Forum logs for 20 Apr 2016

Monday, 16 March, Year 12 d.Tr. | Author:
BingoBoingo: And yet somehow my links ended up on an alf list [00:01]
mircea_popescu: yeah well not everyone is into everything. [00:04]
* BingoBoingo kind of wants, shame it isn't a creeping red fescue http://naturalsociety.com/new-gmo-approved-grass-may-infiltrate-lawn-near/ [00:17]
trinque: $up deedbot- [00:48]
deedbot: deedbot- voiced for 30 minutes. [00:49]
trinque: davout │ deedbot's dead? << seems to be having some trouble with whatever's going on with freenode these days. [01:15]
trinque: but I have the new one eating deeds a bit more work and deedbot- will not be needed anymore [01:16]
punkman: https://www.idontplaydarts.com/2016/04/detecting-curl-pipe-bash-server-side/ [06:07]
mircea_popescu: $up referredbyloper_ [06:38]
deedbot: referredbyloper_ voiced for 30 minutes. [06:38]
mircea_popescu: http://archive.is/Nn8ZM << am i the only one that thinks it a little bizarre someone'd pay 80 btc for a site and then do exactly nothing for the next coupla weeks ?! [06:40]
mircea_popescu: punkman isn't all that buffer-filling shenanigans going to give the show away anyway ? [06:42]
mircea_popescu: "oh look, this 16kb script download takes > 1mb hurrr" [06:43]
shinohai: Good morning #trilema [07:05]
mircea_popescu: hola. [07:07]
shinohai: Trying to work up a hilarious angle on the 123-reg.co.uk lulz this morning. [07:13]
mircea_popescu: *thumbsup* [07:26]
shinohai: This being the UK I am willing to bet the server admin was a Pajeet. [07:31]
mircea_popescu: whassat ? [07:35]
punkman: mircea_popescu: give the show away to who? the guy piping to bash? [07:38]
mircea_popescu: anyone watching the system. [07:39]
mircea_popescu: "gee i wonder why a fresh install of X takes 2gb when the package is 200mb ?" [07:39]
punkman: also makes me think about what would happen in "underhanded bash install script contest". change a couple chars, pwn target? [07:39]
mircea_popescu: possibly. [07:40]
PeterL: mircea_popescu did you clear the orderbook on purpose? [08:39]
mircea_popescu: PeterL http://trilema.com/forum-logs-for-19-apr-2016#2080463 [08:48]
a111: Logged on 2016-04-20 03:11 mircea_popescu: meh. mpex managed to flush the book. everyone is kindly asked to reintroduce whatever orders the had. apologies for the inconvenience. [08:48]
PeterL: so, with the diminished selection of assets listed on mpex, are you going to lower the account fee? [08:49]
solrodar: PeterL: I doubt there have been any real account fees paid for years [09:19]
solrodar: my guess is the whole thing's now just a cover for money laundering [09:19]
shinohai: >.> [09:26]
solrodar: did nobody else find it implausible that he claimed to be able to find at least one person every month willing to spend $20,000 for access to an exchange where nothing was traded but shares of the exchange itself? [09:32]
solrodar: it looked like the world's first unintentional ponzi scheme [09:33]
solrodar: and then a few weeks after someone notices these alleged payments aren't on the blockchain [09:33]
solrodar: "we're going private" [09:33]
shinohai: http://i.imgur.com/AbeOik6.jpg [09:46]
PeterL: solrodar so what was it, mp buying his own product to increase volume? [09:53]
solrodar: could be some of that as well [09:56]
danielpbarron: $v F1F7155A2B45099761F3C62B891DA26ADF8C691BDED7D4798A407B7227F7F2FB [09:56]
deedbot: danielpbarron rated solrodar -1 << I find it implausible that this guy has anything interesting to say [09:56]
solrodar: but what I was thinking is, he's got some private arrangement whereby people pay him "account fees" and then get most of that money back in some other manner [09:57]
PeterL: oh, no, solrodar has iritated the priest of the cult of mircea_popescuism [09:57]
solrodar: after it passes through his "payment network" [09:58]
* solrodar afk [09:58]
PeterL: could be all sorts of things [09:58]
mod6: mod6 here [10:21]
mod6: still beat down status [10:21]
* shinohai passes mod6 some "herbal remedy" <\\V\\V>ڪ [10:27]
punkman: http://www.bloomberg.com/news/articles/2016-04-19/intel-cuts-12-000-jobs-forecast-misses-as-pc-blight-takes-toll [10:35]
shinohai: later tell BingoBoingo http://ix.io/xap [10:45]
gribble: The operation succeeded. [10:45]
mod6: hah, "hey someone put some unicode in this j.." [10:50]
mats: solrodar: 'i don't understand this financial instrument so the operator must be laundering money' [10:56]
mats: you think he paid out 250k btc from account fees? [10:57]
mats: ??? [10:57]
PeterL: did he file the AML paperwork for all those account fees? No, therefore must be money laundering!!! [10:58]
mats: and what are you even talking about, re: 'these alleged payments aren't on the blockchain' ? [11:02]
mats: i assure you, i am a real person, and have received many withdrawals from mpex over the last two years. that are on the blockchain. [11:03]
PeterL: I think he meant the 50 btc account creation fee [11:03]
phf: mats: the "mp is a scammer" crowd is riding on an 18btc bitbet incident, and will continue to do so no matter what happens. remember how everyone was losing their shit, "all bitbet money is gone!!1" [11:05]
shinohai: Probably just "I didn't get named BitBet receiver" butthurt. [11:06]
phf: meanwhile couple of months ago mp explicitly said that he's starting a process of divesting from bitcoin, that was before bitbet, before ~anybody~ had voiced any kind of objections to anything [11:07]
phf: that decision had causes that in turn took several months to form and manifest as these things usually go [11:07]
phf: so now idiots are going to look at the past couple of months and point at patterns, "where's the 50btc account creations!" [11:08]
phf: i'm sure a person who's ready to drop 50btc on a trading account will do ~minimal~ amount of due diligence to confirm that mpex is a viable trading platform ~at the moment~ [11:09]
