548900+ entries in 0.357s
mircea_popescu: i suppose it is a
testament
to my scholarship
that i can actually get away with outsourcing
the actual readsing ?
assbot: Why Dogecoin is a scam, why
the people pushing it are assholes, why Business Insider is a contemptible piece of shit, why anyone who ever worked for it will be dancing in
the street for nickels and why Kevin Rose is a fuckwit. Plus other considerations. pe
Trilema - Un blog de Mircea Popescu. ... (
http://bit.ly/1uKxtRQ )
mircea_popescu: how about
the production of communications
to include voice << im not sure right off what
that'd do.
diametric: Somehow I've never seen
this before.
mats: i was hoping you'd have a better idea
tbh.
mats: perhaps. do you see meaningful expansions ? << how about
the production of communications
to include voice, where
there is a written record of it? e.g.
transcript of wiretap log
mats: mats is like
the oracle on
the
topic now! << not sure if you're criticizing what i said
there
decimation: yeah.
the public can make bitbets;
that's probably mostly what
they want anyway
decimation: at
the present
time, it seems
to me
that bitcoin derivatives do not serve a real market need
mircea_popescu: <mats> mircea_popescu: are you interested in loosening
the requirements? << perhaps. do you see meaningful expansions ?
mircea_popescu: buttonwood_>
There has
to be a better way
to do otc
trades
tho. << nope. isn't, for provable reason. isn't happening in practice, either.
mircea_popescu: <mats> multi sig and escrow are essentially bad patches
to a broken
trust model << mats is like
the oracle on
the
topic now!
mircea_popescu: <buttonwood_> I've been really interested in applying
the smart contract
technology
to bitcoin-otc. Are
there currently any active smart contracts or oracles doing escrow
trading on bitcoin irc ? << no. people do
thje
trust calculation by hand.
mats: i figured i'd give it a shot before proceeding, i've put at least 400 hours in already and i believe
there may not be quite a lot of useful data at
the end if i continue
this way
mats: mircea_popescu: are you interested in loosening
the requirements?
mats: mircea_popescu: but
there are 11 judges with 0 cases and 12 with just 1. i'm now going
through and reading every case involving USG, i've cleared
ten judges
this way, but
the number of cases i've been able
to find have been fairly low in number
mats: mircea_popescu: i've observed
that
the number of cases meeting
the parameters in
the courts-circus project are very scarce: i've read every case involving
the judges nubbins` listed, with
the keyword privacy on Westlaw, which yielded about 90 cases
Pierre_Rochard: ^ exactly,
the difference between a computer oracle and a human arbitrator approaches nil over
time
ben_vulpes: buttonwood_: ^^
that bet right
there is a good example of best-in-class contracts.
[]bot: Bet placed: 2 BTC for No on "Gold
to drop under $1000 before March 2015"
http://bitbet.us/bet/1101/ Odds: 16(Y):84(N) by coin, 18(Y):82(N) by weight.
Total bet: 12.30387905 BTC. Current weight: 61,285.
Pierre_Rochard: buttonwood_ Mainly because of
the
trust involved if you were
to sell me an option. < I’d say it’s mainly because
the volatility and discount rate are going
to be completely out of whack until fiat is destroyed
ben_vulpes: anyways, buttonwood_ are you still interested in working
this
thread?
ben_vulpes: anyways,
there was one. are you aware of why it shut down?
mats: 21:15:50 <+mats> multi sig and escrow are essentially bad patches
to a broken
trust model << in
the way you mean
to use it,
that is
mats: man,
this sicilian lentil soup is fucking amazing
ben_vulpes: you're clearly in a good position
to be
talking about how business should be done around here.
ben_vulpes: trust is built over
time, and can't be abstracted out.
ben_vulpes: buttonwood_: you can't
trust people
that you don't
trust. end of story.
ben_vulpes: <mats> use
the ratings and
talk
to
the raters <<
this. again and again,
this.
