1 2011-08-27 00:01:56 <lfm> as soon as you win a bid you have to put the btc in escrow. When you verify receipt, you release the escrow
  2 2011-08-27 00:03:02 <zipperhead> Interesting idea.
  3 2011-08-27 00:04:10 <marktraceur> Hey, folks, I have a bit of a question. I'm using python and the python-bitcoinrpc library in a web app. We're running into conversion errors with Decimal objects, saying they aren't JSON-serializable. Has anyone figured out what the solution to that problem is? We tried converting to long ints, then back to floats, but then it gives a blank JSONRPCError.
  4 2011-08-27 00:04:45 <lfm> marktraceur: use doubles, not floats
  5 2011-08-27 00:04:53 <marktraceur> lfm: Will try, thanks
  6 2011-08-27 00:05:14 <luke-jr> marktraceur: and use ints internally
  7 2011-08-27 00:05:22 <lfm> marktraceur: and convert to 64 bit ints for any calculations
  8 2011-08-27 00:05:26 <luke-jr> lfm: Python only has one kind of floating-point
  9 2011-08-27 00:05:44 <lfm> oh ok
 10 2011-08-27 00:06:14 <marktraceur> lfm: You don't even know python! :P
 11 2011-08-27 00:06:24 <marktraceur> No, that obviously didn't work
 12 2011-08-27 00:06:30 <luke-jr> marktraceur: really, the Decimal nonsense is just overhead
 13 2011-08-27 00:06:31 <zipperhead> FFI: http://docs.python.org/library/decimal.html
 14 2011-08-27 00:07:27 <luke-jr> marktraceur: just pass all amounts from JSON-RPC into round(INPUT * 1e8), and when sending do OUTPUT / 1e8
 15 2011-08-27 00:08:07 <luke-jr> you can do int(round(INPUT*1e8)) for a tiny performance boost
 16 2011-08-27 00:08:17 <lfm> you still would need to be real carefull doing calculations on them
 17 2011-08-27 00:08:32 <carter_> does anyone know of a good article that explains how to write a miner?
 18 2011-08-27 00:08:33 <marktraceur> OK, more different question
 19 2011-08-27 00:08:45 <luke-jr> carter_: https://en.bitcoin.it/wiki/Getwork
 20 2011-08-27 00:09:02 <marktraceur> I'm passing a float into the sendtoaddress function, now, and it seems to work
 21 2011-08-27 00:09:11 <marktraceur> But then, like I said, it raises an empty JSONRPCError
 22 2011-08-27 00:09:54 <luke-jr> marktraceur: it might be helpful to read over some of Spesmilo's code
 23 2011-08-27 00:10:52 <marktraceur> On it, will see what's what
 24 2011-08-27 00:12:40 <marktraceur> Hm, it doesn't seem that Spesmilo uses ServiceProxy?
 25 2011-08-27 00:13:21 <luke-jr> Spesmilo uses the standard JSON-RPC API
 26 2011-08-27 00:13:26 <luke-jr> which bitcoinrpc also supports
 27 2011-08-27 00:13:46 <marktraceur> Right
 28 2011-08-27 00:13:47 <luke-jr> oh, that IS ServiceProxy
 29 2011-08-27 00:13:51 <luke-jr> see core_interface.py
 30 2011-08-27 00:14:51 <marktraceur> Aha!
 31 2011-08-27 00:18:29 <coderrr> http://www.marketwatch.com/story/bitcoin-buyers-throw-good-money-after-bad-2011-08-26
 32 2011-08-27 00:20:32 <marktraceur> luke-jr: Is it sending the amount time 1e8 to the server?
