1 2013-03-22 00:13:29 <muhoo> hmm, bitcoinj does not use change addresses
2 2013-03-22 00:13:39 <muhoo> or rather, multibit does not
3 2013-03-22 00:14:11 <sipa> what?
4 2013-03-22 00:14:25 <muhoo> multibit. i just sent money, got the change back to the send address
5 2013-03-22 00:15:02 <muhoo> looks like the source code of bitcoinj does have something like change addresses, will experiment to make sure.
6 2013-03-22 00:16:52 <muhoo> yes Wallet.getChangeAddress(), randomly picks an address from the keychain to use for change. hmm. maybe i just lucked out and multibit picked the same address by chance.
7 2013-03-22 00:17:27 <muhoo> ugh, actually, it just picks the *first* address in the keychain, not random.
8 2013-03-22 00:18:27 <muhoo> sipa: http://code.google.com/p/bitcoinj/source/browse/core/src/main/java/com/google/bitcoin/core/Wallet.java#2006
9 2013-03-22 00:18:46 <denisx> what was the adress of the list of the most stable bitcoin nodes again?
10 2013-03-22 00:20:15 <muhoo> denisx: do you mean the seed nodes?
11 2013-03-22 00:20:22 <muhoo> for dns?
12 2013-03-22 00:20:30 <denisx> no
13 2013-03-22 00:21:21 <sipa> denisx: http://bitcoin.sipa.be/seeds.txt ?
14 2013-03-22 00:21:30 <denisx> sipa: yes
15 2013-03-22 00:24:14 <denisx> sipa: is there also nicer version with filter features or something?
16 2013-03-22 00:26:54 <sipa> denisx: it's just a text file :)
17 2013-03-22 00:28:17 <TheLordOfTime> so, just a question about the entire backend. Bitcoin confirmations... what actually has to confirm the transactions, other nodes? some supernodes? something else?
18 2013-03-22 00:28:36 <TheLordOfTime> and does it attempt to confirm every time a block is received/processed, or... what?
19 2013-03-22 00:29:36 <sipa> TheLordOfTime: 'confirmations' is really just how deep the transaction is buried in the block chain
20 2013-03-22 00:29:51 <sipa> the only time it requires 'action' is when a miner puts it in a block
21 2013-03-22 00:30:04 <sipa> every block on top of that block causes an implicit extra confirmation
22 2013-03-22 00:31:06 <muhoo> would it be horribly NIH of me to scrap bitcoinj and just write my own protocol implementation in clojure, using clojure tools like immutability and clojure concurrency primitives, clojure wrappers around netty, etc?
23 2013-03-22 00:31:42 <sipa> muhoo: i think you underestimate the work
24 2013-03-22 00:31:43 <CodeShark> TheLordOfTime: "confirmation" doesn't mean that it is certain - it just means it got harder for someone to remove
25 2013-03-22 00:31:58 <TheLordOfTime> ah i see. that explains it
26 2013-03-22 00:32:14 <TheLordOfTime> ah, and the transaction in question has 1/6 confirmations thus far :P
27 2013-03-22 00:32:26 <TheLordOfTime> so, the deper into the block chain this goes, the more "confirmed" it is?
28 2013-03-22 00:32:30 <muhoo> sipa: i don't. it seems quite a lot. but i'd probably lift the serialization/deserialization stuff, because i'd rather not have to write that.
29 2013-03-22 00:32:30 <sipa> yes
30 2013-03-22 00:32:50 <sipa> muhoo: right, depends on the scope of course
31 2013-03-22 00:33:03 <phantomcircuit> you should do it
32 2013-03-22 00:33:10 <phantomcircuit> because then someone would have done it
33 2013-03-22 00:33:11 <sipa> and you perhaps have a better idea than most who consider starting something like that :)
34 2013-03-22 00:33:44 <muhoo> well i've been working with the bitcoinj codebase for about 5 months now. and dealing with java is making me crazy.
35 2013-03-22 00:34:00 <sipa> haha
36 2013-03-22 00:34:32 <muhoo> mike's code is nice, but as nice as it is, it's still java.
37 2013-03-22 00:35:27 <CodeShark> java the language isn't too entirely horrible...could be worse. At least it has a pretty decent set of libraries with a more or less consistent API
38 2013-03-22 00:35:46 <muhoo> i'm not an OOP guy. i prefer FP.
39 2013-03-22 00:36:01 <CodeShark> it does suck that Java forces you to make everything a class :p
40 2013-03-22 00:36:02 <muhoo> so it's mostly personal taste
41 2013-03-22 00:36:04 <CodeShark> that's idiotic :P
42 2013-03-22 00:36:19 <CodeShark> it's OOP taken to logical absurdity
43 2013-03-22 00:36:26 <sipa> not at all
44 2013-03-22 00:36:28 <CodeShark> into applications where it no longer makes sense.
45 2013-03-22 00:36:35 <sipa> it even has non-object types!
46 2013-03-22 00:36:37 <muhoo> well i didn't mean to start a language flamewar
47 2013-03-22 00:36:40 <CodeShark> lol sipa
48 2013-03-22 00:36:47 <sipa> ever used smalltalk?
