1 2012-01-05 00:18:19 <CIA-100> DiabloMiner: Patrick McFarland master * rf7c2932 / src/main/java/com/diablominer/DiabloMiner/DiabloMiner.java : Improved error handling for argument parsing - http://git.io/fMxVXg https://github.com/Diablo-D3/DiabloMiner/commit/f7c2932f3335af4441dd6baaddbc7b599aae33ee
  2 2012-01-05 01:47:27 <gmaxwell> gavin:
  3 2012-01-05 01:49:52 <BlueMatt> gmaxwell: hes offline...
  4 2012-01-05 01:56:35 <gmaxwell> BlueMatt: he reads the logs.
  5 2012-01-05 02:06:10 <BlueMatt> gmaxwell: ;;later tell?
  6 2012-01-05 02:06:46 <gmaxwell> BlueMatt: eh, can't pass a patch that way. I'll just email him a patch.
  7 2012-01-05 02:11:45 <gmaxwell> Weird txn: https://blockexplorer.com/tx/dbef465eee9ec8c29884ab56547aa6de0ac294c1f05830bd927f0d060b54e415
  8 2012-01-05 02:11:49 <gmaxwell> (look at the fees)
  9 2012-01-05 02:17:13 <BlueMatt> gmaxwell: ok, some idiot paid a huge fee, so?
 10 2012-01-05 02:19:24 <gmaxwell> yea, yea, not _that_ interesting.
 11 2012-01-05 02:19:40 <gmaxwell> someone is saying there is a block with 305 in fees, I'm looking for it now.
 12 2012-01-05 02:24:36 <BlueMatt> gmaxwell: if the client made him pay that fee, it would be crazy, but clearly the guys an idiot...
 13 2012-01-05 02:25:50 <gmaxwell> yea, though I wonder what brand of idiocy caused that.
 14 2012-01-05 02:26:22 <gmaxwell> for example, is the fee field poorly localized so that , and . are being confused in some way?
 15 2012-01-05 02:27:02 <BlueMatt> hmm, yea
 16 2012-01-05 02:27:24 <gmaxwell> I'm sort of tempted to write a patch to cap the fee at some high but sane amount to prevent foot-gunning.
 17 2012-01-05 02:27:29 <gmaxwell> e.g. 1 BTC or something like that.
 18 2012-01-05 02:28:02 <BlueMatt> free market ;)
 19 2012-01-05 02:29:30 <gmaxwell> oh, well, if people _want_ to do that I don't care... I would just feel sad if people accidentally did it due to a typo, confusing interface, etc.
 20 2012-01-05 02:30:45 <BlueMatt> I suppose putting it in the satoshi client wouldnt be objectionable as long as we accept whatever crap we get
 21 2012-01-05 02:31:04 <doublec> I've seen some 48 btc fee transactions
 22 2012-01-05 02:31:04 <gmaxwell> oh yea, accept is fine.
 23 2012-01-05 02:31:06 <gmaxwell> https://blockexplorer.com/tx/1d7749c65c90c32f5e2c036217a2574f3f4403da39174626b246eefa620b58d9
 24 2012-01-05 02:31:11 <doublec> nothing as high as 300 though
 25 2012-01-05 02:31:11 <nanotube> maybe the guy was washing btc by putting stuff into fees and mining his own block.
 26 2012-01-05 02:31:15 <gmaxwell> How about 171 btc?
 27 2012-01-05 02:31:23 <BlueMatt> holy shot
 28 2012-01-05 02:31:30 <BlueMatt> s/ho/hi/
 29 2012-01-05 02:31:32 <doublec> wow, what block is that?
 30 2012-01-05 02:31:50 <gmaxwell> https://blockexplorer.com/block/00000000000009736897988613310282732f87d22a22533940d3d986257d90e1
 31 2012-01-05 02:34:12 <BlueMatt> nanotube: heh, so that narrows it down to what, like 3 people?
 32 2012-01-05 02:34:39 <gmaxwell> ... do I have nanotube on ignore?
 33 2012-01-05 02:34:50 <gmaxwell> oh there is nanotube
 34 2012-01-05 02:34:53 <BlueMatt> why the hell would you do that?
 35 2012-01-05 02:35:17 <gmaxwell> BlueMatt: want to test to see if people will fork the chain to steal fees? :)
 36 2012-01-05 02:35:22 <nanotube> BlueMatt: not really, you can mine a block solo within a week with a fairly modest amount of hash power.
 37 2012-01-05 02:35:55 <nanotube> gmaxwell loves me, he wouldn't ignore me :)
 38 2012-01-05 02:36:10 <BlueMatt> gmaxwell: heh, I wouldnt put it past luke or Tycho...
 39 2012-01-05 02:36:15 <BlueMatt> and probably slush too...
 40 2012-01-05 02:36:47 <BlueMatt> nanotube: meh, what would I know, I have no hash power...
 41 2012-01-05 02:37:17 <nanotube> well, bc,calc knows ;)
 42 2012-01-05 02:37:21 <gmaxwell> blocks since 155000 with >1 btc in fees: http://pastebin.com/aFm7J1we
 43 2012-01-05 02:37:50 <BlueMatt> 221, my god...
 44 2012-01-05 02:38:02 <nanotube> mm that's quite a few
 45 2012-01-05 02:38:19 <gmaxwell> BlueMatt: thats the payment not the fee.
 46 2012-01-05 02:38:27 <gmaxwell> 221 is the 171 btc fee block
 47 2012-01-05 02:38:58 <BlueMatt> gmaxwell: oh
 48 2012-01-05 02:39:10 <BlueMatt> still, it might be btc-washing
 49 2012-01-05 02:39:10 <gmaxwell> two txn with exactly 42.94967296 in fees.
 50 2012-01-05 02:40:54 <gmaxwell> This address is involved in many of them: https://blockexplorer.com/address/18acFQe4qGB7wEhApeSHcx49Aeh76y137B
 51 2012-01-05 02:41:35 <BlueMatt> uh oh...
 52 2012-01-05 02:41:45 <BlueMatt> someone is badly btc-washing...
 53 2012-01-05 02:42:14 <gmaxwell> could just be some broken ass code.
 54 2012-01-05 02:42:24 <nanotube> well, the whole concept is not that great - you know that any block that had ridiculous amount of fees is probably self-mined
 55 2012-01-05 02:42:30 <nanotube> so it doesn't do much for 'washing'
 56 2012-01-05 02:43:43 <gmaxwell> this persons txns are nuts
 57 2012-01-05 02:43:52 <gmaxwell> they keep sending fees from some address to the same address.
 58 2012-01-05 02:44:00 <gmaxwell> e.g. look at one of their other addresses: https://blockexplorer.com/address/15kDcr9wDhZmvPufKP7vYyELSoLg4bKbAR
 59 2012-01-05 02:44:29 <gmaxwell> like someone trying hard to 'wash' coins by sending the coins to themselves.
 60 2012-01-05 02:44:43 <gmaxwell> while simultaniously being set to 10 btc fee per kilobyte.
 61 2012-01-05 02:45:56 <BlueMatt> they are probably just having fun...
 62 2012-01-05 02:46:24 <gmaxwell> well, if they want to lose lots of btc having fun they can send them to me!
 63 2012-01-05 02:48:26 <nanotube> heh let's find this person and ask him! blockchain analyzer bot, go! :)
 64 2012-01-05 02:51:05 <gmaxwell> Luke doesn't have enough hash power that forking is net profitable for 171 btc fee, I think.
 65 2012-01-05 02:51:39 <nanotube> hehe maybe tycho
 66 2012-01-05 02:51:45 <gmaxwell> tycho, sure.
 67 2012-01-05 02:57:28 <gmaxwell> BlueMatt: typedef std::basic_string<char, std::char_traits<char>, secure_allocator<char> > SecureString;
 68 2012-01-05 02:57:53 <BlueMatt> gmaxwell: and?
 69 2012-01-05 02:59:48 <gmaxwell> oh it's only used for the walletpass, nevermind.
 70 2012-01-05 03:00:10 <gmaxwell> I didn't actually check to see where it was used, I'd just assumed that since it was in another file it was an oversight.
 71 2012-01-05 03:00:18 <gmaxwell> (I'm giving your patch a final review before I ACK)
 72 2012-01-05 03:00:22 <gmaxwell> Sorry.
 73 2012-01-05 03:00:41 <BlueMatt> its fine
 74 2012-01-05 03:20:50 <BlueMatt> gmaxwell: thanks, feel like doing https://github.com/bitcoin/bitcoin/pull/593 /
 75 2012-01-05 03:20:51 <BlueMatt> ?
 76 2012-01-05 03:21:24 <BlueMatt> gmaxwell: Ill pay you a few btc...
 77 2012-01-05 03:22:42 <gmaxwell> I don't even have bitcoin on anything with X on it. :)
 78 2012-01-05 03:22:52 <gmaxwell> I'll test lots of stuff, stuff that depends on the UI.. meh.
 79 2012-01-05 03:23:41 <BlueMatt> god, how hard can it be to get damn testers on a patch that has URL support...