 33 2011-08-27 00:20:36 <marktraceur> times*
 34 2011-08-27 00:20:50 <luke-jr> marktraceur: no, you want to divide by 1e8 to send to the server
 35 2011-08-27 00:21:06 <luke-jr> marktraceur: core_interface works with a sanitized integer API, and converts it to float if necessary
 36 2011-08-27 00:21:09 <marktraceur> Ah, yes, I see, it's just spread out
 37 2011-08-27 00:21:24 <luke-jr> marktraceur: Spesmilo still has support for an old patch which makes JSON-RPC work in ints too
 38 2011-08-27 00:21:29 <luke-jr> auto-detected
 39 2011-08-27 00:21:30 <marktraceur> OK
 40 2011-08-27 00:21:39 <marktraceur> But this looks pretty much like what I'm doing
 41 2011-08-27 00:22:08 <phantomcircuit> luke-jr, old patch we use that on britcoin and have a branch that works for all the latest tag release
 42 2011-08-27 00:22:47 <luke-jr> phantomcircuit: I think you mean strings, which is even more broken by design
 43 2011-08-27 00:23:30 <phantomcircuit> it's a lexical cast iirc
 44 2011-08-27 00:24:36 <phantomcircuit> if it fails it throws an exception so it isn't safe for interfacing with broken scripts
 45 2011-08-27 00:33:40 <marktraceur> luke-jr: FYI, I found the issue
 46 2011-08-27 00:33:45 <marktraceur> Must be a rounding error somewhere
 47 2011-08-27 00:33:55 <marktraceur> Because transferring 0.9 times the total balance works
 48 2011-08-27 00:34:00 <marktraceur> But not the total balance
 49 2011-08-27 00:35:39 <luke-jr> O.o
 50 2011-08-27 00:35:46 <luke-jr> maybe there's a fee?
 51 2011-08-27 00:40:42 <carter_> anybody using https://github.com/jb55/node-bitcoin?
 52 2011-08-27 00:43:03 <DiabolicalGamer> Good Evening Everyone
 53 2011-08-27 00:51:54 <marktraceur> luke-jr: That might be it! I'll see.
 54 2011-08-27 01:34:13 <CIA-101> poolserverj: shadders * 2aa651a01704 r34 / (2 files in 2 dirs): - disable updateCounters in prep for 0.2.9 release. Not ready for use yet.
 55 2011-08-27 01:34:14 <CIA-101> poolserverj: shadders * 787388e6598f r37 /.hgtags: Added tag 0.2.9 for changeset 4f67457d4d5e
 56 2011-08-27 02:42:25 <Raccoon> does anyone here use the mtgox websocket feed?
 57 2011-08-27 02:43:33 <Raccoon> i'm curious how to read this data
 58 2011-08-27 02:43:36 <Raccoon> eg
 59 2011-08-27 02:43:43 <Raccoon> {"channel":"24e67e0d-1cad-4cc0-9e7a-f8523ef460fe","depth":{"currency":"USD","item":"BTC","price":"8.69997","price_int":"869997","type":1,"type_str":"ask","volume":"-20.98","volume_int":"-2098000000"},"op":"private","origin":"broadcast","private":"depth"}
 60 2011-08-27 02:44:08 <Raccoon> {"channel":"24e67e0d-1cad-4cc0-9e7a-f8523ef460fe","depth":{"currency":"USD","item":"BTC","price":"8.69997","price_int":"869997","type":1,"type_str":"ask","volume":"20.98","volume_int":"2098000000"},"op":"private","origin":"broadcast","private":"depth"}
 61 2011-08-27 02:44:41 <Raccoon> both are ASKs, but one has a negative volume change, and the other a positive
 62 2011-08-27 02:45:20 <Raccoon> is it possible to sell negative bitcoins now? :p
 63 2011-08-27 02:47:53 <ThomasV> I sell negative bitcoins for positive dollars
 64 2011-08-27 03:06:17 <CIA-101> DiabloMiner: Patrick McFarland master * r682c586 / src/main/java/com/diablominer/DiabloMiner/DiabloMiner.java : Make kernel arrays an option, default to off - http://git.io/8N-Vtw
 65 2011-08-27 03:47:53 <CIA-101> DiabloMiner: Patrick McFarland master * r0e65699 / src/main/resources/DiabloMiner.cl : Use bitselect on non-AMD platforms - http://git.io/rohAXQ
 66 2011-08-27 04:38:35 <CIA-101> bitcoinjs/node-bitcoin-p2p: booo master * rca58c65 / bin/bitcoinjs : bin/bitcoinjs: use absolute paths - http://git.io/Otm4kQ
 67 2011-08-27 04:38:36 <CIA-101> bitcoinjs/node-bitcoin-p2p: Stefan Thomas master * r9b8bc24 / bin/bitcoinjs :
 68 2011-08-27 05:01:34 <FellowTraveler> Hi all.