49 2013-03-22 00:36:52 <sipa> _everything_ is an object
50 2013-03-22 00:37:05 <CodeShark> yeah, smalltalk is even more extreme in that sense
51 2013-03-22 00:37:10 <iwilcox> muhoo: Inevitable :)
52 2013-03-22 00:37:51 <muhoo> iwilcox: there used to be a saying, never discuss politics or religion in polite company. i gues on the internet, never discuss language, editors, or os'es, since those are the politics and religion of the internet.
53 2013-03-22 00:38:01 <CodeShark> java the stack machine sucks, though :)
54 2013-03-22 00:38:21 <iwilcox> JIT rocks, though
55 2013-03-22 00:38:38 <muhoo> i think the JVM is an impressive peice of kit. i just prefer using clojure to work with it :_)
56 2013-03-22 00:38:46 <CodeShark> JIT has made it just barely tolerable
57 2013-03-22 00:39:25 <CodeShark> there are so many times I would love to tell the runtime environment I'm done with something and that it should clean up and give me back some more resources
58 2013-03-22 00:39:46 <CodeShark> I'm not a big fan of automatic garbage collection
59 2013-03-22 00:40:17 <CodeShark> I can tolerate it in dynamically typed languages
60 2013-03-22 00:40:22 <CodeShark> because it affords other powerful features
61 2013-03-22 00:41:14 <iwilcox> I thought you got enough control these days with the WeakReference or whatever stuff.
62 2013-03-22 00:41:53 <CodeShark> I like the RAII paradigm
63 2013-03-22 00:42:02 <CodeShark> { give me something } I'm done with it
64 2013-03-22 00:42:43 <CodeShark> { new Something(); } hey, I don't need it anymore...
65 2013-03-22 00:49:01 <BlueMatt> anyone have slides of an introduction to bitcoin in 15 minutes I could steal?
66 2013-03-22 00:49:45 <donpdonp> BlueMatt: depends on the quality you're looking for :) mine is at http://www.slideshare.net/donpdonp/bitcoin-101
67 2013-03-22 00:52:32 <BlueMatt> thanks, Ill probably cobble together something by stealing slides from anywhere I can find and giving some bastardized hodgepodge
68 2013-03-22 01:01:31 <CodeShark> sipa: r.e. smalltalk, it represents a pretty significant advance in the way people thought about writing software - it really was innovating with the whole object oriented approach. Java, on the other hand, just poorly copies certain features of C++ where they don't really belong
69 2013-03-22 01:01:49 <CodeShark> like public static void main...
70 2013-03-22 01:01:51 <CodeShark> wtf?!!?!
71 2013-03-22 01:04:26 <CodeShark> I don't often have occasion to reuse application entry points and subclass them :p
72 2013-03-22 01:05:00 <kadoban> that's a rather bad example of what's wrong with java, isn't it? it's super verbose for no reason, but it's not like it affects anything
73 2013-03-22 01:05:29 <CodeShark> I don't mind verbose if it's descriptive and conceptually deep
74 2013-03-22 01:05:46 <CodeShark> the issue here is more conceptual
75 2013-03-22 01:07:39 <gmaxwell> I wonder if there have been any websites that displayed confirmation counts in megajoules
76 2013-03-22 01:08:48 <gmaxwell> would make confirmations more comparible for things like btc vs ltc.
77 2013-03-22 01:24:15 <realazthat> hey
78 2013-03-22 01:24:48 <realazthat> so the ecdsa used, is it NIST*p? or secp256k1? or what?
79 2013-03-22 01:28:31 <lianj> realazthat: secp256k1, but readup on the wiki
80 2013-03-22 01:29:29 <realazthat> lianj: I am reading https://en.bitcoin.it/wiki/Elliptic_Curve_Digital_Signature_Algorithm
81 2013-03-22 01:29:42 <realazthat> but unless I am missing something, it doesn't give that info
82 2013-03-22 01:30:16 <lianj> hm someone should add it
83 2013-03-22 01:30:49 <realazthat> ok found some more info on another page
84 2013-03-22 01:30:54 <realazthat> https://en.bitcoin.it/wiki/Secp256k1
85 2013-03-22 01:31:28 <realazthat> the issue I am having with Secp256k1, is that the lib I am using (python-ecdsa) is for some reason expecting the sig to be 64 bytes
86 2013-03-22 01:31:38 <realazthat> and its 70 (usually 71,72,73)
87 2013-03-22 01:31:51 <realazthat> so I have some wrong idea here
88 2013-03-22 01:34:03 <realazthat> ok overcame that; the sig is in der format apparently
89 2013-03-22 02:19:56 <Billdr> I'm reading over the API calls. "sendfrom" says "<amount> is a real and is rounded to 8 decimal places." Is "real" a numeric type?
90 2013-03-22 02:25:44 <Luke-Jr> Billdr: JSON only has one Number type. "real" is a mathematical English term.