mats: use
the ratings and
talk
to
the raters
mats: offloading due diligence and
trying
to use
technology as a crutch
to paper over deficiencies in your business relationship sounds suspiciously like
the old world way of doing
things
mats: sack up and
take responsibility for your actions.
mats: multi sig and escrow are essentially bad patches
to a broken
trust model
mats: can't be done. deal with people you
trust or don't do business, full stop.
mats: look, you can't offload
the complex business of
trust
to anything but a human.
mats: all extant projects purporting
to be 'smart contract
technology' is a scam. like ethereum.
mats: 'smart contract
technology'.
mats: its not a real
thing, buttonwood_.
decimation: there's a semi-anonymous column
there by
the same name?
decimation: buttonwood_: do you work for
the economist?
mircea_popescu: what's being verified is a) control of
the address in question and b) bitcoin feeding it included in a block.
mircea_popescu: A. Here's proof i control address X, and here's a payment of 500 btc
to it *included in block Y*
jurov: right. and even if node doesn't create
txs itself, i'm asking about verifying whatever
txs flying around
ben_vulpes: and all of
this because i asked about what should bear responsibility for creating and signing
transactions.
jurov: whether
they are mined or not
jurov: maybe.
the issue was
the need
to
track unspent outputs, regardless of
mircea_popescu: maybe
the reason we've
talked so long about
this is because we're saying
the same
thing/
jurov: check
that it is verified is done by checking it's in block somewhere
mircea_popescu: you can't have "these are
the valid and
these are
the invalid
txn of block 6"
jurov: check
that is it valid is done by checking all
the inputs' singatures
jurov: you can have valid unconfirmed
transactions, even valid
tx
that spend unconfirmed
transactions
mircea_popescu: see ?
that's
the whole
thing.
that
these
two are separate.
jurov: "valid" and "confirmed" are
two
things
mircea_popescu: the reason ytou know block 5 is a valid block IS
that
they were valid outputs, or was at
the
time, but
this is a separate
topic.
mircea_popescu: but
the reason you know
they're valid outputs is
that block 5 is a valid block.
jurov: ^yes now
that
transaction would list both inputs from block 5
mircea_popescu: 1 btc from block 5, 1.5 btc also from block 5, 2 btc from block 6 -> 4.5 btc
to 1derp
jurov: regardless if
they are in blocks
jurov: no,
they're verifiable by
themselves
mircea_popescu: but if
they are different outputs, included in variuous blocks...
they're verifiable by
those blocks... ?
jurov: makes a withdrawal out of
them
mircea_popescu: <jurov> now you do withdrawal, you need
to gather and sign all of
them despite it's
the same 1Fx address << it's whjat i
took
this
to introduce.
ben_vulpes: much like if you want a > 1m
transaction
that's your problem
mircea_popescu: jurov weren't you proposing
that *somehow*
there's ambiguous outputs as
the entire point of what you were saying ?
ben_vulpes: if i want
to spend my low-priority outputs,
that's my problem.
jurov: nor
there was successful attempt
to make it so so far
mircea_popescu: ben_vulpes nah. im just saying, if it's ambiguous,
then it's ambiguous.
ben_vulpes: perhaps i don't understand what all gets signed. more
than likely.
ben_vulpes: how is my signature even valid in
that scenario?
ben_vulpes: but you're saying
that if i cook up a
transaction
that has 1march29th as an input,
the miners should actually rewrite
that
to use 1march15th as an input?
jurov: but it hasnt' anything with bitcoin
then
jurov: well mircea,
then make a spec. many people unsuccessfully
tried various mixing proposals
that were supposed
to do what you propose
mircea_popescu: ben_vulpes is
this "bitcoin as it should be" or "bitcoin as it is
to be inferred from current codebase"
ben_vulpes: i'm
talking about
the reference implementation.
mircea_popescu: <mircea_popescu> jurov i
thought we were discussing bitcoin as a spec, rather
than bitcoin as a hack. <<