 69 2011-08-27 05:07:06 <SomeoneWeird> Hi FellowTraveler
 70 2011-08-27 05:36:13 <cjdelisle> hey FellowTraveler, interesting project you are doing.
 71 2011-08-27 05:36:57 <FellowTraveler> hi cjdelisle, thanks. I just posted a couple videos here:  http://vimeo.com/28141679
 72 2011-08-27 05:39:44 <cjdelisle> can I get the flvs of those? I like to put stuff on disk
 73 2011-08-27 06:00:02 <phantomcircuit> cjdelisle, you like to put stuff on disc as flash video o.O
 74 2011-08-27 06:00:40 <cjdelisle> well... whatever format it happens to be in when I get it
 75 2011-08-27 06:00:44 <cjdelisle> usually .flv
 76 2011-08-27 06:00:59 <cjdelisle> aslong as mplayer can read it
 77 2011-08-27 06:02:09 <cjdelisle> That's a really slick ui, it's kind of unfortunate that most people have a binary idea of trust.
 78 2011-08-27 06:02:57 <cjdelisle> Either they are willing to trust a bank with their money or they don't even want to hear the word "server".
 79 2011-08-27 06:03:53 <cjdelisle> For the trusting people, that kind of an interface would be a huge hit even if it was on a website. emailable chacks would be a big deal.
 80 2011-08-27 06:10:14 <TuxBlackEdo> hey does my /etc/crontab look right on line 3 http://pastebin.com/75xpshiW
 81 2011-08-27 06:10:52 <TuxBlackEdo> should run every 6 hours right?
 82 2011-08-27 06:12:08 <FellowTraveler> cjdelisle, OT is designed around the concept of "low trust servers".  Meaning the server shouldn't be ABLE to change your balance, or steal your bitcoins, or "get hacked" and cause any harm.
 83 2011-08-27 06:12:37 <FellowTraveler> Unfortunately, many in the bitcoin community still have not learned their lesson on that point  as evidence see recent events at MtGox and MyBitcoin.
 84 2011-08-27 06:12:39 <arcatan> TuxBlackEdo: yeah (i don't know if cron likes zero-prefixed minutes, though)
 85 2011-08-27 06:13:20 <cjdelisle> yeap
 86 2011-08-27 06:13:46 <cjdelisle> Do you think that the server could be replaced by some kind of blockchain?
 87 2011-08-27 06:13:53 <TuxBlackEdo> arcatan, it it bad to just run the crond.hourly every six hours instead? the only other bash script in there is mcelog.cron, is it ok to run that every six hours instead of every hour?
 88 2011-08-27 06:15:13 <FellowTraveler> cjdelisle, a blockchain is a useful tool which runs p2p.  Chaumian blinding (untraceable cash) requires a server.  So the best I can do, while providing chaumian cash, is invent low-trust servers that are able to interact with the blockchain, yet that cannot individually steal people's bitcoins.
 89 2011-08-27 06:15:37 <cjdelisle> hmm indeed
 90 2011-08-27 06:16:30 <FellowTraveler> I have written extensively on my wiki and on the bitcoin board about this idea of using voting pools between OT servers, in order to eliminate the current risk where any single server can just disappear with everyone's money.