91 2013-03-22 02:29:22 <keystrike> getting closer and closer to the v2 block transition
92 2013-03-22 02:32:29 <denisx> keystrike: btcguild alone has more than 5% v1 blocks
93 2013-03-22 02:33:13 <denisx> or did btcguild switch to v2 recently so these are generated no more
94 2013-03-22 02:35:20 <doublec> 86.7% last I looked
95 2013-03-22 02:40:41 <Graet> denisx, they switched
96 2013-03-22 02:40:52 <Graet> patched thier getwork servers
97 2013-03-22 02:41:11 <denisx> ok, then it will soon happen
98 2013-03-22 02:41:22 <Graet> yes :)
99 2013-03-22 02:42:04 <gmaxwell> Graet: so how much hashrate does ozcoin have on the avalon 'default' account? :P
100 2013-03-22 02:43:12 <Graet> varies
101 2013-03-22 02:46:34 <ashod> what is used more often ?, blockexplorer.com or blockchain.info
102 2013-03-22 02:46:51 <SomeoneWeird> probably the latter
103 2013-03-22 02:47:56 <gmaxwell> ashod: What are you working on in any case? You're not a grad student working for an academic on yet another throughly confused and inaccurate paper on bitcoin constructed by screen scraping blockexplorer are you?
104 2013-03-22 02:48:28 <SomeoneWeird> lmfao
105 2013-03-22 02:49:11 <Graet> gmaxwell, avalons were mining for 2 weeks before i was told we wre default pool, and then i found 2 of the 3 were set to ports that werent open... so it wasnt a huge advantage....
106 2013-03-22 02:49:17 <ashod> tell me who you are fist gmaxwell, then i'll give you my info
107 2013-03-22 02:49:44 <Graet> lol
108 2013-03-22 02:49:56 <jgarzik> ashod: http://bitcoin.org/en/development
109 2013-03-22 02:49:59 <SomeoneWeird> ashod, he's gregory maxwell
110 2013-03-22 02:50:24 <Duly> i could hear the hand slapping the face
111 2013-03-22 02:50:27 <Duly> in my mind
112 2013-03-22 02:50:32 <SomeoneWeird> oshieet new bitcoin site
113 2013-03-22 02:50:55 <ashod> and I ashod apakian
114 2013-03-22 02:51:07 <ashod> i'm asking who he is
115 2013-03-22 02:52:10 <ashod> So gregory maxwell baby ? - tell me a bit about yourself
116 2013-03-22 02:52:14 <gmaxwell> ashod: Sure, and??? and you got something like three people telling you who I was??? it's no secret.
117 2013-03-22 02:52:52 <gmaxwell> ashod: Well??? I like long walks on beaches, fresh fruit, and baiting trolls on IRC.
118 2013-03-22 02:53:10 <ashod> come across any trolls lately greg ?
119 2013-03-22 02:53:53 <gmaxwell> A few!
120 2013-03-22 02:54:04 <digitalmagus> LOL
121 2013-03-22 02:54:07 <ashod> You don't like trolls do you ?
122 2013-03-22 02:54:36 <Graet> no-one really does in the -dev chanel
123 2013-03-22 02:55:10 <ashod> of course, who does
124 2013-03-22 02:55:26 <ashod> i don't like trolls either, -
125 2013-03-22 02:55:49 <gmaxwell> Well, I certantly spend a lot of time interacting with them for someone who theoretically doesn't like them. ??? In any case, enough about me. You've been asking a lot of random questions about things and it doesn't seem you've been getting very far. You've been getting lots of answers but I don't think we really understand how to help you, because the questions keep repeating. If we had some idea of what you were up to we could help more.
126 2013-03-22 02:56:23 <SomeoneWeird> +++++1
127 2013-03-22 02:56:29 <jgarzik> gmaxwell: ever do any simulations or tests with UDP block relaying?
128 2013-03-22 02:56:40 <jgarzik> gmaxwell: It seems worthwhile... if retries are not a problem
129 2013-03-22 02:56:51 <ashod> well ask away ; unlike a troll, I will answer your questions
130 2013-03-22 02:56:55 <jgarzik> ..which seems like something only answerable with tests
131 2013-03-22 02:56:59 <gmaxwell> And if my presumptions insulted you, I apologize. Though I must say, being accused of potentially being a grad student is ... not a terribly grave insult in my culture.
132 2013-03-22 02:57:28 <jgarzik> UDP seems _really_ nice for transaction broadcasting.
133 2013-03-22 02:57:51 <ashod> do you think I'm a grad student ?
134 2013-03-22 02:57:56 <jgarzik> And with our P2P network, seems like it would be easy to rig up an automatic UDP hole punch.
135 2013-03-22 02:58:32 <B0g4r7> I ran bitcoin-qt and sent some BTC while it was still synchronizing (it was 12 hours out of date), and now the transaction shows in my client, but not on blockchain.info...
136 2013-03-22 02:58:34 <gmaxwell> jgarzik: Lack of (or mismatched) flow control has implications for block relaying. For transactions not so much. No simulations, but here is why I thought it was interesting: no limit on the number of concurrent 'connections'.
137 2013-03-22 02:58:35 <SomeoneWeird> ashod, who cares
138 2013-03-22 02:58:46 <B0g4r7> How can I fix? I already tried restarting bitcoin-qt.
139 2013-03-22 02:58:59 <SomeoneWeird> it's blockchains fault
140 2013-03-22 02:59:05 <SomeoneWeird> bc.info*
141 2013-03-22 02:59:11 <B0g4r7> It is huh? They're b0rked?
