log☇︎
28200+ entries in 0.559s
mircea_popescu: they don't really exist, unless you're a part of the japanese national wot, which...
a111: Logged on 2017-02-27 12:30 mircea_popescu: (story there was, olympus "hired" aka finally accepted its first foreign devil ceo (michael woodford) in 2011 ; and fired him two weeks later as the dude was principally dedicated to the job of, how can we finance usg out of this japanese corp. the usg however didn't go home, but started "legal proceedings", which eventually resulted in the above theft, plus whatever office supplies woodford managed to take home. apparently 6
asciilifeform: it isn't clear to me, that it does. the typical verification time is ~same
mircea_popescu: (story there was, olympus "hired" aka finally accepted its first foreign devil ceo (michael woodford) in 2011 ; and fired him two weeks later as the dude was principally dedicated to the job of, how can we finance usg out of this japanese corp. the usg however didn't go home, but started "legal proceedings", which eventually resulted in the above theft, plus whatever office supplies woodford managed to take home. apparently 6 ☟︎☟︎
mircea_popescu: i wasn't aware, but hey.
phf: i couldn't get past title. "ten tricks international oil conglomerates that run everything hate"
trinque: bizarre. "no, we really don't think this company which extracts what ~everything runs on~ is as valuable as ... pick your non-essential toy company"
ben_vulpes: didn't see the final part
mircea_popescu: no way out of it, can't allocate more.
mircea_popescu: anyway, no the ramdisk thing doesn't work indefinitely, soon enough the block index will exceed the commodously available ram
jurov: i don't get why that bothers you so much
asciilifeform: the tx index ain't ever getting ~smaller~.
asciilifeform: but take limit of t-->infinity. today it is 30GB, next year - 100
asciilifeform: jurov: it can't run during backup. or during restore. so -- downtime.
asciilifeform: if jurov wants to run his node off ramdisk -- more power to him. but don't try to spin the resulting bitrot as 'hallucination'.
asciilifeform: jurov: people flush write caches purely from hallucination ? maybe i don't need a disk at all then ?
asciilifeform: so it isn't simply a matter of 'make a ramdisk', no.
mircea_popescu: asciilifeform i don't really do blockchain.info style "public support". i think i have the stuff somewhere, i'll have to dig for it. basically it's, blk* live on /sda ; blkindex.dat and friends live on /sdb which happens to be a ramdisk.
mircea_popescu: jurov shutdown of node can readily take > 15 minutes ; and can't even be initiated if the node is, eg, in db lock because block eating.
jurov: Suppose i shutdown bitcoind and backup .bitcoin using rsync (so that all files with recent mtime are backed up).. you say this won't work?
jurov: "entire fs is subject to random rot if my app won't checkpoint its files by calling fsync all the time" is NOT true. where did you get such silly notion?
mircea_popescu: we aren't actually following a purpose here, like "have a good bitcoin". we're merely proceeding from cause : db is broken and THEREFORE must be fixed. not BECAUSE it would bla bla ; but therefore.
a111: Logged on 2017-02-26 19:26 mircea_popescu: now then : a fix for the db would significantly improve a few classes of block verification delays ; and it would alleviate blackhole-like behaviour due to that, node's frozen checking a new block. there's at least 3 different dos vectors for other nodes, and a) the foregoing wouldn't help ; b) if it helped the enemy could easily upregulate the crapflood to compensate.
mircea_popescu: one's like "you were in a coma for the past thirty years, here's what happened that you don't remember" ; the other's like "you had a hallucinatory episode, your history for the past x period is bad and you'll have to rewrite it".
mircea_popescu: yes, but blocks once written don't change. that's the non-rewritable part.
mircea_popescu: it doesn't ?
asciilifeform: it doesn't aggregate accesses
mircea_popescu: the other problem is that a good db fix is a very large project, because bitcoin is written insanely. and our fs db isn't moving, last i heard a month ago someone was going to try and profile an extx ☟︎
mircea_popescu: now then : a fix for the db would significantly improve a few classes of block verification delays ; and it would alleviate blackhole-like behaviour due to that, node's frozen checking a new block. there's at least 3 different dos vectors for other nodes, and a) the foregoing wouldn't help ; b) if it helped the enemy could easily upregulate the crapflood to compensate. ☟︎
mircea_popescu: and i'll point out that the problem here isn't the work, or the thought, but the fucking packaging. you get overexcited and oversignal. it detracts from very valuable stuff.
