1 2011-08-31 00:00:14 <CIA-101> bitcoin: aldiyen * rbe719b81ac74 Phoenix-Miner-personal/WorkQueue.py: WorkQueue: use RPCClient.maxErrCount, not RPCClient.getMaxErrCount()
  2 2011-08-31 00:00:15 <CIA-101> bitcoin: aldiyen * rf90b89e94d03 Phoenix-Miner-personal/ (3 files in 2 dirs): Fixed tracking of work acceptance when retrying after failed communication
  3 2011-08-31 00:36:47 <MobiusL> ;;bc,stats
  4 2011-08-31 00:36:49 <gribble> Current Blocks: 143272 | Current Difficulty: 1777774.4820015 | Next Difficulty At Block: 145151 | Next Difficulty In: 1879 blocks | Next Difficulty In About: 1 week, 5 days, 6 hours, 53 minutes, and 55 seconds | Next Difficulty Estimate: 1902503.54817275
  5 2011-08-31 00:50:32 <vsrinivas> eep, that's a solid difficulty increase.
  6 2011-08-31 00:52:34 <luke-jr> except that it isn't solid :p
  7 2011-08-31 01:05:14 <CIA-101> bitcoin: aldiyen * r32dfb6c8920d Phoenix-Miner-personal/ (Miner.py WorkQueue.py minerutil/RPCProtocol.py phoenix.py):
  8 2011-08-31 01:05:15 <CIA-101> bitcoin: Miner to log, rather than referencing the logger directly
  9 2011-08-31 01:25:13 <CIA-101> bitcoin: Con Kolivas * ra593afdbeba1 cgminer/ (main.c miner.h util.c): Use the curses_lock to protect the curses_active variable and test it under lock.
 10 2011-08-31 02:48:09 <copumpkin> anyone have unscrubbed logs (with IPs still in them) of this channel?
 11 2011-08-31 02:48:37 <copumpkin> the logs in the topic scrub all the identifying information from hosts
 12 2011-08-31 02:53:35 <luke-jr> copumpkin: I do.
 13 2011-08-31 02:54:00 <copumpkin> nanotube is already getting me the info I needed, thanks :)
 14 2011-08-31 02:54:16 <luke-jr> k
 15 2011-08-31 04:05:11 <CIA-101> bitcoin: Con Kolivas * r5aca762a97bd cgminer/main.c: Go back to requesting work 2/3 of the way through the current scantime with CPU mining as reports of mining threads running out of work have occurred with only 5 seconds to retrieve work.
 16 2011-08-31 04:21:25 <luke-jr> &
 17 2011-08-31 04:21:59 <TuxBlackEdo> sup luke-jr
 18 2011-08-31 04:22:08 <luke-jr> [02:05:12] <CIA-101> bitcoin: Con Kolivas * r5aca762a97bd cgminer/main.c: Go back to requesting work 2/3 of the way through the current scantime with CPU mining as reports of mining threads running out of work have occurred with only 5 seconds to retrieve work.
 19 2011-08-31 04:22:11 <luke-jr> ^ this comment sounds stupid
 20 2011-08-31 05:46:59 <TuxBlackEdo> * TuxBlackEdo has kicked Diablo-D3 from #eligius (Diablo-D3)
 21 2011-08-31 06:32:54 <d33tah> anyone here familiar with vanitygen?
 22 2011-08-31 06:56:53 <phantomcircuit> d33tah, kind of
 23 2011-08-31 09:41:45 <piuk> When a new block is found how is it advertised to the network?
 24 2011-08-31 09:42:12 <piuk> e.g. which message from https://en.bitcoin.it/wiki/Protocol_specification?
 25 2011-08-31 09:44:26 <SomeoneWeird> https://en.bitcoin.it/wiki/Protocol_specification?#Block_Headers
 26 2011-08-31 09:47:43 <piuk> so the client has to poll get headers first?
 27 2011-08-31 09:49:20 <SomeoneWeird> no idea
 28 2011-08-31 09:49:22 <SomeoneWeird> :)
 29 2011-08-31 09:49:43 <piuk> ok ty :)
 30 2011-08-31 12:50:46 <jgarzik> luke-jr: got the patches going locally.  just need to push them out.
 31 2011-08-31 13:46:23 <gavinandresen> Anybody up for helping me figure out if I'll massively break things if I change CWalletTx::GetTxTime()  ?
 32 2011-08-31 13:47:12 <imsaguy2> sure, watch do you need?
 33 2011-08-31 13:47:15 <imsaguy2> what*
 34 2011-08-31 13:47:58 <gavinandresen> I really want to remove the CRITICAL_BLOCK(cs_main) from that method... so I'm thinking of making it always just return nTimeReceived
 35 2011-08-31 13:52:01 <imsaguy2> sorry gavinandresen, I don't know it that well.
 36 2011-08-31 13:52:16 <gavinandresen> ok
 37 2011-08-31 13:56:23 <ThomasV> when I run bitcoind it often becomes unresponsive ; deadlocks ?
 38 2011-08-31 13:56:33 <gavinandresen> yes
 39 2011-08-31 13:56:43 <ThomasV> is that a known issue ?
 40 2011-08-31 13:56:56 <ThomasV> is there a fix somewhere ?
 41 2011-08-31 13:56:58 <gavinandresen> yes, I've been working on them for a while....
 42 2011-08-31 13:57:01 <BlueMatt> yes; no
 43 2011-08-31 13:57:48 <ThomasV> gavinandresen: the gui does not seem to have that problem, but I rarely leave it running for hours
 44 2011-08-31 13:59:52 <ThomasV> btw, I see that there is a python bitcoin library ; has somebody bothered to write a gui using it ?
 45 2011-08-31 14:00:27 <BlueMatt> iirc that lib isnt too complete...
 46 2011-08-31 14:00:37 <luke-jr> ThomasV: you mean wallet?
 47 2011-08-31 14:03:11 <ThomasV> luke-jr: no, I mean gui
 48 2011-08-31 14:03:23 <luke-jr> ThomasV: Spesmilo isn't very new
 49 2011-08-31 14:04:12 <ThomasV> oh that is what I was looking for
 50 2011-08-31 14:05:21 <luke-jr> ThomasV: it has some blocking issues when the bitcoind is remote though, I think
 51 2011-08-31 14:05:27 <luke-jr> need to look into that someday&
 52 2011-08-31 14:06:35 <ThomasV> hmm, how about if I use connect_to_remote on a local machine ? I need to call remote because my atadir is not in my homedir
 53 2011-08-31 14:08:12 <BlueMatt> gavinandresen: what is the status of the test harness? afaict its in the shell of what it will be mode?
 54 2011-08-31 14:09:09 <gavinandresen> I'll send you an email I've sent a couple of other people who said they were interested in working on it.
 55 2011-08-31 14:09:57 <BlueMatt> well Im interested in using it, but that works...
 56 2011-08-31 14:10:39 <BlueMatt> thanks
 57 2011-08-31 14:11:18 <luke-jr> ThomasV: no idea what "connect_to_remote" is
 58 2011-08-31 14:11:58 <luke-jr> ThomasV: the problem is network latency blocking GUI
 59 2011-08-31 14:12:17 <ThomasV> luke-jr: yes, I just realize that your gui is not using that bitcoinrpc library
 60 2011-08-31 14:12:29 <luke-jr> ThomasV: it is, if that's what you have installed
 61 2011-08-31 14:13:17 <ThomasV> no...