 80 2012-01-05 03:24:24 <gmaxwell> hit the forum, at worst you might have to build it for someone.
 81 2012-01-05 03:29:47 <luke-jr> BlueMatt: learning the hard way what happens when you snub the things important to others? :P
 82 2012-01-05 03:29:48 <gmaxwell> BlueMatt: what was the motivation behind using the boost ipc instead of our existing rpc?
 83 2012-01-05 03:30:06 <luke-jr> gmaxwell: Bitcoin-Qt doesn't support JSON-RPC by default
 84 2012-01-05 03:30:33 <gmaxwell> luke-jr: yes, but it could.. e.g. fill bitcoin.conf with random username and password if there isn't one.
 85 2012-01-05 03:30:54 <BlueMatt> that
 86 2012-01-05 03:30:56 <BlueMatt> luke-jr: only person I snubbed was you...
 87 2012-01-05 03:30:57 <luke-jr> gmaxwell: it's a security risk.
 88 2012-01-05 03:31:19 <luke-jr> BlueMatt: and I would be gungho on getting that tested, if it didn't intentionally exclude what I cared about
 89 2012-01-05 03:31:29 <gmaxwell> I suppose, is the boost IPC not also a security risk?
 90 2012-01-05 03:31:44 <luke-jr> gmaxwell: no, since it can only open a dialog
 91 2012-01-05 03:31:57 <BlueMatt> gmaxwell: and only one/second
 92 2012-01-05 03:32:07 <gmaxwell> For now what underlying mechenism does it use? can another user on the same system reach it?
 93 2012-01-05 03:32:28 <gmaxwell> (sorry for the dumb questions, these are the things I'd try testing out if I was actually going to run it. :) )
 94 2012-01-05 03:33:28 <BlueMatt> if boost::ipc is poorly written...
 95 2012-01-05 03:33:56 <gmaxwell> does the confirmation UI dialog delay agreement like firefox's addon install dialog?
 96 2012-01-05 03:34:39 <gmaxwell> BlueMatt: oh good, and since we know all our libraries are unable to cause us unexpected security problems& ;)   /me googles
 97 2012-01-05 03:34:50 <BlueMatt> gmaxwell: good point...
 98 2012-01-05 03:34:58 <BlueMatt> it has a confirmation window before sending
 99 2012-01-05 03:35:11 <BlueMatt> (that lists the list of sends)
100 2012-01-05 03:36:09 <gmaxwell> BlueMatt: seems that the best UI practice these days for security-important webapps is to grey out the agree button for a small timeout to avoid the click-click-click.
101 2012-01-05 03:36:37 <gmaxwell> What does this do if I'm running multiple copies of bitcoind? (e.g. -testnet and regular.. or just two regular with different datadir?)
102 2012-01-05 03:43:43 <gmaxwell> looks like it would get handled by one at random. .. maybe even by two if there is a race in the message queue implementation, its not obvious to me from the docs.
103 2012-01-05 03:43:48 <BlueMatt> gmaxwell: first one opened gets the url
104 2012-01-05 03:44:08 <BlueMatt> oh, wait no
105 2012-01-05 03:44:17 <gmaxwell> I'd suggest adding the datadir path and testnet to the queue name.
106 2012-01-05 03:44:20 <gmaxwell> I don't .. think so.
107 2012-01-05 03:44:30 <BlueMatt> open_or_create...
108 2012-01-05 03:44:37 <gmaxwell> I think its the first one to poll.
109 2012-01-05 03:45:55 <BlueMatt> ...yea its very random...
110 2012-01-05 03:46:11 <gmaxwell> +    char strBuf[257];
111 2012-01-05 03:46:37 <gmaxwell> oops.
112 2012-01-05 03:47:46 <BlueMatt> what about it, the size is 256
113 2012-01-05 03:48:01 <BlueMatt> mq = new message_queue(open_or_create, "BitcoinURL", 2, 256);
114 2012-01-05 03:48:26 <gmaxwell> That was a mispaste. sorry.
115 2012-01-05 04:00:12 <luke-jr> not a show-stopper IMO, but ideally it could detect the address version&
116 2012-01-05 04:00:25 <luke-jr> BitcoinURI<versionnumber>
117 2012-01-05 04:00:32 <luke-jr> also note that bitcoin: URIs are not URLs :p
118 2012-01-05 04:01:19 <BlueMatt> yea but people use them interchangeably...
119 2012-01-05 04:02:09 <gmaxwell> luke-jr: what would the detection do?
120 2012-01-05 04:03:55 <luke-jr> BlueMatt: only clueless people
121 2012-01-05 04:04:21 <luke-jr> gmaxwell: well, BitcoinURI_0 would only be opened by mainnet, and BitcoinURI_114 would only be testnet
122 2012-01-05 04:04:25 <luke-jr> or whatever version testnet is
123 2012-01-05 04:04:46 <luke-jr> would need to be sure to listen for the OP_EVAL versions too
124 2012-01-05 04:05:48 <gmaxwell> meh, what happens if you run two mainnet clients with different datadirs (e.g. for different wallets?)
125 2012-01-05 04:05:57 <gmaxwell> thats why I was suggesting using the datadir in the name.
126 2012-01-05 04:09:23 <luke-jr> gmaxwell: that doesn't make sense tho
127 2012-01-05 04:09:35 <luke-jr> a URI doesn't have a datadir
128 2012-01-05 04:09:46 <luke-jr> it's equally valid for either
129 2012-01-05 04:12:23 <gmaxwell> luke-jr: ... not as part of the URL, but just part of the bitcoin invocation.
130 2012-01-05 04:12:44 <gmaxwell> e.g. if you're running multiple copies of bitcoin it's up to you to tell the handler about the right one.
131 2012-01-05 04:30:00 <BlueMatt> gmaxwell: ok, only opens the last bitcoin instance now
132 2012-01-05 04:30:01 <BlueMatt> (would prefer the first, but I couldnt find a way in boost::message_queue
133 2012-01-05 04:39:43 <BlueMatt> how long is a secret again?
134 2012-01-05 04:40:39 <BlueMatt> 32
135 2012-01-05 04:43:26 <luke-jr> gmaxwell: the bitcoin invocation is implicit. it happens when you click a link.
136 2012-01-05 05:50:03 <CIA-100> bitcoin: Mqrius * rec842b63b87e supybot-bitcoin-marketmonitor/MarketMonitor/ (config.py plugin.py): Transaction collapse, Loop sleep http://tinyurl.com/84ys4te
137 2012-01-05 05:50:04 <CIA-100> bitcoin: Daniel Folkinshteyn * rb2a86e0a7203 supybot-bitcoin-marketmonitor/MarketMonitor/ (config.py plugin.py): MarketMonitor: minor style changes http://tinyurl.com/7nz7bh5
138 2012-01-05 06:00:05 <CIA-100> bitcoin: Daniel Folkinshteyn * ra416afc6c789 supybot-bitcoin-marketmonitor/MarketMonitor/plugin.py: MarketMonitor: expand field width for market name http://tinyurl.com/788ch9r
139 2012-01-05 09:47:30 <epscy> bitcoin uris?
140 2012-01-05 10:00:16 <CIA-100> bitcoin: Con Kolivas * rec8501a79cd8 cgminer/main.c: Should the donation pool fail, don't make the fallover pool behave as though the primary pool is lagging. http://tinyurl.com/7c2avn3
141 2012-01-05 10:50:14 <CIA-100> bitcoin: p2k * r61ddb2f4e1d8 ecoinpool/ (7 files in 5 dirs): Smaller Improvements http://luke.dashjr.org/programs/bitcoin/w/ecoinpool.git/commitdiff/61ddb2f4e1d87f603b594a78867fc912bf3ffd07
142 2012-01-05 11:00:21 <CIA-100> bitcoin: p2k * re925e9204cb4 ecoinpool/ (.gitignore README.md): Smaller Improvements http://tinyurl.com/7wbvwcr
143 2012-01-05 12:37:32 <CIA-100> bitcoin: Gavin Andresen master * r7486c64 / src/serialize.h : Merge pull request #740 from TheBlueMatt/perf ... https://github.com/bitcoin/bitcoin/commit/7486c64dd8436febbe59e82dbb875e83ad6b5194
144 2012-01-05 12:37:34 <CIA-100> bitcoin: Matt Corallo master * rf7a9a11 / src/serialize.h : Fix horrific performance found by gmaxwell. - http://git.io/YCoqwg https://github.com/bitcoin/bitcoin/commit/f7a9a11391b18e8d18fce4482a4fcb6a7d6c9796
145 2012-01-05 13:16:08 <scorpius_> the links at the bottom of this page are broken. can someone fix these? http://www.weusecoins.com/mining-guide.php
146 2012-01-05 13:16:22 <scorpius_> links to github
147 2012-01-05 13:24:30 <da2ce7> 2|It would be great to maintain patches for the changes listed in: https://en.bitcoin.it/wiki/Hardfork_Wishlist
148 2012-01-05 13:24:59 <da2ce7> 2|and have development chanins to test them.