mircea_popescu: and this isn't the first time we run into the problem of ... let's call them sloppy run "experiments". but the second, this week.
asciilifeform: and 'nodes charge' probably isn't practical
asciilifeform: (ssd ain't cheap, and its price is ~rising~, don't take my word for it, go and see)
asciilifeform: danielpbarron: it ain't 'a cost', like buying a hammer, it's a monotonically-increasing bleed
mircea_popescu: danielpbarron the analogy doesn't hold. currently the tool gives miners cake while nodes pay for the electricity. there's some people cheering on the sides, which i suppose makes the nodes all warm inside ?
asciilifeform: node -- bleeds and bleeds. and -- turns out -- operating proper nodes, ain't cheap, esp. with bitcoin's retarded non-O(1) verification, with the db idiocy
mircea_popescu: danielpbarron that part wasn't much explained. all mining is technically wasted space for the node, not like they get money for it.
danielpbarron: isn't the vacant block easier for you to validate? shouldn't you prefer most blocks to be vacant except when you have a transaction to send?
mircea_popescu: i don't see this "so regularly"
mircea_popescu: apparently you can't make the flock be good christians through tithe control.
mircea_popescu: there isn't an administrative solution to the problem you perceive. if the "godfee" is low, it won't matter, and if it;s high it won't work.
mircea_popescu: that doesn't pay all users, it pays all holders.
asciilifeform: eh we don't even have a cipher of known hardness, nor any approach to one.
mircea_popescu: this was the pretense of shared hosting. it didn't work irl.
asciilifeform: your tx-s still won't be groupable by enemy ~to each other~, only to ~all of his backed tx~
asciilifeform: 1 more upstack : it is possible to make a repudiatable fallback. ( how : you publish the privkey of the fallback addr, after, of course, you've successfully moved its contents to a new one. ) now it is not enough for enemy to find some d00d who knows the privkey to said fallback -- he also has to know ~who had it at time t~, because today ~everyone~ has it.
mircea_popescu: won't take 200 years the 2nd pass.
mircea_popescu: anyway, there might be others, i make no pretense to exhaustivity, hence why this is a very early phase of the design. we don't well know the space yet.
mircea_popescu: (in case it wasn't obvious, diff between political and technological is based on whether people have an incentive to emulate the fix anyway)
asciilifeform: 'it won't grind to a halt in few years' is not 'political incentive' ?
mircea_popescu: this is what i'm saying, anyway. " what we don't really have is the stuff that we really need, such as debottlers."
mircea_popescu: idiot example #2 : a trb which allows txn to be blocked by others than their issuers is ALSO a "way to do things" which doesn't in fact work, and therefore, exactly equivalent to the peter todd & prb idiots item
asciilifeform: that wasn't a legit bitcoin tx tho
mircea_popescu: never mind that. the problem is that if your tx being included depends on you having a miner, you don't actually have a system. just like the 3bullshit isn't a system.
mircea_popescu: it was an instance of "here's something i made that doesn't work, mommy loves me" complex.
asciilifeform: wasn't that ^ thing simply an instance of 'i'ma put some coin on the floor for a miner to take' ?
asciilifeform: mircea_popescu: as in, 'fuck you, i won't relay this, your addr ends with my auschwitz tatoo number' ?)
asciilifeform: and isn't that what the proof of work thing was originally about..?
asciilifeform: i am at a loss as to how this problem is solvable in general case. aside from mircea_popescu's answer to my 'panopticon' thread, 'don't get caught in a jar!'
asciilifeform: i can't even count how many arms and legs exist
mircea_popescu: what we don't really have is the stuff that we really need, such as debottlers.
asciilifeform: what we afaik don't have is 'incentive for node-keeping' algo (though mircea_popescu's partially solves this, by requiring miners to have healthy nodes)
asciilifeform: there may exist some way to solve 'castle problem' that doesn't require a tx to stand alone 'for all time'.
mircea_popescu: wasn't so obvious back when ppl cpumined on the single windows binary
mircea_popescu: it inadvertently forces a node-miner tandem ("you don't like the mempopol, fucking mine it already")
trinque: would release but the thing doesn't need more legs
mircea_popescu: trinque my concern was more in the vein of, i don't want deedbot to answer with an item crafted as described to a !!key command
asciilifeform: doesn't that make setting up the temperature pump harder, rather than easier ?
