log☇︎
12800+ entries in 0.116s
mp_en_viaje: so today, i took elevator in neue rathaus (it's a thing in munchen, irrelevant specifically). this involved paying a fee. the man i paid it to made change from 52 for 12 by handing me a 20 euro and a 100 euro bill.
mp_en_viaje: that'll pass. now, i propose to you that this is a parameter in the first place, in that it consists of a matter of degrees, however combined. it's entirely quantitative, there's no qualitative portion to it. and it is secret, because her belief is that another wouldn't know which valus the parameter takes, nothing else.
Mocky: ok, a tx characteristic upon which a miner decides not to mine said tx
mp_en_viaje: the "secret mempool", the "claiming saddam has things he doesn't will forward our goals", it's an endless list. remarkably enough, it's a COMPLETE list of contemporary thought.
mp_en_viaje: there can't be, you see, such a thing as a secret parameter. and this applies equally well to the "bombing stationary object" problem touched above, and to the "i will make website ban so and so after so and so" anti-mp historical fetlifish computer "engineering", and to the romcom notion of "deeply personal experience" and so following.
mp_en_viaje: my point was that the contemporary notion of personal identity is actually predicated on a supposed set of secret parameters, something much akin to having it predicated on spherical cubes, or alf's 2's that are also somtimes 1s.
mp_en_viaje: asciilifeform, isn't that quite the mental image btw ? a plutonium lake.
asciilifeform: entirely so, fails in principle, like proggy that needs a 2 that also equals 1 sometimes
asciilifeform: imho patents were a joke from day1, what with 'exam takers' immediately finessing the problem of 'reveal enuff to win if litigated, but not enuff to build clone'
asciilifeform: i dun even have sharp enuff drill to get a picture where they ~have~ a 'notion of personal identity'
asciilifeform: typical homo programmicus living today, i suspect, will live & die without ever encountering a 1980s-style sys where 'understand from first principles' even conceptually seems like a possibility
asciilifeform: the programmers actually built themselves a parallel world where 'science dunwork' , and moved in.
mp_en_viaje: ask any dood you feel like, see if he believes that "secret parameters" are a thing or not.
mp_en_viaje: and i don't mean JUST bitcoin. think of the problem in terms of http://btcbase.org/log/2019-03-31#1906304 : fetlife hires engineer to add "impossible to guess" parameters to its laughingstock of a codebase. mp is unincluded in the ☝︎
mp_en_viaje: a a
asciilifeform: mp_en_viaje: same convention. i was parsing 'c' as positional 'a, b, c'
asciilifeform: still remains the case that good % of what is typically thought of as the protocol, is actually promise ( i.e. it is trivial for a constructed noad to feed 'allcomers' 1 mempool, with stale shit, while he himself mines from entirely other set, that he dun relay to anyone , etc )
asciilifeform: the 'who lives at what ip' is entirely red herring, only the actual 'smoke from the fire' is at all informative, re heuristic 'who is miner' -- who seems to show up with new blox before others ? who takes shortest interval b/w getting a tx and it ends in a block ? etc
asciilifeform: was aboutta add -- from pov of external observer (i.e. has no moles) a 'miner cartelnet' will simply look like 'miner with n public intakes'
mp_en_viaje: there is tension in this question -- inasmuch as declaring your computer TWO computers arbitrarily, and then discussing the safety of the talk over this imaginary boundry is a purely abstract exerise.
asciilifeform: a good % of what those folx do, i suspect can only be explained via 'spam logic'
asciilifeform: also gotta note, i did not posit that it 'works' (in the sense of 'they make bank'), only that 'they think it worx' ( a la spam )
mp_en_viaje: http://btcbase.org/log/2019-04-03#1906661 << i do not believe so ; not in the sense he does not do a lot. ☝︎
a111: Logged on 2019-04-03 17:13 asciilifeform: http://btcbase.org/log/2019-04-03#1906628 << to expand on this: say you issued tx1 where input i and output o1. he sees it, it goes to back of queue, as uninteresting, he does not mine it himself, but does relay to competitors. but if you also issue a and tx2, where input i and output o2, o2 != o1 , ~then~ tx2 goes to front of his queue, as by mining it he can throw caltrop to the competing miners , invalidating their chain
a111: Logged on 2019-04-03 14:58 asciilifeform: so far the closest thing i have to a hypothesis, is that the fattest chinese want you to issue a set of conflicting tx so the thinner ones end up with longer invalidated chains
a111: Logged on 2019-04-03 14:54 asciilifeform: http://btcbase.org/log/2019-04-03#1906616 << this is a well-known effect ( most famously, possibly involved in how 'bitbet' burned down ) and i've personally observed it erry single time i sent coin in past coupla yrs . tx dun move until you send a conflicting tx . why the miners do this -- i still do not know
mp_en_viaje: http://btcbase.org/log/2019-04-03#1906632 << i have no confidence in tghe truetype/whatevr "standards", a lot of actual experience with both empire and alt-empire standards guts has cured me for three lifetimes. ☝︎
mp_en_viaje: but in a coffeehouse it's apparently reglement verboten to have a chair with a back higher than your hips, or padded.