149 2012-01-05 15:10:16 <CIA-100> bitcoin: p2k * rab3c6342d519 ecoinpool/ (8 files in 6 dirs): Smaller Improvements http://tinyurl.com/83z7866
150 2012-01-05 15:26:26 <JImmy06> hey guys
151 2012-01-05 15:26:53 <JImmy06> i just read a log where someone was going on about centos and jansson lib
152 2012-01-05 15:27:41 <JImmy06> libjansson.so
153 2012-01-05 15:30:52 <jimmy06> jrmithdobbs: You there m8 ?
154 2012-01-05 16:00:59 <ryannathans> my client won't download any more blocks
155 2012-01-05 16:01:15 <ryannathans> 5.0.1
156 2012-01-05 16:01:35 <ryannathans> stuck on 160504
157 2012-01-05 16:03:35 <pierre`> natural selection
158 2012-01-05 16:04:47 <sipa> ryannathans: tried restarting?
159 2012-01-05 16:05:18 <ryannathans> sipa: many times
160 2012-01-05 16:05:39 <ryannathans> it's been like this for days
161 2012-01-05 16:05:45 <ryannathans> 8 connections to btc network
162 2012-01-05 16:05:49 <ryannathans> sorry
163 2012-01-05 16:05:50 <sipa> has your system crashed while downloading?
164 2012-01-05 16:05:53 <ryannathans> 76 connections to btc network
165 2012-01-05 16:05:55 <ryannathans> sipa: no
166 2012-01-05 16:10:10 <luke-jr> ryannathans: OS?
167 2012-01-05 16:13:39 <ryannathans> luke-jr: windows 7 pro 64bit
168 2012-01-05 16:13:57 <luke-jr> ryannathans: you tried restarting the client?
169 2012-01-05 16:14:07 <ryannathans> luke-jr: over a period of 2 days
170 2012-01-05 16:14:14 <ryannathans> computer and client many times
171 2012-01-05 16:16:13 <luke-jr> please shutdown the client entirely, reboot the computer (be sure it doesn't auto-start after rebooting!), ZIP up your bitcoin data directory (minus wallet.dat), and post it on a GitHub Issue
172 2012-01-05 16:20:10 <ryannathans> luke-jr: roger that.
173 2012-01-05 16:21:03 <ryannathans> luke-jr: err you mean the %appdata% one?
174 2012-01-05 16:21:08 <luke-jr> probably
175 2012-01-05 16:21:12 <luke-jr> the directory with wallet.dat
176 2012-01-05 16:21:19 <luke-jr> just don't include wallet.dat in the zip
177 2012-01-05 16:21:43 <ryannathans> these blk0001.dat and blkindex.dat better bloody compress a heap
178 2012-01-05 16:23:00 <da2ce7> ryannathans: about 30% rite from memoroy?
179 2012-01-05 16:23:33 <ryannathans> depends what i compress them with ;D
180 2012-01-05 16:25:12 <CIA-100> bitcoin: Matt Corallo 0.4.x * r99e9601e805c bitcoind-stable/src/serialize.h: Fix horrific performance found by gmaxwell. http://tinyurl.com/6uv7zyr
181 2012-01-05 16:27:08 <luke-jr> ryannathans: just not RAR
182 2012-01-05 16:27:40 <ryannathans> I hate rar
183 2012-01-05 16:44:29 <ryannathans> 715 MB, really?
184 2012-01-05 16:44:32 <ryannathans> 90kb/s up
185 2012-01-05 16:44:53 <ryannathans> so that's...
186 2012-01-05 16:45:10 <ryannathans> 2 1/2 hours
187 2012-01-05 16:45:28 <ryannathans> i can't upload now, low on bandwidth
188 2012-01-05 16:48:23 <gmaxwell> If you can't handle running a bitcoin node, use a thin client.
189 2012-01-05 16:48:55 <ryannathans> this client?
190 2012-01-05 16:48:57 <ryannathans> thin*
191 2012-01-05 16:49:30 <gmaxwell> e.g. http://ecdsa.org/electrum/
192 2012-01-05 16:49:59 <luke-jr> gmaxwell: the topic is him uploading his .bitcoin for a bug report
193 2012-01-05 16:50:14 <jeremias> does electrum support multiple addresses btw
194 2012-01-05 16:50:22 <gmaxwell> oh. I'm sorry I missed the backscroll.
195 2012-01-05 16:50:48 <luke-jr> jeremias: wtf?
196 2012-01-05 16:51:36 <gmaxwell> luke-jr: did you get the debug log from him?
197 2012-01-05 16:51:45 <luke-jr> gmaxwell: no
198 2012-01-05 16:51:49 <luke-jr> that's part of .bitcoinj
199 2012-01-05 16:51:51 <luke-jr> that's part of .bitcoin
200 2012-01-05 16:52:00 <gmaxwell> luke-jr: it's probably a case of a bit error in an already validated block which prevents 160505 from confirming.
201 2012-01-05 16:52:19 <luke-jr> gmaxwell: how is that possible?
202 2012-01-05 16:52:49 <ryannathans> where do you want it uploaded
203 2012-01-05 16:52:53 <gmaxwell> memory corruption / sata bus corruption / memory corruption in the integrated electronics on the hdd.
204 2012-01-05 16:53:06 <luke-jr> ryannathans: see if GitHub takes it >:D
205 2012-01-05 16:53:08 <ryannathans> gmaxwell: would be the first time..
206 2012-01-05 16:53:22 <gmaxwell> luke-jr: there was another case of a bit error in a local node's copy of the chain reported on the forums once.
207 2012-01-05 16:53:26 <gmaxwell> ryannathans: first time?
208 2012-01-05 16:53:30 <ryannathans> luke-jr: yes
209 2012-01-05 16:53:57 <gmaxwell> ryannathans: your system could be creating errors for one in a million bits and you wouldn't tend to notice.
210 2012-01-05 16:53:57 <ryannathans> gmaxwell: ?
211 2012-01-05 16:54:38 <gmaxwell> But bitcoin notices because it validates everything.
212 2012-01-05 16:55:00 <gmaxwell> Unfortunately it doesn't recover well if something already validated becomes corrupted after the fact.
213 2012-01-05 16:55:00 <luke-jr> gmaxwell: is there an easy way to test without him uploading it? :P
214 2012-01-05 16:55:09 <gmaxwell> see what the debug log says.
215 2012-01-05 16:55:21 <ryannathans> here? https://github.com/bitcoin/bitcoin/issues
216 2012-01-05 16:55:25 <gmaxwell> If it says a block we _know_ is good (based on the hash) isn't validating then that must be the problem.
217 2012-01-05 16:55:49 <ryannathans> put it this way, i have more than 1TB of games, none of those have anything wrong, all the hashes are good as gold
218 2012-01-05 16:55:56 <ryannathans> or ever had anything wrong
219 2012-01-05 16:55:57 <luke-jr> ryannathans: yes; post the debug.log first if it's too big to upload right away
220 2012-01-05 16:56:05 <ryannathans> kay, on that page linked?
221 2012-01-05 16:56:24 <luke-jr> yeah
222 2012-01-05 16:56:44 <gmaxwell> ryannathans: so? bitflips happen, even on perfectly reliable hardware.
223 2012-01-05 17:01:48 <ryannathans> do i upload this crap after or while i submit the issue
224 2012-01-05 17:02:31 <luke-jr> dunno
225 2012-01-05 17:02:37 <luke-jr> never used GitHub issues
226 2012-01-05 17:02:39 <luke-jr> :D
227 2012-01-05 17:02:51 <ryannathans> i'll find out, xD
228 2012-01-05 17:04:11 <ryannathans> https://github.com/bitcoin/bitcoin/issues/745
229 2012-01-05 17:05:32 <gribble> New news from bitcoinrss: ryannathans opened issue 745 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/issues/745>
230 2012-01-05 17:05:47 <ryannathans> hehe
231 2012-01-05 17:08:27 <luke-jr> ryannathans: your block 152704 is probably corrupt
232 2012-01-05 17:08:37 <luke-jr> and/or the index?
233 2012-01-05 17:08:54 <luke-jr> try renaming blkindex to something else
234 2012-01-05 17:09:44 <ryannathans> what'll it be doing now?
235 2012-01-05 17:10:05 <ryannathans> downloading all teh blocks again?
236 2012-01-05 17:10:27 <luke-jr> hopefully just reindexing
237 2012-01-05 17:10:30 <luke-jr> no idea tho
238 2012-01-05 17:10:46 <gmaxwell> oh it won't reindex. you shouldn't just delete the index or you'll waste a lot of space.
