assbot: Logged on 04-07-2015 06:06:14; midnightmagic: I'm familiar with the DER-encoding change they made, and I'm aware of, if not familiar with, every major bug in openssl since 2001 or so. Could I draw a line between releases that had bugs and releases that fixed them? No. Not even close.
assbot: Logged on 04-07-2015 05:54:09; midnightmagic: that is the inverse of what will happen as more-efficient mining equipment arrives.
mircea_popescu: i mean, other than painting the picture of the vermin in unflattering tones, which it does. in the field it can do precisely jack.
assbot: Logged on 04-07-2015 05:26:45; decimation: I would also note that the bitcoind github commits and comments lie about the IsSuperMajority machine. They say that the mandatory rejection won't take effect until 95% of the blocks are incremented - but in fact it's only 950
mircea_popescu: but at any rate : setting the "nversion" to maxint has at least the important symbolic significance of saying "this is the last version".
mircea_popescu: http://log.bitcoin-assets.com/?date=04-07-2015#1186772 << and watch the PR implementation crash and burn over division by negative zero and things.
mircea_popescu: i believe the lesson was learned.
assbot: Logged on 04-07-2015 04:51:34; asciilifeform: (for as long as it carries under 'technical', the weasels can whine, wheedle, emit 'reasonable reasonings', even persuade the persuadable)
mircea_popescu: http://log.bitcoin-assets.com/?date=04-07-2015#1186723 << it has historically proven to have been a massive mistake, perhaps the largest mistake on record, that people in the early f/oss made to argue on supposed technical merits and play-pretend the "impartial scientists".
mircea_popescu: fitting for the empire of stupid, but still.
mircea_popescu: nevertheless, being critted for > 9k ironies over something like this is beyond comedic.
mircea_popescu: http://log.bitcoin-assets.com/?date=04-07-2015#1186697 << i dun think it does anything useful ; on the other hand it doesn't do anything whatsoever that wasn't either already done, or as good as already done. but whatever, let the flies buzzing in front of the truck get their fly trophies for opening the road to trucks.
mircea_popescu: http://log.bitcoin-assets.com/?date=04-07-2015#1186389 << for a long time in 2012/2013 there were "gencoin only" miners, at the peak doing like 15% ish of total hashing
mircea_popescu: http://log.bitcoin-assets.com/?date=04-07-2015#1186358 << also touched upon in here a few times. definitely the right way to do this.
mircea_popescu: (an important point about co2 extinguishers is that they also cool when deployed. this effect is significant. heavier gases - not so much)
assbot: Logged on 04-07-2015 02:07:07; DanyAlos: I was looking for #bitcoin-assets on this search engine (http://irc.netsplit.de/channels), and realized that it is not listed. Is there any particular reason for not being there?
trinque: nice way to read the source
mircea_popescu: http://log.bitcoin-assets.com/?date=04-07-2015#1186287 << co2. still the best solution in any perspective.
ben_vulpes: midnightmagic: that's some wack ass disingenuous shit. you jumped in on the n block consensus line.
decimation: at any rate, I'm going to sleep, I'm sure others will comment tomorrow on this discussion.
thestringpuller: midnightmagic: also mircea_popescu 's series on the subject (http://trilema.com/2015/lets-address-some-of-the-more-common-pseudo-arguments-raised-by-the-very-stupid-people-that-like-the-gavin-scamcoin-proposal/) outlines and details the idiocy of every redditard comment that was, is, and will be on the subject
thestringpuller: midnightmagic: there really isn't news. just gavin/hearn having a temper tantrum like a little kid because they want more subsidies for poor people.
decimation: I think there's a general ambivalence about what bitcoin-devs want or do
thestringpuller: midnightmagic: cause I fucked up my shoulder being on the computer all day at work so I read comic books instead after work.
ben_vulpes: dumpblock, eatblock, exposure of the ancient blockchain is quiet now?
