1 2013-05-09 00:00:10 <gmaxwell> or better, http://bitcoin.sipa.be/pruning-size.png
2 2013-05-09 00:00:19 <ezdiy> UTXO is irrelevant because utxo set size is not accounted for in fee schedule (yet, hopefully it will be in future)
3 2013-05-09 00:01:16 <cjd> Rothgar: I read http://bitcoin.sipa.be/pruning-size.png to mean 100MB of UTXO set
4 2013-05-09 00:01:20 <gmaxwell> Rothgar: the interesting thing in this graph is showing how pruning dramatically reduces the data set size and growth, and but in later times very tiny outputs undermining that improvement http://bitcoin.sipa.be/pruning-size.png
5 2013-05-09 00:01:29 Luke-Jr has joined
6 2013-05-09 00:01:38 <gmaxwell> cjd: yea thats a few months old, its grown rapidly. :(
7 2013-05-09 00:02:08 <Rothgar> gmaxwell: Thank you!!!
8 2013-05-09 00:02:13 <gmaxwell> the coins database is 210 mb on disk now. (I assume the chart is the seralized size but not the on disk size)
9 2013-05-09 00:02:18 caedes has joined
10 2013-05-09 00:02:18 caedes has quit (Changing host)
11 2013-05-09 00:02:18 caedes has joined
12 2013-05-09 00:02:34 <cjd> 210 sounds like a good number
13 2013-05-09 00:02:46 <ezdiy> gmaxwell: huh, interesting. is there proof that same wallets are responsible for mergin and then re-splitting the coins?
14 2013-05-09 00:02:49 <cjd> really, in a news article people just need an order-of-magnitude idea
15 2013-05-09 00:03:29 <Rothgar> almost 2 orders of magnitude.
16 2013-05-09 00:03:30 <phantomcircuit> 21 * 10 ^ 14 satoshis, assuming the most ridiculous scenario possible with 1 satoshi per txout and 400 bytes per transaction, that's only 750 petabytes
17 2013-05-09 00:03:43 <gmaxwell> ezdiy: the utxo set grows because people split coins and then never (or not quickly at least) remerge them.
18 2013-05-09 00:03:57 <gmaxwell> phantomcircuit: 400 bytes is way too big.
19 2013-05-09 00:04:04 <phantomcircuit> gmaxwell, ok 240 bytes
20 2013-05-09 00:04:09 <ezdiy> gmaxwell: i suspect thats because there is no incentive to do so
21 2013-05-09 00:04:09 <phantomcircuit> 350 PB
22 2013-05-09 00:04:09 <gmaxwell> phantomcircuit: the utxo set would be about 45 PB if split all the way with the current encoding.
23 2013-05-09 00:04:11 <phantomcircuit> whatever
24 2013-05-09 00:04:21 <gmaxwell> ezdiy: worse, it actually costs money to do it.
25 2013-05-09 00:04:26 <ezdiy> gmaxwell: yup
26 2013-05-09 00:04:38 <gmaxwell> ezdiy: e.g. if you have a 1e-8 output, it will cost you way more than that to use it in a transaction.
27 2013-05-09 00:04:46 richcollins has quit (Quit: richcollins)
28 2013-05-09 00:04:47 <phantomcircuit> log(21 * 10 ^ 14) = 15
29 2013-05-09 00:04:53 <cjd> ahh but forgetting about them is free
30 2013-05-09 00:05:06 <phantomcircuit> so assuming this ridiculous extreme happens and everybody has ssds
31 2013-05-09 00:05:09 <gmaxwell> In my mind that one fact is about 80% of the reason to make the output value change.
32 2013-05-09 00:05:17 <phantomcircuit> and a few hundred petabytes lying around
33 2013-05-09 00:05:19 <cjd> You're welcome to spend that satoshi as long as someone else mines it
34 2013-05-09 00:05:21 <phantomcircuit> it wont much matter
35 2013-05-09 00:05:37 <ezdiy> gmaxwell: which is why i'd like to see txfee = MAGIC * tx_kb * txout_n / txin_n eventually implemented
36 2013-05-09 00:05:43 <gmaxwell> cjd: huh?
37 2013-05-09 00:05:43 <ezdiy> or hear why its not possible
38 2013-05-09 00:05:46 <gmaxwell> ezdiy: irrational.
39 2013-05-09 00:06:06 <gmaxwell> ezdiy: I mean for miners to apply that on a standing basis, in general.
