mircea_popescu: http://log.bitcoin-assets.com/?date=04-07-2015#1187155 << this isn't the right aproach. can you think of any reason TO include it ?
gernika: Will do. Need to sign up for the mailing list.
ben_vulpes: so's this phr0k the nail in 20mb blocks' coffin?
punkman: shade's good for the bunnies too, such synergy
punkman: used to have one of these on 2nd floor roof, vine roots at ground floor
cazalla: punkman, that's lovely, just wish i had the room :\
cazalla: punkman, no pergola but we have a clothesline i could commandeer but doubt the missus would like that heh
assbot: Logged on 04-07-2015 22:03:31; mircea_popescu: 3. in order to avoid the situation that happened tonight, a supposed "voting" mechanism was deployed where people switched an irrelevant byte in the blocks they mined, to go from 2 to 3, to indicate they are ready and willing to follow the new rules.
cazalla: might've been the uv damage that killed em, i don't really know other than summer 2013 provide no tomatos
mircea_popescu: otherwise, stews it. which...
mircea_popescu: a gallon of water makes half hour's mist, protects the plant if there's any air draft at all
mircea_popescu: people in heatwave places have these fine nozzles
cazalla: mircea_popescu, nope, good water in the evening and morning but the 4-5 days of 40-44 degree heat toppled em
punkman: cazalla: I fucked the up with too much sunlight many times.
mircea_popescu: the one i originally advertised for instance.
assbot: Logged on 04-07-2015 22:58:41; punkman: I tried heirloom tomatos for a bit. I now know why everyone's using the same boring/bland strain.
cazalla: http://log.bitcoin-assets.com/?date=04-07-2015#1187574 <<< never had a problem with them down under.. 4-5 day heat wave is the only thing that fucked my tomatoes 2 seasons ago
punkman: yeah but after reorg it throws the things away right? you'd have to shitblock at just the right time
mircea_popescu: almost as if someone was holding off hitting the muppets over the head with a 2x4 until b-a crowd had its ducks lined
mircea_popescu: this is exactly the sortt of situation the entire deterministic thing he was crowing about is meant for.
mircea_popescu: punkman well now we have it! alf wrote the shitblock!
punkman: we need something that stores these forks for inspection and replaying. I think ben_vulpes had a thread about that
mircea_popescu: i still dunno why they do it. i've not yet managed to unstick a stuck one from last night
mircea_popescu: mod6 the blockchain is not compromised.
mircea_popescu: hanbot it still seems thermodynamically impossible.
punkman: I tried heirloom tomatos for a bit. I now know why everyone's using the same boring/bland strain.
punkman: for a short time I believed in the mythical farmer that actually knows what he's doing. But after watching the experts debug these things, my only conclusion is they sometimes get lucky and that's the best I could hope for.
mircea_popescu: the town was fully of blooming magnolias and i never saw a single artichoke plant outside of my garden
punkman: I get very frustrated at things like "o hey we've sprayed 12 different solutions at it and the damn lemon tree still has fungus"
mircea_popescu: <punkman> wait until you try to debug plants << he has a point there. tho in fairness iirc alf debugged mice at some point.
mircea_popescu: there's no reason to expect etc etc.
mircea_popescu: and yes, most men died before seeing the thing they were working for. historically, to date.
mircea_popescu: there is no reason to expect any better in this new field. it is, after all, a field.
mircea_popescu: asciilifeform just sayin, humanity was built by originally trudging the muck with no tools.
punkman: surprised we aren't seeing more of these
ascii_modem: the male ones at least
mircea_popescu: yea, i get it, may mean that this makes your fate be to muddle in the dirt, piling up data for your kid, who maybe might do it.
mircea_popescu: until we know exactly wtf to do about relayers, about miners now, there is no point. the prototype's still prototypin'.
mircea_popescu: but this situation is why im so cold on your ada idea. sure, you go put all the god damned effort into writing a fully correct, fully specified implementation oif an... broken idea.
ascii_modem: one one hand, i don't give a shit, yes. on the other, a human turd pressed a button, my nodes are halted, and there is no guarantee that they don't have a bottomless magazine of these
mircea_popescu: punkman right. hence all the "We don't give a shit" verbiage here.