mp_en_viaje: i have nfi how this works, i NEVER saw a car here with wooden bench inside and fuck you. they understand what chairs are, every bank, every god damned nothing-in-particular shop has proper chairs in there. they have furniture shops selling actual chairs.
diana_coman now admits she has purposefully kept away from Munchen for quite a few years now in order to not spoil earlier pleasant recollections of it...
mp_en_viaje: the food was ok, but the service kinda meh. i dunno, they have a problem here it seems with not hirin genough waiters.
mp_en_viaje: ie, the circular connector to frankfurt is frankfurtring, and so this being also a ring but not really connecting to a town... they gave it god's name as it were.
mp_en_viaje: in real civilised places, that community acre is more like a few sqmiles, and they are surrounded by road. in munchen, road is called bavariaring.
mp_en_viaje: in this particular case, they picked a place on fucking bavariaring. how shall i explain what this is... basically every real town has a spot of land kept fallow on purpose to do things like i dunno, let a passin circus pitch tents there, right ? or whatever other large but dirty activity like that, yes.
diana_coman: that being said, I actually had a nice time at various biergartens fwiw - obv, with people "brought from home" as it were so I don't know it had much to do with the *garten itself
mp_en_viaje: diana_coman, very fucking to be expected indeed. the problem is still, almost a decade later, that bitcoin "enthusiasm" is pretty much code for "we'll shock you with how inept we are".
asciilifeform always pictured 'biergarten' as a kindergarten w/ beer
diana_coman: myeah, it could do with a whole lot less mystery
a111: Logged on 2019-04-03 06:36 diana_coman: ersten und originalen Stammtisch - at least it's at a biergarten I hope
mp_en_viaje: http://btcbase.org/log/2019-04-03#1906614 << actually, it was a morongarten. ☝︎
asciilifeform: diana_coman: must admit, currently i have massive hole of mystery where 'how and where does trb tx propagage' oughta be. is why last autumn i sat down & wrote (most of) a fairly gnarly proggy to try & map out node zoology & mempool dynamics. but sadly on hold presently, not enuff hands
asciilifeform: for all i know, even mempool observatory would not necessarily tell you, there is 0 to stop a usg node from relaying one notional mempool to 1 peer and another to another, based on ???
asciilifeform: granted , it is impossible to say what exactly the 9000 garbage 'nodes' are running, it could be a published prb , or just about anyffin else entirely.
asciilifeform: therefore i suspect specifically the miners, for whom by a certain set of prior this idiocy appears +ev
asciilifeform: that being said, none of the prb's asciilifeform dissected to date, appeared to contain any mechanism that would preferentially propagate a tx in an n-spend set vs 'ordinary'
asciilifeform: BingoBoingo: prb ~does~ iirc have a lulzy notion that it can somehow prefer 1 tx in an n-spend tuple ~in mempool~ as 'the real tx' (the protocol offers no such mechanism, of course, tx dun contain anyffin to tie it to e.g. leading block when-issued , only a promisetronic timestamp that can equal whatever the issuer wants)
BingoBoingo: Well, if the USG has a fleet I don't know why they wouldn't prefer to propagate double spends
BingoBoingo: I suspect a large portion of the junk nodes might simply not be relaying transactions.
asciilifeform: ( trivially, generating a tx 'to self' only costs you anyffin if ~someone else~ mines it )
asciilifeform: ... if indeed this is happening, it may also explain some of the churn liquishit that they pad blocks with -- if he cannot find a tx1-tx2 pair in the wild, then... why not generate himself.