142 2013-03-22 02:59:29 <SomeoneWeird> no, they just havn't received the transaction yet
143 2013-03-22 02:59:31 <ashod> well let's see if I can be of any help to bitcoin - . I think I have enough experience in software, to be of some use to you guys
144 2013-03-22 02:59:42 <gmaxwell> B0g4r7: any chance you've double spent.
145 2013-03-22 02:59:50 <B0g4r7> Usually it's instant.
146 2013-03-22 03:00:01 <B0g4r7> I don;t know. How would that happen?
147 2013-03-22 03:00:23 <B0g4r7> Nothing was spent or received in the unsyncronized blocks.
148 2013-03-22 03:00:29 <gmaxwell> ashod: That would be fantastic, as always. So what are you trying to dig into? you've been asking a bunch of questions in the area of history analysis and the online block explorers. What questions are you trying to answer or what tools are you trying to build?
149 2013-03-22 03:00:46 <ashod> well I'm building 2 things
150 2013-03-22 03:01:46 <B0g4r7> Usually it shows instantly on bc.info, even if unconfirmed. There's been a block now and still nothing.
151 2013-03-22 03:02:01 <ashod> let me ask you @gmaxwel, what do you think I'm really trying to do here. you seem to have an opinion of this
152 2013-03-22 03:02:07 <B0g4r7> tx id af9314b262f2fd3fc9438bf7f387e106fcc34b68cee3b6c194467e5cc3178a26
153 2013-03-22 03:03:01 <hmmmstrange> ACTION is building 3 things
154 2013-03-22 03:03:22 <SomeoneWeird> ashod, he just asked you, so how would he know?
155 2013-03-22 03:03:36 <ashod> Without offending anyone, - why does this channel exist ? - Is bitcoin incomplete ? - or do some bugs still exist, that require solving ? or is there an ongoing commitment to improving the performance of bitcoin. - I ask this, because it seems bitcoin is working..
156 2013-03-22 03:03:39 <Saberu> for Scrypt mining (Litecoins) is the memory usage proportional to the speed of the hashing????so if you are hashing 1000khash would use 10x more memory than 100khash?
157 2013-03-22 03:03:56 <SomeoneWeird> all of the above
158 2013-03-22 03:04:04 <Graet> bitcoin is beta, maybe you should read up on iy a bit ashod :)
159 2013-03-22 03:04:07 <Duly> yeah this is the Jiffy Lube for the bitcoin car
160 2013-03-22 03:04:44 <Saberu> anyone have an idea about my question? thanks
161 2013-03-22 03:04:57 <Graet> Saberu, this is #bitcoin-dev - no-one cares
162 2013-03-22 03:04:58 <ashod> What is required for it to be no longer be in beta ? -
163 2013-03-22 03:05:16 <Saberu> ok
164 2013-03-22 03:05:36 <ashod> Yes, you need to throw a lot of tech jargon, otherwise you're not worthy
165 2013-03-22 03:05:47 <gmaxwell> ashod: It is far from complete, ... and, well, obviously the team here cares little about performance having only improved it by an order of magnitude in the last major release....
166 2013-03-22 03:05:48 <B0g4r7> So no idea how I can cause my transaction to be broadcast again?
167 2013-03-22 03:06:29 <ashod> @gmaxwell So everyone is happy with the performace at the moment ?
168 2013-03-22 03:06:55 <pigeons> much happier than a few months ago
169 2013-03-22 03:07:26 <ashod> yet, @gmaxwell, - it's far from complete ... - Why so
170 2013-03-22 03:07:31 <gmaxwell> ashod: I suspect that the sarcasm in my last statement may have been lost on you. No, we're not happy with it now.
171 2013-03-22 03:07:35 <Duly> it speaks like a journalist or reporter not like a dev
172 2013-03-22 03:07:43 <SomeoneWeird> >.<
173 2013-03-22 03:08:24 <jgarzik> nice
174 2013-03-22 03:08:37 <gmaxwell> ashod: Unfortunately if I start compactly enumerating the missing things you're likely to accuse me of 'throw(ing) a lot of tech jargon'
175 2013-03-22 03:08:44 <jgarzik> http://bitcoin.org/en/choose-your-wallet now recommends Bitcoin-Qt or Android's Bitcoin Wallet
176 2013-03-22 03:08:46 <jgarzik> ACTION cheers
177 2013-03-22 03:09:11 <ashod> np, I didn't pick up on the sarcasm .. - @gmaxwell, is performance the main issue keeping it in beta -- e.g if you could speed things up by 50% , would that be cause to leave beta status
178 2013-03-22 03:09:38 <ashod> it's fine greg, there is no tech jargon you can throw at me, that will phaze me
179 2013-03-22 03:10:16 <ashod> in face greg, you and I have crossed paths before
180 2013-03-22 03:10:22 <gmaxwell> ashod: No, improved performance is important??? and there are more improvements in the pipeline... but not sufficient.
181 2013-03-22 03:10:40 <Ry4an> ashod: here is a list of specific things people hope to complete on the road to 1.0: https://github.com/bitcoin/bitcoin/issues
182 2013-03-22 03:11:05 <gmaxwell> Ry4an: and far from complete!