phf: i actually did some maths and was ready to drop 50btc on an mpex account, but held off for the main reason that i don't see a solution (and there wasn't one found) to minor collusion problem. that conversation happened ~last october~ (or whereabouts), it was framed as a hypothetical, but it was a long precursor to chinese miner conferences (that in turn was serious grounds to suspect potential cartel) and all that other stuff. [11:16]
phf: so whether or not mp is a scammer, the greater framing of this conversation, makes certain allegations not only irrelevant, but also inane [11:19]
phf: ffs. [11:20]
solrodar: shinohai: not butthurt in the slightest, davout was clearly the better candidate [11:20]
mats: it got a little harder when mpex stopped offering options, but s.mpoe had a lot of volume for a long time precisely because it was valuable and served as a symbol for the exchange [11:21]
solrodar: $v AC0E2634089EEF76EBA906B010CE7E1F1EC0232B09CBD110CE9C2FBA4241D24A [11:22]
deedbot: solrodar rated davout 1 << competently completed BitBet receivership in only two weeks (tm) [11:22]
mats: i also don't know what the fee has to do with anything [11:23]
mats: 'its so high nobody could possibly ever want to pay for it unless it was to launder money' ??? [11:23]
phf: mats: it's the muckraking from b-a spilling over to here [11:24]
mats: there is nothing unbelievable about a fee like that being tenable when it is 1) an exchange that has never been hacked 2) operated by someone with a serious presence in the wot [11:25]
solrodar: 3) and on which practically nothing was traded except the shares of the exchange itself [11:26]
solrodar: whose main source of income was new memberships [11:26]
solrodar: see the circle here? [11:26]
mats: why does 3 even matter? [11:27]
solrodar: because there's no actual economic activity in the system [11:27]
mats: you think folks wouldn't pay to trade in the stock of a single company? [11:27]
mats: because you can't possibly conceive of a reason why ~you~ would do it, the activity must be fraudulent? [11:27]
solrodar: never said fraudulent [11:27]
solrodar: evidently investors did actually get their money back [11:28]
mats: you used the phrase money laundering. [11:28]
solrodar: only a USG agent would assume that was a bad thing :P [11:28]
phf: solrodar: aren't you also implying mpex is a ponzi scheme? [11:29]
mats: only a poor and unimaginative fella would say the things you did this morning [11:29]
solrodar: it would have been a ponzi scheme if it was closed [11:30]
solrodar: but you know how ponzi schemes usually claim to have some undisclosed source of income [11:30]
solrodar: my suggestion is that that was actually the case [11:30]
solrodar: income from people "buying accounts" when they weren't really buying accounts [11:31]
solrodar: but a tumbling service [11:32]
phf: heh that's actually a good idea [11:33]
phf: but membership purchase is the core of ponzi, where's undisclosed source of income is the diversion (like herbalife sales) [11:34]
phf: can also have pure ponzi, like mmm [11:34]
solrodar: yeah, so I'm not saying it was a ponzi, or any other kind of scam, just that he wasn't being entirely up-front about what was going on [11:45]
solrodar: and he admitted as much when he said "By now its opacity verges on the ridiculous." [11:47]
phf: solrodar: you're reading tea leaves [11:47]
solrodar: maybe [11:47]
solrodar: so you were considering buying an mpex account in october, what would you have done with it? [11:49]
phf: nah, i wasn't considering in october, by october i've decided i'm not going to [11:49]
solrodar: I'm interested in how the maths could have made sense [11:51]
solrodar: were you hoping to get your money back from day-trading, dividends, long-term investments, or what? [11:51]
phf: long-term investment [11:51]
phf: mpex value hinges on long term viability of bitcoin as a way to store and accumulate wealth [11:52]
phf: hence all the talk of berkshire, intentional cultivation of "bitcoin company" values, ideological separation from fiat, tmsr, etc. [11:52]
phf: from that perspective 50btc is just that, if there's a return on it 5 years from now, and continued return after that, you get +ev. more companies get listed today or in a year, list own company, etc. [11:52]
solrodar: and all those things mean the share price of mpex itself would rise [11:53]
solrodar: fair enough [11:53]
phf: from that perspective it's not strange at all that s.bbet was "zero assets", or s.nsa doesn't pay for ascii's maryland house [12:20]
phf: both are fledgling processes, self contained by implicit tmsr rules. as economic activities they could never the less grow and gradually eat more of the outside world [12:20]
phf: enough of these self contained but intertwined economic activities (s.nsa and s.bbet hosted on s.host selling parts to ...) and you get something interesting [12:20]
solrodar: you would [12:26]
solrodar: but when I first came across it a year or so ago, I got the impression that 2012 was the what bliss in that dawn to be alive [12:27]
solrodar: and it's been winding down ever since [12:27]
davout: and here goes bitbet: https://blockchain.info/fr/tx/a0f12a33413ec3a38e3ae931068febec133b083d72810fc586a8b1c5f0d7922a [12:36]
davout: "Estimated BTC Transacted: 0.0001 BTC" <<< win [12:39]
ben_vulpes: omg not all at once davout [13:07]
deedbot: [Qntra] "Spring Into Money" Exchange Hacking Event Continues - http://qntra.net/2016/04/spring-into-money-exchange-hacking-event-continues/ [13:22]
BingoBoingo: Sorry shinohai fxd [13:26]
shinohai: Awesum fix BingoBoingo "Spring Into Money" kek [13:37]
shinohai: I am sure Argentine Bitcoiners are laughing at their little Chilenos today. [13:39]
mircea_popescu: trinque incidentally, i received confirmation from a rather unlikely quarter that your "SQL is the DEF" idea is spot on : apparently people in games [modders, people who do the auction websites, etc etc,] have long ago and unviersally agreed THAT is how to DE. pic : http://logs.minigame.bz/2016-04-20.log.html#t07:32:45 [13:40]