ascii_modem: so far there is no structure remotely enough to take up the weight if the duct tape is ripped out
mircea_popescu: atm the split is .999 - .001 or somesuch. this will change monotoniucall;y the other way
mircea_popescu: really the % of income "just known" vs "information acquired" is the proper measure.
ascii_modem: also touches on the mempool thread
mircea_popescu: hard to convey what the changes in dynamic such a thing brings.
mircea_popescu: if as much as 66% comes that way and the other 33% has to be found out about... o baby.
mircea_popescu: tomorrow, they will not KNOW immediately through act of god where that 25 btc comes from
mircea_popescu: ascii_modem more specifically put : currently everyone is mining with a "god given" known tx of 25 btc + whatever they pick off the ground.
mircea_popescu: but when that begins all this may fucking collapse under the weight of... if money's the incentive to mine mining will not exactly allign with "securing the network"
ascii_modem: this is the exam thread all over again
mircea_popescu: or else by fucking up other people's miners. i ordered a listing of "all the ways you could exploit a miner with crafted comms for a hash advantage"
mircea_popescu: specifically : atm there is an arbitrage open, where you can either make money by "brute forcing" it ie hashing,
mircea_popescu: anyway, the main sufferance in my head atm is that reading the chinese stuff (in translation) clarifies in my head an objection that may well be a second major flaw to the protocol, after the "relay nodes gotta do it for the glory" : its altogether unclear a purely financial incentive is the correct solution for miners.
punkman: in other news, ATMs getting ddos'd even at 1am. rejects transaction, spits out card, and you gotta start over. perhaps intentional rate-limiting.
ascii_modem: so far i'm not sure any of you apprehend just what a steaming turd the thing is
mircea_popescu: in which case they'll prolly change it rather than start from the original. unles it's VERY dumb.
punkman: I think we need to put a lot more comments on the code, kinda clunky to do through patches to ML though
ascii_modem: there are different schools of thought re how to furnish a room
ascii_modem: right now we got an anthill in the living room
mircea_popescu: gotta make the fambly unit work.
mircea_popescu: gotta make the fambly unit work alfie.
mircea_popescu: this is a major problem, and the fact we're looking into it one of the few things giving me solace and better rest at night
ascii_modem: how many other ways for fuckers to wedge us
mircea_popescu: anyway, i am very much amused at the way discussion with kakobrekla turned out. 600 hours "oh miners will just so and so " "nobody cares what miners do" ; 1800 "oh miners just so and so'd!!1" "and it turns out nobody cared".
trinque: cool, just wanted to post the link
mircea_popescu: then perforce ?
ascii_modem: only the log
mircea_popescu: i dun think i ever used the original.
mircea_popescu: the original point of this thing i run here was to give all comers the historical chain as is in my custody.
mircea_popescu: now, since you're synced to my level, and you have a full validating node, i see no reason to not open the listen port. do you ?
ascii_modem: gonna plug in the one on dulap when i get home
mircea_popescu: specifically, what happens if you put the new chain into a node
mircea_popescu: the only problem is that the clients ain't reorging properly, which needs looking into/
ascii_modem: my node is dead in the water
mircea_popescu: well, just let the chain reorg an' carry on, imo.
mircea_popescu: ascii_modem where were the daily forks b/w january (?) and now ? << costs a little to make one, nobody cared enough. (also lots of manual intervention and general ducttaping at the miner and relayer level - bitcoin is quickly becoming an excellent makework tool, keeping idiots both employed and in a delusion of importance)
mircea_popescu: none of this in any way invalidates anything about the new chain, nor is it a point of concern that i can distinguish
mircea_popescu: this utterly proves that a) the BIP system provides no benefit and should be disused ; b) the peoiple involved with all this nonsense should be muted ; c) voting does not work. nothing here is novel.
mircea_popescu: 4. the people who voted did so as all people voting ever do : blankly, mouth only. then, someone mined a v2 block on top of a v3 block (which to them look identical) and the supposed 95% majority that promised by that vote to distinguish failed to distinguish. well, half of them or so. the other half distinguished
mircea_popescu: and we'll never see an odd hash block for as long as they keep at it.