 62 2011-08-31 14:13:47 <luke-jr> bitcoinrpc is API-compatible with python-jsonrpc
 63 2011-08-31 14:13:58 <jgarzik> gavinandresen: let's go ahead and release 0.4-rc1
 64 2011-08-31 14:14:10 <BlueMatt> jgarzik: oh fix the deadlocks first
 65 2011-08-31 14:14:10 <ThomasV> luke-jr: https://en.bitcoin.it/wiki/Bitcoin-python
 66 2011-08-31 14:14:26 <jgarzik> BlueMatt: that's what the bug fix-only cycle is for
 67 2011-08-31 14:14:43 <jgarzik> BlueMatt: -rc release means "we've stopped taking new features; only take bug fixes"
 68 2011-08-31 14:14:54 <jgarzik> BlueMatt: _that_ has been true de facto for a while
 69 2011-08-31 14:14:55 <BlueMatt> jgarzik: mmm....well...
 70 2011-08-31 14:14:59 <gavinandresen> jgarzik: I'm debugging GUI deadlocks now, hope to commit fixes before end of the day
 71 2011-08-31 14:15:07 <luke-jr> ThomasV: oh, nope, never heard of that one
 72 2011-08-31 14:15:11 <luke-jr> ThomasV: it looks broken by design
 73 2011-08-31 14:15:18 <BlueMatt> jgarzik: can I get 464 pulled first?
 74 2011-08-31 14:15:33 <jgarzik> IMNSHO we can all agree that we've stopped taking new features, and are focusing on bug fixes, right?
 75 2011-08-31 14:15:33 <luke-jr> jgarzik: -beta means that
 76 2011-08-31 14:15:43 <luke-jr> jgarzik: -rc means "all obvious bugs fixed, we think it's good to go"
 77 2011-08-31 14:15:44 <BlueMatt> jgarzik: pretty much
 78 2011-08-31 14:15:50 <BlueMatt> luke-jr: not in kernel-land
 79 2011-08-31 14:15:54 <jgarzik> luke-jr: not in kernel land
 80 2011-08-31 14:15:56 <jgarzik> heh
 81 2011-08-31 14:16:28 <gavinandresen> I'd rather not tell people to start testing before the deadlock changes are in.
 82 2011-08-31 14:16:32 <luke-jr> well, Linus just recently fixed the other non-standard versioning thing; maybe using betas is next
 83 2011-08-31 14:16:39 <gavinandresen> (I've changed a fair bit of code)
 84 2011-08-31 14:16:59 <CIA-101> bitcoin: Jeff Garzik master * r8af99d1 / (14 files in 4 dirs):
 85 2011-08-31 14:16:59 <luke-jr> in any case, I don't see why Linux versioning has any influence on Bitcoin clients
 86 2011-08-31 14:17:16 <jgarzik> gavinandresen: ok
 87 2011-08-31 14:17:25 <BlueMattBot> jgarzik did you mean me? Unknown command 'sure...'
 88 2011-08-31 14:17:25 <jgarzik> BlueMattBot: sure...  contrib/* changes are easy
 89 2011-08-31 14:17:34 <BlueMattBot> jgarzik did you mean me? Unknown command 'no,'
 90 2011-08-31 14:17:34 <jgarzik> BlueMattBot: no, I did not mean you
 91 2011-08-31 14:17:35 <BlueMattBot> Use 'BlueMattBot: DONT SPAM the Chan help' to get help!
 92 2011-08-31 14:18:17 <luke-jr> not that I mind solo miners duplicating work&
 93 2011-08-31 14:18:57 <luke-jr> jgarzik: btw, you realize the 2 patches I meant yesterday are regarding pushpool, right?
 94 2011-08-31 14:20:16 <CIA-101> bitcoin: aldiyen * rf16d460d3e16 Phoenix-Miner-personal/ (Miner.py WorkQueue.py minerutil/RPCProtocol.py): Changed submit error handling to make the software finish the current work's noncerange, but still discard the queue (some possibly invalid work during a network outage is better than no work)
 95 2011-08-31 14:20:33 <jgarzik> luke-jr: yes
 96 2011-08-31 14:20:50 <jgarzik> luke-jr: I have some local tests I run on pushpool, before pushing out merged changes
 97 2011-08-31 14:21:06 <luke-jr> i c
 98 2011-08-31 14:21:41 <ThomasV> luke-jr: why ?
 99 2011-08-31 14:21:58 <luke-jr> ThomasV: ?
100 2011-08-31 14:22:24 <ThomasV> <luke-jr> ThomasV: it looks broken by design
101 2011-08-31 14:22:43 <luke-jr> ThomasV: because it uses float BTC instead of integer Satoshi like any sane API should
102 2011-08-31 14:23:37 <CIA-101> bitcoin: Jeff Garzik master * rce39dfe / locale/pl/LC_MESSAGES/bitcoin.po :
103 2011-08-31 14:29:04 <CIA-101> bitcoin: Jeff Garzik master * r927aaa0 / (3 files): Add reference python miner, in contrib/pyminer/ - http://git.io/P4XMQw
104 2011-08-31 14:32:16 <phantomcircuit> jgarzik, whens the last time you took a look at python-jsonrpc
105 2011-08-31 14:32:23 <phantomcircuit> python-bitcoinrpc
106 2011-08-31 14:32:44 <CIA-101> bitcoin: Jeff Garzik master * r21189a4 / (doc/README src/rpc.cpp):
107 2011-08-31 14:32:45 <CIA-101> bitcoin: Merge pull request #463 from TheBlueMatt/encreadme
108 2011-08-31 14:33:07 <jgarzik> phantomcircuit: a while ago
109 2011-08-31 14:33:17 <jgarzik> phantomcircuit: any bugaboos lurking?
110 2011-08-31 14:33:34 <phantomcircuit> not that i've seen
111 2011-08-31 14:33:36 <phantomcircuit> just wondering
112 2011-08-31 14:34:03 <luke-jr> jgarzik: sometimes it seems HTTPS doesn't work right, but I couldn't see a problem with the latest code, so I haven't reported it yet
113 2011-08-31 14:42:26 <luke-jr> git://gitorious.org/~Luke-Jr/bitcoin/luke-jr-bitcoin.git base58_liberal_parsing bugfix_getwork_newblk_race rpc_keepalive gitignore getwork_dedupe minfee_modes
114 2011-08-31 14:42:34 <BlueMatt> thanks jgarzik
115 2011-08-31 14:42:35 <luke-jr> various branches to probably merge
116 2011-08-31 14:43:45 <Mad7Scientist> WIll debug.log ever stop growing?
117 2011-08-31 14:43:48 <Mad7Scientist> It is over 200MB now
118 2011-08-31 14:44:05 <mtrlt> it won't.
119 2011-08-31 14:44:08 <Mad7Scientist> If I leave bitcoin running for too long it will be bigger than the bitcoin database itself
120 2011-08-31 14:44:18 <BlueMatt> who is Alex Waters
121 2011-08-31 14:44:41 <Mad7Scientist> I guess I have to shut down bitcoin to delete the file
122 2011-08-31 14:44:41 <phantomcircuit> Mad7Scientist, there's a bug that it'll only grow to 2 GB
123 2011-08-31 14:44:48 <Mad7Scientist> lol
124 2011-08-31 14:44:49 <phantomcircuit> on 32 bit systems only though
125 2011-08-31 14:44:52 <mtrlt> lol
126 2011-08-31 14:45:10 <Mad7Scientist> when I restart bitcoin does it start over?
127 2011-08-31 14:45:27 <luke-jr> Mad7Scientist: if it's sufficiently large, it shoudl
128 2011-08-31 14:45:28 <Mad7Scientist> I would have thought the bug would be 4GB
129 2011-08-31 14:46:14 <luke-jr> Mad7Scientist: signed.