mircea_popescu: PeterL i dun know that the "selection of assets" was ever the criteria. [13:40]
trinque: mircea_popescu: yep, it suffers from having been given an inane syntax, the smell of which drives people away from the relational model [13:41]
mircea_popescu: http://btcbase.org/log/2016-04-20#1453984 << new version of "i don't understand how money works so jews must have pact with devil". which is WHAT is the problem with the sort of lowest-possible-effort, deliberately ignorant effort, PeterL [13:42]
a111: Logged on 2016-04-20 14:56 mats: solrodar: 'i don't understand this financial instrument so the operator must be laundering money' [13:42]
mircea_popescu: it'll never wash. if it washes at some point in some circle, that's proof positive the circle in question is made of shit. maybe more or less invidious shit, depending, but shit nevertheless. [13:43]
trinque: http://logs.minigame.bz/2016-04-20.log.html#t07:47:08 << ah csv, a shitty relational table! [13:44]
mircea_popescu: you know ? [13:44]
mircea_popescu: in any case - practice prevails. this ("modder people use sql to de") reads to me exactly like "porn people use vhs" did in 1980. it's not that it's over for betamax. it's that "what is this and where's the vhs." [13:45]
PeterL: vhs was king until something better came along [13:46]
mircea_popescu: [vocabulary : def = data exchange format de = data exchange pic = point in case] [13:48]
mircea_popescu: check out phf doing his logreading homework eh. [13:52]
mircea_popescu: http://btcbase.org/log/2016-04-20#1454005 <<< it was explained originally and has to be explained periodically [because the flies have infinite hitpoints only in aggregate, as the fly swarm, otherwise live for half a year as individuals] that the principal point of the fee is specifically to keep fees away. [13:54]
a111: Logged on 2016-04-20 15:23 mats: 'its so high nobody could possibly ever want to pay for it unless it was to launder money' ??? [13:54]
mircea_popescu: see old discussions about "qualified investor" and so on and so forth. [13:54]
mircea_popescu: to keep flies* away [13:55]
mircea_popescu: http://btcbase.org/log/2016-04-20#1454008 << look, i understand this "teach plox throw the curve" talk has served you well through life. neveretheless, the fact that you don't see more things, and things made by orther people on mpex is that you personally suck. [13:56]
a111: Logged on 2016-04-20 15:26 solrodar: 3) and on which practically nothing was traded except the shares of the exchange itself [13:56]
mircea_popescu: it doesn't attach to me, or to mpex. i'm doing things. it attaches to you, personally : you're not good enough. and you sure as hell fore that reason do not belong opining in here. go learn how to do things that mpex may consider listing. [13:56]
mircea_popescu: it's a sad property of EVERYONE IN BITCOIN that mp is in the classical position of zeus, where if the whole mt olympus picks up the other side of the ring, he can still throw them all over the fucking sky. [13:57]
mircea_popescu: this isn't something to whine about. this is something to motherfucking fix. starting years ago, but 2nd best - today. [13:57]
mircea_popescu: http://btcbase.org/log/2016-04-20#1454022 << you are also using words you do not master the meaning of. successful closure is sufficient and definitive proof something WAS NOT a pyramid scheme. the reason should be obvious, if you're not lazy and retarded. you however are, no matter what your mommy may have mendaciously told you, both lazy and retarded, so let's explain : [14:05]
a111: Logged on 2016-04-20 15:30 solrodar: but you know how ponzi schemes usually claim to have some undisclosed source of income [14:05]
mircea_popescu: the reason ponzi collapses is not because ~someone wants to~. that's the angle ponzi masters ~push~, and lazy, retarded ignoramuses like yourself lap up. [14:06]
mircea_popescu: the reason ponzi collapses is because ~it can not continue~. that's it. if he could have, maddof would have closed down his ponzi cleanly. he can not. if it could have, the usg would have closed down its social securities, or "us financial system" or "real estate values" ponzies cleanly [14:07]
mircea_popescu: it can not. that's the only important attribute of a ponzi - that it can not be closed down cleanly. [14:07]
mircea_popescu: if it weren't for this, it wouldn't even be a bad thing, everyone'd do it. [14:07]
PeterL: I suppose a ponzi could be closed cleanly if done while it is still very small? If the operator kiscks in enough money to close it before it grows bigger than his pockets? [14:08]
mircea_popescu: if it's not larger than the operator then what is it ? [14:09]
PeterL: or is it only a ponzi once it gets bigger than the operator? [14:09]
mircea_popescu: this is some version of the "tree in the forest" argument. [14:09]
PeterL: could be that [14:09]
mircea_popescu: http://btcbase.org/log/2016-04-20#1454029 <<< i'm sorry, you want free access to IP, go ask apple. and for that matter intel. [14:12]
a111: Logged on 2016-04-20 15:45 solrodar: yeah, so I'm not saying it was a ponzi, or any other kind of scam, just that he wasn't being entirely up-front about what was going on [14:12]
wywialm: hello [14:17]
mircea_popescu: hola [14:17]
wywialm: i wonder if F.MPIF is open for new profit centers [14:18]
PeterL: didn't f.mpif close? [14:18]
mircea_popescu: wywialm well it's a sad time for f.mpif, what with bitbet going. but i guess maybe ? what've you got ? [14:18]
wywialm: if you perhaps remember, i once announced work on a derivatives exchange [14:19]
mircea_popescu: vaguely [14:19]
wywialm: the work approaches completion, and in parallel i cooperated with developing market making algorithms on both futures and options [14:21]
mircea_popescu: http://btcbase.org/log/2016-04-20#1454044 << the principle is sound and very close to home. the unforeseen obstacle is, of course, people. specifically it turns out it's a lot harder than expected to distinguish the people who'd like to be involved from the people who have any busienss being involved. the sad realisation is that the world changed A LOT over the past few decades, and not for the better. [14:21]