40 2013-05-09 00:06:07 <phantomcircuit> gmaxwell, something between a full node and an SPV node, which forgets about dust
41 2013-05-09 00:06:07 <ezdiy> gmaxwell: _n as number of inputs, not the sum of value
42 2013-05-09 00:06:26 <cjd> gmaxwell: I knew you'd love that idea... Just throwing out old dust and dropping any tx that pays it, forcing the work onto someone with a more compliant node
43 2013-05-09 00:06:46 <cjd> *spends
44 2013-05-09 00:06:47 <phantomcircuit> jgarzik, lol
45 2013-05-09 00:06:55 <ezdiy> gmaxwell: you're contradicting yourself, miners are super irrational now (if you mean greed by rationality)
46 2013-05-09 00:06:55 <jgarzik> ?
47 2013-05-09 00:06:57 <gmaxwell> cjd: no can do. otherwise I just start making blocks that inflate the currency by spending dust that never existed.
48 2013-05-09 00:06:58 <phantomcircuit> jgarzik, they are pretty historical
49 2013-05-09 00:06:59 <phantomcircuit> er
50 2013-05-09 00:07:02 <phantomcircuit> hysterical
51 2013-05-09 00:07:45 <cjd> gmaxwell: nah, I'll keep track of the total dust which I vacuumed up so you can mine a block to steal it but you can't inflate anymore than actually existed :)
52 2013-05-09 00:07:59 <gmaxwell> ezdiy: I'm not, let them put their pro-bitcoin chairity in supporting free transactions. For that you'd want your formula to be for _priority_ not fee.
53 2013-05-09 00:08:12 <sipa> gmaxwell: i should update it, indeed
54 2013-05-09 00:08:14 <ezdiy> gmaxwell: agreed
55 2013-05-09 00:08:21 <gmaxwell> cjd: stealing lost coin is still a kind of inflation.
56 2013-05-09 00:08:30 <ezdiy> gmaxwell: so, whats the stopper again?
57 2013-05-09 00:08:46 <gmaxwell> ezdiy: yes I tried proposing that about 6 months ago and basically got fed up with the issue when everyone wanted to shed paint like nuts over it. Knock yourself out. :P
58 2013-05-09 00:08:49 <ezdiy> it would be nice if people would be able to spend their 0.000001s
59 2013-05-09 00:08:50 <Rothgar> Thank you, I'm going to rewrite the fee section to make sure it is correct and then I'm going to mention how the 5430 satoshi number can be changed as market conditions make that valuable. I'm also going to mention how the UtxO set can remove the need to store over 6GB on hard disk.
60 2013-05-09 00:09:04 <ezdiy> gmaxwell: :(
61 2013-05-09 00:09:37 <cjd> gmaxwell: sort of but it doesn't violate the basic premise of bitcoin (too much) there's still only 21 million when you're done
62 2013-05-09 00:10:02 <gmaxwell> Rothgar: you might even include http://bitcoin.sipa.be/pruning-size.png and you can say that but a proliferation of very tiny outputs that can't be spent (either because they are unspendable "data storage" outputs, or because they're just not valuable enough) undermine that improvement.
63 2013-05-09 00:10:09 Ogig has quit (Remote host closed the connection)
64 2013-05-09 00:10:14 Anduck has quit (Ping timeout: 255 seconds)
65 2013-05-09 00:10:24 <cjd> I think it's hilarious because it's gurilla demerrage and noone can really stop it
66 2013-05-09 00:10:45 <gmaxwell> cjd: uh part of the premise is that people can't steal your coins.
67 2013-05-09 00:11:06 <gmaxwell> cjd: what do you mean noone can stop it? Bitcoin nodes do full validation, and so long as they do your blocks will go nowhere.
68 2013-05-09 00:11:07 <cjd> damn the fine print
69 2013-05-09 00:11:10 nsillik has quit (Quit: nsillik)
70 2013-05-09 00:11:28 <Rothgar> gmaxwell: I have these two sentences: The issues with these one satoshi payments exaggerate the blockchain bloat problem mentioned above. It even reduces the effectiveness of the unspent transaction output solution to the bloat problem.
71 2013-05-09 00:11:29 <cjd> oh, I would never mine a block which had any dust theft in it
72 2013-05-09 00:11:34 <cjd> I'd drop those transactions
73 2013-05-09 00:11:45 <cjd> but I would allow someone else to mine a dust theft block
74 2013-05-09 00:12:19 <ezdiy> lol, dust theft?