239 2012-01-05 17:10:49 <ryannathans> it says it's downloading blocks 0 to 160764
240 2012-01-05 17:10:54 <luke-jr> gmaxwell: blah
241 2012-01-05 17:10:58 <gmaxwell> delete the index and the blocks
242 2012-01-05 17:12:08 <gmaxwell> key lines are:
243 2012-01-05 17:12:09 <gmaxwell> received block 0000000000000c0858f9
244 2012-01-05 17:12:12 <gmaxwell> InvalidChainFound: invalid block=0000000000000c0858f9  height=160505  work=443392239434410037
245 2012-01-05 17:12:27 <ryannathans> i don't see how this kills everything
246 2012-01-05 17:12:33 <ryannathans> shouldn't it just redownload the block
247 2012-01-05 17:12:57 <gmaxwell> 160505 isn't corrupt, some prior block you already validated is.
248 2012-01-05 17:13:10 <ryannathans> how the f did it validate?
249 2012-01-05 17:13:17 <gmaxwell> it wasn't corrupt when it validated.
250 2012-01-05 17:13:55 <gmaxwell> it's block 152704 thats bad.
251 2012-01-05 17:14:04 <gmaxwell> oh luke said that
252 2012-01-05 17:14:26 <gmaxwell> Yea, it can't tell that its bad, it can just tell that it has never seen a transaction that block 160505 is spending.
253 2012-01-05 17:14:29 <ryannathans> 15% re downloaded
254 2012-01-05 17:14:49 <Cryo> what makes fetching current blocks so slow?
255 2012-01-05 17:15:02 <luke-jr> Cryo: mlock
256 2012-01-05 17:15:10 <gmaxwell> Cryo: a number of things, including a bug I just found which was fixed today.
257 2012-01-05 17:15:18 <Cryo> is it the number of connections to the network?
258 2012-01-05 17:15:27 <gmaxwell> No. It has nothing to do with the network.
259 2012-01-05 17:15:40 <Cryo> ok, just mlock? weird.
260 2012-01-05 17:15:47 <ryannathans> who owns bitcoin.org
261 2012-01-05 17:15:48 <luke-jr> Cryo: turns out mlock is VERY slow
262 2012-01-05 17:15:55 <luke-jr> sirius
263 2012-01-05 17:15:58 <luke-jr> tcatm admins it
264 2012-01-05 17:16:03 <gmaxwell> Not just mlock, there are other things that make it slow.
265 2012-01-05 17:16:13 <luke-jr> gmaxwell: mlock made a big difference tho :P
266 2012-01-05 17:16:14 <gmaxwell> But mlock was easily fixed once we realized it.
267 2012-01-05 17:16:18 <gmaxwell> and that.
268 2012-01-05 17:16:31 <gmaxwell> Cryo: http://people.xiph.org/~greg/bitcoin-sync.png
269 2012-01-05 17:17:13 <Cryo> awesome
270 2012-01-05 17:17:13 <gmaxwell> (though thats on a quite fast system, it won't be that fast for most people, at least not until some other things are improved)
271 2012-01-05 17:17:41 <luke-jr> gmaxwell: think I should tag stable versions early, to get this fix out there?
272 2012-01-05 17:17:44 <gmaxwell> luke-jr: did you pull the recent deadlock fixes into stable too?
273 2012-01-05 17:17:51 <ryannathans> gmaxwell: I don't understand this graph
274 2012-01-05 17:18:07 <luke-jr> gmaxwell: yes
275 2012-01-05 17:18:08 <cjdelisle> wow that's pretty awesome, congrats
276 2012-01-05 17:18:16 <luke-jr> gmaxwell: also the orphans-fill-up-memory fix
277 2012-01-05 17:18:20 <gmaxwell> ryannathans: X is the number of seconds after the node started fetching blocks. Y is how many blocks it has fetched.
278 2012-01-05 17:18:50 <gmaxwell> ryannathans: with the fix it finishes the whole chain download in about 1780 seconds. Without the fix it would take many hours.
279 2012-01-05 17:19:00 <ryannathans> ocrap
280 2012-01-05 17:19:08 <ryannathans> the colours did my head in, i'm colour blind
281 2012-01-05 17:19:17 <luke-jr> gmaxwell: think I should consider the "About Qt" menu item to be a fix for 0.5.x?
282 2012-01-05 17:19:20 <ryannathans> i thought 'no mlock' was at the bottom
283 2012-01-05 17:19:33 <gavinandresen> gmaxwell: very nice detective work, by the way-- I made an effort to find the slowness a couple months ago and gave up.
284 2012-01-05 17:19:44 <gmaxwell> ryannathans: oh wow, your colorblinedness is bad enough that you can't distinguish that?
285 2012-01-05 17:19:53 <ryannathans> yup
286 2012-01-05 17:19:57 <luke-jr> gavinandresen: your thoughts on 'About Qt' and early stable releases? :P
287 2012-01-05 17:20:09 <luke-jr> I should have pokes wumpus about 'About Qt' the other night
288 2012-01-05 17:20:16 <luke-jr> he's almost never here :<
289 2012-01-05 17:20:19 <gmaxwell> gavinandresen: yea, it was .. unusually hard. I think at lot of the mlock slowness comes from constantly trashing the TLB, so it didn't show up on most profiling tools.
290 2012-01-05 17:20:29 <gavinandresen> early stable releases:  seems like a 0.5.2 is a good idea, as long as somebody else volunteers to spearhead the release process.
291 2012-01-05 17:20:32 <Cryo> mmm wumpus
292 2012-01-05 17:20:55 <ryannathans> 37% synced
293 2012-01-05 17:21:01 <gavinandresen> (I'd be happy to build and gitian-sign a 0.5.2)
294 2012-01-05 17:21:02 <luke-jr> gavinandresen: BlueMatt knows what to do for that, right?
295 2012-01-05 17:21:09 <ryannathans> all this for 12 btc.
296 2012-01-05 17:21:10 <ryannathans> lol.
297 2012-01-05 17:21:18 <gavinandresen> luke-jr: it should be documented pretty well in doc/release_process.txt
298 2012-01-05 17:21:40 <gavinandresen> luke-jr: I'd be happy to do a mac build, too.
299 2012-01-05 17:22:00 <luke-jr> gavinandresen: would you build 0.4.3 as well? ;)
300 2012-01-05 17:22:05 <gavinandresen> luke-jr: nope
301 2012-01-05 17:22:09 <luke-jr> aww
302 2012-01-05 17:22:16 <luke-jr> oh well, 0.4.x users can stick to git :p
303 2012-01-05 17:22:24 <gavinandresen> gotta draw the line somewhere, I've got lots to do
304 2012-01-05 17:22:40 <Cryo> why isn't there a buildbot? :)
305 2012-01-05 17:23:04 <luke-jr> Cryo: there is, but it doesn't do releases for some reason
306 2012-01-05 17:23:12 <ryannathans> current block number?
307 2012-01-05 17:23:21 <luke-jr> would be nice to upload sources and get it automatically built
308 2012-01-05 17:23:22 <luke-jr> ;;bc,block
309 2012-01-05 17:23:23 <gribble> Error: "bc,block" is not a valid command.
310 2012-01-05 17:23:24 <gribble> 160764
311 2012-01-05 17:23:24 <luke-jr> ;;bc,blocks
312 2012-01-05 17:23:39 <luke-jr> then I could even make next-test builds
313 2012-01-05 17:23:41 <luke-jr> :p
314 2012-01-05 17:24:00 <luke-jr> I'll wait until I see BlueMatt and let him decide
315 2012-01-05 17:24:47 <ryannathans> the new bitcoin gui is filth, it'd be so much better if it wasn't so slow
316 2012-01-05 17:25:01 <gmaxwell> The gui is slow?
317 2012-01-05 17:25:11 <ryannathans> the new one
318 2012-01-05 17:25:15 <ryannathans> qt
319 2012-01-05 17:25:19 <luke-jr> gavinandresen: if you want to preemptively build 0.5.2 assuming BlueMatt is OK with it, let me know and I'll push the latest fixes to the 0.5.x branch
320 2012-01-05 17:25:30 <Cryo> qt beats java
321 2012-01-05 17:25:31 <luke-jr> ryannathans: are you volunteering to maintain wxBitcoin?
322 2012-01-05 17:25:42 <ryannathans> what's wxBitcoin
323 2012-01-05 17:25:46 <luke-jr> ryannathans: the old client
324 2012-01-05 17:26:14 <luke-jr> exactly.
325 2012-01-05 17:26:19 <luke-jr> Nobody wants to maintain that crap.
326 2012-01-05 17:26:30 <ryannathans> I dislike nokia
327 2012-01-05 17:26:39 <luke-jr> me too, but Qt predates Nokia.
328 2012-01-05 17:26:45 <da2ce7> ok I'm off cya guys
329 2012-01-05 17:26:50 <gmaxwell> ryannathans: and please be more polite with your criticisms, the people who worked very hard on it are here and many people like it a lot. (and they're actually maintain it, no one wnats to maintain the old gui). AFAIK, you're not paying for any of this development work, so while constructive criticism is helpful the overtly negative language is discouraging and not appropriate.
330 2012-01-05 17:27:06 <ryannathans> gmaxwell: I like the new gui
331 2012-01-05 17:27:13 <ryannathans> gmaxwell: if it was faster it would be sicko
332 2012-01-05 17:27:17 <Cryo> that was the most polite stfu I've ever seen. bravo.