130 2011-08-31 14:46:26 <Mad7Scientist> bitcoin   11835  rep    6w      REG        8,7 276623109   1324676 /home/rep/.bitcoin/debug.log
131 2011-08-31 14:47:00 <Mad7Scientist> COMMAND     PID USER   FD      TYPE     DEVICE      SIZE      NODE NAME
132 2011-08-31 14:47:26 <Mad7Scientist> If I delete the file it will go to /lost+found won't it
133 2011-08-31 14:47:33 <gavinandresen> deadlock fixes are done for now:  https://github.com/bitcoin/bitcoin/pull/480
134 2011-08-31 14:47:38 <phantomcircuit> personally id just create a named pipe
135 2011-08-31 14:47:43 <gavinandresen> (I'll rebase when jgarzik is done pulling stuff)
136 2011-08-31 14:47:48 <Mad7Scientist> or disappear until bitcoin closes the file
137 2011-08-31 14:48:05 <Mad7Scientist> How do I create a named pipe?
138 2011-08-31 14:48:08 <BlueMatt> gavinandresen: wow that is alot of change
139 2011-08-31 14:48:23 <phantomcircuit> stuff doesn't go to /lost+found unless you kill the system
140 2011-08-31 14:48:49 <luke-jr> BlueMatt: looks like a lot is renaming variables
141 2011-08-31 14:48:50 <gavinandresen> BlueMatt: yup.  Doesn't look as bad if you git diff --ignore-whitespace-change
142 2011-08-31 14:49:00 <Mad7Scientist> what are the  4 kinds of files again? normal, device, fifo and socket right?
143 2011-08-31 14:49:34 <BlueMatt> luke-jr: true...
144 2011-08-31 14:51:00 <phantomcircuit> downloading the block chain on this rackspace server is damn fast
145 2011-08-31 14:51:05 <phantomcircuit> <3 SAN
146 2011-08-31 14:51:18 <Mad7Scientist> the whole thing probably fits in the cache
147 2011-08-31 14:51:19 <phantomcircuit> lol
148 2011-08-31 14:51:47 <phantomcircuit> this is basically waht you get when you completely ignore fsync
149 2011-08-31 14:52:42 <phantomcircuit> i removed the foreign key constraints from a postgresql schema for intersango-v2 and saw a 20x speed up
150 2011-08-31 14:52:43 <jgarzik> gavinandresen: done for the day
151 2011-08-31 14:52:45 <phantomcircuit> i was like
152 2011-08-31 14:52:46 <phantomcircuit> :(
153 2011-08-31 14:52:50 <luke-jr> technically, Linux goes overboard for fsync
154 2011-08-31 14:52:57 <luke-jr> it's NOT supposed to wait for it to be on the disk
155 2011-08-31 14:53:06 <jgarzik> gavinandresen: and I tried to avoid touching any code for the most part, since it sounded like you were touching a lot of areas of code
156 2011-08-31 14:53:18 <jgarzik> mostly maintenance pulls
157 2011-08-31 14:53:21 <luke-jr> jgarzik: you forgot to merge my still
158 2011-08-31 14:53:23 <luke-jr> stuff*
159 2011-08-31 14:53:28 <phantomcircuit> luke-jr, what's it supposed to do?
160 2011-08-31 14:53:56 <jgarzik> luke-jr: no, Mr. Impatience, I already told you that I run local pushpool tests.  They take many hours to complete, because I'm CPU mining...
161 2011-08-31 14:54:16 <jgarzik> luke-jr: I don't push until those tests complete
162 2011-08-31 14:54:28 <luke-jr> jgarzik: I meant the bitcoind fixes that time
163 2011-08-31 14:54:31 <BlueMatt> wow you actually test code, what is this foreign idea?
164 2011-08-31 14:54:46 <Mad7Scientist> Are there any test networks or alternate networks where you can CPU mine and expect to win frequently even if the coins aren't worth anythinG?
165 2011-08-31 14:54:58 <BlueMatt> testnet in a box
166 2011-08-31 14:55:26 <BlueMatt> now to go (finally) fix the blockchain and nightly build uploaders...
167 2011-08-31 14:55:30 <phantomcircuit> gavinandresen, didn't notice your comment, adding that is trivial https://github.com/bitcoin/bitcoin/pull/421
168 2011-08-31 14:56:31 <luke-jr> jgarzik: but on pushpool-- could you push the new tree to another branch-name or something perhaps, so I can begin working on my new code on top of it? :P
169 2011-08-31 15:02:26 <Mad7Scientist> Is it easy to set up a simple mining pool yourself?
170 2011-08-31 15:03:17 <luke-jr> Mad7Scientist: no need to setup a pool
171 2011-08-31 15:03:29 <luke-jr> Mad7Scientist: just run a new testnet client pair without any outside connecitivity
172 2011-08-31 15:23:23 <pakis> acc
173 2011-08-31 15:24:48 <CIA-101> bitcoin: Gavin Andresen master * rb0243da / src/util.cpp : Highlight mis-matching locks - http://git.io/KaMKuQ
174 2011-08-31 15:24:49 <CIA-101> bitcoin: Also change master RPC call handler to acquire cs_main and cs_wallet locks
175 2011-08-31 15:24:50 <CIA-101> bitcoin: Gavin Andresen master * r471426f / (doc/coding.txt src/ui.cpp src/wallet.cpp):
176 2011-08-31 15:24:51 <CIA-101> bitcoin: Fixed potential deadlocks in GUI code.
177 2011-08-31 15:24:52 <CIA-101> bitcoin: Gavin Andresen master * rcb6c4b8 / (11 files in 2 dirs):
178 2011-08-31 15:24:53 <CIA-101> bitcoin: Simplify mutex locking, fix deadlocks. Fixes issue #453 - http://git.io/Ep5DSg
179 2011-08-31 15:25:48 <upb> wow
180 2011-08-31 15:27:52 <ThomasV> pulling
181 2011-08-31 15:28:19 <CIA-101> bitcoin: Gavin Andresen master * r541e585 / doc/README : Fix typo ("you own security") - http://git.io/toE6mQ
182 2011-08-31 15:28:27 <BlueMatt> sorry
183 2011-08-31 15:34:23 <phantomcircuit> gavinandresen, nice fox
184 2011-08-31 15:34:42 <gavinandresen> fox?
185 2011-08-31 15:34:46 <phantomcircuit> was a joke
186 2011-08-31 15:34:47 <phantomcircuit> lulz
187 2011-08-31 15:35:14 <gavinandresen> you kids these days....
188 2011-08-31 15:35:27 <phantomcircuit> kids im practically a teenager!
189 2011-08-31 15:35:28 <gavinandresen> kidz theze daze?
190 2011-08-31 15:35:31 <phantomcircuit> wait
191 2011-08-31 15:36:03 <BlueMatt> gavinandresen: are you sure its safe to not have an account lock (or a lock which is used when account stuff is being done)
192 2011-08-31 15:36:41 <ThomasV> no deadlock so far (compiled and been running for 5 minutes :-)
193 2011-08-31 15:36:49 <gavinandresen> everything that mucks with accounts should grab the wallet mutex.  Did I screw that up somewhere?
194 2011-08-31 15:37:14 <BlueMatt> gavinandresen: oh, but it grabs the lock outside of wallet...sorry
195 2011-08-31 15:37:28 <gavinandresen> yup.  Making wallet self-contained is a job for another day
196 2011-08-31 15:37:29 <BlueMatt> would it not be safer to also grab the lock in the actual wallet function
197 2011-08-31 15:37:37 <BlueMatt> ah ok
198 2011-08-31 15:38:12 <BlueMatt> ok, nightly builds and jenkins running right again, now I just have to do nightly blockchains
199 2011-08-31 15:38:14 <gavinandresen> (ideally, it would be like keystore, and have a nice clean API so you could use it without worrying about locking)
200 2011-08-31 15:38:54 <pakis> hello
201 2011-08-31 15:39:15 <tcatm> gavinandresen: did you get my email about bitcoin.org?