midnightmagic: thestringpuller: not at all, I mainly wanted to know why you all were being so quiet lately. :-P
ben_vulpes: nah we seem to have defused the antagonism for now
midnightmagic: The Satoshi-Halo-Wearers.
decimation: 'they' being gavin et al?
midnightmagic: Maybe my appraisal is just wrong. Could be. Seems to me they're just taking a breather to set up the schism.
ben_vulpes: <midnightmagic> ben_vulpes: Dude man, with the shit you guys say in here, I have no idea when you're ribbing someone, or promising a spear in the gut. :( << the ambiguity has to be completely intolerable
ben_vulpes: anyways, chickens run around with their heads cut off for minutes. what of it? they still lost to the butcher.
midnightmagic: ben_vulpes: Dude man, with the shit you guys say in here, I have no idea when you're ribbing someone, or promising a spear in the gut. :(
decimation: why not solidify the questionable openssl code first, before lightly restricting certain signature forms?
ben_vulpes: midnightmagic: re #trollfail it's supposed to be lighthearted elbows-in-ribs
decimation: all I can say, for myself, is that folks write stuff on this channel, I would read; comment - as would others
decimation: umm, why would we give a shit about what they are saying otherwise? I donno bro, you're the one accusing here.
ben_vulpes: moreover, if the decisions made here trickle down to "bitcoin core", why should we pursue them further?
midnightmagic: .. why would they do that?
decimation: if bitcoin core devs agree with what people here are writing, why don't they venture here to make their case?
trinque: gcovr outputs a nice html version of the gcov output
ben_vulpes: so we won, and you want to 3/10 troll on the topic?
midnightmagic: You guys *heads* were exploding in here, and then when "the rest" of the bitcoin-core devs took up the flag you all went quiet again. wtf?
midnightmagic: And here they are, going on little half-drunken joke-rants about how maybe they should just remove everyone else's commit access and unilaterally take control again.
midnightmagic: Unless you are implying people outside the bitcoin world are voracious readers of the -ass logs, as far as I can tell in all the articles, reddit posts, twitter feeds, etc, I don't see more than a passing mention. But even if that weren't so, really I'm a little disappointed the wind all went out of your sails, as it were.
ben_vulpes: <midnightmagic> Yeah, I mean aside from the grumbling in here which nobody but people *in here* seems to read. << how could you possible know
decimation: he did a few times, rather unimpressively
midnightmagic: For a while you were writing reddit posts and qntra articles and all sorts of stuff. Half the time you got Gavin himself "derping" as you put it, in your comments.
midnightmagic: Yeah, I mean aside from the grumbling in here which nobody but people *in here* seems to read.
midnightmagic: it's all a question of how much you trust your ability to make code that converges on consensus. are you so awesome you can either sit on an old openssl, or write your own replacements? are you so godlike you can write testcases for every corner-case, bugs and all? I know I'm not. Maybe you guys are. I dunno.
decimation: his code sucks, I don't deny, but it's the closest thing to a spec that we have
ben_vulpes: it's all well and good to say "strip out the components", i just don't buy that that's possible.
midnightmagic: ehh. satoshi's code, bugs and all. we could also stick with bdb and accept the quirks like the old accidental fork post-leveldb.
decimation: again, nobody is forcing anybody use a version of openssl they don't want to use.
midnightmagic: because, like I said, I think some people in there appear to be adding exploitable code with absurd frequency.
midnightmagic: or, do what sipa did and write a secp256k1 lib because the openssl people don't give a shit they're wrecking dependencies.
midnightmagic: I'm not pretending that; I'm explicitly saying, divergence implies there is no reason to even *use* openssl at that point. strip out the components, use them, skip openssl entirely, and, I guess, trust in your ability to monitor the progenitor of your codebase for bugs that *explicitly affect your consensus-critical code*.
ben_vulpes: decimation: no but you see we need to slurp the spitoon because if we don't something terrifying that we can't reason about won't happen!