333 2012-01-05 17:27:19 <gavinandresen> frickin frackin binary encodings....   so I wrote the addmultisigaddress RPC command to take any of three public key encodings because I couldn't make up my mind:  bas58 hash-of-address (if key is in your wallet), full public key encoded with base58, and full public key encoded with hex.
334 2012-01-05 17:27:28 <gavinandresen> ... and now I'm regretting it.
335 2012-01-05 17:27:38 <gmaxwell> ryannathans: Whats slow about it?
336 2012-01-05 17:27:50 <ryannathans> where to start.
337 2012-01-05 17:28:05 <gmaxwell> ryannathans: with whatever you find most annoying / disruptive.
338 2012-01-05 17:28:16 <luke-jr> gmaxwell: Spesmilo is significantly slower on Windows than Linux; I assumed it was a Python threading issue, but maybe it's Qt O.o
339 2012-01-05 17:28:27 <ryannathans> if i right click the task bar icon
340 2012-01-05 17:28:33 <ryannathans> then click elsewhere, eg another window
341 2012-01-05 17:28:42 <luke-jr> ryannathans: instant for me
342 2012-01-05 17:28:43 <ryannathans> it takes up to 10 seconds to close
343 2012-01-05 17:28:53 <gmaxwell> 10 seconds? 0_o
344 2012-01-05 17:29:07 <ryannathans> the tray menu
345 2012-01-05 17:29:08 <gmaxwell> what thread does that crazy windows registry rng stuff run in?
346 2012-01-05 17:29:20 <ryannathans> ?
347 2012-01-05 17:29:28 <gmaxwell> That wasn't directed to you, obviously.
348 2012-01-05 17:29:47 <luke-jr> ryannathans: is it just Bitcoin-Qt, or are other Qt apps slow too?
349 2012-01-05 17:29:56 <ryannathans> just bitcoin-qt
350 2012-01-05 17:29:59 <ryannathans> mumble is fine
351 2012-01-05 17:30:18 <luke-jr> BTW
352 2012-01-05 17:30:34 <luke-jr> ButterFlyLabs lost a BitForce purchase via Bitcoin x.x
353 2012-01-05 17:30:46 <plato> 10:29 < plato> has there been any discussion of using torrents to distribute old blocks?
354 2012-01-05 17:30:54 <luke-jr> hmm, maybe I wasn't supposed to tell people about that
355 2012-01-05 17:31:16 <luke-jr> plato: that would be retardedly redundant.
356 2012-01-05 17:31:39 <ryannathans> i wouldn't be here still on 40%
357 2012-01-05 17:31:45 <gmaxwell> luke-jr: oh you canceled yours?
358 2012-01-05 17:32:08 <luke-jr> gmaxwell: no
359 2012-01-05 17:32:13 <plato> luke-jr: thanks
360 2012-01-05 17:32:33 <luke-jr> gmaxwell: the guy 'never paid', then weeks later decided to pay, but BFL no longer has the wallet he paid to
361 2012-01-05 17:32:40 <gmaxwell> plato: no, there doesn't appear to be any need. We already have a p2p network which should be better for block distribution than torrent.
362 2012-01-05 17:32:51 <gmaxwell> luke-jr: doh!
363 2012-01-05 17:32:52 <ryannathans> should
364 2012-01-05 17:32:53 <luke-jr> gmaxwell: I'm trying to convince them to give me raw access to the HD so I can do a scam for privkeys
365 2012-01-05 17:33:06 <gmaxwell> luke-jr: would be a good idea.
366 2012-01-05 17:33:29 <gmaxwell> ryannathans: The network part is just fine. The slow part of fetching blocks is the validation, not the transfer.
367 2012-01-05 17:33:35 <luke-jr> also, when I tested their stuff a few weeks ago, it was mining to a local wallet
368 2012-01-05 17:33:41 <luke-jr> they didn't bother to save that either&
369 2012-01-05 17:33:43 <ryannathans> socket recv error 10054
370 2012-01-05 17:33:47 <ryannathans> wtf
371 2012-01-05 17:33:54 <b4epoche> whoa?  Butterfly takes bitcoin?
372 2012-01-05 17:33:55 <luke-jr> it was only 0.06 BTC tho
373 2012-01-05 17:33:58 <luke-jr> b4epoche: yes
374 2012-01-05 17:34:06 <Joric> where i may get transaction hash in bitcointools? https://github.com/gavinandresen/bitcointools/blob/master/deserialize.py guess it's somewhere in the stream? in parse_Transaction? (line 79)
375 2012-01-05 17:34:16 <b4epoche> what conversion rate do they use?
376 2012-01-05 17:34:18 <gmaxwell> ryannathans: Where are you getting that?
377 2012-01-05 17:34:22 <luke-jr> b4epoche: MtGox
378 2012-01-05 17:34:26 <ryannathans> debug.log, spammed
379 2012-01-05 17:34:40 <ryannathans> it's stopped at 40%
380 2012-01-05 17:35:27 <ryannathans> i'm going to sleep
381 2012-01-05 17:35:31 <b4epoche> interesting.  I'm surprised a 'real' business selling hardware is accepting bitcoin.
382 2012-01-05 17:35:33 <gmaxwell> ryannathans: mouseover and get the count, is it actually stopped are just slow?
383 2012-01-05 17:35:34 <ryannathans> night, i'll be back tomorrow maybe
384 2012-01-05 17:35:55 <ryannathans> gmaxwell: in client all appears well
385 2012-01-05 17:36:05 <ryannathans> in debug.log, last 100 lines are that error
386 2012-01-05 17:36:17 <gmaxwell> That might be harmless.
387 2012-01-05 17:36:29 <ryannathans> o well, i'll let this dl overnight
388 2012-01-05 17:36:31 <ryannathans> night.
389 2012-01-05 17:36:51 <gmaxwell> K. Cya.
390 2012-01-05 17:37:26 <gmaxwell> Apparently crud like "peerblock" can cause that (according to random forum posts)
391 2012-01-05 18:39:24 <sadsadsa> mhm
392 2012-01-05 18:40:17 <Joric> gavinandresen, gmaxwell where i may get transaction hash? i was trying to read raw transaction data and sha256 it but it's different
393 2012-01-05 18:50:51 <lianj> Joric: two times sha256
394 2012-01-05 18:51:14 <ciscoftw> sup with the #bitcoind chan?
395 2012-01-05 18:52:32 <Joric> lianj, yeah found an explanation https://bitcointalk.org/index.php?topic=37532.0
396 2012-01-05 19:48:30 <CIA-100> bitcoin: Steve Brecher * r52d134ce2e4a cgminer/main.c: Add percent signs to reject ratio outputs http://tinyurl.com/82v5haf
397 2012-01-05 19:48:31 <CIA-100> bitcoin: Con Kolivas * rad54c0161a70 cgminer/main.c: Merge pull request #67 from Proofer/RejectRatioPercentSigns http://tinyurl.com/77hulrb
398 2012-01-05 20:05:11 <CIA-100> libbitcoin: Kamil Domanski * rd65c930e7889 /include/bitcoin/types.hpp: struct node_address got lost in merger http://tinyurl.com/7boagz2
399 2012-01-05 21:11:26 <tcatm> helo: creating new humans. could also easily be verified using automated genetic tests
400 2012-01-05 21:18:11 <Eliel> I think there are enough babies born in the world that it might work from that perspective :P
401 2012-01-05 21:18:35 <Eliel> no idea how to make that secure though :D
402 2012-01-05 21:19:38 <Eliel> helo: you'd need something that's relatively easily done by a human but easy and fast to verify with a computer. I'm doubtful if something like that exists
403 2012-01-05 21:20:33 <tcatm> it would be even better if humans could verify it easily, too
404 2012-01-05 21:20:52 <makomk> Eliel: CAPTCHAs would be, except they're fundamentally not decentralized in the way Bitcoin requires
405 2012-01-05 21:21:10 <helo> turing test :)
406 2012-01-05 21:22:18 <Backburn2> spoken word (like navajo) recorded
407 2012-01-05 21:22:46 <tcatm> easy to duplicate
408 2012-01-05 21:23:02 <Backburn2> worked in WWII :)
409 2012-01-05 21:23:30 <helo> last block's hash embedded in video recording?
410 2012-01-05 21:23:49 <helo> no lol
411 2012-01-05 21:24:25 <tcatm> that might work. we could even re-use our GPUs for realistic rendering of fake videos :)
412 2012-01-05 21:24:34 <Backburn2> lol
413 2012-01-05 21:30:13 <CIA-100> bitcoin: Con Kolivas * r17e8d60b5f36 cgminer/main.c: Adjust utility width to be constant maximum as well. http://tinyurl.com/7eeczwn
414 2012-01-05 21:39:27 <helo> ahh, last block recited by both participants in unison :)
415 2012-01-05 21:40:11 <CIA-100> bitcoin: Kano * r7187dfa04910 cgminer/main.c: If api-description is specified, save it when writing the config file http://tinyurl.com/6lnucl7
416 2012-01-05 21:40:12 <CIA-100> bitcoin: Con Kolivas * r95522262eddd cgminer/main.c: Merge pull request #66 from kanoi/master http://tinyurl.com/7sjjgla
417 2012-01-05 21:41:45 <luke-jr> BlueMatt: hey
418 2012-01-05 21:42:00 <luke-jr> BlueMatt: we were talking about 0.5.2 earlier: can you do the release stuff for it?