75 2013-05-09 00:12:41 <gmaxwell> Rothgar: OK. One of the posts on bitcoin talk made a nice point like "It's _INSANE_ that bitcoin allows you to make a payment that costs more to redeem than its worth"
76 2013-05-09 00:12:46 Luke-Jr has quit (Read error: Connection reset by peer)
77 2013-05-09 00:13:09 <gmaxwell> lemme see if I can find that post.
78 2013-05-09 00:13:12 <cjd> ezdiy: based on a hypothetical client which classifies tiny/old transactions as dust and forgets about them, keeping only a tally of the total amount of BTC worth of dust which has been forgotten
79 2013-05-09 00:13:27 <ezdiy> cjd: oh that
80 2013-05-09 00:13:38 <ezdiy> cjd: well, every 0.8.2 client kinda does that by default :)
81 2013-05-09 00:13:43 <cjd> no
82 2013-05-09 00:13:47 <gmaxwell> ezdiy: NO. :(
83 2013-05-09 00:13:51 <sipa> you cannot forget dust
84 2013-05-09 00:13:52 <cjd> hehehe
85 2013-05-09 00:13:54 * gmaxwell gets out the robotic stabbing machine
86 2013-05-09 00:13:58 <sipa> doing so would require a (soft) fork
87 2013-05-09 00:14:04 * Rothgar runs
88 2013-05-09 00:14:05 <ezdiy> gmaxwell: chill out. i know it wont just let the tx through
89 2013-05-09 00:14:11 <cjd> I knew you would love that ides :D
90 2013-05-09 00:14:13 <ezdiy> it wont steal or block the dust or anything
91 2013-05-09 00:14:22 <gmaxwell> ezdiy: there is a very important difference between _spending_ a dust output and _creating_ one.
92 2013-05-09 00:14:23 <ezdiy> but from pow of network user using really old client ...
93 2013-05-09 00:14:28 nizeguy has quit (Ping timeout: 260 seconds)
94 2013-05-09 00:14:37 <MC1984> my mums got a robotic stabbing machine
95 2013-05-09 00:14:37 <gmaxwell> ezdiy: the 0.8.2 change strongly discourages creating them, but does nothing relative to spending them.
96 2013-05-09 00:14:49 <sipa> and there is also a very important difference between relay policy and block validity rules
97 2013-05-09 00:14:49 <MC1984> she fixed my cloths with it when younger
98 2013-05-09 00:14:57 <cjd> I shold be clear, what I said above is MY INSANE IDEA ONLY, 0.8.2 does NOTHING LIKE THAT
99 2013-05-09 00:15:02 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|sipa: what about just 'not checking' blocks that mine very old dust... so if sombody theifed the dust, the chain would not get blocked.
100 2013-05-09 00:15:03 <gmaxwell> cjd: is talking about the spending side, which has ugly implications like people being able to argue that it constitutes theft.
101 2013-05-09 00:15:06 MarceFe has joined
102 2013-05-09 00:15:08 <MarceFe> Hello! Anyone knows some good repository for stratum server? I tried https://github.com/slush0/stratum.git but don't work!
103 2013-05-09 00:15:16 <sipa> da2ce7|2: you cannot compute block fees anymore then
104 2013-05-09 00:15:30 <cjd> gmaxwell: I know, but they won't get anywhere because I'll just tell them to get their tx mined somewhere else :]
105 2013-05-09 00:15:43 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|oh... more coins in than exits problem.
106 2013-05-09 00:15:57 <sipa> da2ce7|2: no, fees are defined as inputs-outputs
107 2013-05-09 00:16:11 <sipa> da2ce7|2: what you propose means some transactions would get unknown inputs
108 2013-05-09 00:16:25 quaz0r has quit (Ping timeout: 245 seconds)
109 2013-05-09 00:17:23 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|then what about making a bitcoin-rule that rejects tansactions say less than < 0.0000001 that are older than 4 years. So all the nodes can force-prune those outputs.
110 2013-05-09 00:17:29 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|*input
111 2013-05-09 00:17:36 <sipa> da2ce7|2: that would be considered theft :)
112 2013-05-09 00:17:41 <sipa> (by some)
113 2013-05-09 00:18:06 <cjd> da2ce7|2: that's what I was saying, except instead of making a network rule and handling the drama, I'll just implement it in *my* client and be like "go get mined elsewhere"
114 2013-05-09 00:18:21 Luke-Jr has joined
115 2013-05-09 00:18:29 <gmaxwell> cjd: but that isn't how bitcoin works. Say that happens. Now there is a block on the network which you cannot validate.