202 2011-08-31 15:39:35 <gavinandresen> tcatm: yep, sorry I didn't respond, it's just not high on my priority list
203 2011-08-31 15:40:47 <pakis> i still dnt have power in my house
204 2011-08-31 15:41:07 <tcatm> gavinandresen: who should decide whether to make such changes?
205 2011-08-31 15:41:29 <gavinandresen> Not it!
206 2011-08-31 15:42:09 <tcatm> BlueMatt: I'm having a little trouble with memory consumption on my server. what software do you use to run the dnsseed?
207 2011-08-31 15:42:29 <BlueMatt> tcatm: its php, I can probably drop mem usage a bit...one sec
208 2011-08-31 15:42:55 <gavinandresen> tcatm has a nice new design for bitcoin.org.  At least, I think it is nice, but I've got no design sense.
209 2011-08-31 15:43:20 <ThomasV> can we get a glimpe ?
210 2011-08-31 15:43:26 <ThomasV> *glimpse
211 2011-08-31 15:44:56 <tcatm> gavinandresen: if no one volunteers I'd offer to manage the website, add content and translations
212 2011-08-31 15:45:37 <gavinandresen> I think jgarzik touched the website last.  It'd be nice if it was put under source control....
213 2011-08-31 15:45:47 <BlueMatt> you can do it on github...
214 2011-08-31 15:46:15 <tcatm> gavinandresen: yes, that's what I have in mind. have a git repo and a makefile that takes a version number (like 0.3.24) and maybe other variables and produces a static website
215 2011-08-31 15:46:28 <BlueMatt> you could just host it on github?
216 2011-08-31 15:46:40 <gavinandresen> Great!  what do you need to make that happen?
217 2011-08-31 15:46:40 <tcatm> yep we could use github pages
218 2011-08-31 15:47:11 <tcatm> gavinandresen: ftp/sftp/scp/whatever access to bitcoin.org so I can upload files
219 2011-08-31 15:47:25 <gavinandresen> bitcoin.org is a sourceforge static page
220 2011-08-31 15:47:31 <BlueMatt> http://tcatm.github.com/bitcoin-js-remote/
221 2011-08-31 15:48:12 <gavinandresen> ... talk with jgarzik about how it works exactly (I can't recall)
222 2011-08-31 15:48:48 <Incitatus> hey magicaltux, we got conflicting information about your merchant API service. We got an email on the 22nd saying it would be up in a week and you told me it wouldn't be up but that we can use account to account transfers
223 2011-08-31 15:48:54 <BlueMatt> you can forward the dns to github pretty simply afaict
224 2011-08-31 15:49:17 <Incitatus> magicaltux, if it's going to be up soon we will wait and not spend time on doing code for a reference system
225 2011-08-31 15:49:43 <ThomasV> thairentboi: does your nick have a meaning ?
226 2011-08-31 15:51:08 <BlueMatt> (iirc bitcoin.org dns forwards to sourceforge, assuming its done the same way as other projects)
227 2011-08-31 15:51:19 <BlueMatt> so you could just move to github...
228 2011-08-31 15:52:30 <thairentboi> ha ... only LOVE
229 2011-08-31 15:53:02 <jarpiain> Incitatus: might have better luck asking at #mtgox
230 2011-08-31 15:53:14 <jarpiain> Incitatus: but it's 3 AM in tokyo now
231 2011-08-31 16:00:50 <CIA-101> bitcoin: Gavin Andresen master * re3c26b2 / (src/keystore.cpp src/keystore.h): SetCrypted() obtains keystore lock, to be safe. - http://git.io/WtwRCQ
232 2011-08-31 16:08:05 <pakis> its 2pm in ny
233 2011-08-31 17:08:46 <kinlo> does the bitcoinclient store *all* blocks, including orphans?
234 2011-08-31 17:08:58 <kinlo> are orphans distributed trough the network?
235 2011-08-31 17:11:13 <gmaxwell> kinlo: they aren't distributed once they're orphaned.
236 2011-08-31 17:11:20 <gmaxwell> But if you've heard of one you'll remember it
237 2011-08-31 17:11:24 <BlueMatt> but they are still stored (once youve gotten them)
238 2011-08-31 17:11:33 <gmaxwell> (because the software doesn't delete from the mmaped file)
239 2011-08-31 17:11:37 <kinlo> well
240 2011-08-31 17:11:46 <kinlo> I've started a new client with empty blockchain
241 2011-08-31 17:11:55 <kinlo> and the debug.log is full with blocks being orphanned
242 2011-08-31 17:12:15 <gmaxwell> that sounds ... odd.
243 2011-08-31 17:12:17 <kinlo> so that looks to me like he downloaded blocks and marked them orphan due to the fact that there is a longer chain
244 2011-08-31 17:12:37 <gmaxwell> Someone attacking (stupidly) perhaps?
245 2011-08-31 17:12:38 <kinlo> it does sound odd
246 2011-08-31 17:12:47 <kinlo> so that's why I mention it
247 2011-08-31 17:13:20 <BlueMatt> the client often downloads orphan blocks != blocks in an orphan chain
248 2011-08-31 17:13:40 <BlueMatt> during download, the client ends up getting orphaned blocks all over the place, or so Ive seen
249 2011-08-31 17:13:52 <kinlo> grep -c "ProcessBlock: ORPHAN BLOCK" .bitcoin/debug.log
250 2011-08-31 17:13:53 <kinlo> 7238
251 2011-08-31 17:14:10 <kinlo> new client, it's still downloading initial set of blocks
252 2011-08-31 17:14:36 <gmaxwell> BlueMatt: I don't know how it could, when it does the request the neighbor shouldn't push any blocks that aren't on the best path.
253 2011-08-31 17:14:38 <kinlo> or am I misinterpreting debug log?
254 2011-08-31 17:15:12 <gmaxwell> duhhh
255 2011-08-31 17:15:22 <gmaxwell> yea it's new blocks which aren't yet connected to the chain perhaps?
256 2011-08-31 17:15:27 <BlueMatt> gmaxwell: I never bothered to look, I just know when I was doing some hacking it gets blocks that are on the best path, but missing a block or two
257 2011-08-31 17:15:52 <gmaxwell> I'm not sure how that happens, but it's not unreasonable.
258 2011-08-31 17:16:09 <BlueMatt> it happens quite a bit...but as I said I never bothered to look
259 2011-08-31 17:16:41 <gmaxwell> kinlo: e.g. it (somehow) gets 1,2,3,4,5   8,9,10 ... all are on the best path but 8+ are orphan until it has the missing blocks to connect them to the best contigious path.
260 2011-08-31 17:17:03 <kinlo> oh, so out of order blocks
261 2011-08-31 17:17:14 <kinlo> that could make sense
262 2011-08-31 17:18:33 <kinlo> too bad, it would be usefull to investigate orphan blocks for research reasons
263 2011-08-31 17:19:24 <gmaxwell> kinlo: collect them from longstanding nodes.
264 2011-08-31 17:19:38 <gmaxwell> There isn't anything private in the block data... people should be willing to share it.
265 2011-08-31 17:19:41 <kinlo> ah, they are still available in the network? :)
266 2011-08-31 17:19:46 <kinlo> cool
267 2011-08-31 17:20:00 <kinlo> well, don't have any research project going on for now, but it's cool to know I can get them
268 2011-08-31 17:20:04 <gmaxwell> well, you can fetch them IIRC, but you'd need to know about them.
269 2011-08-31 17:20:19 <gmaxwell> But you can just ask people for their whole block data file.