ascii_modem: where were the daily forks b/w january (?) and now ?
mircea_popescu: this, again, has nothing to do with us. if tomorrow miners decide as a social game to never communicate any blocks they find that hash to odd, it's THEIR problem
mircea_popescu: 3. in order to avoid the situation that happened tonight, a supposed "voting" mechanism was deployed where people switched an irrelevant byte in the blocks they mined, to go from 2 to 3, to indicate they are ready and willing to follow the new rules.
mircea_popescu: it'd have been ideal to not import the satoshism in question into bitcoin, but before our time.
mircea_popescu: this is neither a bad idea nor in any sense avoidable. because, again, without it we'll fork daily.
mircea_popescu: 2. a voluntary, user-enforced change to the protocol was proposed (iirc by petertodd) where only DER sigs are accepted anymore.
mircea_popescu: so 1. it was observed that the idiots running openssl, who for historical reasons are actually wholesale pulled into bitcoin, decided to fuck up the way they do signatures, which is to say, make items signed come out in rnadom shapes. this is obviously very dangerous for bitcoin
mircea_popescu: lemme lay out the issue here in detail, i think it's mebbe getting fudged.
mircea_popescu: this, obviously, would also be where to bury who knows what. but if it's there, nobody's found it yet that i know
mircea_popescu: the difference between v3 and v2 is , as best anyone can determine, actual fucking validation for the openssl nonsense. hardly avoidable.
mircea_popescu: atm the situation is that block 363730 is forked. one chain, 6 blocks long, proceeds atop a v2 block. the current main chain proceeds from 363730 on v3 blocks.
ascii_modem: innit in there now? or am i confused
mircea_popescu: i suspect this obviousness is a fruit of your eye, not of the thing you're lookling on. like people perceive souls in cats' eyes.
mircea_popescu: "the enemy" is barely cogent enough to be enmitous at all.
ascii_modem: the enemy accomplished his objective, which probably had something to do with discrediting chinese mines
mircea_popescu: the dead don't know they're dead ; the ridiculous don't know they're ridiculous. the world is well constructed for one's amusement.
mircea_popescu: but whatever, phantomcircuit is still the guy who ran "a security companmy" that oversaw 3 bitcoinica thefts, and still the guy who hacked a python one liner to send every customer an email with the full list of customer emails
mircea_popescu: there are some people that go around introducing themseves as "bitcoin devs" who just spectacularly failed the least controversial soft fork to date.
mircea_popescu: there is no physical punishment for being extremely ridiculous.
mircea_popescu: apparently in the world we live in, the best way to ensure nobody reads something is to publish it as code anyway
jurov: or b1ockchain.info or other similar derivatives
mircea_popescu: also lulzy for all the people using blockchain.info to validate txn.
pete_dushenski: mats: aha thanks. just goes to show that there are shades of grey in the braindead morass of veecee moolah, not all are created equal.
mircea_popescu: alternatively they could just shut down and stfu already.
mod6: <+mircea_popescu> here's what's the idea : after botching the soft fork, it seems illogical that the entire bip system should continue at all. << Very much agreed. If someone wants to change something they should write to The Foundation's btc-dev mailing list and submit a patch. And if they can't because they're not in the WoT, well they're not in the WoT. They can make a personal appeal here in person.
gribble: The operation succeeded.
pete_dushenski: ;;later tell travispatron you sucked before, you suck now, and you will always suck. enjoy the periphery, derp.
mircea_popescu: for the unsubstantiated delusionsm of a few random geeks, working to make the world believe they're knowledgeable and important, bitcoin hiccups every so often.
pete_dushenski: mircea_popescu: it's the broken window/car accident economy
mircea_popescu: right. splendid. how about go home and the problem goes away with you.
gribble: The operation succeeded.
mircea_popescu: this is EXACTLY how usg works throughout. first, cause black people to be poor and depenedent. then cause them to be violent. then go on press conferences about how "they're doing things about black violence"
mircea_popescu: then they go on reddit to "explain" why what they fucked up isn't their fault, and "what really happened" , and collect cheap applause from the celenterates therein living.