116 2013-05-09 00:19:02 <gmaxwell> cjd: so now you either believe random stuff in blocks (which potentially undermines the incentive structure behind making honest blocks) or you reject the block and the chain forks.
117 2013-05-09 00:19:02 <cjd> I give it the I can't prove there's anything wrong with this stamp of approval :D
118 2013-05-09 00:19:07 <MarceFe> Hello! Anyone knows some good repository for stratum server? I tried https://github.com/slush0/stratum.git but don't work!
119 2013-05-09 00:19:16 <phantomcircuit> gmaxwell, either nobody can validate it, or you can ask the network for the transaction and someone still has it
120 2013-05-09 00:19:24 <phantomcircuit> (like say the person who wants to spend the 1 satoshi)
121 2013-05-09 00:19:26 <sipa> cjd: and even worse, you can't know whether it was because you overpruned something, or because the transaction in the block genuinely spent something that didn exist
122 2013-05-09 00:19:34 <gmaxwell> phantomcircuit: no, you can't randomly access transactions.
123 2013-05-09 00:20:02 andyh2 has quit (Quit: Leaving...)
124 2013-05-09 00:20:04 <sipa> cjd: so basically you reduce your ability to check for valid inputs to 0
125 2013-05-09 00:20:09 <cjd> well, we could always add a "prove that transaction exists" RPC but that would be annoying
126 2013-05-09 00:20:27 nospinzy has quit (Ping timeout: 264 seconds)
127 2013-05-09 00:20:28 <gmaxwell> cjd: basically your mode of operation fails the catagorical imperativeâ it only works so long as its just you or just you plus a few others.
128 2013-05-09 00:20:30 <sipa> cjd: that'd be incompatible with pruning in the first place
129 2013-05-09 00:20:38 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|cjd: but how do you prove that it hasn't been already spent?
130 2013-05-09 00:20:46 <gmaxwell> cjd: prove that it exists is hard in any case, e.g. I prove that it existed but really it was already spent.
131 2013-05-09 00:20:53 Irencus has quit ()
132 2013-05-09 00:21:02 <phantomcircuit> gmaxwell, well im guessing that removing spent outputs would be a much larger benefit than removing dust for the time being
133 2013-05-09 00:21:03 <gmaxwell> A committed utxo can be used to address that, but not if you're forgetting random stuff.
134 2013-05-09 00:21:18 <sipa> phantomcircuit: that's exactly what happens in the UTXO set
135 2013-05-09 00:21:22 <gmaxwell> phantomcircuit: we already completely remove spent outputs from the utxo set..
136 2013-05-09 00:21:24 <cjd> sipa: I'll keep a uint64_t total_dust and reject any block which spends unverifiable transactions in excess of that number
137 2013-05-09 00:21:28 <gmaxwell> (because .. thats what it is!)
138 2013-05-09 00:21:31 anarchy5 has joined
139 2013-05-09 00:21:32 <phantomcircuit> i actually meant entirely
140 2013-05-09 00:21:46 <sipa> phantomcircuit: only thing missing is actually removing old block files
141 2013-05-09 00:21:52 <sipa> phantomcircuit: it's trivial implementation-wise
142 2013-05-09 00:22:20 <sipa> except it would damage the network significantly if there is no better mechanism in place to find old blocks via the P2P network
143 2013-05-09 00:22:30 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|is there some lock-rule that we could put on outputs < 0.000000001 that says 'after 4years anyone can spend this output' and include that by default on small outputs?
144 2013-05-09 00:22:50 <phantomcircuit> da2ce7, no
145 2013-05-09 00:23:01 <cjd> da2ce7|2: that kind of protocol rule == endless bleeding on the forum
146 2013-05-09 00:23:04 <cjd> never work
147 2013-05-09 00:23:04 <sipa> da2ce7|2: why would anyone add such a lock-rule (assuming it was possible) ?
148 2013-05-09 00:23:15 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|becasue of network health.
149 2013-05-09 00:23:26 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|I would add it to my tx.
150 2013-05-09 00:23:29 <sipa> how about we make a deal never to do double spends
151 2013-05-09 00:23:38 <sipa> because of network health?