decimation: understand the code, make sure it doesn't happen
midnightmagic: well, then we have the forking risk I mentioned above: what happens when the openssl people make a fix or the internet finds a bug in the component that we depend on? If we sit on 0.9.8 or whatever the version was before those idiots got their hands on it and started adding malicious exploits, what happens?
ben_vulpes: midnightmagic: you were probably just on the cusp of killing hitler too.
decimation: midnightmagic: the cure to that problem is not forcing changes of an uncertain nature, but to gain certainty in the codebase
midnightmagic: decimation: Nobody, of course. Then we go back to static builds and what happens when an actual bug hits and the fork is so old that the fix doesn't backport? How divergent are we willing to accept?
ben_vulpes: this, i gather, is news to you.
decimation: why isn't anyone seriously attempting to extract the open-ssl code paths used by bitcoin?
midnightmagic: I'm familiar with the DER-encoding change they made, and I'm aware of, if not familiar with, every major bug in openssl since 2001 or so. Could I draw a line between releases that had bugs and releases that fixed them? No. Not even close.
ben_vulpes: midnightmagic: art thou familiar with the 1.0.1g issue?
midnightmagic: Thus, backing up further, there is the legitimacy of BIP66 at all. If it is not legitimate, we have a consensus code failure every time openssl decides they want to change behaviour.
ben_vulpes: just the network. that, only that, no more, no less.
ben_vulpes: there's* just no window.
midnightmagic: ben_vulpes: On that at least, we agree. I agree with that: the current hashrate is as illegitimate as a vote of private keys would be in determining a softfork. What else is there?
ben_vulpes: this moronic 'acceptance-in-blockchain' algo doesn't work, because hashpower can revert and rewrite an arbitrarily-lengthed blockchain.
midnightmagic: Correct. I am saying your complaint about it being unreasonable is illogical by any measure of the mining work done: there *is* no other meaningful window, or measurement, of the bitcoin network without shifting to PoS or DPoS. But if you want to do that, fork bitshares.
ben_vulpes: midnightmagic: you make this mistake of letting the plants control the conversation. "what's the right magic number of blocks to signal fork acceptance?" answer: there isn't. there is only the long-term behavior of the network.
ben_vulpes: i don't accept it either.
decimation: nor does this argument hold for 1000, because the same argument can be made for 50000
midnightmagic: Regardless, a 1000-block window is not unreasonable if one accepts that mining hashpower is the vote that counts.
midnightmagic: ben_vulpes: You would call a bite any response. That is the fundamental nature of #trollfail.
decimation: no, but it is for profitability in the face of non-zero electric rates
decimation: unless someone can tell me exactly how they are going to 'get efficient'
decimation: I think it's likely to level out in the coming year or two, maybe longer
midnightmagic: that is the inverse of what will happen as more-efficient mining equipment arrives.
decimation: midnightmagic: the "amount of work" argument utterly fails to impress
ben_vulpes: another value for derpfiguration.
midnightmagic: decimation: 1000 blocks is a couple percent of all work done ever on the entire blockchain since inception, and the current hashrate could rewrite the entire history up to something like august 2014 in somewhere close to the span of time that non-vote took place over, times a very small number. :( unfortunately.
assbot: Use CTransaction/CBlock version numbers for smoother upgrades by gavinandresen · Pull Request #1525 · bitcoin/bitcoin · GitHub ... ( http://bit.ly/1LNKrVq )
decimation: hardly 'consensus' in view of the 363000 block history of bitcoin
decimation: 950 out of the last 1000 blocks
decimation: I would also note that the bitcoind github commits and comments lie about the IsSuperMajority machine. They say that the mandatory rejection won't take effect until 95% of the blocks are incremented - but in fact it's only 950
mod6: i put it out there on derp-media too
decimation: as more than 950 blocks have passed since the first instance of the IsSuperMajority machine being used
trinque: the golang part gets into some dumb state where it wont reconnect.
ben_vulpes: those hafta be the 2 derpiest typos of my life.