419 2012-01-05 21:42:09 <luke-jr> BlueMatt: Gavin said he can do one of the gitian builds
420 2012-01-05 22:14:05 <BlueMatt> luke-jr: oops, did I still forget to do that?
421 2012-01-05 22:14:10 <BlueMatt> will do now...
422 2012-01-05 22:14:17 <luke-jr> BlueMatt: no, I mean a new release ;P
423 2012-01-05 22:14:21 <luke-jr> 0.5.2
424 2012-01-05 22:14:31 <BlueMatt> oh, well whatever, Ill go build it
425 2012-01-05 22:14:52 <luke-jr> BlueMatt: whole release process?
426 2012-01-05 22:15:02 <BlueMatt> what do you want me to do?
427 2012-01-05 22:15:13 <luke-jr> hmm, not sure, lemme review it
428 2012-01-05 22:15:33 <luke-jr> I guess it's build + gitian stuff
429 2012-01-05 22:15:38 <luke-jr> not sure who can update bitcoin.org
430 2012-01-05 22:15:54 <BlueMatt> tcatm handles that usually
431 2012-01-05 22:17:08 <tcatm> yep. either make a pull request or send me a signed email
432 2012-01-05 22:17:24 <BlueMatt> luke-jr: is there no 0.5.2 tag?
433 2012-01-05 22:17:35 <luke-jr> BlueMatt: making tags now
434 2012-01-05 22:17:40 <luke-jr> BlueMatt: beginning with v0.4.3rc1
435 2012-01-05 22:18:04 <BlueMatt> wait, are these rcs or stables?
436 2012-01-05 22:18:49 <luke-jr> let's do a single RC just in case
437 2012-01-05 22:18:53 <luke-jr> stable RC
438 2012-01-05 22:19:05 <BlueMatt> ok, can you push those now, I have to go, but Ill start a build
439 2012-01-05 22:19:06 <luke-jr> just in case gmaxwell's optimization has some bad effect
440 2012-01-05 22:19:19 <luke-jr> v0.4.3rc1 and v0.5.0.3rc1 done
441 2012-01-05 22:19:24 <luke-jr> v0.5.2rc1 soon
442 2012-01-05 22:19:47 <BlueMatt> wait, you want me to build for 3 releases?
443 2012-01-05 22:20:14 <luke-jr> v0.5.2rc1 is probably the important one
444 2012-01-05 22:20:23 <CIA-100> bitcoin: Luke Dashjr 0.5.x * r7de7913abdba bitcoind-stable/src/ (5 files in 2 dirs): Merge branch '0.5.0.x' into 0.5.x http://tinyurl.com/7j5tq8q
445 2012-01-05 22:20:47 <luke-jr> v0.5.2rc1 tagged & pushed
446 2012-01-05 22:21:45 <luke-jr> gavinandresen: ^
447 2012-01-05 22:22:11 <gmaxwell> I don't expect bad effects, but we might get an "OH WAIT FIX THIS TOO" if we beat on the drum some for people to try it.
448 2012-01-05 22:22:51 <luke-jr> worst case we rename the file and put it on the site in a few days
449 2012-01-05 22:24:49 <gmaxwell> Sounds like a plan to me.
450 2012-01-05 22:27:01 <Joric> is there anything that could export blocks from berkeley db to json (as on blockexplorer)?
451 2012-01-05 22:27:40 <sipa1024> Joric: getblock was merged for 0.6
452 2012-01-05 22:28:21 <Joric> sipa1024, is it possible to getblock via http?
453 2012-01-05 22:29:33 <Joric> apparently not
454 2012-01-05 22:29:38 <luke-jr> Joric: all JSON-RPC is HTTP
455 2012-01-05 22:29:38 <sipa> how do you mean?
456 2012-01-05 22:30:36 <Joric> oh okay then
457 2012-01-05 22:31:51 <sipa> Joric: not sure i understood your question
458 2012-01-05 22:32:27 <graingert> do you mean via a get request
459 2012-01-05 22:32:39 <Joric> yes i need http request instead of sockets
460 2012-01-05 22:32:48 <luke-jr> does anyone have an OpenCL-enabled system I can SSH into to test some code?
461 2012-01-05 22:32:56 <sipa> Joric: http uses sockets...
462 2012-01-05 22:33:00 <luke-jr> Joric: JSON-RPC is HTTP POST
463 2012-01-05 22:38:08 <TD> gavinandresen: hiya. what's the latest version of the op_eval replacement proposal?
464 2012-01-05 22:38:18 <TD> i had to stop following it for a few days and now i'm not sure what the current thinking is
465 2012-01-05 22:38:26 <gavinandresen> TD: https://en.bitcoin.it/wiki/BIP_0016
466 2012-01-05 22:39:07 <luke-jr> gavinandresen: v0.5.2rc1 tagged, if you wanted to build it
467 2012-01-05 22:40:56 <gavinandresen> luke-jr: I may have some time tomorrow to start builds, I'm about to quit for today.
468 2012-01-05 22:42:47 <TD> gavinandresen: should OP_EQUAL be OP_EQUALVERIFY in these examples?
469 2012-01-05 22:43:02 <gavinandresen> Nope, the script has to leave a true value on the stack
470 2012-01-05 22:43:34 <TD> right, sorry
471 2012-01-05 22:43:37 <gavinandresen> ... although I suppose it could be EQUALVERIFY, the script hash will always be true.....
472 2012-01-05 22:43:54 <gavinandresen> or, rather, the signature(s)
473 2012-01-05 22:45:11 <gavinandresen> Implementation that will probably be a pull request tomorrow is here:  https://github.com/gavinandresen/bitcoin-git/tree/pay_to_script_hash
474 2012-01-05 22:46:49 <BlueMatt> luke-jr: ok, 0.5.2rc1 building
475 2012-01-05 22:46:50 <gavinandresen> Lots of eyeballs on the logic in VerifyScript very welcome:  https://github.com/gavinandresen/bitcoin-git/blob/pay_to_script_hash/src/script.cpp#L1451
476 2012-01-05 22:48:18 <graingert> has anyone proved Bitcoin yet?
477 2012-01-05 22:48:50 <BlueMatt> proved as in?
478 2012-01-05 22:49:07 <graingert> Event-B formal methods etc
479 2012-01-05 22:49:56 <BlueMatt> dont think so
480 2012-01-05 22:50:12 <CIA-100> bitcoin: p2k * r43babbfd6d62 ecoinpool/apps/ecoinpool_mysql_replicator/src/ (ecoinpool_mysql_replicator_sup.erl mycouch_replicator.erl): Improved Replicator http://tinyurl.com/72u4rqo
481 2012-01-05 22:50:13 <CIA-100> bitcoin: p2k * r63f872c98ea2 ecoinpool/apps/ecoinpool_mysql_replicator/src/mycouch_replicator.erl: Fixed a typo http://tinyurl.com/7om3h4b
482 2012-01-05 22:51:02 <TD> has anyone proved a program of any interesting complexity, ever?
483 2012-01-05 22:51:56 <graingert> TD: an entire operating system was proved by the russians
484 2012-01-05 22:52:16 <TD> what language was it written in?
485 2012-01-05 22:52:19 <luke-jr> does it do anything beyond calculator? P:P
486 2012-01-05 22:52:28 <graingert> bitcoin sounds like the sort of thing that might actually be worth proving, if posible
487 2012-01-05 22:52:31 <graingert> possible*
488 2012-01-05 22:52:38 <graingert> brb
489 2012-01-05 22:54:15 <luke-jr> I started writing a MIPS emulator that followed every possible branch&
490 2012-01-05 22:54:19 <luke-jr> anyone want to finish it? :P
491 2012-01-05 22:55:03 <CIA-100> bitcoin: Luke Dashjr * r5bbfbd8547be gentoo/dev-python/bitcoinrpc/ (Manifest bitcoinrpc-9999.ebuild): dev-python/bitcoinrpc-9999: enable Python3 support http://tinyurl.com/6q5m9qq
492 2012-01-05 22:55:05 <CIA-100> bitcoin: Luke Dashjr * r347c488a33da gentoo/net-p2p/ (4 files in 2 dirs): net-p2p/bitcoin{-qt,d}: bump to 0.5.2_rc1 http://tinyurl.com/728dny2
493 2012-01-05 22:56:14 <sipa> when did the exchange rate get to $7 again? :o
494 2012-01-05 22:56:36 <luke-jr> today
495 2012-01-05 22:56:40 <luke-jr> it's been going up for a week now
496 2012-01-05 22:56:56 <sipa> yay for speculative bubbles
497 2012-01-05 22:57:06 <luke-jr> dunno, it might be real this time
498 2012-01-05 22:57:13 <sipa> who knows :)
499 2012-01-05 22:57:23 <TD> probably anticipation of the tv show?