a111: Logged on 2016-04-20 16:20 phf: both are fledgling processes, self contained by implicit tmsr rules. as economic activities they could never the less grow and gradually eat more of the outside world [14:21]
mircea_popescu: wywialm good for you but what's this to do with f.mpif ? [14:22]
mircea_popescu: http://btcbase.org/log/2016-04-20#1454047 << no. 2012 was the year every dog with a flea in his beard could pretend to humanity. then 2013 brought the requirement to actually not be poor, and 2016 brought the requirement to not be stupid, and common folk are all butthurt over being left behind [14:23]
a111: Logged on 2016-04-20 16:27 solrodar: but when I first came across it a year or so ago, I got the impression that 2012 was the what bliss in that dawn to be alive [14:23]
mircea_popescu: squarely ignoring the fundamental part - that they should be left behind, that they're the scum in the engine and the gout in the knight. [14:23]
wywialm: :) i could invite F.MPIF to join the market making operations [14:23]
deedbot: [Qntra] BitBet Settlement Transaction Broadcast And Confirmed - http://qntra.net/2016/04/bitbet-settlement-transaction-broadcast-and-confirmed/ [14:23]
mircea_popescu: so basically "i have this untested thing we baked together with some greenhorns that aren't in the wot but wouldn't mind using me - wouldn't you like to give us money ?" [14:24]
mircea_popescu: this is a pretty miserable offer. capitalism is about who gets what, not about who needs what. [14:24]
mircea_popescu: http://btcbase.org/log/2016-04-20#1454051 << he condensed it to a pretty tiny size. [14:26]
a111: Logged on 2016-04-20 17:07 ben_vulpes: omg not all at once davout [14:26]
punkman: phf: meanwhile couple of months ago mp explicitly said that he's starting a process of divesting from bitcoin, that was before bitbet, before ~anybody~ had voiced any kind of objections to anything << "divesting" came after "A miner problem" [14:27]
PeterL: is there any benefit from splitting a transaction into multiple parts? seems like it would be most efficient to just do it all at once? [14:27]
punkman: phf: so now idiots are going to look at the past couple of months and point at patterns, "where's the 50btc account creations!" << and this was 1 year ago [14:28]
wywialm: i did not say that i need this money. also, it is not an offer for investing right now, just a check whether you will find anything interesting in it for you. if yes, i could go into greater detail how this is tested, how is going to be tested, etc. [14:28]
mircea_popescu: what has happened since you first mentioned it to lukewarm reception is that the person doing it has still not done his half year of log reading and humble wot beginings. [14:29]
mircea_popescu: the notion that i'd support such a thing is plainly outrageous. [14:30]
wywialm: for all practical purposes, i represent both the exchange and the market making, and it's more than a half year since my announcement. none of this, of course, is meant to imply that you should support anything i have to offer, i just try to clarify [14:33]
mircea_popescu: so you made it ? [14:34]
PeterL: wywialm what is traded on your exchange? [14:35]
wywialm: i designed the exchange, yes - as i said then, all the economic and transactional aspects are designed by me. [14:36]
wywialm: Quedex aims to offer FX futures and options, initially at least [14:36]
mircea_popescu: i dunno why this never came through. [14:37]
mircea_popescu: well, so explain the design ? [14:37]
shinohai: http://btcbase.org/log/2016-04-20#1454120 <<< sounds like a Mexican cheese brand kek [14:39]
a111: Logged on 2016-04-20 18:36 wywialm: Quedex aims to offer FX futures and options, initially at least [14:39]
mircea_popescu: later tell mike_c hey, danielpbarron can't log into eulorum because something with cookies. anything change ? [14:39]
gribble: The operation succeeded. [14:39]
mircea_popescu: shinohai you ever had bimbo bread ? [14:39]
shinohai: I have a loaf in my fridge lol [14:40]
shinohai: Bodegas are the only place to get fresh ingredients for anything around here, except when farmer's markets are open. [14:42]
mircea_popescu: pretty shitty wonderbread, but always good for a lulz, make teh galz have it. [14:42]
wywialm: to recall main points from the previous exposition: the exchange employs gpg messaging (following the MPEx standards) and allows to trade financially-settled futures and options initially on BTCUSD exchange rate. [14:42]
mircea_popescu: wywialm nice, but this isn't what i have in mind when i hear "economic design". [14:42]
mircea_popescu: explain that part. how are they designed, these instruments. [14:43]
wywialm: it utilises two matching models: continuous trading and auction, the latter being triggered around market settlement and around significant market moves [14:43]
wywialm: all futures and options have fixed USD notional value, each of them is margined separately, though portfolio-based margining is coming soon [14:44]
wywialm: i.e. each contract corresponds to $100 [14:44]
wywialm: so that bitcoin is the home currency [14:45]
mircea_popescu: on what basis can you offer margin ? [14:45]
mod6: <+mircea_popescu> shinohai you ever had bimbo bread ? << first time I saw this in a supermarcado i was like "ha! they don't even know..." [14:48]
mircea_popescu: yeah lol [14:49]
wywialm: i'm not sure that i understand what you mean. i offer margin as i believe that the liquidation mechanism (including auction) will make sure that the losses do not exceed the deposits [14:49]
mircea_popescu: was one of the lulzy moments in costa rica, day after landing there, me spends with local bimbos, next day, supermarket, bimbo bread, o.O [14:49]
mod6: :D [14:50]