ben_vulpes: http://therealbitcoin.org/ml/btc-dev/2015-July/000114.html << ahaha fattest fingers.
decimation: where did this consensus shit enter into the code base?
mod6: ok mp says he can see signing that statement ben_vulpes. go ahead, he'll even sign later when he gets on his other box.
decimation: err, no that was the bdb lock thing
assbot: Logged on 21-01-2015 02:48:11; Luke-Jr: it can't, if you don't have the 0.8.1 hardfork patched in..
gribble: The operation succeeded.
ben_vulpes: looks like the shitgnomes have been working overtime to address my complaints.
ben_vulpes: an informal signoff from y'bosses would be nice before i deposit this in the white porcelain turdotron
decimation: asciilifeform: I can already hear jwz saying that he doesn't want politics, just to do the right thing
asciilifeform: (for as long as it carries under 'technical', the weasels can whine, wheedle, emit 'reasonable reasonings', even persuade the persuadable)
asciilifeform: political, rather than technical decision
decimation: although as I've noted, they already incremented this machine
asciilifeform: this is why shitgnomes are to be written out of the script permanently and unconditionally
mod6 doublechecks the SoBAs
mod6: otherwise we had issues on 168`001 iirc
ben_vulpes: what was the version number of openssl we determined necessary to validate the full chain?
ben_vulpes: well this signature issue is tricky as shit already - openssl is already blowing up validating the chain with anything other than...
decimation: but the way it's being forced is pretty stupid
asciilifeform: 'who comes to us with a fork, shall die by the fork'
asciilifeform: 'the fella shooting our way'
asciilifeform: ben_vulpes: we're doing war, not mathematics
ben_vulpes: this is almost as bad as defining a word with the word to be defined
trinque: would the logs be interesting in any way?
ben_vulpes: there's a ph0rk in progress
ben_vulpes: no no the conformal impl
trinque: I have others
mod6: the hope was to find if he can verify if v0.5.3.1 rejected or accepted the blocks in question; 363738
mircea_popescu: anyway, having been duly amused for the evening, i retire to my eulorean empire.
asciilifeform: i doubt anything can ever beat the 'i'll use c++, and on vs for good measure' moment
mircea_popescu: but the "we'll make wallet a plaintext file" and the "we'll target windows" and the "you know what's better than boost ? qt!" were pretty epic turn points too
ben_vulpes: meaning adopt the power ranger braindamage again?
mircea_popescu: nah, the "we'll move to this new db system one schmuck that gofer'd coffee at google wrote with his chest hairs and nobody tested in any way" was the stupidest sahit
decimation: if there's any reason why bitcoin will never be adopted, it's this kind of shit
decimation: this is the dumbest fucking thing I've seen in all my years with bitcoin
mircea_popescu: "Current status: F2Pool still broken; Antpool fixed (but no promise they won't intentionally re-break in the future)." is really all that one needs to lol
mircea_popescu: Explorers on the good chain:
ben_vulpes: i want a man in the ground, i don't make that dependent on his wife squirting or w/e
mircea_popescu: on the record having lied for profit and all that good stuff, but why should anything matter.
gribble: The operation succeeded.
ben_vulpes: "Most likely this is caused by broken-by-design-for-profit mining code, but none of their stuff is open source AFAIK. Maybe more details will be known with time."
ben_vulpes: reddit, for all the silencing and muting of the actual bitcoin foundation, seems to be turning the corner on sense
ben_vulpes: "Updating to the latest when what you have works is how you break things.
ben_vulpes is also curious to see how much "political capital" the derpdation burns today
mircea_popescu: i dun see it. the rule is "come to school dressed". if you all agree to wear skirts or all agree to wear cardigans, it's still school.
ben_vulpes: mircea_popescu: anyone can narrow the rules, but whether that's "bitcoin" is an open question.
decimation: I'm not sure if there are enough blocks to trigger it
mircea_popescu: otherwise, no.