mircea_popescu: an abundance of instances can be documented where they fucked things up
mircea_popescu: there can not be a single instance documented where they did anything useful to any degree or in any sense
mircea_popescu: these are people who have been so far causing nothing but problems for bitcoin.
mircea_popescu: here's what's the idea : after botching the soft fork, it seems illogical that the entire bip system should continue at all.
pete_dushenski: so the numbers are meaningless ?
pete_dushenski: jurov: perhaps ima confoozed, but don't orphaned blocks have to be replaced with valid blocks in the main chain ?
jurov: no it reports them as orphaned
mircea_popescu: are they ? i've not looked.
pete_dushenski: mircea_popescu: ok, but assuming those 6 invalid orphan blocks are off on their own, how is it that some sites (eg. bc.i) are reporting that this foundation of sand is still being built on ?
assbot: Logged on 04-07-2015 04:15:35; ben_vulpes: simple rehash of the "let's probe network cohesion strength" fork
mircea_popescu: http://log.bitcoin-assets.com/?date=04-07-2015#1186531 << the clients do seem to run into similar problems eh.
mircea_popescu: then again plenty of people do not even wait for one.
pete_dushenski: mircea_popescu: you think there was a big transaction that got double-spent ? or...
mircea_popescu: incidentally, am i the only one finding it a wee bit suspicious that the orphan chain ended up 6 blocks long, ie, exactly the size normally deemed "safe" ?
trinque: danielpbarron: s-stop the surveillance!
trinque: jurov: according to this you lose a lot of detail, i.e only get statistics for the line of the macro call http://codingfreak.blogspot.com/2009/02/gcov-analyzing-code-produced-with-gcc.html
trinque: mircea_popescu: I'm sure I'm about to see a nightmare when I pull the deps into the gcov thinger
mircea_popescu: <trinque> might be able to just rip those bits off from the dep, then cut off the dep <<< ahahaha you.
mircea_popescu: alternatively, feed it the last few blocks through the mechanism discussed that alf created recently i guess.
mircea_popescu: n ProcessMessage()\n\nProcessMessage(block, 947244 bytes) FAILED\n" which is the confirmation. then increase the number o' locks.
mircea_popescu: trinque, asciilifeform, wimc : if client fails to progress past 363736, the likely culprit is the db locks limit. look in derp.log (assuming standard derp) for "REORGANIZE\nREORGANIZE: Disconnect 6 blocks; 000000000000000006a3..000000000000000013fe\nREORGANIZE: Connect 7 blocks; 000000000000000006a3..0000000000000000014e\n\n\n************************\nEXCEPTION: 11DbException\nDb::get: Cannot allocate memory\nbitcoin i
assbot: Logged on 15-02-2015 22:34:28; asciilifeform: 2) tearing out 'boost' while keeping the project in cpp will turn it into an unreadable morass of crud that makes the existing turd look like the finest sausage
trinque: anyhow this is what I wanna see with the gcov thing
jurov: trinque and anyone other: mandatory reading
trinque: I was surprised to find that bitcoin doesn't actually look that *big* compared to other C++ projects I've seen
trinque: speaking of which, I'm adding the coverage flag to the deps too, and will regen that gcovr html thereafter
assbot: Shout out to the American Core Devs that are spending their 4th ensuring the health of the blockchain : Bitcoin ... ( http://bit.ly/1TcE1R7 )
punkman: http://www.reddit.com/r/Bitcoin/comments/3c404r/shout_out_to_the_american_core_devs_that_are/
mircea_popescu: they should. eventually.
trinque: ass-goblins in the intertubes
mircea_popescu: asciilifeform the dispute was 730/736
trinque: with the whip iirc
williamdunne: jurov: Quite a bit further IIRC
shinohai: @ mod6 yes, other one still not built yet
jurov: lol, that's actually true that dick continues deeper in the body
trinque: you have to measure inflation from the balls!
trinque: I'm sure they used to mean something else, but we're in bizarro land where the length of a meter changes day to day
trinque: williamdunne: I tend to take stock market prices as inflation indicators these days
williamdunne: http://www.followingthetrend.com/wp-content/uploads/2015/06/Figure-15-1-Benchmarking-Randomness.png
asciilifeform out to meatspace for most of the remaining day
asciilifeform: mod6: i've been synced for ~1hour and it looks like it's firmly stuck there
mod6: <+phf> apparently gcc doesn't always include all the necessary pthread bits (not just openbsd but other unixes), which results in segfault on launch << on my obsd 5.6 on x86-64 that's the same problem i kept running into; segfault at execution time.