183 2013-03-22 03:11:17 <gmaxwell> ashod: we have?
184 2013-03-22 03:11:27 <ashod> yes, a couple of times actually
185 2013-03-22 03:11:58 <gmaxwell> I have no clue, but thats not unusual.
186 2013-03-22 03:12:51 <gmaxwell> In any case, I have a list of middle term features of varrying importance, its due for an update at some point soon, https://en.bitcoin.it/wiki/User:Gmaxwell/features
187 2013-03-22 03:13:34 <ashod> Yes, in fact , we got a long really well back then -
188 2013-03-22 03:13:50 <ashod> So when I heard you we're working on this project, I thought I would look into it
189 2013-03-22 03:15:47 <gmaxwell> I'm flattered.
190 2013-03-22 03:16:29 <ashod> well I do respect you as a coder..
191 2013-03-22 03:17:08 <ashod> and as we both know , developing codecs, is probably the most challenging thing one can work on - , god knows, we spent enough time doing that
192 2013-03-22 03:18:04 <ashod> might get in touch with michael militzer again and work on h265.
193 2013-03-22 03:18:28 <keystrike> disk corruption robust wallet storage <- nice
194 2013-03-22 03:18:46 <nanotube> jgarzik: heh yes, though i must say boostrapping qt from scratch is not really in the category of 'fast'. :)
195 2013-03-22 03:20:10 <nanotube> jgarzik: maybe if it started up in spv mode while downloading the blockchain from the tip back, as i think gmax has suggested at some point in the past...
196 2013-03-22 03:20:59 <gmaxwell> nanotube: you don't actually have to go backwards, though that has some useful anti-DOS properties. You can get the headers first forward then fill the blocks in, fetching in whatever order.
197 2013-03-22 03:21:21 <gmaxwell> (though obviously validating must move forward)
198 2013-03-22 03:22:38 <nanotube> yea, at any rate that'd be a cool way to shorten the perceived startup time.
199 2013-03-22 03:23:21 <nanotube> which even with .8 code can take a few hours
200 2013-03-22 03:26:00 <gmaxwell> nanotube: thats largely network behavior limited right now. Doing headers first would allow parallel block fetching which would also largely solve that.
201 2013-03-22 03:27:02 <nanotube> hmm well, isn't it also cpu-limited though, on the verification side?
202 2013-03-22 03:27:04 <gmaxwell> Right now what happens is you'll get an overloaded or slow peer connected to first and then pull from it serially... very slow. Without the network issues my laptop syncs in about an hour??? and much of that is on ECDSA which sipa's new code makes 4x faster.
203 2013-03-22 03:27:28 <gmaxwell> nanotube: less cpu limited now that we have parallel signature checking, at least.
204 2013-03-22 03:27:28 <nanotube> i know my 0.8.0 rexndexing took about 1.5 hours on a single core.
205 2013-03-22 03:28:04 <nanotube> (because my laptop cpu was getting uncomfortably hot running on both :) )
206 2013-03-22 03:28:29 <nanotube> but yes, that doesn't sound half bad, with parallel pulls and 4xfaster ecdsa.
207 2013-03-22 03:48:22 <sivu> hmm.. bitcoinj gurus around?
208 2013-03-22 03:50:47 <iwilcox> Howcome addresses these days all seem to match 1[A-Z0-9] but never have the lowercase letters used in that first char?
209 2013-03-22 03:52:01 <gmaxwell> iwilcox: because the version byte is a byte. and 2^8 > 58 Basically the first character also expresses part of the version data.
210 2013-03-22 03:52:19 <gmaxwell> What major rule proofs and I currently drawing a blank on here? https://en.bitcoin.it/wiki/User:Gmaxwell/features#Proofs
211 2013-03-22 03:52:26 <iwilcox> Sorry, in that first char after the 1
212 2013-03-22 03:52:43 <iwilcox> Oh, right.
213 2013-03-22 04:19:44 <peawormsworth> I would like to make my own snapshot of the block data from an up to date bitcoin-qt install... and then apply this to other new bitcoin-qt installs. But I am getting errors. Is there a way to put the data into a state for easy transfer?
214 2013-03-22 04:20:59 <peawormsworth> fyi: im using 0.8 client. i am copying blkindex.dat, blk0001.dat and the content of the blocks directory. I get: "cant open database file blkindex.dat, error 22"
215 2013-03-22 04:23:36 <gmaxwell> Bitcoin 0.8 should never say anything about "blkindex.dat"
216 2013-03-22 04:24:41 <peawormsworth> o crap... ur right. I seem to have 0.7 and the original source is from 0.8. That could be an issue. thx.
217 2013-03-22 04:25:43 <nanotube> gmaxwell: maybe 'proof or block hash too high' ?
218 2013-03-22 04:26:35 <gmaxwell> I talk good.
219 2013-03-22 04:27:16 <gmaxwell> oh I see what you're saying.
220 2013-03-22 04:28:41 <nanotube> er, s/or/of/ :)
221 2013-03-22 04:30:53 <gmaxwell> nanotube: So??? Added a bit. I think I was working from the assumption that nodes would check all the things a spv node can check, but at least further out in the future that might be considered expensive.
