asciilifeform: even in shinohai's summary, http://btcbase.org/log/2016-06-23#1488807 -- it was quite unclear what is 'the hacker' ☝︎

asciilifeform: then, next, plays cloak'n'dagger games with reddit, etc.

asciilifeform: shinohai: lulzy. buterin drains own shit bottle, rebrands as 'whitehat attacker', then signs message. voila, 'nobody' gives a fuck re original message.

asciilifeform: will make my filtration work considerably easier.

asciilifeform: i can't wait for prb to formally break protocol.

asciilifeform: let'em merge 'rm -rf /' for all i care.

asciilifeform: for srs, if you don't intend to quit bad habit, 'own' it.

asciilifeform: saying 'i read on reddit...' without link is quite pointless, you already admitted to habitually fucking farm animals, why not then share whether goat snatch is tight or not, it won't hurt

asciilifeform: another one who assumes that everyone reads motherfucking reddit ?"!

asciilifeform: i can't speak for others, but i ain't dredging through kilometre-deep pile of liquid shit.

asciilifeform: shinohai et al: imho if yer gonna mention 'and on reddit...' it is good form to post link, so that folks who refuse to reddit can follow thread.

asciilifeform: shinohai: and which block is that in

asciilifeform: shinohai: link ?

asciilifeform: ;;seen mike_c

asciilifeform: still in limbo?

asciilifeform: whatever happened to bitcoinalpha

asciilifeform: (possibly not awake just now)

asciilifeform: hm he oughta be self-powered iirc

asciilifeform: $up diametric

asciilifeform: i misread as 'lenin-ing together'

asciilifeform: 'Together women accomplish amazing things. Google supports #LeanInTogether' << google usa front page.

asciilifeform: at any rate, 'hash the blocks' is a pretty obvious thing imho, like 'train cars ought to have linked brakes' - both nevertheless omitted for various jwzistic reasons from the originals

asciilifeform: it is quite impossible: boecks will appear and claim to have done each item, first!11111

asciilifeform: apppaaarently

asciilifeform: 'The idea behind the original Hashimoto is to use the blockchain as a dataset, performing a computation which selects N indices from the blockchain, gathers the transactions at those indices, performs an XOR of this data, and returns the hash of the result.' << sounds like braindamaged version of mircea_popescu's algo

asciilifeform: 'Mirror makes structured products designed to help private market investors get exposure and hedge risk in late-stage companies.'

asciilifeform: http://vaurum.com/hashimoto.pdf << goes to some spam co

asciilifeform: https://web.archive.org/web/20160317002641/http://vitalik.ca/ethereum/dagger.html << appears

asciilifeform: http://vitalik.ca/ethereum/dagger.html << lol 404

asciilifeform: https://github.com/ethereum/wiki/blob/master/Dagger-Hashimoto.md << algo.

asciilifeform: http://cryptomining-blog.com/tag/ethminer << claims gpu only..

asciilifeform: iirc possibly our first one, in fact.

asciilifeform: it was

asciilifeform: (which is where i even started looking...)

asciilifeform: anyway i've only recently grudgingly swallowed hypothesis that ORIGINAL btc pow isn't kleptographic.

asciilifeform: it leverages what they have. (seats on standards committees, stooges in open sores, etc.)

asciilifeform: (and countless others.)

asciilifeform: e.g., the dual_ec_drbg incident.

asciilifeform: very much inline with nsa 'NOBUS' ('nobody but us' doctrine.)

asciilifeform: kleptohash is interesting concept.

asciilifeform: if the ramp ever reaches 1P, we now know how.

asciilifeform: but worth thinking about.

asciilifeform: why not 50P.

asciilifeform: if it existed, why would they stop at faking 50T

asciilifeform: (if there isn't a whole directorate at nsa concerning solely this, they are truly for the dogs)

asciilifeform: (lets you simulate titanic hashrate with barely any effort, if you know the boojum)

asciilifeform: incidentally i wouldn't rule out a kleptographic hash algo

asciilifeform: (crackpot alts)

asciilifeform: the ONE thing they're good for.

asciilifeform: mircea_popescu: very easy to change if they bought up all the old fpga farms.

asciilifeform predicted that a use would be found for them.

asciilifeform: hey hey ho ho THAT's where the fpga farms went.

asciilifeform: prolly running on fpga.

asciilifeform: looks like something vaguely like scrypt.

asciilifeform: if it doesn't, where did they get 50TH of it...

asciilifeform probably condemned to actually read about the thing isnthe.

asciilifeform: i thought eth used btc workfunction

asciilifeform: waiwut

asciilifeform: (esp if you pwn a few boxes connected to existing asic...)

asciilifeform: asic is cheaper.

asciilifeform: anyway nsa needs 5-10mil. boxes with reasonably fresh gpu.

asciilifeform: konsoomers never notice, only poor schmucks attempting linux etc.

asciilifeform: nope, every month they sell new variant.

asciilifeform: in circulation.

asciilifeform: there are 100+ wholly incompatible chipsets.

asciilifeform: nope.

asciilifeform: then you need, say, 5-10mil.

asciilifeform: and distribute.

asciilifeform: BUT let's say you're nsa and write infector for 100+ variants of gpu.

asciilifeform: just that i never turned up (despite considerable effort) a single example.

asciilifeform: can, physically.

asciilifeform: we're doing cpu.

asciilifeform: because drivers.

asciilifeform: there are no working gpu bots!

asciilifeform: CARD mircea_popescu

asciilifeform: CARD

asciilifeform: woah i got it exact.

asciilifeform: 100 mil.

asciilifeform: ;;calc (50*10**12) / (500*10**3)

asciilifeform: let's be generous and give half a megahash (unheard of) per chump

asciilifeform: worse

asciilifeform: more like 100M.

asciilifeform: nope.

asciilifeform: much less botted.

asciilifeform: i'm not certain that qty of boxes has ever been ~sold~

asciilifeform: now calculate how many, say, 4GH winblowz chumps add up to 50TH.

asciilifeform spent years trawling through malware samples collected at tremendous expense, in search of ONE

asciilifeform: there is not nor have ever been a working gpu bot.

asciilifeform: but at 2015 prices it would cost roughly what a toyota costs.

asciilifeform tried to calculate approx cost of nuking eth via 51%, but realized that no one has any public idea re what miner costs presently

asciilifeform: (e.g., say mircea_popescu issues a tx with a fee consisting of ~whole thing!!)

asciilifeform: of course this only works if the 51 is made from solid shitanium, rather than fragmentable (with incentives to defect) cartel

asciilifeform: aha, race is to the swift, yeah.

asciilifeform: i suppose this depends on 'how soft' is the soft fork (are miners merely rejecting 'tainted' tx, or also deliberately orphaning any block which contains one ?)

asciilifeform: (which theoretically could be done for a few btc if you were willing to wait a year)

asciilifeform: what i asked after was the rough equivalent of mining a high-S btc tx.

asciilifeform: unless there is something radically different re ethercrud vs btc, 51% only gives you doublespend

asciilifeform: mircea_popescu: i didn't say 'cost to 51%'. but to get arbitrary valid tx mined in the face of 'soft fork'

asciilifeform: thx for fixing log, phf

asciilifeform: ok