270 2011-08-31 17:20:28 <kinlo> rather large file
271 2011-08-31 17:20:56 <kinlo> but doesn't matter, should be usefull for someone with too much time willing to write some kind of "orphan" explorer
272 2011-08-31 17:21:05 <gmaxwell> xz will get you a good 30% compression iirc. :)
273 2011-08-31 17:21:28 <kinlo> but I do think it would be cool to have a look at those transactions that are doublespend in the orphan blocks vs the real chain
274 2011-08-31 17:21:41 <gmaxwell> If there are any.
275 2011-08-31 17:21:54 <kinlo> well, the mybitcoin case should be visible that way
276 2011-08-31 17:22:08 <gmaxwell> People have looked.
277 2011-08-31 17:22:26 <kinlo> nothing found?
278 2011-08-31 17:22:27 <gmaxwell> They were vague enough that you can't say they lied if no such block ever existed.
279 2011-08-31 17:22:31 <gmaxwell> Right.
280 2011-08-31 17:23:02 <kinlo> would help if the mybitcoin people showed their blk001.dat then
281 2011-08-31 17:23:07 <gmaxwell> Yup.
282 2011-08-31 17:23:33 <gmaxwell> and it would be moderately computationally hard for them to forge that orphan.
283 2011-08-31 17:24:12 <kinlo> well, they should have done it immediatly, it's relatively easy to create such a block if you have weeks time
284 2011-08-31 17:25:09 <gmaxwell> it's not just the block, you need the candidate txn which you control the keys for
285 2011-08-31 17:25:25 <gmaxwell> and if you use one of your own txn you take the risk that people sluth it out and track it back to you.
286 2011-08-31 17:26:02 <kinlo> true
287 2011-08-31 17:26:25 <gmaxwell> e.g. "this 1000btc" was the ripoff. and them mtgox says "oh really? _that_ was a transfer out to your own account at mtgox mr mybitcoin."
288 2011-08-31 17:26:47 <kinlo> who's behind mybitcoin?
289 2011-08-31 17:26:53 <BlueMatt> no one knows...
290 2011-08-31 17:27:55 <laetus> bruce wagner just joined #bitcoin O.o......
291 2011-08-31 17:28:15 <asoltys> hi, just tried bitcoin yesterday. the initial block chain download was a real put-off.  i read that a "bitcoin lite" client is being worked on that won't require you to download the entire chain.   can someone point me to more info on that?
292 2011-08-31 17:28:42 <laetus> 21:27:07 -!- bruce_wagner [~brucewagn@p3208-ipngn2301marunouchi.tokyo.ocn.ne.jp] has joined #bitcoin
293 2011-08-31 17:28:46 <BlueMatt> its not really in a remotely usable state right now...
294 2011-08-31 17:28:59 <laetus> anyone know if this is a VPN or proxy in japan?
295 2011-08-31 17:29:31 <BlueMatt> asoltys: not even sure if its written down yet...bitcoinj implements it, but no desktop client afaik
296 2011-08-31 17:30:02 <asoltys> BlueMatt: ok, thanks
297 2011-08-31 17:30:11 <kinlo> the idea about creating some kind of "aggregate block" should help a lot too - afaik it was described in the paper
298 2011-08-31 17:30:31 <BlueMatt> asoltys: you could google a bit to see if there is a bitcoinj desktop client, but other than that, feel free to ask :)
299 2011-08-31 17:34:12 <iddo> bitcoinj client will store headers of all blocks since genesis block? or it's ok to trust more recent blocks and trim the older blocks?
300 2011-08-31 17:34:35 <tcatm> BlueMatt: do you know whether anyone else than gavin could create the bitcoin.github.com repo?
301 2011-08-31 17:35:59 <iddo> i think in Satoshi's article it's assumed that in simple verification mode all headers since genesis block are kept?
302 2011-08-31 17:36:21 <BlueMatt> tcatm: if you cant, Id assume only gavin can, but you could ask sipa, jgarzik (though they are also both not here atm...)
303 2011-08-31 17:36:51 <laetus> I'm not a fan of bruce wagner.. but I must admit that he has some balls.   no pun intended
304 2011-08-31 17:38:09 <iddo> laetus: why people don't like him?
305 2011-08-31 17:38:41 <BlueMatt> some bs accusations about child relations
306 2011-08-31 17:38:50 <b4epoche> http://buttcoin.org/has-bruce-wagner-pulled-off-the-financial-biggest-scam-on-the-bitcoin-community
307 2011-08-31 17:38:51 <BlueMatt> none of it even really realistic
308 2011-08-31 17:39:18 <laetus> http://www.reddit.com/r/Bitcoin/comments/jztwc/know_you_btc_shills_bruce_wagner/
309 2011-08-31 17:39:24 <iddo> but if the accusations are false, what is the real reason that they don't like him?
310 2011-08-31 17:39:39 <b4epoche> ( )( )
311 2011-08-31 17:39:42 <b4epoche> |\n1658328
312 2011-08-31 17:40:08 <laetus> the child porn allegations are kind of iffy
313 2011-08-31 17:40:26 <laetus> but looking at some of his posting history.. it's not super clean
314 2011-08-31 17:41:00 <laetus> but for some reason I never really liked him, even before this shitstorm happened
315 2011-08-31 17:41:17 <laetus> well... his posts on his own blog are curious to say the least
316 2011-08-31 17:41:35 <laetus> I wouldn't say he is guilty of having sex with underage children though
317 2011-08-31 17:41:36 <b4epoche> the mortgage stuff is proven
318 2011-08-31 17:41:41 <laetus> yes that is true
319 2011-08-31 17:41:57 <b4epoche> and his explanation seems feasible until you read the court documents
320 2011-08-31 17:42:13 <laetus> also.. some sites that offer escort guys.. has the same telephone number as the one on onlyonetv
321 2011-08-31 17:43:37 <b4epoche> anyway, he seems pretty shady
322 2011-08-31 17:43:45 <BlueMatt> meh, whatever
323 2011-08-31 17:44:18 <laetus> which is weird.. because con artists benefit the most from not looking shady at all
324 2011-08-31 17:44:53 <tcatm> bitcoin.org website repo created and gavin emailed :)
325 2011-08-31 17:45:07 <BlueMatt> where?
326 2011-08-31 17:45:14 <BlueMatt> also, shit: https://github.com/bitcoin/bitcoin/commit/59ed5cb398f3817cebd8f5c6aeba3424de2acefe#commitcomment-562434
327 2011-08-31 17:45:26 <tcatm> https://github.com/tcatm/bitcoin.org
328 2011-08-31 17:45:53 <b4epoche> He doesn't seem to be the kind of fellow you really want being a 'face' of bitcoin...
329 2011-08-31 17:46:22 <tcatm> once gavin creates a bitcoin/bitcoin.github.com repo I'll merge it there and github will automatically run the scripts
330 2011-08-31 17:46:29 <laetus> he kind of already was ....   how to handle this in a controlled way?
331 2011-08-31 17:46:49 <bighairyballs> hey
332 2011-08-31 17:48:07 <wagnerwheels> lawl
333 2011-08-31 17:48:28 <b4epoche> so BlueMatt doesn't that seem like an easy fix?
334 2011-08-31 17:49:05 <I_am_your_wagner> lawl
335 2011-08-31 17:49:23 <phantomcircuit> BlueMatt, ^ +b
336 2011-08-31 17:52:39 <BlueMatt> b4epoche: what is easy to fix?
337 2011-08-31 17:53:03 <b4epoche> BlueMatt:: that comment
338 2011-08-31 17:53:28 <BlueMatt> b4epoche: oh, yes, but it needs done...
339 2011-08-31 17:53:53 <b4epoche> although I'm not sure why it matters if the client isn't used for mining
340 2011-08-31 17:54:07 <BlueMatt> it is though
341 2011-08-31 17:54:22 <b4epoche> I thought there was no more client mining
342 2011-08-31 17:54:41 <BlueMatt> client still generates the blocks/txes on which to mine
343 2011-08-31 17:55:47 <b4epoche> I thought the coinbase tx would have been created in the miner (not the client)
344 2011-08-31 17:56:05 <BlueMatt> no, getdata generates the coinbase tx
345 2011-08-31 17:57:16 <b4epoche> so a miner has to generate the coinbase tx with the client?