152 2013-05-09 00:23:45 <cjd> my idea works because I can implement it alone and for large transactions, I'm a fully validating node but for dust, I'm effectively SPV
153 2013-05-09 00:23:48 <gmaxwell> Rothgar: you might want to take a look at my forum posts over the past two days, almost all of them have been on this subject.
154 2013-05-09 00:24:10 <Rothgar> gmaxwell: I will thanks
155 2013-05-09 00:24:14 <gmaxwell> Rothgar: both because I make some interesting points, but also because it samples some of the concerns people have had and at least one take on them.
156 2013-05-09 00:24:22 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|er... it could be just multi-sig type lock-time, that anyone can spend.
157 2013-05-09 00:24:33 mollison has quit (Remote host closed the connection)
158 2013-05-09 00:25:13 <gmaxwell> Rothgar: jgarzik also made a number of posts on the subject.
159 2013-05-09 00:25:56 <Rothgar> OK: and you're both developers. Great.
160 2013-05-09 00:26:18 * da2ce7 2!~kvirc@opentransactions/dev/da2ce7|has had enough of stupid ideas... will go back to being lurking to save sounding stupid. lol
161 2013-05-09 00:26:54 <cjd> It is better to keep your mouth shut and be thought a fool than to open it and remove all doubt
162 2013-05-09 00:26:57 <cjd> err
163 2013-05-09 00:26:58 <cjd> oops
164 2013-05-09 00:27:05 <jgm> Someone should code up a new piece in bitcoin-qt that takes all of the dust inputs and sends them to a single address. Then a few weeks from now (when most reachable dust will have been gathered) randomly return it in one go to one of those who sent dust in.
165 2013-05-09 00:27:49 * jgm wonders how much dust really is reachable
166 2013-05-09 00:27:57 <cjd> meh protocol rules -> endless discussion
167 2013-05-09 00:28:07 <cjd> cjdcoin/dustbuster -> endless hilarity
168 2013-05-09 00:28:13 <Rothgar> da2ce7|2: If/when ECDSA is broken that many of those satoshis will be spendable.
169 2013-05-09 00:28:32 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|Rothgar: SHA256???
170 2013-05-09 00:28:45 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|only the ones we know the public key for,
171 2013-05-09 00:29:03 <phantomcircuit> Rothgar, lol that's actually not true, you would need to break ECDSA SHA256 and RIPEMD160
172 2013-05-09 00:29:10 <gmaxwell> Rothgar: no. the data ones are almost certantly unspendable even given a complete ecdsa break.
173 2013-05-09 00:29:43 <gmaxwell> and "breaks" to hash functions are really really not likely to be of a form that would permit spending those things.
174 2013-05-09 00:29:45 Edward_Black has quit (Ping timeout: 264 seconds)
175 2013-05-09 00:29:53 icellan has quit (Quit: icellan)
176 2013-05-09 00:30:01 quaz0r has joined
177 2013-05-09 00:30:03 <Rothgar> phantomcircuit: Even if there has already been a spend from that address and the public key is exposed?
178 2013-05-09 00:30:07 owowo has quit (Remote host closed the connection)
179 2013-05-09 00:30:19 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|or even arbitrary sized quantum computing, that just reduces SHA256 >~ SHA128...
180 2013-05-09 00:30:22 <phantomcircuit> Rothgar, effectively none of them have that property
181 2013-05-09 00:30:40 owowo has joined
182 2013-05-09 00:30:48 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|(good for mining, but not much else)
183 2013-05-09 00:30:50 <Rothgar> most the satoshi dice players make addresses with that property.
184 2013-05-09 00:30:57 <gmaxwell> Rothgar: by definition the data storage ones do not.
185 2013-05-09 00:31:20 <gmaxwell> Rothgar: SD isn't even an unspendability concern here. They increased their txout size to a level which is worth redeeming.
186 2013-05-09 00:31:26 eqolo has joined
187 2013-05-09 00:31:38 <gmaxwell> (after months of being haphazardly filtered by people on their own inititivate)
188 2013-05-09 00:32:07 <jgm> cjd: simplest dustbuster is to just prune any UTXOs that are less than a certain amount. Not that it would make anyone particularly happy, but would be a good precedent to set IMO
189 2013-05-09 00:32:36 <gmaxwell> jgm: people would rightfully call that theft.