mircea_popescu: decimation provided their "mine support" actually materializes
decimation: the problem is that 0.5.3 blocks generated by 0.5.3 won't validate if those IfSuperMajority rules check
mod6: <+ben_vulpes> Bitcoin 0.5.3 is the canonical reference implementation. If a fork occurs and one side validates on the 0.5.3 codebase while the other does not, the chain that validates under 0.5.3 is the only valid chain. << I have nothing further to add to this at this time.
ben_vulpes: too bad, fuck them.
mircea_popescu: hard fork is enlarging the rules.
decimation: they claim this is a softfork
mircea_popescu: well, any people can at any time decide to narrow the rules. this makes some sense.
mircea_popescu: i have nfi how the power rangers imagine they'll manage a hardfork when they can't as much as get a "relevant support" softwork that's fairly uncontroversial.
ben_vulpes: frantic action is for the impotent.
mircea_popescu: anyway, todd is right. the only way this affects us is that we don't really give much of a shit.
mircea_popescu: Bitcoin Core (after 0.10.0) rejects these invalid blocks, but a lot of other stuff doesn't. SPV Bitcoinj wallets do no validation what-so-ever, blindly following the longest chain. blockchain.info doesn't appear to do validation as well; who knows what else? <<
ben_vulpes: Bitcoin 0.5.3 is the canonical reference implementation. If a fork occurs and one side validates on the 0.5.3 codebase while the other does not, the chain that validates under 0.5.3 is the only valid chain.
mircea_popescu: BIP66 protocol rule changes have gone active in part thanks to Antpool and F2Pool's support of it - but their pool appears to not actually be enforcing the new rules, and is now mining invalid blocks << lulzy.
mircea_popescu: gotta explain to the masses 1) how irresponsible it is to empower scammers, whjether they call themselves "pirate", "Gavin" or anything else
ben_vulpes: "Except BIP 66 received 95% support from the relevant group (miners). " << ehuehuehehueeueueueueueeeee
mircea_popescu: ben_vulpes wait, the usg dept of stupid is now on the record that miners don't, after all, decide ?
ben_vulpes: simple rehash of the "let's probe network cohesion strength" fork
ben_vulpes: "The majority of hashing power is mining an invalid chain - it's not going to "win" - they're just wasting their effort." << euheuheuheuheuheheehueheuheuheehu
mircea_popescu: ben_vulpes tjhat's the only position.
ben_vulpes: "However, this also means they're not checking the new BIP66 rule, and are now mining invalid blocks because of it. (another miner happened to create an invalid, non-BIP66 respecting block) If you're not using Bitcoin Core, you might be accepting transactions that won't be on the longest valid chain when all this is fixed." << curious to see if 0.5.whatever comes out on top at the end of this
mircea_popescu: the "lonmger chain" 6.
mircea_popescu: note the orphaned is 8 blocks long
mircea_popescu: https://blockchain.info/blocks << apparently they orphaned everything past bw.com's https://blockchain.info/block-height/363730
mircea_popescu: [ This is the only occurrence ]
asciilifeform: 0.8 sees the phork
mircea_popescu: i guess they eventually settled on pretending it's an accident or what ?
mircea_popescu: anyone actually see these ?
mircea_popescu: none of my nodes verify the imaginary 737-738 blockchain.info claims to exist.
mircea_popescu: asciilifeform listen, seems the chain actually forked.
asciilifeform: is the benefit.
assbot: Logged on 04-07-2015 03:50:47; mircea_popescu: <asciilifeform> ben_vulpes, mod6, mircea_popescu, et al: anybody interested in multiprocessorizing bitcoin sig check ? << i don't see the benefit.
mircea_popescu: <ben_vulpes> of the active users. << well, like a dozen or so people, spent a month in game last month. so like, 2 hrs / day or somesuch
mircea_popescu: news at 12:55, which happens to be the time here.
mircea_popescu: <asciilifeform> ben_vulpes, mod6, mircea_popescu, et al: anybody interested in multiprocessorizing bitcoin sig check ? << i don't see the benefit.