500 2012-01-05 22:57:44 <BlueMatt> heh, for some reason I doubt that
501 2012-01-05 22:58:02 <BlueMatt> who is going to google (and buy) a random-seeming fake-seeming currency on a soap?
502 2012-01-05 22:58:24 <BlueMatt> or more likely yahoo search judging by the average watcher of soaps...
503 2012-01-05 23:00:14 <CIA-100> bitcoin: p2k * r337322720005 ecoinpool/apps/ecoinpool_mysql_replicator/src/mycouch_replicator.erl: Unique Replicator Trigger Names http://tinyurl.com/7vl6mhg
504 2012-01-05 23:00:56 <TD> i think it's technically a legal drama rather than a soap
505 2012-01-05 23:01:02 <TD> and good question
506 2012-01-05 23:01:30 <TD> apparently this show often bases its plotlines on current affairs. it has ~12 million viewers. so if you assume some of them are also interested in current affairs, and take 1% of that
507 2012-01-05 23:01:41 <TD> then it's still some non-trivial number of people finding out for the first time
508 2012-01-05 23:01:47 <BlueMatt> mmm, thats true...
509 2012-01-05 23:01:48 <TD> that said, it's all just speculation isn't it
510 2012-01-05 23:02:09 <TD> people think it might lead to new buyers, so they buy
511 2012-01-05 23:02:18 <TD> new buyers don't care what the price is because it's just a proxy currency
512 2012-01-05 23:03:08 <BlueMatt> but I would assume the % that buy  after googling is very, very low...but then again I have no clue
513 2012-01-05 23:03:33 <BlueMatt> (seems to me like most non-highly-technical users wont buy bitcoins, usually)
514 2012-01-05 23:03:54 <BlueMatt> but it would be interesting to actually know...
515 2012-01-05 23:05:04 <CIA-100> bitcoin: Luke Dashjr * r6b79417f8edb gentoo/net-p2p/ (5 files in 2 dirs): net-p2p/bitcoin{-qt,d}: add 0.4.3_rc1 and 0.5.0.3_rc1 (bitcoin-qt only) http://tinyurl.com/7tm65pq
516 2012-01-05 23:05:13 <BlueMatt> oh, well speculation always runs rampant in a market like bitcoin where liquidity is very low...
517 2012-01-05 23:06:08 <TD> it runs rampant in markets where liquidity is high too
518 2012-01-05 23:06:20 <TD> i don't think you can have any kind of object that's a proxy for other kinds without speculation on the proxies value
519 2012-01-05 23:06:23 <BlueMatt> heh, ok speculation always controls price...
520 2012-01-05 23:08:24 <TD> hmm
521 2012-01-05 23:08:27 <TD> c++ trivia question (i forgot)
522 2012-01-05 23:08:35 <TD> what happens if you use vector::back() on an empty vector
523 2012-01-05 23:09:44 <BlueMatt> undefined, according to msdn
524 2012-01-05 23:10:10 <JFK911> that's the problem with c++, everything works differently according to who made your tools
525 2012-01-05 23:10:11 <TD> at least msdn is kind enough to say it's undefined
526 2012-01-05 23:10:13 <CIA-100> bitcoin: p2k * r6cba277ecc1f ecoinpool/apps/ecoinpool_mysql_replicator/src/mycouch_replicator.erl: Another Fix http://tinyurl.com/885zebo
527 2012-01-05 23:10:31 <TD> ms always had thorough docs
528 2012-01-05 23:10:42 <BlueMatt> yea, msdn is pretty good
529 2012-01-05 23:10:43 <TD> cplusplus.com doesn't mention this case at all
530 2012-01-05 23:10:44 <JFK911> true
531 2012-01-05 23:10:52 <JFK911> people who bitch about how windows works, didn't read.
532 2012-01-05 23:10:57 <TD> so for https://github.com/gavinandresen/bitcoin-git/blob/pay_to_script_hash/src/script.cpp#L1451
533 2012-01-05 23:11:08 <TD> is it possible to have a scriptSig that pushes nothing onto the stack at all?
534 2012-01-05 23:11:12 <TD> probably that won't get mined, right
535 2012-01-05 23:11:31 <TD> sorry, never mind. i mean will it get excluded by other checks
536 2012-01-05 23:11:48 <TD> empty scripts pass EvalScript, AFAICT.
537 2012-01-05 23:12:17 <luke-jr> TD: are you OK with the direction this is going? the plan with this version is that Bitcoin 2.0 will eliminate scriptPubKey entirely, and replace it with a script hash
538 2012-01-05 23:12:43 <TD> i kind of like bitcoin v1.0
539 2012-01-05 23:13:01 <TD> ok, never mind. i don't think it's possible to reach line 1473 with an empty stackCopy
540 2012-01-05 23:13:12 <Backburn2> all this talk of forking is getting me hot
541 2012-01-05 23:13:17 <luke-jr> 2.0 won't be for years, but if we're heading that direction, I do plan to stop mining non-P2SH scriptPubKeys when clients migrate
542 2012-01-05 23:13:28 <gavinandresen> TD: yup, the OP_HASH will fail in the second EvalScript before then
543 2012-01-05 23:13:38 <TD> yeah
544 2012-01-05 23:14:47 <TD> luke-jr: bitcoin 2.0 is probably going to be like xml vs html, if it ever exists at all
545 2012-01-05 23:14:59 <luke-jr> TD: it's inevitable.
546 2012-01-05 23:15:04 <CIA-100> bitcoin: Luke Dashjr maintree * ra95dc02f6701 gentoo/net-p2p/ (5 files in 2 dirs): Merge branch 'master' into maintree http://tinyurl.com/6whejgg
547 2012-01-05 23:15:18 <TD> yeah, like the year of desktop linux ;)
548 2012-01-05 23:15:27 <TD> old+crappy > new+cool+incompatible
549 2012-01-05 23:15:30 <luke-jr> TD: Bitcoin 1.0 *cannot* scale.
550 2012-01-05 23:16:35 <BlueMatt> s/allowing people to use custom scripts all over the place/allowing TD to get rid of constant addresses, or atleast get the same result/
551 2012-01-05 23:17:16 <TD> i still think it's a minor optimization designed for soon-to-be-outmoded forms of interaction, but ok, whatever. i made my case several times already :)
552 2012-01-05 23:17:46 <TD> anyway, i guess i'd have to think whether all the contracts still work with spender-provides-script
553 2012-01-05 23:17:48 <luke-jr> TD: it certainly isn't the only thing planned for Bitcoin 2.0 :P
554 2012-01-05 23:17:55 <TD> ugh
555 2012-01-05 23:18:15 <BlueMatt> luke-jr: can we focus on bitcoin 1.0 first?
556 2012-01-05 23:18:40 <gavinandresen> Yes, there's still a lot to do before I'd call any bitcoin software "1.0"
557 2012-01-05 23:19:14 <luke-jr> BlueMatt: yes, but before tolerating P2SH, I want to be sure the direction is OK :P
558 2012-01-05 23:19:25 <luke-jr> gavinandresen: not referring to software
559 2012-01-05 23:19:30 <luke-jr> Bitcoin is the protocol :P
560 2012-01-05 23:20:12 <CIA-100> bitcoin: p2k * r034fc8eaf054 ecoinpool/apps/ecoinpool/priv/main_db_workers.json: Workers Filter Function Added http://tinyurl.com/6w9kzy2
561 2012-01-05 23:20:24 <lianj> the protocol also says script length is an var_int & :D
562 2012-01-05 23:21:47 <BlueMatt> can I ask what people prefer for https://github.com/bitcoin/bitcoin/pull/454 -keepnode or replace -addnode ?
563 2012-01-05 23:21:55 <gavinandresen> The block size is the only really hard limit I see to scaling up, and I bet if we thought hard we could come up with backwards-compatible schemes to get around it.
564 2012-01-05 23:22:09 <luke-jr> BlueMatt: I'd do keepnode for purity, but I don't strongly care either way
565 2012-01-05 23:22:52 <luke-jr> jgarzik: you forgot reminding you to merge coinbaser daily.
566 2012-01-05 23:23:00 <jgarzik> ;)
567 2012-01-05 23:23:16 <gavinandresen> Seems to me addnode should be persistent.
568 2012-01-05 23:23:41 <luke-jr> IMO, making it -addnode implies a bugfix, and -keepnode implies a new feature
569 2012-01-05 23:24:06 <makomk> Was -addnode intended for bootstrapping?
570 2012-01-05 23:24:14 <luke-jr> makomk: I think so.