222 2013-03-22 04:32:03 <nanotube> i see it
223 2013-03-22 04:32:25 <nanotube> yea that covers it
224 2013-03-22 04:38:25 <nanotube> would spending an input with value X and having an output with value X+e fall under 'invalid script' ?
225 2013-03-22 04:42:04 <nanotube> or maybe 'nonexisting input' hmmm
226 2013-03-22 04:42:14 <nanotube> gmaxwell: ^
227 2013-03-22 04:42:54 <gmaxwell> that would be a case of invalid script, but good catch... that one requires all the inputs as proof.
228 2013-03-22 04:44:42 <peawormsworth> i am still not able to copy blk*.dat and block/* files from one bitcoin-qt install to another new install. it now requests that i reload the chain.
229 2013-03-22 04:45:51 <peawormsworth> gmaxwell: you once told me: "This won't be an issue starting with the 0.8 release, because we don't store the blockchain in .dat files any more." ... so I dont need to use "detachdb=1" option to transfer the chain data. But I cant seem to find a solution for creating my own snapshots.
230 2013-03-22 04:47:56 <gmaxwell> We don't store the _index_ that way anymore. You still need to copy the indexes if you don't want them rebuilt... though it can now rapidly reindex from just the chain data.... and the new database doesn't need detachdb.
231 2013-03-22 04:52:36 <peawormsworth> I am copying the files: blk0001.dat, blkindex.dat and block/*. and the 2nd install is rejecting and saying "reload the block chain?" which starts a full reload i think, which seems to defeat my goals. Is there a file I am forgetting?
232 2013-03-22 04:56:49 <warren> peawormsworth: what version is the new client?
233 2013-03-22 04:57:49 <peawormsworth> 0.8.0
234 2013-03-22 04:58:22 <warren> peawormsworth: you are aware that 0.8 needs to reload the chain anyway?
235 2013-03-22 04:59:02 <peawormsworth> warren: do u mean there is no way to "jump start" a new install of bt-qt in order to avoid the hours of download?
236 2013-03-22 04:59:28 <warren> peawormsworth: you can load it from the 0.7 blocks, but it needs to reindex them using the new db, which is much faster than 0.7's db
237 2013-03-22 04:59:35 <gmaxwell> warren: he's copying 0.8 to 0.8 AFAIK
238 2013-03-22 04:59:39 <warren> oh
239 2013-03-22 04:59:41 <warren> sorry
240 2013-03-22 04:59:51 <warren> argh
241 2013-03-22 05:00:01 <warren> not getting enough sleep
242 2013-03-22 05:00:09 <peawormsworth> yes. i have an up to date 0.8. and i am trying to "jump start" a 2nd copy of 0.8
243 2013-03-22 05:01:04 <gmaxwell> peawormsworth: just copy everything except wallet.dat from a shut down (but not detached, no need) node... should be good to go.
244 2013-03-22 05:02:09 <peawormsworth> yes. i will try again. maybe i just made a silly mistake. FYI: the specific error in debug.log is "errors in block header" and "failed to read block"
245 2013-03-22 05:02:17 <warren> gmaxwell: is there a way to safely clone the blocks + db from a running bitcoind?
246 2013-03-22 05:03:28 <gmaxwell> warren: you can just copy the blockfiles, ??? and reindex on the other side. You can not safely copy in in-flight leveldb.
247 2013-03-22 05:03:47 <gmaxwell> We could quiet the db for copying... but I dunno if anyone would bother.
248 2013-03-22 05:03:55 <gmaxwell> e.g. a pausedb and unpausedb rpc calls.
249 2013-03-22 05:04:04 <Transisto> Can I do something so bitcoind to get more connection ? been stuck at ~36 for more than 24h (I know it's sufficient, I just want to seed as much as I can)
250 2013-03-22 05:04:11 <gmaxwell> and eventually running out of ram if you don't unpause
251 2013-03-22 05:04:30 <peawormsworth> yeah... but what would be the difference then between shutting it down? if no updates to db are allowed? maybe for read only purposes?
252 2013-03-22 05:05:23 <warren> peawormsworth: I've corrupted the leveldb with an unclean server shutdown a few times
253 2013-03-22 05:05:27 <warren> on Windows and Linux
254 2013-03-22 05:05:57 <warren> For that reason I'd personally keep recent offline cold spares
255 2013-03-22 05:05:59 <peawormsworth> warren: ah yes. and also gmaxwell points to the idea of moving to ram only. So i should be quiet now. sry.
256 2013-03-22 05:06:47 <gmaxwell> The idea would be to make the database readonly so it could be safely copied out. Bitcoind has everything needed for that, would probably only be a few lines of code....
257 2013-03-22 05:07:04 <gmaxwell> but whats the use case for copying it while running?
258 2013-03-22 05:07:16 <warren> ah, you could make cold spare index copies for faster startup in the event of corruption (which happens...)
259 2013-03-22 05:07:39 <gmaxwell> warren: we'd rather fix the corruption. :P
260 2013-03-22 05:07:48 <warren> I pointed out one of the corruption bugs
261 2013-03-22 05:08:01 <gmaxwell> I know, one you pointed out is a BTRFS bug though.