gribble: The operation succeeded.
mod6: <+asciilifeform> phf: you may be the first to achieve openbsd build! consider posting recipe << please do!
asciilifeform: the miners ~are~ zombies
gribble: The operation succeeded.
asciilifeform: ;;later tell kakobrekla my current understanding of the ph0rk situation is that your hypothesis re: miners being dumb as bricks is essentially correct...
punkman: heh "I shook hands with the man I did most of the talking with and they took off... I was a little shooken up but realized I made a big mistake! I forgot to shake the black police detective's hand... shit!!! "
gribble: The operation succeeded.
asciilifeform: ;;later tell mircea_popescu i built stator on dulap just now. unchanged from public recipe other than the 'realpath' thing (mentioned earlier.) same process ought to work on your other centos boxes. it will run as a public node as soon as i shove a blockchain in there.
phf: asciilifeform: i have a small backlog of things i want to send to the list, i'll wrap it up sometime over the weekend
phf: apparently gcc doesn't always include all the necessary pthread bits (not just openbsd but other unixes), which results in segfault on launch
phf: asciilifeform: it pretty much builds out of the box. the only tricky part is the " -Wl,--whole-archive -lpthread -Wl,--no-whole-archive" business
asciilifeform: send it to the ml ?
asciilifeform: phf: you may be the first to achieve openbsd build! consider posting recipe
asciilifeform: 'nd that they sell over 30k student emails for $20, which I had bought;' << l337 sp4mz0r ?
punkman: asciilifeform: what do you think about making no-verifysig-skip the default for next release?
asciilifeform: as temporary workaround, replace the realpath thing (whole right side of equal sign) with the absolute path of 'ourlibs'
asciilifeform: which is used in the script
asciilifeform: 363673... almost there
asciilifeform: the buggers like to monkey with versioning? fine - let's nail their ears to their heads.
asciilifeform: and from general principle, there is no reason ~not~ to make life maximally difficult for the enemy if this costs us nothing
assbot: Logged on 04-07-2015 07:21:18; 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".
cazalla: now and then i've thought of the few teacher's aids that would come for a school term as part of their studies.. they were so young and pretty but would be old bags now
punkman: "Two other male students, both also 17 years old, came forward as victims after Fichter was arrested." << lol assholes
cazalla: wow GG America, well and truly lost the plot now http://myfox8.com/2015/07/03/florida-teacher-who-had-sex-with-three-teens-sentenced-to-22-years-in-prison/
punkman: in other news, stator (with openssl1.0.1g) patched to not skip VerifySignature, has not barfed up to block 146k (previous attempt barfed earlier than that)
shinohai: It appears f2p pool "almost" mined them out xD
kakobrekla: fees not large enough to offset the finite speed of light
mircea_popescu: check out all the miners that were mining off thin air, incidentally. because the bandwidth doubled so much.
kakobrekla: so if nothing else, cpu mining code needs to be updated in the ref implementation now or what
assbot: If you are using any wallet other than Bitcoin Core 0.10.x or 0.9.5 (or something backed by one of those versions), then you should not trust incoming transactions until they have ~30 confirmations. : Bitcoin ... ( http://bit.ly/1G0jGVG )
punkman: "We are using 10.2 but got this message anyway: "Your node software is out of date and may accept an invalid blockchain fork. Do not trust confirmation."" "There was a mistake made and an alert that showed up on v0.10.2 was sent out accidentally; should be now fixed."
assbot: Logged on 04-07-2015 04:30:43; mircea_popescu: Uh oh all these block explorers just flipped over to the invalid chain:
mircea_popescu: he's still alive. never trust the living. for all you know he ends up like donald trump, doing reality shows in his 70s
midnightmagic: (but Perelman's a safe bet because he's a recluse and there's no possibility of that image ever being shattered)
mircea_popescu: "and then we came THIS CLOSE to selling out to amazon for 5 trtillion"
mircea_popescu: to most everyone else, gavin's just another washed up website designer with a great story from five years ago.