asciilifeform: elementary napkin gametheoretical arithm suggests that it is always +ev to mine a tx for which you know a conflicting one has been issued , vs. 1 for which you do not know of one , ~if~ you think you have better connectivity (i.e. more likely to be the 1st to see the split) than other miners ( and i suspect that each of'em thinks he has best connectivity, somehow )
asciilifeform: tho it is not necessarily evidence against it, could simply mean that 'fat miner' lost the race in that particular case to a runner-up
asciilifeform: * a tx2
a111: Logged on 2019-04-03 14:58 asciilifeform: so far the closest thing i have to a hypothesis, is that the fattest chinese want you to issue a set of conflicting tx so the thinner ones end up with longer invalidated chains
asciilifeform: http://btcbase.org/log/2019-04-03#1906628 << to expand on this: say you issued tx1 where input i and output o1. he sees it, it goes to back of queue, as uninteresting, he does not mine it himself, but does relay to competitors. but if you also issue a and tx2, where input i and output o2, o2 != o1 , ~then~ tx2 goes to front of his queue, as by mining it he can throw caltrop to the competing miners , invalidating their chain ☝︎☟︎
feedbot: http://qntra.net/2019/04/english-judge-i-cannot-think-of-any-more-obviously-fundamental-human-right-than-the-right-of-a-man-to-have-sex-with-his-wife/ << Qntra -- English Judge: "I Cannot Think Of Any More Obviously Fundamental Human Right Than The Right Of A Man To Have Sex With His Wife"
asciilifeform: imho the 'machine has 9000 fonts and each 1 is a gnarly proggy for turing-complete ball of 1m+loc c turd' is quite similar to the 'american shop where 9000 brands of toothpaste and each 1 fulla corn syrop' thing
asciilifeform: so far the closest thing i have to a hypothesis, is that the fattest chinese want you to issue a set of conflicting tx so the thinner ones end up with longer invalidated chains ☟︎☟︎
asciilifeform: it's riotously obnoxious and erry time i encounter it , cements in my head the 'mining is a bug'
asciilifeform: http://btcbase.org/log/2019-04-03#1906616 << this is a well-known effect ( most famously, possibly involved in how 'bitbet' burned down ) and i've personally observed it erry single time i sent coin in past coupla yrs . tx dun move until you send a conflicting tx . why the miners do this -- i still do not know ☝︎☟︎
mp_en_viaje: looks like it, but if not, i have a lowen thing spotted also.
diana_coman: ersten und originalen Stammtisch - at least it's at a biergarten I hope ☟︎
mp_en_viaje: what a productive morning.
mp_en_viaje goes to drown trilema.com a little ; #trilema is well drowned for the time being.
mp_en_viaje: meanwhile, in synthetic language, character is the meaning unit, and it is fucking important the function named lil allings with the word www. OR FUCKING ELSE, gott knows we have enough byte errors even without importing a whole new pile of idiocy capable of multiplying them 3x by variable-length that confuses the eye.
a111: Logged on 2016-12-21 18:32 mircea_popescu: because no, words don't "have meanings". your meanings for ANY WORD are a function of ALL THE OTHER WORDS YOU KNOW. which is why my definitions regularily blow out english dictionaries, wikipedia and other sources of "wisdom" out of the water - i know more words, and in this knowledge i know all the words i know ~better~. infinitely and irreproducibly so.
mp_en_viaje: the united kingdom looks more and more like a parody channel on chinese television, "the island of buran" or something.
mp_en_viaje: but at least i found out theresa may has meanwhile solved the problem -- she'll just ask for a new delay in front of some cheap looking, chinese-made union jacks and behind the derpiest drawing of the queen's crest i ever did see.
mp_en_viaje: the thousand+ turned out to be the exact dozen-and-a-half i remember from the 80s, ie EXACTLY http://trilema.com/2017/fake-news-are-just-one-tail-of-the-failed-female-state/ problem.
mp_en_viaje: http://btcbase.org/log/2019-04-02#1906572 << this is true, and good point, most of the time tha tq pops up you're stuck on a reorg or somesuch. trad bitcoin is ~horrible~ re propagation, will not propagate for all sorta bs reasons. ☝︎
a111: Logged on 2019-04-02 19:35 asciilifeform: http://btcbase.org/log/2019-04-02#1906560 << it aint that i've anyffin against 'civilized' typography (tho prolly oughta admit, errything outside of graphical www on my boxen is displayed in exactly fixed-bitmap lettering) . simply gotta point out that 'fit 3 l where 1 w goes' is a luxury of sorts, costs moar moving parts than folx who never tried to implement it are typically aware of
mp_en_viaje: http://btcbase.org/log/2019-04-02#1906570 << no argument that it's expensive (incidentally /me is fully aware, did professional stint in typo, asylum was printed on trad webpress process after a setting by yours truly for example). but by the same hand i'm aware it must be done and is worth doing. ☝︎
a111: Logged on 2019-04-02 18:21 mp_en_viaje: http://btcbase.org/log/2019-04-01#1906511 << problem with fixed width font, in the specific, is that it makes typography (eg, someting like a blog) look like shit, and in the abstract, that it's yet another http://btcbase.org/log/2018-11-04#1869323 : there's no objective reason "all letters 8 px wide", and some pretty great reason w and l don't take same space.
