ThickAsThieves but see, if you delete
the program, you can't say you've broken
the code.
kolinko: the problem is
that
the overlord can disappear at any
time
mircea_popescu: this is
the main advantage of an overlord
that
the decentralize fanboys
tend
to ignore. if you have a good overlord,
the system is unbreakable period.
mircea_popescu: and for
that matter :
the spf system is not breakable.
mircea_popescu: hence my comment about
them being arsed
to put up with
the cost of
the defense.
mircea_popescu: but
this also happens
to be
the definition of a sybil vulnerability.
kolinko: you're nitpicking. what I meant is
that it doesn't matter who created
the list and how. what matters is whether
the list really contains gpg keys of
the people, and whether
the people on
the list want
to participate
mircea_popescu: ThickAsThieves nah, he's just
trying
to solve a fundamental problem
through implementation and got blindsided by a
theoretical exposition of it.
mircea_popescu: and anyway, if "you" create it
then you're just counting from 1
to 15.
ThickAsThieves: why are we making lists and giving
things
to 15 people? i guess i better read
the logs, is
this OT or something?
kolinko: in other words - you don't
trust
the list of 15 people because
the list of 14 people said
that you should
trust it. you
trust
the list because you can see who's on it, and
they look
trustworthy.
kolinko: you can create
the list of 15 all at once. people wouldn't
trust
the list because
they saw it being created.
they
trust
the list, because
they can review it one by one, and see
that all
the individuals are indeed
trustworthy
mircea_popescu: kolinko if you have a process
to go from 1 party
to
two,
then you have a sybil problem. and
to have 15 you necessarily must have had
that process.
mircea_popescu: what if one wants
to join ? were
they 8 at some point ? 6 ?
mircea_popescu: people
tend
to easily forget what identity means. here it is : =. every
time you're using an equality sign, numerical, logical or otherwise you're relying on identity. cuz
that's what it is.
kolinko: what I meant is - if you have a system of 15 parties, each one of
them signing messages with
their own gpg keys,
that can be quite secure against
the sybil attack.
the attacker would have
to steal keys of 8 independent parties.
mircea_popescu: now, if you want
to go back and get diddled by
THE SAME GUY AS LAST
TIME, you're in for it.
mircea_popescu: anyway, sybil attacks are specifically not a problem in anon systems. if you go for a romp in a gay cinema,
to be fucked by a random man, it makes no difference
to you who fucks you, so you can't be sybil'd by definition.
kolinko: jurov:
the idea is
that you agree on an algorithm beforehand, and everyone uses
the same one
jurov: kolinko it currently does not happen,
that's
the problem
mircea_popescu: and
the arsed problem is not something
to brush aside.
the better your sybil protection,
the more expensive
their pointless expense.
mircea_popescu: you are wrong in
the first. sybil attacks are a problem on all systems
that rely on identity
to any degree.
jurov: lol how would panel of 15
trustworthy individuals come up with bitcoin vwap
to be used for option exercises?
kolinko: as for why would be
they arsed - for provision/payment of course
kolinko: well, sybil attack is really only a problem in systems
that rely on anonymous parties
mircea_popescu: circle cvadrature is not a question of "get better draing
tools"
mircea_popescu: these aren't simple problems, and
they're unsimple for fundamental reasons, it's not a matter of implementation.
mircea_popescu: sybil attacks, and why would
they be arsed
to care about your needs.
kolinko: well, but if we had a panel of 15
trustworthy individuals, with a consensus of more
than 50% of
them required
to release
the fund
mircea_popescu: the problem with
this approach is
that if
they're
to be
trusted
they're sops by definition, and vice-versa.
kolinko: the idea for
the distributed contract was
that if I want
to build a business
that relies on options, such a business shouldn't rely on a single individual/company
kolinko: mircea: I'll read
the misc in a sec. as for your volume - I know it well - I'm your user since 2012 iirc
kolinko: mircea_popescu - confirmation
that I'm
the appcodes kolinko:
twitter.com/kolinko
mircea_popescu: a "single point of failure" is a
trustworthy individual by definition anyway.
mircea_popescu: and when you're done with
that,
to put
things in perspective for you, mpex carried about 1mn
total btc worth of options over a coupla years.
assbot: MPOE, February 2014 Statement pe
Trilema - Un blog de Mircea Popescu.
kolinko: (the idea is
that a smart contract would be protected by a set of
trustworthy individuals - no reliance on a single point of failure)
kolinko: and I'm
thinking of building a bitcoin smart contract for put/call options
kolinko: cool, will do
that in a few minutes
mircea_popescu: if you prove
that i'll rate you,
then you can self voice
kolinko: I founded Orisi
two months ago (orisi.org) - it's a framework for building smart contracts
mircea_popescu: o hay beautyon_ ! Le never fails
to unintentionally entertain, but really,
that's one you should prolly blog.
nubbins`: what's
the word for when something is made
to look like a human?
assbot: imgur:
the simple image sharer
fluffypony: nubbins`: at least
they're being useful and helping with cleaning
mircea_popescu: anyway
this dga fellow should be here. someone invite him
mircea_popescu: lions in
the fucking forest. huzzah for all
teh lernin'.
assbot: Judy Garland: Lions and
tigers and bears! Oh my! (The Wizard Of Oz, 1939) - YouTube
fluffypony: and indian girls looking for free flight
tickets
fluffypony: sometimes
there is even pronz in
the channel
mircea_popescu: i can arrive at a formulation of what
the op
thinks
that i can integrate, it just
takes an op
that can answer questions for long enough :D
mircea_popescu: so basically you're saying
the
technical
tradeoff in question isn'
treally
trading anything worth
the menton.
fluffypony: mircea_popescu: well Bitcoin's solution
to
the problem is
to use SHA2
mircea_popescu: fluffypony ok, so if we accept
this avenue exists,
then a blacklisting scheme would not logically be expected
to resolve it, just move
the problem.
fluffypony: no, we
theorised
that
the attack was possible
mircea_popescu: fluffypony so are you basically saying he's just
talking nonsense ?
mircea_popescu: "
technical
tradeoffs (slow block verification leading
to increased susceptibility
to block-flooding DoS attacks"
fluffypony: in
the 0.03 seconds it
takes
the peer
to verify
the block you won't even have completed
the handshake
to send another block
jurov: mircea_popescu: so you send bad block
to everyone at once, yes?
mircea_popescu: can i craft blocks and announce
them so
that
the entire fucking network ends up graylisting each other