262 2013-03-22 05:08:09 <warren> oh really? ok.
263 2013-03-22 05:08:16 <gmaxwell> BTRFS corrupts sparse files.
264 2013-03-22 05:08:25 <warren> I had other weird leveldb failures on windows 7
265 2013-03-22 05:08:38 <gmaxwell> Yea, not the only report either.
266 2013-03-22 05:08:54 <gmaxwell> So I know they happen. I do believe all linux reports I've seen have been on btrfs.
267 2013-03-22 05:09:23 <gmaxwell> as far as fast recovery, if you're doing something where fast recovery is urgent you should really have multiple nodes.
268 2013-03-22 05:09:49 <warren> point being, if your business relies upon bitcoind uptime, you can ensure WAYYY faster restart times even in the ultra rare case of corruption if you regularly make cold spare copies of the index.
269 2013-03-22 05:10:00 <warren> good point
270 2013-03-22 05:10:16 <jgarzik> Man, the blocks are flying off the assembly line today.
271 2013-03-22 05:10:31 <weex> diff going up a bit i'd say
272 2013-03-22 05:10:43 <warren> jgarzik: and p2pool somehow made up for its entire bad luck reputation
273 2013-03-22 05:33:45 <berni23> fuck me sideways: http://www.derivativesintelligence.com/Article/3176278/News/Big-Put-Trade-Rocks-Fx-Mart.html
274 2013-03-22 05:34:43 <gmaxwell> berni23: does it mention bitcoin?
275 2013-03-22 05:34:58 <berni23> wrong chanell
276 2013-03-22 05:35:00 <berni23> no
277 2013-03-22 05:35:05 <warren> gmaxwell: would be awesome to have a $900M bitcoin PUT
278 2013-03-22 05:35:15 <berni23> but it influeces btc
279 2013-03-22 05:35:39 <petertodd> berni23: then say it in #bitcoin, this is the -dev channel
280 2013-03-22 05:35:52 <berni23> i know
281 2013-03-22 05:35:56 <berni23> [07:34:24] <berni23> wrong chanell
282 2013-03-22 05:36:00 <berni23> sooooory
283 2013-03-22 06:04:42 <bfdshreshre> http://redd.it/1asd4s 50btconly$1099 last day
284 2013-03-22 06:04:43 <bfdshreshre> http://redd.it/1asd4s 50btconly$1099 last day
285 2013-03-22 06:04:44 <bfdshreshre> http://redd.it/1asd4s 50btconly$1099 last day
286 2013-03-22 06:04:45 <bfdshreshre> http://redd.it/1asd4s 50btconly$1099 last day
287 2013-03-22 06:04:46 <bfdshreshre> http://redd.it/1asd4s 50btconly$1099 last dayhttp://redd.it/1asd4s 50btconly$1099 last day
288 2013-03-22 06:05:07 <bfdshreshre> http://redd.it/1asd4s 50btconly$1099 last day
289 2013-03-22 06:05:08 <bfdshreshre> http://redd.it/1asd4s 50btconly$1099 last day
290 2013-03-22 06:05:09 <bfdshreshre> http://redd.it/1asd4s 50btconly$1099 last day
291 2013-03-22 06:05:36 <bfdshreshre> http://redd.it/1asd4s 50btconly$1099 last day
292 2013-03-22 06:05:37 <bfdshreshre> http://redd.it/1asd4s 50btconly$1099 last day
293 2013-03-22 06:15:32 <Transisto> asking again
294 2013-03-22 06:15:35 <Transisto> Can I do something so bitcoind to get more connection ? been stuck at ~36 for more than 24h (I know it's sufficient, I just want to seed as much as I can)
295 2013-03-22 06:25:46 <Transisto> everything is working fine, I just wish to report I get : "errors" : "EXCEPTION: St9bad_alloc \\nstd::bad_alloc \\nbitcoin in ProcessMessages() \\n"
296 2013-03-22 06:26:05 <sivu> out of memory?
297 2013-03-22 06:26:24 <Transisto> could be it has 1gb
298 2013-03-22 06:27:02 <gmaxwell> thats not enough memory for running many connections.
299 2013-03-22 06:27:12 <gmaxwell> future versions of bitcoind may do better.
300 2013-03-22 06:27:37 <sivu> i wonder how much memory does blockchain use for keeping 1400+ connections
301 2013-03-22 06:27:43 <Transisto> should I bother to lower the connectoin# ?
302 2013-03-22 06:28:43 <gmaxwell> Transisto: you should if you're running out of memory.
303 2013-03-22 06:31:46 <Transisto> I don't see how connection to ram relate much unless it's caching every incoming tx from each node separately
304 2013-03-22 06:33:10 <gmaxwell> Transisto: the client maintains buffers for each peer so that messages in and out which block at the network can be processed atomically by the node.
305 2013-03-22 06:33:38 <gmaxwell> its not the only way to do things??? or even a good way??? but its the way it does it right now.