 91 2011-08-27 06:16:52 <FellowTraveler> And yes, for that to work, requires interaction with the blockchain as well as the BTC script language.
 92 2011-08-27 06:17:01 <arcatan> TuxBlackEdo: mcelog checks for hardware errors. i've never used it, but i'm under the impression that it's supposed to be run even more often than hourly.
 93 2011-08-27 06:17:45 <arcatan> though, looking at its homepage, they recommend running it as a daemon and not as a cronjob
 94 2011-08-27 06:17:48 <arcatan> http://mcelog.org/index.html
 95 2011-08-27 06:18:53 <cjdelisle> I could ask a bunch of questions but I'm tired and I don't fully remember the stuff I read so I'll spare you for now ;)
 96 2011-08-27 06:19:40 <FellowTraveler> cjdelisle also, FYI, the test GUI is not what I'd call slick, it's just for testing. What's cool about OT isn't the ugly test GUI. It's the ricardian contracts, the chaumian blinding, the destruction of acct history, the markets, cheques, and other instruments, and someday soon, the bitcoin voting pools.
 97 2011-08-27 06:21:13 <cjdelisle> Well by click I don't mean pretty, I mean there is a hell of a lot that a real person can do with that even if they do trust mtgox with their savings.
 98 2011-08-27 06:21:17 <cjdelisle> *slick
 99 2011-08-27 06:21:38 <cjdelisle> (can and might reasonably want to)
100 2011-08-27 06:23:21 <FellowTraveler> I prefer to use sites that I don't have to trust, instead of sites that I do have to trust
101 2011-08-27 06:23:24 <FellowTraveler> when it comes to storing my money.
102 2011-08-27 06:24:06 <FellowTraveler> What steps are MtGox actively taking to insure that I don't have to trust them, in order to safely store my BTC on their site?
103 2011-08-27 06:24:20 <cjdelisle> Yup. That is a complex consept which a lot of people will likely have trouble grasping.
104 2011-08-27 06:25:25 <FellowTraveler> I think once everyone's money disappears a few times, they will start to grasp it.
105 2011-08-27 06:25:32 <FellowTraveler> I'll keep merrily working away in the meantime.
106 2011-08-27 06:25:41 <cjdelisle> Once it's commonplace, it's obvious... "Hey, why don't you use OT, what do you want my money for so bad?" but until then it will be an uphill climb.
107 2011-08-27 06:29:18 <phantomcircuit> FellowTraveler, still haven't solved the issuer inflating the monetary base problem though?
108 2011-08-27 06:29:51 <FellowTraveler> phantomcircuit, I believe the solution there is a real-time auditing protocol.
109 2011-08-27 06:30:24 <phantomcircuit> im too tired to think about how that would help
110 2011-08-27 06:30:40 <FellowTraveler> Here's why:
111 2011-08-27 06:31:08 <FellowTraveler> Because even if the server inflates currency (say, using a dummy account), it cannot SPEND that currency without allowing it to slip into the general population of accounts.
112 2011-08-27 06:31:13 <phantomcircuit> http://www.c4l.co.uk/colocation/1u-colocation
113 2011-08-27 06:31:14 <phantomcircuit> lol
114 2011-08-27 06:31:15 <FellowTraveler> Once that happens, it will turn up on an audit.
115 2011-08-27 06:31:18 <phantomcircuit> connection 10MB
116 2011-08-27 06:31:21 <phantomcircuit> the hell does that mean
117 2011-08-27 06:31:23 <FellowTraveler> Therefore, real-time auditing will instantly stop any such business.
118 2011-08-27 06:31:51 <phantomcircuit> ok now how would you implement that?
119 2011-08-27 06:32:23 <FellowTraveler> Basically that means for Bitcoin-based currencies on OT (where the "issuer" becomes a voting pool) that means the other servers in the pool must share info on transactions, in order to perform real-time auditing.