571 2012-01-05 23:24:20 <TD> gavinandresen: in CTransaction::AreInputsStandard, nUnused is indeed now entirely unused ;)
572 2012-01-05 23:24:37 <luke-jr> lol
573 2012-01-05 23:24:44 <TD> gavinandresen: also i think _here_ as opposed to before, it is possible to call back() on an empty stack
574 2012-01-05 23:25:05 <TD> gavinandresen: if somebody gives you a tx with an empty scriptSig that connects to a p2sh txout, after EvalScript is run stack will be empty but you'll try reading the last element
575 2012-01-05 23:25:06 <BlueMatt> makomk: hence why I originally went with keepnode, but I dont feel too strongly either way, so I just figured Id ask and if anyone had a strong preference, Id just go with that one (hopefully to get it merged...)
576 2012-01-05 23:25:36 <gavinandresen> TD:  good catch, I'll add a unit test and fix.
577 2012-01-05 23:25:43 <TD> thanks
578 2012-01-05 23:26:23 <luke-jr> I'd like to rephrase BlueMatt's question:
579 2012-01-05 23:26:37 <luke-jr> Is "keepnode" a bugfix, that I should backport to 0.4 and 0.5 when it gets merged? ;)
580 2012-01-05 23:26:59 <luke-jr> (if so, then it should be called "addnode"; if not, I'd prefer "keepnode")
581 2012-01-05 23:27:28 <gmaxwell> makomk: an addr.txt file is better for bootstrapping.
582 2012-01-05 23:27:37 <BlueMatt> someone has to have a somewhat-strong opinion...
583 2012-01-05 23:27:43 <gmaxwell> The fact that keepnode isn't addnode is pretty counter-intutive.
584 2012-01-05 23:27:58 <gavinandresen> I strongly agree.  There.
585 2012-01-05 23:28:07 <luke-jr> gavinandresen: & with which name? :P
586 2012-01-05 23:28:16 <gavinandresen> (I don't really, but I'll pretend if it helps)
587 2012-01-05 23:28:37 <gmaxwell> Perhaps addnode should be changed to 'bootstrap', and keepnode should become addnode?  (or stay keepnode and addnode should go away)
588 2012-01-05 23:29:16 <gmaxwell> I don't see a real reason to backport this, however.
589 2012-01-05 23:29:18 <luke-jr> somehow it seems to fit in with sipa's role :P
590 2012-01-05 23:29:19 <BlueMatt> if addnode becomes bootstrap, addnode nodes should have their connection dropped after we get connected to the network
591 2012-01-05 23:29:41 <gmaxwell> BlueMatt: they do  eventually. :)
592 2012-01-05 23:30:18 <gavinandresen> I say make addnode do what 99.8% of people will think it does when they see it:  adds a node (persistently) to the connection list.
593 2012-01-05 23:30:21 <BlueMatt> or just get addnode nodes added to pnAddrSeed
594 2012-01-05 23:30:26 <TD> gavinandresen: CTransaction::ClientConnectInputs now also has a dead nUnused
595 2012-01-05 23:30:41 <luke-jr> if we do keepnode, I'm going to change wiki pages s/add/keep/, which is not backward compatible
596 2012-01-05 23:30:43 <luke-jr> :P
597 2012-01-05 23:30:47 <gmaxwell> gavinandresen: yea, thats what keepnode does, and thats what people will think it will do.
598 2012-01-05 23:30:57 <gavinandresen> Is there a reasonbly-easy-to-use tool to find dead code/vars in C++ ?
599 2012-01-05 23:30:58 <luke-jr> let's just go with addnode and move on
600 2012-01-05 23:31:07 <luke-jr> gavinandresen: usually GCC warns I think O.o
601 2012-01-05 23:31:13 <luke-jr> it *can* at least
602 2012-01-05 23:31:23 <gmaxwell> we may overload some nodes that are being frequently addnoded right now though. ::shrugs:: but okay.
603 2012-01-05 23:31:23 <TD> gavinandresen: gcc can do it
604 2012-01-05 23:31:33 <TD> gavinandresen: -Wwarn-unused or something like that
605 2012-01-05 23:31:36 <luke-jr> gmaxwell: hmm, that might be a good point
606 2012-01-05 23:31:52 <TD> i suspect there is a lot of dead code in bitcoin
607 2012-01-05 23:32:01 <BlueMatt> TD: dont think that works if its ie in a header
608 2012-01-05 23:32:04 <BlueMatt> extern variable
609 2012-01-05 23:32:04 <luke-jr> there's an API in there that nobody uses yet
610 2012-01-05 23:32:12 <TD> right
611 2012-01-05 23:32:18 <luke-jr> BlueMatt: extern stuff is by definition always used ;)
612 2012-01-05 23:32:32 <luke-jr> by the magic "unknown external plugin"
613 2012-01-05 23:33:44 <TD> gavinandresen: what are the slashes in /P2SH/ for ?
614 2012-01-05 23:34:23 <gavinandresen> To make the string a little longer-- I was worried just P2SH was short enough somebody would worry about false positives
615 2012-01-05 23:34:38 <gavinandresen> (PayToScriptHash seemed too long....)
616 2012-01-05 23:35:02 <gmaxwell> TD: lcov is your friend.
617 2012-01-05 23:36:02 <gmaxwell> It's great when you get jenkins to run your lcov tests, e.g. https://mf4.xiph.org/jenkins/job/opus-coverage/ws/coverage/index.html  then you can see when someone just added a buch of code that isn't getting run by your basic built time sanity tests.
618 2012-01-05 23:36:42 <gmaxwell> Also, I've found a fair number of bugs from going "wtf isn't this line getting executed="
619 2012-01-05 23:37:01 <gmaxwell> "Oh, because there is a bug ten lines up"
620 2012-01-05 23:40:00 <gavinandresen> mmmm....   there's no way to put a #pragma inside a #define, is there?   IMPLEMENT_SERIALIZE defines some variables that are usually unused....
621 2012-01-05 23:40:51 <gavinandresen> ... then again, there are only 23 IMPLEMENT_SERIALIZEs....
622 2012-01-05 23:41:30 <k9quaint> ok, what ever mouse button you guys have been clicking, keep it up
623 2012-01-05 23:41:37 <k9quaint> BTC price has recovered :)
624 2012-01-05 23:41:54 <BlueMatt> gavinandresen: done, -addnode removed
625 2012-01-05 23:44:02 <TuxBlackEdo> since #bitcoin is being flooded by newbies i had to come here
626 2012-01-05 23:44:10 <TuxBlackEdo> and i wanted to ask gavinandresen a question
627 2012-01-05 23:44:15 <BlueMatt> k9quaint: what, you dont have access to the  btcPrice++; button?
628 2012-01-05 23:44:33 <makomk> gavinandresen: yeah, -Wall is a bit spammy as a result...
629 2012-01-05 23:45:00 <gmaxwell> TuxBlackEdo: it always happenes when the price changes. :(
630 2012-01-05 23:45:01 <edcba> ;;bc,mtgox
631 2012-01-05 23:45:02 <gribble> {"ticker":{"high":7.22,"low":5.65291,"avg":6.24092784,"vwap":6.287750074,"vol":191829,"last_all":7,"last_local":7,"last":7,"buy":7,"sell":7.03488}}
632 2012-01-05 23:45:03 <BlueMatt> TuxBlackEdo: thats what happens when the price goes up...
633 2012-01-05 23:45:48 <gavinandresen> TuxBlackEdo: the answer is eleven.
634 2012-01-05 23:45:50 <TuxBlackEdo> gavinandresen, what would you do if someone big like valve would update steam with a mining screen saver in an attempt to take over the bitcoin network
635 2012-01-05 23:46:00 <TuxBlackEdo> i always wondered that
636 2012-01-05 23:46:05 <Raccoon> remind me, how many unique bitcoin addresses are algorithmically possible?  mind, i'm not talking about the bit-length of an address or key, but how many unique addresses the algorithm is capable of producing.
637 2012-01-05 23:46:42 <TuxBlackEdo> like in an attempt to launch multiple double spending attacks against all existing bitcoin businesses like mtgox/tradehill and others
638 2012-01-05 23:46:50 <gavinandresen> TuxBlackEdo: that seems unlikely, but I'd certainly make some noise about how they were stealing from their users.
639 2012-01-05 23:47:11 <k9quaint> bluematt: its a private variable :(
640 2012-01-05 23:47:15 <TuxBlackEdo> hm ok, that's all i wanted to ask /me disappears into lurker mode again
641 2012-01-05 23:49:29 <k9quaint> if valve did that, I would hack into steam and sub my workerID in for theirs :)
642 2012-01-05 23:50:17 <k9quaint> then I would split the BTC with gavin so he wouldn't complain about the theft :P
643 2012-01-05 23:50:46 <BlueMatt> have fun pulling that one off...
644 2012-01-05 23:51:03 <Raccoon> ...
645 2012-01-05 23:51:17 <k9quaint> if Sony did it, it would be a cinch :P
646 2012-01-05 23:51:33 <BlueMatt> ouch...
647 2012-01-05 23:58:32 <enquirer> and it's no brainer to hack sony
648 2012-01-05 23:59:13 <k9quaint> I don't think one can even call it hacking anymore