mircea_popescu: wywialm a) you're sure about something you've not explained b) you are forcing me to read between the lines which is the opposite of competence c) if what you mean is that you steal everyone's deposited cash to do mutually-financed margin, what happens when a position split 50/50 (allowing you, possibly, somewhere close to 2x margin) suddenly moves in choppy trading to 85/15 split [14:50]
mircea_popescu: and c-1) have you looked into the history of icbit.se which did the exact same thing and its trampling is discussed in early trilema ? [14:51]
shinohai: mod6 they sell "crumbs" too so I wonder if a trail of those leads bimbos to your door? [14:52]
wywialm: beggining from the last, c-1) yes, and as far as i remember, you removed a 'scam' warning from them [14:52]
wywialm: a) i'll try to proceed to explain everything i know about b) if that's the case, then indeed i'll try to speak more clearly [14:54]
mircea_popescu: they paid me. this doesn't make the model more workable. [14:54]
mircea_popescu: and don't tell me you're about to skip c ?! [14:55]
wywialm: no, certainly not. c) what do you mean by 'position split' and what is the relation of it to leverage? [14:57]
wywialm: the long positions and short positions always necessarily must match [14:58]
wywialm: at any rate, margin is unavoidable when one trades in futures contracts - the short leg's losses can always exceed any deposit [15:02]
mircea_popescu: ... [15:03]
mircea_popescu: mkay, let's detail this for the benefit of the kids reading logs. [15:04]
wywialm: ok, let me take it directly c) no, in no case the actual deposits are used as a coverage. The risk management has following steps: [15:05]
wywialm: first, if the position holder's losses approach his maintenance margin level, his entire position is liquidated on the market [15:05]
wywialm: second, if it happens that the market will move significantly impacted by this liquidation, the auction is triggered [15:06]
wywialm: this means that continuous trading is halted, orders are gathered and order book is balanced [15:07]
mircea_popescu: exchange trades symbol X. participant Be[ar] comes in and deposits 1000 coins. participant Bu[ll] comes in and deposits 1000 coins. [15:08]
mircea_popescu: exchange now has 2000 coins cash. [15:08]
mircea_popescu: exchange advances 2:1 margin to Be, backed by "its" cash. [15:08]
mircea_popescu: exchange advances 2:1 margin to Bu, backed by idem. [15:08]
mircea_popescu: Be puts in one order, at 9.9, ie 202 X. [15:08]
wywialm: if nevertheless, the auction price is such that the person's losses exceed his maintenance margin, the exchange covers the losses [15:08]
mircea_popescu: Bu puts in one order, at 10.1, ie, 198 X [15:08]
mircea_popescu: MP comes in and sells 1 X at 10.1. the price for X is now 10.1 [15:08]
mircea_popescu: Be is now underwater, exchange takes its order off. [15:08]
mircea_popescu: MP comes in and buys the contract back, at 5. [15:08]
mircea_popescu: Bu is now way the fuck underwater, exchange takes his order off also. [15:08]
mircea_popescu: This in the happy case where the exchange doesn't attempt to autoliquidate positions, creating all the noise surges you could wish for. [15:08]
mircea_popescu: "the exchange covers the losses" out of what ? you're not fabulously rich afaik. [15:09]
mircea_popescu: and no, you can't fucking liquidate on the market, think about it for a moment. the very reason you are liquidating in the first place is that the market's no good. [15:09]
wywialm: from the insurance fund, replenished from commissions [15:10]
mircea_popescu: this is the first time i hear of such. [15:11]
mircea_popescu: hence a)/b). [15:11]
wywialm: well, liquidations don't happen when market is no good, but only when one participant is no good [15:11]
mircea_popescu: in any case : this is a rehash of that old http://trilema.com/2014/the-woes-of-altcoin-or-why-there-is-no-such-thing-as-cryptocurrencies/ which'd prolly make good reading. [15:11]
mircea_popescu: essentially, you see what the big boys are doing, and figure you'll just do it yourself. except - the big boys can do things BECAUSE they are the big boys, ie, there's bernanke there to "save the economy". you aren't, and i can lean on you. moreover, the big boys don't ever do this 2:1 thing. they do a little margin, and as private financing, which is exactly how you should be doing it also. [15:12]
wywialm: well, certainly the fiat currency system makes the working of derivatives exchanges much easier [15:15]
wywialm: returning for a while to the 'market's no good' argument [15:16]
wywialm: if the liquidated position is small enough, then liquidating on the market is seamless if it's large, here comes the auction. the orders are gathered during a certain period of time, which may be extended, without time priority [15:18]
mircea_popescu: we're not discussing here the ideal case where "the liquidated position is small enough". we're discussing here interesting cases, when you have a noisy market and a bunch of agents that act irrationally. [15:19]
wywialm: if price moves far enough, the auction serves as a circuit-breaker. Essentially, the only risk lies not in a noisy market or irrational, but in a large, certain but unforseen, and fundamental jump in the price [15:21]
mircea_popescu: except you're small, so "large" doesn't mean much. [15:21]
mircea_popescu: what happens if A is leveraged 1:1 and owns 100 contracts and B is leveraged 2:1 and "owns" 400 contracts, if that " the auction is triggered" undocumented bug springs into auction ? can A's wholly-owned contracts be touched by B's insanity ? [15:23]
wywialm: large here means 'large relative to maintenance margin percents', so our size doesn't matter much in this case. if the maintenance margin is 10% then this means that a 10% price jump won't hurt us regardless of our reserves. [15:24]
mircea_popescu: not that you ever explained in your design description what "maintenance margins" are, but this is a magic trick without substance : if you need to move 5% to break you'll move 5% and if you need to move 15% you'll move 15%. [15:25]