346 2011-08-31 17:57:23 <gmaxwell> ~.
347 2011-08-31 17:57:29 <b4epoche> although I'm not sure how all this works with pooled mining
348 2011-08-31 17:57:33 <BlueMatt> yea, well the client has to generate the whole blocks
349 2011-08-31 17:58:01 <BlueMatt> because it has to check the txes
350 2011-08-31 17:58:24 <b4epoche> oh, I guess I assumed the mining clients were 'smarter'
351 2011-08-31 17:58:49 <BlueMatt> no, the mining clients just take headers and hash like crazy
352 2011-08-31 17:59:23 <b4epoche> I suppose that pooled mining is why I never saw that stuff&  or was, say, diablo-miner talking to the client?
353 2011-08-31 17:59:55 <b4epoche> it was just talking to the pool and the pool was doing the tx checking, etc., right?
354 2011-08-31 18:00:03 <BlueMatt> a pool almost always uses just a proxy on top of bitcoind
355 2011-08-31 18:00:07 <BlueMatt> so...it also effects pools
356 2011-08-31 18:00:21 <BlueMatt> in fact, I dont know if any pools do it any other way
357 2011-08-31 18:02:10 <Tuxavant> can someone with ops have a talk with phantomcircuit regarding #bitcoin please
358 2011-08-31 18:02:24 <BlueMatt> whats he doing?
359 2011-08-31 18:02:25 <phantomcircuit> im pro moderator
360 2011-08-31 18:02:32 <phantomcircuit> lol i accidentally the channel
361 2011-08-31 18:02:46 <BlueMatt> oh, yea you kicked everyone, didnt you?
362 2011-08-31 18:02:50 <phantomcircuit> yeah
363 2011-08-31 18:02:57 <phantomcircuit> was trying to clear bans and cleared users
364 2011-08-31 18:02:58 <phantomcircuit> lol
365 2011-08-31 18:03:00 <b4epoche> can we just add our own verb?
366 2011-08-31 18:03:12 <phantomcircuit> b4epoche, verbs are
367 2011-08-31 18:03:13 <b4epoche> i accidentally <ver> the channel
368 2011-08-31 18:03:13 <iddo> heh
369 2011-08-31 18:03:18 <b4epoche> +b
370 2011-08-31 18:03:22 <BlueMatt> phantomcircuit: dont clear bans that arent yours
371 2011-08-31 18:03:25 <BlueMatt> ie that you didnt add
372 2011-08-31 18:04:06 <phantomcircuit> usually i wouldn't but im pretty sure they're 90% me
373 2011-08-31 18:04:43 <BlueMatt> what client are you using?
374 2011-08-31 18:04:49 <phantomcircuit> xchat
375 2011-08-31 18:04:55 <BlueMatt> window->ban list
376 2011-08-31 18:12:09 <pakis> hi
377 2011-08-31 18:17:48 <phantomcircuit> BlueMatt, i now how to irc i just made a typo
378 2011-08-31 18:18:11 <Blitzboom> what is the first thing the devs will do when a fatal bitcoin flaw is found?
379 2011-08-31 18:18:12 <b4epoche> lrn2irc
380 2011-08-31 18:18:16 <Blitzboom> 1) sell their bitcoins
381 2011-08-31 18:18:22 <Blitzboom> 2) fix the problem
382 2011-08-31 18:18:27 <Blitzboom> or 3) warn everybody
383 2011-08-31 18:18:40 <phantomcircuit> b4epoche, i will irc the shit out of you
384 2011-08-31 18:18:54 <kinlo> Blitzboom: 2), it will create a seperated chain, learn the bitcoin history, it has happened before
385 2011-08-31 18:19:13 <Blitzboom> i didnt say what kind of flaw
386 2011-08-31 18:19:31 <Blitzboom> and yes, the challenge willbe to get everybody to upgrade
387 2011-08-31 18:19:38 <Blitzboom> then insert this to 2)
388 2011-08-31 18:20:03 <Blitzboom> 2) release new version as soon as possible
389 2011-08-31 18:20:15 <ThomasV> 2) ?
390 2011-08-31 18:20:24 <kinlo> Blitzboom: do you have found such a fatal flaw? :)
391 2011-08-31 18:20:32 <Blitzboom> just asking
392 2011-08-31 18:20:35 <Blitzboom> i bet it would be 1)
393 2011-08-31 18:20:44 <Blitzboom> would be the most rational thing
394 2011-08-31 18:22:52 <BlueMatt> phantomcircuit: check your spelling before hitting enter then ;)
395 2011-08-31 18:22:54 <BlueMatt> its all good though
396 2011-08-31 18:23:23 <Eliel> Blitzboom: a serious misuse of inside information though.
397 2011-08-31 18:23:26 <denisx> lolz: http://pastebin.com/BKcmMd47
398 2011-08-31 18:23:41 <BlueMatt> Blitzboom: all the devs I know dont have enough btc to freak out enough -> they would 2)
399 2011-08-31 18:23:43 <Blitzboom> Eliel: depends
400 2011-08-31 18:23:48 <Eliel> only people whose sense of what's right is skewed would do it.
401 2011-08-31 18:24:02 <Blitzboom> Eliel: depends on your political and economic ideology
402 2011-08-31 18:24:13 <Blitzboom> in a free market, it would be considered a simply price signal to the market
403 2011-08-31 18:24:15 <phantomcircuit> BlueMatt, KICK ALL THE THINGS
404 2011-08-31 18:24:24 <Blitzboom> it gives the market new information
405 2011-08-31 18:24:42 <BlueMatt> phantomcircuit: ????
406 2011-08-31 18:24:54 <phantomcircuit> lol you dont get it
407 2011-08-31 18:25:07 <CIA-101> bitcoin: Florian Schmaus * r4317e5083f30 gentoo/app-misc/poclbm/ (Manifest poclbm-9999.ebuild): fixed sed in poclbm live ebuild
408 2011-08-31 18:25:13 <Blitzboom> BlueMatt: i dont understand why you people dont have enough bitcoins to support yourself tor thex next few decades
409 2011-08-31 18:25:30 <Blitzboom> (theoretically at the current price)
410 2011-08-31 18:25:47 <gmaxwell> phantomcircuit: "OP all the things!" is more funny.
411 2011-08-31 18:25:53 <pigeons> Blitzboom: it's happened before and the blockchain was forked and everybody moved to the new version
412 2011-08-31 18:26:09 <Blitzboom> pigeons: right, but there hasnt been any established market
413 2011-08-31 18:26:16 <Blitzboom> that would be the interesting part
414 2011-08-31 18:26:23 <pigeons> true
415 2011-08-31 18:26:33 <BlueMatt> Blitzboom: gavin has enough, cant speak for sipa, jgarzik or tcatm but I know they dont have too much to freak out...me...well I think I have like 5
416 2011-08-31 18:26:40 <BlueMatt> phantomcircuit: no I get it...but Things?
417 2011-08-31 18:26:45 <phantomcircuit> gmaxwell, lol i used to randomly +o entire channels
418 2011-08-31 18:26:54 <phantomcircuit> and then wait a second and -o
419 2011-08-31 18:26:57 <gmaxwell> phantomcircuit: that was always fun on efnet where it was hard to undo.