190 2013-05-09 00:32:40 <cjd> jgm: that's exactly my idea, actually 1/amount * age
191 2013-05-09 00:32:48 <Rothgar> but the satoshi dice addresses often have many spends from them which make them vulnerable if ECDSA was broken and nothing else.
192 2013-05-09 00:32:54 <sipa> jgm: that completely removes your ability to validate inputs in transactions...
193 2013-05-09 00:33:08 <cjd> so you drop the transactions on the floor :D
194 2013-05-09 00:33:20 <gmaxwell> Rothgar: the dice addresses aren't the ones recieving dust.
195 2013-05-09 00:33:22 Edward_Black has joined
196 2013-05-09 00:34:18 <Rothgar> but the people who play receive dust and often have many spends from those addresses.
197 2013-05-09 00:34:32 <gmaxwell> Rothgar: and in any case, SD is more or less orthorgonal to this discussion. Their outputs are (now, finally) almost all above the threshold (only they produce some that are slightly below, but evhorees didn't seem to have a problem with the tiny increase required to fit)
198 2013-05-09 00:34:37 <Rothgar> or at least they did.
199 2013-05-09 00:35:06 <Rothgar> yes, thanks
200 2013-05-09 00:35:14 <jgm> gmaxwell: there are over 1 million UTXOs in satoshis (<1uBTC), totalling just over 0.1BTC. And it's not like anyone is gaining from it (apart from deflation, I suppose)
201 2013-05-09 00:35:36 volante has quit (Quit: Leaving)
202 2013-05-09 00:35:43 <jgm> sipa: okay so that I didn't know. More of a problem, definitely
203 2013-05-09 00:36:02 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|I would love to make a *coin that doubles the generation rate every 4yr, and halves the minium input that is spendable every 4 years.... inflatacoin. For use with small-short-term transactions.
204 2013-05-09 00:36:08 <gmaxwell> jgm: yes, it's a mess and we should have had these rules a yea ago. But it's also only ... whatever, like 100mbytes of data now or whatever it is. It's just important that it stops getting worse.
205 2013-05-09 00:36:09 <jchp> how evil it would it be to make a p2pool fork that automatically accepted double spends on SD (where only a single txout is changed)?
206 2013-05-09 00:36:38 <gmaxwell> jchp: p2pool doesn't do anything with transactions the partnerned bitcoind does.
207 2013-05-09 00:36:40 <da2ce7> 2!~kvirc@opentransactions/dev/da2ce7|*doubbles the mini.
208 2013-05-09 00:36:45 <jchp> oh right, the bitcoind client then
209 2013-05-09 00:36:54 <phantomcircuit> re-roll
210 2013-05-09 00:37:01 <phantomcircuit> yeah
211 2013-05-09 00:37:07 <phantomcircuit> if only someone did that already
212 2013-05-09 00:37:09 [\\\] has quit (Ping timeout: 256 seconds)
213 2013-05-09 00:37:21 <cjd> jchp: wouldn't work (unless my idea became the defacto mining client)
214 2013-05-09 00:37:22 <gmaxwell> jchp: I'm sure if you ask around you can find patches that do exactly that, but I think people are less mad at SD now because they're not quite as network horiffic as they used to be.
215 2013-05-09 00:37:38 <gmaxwell> cjd: I assumed he meant unconfirmed transactions.
216 2013-05-09 00:37:44 <cjd> oh ic
217 2013-05-09 00:37:49 <cjd> hehehe
218 2013-05-09 00:37:55 CodesInChaos has quit (Ping timeout: 256 seconds)
219 2013-05-09 00:38:13 <gmaxwell> SD has switched to requiring confirmation on high value txn in any case, due to the aformentioned blocking and stuff enabling double spends.
220 2013-05-09 00:38:15 macboz has joined
221 2013-05-09 00:39:29 <eqolo> I have a small amount of Liberty Reserve as I would like to exchange to Bitcoin. Is this possible? / msg me
222 2013-05-09 00:39:46 <phantomcircuit> eqolo, #bitcoin-otc
223 2013-05-09 00:39:48 <phantomcircuit> kthx
224 2013-05-09 00:39:48 <gmaxwell> eqolo: #bitcoin-otc
225 2013-05-09 00:39:59 <phantomcircuit> or that
226 2013-05-09 00:39:59 eqolo has joined
227 2013-05-09 00:40:02 <eqolo> ok sorry
228 2013-05-09 00:40:03 <gmaxwell> (and, fwiw, people that solicit PM trades are often scammers)
229 2013-05-09 00:40:24 <eqolo> I could send you Liberty Reserve first np
230 2013-05-09 00:40:29 <eqolo> if you are good user
231 2013-05-09 00:41:07 <jchp> yeah, but if they were attacked frequently enough, they'd be forced to only send after a confirmation, which kills a lot of their appeal (opening the door to competitors that do deposit-based betting)
232 2013-05-09 00:41:25 <jgarzik> eqolo: not on this channel.