120 2011-08-27 06:33:00 <FellowTraveler> For normal (gold based) issuers, the OT server must share info with the Issuer to perform that audit.
121 2011-08-27 06:33:18 <FellowTraveler> Either way, you either do it once per day, with rollback ability  or you do it in REAL TIME, so no one can ever get out of whack.
122 2011-08-27 06:33:31 <phantomcircuit> ok and now if the servers and issuer work in collusion...?
123 2011-08-27 06:34:48 <FellowTraveler> 1) there is nothing to stop physical (gold) issuers from stealing, if they are not physically audited  it is the job of the market to decide which issuers they trust, based on auditing procedures, jurisdiction, etc.
124 2011-08-27 06:35:17 <phantomcircuit> yeah that seems fairly obvious
125 2011-08-27 06:35:21 <FellowTraveler> 2) For Bitcoin, the voting pool acts as the "issuer".  In THAT case, then you would need 9 out of 10 servers (or whatever the vote threshold is set to) to collude
126 2011-08-27 06:35:31 <phantomcircuit> it is possible to make sure a bitcoin issuer has not yet stolen funds fairly trivially
127 2011-08-27 06:35:46 <FellowTraveler> If there are, say, 50 servers in the pool, and 45 of them collude to steal from the pool, then YES they would be able to do that.
128 2011-08-27 06:35:57 <phantomcircuit> i assume the servers need to be authorized by the issuer
129 2011-08-27 06:36:06 <FellowTraveler> yes.
130 2011-08-27 06:36:20 <FellowTraveler> But in today's world, on sites such as mybitcoin and mtgox, you only need ONE server to "collude" or "get hacked" and you are in a world of hurt.
131 2011-08-27 06:36:26 <phantomcircuit> so that seems fairly trivial in reality for all the servers authorized by an issuer to be compromised
132 2011-08-27 06:36:33 <FellowTraveler> I think that 45 out of 50 is an improvement.
133 2011-08-27 06:36:44 <phantomcircuit> yes and no
134 2011-08-27 06:37:11 <FellowTraveler> Just because 50 servers have had a currency issued, on all 50, by the same issuer, doesn't mean that hacking that issuer gives you power over those 50 servers.
135 2011-08-27 06:37:15 <FellowTraveler> You'd still have to hack all 45 of them.
136 2011-08-27 06:37:57 <FellowTraveler> And you'd still have to trust that 45 of them aren't secretly owned by the same entity.
137 2011-08-27 06:38:15 <phantomcircuit> that last point was what i was thinking
138 2011-08-27 06:38:33 <FellowTraveler> Use a large pool?
139 2011-08-27 06:38:36 <FellowTraveler> That's the bitcoin solution it seems.
140 2011-08-27 06:38:52 <FellowTraveler> 1000 servers.
141 2011-08-27 06:38:53 <phantomcircuit> the thing i would be afraid of would be placing trust in something which can appear much more secure than it really is
142 2011-08-27 06:39:00 <FellowTraveler> That is fair.
143 2011-08-27 06:39:02 <phantomcircuit> at least with something like mtgox people understand their risk
144 2011-08-27 06:39:30 <FellowTraveler> Perhaps people should only trust a pool where all the server owners are publicly known, in their jurisdictions
145 2011-08-27 06:39:48 <FellowTraveler> instead of pools full of anonymous servers.
146 2011-08-27 06:40:05 <phantomcircuit> it's much easier to take people than you'd think
147 2011-08-27 06:40:16 <phantomcircuit> at least enough that it would stand up to internet police
148 2011-08-27 06:41:22 <MrTiggr> :D
149 2011-08-27 06:41:32 <Blitzboom> FellowTraveler: i remember you were the only socialist in the forum & has your ideological stance changed?