420 2011-08-31 18:27:01 <phantomcircuit> lol
421 2011-08-31 18:27:22 <Blitzboom> BlueMatt: if you have too litte bitcoins, your incentive will suck :D
422 2011-08-31 18:27:30 <Blitzboom> monetary incentive
423 2011-08-31 18:28:06 <BlueMatt> Blitzboom: personally, I am a low-risk investor, if you want to do high-risk stuff, fine, but thats just not me...that said if I do good coding for bitcoin I have other incentives
424 2011-08-31 18:28:08 <Blitzboom> thats why i want key persons to be heavily invested financially
425 2011-08-31 18:28:46 <BlueMatt> eg get people like TD, jgarzik, etc to see me and maybe get my foot in the door to places where I could get a job in 4 (or more) years
426 2011-08-31 18:29:01 <BlueMatt> or a job over the summer, or whatever
427 2011-08-31 18:29:14 <Blitzboom> yeah, thats a good investment too
428 2011-08-31 18:29:39 <BlueMatt> that, and if Im bored, what else am I gonna do?
429 2011-08-31 18:29:45 <BlueMatt> anyway...Ima go eat, bb later
430 2011-08-31 18:29:51 <Blitzboom> play some games on your GPU? :P
431 2011-08-31 18:30:03 <BlueMatt> heh, well that too
432 2011-08-31 18:31:40 <Blitzboom> the least bitcoin would bring are free GPUs for earlier miners and very cheap GPUs on ebay
433 2011-08-31 18:31:45 <Blitzboom> :D
434 2011-08-31 18:57:01 <edcba> ;;bc,mtgox
435 2011-08-31 18:57:02 <gribble> {"ticker":{"high":8.85,"low":8.02,"avg":8.487661654,"vwap":8.418129799,"vol":17929,"last":8.34092,"buy":8.34092,"sell":8.35898}}
436 2011-08-31 19:00:08 <md2k7> Mt.Gox have changed their SSL CA to VeriSign, with an intermediate certificate :-(
437 2011-08-31 19:00:19 <md2k7> I can't seem to get the connection right from python
438 2011-08-31 19:00:27 <md2k7> (or wget, or openssl, for that matter)
439 2011-08-31 19:01:18 <md2k7> does anyone know how to specify intermediate certificates?
440 2011-08-31 19:02:06 <md2k7> I was using ca_certs=... file of ssl.wrap_socket()
441 2011-08-31 19:09:54 <tcatm> any ideas how one could visualize "bitcoin" in a logo/image?
442 2011-08-31 19:12:18 <sytse> md2k7: you should only need to trust one certificate in the chain, right?
443 2011-08-31 19:12:57 <md2k7> sytse: errr, I'm not sure what I need to trust, because the "highest" CA certificate isn't working for me
444 2011-08-31 19:14:32 <md2k7> actually, nothing seems to be working, so if you have hints on where to start debugging this, I'd be glad
445 2011-08-31 19:15:01 <aldiyen> you are explicitly specifying a cacerts file path?
446 2011-08-31 19:15:45 <md2k7> $ openssl s_client -CAfile cert -connect mtgox.com:443
447 2011-08-31 19:16:06 <md2k7> or, alternatively, $ wget --ca-certificate cert https://mtgox.com
448 2011-08-31 19:17:20 <md2k7> or in Python: ssl_sock = ssl.wrap_socket(s,ca_certs="cert",cert_reqs=ssl.CERT_REQUIRED)
449 2011-08-31 19:18:19 <md2k7> neither seems to work, not with the Class 3 SGC CA, not with the "G5", not with the server cert
450 2011-08-31 19:18:26 <md2k7> nor with combinations of these
451 2011-08-31 19:18:33 <aldiyen> I think ca_certs needs to be a file object
452 2011-08-31 19:18:35 <aldiyen> not a filename
453 2011-08-31 19:19:17 <aldiyen> or you could just tell it not to check, depending on what you're trying to do
454 2011-08-31 19:19:18 <md2k7> aldiyen: http://docs.python.org/library/ssl.html#client-side-operation
455 2011-08-31 19:19:41 <aldiyen> try specifying the full file path?
456 2011-08-31 19:19:41 <log0s> md2k7: i had some problems with some of the URLs earlier, but when i added www. to the URL, they worked
457 2011-08-31 19:19:46 <md2k7> not to check obviously misses the whole purpose of SSL
458 2011-08-31 19:20:06 <aldiyen> SSL is intended to encrypt the communication
459 2011-08-31 19:20:41 <aldiyen> it's nice to know that the other person is who they say they are, but to say that that is the whole purpose of it is inaccurate
460 2011-08-31 19:21:30 <log0s> the purpose is both...to encrypt the connection and to be reasonably sure that you are connected to who you want to be connected to and not some attacker
461 2011-08-31 19:22:47 <md2k7> well, sending your Mt.Gox password encrypted to an attacker who negotiated the decryption key is barely better than sending it unencrypted
462 2011-08-31 19:23:45 <log0s> yes, the encryption is pointless if anyone can MITM without you knowing since you're ignoring any authentication
463 2011-08-31 19:26:39 <ByteCoin> anyone know anything about transaction replacement
464 2011-08-31 19:26:49 <ByteCoin> or nLockTime?
465 2011-08-31 19:27:41 <log0s> it seems the mtgox URLs i was having trouble with earlier before adding www. to them work either way now
466 2011-08-31 19:28:17 <phantomcircuit> ByteCoin, iirc they're disabled currently
467 2011-08-31 19:28:48 <ByteCoin> Yes I know THAT.. but do you know how they're meant to work if they are enabled?
468 2011-08-31 19:29:28 <gmaxwell> ByteCoin: whats to know? nlocktime keeps a txn from being mined prior to a block index or network timestamp.
469 2011-08-31 19:29:51 <gmaxwell> replacement isn't just disabled there isn't even code for it IIRC.
470 2011-08-31 19:30:36 <ByteCoin> Cool! Some people who know stuff! Ok. Are there restrictions on the transaction that can be made to replace a given transaction?
471 2011-08-31 19:31:20 <ByteCoin> I presume that all that's required is that the new transaction uses all the txins of the existing transaction and has a higher sequence number
472 2011-08-31 19:32:10 <ByteCoin> Can anyone think of a use for nLockTime except for situations involving transaction replacement?
473 2011-08-31 19:32:48 <lfm> ByteCoin loan payments
474 2011-08-31 19:32:52 <gmaxwell> I gave a crazy one.
475 2011-08-31 19:32:55 <gmaxwell> on the forums once
476 2011-08-31 19:33:05 <gmaxwell> where you use it to pay miners to mine a fork.
477 2011-08-31 19:33:23 <ByteCoin> gmaxwell: I vaguely remember... got the title or some keywords?
478 2011-08-31 19:34:06 <ByteCoin> Do you recall off hand how that worked?
479 2011-08-31 19:34:09 <gmaxwell> by creating a series of dependant txn with increasing nlocktime and txn fees in each. Nlocktime prevents the first miner from taking them all as a group in one block, the dependency makes it all or nothing.
480 2011-08-31 19:34:36 <ByteCoin> I remember that one. I posted on that thread a fair amount
481 2011-08-31 19:34:51 <ByteCoin> We argued about some detail
482 2011-08-31 19:34:56 <gmaxwell> Yea, I forget what.
483 2011-08-31 19:35:17 <ByteCoin> lfm: How would nLockTime work for loan payments?
484 2011-08-31 19:35:19 <gmaxwell> I'm sure if I think for a bit I could come up with other nlocktime uses.
485 2011-08-31 19:35:45 <lfm> ByteCoin: not well actually now I think of it cuz the money would be tied up.
486 2011-08-31 19:35:57 <ByteCoin> lfm: Precisely
487 2011-08-31 19:36:52 <lfm> It could be a week form of escrow. proves you have paid but they dont get the money right away
488 2011-08-31 19:37:20 <ByteCoin> lfm: But if you can just replace the transaction with one that sends the money back to yourself?