mircea_popescu: <asciilifeform> they'd just all happen to be employees of the house. << if you only divulge winning tickets you get the natural occurence of winners + extras.
mats: so you're saying now is the time to fleece bitpay
punkman: "BIP66 protocol rule changes have gone active in part thanks to Antpool and F2Pool's support of it - but their pool appears to not actually be enforcing the new rules, and is now mining invalid blocks."
decimation: why are they 'inaccurate'
assbot: F2Pool is not properly validating blocks, their fork is winning temporarily. SPV clients and Blockchain.info are inaccurate : Bitcoin ... ( http://bit.ly/1Kzrh5m )
punkman: http://www.reddit.com/r/Bitcoin/comments/3c2cnj/f2pool_is_not_properly_validating_blocks_their/
decimation: the egypt thing appears to have limited support for C++
asciilifeform: mod6: it's 4.something but dies because not found on the gnu ftp
asciilifeform: rather than one which depends on a walk
asciilifeform: tired of looking at the thing through a keyhole
asciilifeform: i want the fucking callgraph
decimation: sure, but you gotta spend time massaging the graph into something legible even at that size
decimation: asciilifeform: I've tried printing such things before, even on a big plotter the results are disappointing
asciilifeform kinda wants a wall-sized callgraph of therealbitcoin
decimation: asciilifeform: maybe the naked blocks are a counterpoint :)
decimation: the big blocks are running nearly a mb or so
decimation: it would sure make it simpler to devise a big asic machine to find hashes if it doesn't need to bother with transactions
decimation: I guess as long as they find a block that matches difficulty it will be valid
asciilifeform: <decimation> weird, the last four blocks had no transactions << betcha somebody's playing with the 'tx selector' thing
decimation: weird, the last four blocks had no transactions
decimation: unfortunately with the effective bandwidth needed by the current blockchain makes global synchronization quite expensive
decimation: at any rate, I'm imagining ways to sew parachutes before they are needed
decimation: imagine three frequencies with three different nodes around the world, all with agreement on headers
decimation: I suppose if a peer is available, you could do that. but then you are only trusting one node's say
asciilifeform: what's wrong with the old-fashioned 'inv' ?
decimation: for the headers thing, I'm imagining a scenario where someone is waiting for a transaction to clear
decimation: oh you just mean the sums
asciilifeform: but will discourage some of the more hamfisted kinds of shitgnomery
decimation: ah, I will read the code
asciilifeform: it is part of the protocol
asciilifeform: that'd be the header, no ?
decimation: it occurs to me that it would be useful to broadcast a digest of a block in addition to the actual block data
asciilifeform: the hash list ?
decimation: one of them on mircea's list
decimation: I'm syncing static node with some random ip, to see if I get the same blkxxxx.dat results
assbot: Logged on 04-07-2015 02:30:46; punkman: updated my mirror with thermonuke branch https://github.com/extempore/real-bitcoin/commits/thermonuke
asciilifeform: http://log.bitcoin-assets.com/?date=04-07-2015#1186324 << very good for folks reading patches. but remember, the patches ~are~ the canonical patches...
mats: DanyAlos: this channel has the flag +s, and so it is unlisted
asciilifeform: and puts a null stub there
asciilifeform: something in 'boost' theoretically can invoke it, or rather - gcc is too dumb to know that it will never happen
asciilifeform: (not in the dns-free variant)
punkman: it complained about getaddrinfo, that's to be expected then?
asciilifeform: punkman: it's an interim thing. still need to shoot glibc in the head
punkman: asciilifeform: it's working, good job on the stator
asciilifeform: also thermal shocks machinery
asciilifeform: decimation: co2 extinguishers, are, interestingly, very expensive (largely on account of their size. you need a great lot of it)
punkman: updated my mirror with thermonuke branch https://github.com/extempore/real-bitcoin/commits/thermonuke