150 2011-08-27 06:41:36 <Blitzboom> offtopic, i know :P
151 2011-08-27 06:41:51 <Blitzboom> just wondering
152 2011-08-27 06:41:55 <phantomcircuit> and i think i should mention that i have experience in faking people for hilarity
153 2011-08-27 06:42:03 <FellowTraveler> BlitzBoom, you must be remembering some other individual, since I do not believe in group rights over individual rights.
154 2011-08-27 06:42:19 <Blitzboom> aah, i did
155 2011-08-27 06:43:40 <FellowTraveler> A group is only a collection of individuals.  If an action results in the violation of the rights of specific individuals, in return for some esoteric "group" benefit, I view that the same as splitting a baby in half.
156 2011-08-27 06:44:00 <FellowTraveler> The purpose of OT is to better help protect people's rights.
157 2011-08-27 06:44:05 <FellowTraveler> I do not trust banks to do so.
158 2011-08-27 06:45:48 <Blitzboom> what does OT stand for?
159 2011-08-27 06:46:14 <Blitzboom> also i agree with your approach
160 2011-08-27 06:46:33 <Blitzboom> trust should not be required
161 2011-08-27 06:46:42 <FellowTraveler> I am glad to see that cryptography and mathematics seem more suited to protecting individual freedoms than to protecting Rousseau's social contract, which perhaps is a sign of which is favored more by natural law.
162 2011-08-27 06:47:06 <FellowTraveler> BlitzBoom FYI OT stands for Open-Transactions, available here:   https://github.com/FellowTraveler/Open-Transactions/wiki
163 2011-08-27 06:47:20 <Blitzboom> ah okay, ive heard of that
164 2011-08-27 06:47:37 <Blitzboom> even saw that video yesterday or so
165 2011-08-27 06:53:25 <FellowTraveler> Seems like, if you just balance the size of the pool, so that the amount of money you could get by stealing it, is much smaller than what 45 server operators out of 50 are willing to lose in transaction fees by continued operation, that would help.
166 2011-08-27 06:53:38 <FellowTraveler> They have to balance the benefit of stealing the pool, against the loss of all the transaction fees that were coming in.
167 2011-08-27 07:16:03 <Graet> FellowTraveler> Perhaps people should only trust a pool where all the server owners are publicly known, in their jurisdictions<FellowTraveler> instead of pools full of anonymous servers.  << i agree, i registered a company to run my pool and comply with local laws
168 2011-08-27 07:17:09 <Graet> other btc businesses are doing it properly pools seem to be lagging in that field
169 2011-08-27 07:24:24 <upb> right, i look forward to this years reports of some companies..:p
170 2011-08-27 07:25:09 <upb> the entity used for eur deposits of an exchange had a revenue of $100 last year
171 2011-08-27 07:25:10 <Graet> interesting times for sure :)
172 2011-08-27 08:48:29 <lfm> interesting, bitcoin defines a year as 364.58 days
173 2011-08-27 08:50:58 <tcatm> lfm: ?
174 2011-08-27 08:52:33 <lfm> 210000 blocks in four years, 144 blocks/day
175 2011-08-27 08:53:04 <tcatm> does it say 21e4 blocks == 4 yours anywhere?
176 2011-08-27 08:53:14 <lfm> roughly
177 2011-08-27 08:53:28 <lfm> thats the period when it adjusts the rewards
178 2011-08-27 09:27:45 <fingster> anybody?
179 2011-08-27 09:28:08 <tcatm> ;seen bluematt
180 2011-08-27 09:28:27 <fingster> Bitcoin Express is submitted to app store, but rejected
181 2011-08-27 09:28:45 <gribble> BlueMatt was last seen in #bitcoin-dev 1 day, 16 hours, 39 minutes, and 51 seconds ago: <BlueMatt> luke-jr: this is coinbase for what, eligius blocks?
182 2011-08-27 09:28:45 <tcatm> ;;seen BlueMatt