mircea_popescu: and the "how i picked 10%" is a testament to the whole "copying big boys" thing. there's a reason they use the magic numbers they do. [15:26]
mircea_popescu: that reason is empiric. [15:26]
wywialm: in fact, the 10% maintenance margin is illustrative only [15:26]
wywialm: but i proceed to explain [15:26]
mircea_popescu: you can't say you designed something and then keep pulling items from your sleeve as discussion of your design progresses. [15:27]
PeterL: perhaps it would be better to write up a full description instead of trying to string it out bit by bit? [15:28]
wywialm: PeterL, yes - the full documentation will appear on the website [15:29]
mircea_popescu: so far, i'm not as much as able to distinguish this from "here's my facebook clone", honestly. [15:30]
mircea_popescu: bit by bit works, but by invalidating THAT, and sorted in order of importance. [15:30]
wywialm: i didn't certainly revolutionize the electronic derivatives trading industry (nor claimed to), just trying to make a robust bitcoin-based market [15:32]
mircea_popescu: but your design is anything but robust. it's web-optimized, metaphorically speaking. [15:32]
wywialm: :) [15:33]
wywialm: let me refer to your example. first A owns his contracts in the same way B owns them. if A has short positions, these contracts are not fully collateralized. if B's losses exceed his margin (here: 50%), and exchange's insurance fund, in that situation A's profit, but not his initial deposit will be diminished accordingly. in that way, the exchange acts only as a intermediate between the counterparties [15:36]
deedbot: [Qntra] Happy 420 Charlie Shrem - http://qntra.net/2016/04/happy-420-charlie-shrem/ [15:36]
wywialm: an intermediate* [15:36]
trinque: BingoBoingo: lol! [15:38]
mircea_popescu: wywialm the fundamental problem with this view is that i do not wish to engage in a financial relationship with deadbeat B, where "your site was only an intermediary". i wish to engage in specified, modellable deals. [15:38]
mircea_popescu: similarly, if oksluts.com opens up offering sluts starting at 1.75 and DD for 1k dollars, [15:39]
mircea_popescu: i am not open to receiving a short hairy jewish girl for my money, even if oksluts.com imagines "it was just an intermediary between parties" [15:39]
wywialm: at present stage, the mechanisms i tried to describe are only capable of making the "short hairy jewish girl" scenario as unlikely as possible. what could make it really impossible, is to issue a CDS on the general liabilities of the exchange. The proceedings from the sale would increase the insurance fund in return for a compensation. If total amount of insurance fund and CDS issued is greater than potential losses, no haircut on profits would [15:45]
wywialm: be possible. [15:45]
mircea_popescu: these are the canonical objections to this sort of "web exchange, mutually-insured on user funds" things : that on one hand they're an algorithmic noise amplification machine with no dampener available other than the power of prayer and that the yessentially implement a lemon market, a sort of competition in deadbeat-ism among the users. which is why they're so popular on reddit, and nowhere else. [15:46]
wywialm: this doesn't refer to the CDS-insured variant, does it? [15:48]
mircea_popescu: depends how that is implemented. [15:48]
mircea_popescu: but finance, much like cryptography, is altogether a crapshoot in the dark. [15:51]
mircea_popescu: both are disciplines concerning thesmelves with the unseen unknown. [15:51]
wywialm: generally, it will look as follows: the CDSs have a specific maturity (possibly matching futures expirations). The person buys the CDS with a discount from the notional value, the proceeds from the sale of the CDS is locked until maturity. If no credit event occurs, full notional value is paid out. if it does, first the insurance fund from the exchange is used, and CDS holders are a last resort [15:55]
wywialm: if the losses exceed the insurance fund, the CDS's notional value is diminished proportionally [15:56]
shinohai: BingoBoingo: https://www.justice.gov/usao-wdla/pr/former-shreveport-chiropractor-son-plead-guilty-operating-illegal-bitcoin-exchange [15:57]
BingoBoingo: shinohai: Guilty pleas are generally boring [15:58]
mircea_popescu: this is actually robust. [15:58]
BingoBoingo: Oh? [15:58]
mircea_popescu: BingoBoingo was talking to wywialm [15:58]
BingoBoingo: ah [15:58]
mircea_popescu: shinohai i don't get it, what the fuck difference does it make that guy's dad used to be a chiropractor ?! [15:59]
shinohai: I guess he was using his medical license to sell Xanax [16:00]
mircea_popescu: heh [16:00]
shinohai: *because [16:00]
shinohai: File under: Bartards [16:01]
wywialm: the exchange has not yet launched and CDS mentioned above are to be implemented yet, but i asked about f.mpif knowing that in case of your slightest interest, any preparations will take a lot of time. [16:02]
mircea_popescu: who'd be running the pc anyway ? [16:02]
wywialm: i could offer my candidature, but of course this may be unsuitable for you. in that case, i could present the details to a person of your choice [16:05]
mircea_popescu: so basically you're looking for a skilled market maker to market make your new exchange ? [16:06]
wywialm: no, as i said before, i already have the automated market making algorithms, i look for expanding its funding [16:08]
mircea_popescu: this is also the first time you mentioned the exchange is going to take positions too ? [16:08]
wywialm: the market making project is separate from the exchange. i designed both, but they are independent from one another [16:10]
mircea_popescu: o.O [16:16]
mircea_popescu: are your futures going to be physically settled ? [16:16]
wywialm: initially, not, but it is definitely planned [16:20]
mircea_popescu: so how does the exchange get its price signal (other than the obvious "what'd be most convenient for the market making project") [16:21]
