327500+ entries in 0.204s
mircea_popescu: this is
the smoking gun re miner collusion
to scam bitbet.
the loing and
the short of it.
kakobrekla: you get higher priority when unconfirmed
time rises, yes.
mircea_popescu: then it just so happens
that
tuesday night it's good enough ?
mircea_popescu: so a
tx is broadcast, saturday. it doesn't even register.
kakobrekla: i could easily get past fee paying
txes with large and old inputs without fee months ago. i reckon
this still stands.
mircea_popescu: for
the record,
there are about 50k
txn with a higher per-byte offering, and all of
those have been in
the mempool for ehhh i dunno, a week more
than half hour ?
mircea_popescu: just now, an hour after
the other one, all it needs is a block.
mircea_popescu: not saturday, right ? and not
the one before it, when
the bet closed
kakobrekla: is not bad no, i guess
the size and age did it just.
mircea_popescu: how about
that kakobrekla ! not bad for a 0 fee
tx "in
this climate" eh ?
kakobrekla: can you share
the
tx hex of b28245044ac784766b58dcef00750608400c43bd63dce0f8c41491bd9405ad3a ?
kakobrekla: anyway, 'some party' cant order ALL
the miners 'not
to include' your
tx really.
mircea_popescu: as long as
the competitor isn't mined, what's
the hurry ?
mircea_popescu: doesn't need
to be in
the next one. understand how miner collusion works.
mircea_popescu: i personally
think
the only reason it's not mined yet is
that
there hasn't been a block.
☟︎ kakobrekla: anyway, given
the current state of
txes
there is a good chance
that evil 0 fee
tx will get forgotten again
kakobrekla: its possible you are connected
to a node pocket -
trb do not like non-trb nodeas (ban hammer) so
they are only left connected (mostly)
to
trb, not prb which you need if you want
the world
to see your
txes
mircea_popescu: anyway. miners must be broken, burned and
the land fucking salted.
☟︎ mircea_popescu: well, i'm going
to make more, and
try
to mine
them and etc, but yes.
kakobrekla: that is
the doublespend of de9bc173174f5ae97e7fe0d2e171647cade189c76f9cbe48f37494c973aca2e4 ?
mircea_popescu: and
the one doublespending it with a fee is... not visible. right ?
kakobrekla: anyway,
tx is unconfirmed at
this
time with 0 fee, it will
take a while
to get included in a block if it even will be. you can
try
to doublespend
those inputs
to a different address and save
the btc.
mircea_popescu: kakobrekla understand : just because
they picked one particular
txn out of a set of
txn
they withheld should NOT be an argument
they may bring.
mircea_popescu: but
the whole KEY of
the
thing is
that
this "wait until you succeed" ~= infinity.
kakobrekla: the problem isnt bitbet being
targeted.
the problem is 0 fee and
the fact you can know if uncofirmed
tx is out of EVERYONES mempool.
mircea_popescu: if spending worked,
the problem wouldn't exist in
the first place.
mircea_popescu: punkman except spending goes
through
the same process generally.
punkman: gotta
try
to spend/move
the old inputs
to be sure old
transaction doesn't suddenly reappear
kakobrekla: one of
txes with erroneous 0 fee also got lost after a 2 or 3 days a couple of days ago
kakobrekla: and
the 29 blocks confirmed one has 0.00022
mircea_popescu: one of
the many. understand
that
there was a set of
these.
kakobrekla: the currently unconfirmed
tx has 0 fee ?
mircea_popescu: however - since bitbet publishes
the payout addresses, it's
trivial for
them
to know what
to sit on.
mircea_popescu: kakobrekla
this is what i was doing for a fucking week.
kakobrekla: afaik would not happen if you would select same inputs
to be used for second
tx or alternatively rebroadcast
the first one
mircea_popescu: the problem is broadly speaking
that bitcoin is not actually fit for commerce, and
that fixing it necessarily requires changing
the pow
to exclude f2pool/antpool/etc.
☟︎ mircea_popescu: this means
that if you are waiting on a payment - you got
to
talk
to a miner. bitbet will not further consider any complaints of
this nature.
mircea_popescu: HOWEVER - bitbet will NOT ever make 2nd
transactions for any reason and in any case.
mircea_popescu: the damage in
this case is ~17 BTC, which whatever - cdb.
mircea_popescu: because
they can sit on a
transaction more or less forever.
mircea_popescu: this adds an entire layer of breakage
to
the already broken bitcoin infrastructure. specifically - due
to chinese miner collusion (and yes
they do control over 50% of
the network) it is impossible
to guarantee payment at ANY point in
the future.
mircea_popescu: once a
transaction
that DIDN'T spend
the same inputs was announced and confirmed,
the OLD
transactions were somehow found, and are now broadcast.
mircea_popescu: specifically : a set of a dozen or so
txn
that were
trying
to pay out winners of a bet went out, only
to be "ignored" by
the network.
they all spent
the same inputs.
this went on for about a week.
mircea_popescu: sooo... it pains me
to say
that bitbet was just on
the receiving end of a concerted miner withholding attack.
mircea_popescu: well...
the deed seals keep getting spent, so someone must be
trinque: it would be good for me
to post an example verifier on
the site
trinque: yeah, just a matter of
taking
the deed bundle
txt, performing sha256 and using
that as
the privkey, generating a pubkey and address from
that
mircea_popescu: maybe
try and verify one on
the basis of
the mirrored data only
to be sure ?
mircea_popescu: aite just make sure some silly
tidbit doesn't get lost and we find ourselves stranded
trinque: the FAQ page specifies how
to verify
them if one has a deed bundle
assbot: Logged on 01-03-2016 21:39:15; mircea_popescu:
that
trump
thing's on fire
Anduck: fwiw, i
tried
to pm with
trinque but he kept ignoring me from
the beginning. my script accidentally used his site
too frequently and now he's doing
this, annoying me occasionally with sudden "fuck you anduck" hilights etc
mircea_popescu: i dunno, originally it was a pretty good marketing
tool.
mircea_popescu: anyway. bill's up, i guess im closing down
this phuctor server and eating
the 2 month bill as cdb ?
mircea_popescu: lol. well,
the plan is working splendidly. who'd bother shooting you, you're rather easy
to mothball anyway.
mircea_popescu: asciilifeform> anyway i specialize in making some selected subset of it actually WORK. <<
this much is
true, yes. but you have a finite number of hands!
mircea_popescu: fucking stupidest idea ever, should have
tried building it on old crack whores first.
mircea_popescu: asciilifeform ikr ? and now we had
to build money on
top of it
trinque: jurov: gotta go do an errand, but when I'm seated at desk again I'll put some more
thought into
the section of my nginx config labeled "Fuck you, Anduck"