asciilifeform: http://btcbase.org/log/2019-04-02#1906560 << it aint that i've anyffin against 'civilized' typography (tho prolly oughta admit, errything outside of graphical www on my boxen is displayed in exactly fixed-bitmap lettering) . simply gotta point out that 'fit 3 l where 1 w goes' is a luxury of sorts, costs moar moving parts than folx who never tried to implement it are typically aware of ☝︎☟︎
a111: Logged on 2019-04-02 11:03 spyked: noob question re bitcoin: is there an expected time for tx propagation to peers' mempools? details: I'm running a fully synced trb node, connected to trusted republican nodes using -addnode. I've sent 2 txen using sendtoaddress ~3hrs ago and none of them are on e.g. blockchain.info so far
mp_en_viaje: http://btcbase.org/log/2019-04-01#1906511 << problem with fixed width font, in the specific, is that it makes typography (eg, someting like a blog) look like shit, and in the abstract, that it's yet another http://btcbase.org/log/2018-11-04#1869323 : there's no objective reason "all letters 8 px wide", and some pretty great reason w and l don't take same space. ☝︎☝︎☟︎
lobbes just had a moment of "I know I hit this wall before, but forgot the cause"; thankfully logs helped me out once again
asciilifeform: http://btcbase.org/log/2019-04-02#1906538 << believe or not, asciilifeform recently turned up a crate containing several ( i have nfi, thought i gave'em all to danielpbarron ) but loox like he has you covered ☝︎
asciilifeform: iirc mircea_popescu had a custom in-house ??? for same ( also sadly not published anywhere afaik )
asciilifeform: spyked: i have a (sadly unfinished) http://btcbase.org/log/2018-10-18#1863409 proggy , for exactly these cases, on back burner. ☝︎
spyked: noob question re bitcoin: is there an expected time for tx propagation to peers' mempools? details: I'm running a fully synced trb node, connected to trusted republican nodes using -addnode. I've sent 2 txen using sendtoaddress ~3hrs ago and none of them are on e.g. blockchain.info so far ☟︎
feedbot: http://qntra.net/2019/04/f-35-deliveries-to-turkey-suspended-as-turkey-determines-russian-missiles-a-better-value/ << Qntra -- F-35 Deliveries To Turkey Suspended As Turkey Determines Russian Missiles A Better Value
BingoBoingo: I acquired two sample bags of the new type today. They appear substantially stronger, with a more uniform quality than the old type which were potato droppers with inconsistently cut handles.
diana_coman: anyway, fonts are not a big concern atm
diana_coman: iirc jurov had an attempt at a text-only client
asciilifeform: ( a multi-MB and massively crufty mil+ loc of c liquishit, with 0day published erry quarter or so, for ~decade nao )
asciilifeform: indeed. i can think of a coupla potential proggies that call for it, tho ( 'tmsr browser' ; hypothetical replacement for emacs ; etc )
diana_coman: that sounds like a farawy problem still, plenty of headaches before getting there as far as I can see.
asciilifeform: diana_coman: at some pt we're prolly doomed to bake a proper font renderer
a111: Logged on 2018-12-20 00:45 mircea_popescu: this is no proof. but it does inform what is, for all of us, a reasonable suspicion. you think you have reason to suspect bolix is more notable than atari.
a111: Logged on 2019-01-28 01:21 asciilifeform: for extra mindfuck : the primary xtal in the bolix ( valpey-fisher vf155 ) is a <1ppm txco.
asciilifeform: this explains an otherwise strange find of asciilifeform's : that there apparently was a small run of (possibly saphire) 'ivories' for orbit market
asciilifeform: or, e.g., accessed memory via a separate iron sequencer, which queued reads/writes and spread between up to 4 separate memories ( ride 1 while the others busy , making the best of slow '80s drams )
asciilifeform: mircea_popescu: it's 100% enuff for a test harness ( thing even has a handy arse port from which it shits microcode instr. in realtime as it runs )
diana_coman: guess so re derping hq; re fonts, I took that mess of a font out, yes; anyway, this was just on the side while stripping down the "login" window of useless password etc.
mircea_popescu: diana_coman, how could it not be, a buncha idle virgin bois derping about "convincing pretense" as to adulthood and whatnot. if it weren't boring as fuck the problem of hiring schoolteachers would be merely difficult, rather than plain impossible.
mircea_popescu: asciilifeform, o wow, that sums to almost a workable test harness!
asciilifeform: at some pt, possibly earlier, will re-type the thing for publication, but i dun have a nicoleci , so may take a while .
asciilifeform: will have a torture board baked at some pt, with socket . but prolly not till later summer at earliest.
asciilifeform: actually i think it was from a now-forgotten crackpot us candidate