wywialm: an average of spot exchanges [16:22]
mircea_popescu: honestly this just reduced in my head to basically "hey, i want to be mp, and for the year to be 2012". so i guess... good luck with it ? [16:23]
phf: http://btcbase.org/log/2016-04-20#1454110 << but "divesting" came before "anybody had voiced any kind of objections to anything", which is what i said [16:24]
a111: Logged on 2016-04-20 18:27 punkman: phf: meanwhile couple of months ago mp explicitly said that he's starting a process of divesting from bitcoin, that was before bitbet, before ~anybody~ had voiced any kind of objections to anything << "divesting" came after "A miner problem" [16:24]
mircea_popescu: poor deedbot [16:26]
trinque: he got a few timeouts when trying to reconnect to chat.freenode.net [16:28]
wywialm: mircea_popescu, i certainly aim to create an options market similarly to what you did, but i'm not you. [16:32]
wywialm: i was only presenting an offer for you to consider or discard, according to your judgement [16:32]
BingoBoingo: wywialm: If you want to actually distinguish yourself you prolly ought to consider offering actual tulip futures with physical delivery. Gives you something to practice your market making with. [16:54]
shinohai: lo [16:55]
BingoBoingo: Seriously. It's a commodity, that gets traded, and covert delivery is easy. [16:58]
danielpbarron: it's prolly harder to do such a thing with bulbs. there already exist plenty of companies doing this, it's a very large market, it's a perishable product, very very heavy, need good connections, etc.. [17:00]
BingoBoingo: Not so perishable. [17:03]
danielpbarron: they keep for a few months and then they better be in the ground already [17:06]
BingoBoingo: Prepped for storage well they'll survive out of the ground longer. [17:07]
shinohai: later tell mod6 or ben_vulpes Should the #b-a part be changed on http://thebitcoin.foundation/ ? [17:08]
gribble: The operation succeeded. [17:08]
mod6: Oh, kinda forgot about that. [17:09]
mod6: <+shinohai> mod6 they sell "crumbs" too so I wonder if a trail of those leads bimbos to your door? << inquiring minds want to know! [17:10]
shinohai: I'll set up a gofundme for research. If the cryptoid horseshit can get funds so can I. [17:11]
mod6: I'll have to discuss with ben on the IRC section, do we even allow people to use the webclient to connect here? [17:11]
trinque: heh, beat me to it [17:13]
mod6: heheh. i updated that to #trilema [17:14]
mod6: thanks trinque [17:14]
mod6: i think we're gonna need our own wiki now too, which is fine i guess. [17:14]
shinohai: thx mod6 I just happened to notice it earlier. [17:15]
mod6: thanks for keeping an eye out. [17:15]
shinohai: On the off chance someone finds it and wants to contribute, want them to show up in proper channel. [17:16]
mod6: werd [17:20]
BingoBoingo: bc,stats [17:32]
gribble: Current Blocks: 408204 | Current Difficulty: 1.7867830767168845E11 | Next Difficulty At Block: 409247 | Next Difficulty In: 1043 blocks | Next Difficulty In About: 6 days, 15 hours, 26 minutes, and 22 seconds | Next Difficulty Estimate: None | Estimated Percent Change: None [17:32]
BingoBoingo: ticker --market all [17:32]
gribble: Bitstamp BTCUSD last: 443.91, vol: 7873.36677405 | BTC-E BTCUSD last: 435.001, vol: 9521.99303 | Bitfinex BTCUSD last: 445.27, vol: 24253.22781182 | CampBX BTCUSD last: 433.0, vol: 1.1791705 | BTCChina BTCUSD last: 444.781108, vol: 39152.19710000 | Kraken BTCUSD last: 443.5, vol: 1757.46015997 | Bitcoin-Central BTCUSD last: 438.739092, vol: 44.60126226 | Volume-weighted last average: (1 more message) [17:32]
BingoBoingo: more [17:32]
gribble: 443.683554541 [17:32]
BingoBoingo: http://www.emergencymedicalparamedic.com/most-horrific-obesity-story/ << Don't blame the mice! [17:35]
shinohai: If true, perhaps best hambeast story I have ever read. [17:42]
trinque: mod6 │ i think we're gonna need our own wiki now << http://wiki.deedbot.org/ I got you [18:24]
trinque: I just dumped cliki2 on there needs a better stylesheet [18:25]
trinque: would probably be cool to hack deedbot auth into that [18:28]
BingoBoingo: Just ask Herr Vulpes how he handled VAN logins [18:29]
ben_vulpes: session up [18:38]
trinque: do what now [18:40]
trinque: what's VAN? [18:40]
ben_vulpes: abortive tmsr~ ad network i build once upon a time [18:41]
trinque: ah I remember you talking about that [18:45]
BingoBoingo: dat GPG login dough [18:45]
ben_vulpes: shat out an otp, successful decryption of which set a session on the user's browser. [18:46]
ben_vulpes: nothing magical [18:46]
phf: he's a witch, burn him [19:04]
trinque: well... we did do the nose [19:06]
mod6: heheh [19:15]
mod6: trinque: awesome! [19:16]
mod6: this matcha tea isn't bad as long as its not clumpy [19:17]
mircea_popescu: wywialm yes but the offer's not clear. asked if you need money, turns out no asked if you need skill, turns out no. well... [20:23]
mircea_popescu: mod6> I'll have to discuss with ben on the IRC section, do we even allow people to use the webclient to connect here? <<< i thought so, no ? [20:24]
mircea_popescu: http://btcbase.org/log/2016-04-20#1454289 << lol! [20:28]
a111: Logged on 2016-04-20 23:06 trinque: well... we did do the nose [20:28]
mircea_popescu: ftr, it was my idea / prodding that pushed him to it! [20:28]
mircea_popescu: http://btcbase.org/log/2016-04-20#1454292 << check it out, usians now drinking matchsticks tea for lack of phosphorus [20:28]
a111: Logged on 2016-04-20 23:17 mod6: this matcha tea isn't bad as long as its not clumpy [20:28]
mircea_popescu: http://qntra.net/2016/04/bitbet-settlement-transaction-broadcast-and-confirmed/ << o hey check it out. frenchy becomes 1st member of tmsr noblesse du robe, makes living off republican legal process. [20:36]
mod6: mircea_popescu: yeah, the web-irc thing seems to be fine. [20:44]