306 2013-03-22 06:34:46 <gfawkes> atomic processes are a PITA to craft tho
307 2013-03-22 06:34:53 <Transisto> 940mb is odd comparing to my windows QT client, 383mb (25 connection)
308 2013-03-22 06:35:31 <Transisto> Thanks /to bed/ zzz
309 2013-03-22 07:19:58 <JyZyXEL> can WinXP be used to gpumine?
310 2013-03-22 07:23:53 <warren> JyZyXEL: you might get better answers from #bitcoin-mining
311 2013-03-22 07:24:05 <JyZyXEL> thank you
312 2013-03-22 07:48:35 <_pr> bitcoin piece just on radio 4 in UK
313 2013-03-22 07:49:19 <_pr> that'll be BBC Radio 4
314 2013-03-22 07:50:56 <TD[away]> nice
315 2013-03-22 07:50:59 <TD> what did they say?
316 2013-03-22 07:51:34 <_pr> jokey, novelty style with some investment guy stating he thought it was a fad
317 2013-03-22 07:51:56 <TD> that's fine
318 2013-03-22 07:53:38 <_pr> had a small amount of info in it but not good quality, mentioned 'inflation-proof' and requiring trust in the bitcoin community
319 2013-03-22 07:54:13 <TD> as it happens, i am just about to start a new forum thread about building a press center for the new website
320 2013-03-22 07:54:30 <TD> i think once journalists have a section of the website just for them, with all the resources they need to write a great story, coverage will improve a lot
321 2013-03-22 07:54:48 <_pr> man, wish they'd got gavin on!
322 2013-03-22 07:55:19 <_pr> TD: agreed
323 2013-03-22 07:55:39 <RoboTeddy> TD: excellent idea
324 2013-03-22 07:55:48 <TD> wanna work on it guys? :)
325 2013-03-22 07:56:24 <_pr> sure
326 2013-03-22 07:56:49 <TD> i'll paste a link to the thread here in a few minutes, once i create it
327 2013-03-22 07:58:39 <RoboTeddy> I think the first few sections of http://evoorhees.blogspot.com/ give a better non-technical explanation than I've seen elsewhere
328 2013-03-22 08:10:38 <TD> https://bitcointalk.org/index.php?topic=156364.0
329 2013-03-22 08:15:58 <peawormsworth> in bt-qt I have a directory: ~/.bitcoin/telnet3 in this there is an additional wallet.dat. What is this second wallet for? And is it something I should secure?
330 2013-03-22 08:16:19 <TD> testnet3 is the test network
331 2013-03-22 08:16:29 <TD> unless you received test coins, it isn't important
332 2013-03-22 08:16:30 <peawormsworth> ah yes. i ran -testnet before. thx.
333 2013-03-22 08:16:48 <TD> it doesn't need to be secure, but if you have test coins send them back to the faucet please. don't just delete the wallet
334 2013-03-22 08:16:59 <peawormsworth> will do.
335 2013-03-22 08:18:18 <RoboTeddy> someone should make an open source satoshi dice that donates all the house winnings to charity automatically
336 2013-03-22 08:19:19 <RoboTeddy> I'd do it but I live in the wrong country
337 2013-03-22 08:19:33 <peawormsworth> maybe an evil version that deletes the private keys to hide/destroys the coins.
338 2013-03-22 08:35:54 <[Tycho]> I wonder why no one just asked me directly if they were waiting for me :) https://bitcointalk.org/index.php?topic=154521.0
339 2013-03-22 08:38:03 <TradeFortress> RoboTeddy, you'd need a good bankroll
340 2013-03-22 08:38:06 <TradeFortress> hmm maybe p2p gambling
341 2013-03-22 08:38:16 <RoboTeddy> TradeFortress: well, you could start with small bet limits
342 2013-03-22 08:38:30 <RoboTeddy> TradeFortress: it's possible to dynamically modify the bet limits in order to keep a constant (low) probability of going bust
343 2013-03-22 08:39:25 <RoboTeddy> TradeFortress: I bet if there were a site where the house winnings went to charity, people would use it instead of SD :D
344 2013-03-22 08:39:41 <RoboTeddy> getting OT here though, so I'm going to stfu
345 2013-03-22 08:39:51 <TradeFortress> eh, i'm a S.DICE shareholder so I certainlly won't do that :P
346 2013-03-22 08:47:21 <ryan-c> I just tripped on this https://bitcointalk.org/index.php?topic=9680.0
347 2013-03-22 08:47:34 <ryan-c> thought my code was broken :(
348 2013-03-22 09:29:22 <[Tycho]> What is "UTXO" ?
349 2013-03-22 09:35:57 <jgarzik> [Tycho]: unspent transaction output
350 2013-03-22 09:36:13 <jgarzik> [Tycho]: or in other words, a bitcoin
351 2013-03-22 09:40:38 <[Tycho]> Thanks.
352 2013-03-22 09:42:51 <ThomasV> how does the reference client compute its fee?
353 2013-03-22 09:43:48 <ThomasV> oh, sorry, found it
354 2013-03-22 09:50:20 <Allaun> i have a minor bug, should i post it here?
355 2013-03-22 09:50:58 <sipa> github.com/bitcoin/bitcoin/issues
356 2013-03-22 09:52:11 <Allaun> ok, thank you