233 2013-05-09 00:41:51 <eqolo> jgarzik: ok
234 2013-05-09 00:41:53 [\\\] has joined
235 2013-05-09 00:42:03 <cjd> jchp: or plunk a few nodes around the network and listen for double spends...
236 2013-05-09 00:42:30 <gmaxwell> cjd: lolisten? ... nothing requires a conflicting transaction to be visible before it shows up in a block.
237 2013-05-09 00:42:44 unknown45682 has joined
238 2013-05-09 00:42:57 <gmaxwell> the MSFT red ballons people argue that miners would someday greedily not relay high fee txn they get givenâ¦
239 2013-05-09 00:43:14 <jchp> if it's directly sent to SD-hostile nodes that don't rebroadcast....
240 2013-05-09 00:43:14 FredEE has quit (Changing host)
241 2013-05-09 00:43:14 FredEE has joined
242 2013-05-09 00:43:36 <gmaxwell> and because of differences in relay policy its not terribly hard to construct a conflicting transaction that some pools will take but most nodes wont.
243 2013-05-09 00:43:38 <cjd> sure but then you plunk a node in p2pool
244 2013-05-09 00:43:44 ngc0202 has joined
245 2013-05-09 00:43:56 <cjd> oh indeed
246 2013-05-09 00:44:04 <cjd> IsStandard() strikes again
247 2013-05-09 00:44:10 <gmaxwell> not just isstandard.
248 2013-05-09 00:44:36 <gmaxwell> I mean, for example luke's weird tonal fee policy. nlocktime could also be potentially abused for this in the future.
249 2013-05-09 00:45:11 <cjd> tonal fee policy?
250 2013-05-09 00:45:11 mollison has joined
251 2013-05-09 00:45:22 <gmaxwell> Unconfirmed transactions are flat out unsafeâ now- many applications don't really require safty... but don't trick yourself, stunts don't make the unsafe safe. :)
252 2013-05-09 00:45:29 <jchp> cjd: 0.00008192/KB ithink
253 2013-05-09 00:45:35 <cjd> :D
254 2013-05-09 00:45:51 <MarceFe> Hey guys! Some documentation for stratum server? I need to know the dependencies, in my develop server run correctly, but in my production server have error :S
255 2013-05-09 00:45:56 <cjd> I see, he will take lower fee tx than others, makes sense
256 2013-05-09 00:46:21 <gmaxwell> cjd: lower in some cases.. higher in othersâ it doesn't take some txn that many other nodes would accept as free.
257 2013-05-09 00:46:22 <MarceFe> Hey guys! Some documentation for stratum server? I need to know the dependencies, in my develop server run correctly, but in my production server have error :S But no log or indices in error mode :S
258 2013-05-09 00:47:31 <cjd> well if his stuff relays it then it should be ok to make lots of connections and hope you hear about the conflicting tx
259 2013-05-09 00:47:55 <cjd> that breaks down when you have evil miners who accept but don't relay
260 2013-05-09 00:48:20 dhill has joined
261 2013-05-09 00:48:28 <jchp> well if you accept incoming connections, blockchain.info will always try to connect to you and will relay most transactions :-P
262 2013-05-09 00:48:50 <cjd> clearly they don't use ipv6
263 2013-05-09 00:49:01 <cjd> I've been open on ipv6 for a while now and no incoming connections
264 2013-05-09 00:49:10 <cjd> except for that one cjdns node which is now known
265 2013-05-09 00:50:38 richcollins has joined
266 2013-05-09 00:52:23 RBecker has quit (Excess Flood)
267 2013-05-09 00:53:08 RBecker has joined
268 2013-05-09 00:53:29 santoscork has quit (Quit: Hibernation Time â¦)
269 2013-05-09 00:53:48 <sipa> MarceFe: repeating the same question over and over again will not help; likely nobody here now knows
270 2013-05-09 00:54:01 eqolo has quit ()
271 2013-05-09 00:55:17 <gmaxwell> cjd: are you getting returned by sipa's dns seed?