mod6: So I dunno when I said I was going to check in on the list of things to do, if that was today or friday or whatever. But I'm just getting started now. Been too damn beat down since Friday to even do anything. [20:45]
mod6: Alas, onwards and upwards. [20:45]
mircea_popescu: cheers. [20:45]
mod6: $up copumpkin [21:30]
deedbot: copumpkin voiced for 30 minutes. [21:30]
copumpkin: 1) deedbot seems to reject my attempts to register with it [21:30]
copumpkin: 2) mpex seems to have forgotten my public key [21:31]
copumpkin: anyone know what's up with either of those? [21:32]
copumpkin: I've been trying echo "STAT" | gpg -u <my key id> --clearsign | gpg --encrypt --armor -r 2FB7B452 [21:32]
mircea_popescu: hola. [21:42]
mircea_popescu: when's the last time you used it ? [21:42]
mircea_popescu: for one thing, seems to be way before the last mpex key update years ago ? [21:43]
copumpkin: not years ago [21:48]
copumpkin: I did have to give you my key again a couple of years ago [21:48]
mod6: copumpkin: 2FB7B452 << this is mp's personal key, try with '02DD2D91' [21:48]
mircea_popescu: oh right, that's my key. [21:48]
copumpkin: oh [21:48]
mod6: http://mpex.biz/faq.html#6 [21:48]
copumpkin: whoops, I pasted the wrong thing [21:49]
mircea_popescu: oddly enough i don't even recognize my key in short form o.O [21:49]
copumpkin: oh that worked [21:49]
mircea_popescu: lol yw! [21:49]
copumpkin: man, I feel stupid now :) [21:50]
copumpkin: it still shows me as having MPOE holdings. Is that expected? [21:50]
mircea_popescu: still there for collectors, ie, people just like you, who might want a screenshot\ [21:50]
copumpkin: :D [21:50]
mircea_popescu: they don't have any monetary value or anything. [21:50]
copumpkin: boo [21:51]
mircea_popescu: well they already paid out. [21:51]
copumpkin: yeah :) [21:51]
mircea_popescu: anyway, how's life these days. [21:51]
copumpkin: pretty good. You? I hear you had a falling out with -assets [21:52]
mircea_popescu: more like moved on. [21:53]
mircea_popescu: but not bad, check out eulora sometime. [21:53]
mircea_popescu: ahahah check out his hostmask asciilifeform ~copumpkin@haskell/developer/copumpkin :D [21:57]
copumpkin: :) [21:57]
mircea_popescu: anything interesting ? [21:57]
BingoBoingo: $up copumpkin [22:05]
deedbot: copumpkin voiced for 30 minutes. [22:05]
mircea_popescu: $gettrust deedbot copumpkin [22:06]
deedbot: L1: 0, L2: 1 by 1 connections. [22:06]
mircea_popescu: just, huh. time has passed you by yo! [22:06]
mircea_popescu: but anyway - you can self-voice for an indefinite period by pm-ing deedbot $up and then $v the otp it gives you [22:07]
copumpkin: yeah, I tried that before, but it refused to register me [22:14]
mircea_popescu: uh ? [22:14]
copumpkin: [21:17:27] <deedbot> Import failed for copumpkin. [22:15]
copumpkin: I just typed $register copumpkin <fingerprint> [22:15]
mircea_popescu: hmmm wtf is tis [22:16]
mircea_popescu: oh, tyou're prolly already regiustered i expect [22:16]
copumpkin: hmm [22:16]
copumpkin: with a different nick? [22:16]
mircea_popescu: well no with this one since it sees your trust level [22:17]
copumpkin: hmm [22:17]
mircea_popescu: say $up [22:18]
copumpkin: [22:18:14] <deedbot> copumpkin may not $up [22:18]
copumpkin: I'm pretty evil [22:18]
mircea_popescu: lemme see one second, how was the key recover thing [22:19]
mircea_popescu: $key copumpkin [22:20]
deedbot: http://wotpaste.cascadianhacker.com/r/0a9db265-acd3-446f-815f-428bf5b23a41/ [22:21]
mircea_popescu: you are already registered and it knows your key. [22:22]
mircea_popescu: say $up in here [22:22]
copumpkin: okay, weird [22:22]
copumpkin: $up [22:22]
deedbot: You must be registered. [22:22]
copumpkin: lol [22:22]
mircea_popescu: da fuck is this weird [22:22]
mircea_popescu: trinque any clue ? [22:22]
copumpkin: I think it dislikes me [22:22]
mircea_popescu: copumpkin it doesn't run on a mac, so why would it. [22:22]
BingoBoingo: Fun with CBZ and grey water irrigation https://archive.is/BM8DS [22:23]
copumpkin: :) [22:24]
BingoBoingo: $up copumpkin [22:44]
deedbot: copumpkin voiced for 30 minutes. [22:44]
asciilifeform: veeeeeeeeeeeery preliminarily: http://nosuchlabs.com [22:53]
asciilifeform: (plz do not bookmark the rss yet! url will change to http://phuctor.nosuchlabs.com soon) [22:53]
asciilifeform: tank is getting refilled. [22:54]
asciilifeform: meanwhile, boiz and gurlz, enjoy (enjirl) the all-new Braindamage section. [22:54]
asciilifeform: later tell hanbot there. [22:55]
gribble: The operation succeeded. [22:55]
BingoBoingo: nice http://nosuchlabs.com/sadmods [22:56]
BingoBoingo: ROLFCOPTER http://nosuchlabs.com/gpgkey/0D9057DA7AEE12C725AA9408D47F4FFC3769BEF7891A0F9C0A9F38420C5C08AB [22:57]
asciilifeform: eh that one's old nyooz [22:57]
asciilifeform: see mircea_popescu's site. [22:57]
asciilifeform: but the basic litmus is now built-in and it turned up a buncha strange. [22:58]
asciilifeform: also thing runs on a real db now and can handle 'slashdotting' etc. [22:58]
* asciilifeform has been awake for quite some time and will bbl. [22:58]
BingoBoingo: Yes, but now linkable [22:59]
mircea_popescu: lol enjirl [22:59]
BingoBoingo: So roughly extrapolating alf time from this and applying to Cardano... By 2020? [23:02]
mircea_popescu: lol merciless. [23:02]
mircea_popescu: people aren't machines! [23:02]
BingoBoingo: Oh, I was thinking more cosmological phenomena or natural process than machine [23:07]
BingoBoingo: to be fair I'm running roughly 10 years behind on everything that matters myself [23:10]
mircea_popescu: :p [23:12]
BingoBoingo: http://www.amplestuff.com/750-lbwideplatformdigitalscale.aspx [23:17]
* BingoBoingo awaits seeing this sold in Walmart bathroom aisle. Starting to suspect instead of mp's watermelon field we'll just palletize obeasts [23:21]
mod6: asciilifeform: congrats! [23:32]
Category: Logs
Comments feed : RSS 2.0. Leave your own comment below, or send a trackback.
Add your cents! »
    If this is your first comment, it will wait to be approved. This usually takes a few hours. Subsequent comments are not delayed.