midnightmagic: No, not to me. He wears the halo that makes large audiences cheer for him instead of throwing spitballs, when he says he should revoke everyone's commit access and be a dictator.
mircea_popescu: he only wears the halo ~to you~. because you're a particular sort of inner build, and because you ate a particular diet.
midnightmagic: Dude wears the Halo. The other scammers don't matter so much. And if I really cared, I'd be doing something other than flicking out belly button lint in the shower. I'm just curious to know why *you guys* stopped.
mircea_popescu: the legions of hell are truely an infinite headcount. why do you care so much about some particular schmucks in the chorus ? why not that tv scammer dude while you're at it, what was his name
mircea_popescu: heck, kenna and the armandi fellow are prolly due for a novel attempt too
midnightmagic: oh, the litlicence guy.
mircea_popescu: eh, they nibble discreetely at teh periphery.
midnightmagic: Because they haven't stopped yet. Hearn's submitting pullreqs he's going to use as propaganda to push people to -XT, and Gavin's going through the bip/pullreq motions when he knows he's going to be voted down. Again.
mircea_popescu: why ? point's well made, everyone's happy to be part of bitcoin by standing on borrowed intelligence. let them.
assbot: Logged on 04-07-2015 06:38:25; midnightmagic: The Satoshi-Halo-Wearers.
assbot: Logged on 04-07-2015 06:33:28; decimation: why not solidify the questionable openssl code first, before lightly restricting certain signature forms?
assbot: Logged on 04-07-2015 06:24:27; 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.
assbot: Logged on 04-07-2015 06:18:37; 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.
assbot: Logged on 04-07-2015 06:14:44; midnightmagic: because, like I said, I think some people in there appear to be adding exploitable code with absurd frequency.
mircea_popescu: http://log.bitcoin-assets.com/?date=04-07-2015#1186857 << not just there. which is why the entire signed patches business.
mircea_popescu: http://log.bitcoin-assets.com/?date=04-07-2015#1186854 << it's a complex mess. satoshi generally wrote stuff that didn't actually work for the intended purpose. it is very far from inconceivable we shall in a year discover a fundamental bug in any of a dozen dozens different essential components.
mircea_popescu: you familiar with my observation re alf's "folk with brains are useless - they want to use the brains" that in terms of "optimal impact", you're always better off waiting ?
midnightmagic: And sipa, it seems, agrees with you, hence the existence of libsecpblah
midnightmagic: I meant to suggest that the amount of effort of doing it oneself compared with the projected risk of it happening while someone else is working on it.. which option is less expensive/reliable?
mircea_popescu: it is in point of fact better to have a static build that specifically ennumerates blocks prior to height X and then proceeds from there, than to have the present situation.
midnightmagic: Almost like a sort of mass-ennui/retirement mentality. Tired of working for something they don't believe in anymore.
assbot: Logged on 04-07-2015 06:11:21; 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?
mircea_popescu: http://log.bitcoin-assets.com/?date=04-07-2015#1186847 << while this problem exists, i think the case that it is strictly preferable to attempt building a functional system than to bemoan one's fate of having been born to parents this poor, stupid and useless needs not be further pressed.
mircea_popescu: whose great-grand parents noticed, while taking the upper-middle class mandatory "tour of europe", that the italians similarly etc etc.
mircea_popescu: similarly is the case of "americans".
mircea_popescu: an exquisitely african thing this, as late as 2000 one could notice that all the egyptians seem willing to do is stand in front of the pyramids with their chests pushed out, or else gesturing importantly. meanwhile... the people who built those things don't look anything like the arab mongrels currently populating the place, if extant statues are to be believed.
assbot: Logged on 04-07-2015 06:06:56; decimation: why isn't anyone seriously attempting to extract the open-ssl code paths used by bitcoin?
mircea_popescu: http://log.bitcoin-assets.com/?date=04-07-2015#1186837 << for the same reason no one's attempting any other flavour of serious works anywhere in the decaying west, on any topic, for any reason. too busy posturing in front of things.
mircea_popescu: pure spaghetti mess, wherein no soul can tell where the pasta ends and the cook's hairs begin