489 2011-08-31 19:37:28 <ByteCoin> Then what does it guarantee?
490 2011-08-31 19:37:29 <gmaxwell> not if you don't control all the input keys.
491 2011-08-31 19:37:37 <lfm> not sure exactly
492 2011-08-31 19:37:48 <ByteCoin> gmaxell: I see
493 2011-08-31 19:37:49 <gmaxwell> (presuming replacement requires signatures by all the inputs)
494 2011-08-31 19:38:40 <ByteCoin> gmaxwell: So it's for collaboratively signed transactions which inevitably have more than one txin
495 2011-08-31 19:39:11 <lfm> ByteCoin: of just joint accounts
496 2011-08-31 19:39:13 <gmaxwell> I think the examples I've seen describing it being used were all of that kind. iirc there is a page in the wiki with an example like that.
497 2011-08-31 19:39:59 <ByteCoin> Ok. You've been great help, both of you.
498 2011-08-31 20:00:03 <ByteCoin> I hope you guys are still around... The hash of the transaction is used to identify the transaction. There are a number of occasions where you have to make a transaction which spends an incompletely signed transaction. I presume that means that the signatures of the txins are zeroed out for the hash. Correct?
499 2011-08-31 20:07:53 <tcatm> BlueMatt: gavin set up the repo. now we only need to change the DNS :) https://github.com/bitcoin/bitcoin.github.com
500 2011-08-31 20:37:27 <phantomcircuit> tcatm, ima be screwing with intersango.com/.us for a bit and the api stuff is definitely going to break so yeah
501 2011-08-31 20:37:31 <phantomcircuit> just so you know
502 2011-08-31 20:38:01 <tcatm> phantomcircuit: do it on a testing/staging server :P
503 2011-08-31 20:38:17 <phantomcircuit> completely new software migration
504 2011-08-31 20:38:25 <phantomcircuit> so it'll be fine
505 2011-08-31 20:38:28 <phantomcircuit> until it's broken
506 2011-08-31 20:57:31 <BlueMatt> tcatm: nice
507 2011-08-31 20:59:23 <tcatm> I'm also thinking about adding a small news section for release notes and other important, dev-related announcements (I'll volunteer to write both)
508 2011-08-31 21:00:13 <ByteCoin> tcatm & BlueMatt: Is there something wrong with my client that I seem to be missing out on most of your conversation?
509 2011-08-31 21:00:53 <tcatm> ByteCoin: we're having a highly asynchron chat with long delays. like 1 msg per hour :)
510 2011-08-31 21:01:09 <ByteCoin> Got it.
511 2011-08-31 21:36:20 <surikator> Hi everyone! If two nodes discover the same block at the same time (i know, very unlikely) who gets the 50 BTC? More importantly, can it not happen that some nodes take the first transaction of the block (the 50 BTC) as having a different receiver as others and screw up transaction consistency?
512 2011-08-31 21:37:17 <imsaguy> surikator, it would come down to who discovers the next block
513 2011-08-31 21:37:20 <imsaguy> one would 'win'
514 2011-08-31 21:37:23 <imsaguy> the other lose
515 2011-08-31 21:39:29 <surikator> so if node A and B simultaneously discover a block with equal length and then node C discovers the next block, what then?
516 2011-08-31 21:40:07 <ByteCoin> Depends which of A or B, C's block was built off
517 2011-08-31 21:40:51 <ByteCoin> If C built off A, B will lose
518 2011-08-31 21:41:18 <ByteCoin> How do you know which C is likely to build off? Probably which ever he got first
519 2011-08-31 21:41:32 <surikator> i see.
520 2011-08-31 21:41:41 <BlueMatt> jgarzik: ping
521 2011-08-31 21:41:50 <surikator> interesting.
522 2011-08-31 21:41:52 <surikator> thanks!
523 2011-08-31 21:42:38 <BlueMatt> ;;seen gavinandresen
524 2011-08-31 21:42:38 <gribble> gavinandresen was last seen in #bitcoin-dev 5 hours, 54 minutes, and 26 seconds ago: <gavinandresen> ... talk with jgarzik about how it works exactly (I can't recall)
525 2011-08-31 21:46:57 <BlueMatt> ;;later tell gavinandresen I thought you said you replaced addressbook cs with wallet and got that one when you needed address book stuff? rpc GetAccountAddress...
526 2011-08-31 21:46:57 <gribble> The operation succeeded.
527 2011-08-31 21:48:28 <imsaguy> thanks ByteCoin
528 2011-08-31 21:50:09 <BlueMatt> anyone have a spare vps?
529 2011-08-31 21:51:02 <phantomcircuit> BlueMatt, what for?
530 2011-08-31 21:51:11 <BlueMatt> blockchain nightly creation
531 2011-08-31 21:51:22 <phantomcircuit> oh
532 2011-08-31 21:51:24 <BlueMatt> preferably with unmetered bw
533 2011-08-31 21:51:24 <phantomcircuit> nop
534 2011-08-31 21:51:59 <phantomcircuit> there isn't really such a thing
535 2011-08-31 21:52:09 <phantomcircuit> an unmetered 100mbps line is like $800/month
536 2011-08-31 21:52:16 <phantomcircuit> gbps around 2k/month
537 2011-08-31 21:52:18 <BlueMatt> it doesnt need 100mbps
538 2011-08-31 21:52:23 <BlueMatt> like maybe 5 is good
539 2011-08-31 21:52:29 <BlueMatt> if you even need that
540 2011-08-31 21:53:26 <tcatm> BlueMatt: how CPU intense is it? I already have a bitcoin node, lots of free diskspace and unmetered bw
541 2011-08-31 21:53:44 <gjs278> can you max the line 24/7
542 2011-08-31 21:53:48 <gjs278> that unlimited of bw
543 2011-08-31 21:53:52 <tcatm> gjs278: yes
544 2011-08-31 21:53:59 <BlueMatt> tcatm: then its just a script which stops bitcoind, tars blk*, gpg signs it, uploads it
545 2011-08-31 21:54:21 <BlueMatt> but the uploads are getting bigger and bigger, Im just saying unmetered bc bitcoin uses a bit and then the uploads can be big
546 2011-08-31 21:54:55 <tcatm> bitcoind is already running on that server so it would be just cp blk* /somewhere else
547 2011-08-31 21:55:23 <lfm> scp?
548 2011-08-31 21:55:24 <BlueMatt> well it would always be better to stop bitcoind first
549 2011-08-31 21:55:43 <BlueMatt> or patch bitcoin to stop writing for a minute
550 2011-08-31 21:56:05 <tcatm> or run a 2nd instance
551 2011-08-31 21:56:07 <lfm> backupblk like backupwallet ?
552 2011-08-31 21:56:16 <gjs278> bitcoind freeze
553 2011-08-31 21:56:27 <BlueMatt> tcatm: thats what I used to do, second instance for a couple minutes
554 2011-08-31 21:56:42 <tcatm> so does the script already exist?
555 2011-08-31 21:56:57 <BlueMatt> I think i lost it...but it might take like...1 minute to write?
556 2011-08-31 21:59:28 <BlueMatt> Im just out of mem on mine, and bw is tight...
557 2011-08-31 22:22:11 <gribble> The operation succeeded.
558 2011-08-31 22:22:11 <tcatm> ;;later tell BlueMatt http://eu1.bitcoincharts.com/blockchain/ should be updated daily, missing signing for now
559 2011-08-31 23:41:40 <lfm> surikator  random string can't be compressed
560 2011-08-31 23:48:23 <luke-jr> lfm: sometimes it can!
561 2011-08-31 23:48:31 <luke-jr> lfm: like 1 MB of nulls
562 2011-08-31 23:49:52 <ByteCoin> It's ill-defined to talk about random strings. You can only talk about random sources.