mircea_popescu: see, the fact that you don't have txn as you describe is what allowed me the "you will die if you fork" threat last year : i don't have to know jack about their chain to murder their chain.
mircea_popescu: somewhat in the same way a retarded child who buys things but always ends up using them as if they were icecream doesn't thereby realise he's retarded.
mircea_popescu: http://btcbase.org/log/2017-02-25#1618154 << it's so funny, at least to me, the sheer wastage of resources usg oligarchs engage in. they keep buying things, which they don't understand, on the expectation that "we;ll find a use for them". they do. it's ALWAYS the same one. somehow the fact that i know in advance what it'll be doesn't inform them as to their horrible strategic position. ☝︎
asciilifeform: which is why i won't even take a position on this , and leave it up to folx who actually worked this scenario in real life , e.g. mircea_popescu .
asciilifeform: mircea_popescu: under the don't-need-antecedent-hash -- yes, you could
asciilifeform: trinque: though in classical bitcoin you don't usually need to regrind
mircea_popescu: canned txn wouldn't be impossible, just less maneuvrable.
asciilifeform: (alternatively, could require that a tx include a 'i want to live in block N' value at birth! and if it doesn't get into N, it is invalid for all time. but this would have serious engineering tradeoff, 'canned' txen as discussed in old mircea_popescu article would be impossible.)
mircea_popescu: similarily, young bride isn't AIMING to become a bitch. she's just getting married.
mircea_popescu: orc dun know what he didn't read on twitter.
asciilifeform: btw not only do i ask for this, but i would even buy a multi-ported ssd. to ACTUALLY read from two holes at same time. because in some applications, you can guarantee consistency using YOUR OWN system, and don't need to rely on some idiot's lock
mircea_popescu: and phf was all like "this isn't in the spec" then read the spec etc
mircea_popescu: no, just, i don't give a shit what happens to any request.
asciilifeform: so probably it isn't '1k/sec' then
mircea_popescu: but you don't ~only. you web. rarely write, often read.
asciilifeform: if you ~only read, you don't need a db
mircea_popescu: (no, the web doesn't have "logins" ffs.)
mircea_popescu: quite literally irresponsible : i don't fucking answer TO YOU SAD LOT
asciilifeform: in other noose, worldwide butthurt re 'oh noez, people won't delete ILLICIT LEAKED SEEKRITZ!111 from crapflare 'bug' !! IRRESPONSIBLE!'
mircea_popescu: o look, stefan molyneux made the list, that chick angie what'sher name didn't.
mircea_popescu: don't begrudge an orc his orcish tongue ; "blockchain principles" is how they say "tmsr made"
mircea_popescu: asciilifeform the guy doesn't know how to say http://trilema.com/2015/a-proper-social-site-for-the-bdsm-community/
shinohai: Why have an api if you can't actually do useful things with it.
asciilifeform: must point out : keyring diddles don't affect folx who don't use keyrings.
mircea_popescu: no fixes till then i don't think.
trinque: of course these things needn't have one particular origin
thestringpuller: http://btcbase.org/log/2017-02-25#1617872 << In world of warcraft alliance and horde factions couldn't talk to each other. So as an alliance player if someone in the horde typed "lol" it would come out "kek". nostalgia :| ☝︎
ben_vulpes: veen: i share these reservations; d'ya think the t-top'd have been better off without a shroud?
mircea_popescu: there's three different ways a node can end up in reorg crisscross, and it doesn't even take owning all its connections.
mircea_popescu: http://archive.is/jaM2c in other things that just happen and aren't anybody's fault, also known as "inept our democracy diplomacy"
mod6: asciilifeform: ya, I've seen the 'std::bad_alloc' from ProcessMessage before -- it doesn't happen to me often, and it doesn't seem to be at the point of OOM exactly. i have noticed that it seems to happen after my node has been up for an extended period of time. say >1week.
mircea_popescu: i like how they manage to package tidal locking and orbital harmonics into "a little known theory from 1989". by the same measure fucking is a little known activity from the 70s. if you're at camp with janine garofalo i guess this isn't even so far fetched a description.