272 2013-05-09 00:55:53 <cjd> host?
273 2013-05-09 00:56:06 <jchp> dig AAAA seed.bitcoin.sipa.be
274 2013-05-09 00:57:08 <cjd> nope
275 2013-05-09 00:57:09 AndChat64721 has quit (Read error: Connection reset by peer)
276 2013-05-09 00:57:23 <cjd> inet addr:64.15.65.115 P-t-P:64.15.65.115 Mask:255.255.255.255
277 2013-05-09 00:57:23 <cjd> inet6 addr: fc88:dfd0:89d4:abfe:de2:a17a:6ed5:6fb1/8 Scope:Global
278 2013-05-09 00:57:23 <cjd> inet6 addr: 2001:470:b040:101:fc88:dfd0:89d4:abfe/0 Scope:Global
279 2013-05-09 00:57:28 <cjd> my tun device
280 2013-05-09 00:57:32 <sipa> i have 282 IPv6 hosts that are being served
281 2013-05-09 00:57:59 <sipa> nope, not in there
282 2013-05-09 00:58:09 <cjd> hmm
283 2013-05-09 00:58:52 <cjd> user@ubnta8:~/wrk/cjdns-pvt$ cat ~/.bitcoin/debug.log | grep 'AddLocal'
284 2013-05-09 00:58:52 <cjd> AddLocal(64.15.65.115:8333,1)
285 2013-05-09 00:58:52 <cjd> AddLocal([2001:470:b040:101:fc88:dfd0:89d4:abfe]:8333,1)
286 2013-05-09 00:59:47 <cjd> note that ipv4 addr is unused because for ipv4 I need to do manual `route add` and I am not setting it atm
287 2013-05-09 01:02:56 <sipa> perhaps use -externalip explicitly
288 2013-05-09 01:03:19 <sipa> (sets the IP which is broadcast)
289 2013-05-09 01:03:19 <BCB> anyone know about btcwire: The bitcoin wire protocol package from btcd (written in go)??
290 2013-05-09 01:03:25 <MarceFe> sipa nobody learn me :/
291 2013-05-09 01:05:08 <cjd> I hate using flags, if it doesn't "just work" I'm more in to writing a patch ;)
292 2013-05-09 01:05:55 <gmaxwell> BCB: what about it?
293 2013-05-09 01:08:15 <BCB> gmaxwell: do you know anything about it
294 2013-05-09 01:08:28 <BCB> have you heard of it
295 2013-05-09 01:08:31 <BCB> do you know the developer
296 2013-05-09 01:08:33 <BCB> it is legit
297 2013-05-09 01:08:36 <BCB> etc
298 2013-05-09 01:08:59 <gmaxwell> BCB: it's just deserialization code right now.
299 2013-05-09 01:09:23 <gmaxwell> BCB: there is a thread about it here? https://bitcointalk.org/index.php?topic=192880.msg2080098
300 2013-05-09 01:09:28 <BCB> cool
301 2013-05-09 01:09:29 <BCB> thx
302 2013-05-09 01:09:30 da2ce7 has quit (2!~kvirc@opentransactions/dev/da2ce7|Ping timeout: 276 seconds)
303 2013-05-09 01:10:01 btcls has joined
304 2013-05-09 01:10:44 <btcls> hi all - [lurking to learn]
305 2013-05-09 01:11:19 <btcls> #bitcoin
306 2013-05-09 01:13:07 jaequery has quit (Quit: Computer has gone to sleep.)
307 2013-05-09 01:14:06 <Corndawg> dev question: How hard would it be to implement a "recipiant paid fee"?
308 2013-05-09 01:14:09 porquilho has quit ()
309 2013-05-09 01:14:48 <Corndawg> maybe a non issue but like if customers wanted to buy something in BTC and the merchant was happy to pay the tx fee knowing it would still be far less than CC fee
310 2013-05-09 01:15:00 grau has joined
311 2013-05-09 01:15:17 <Corndawg> and in future (after bitcoins stopped being mined) might not be such an insignificant fee
312 2013-05-09 01:15:36 Casimir1904 has joined
313 2013-05-09 01:15:49 <Corndawg> assuming miners stopped being so altruistic and only mind if fee is (considered by users) exorbinant
314 2013-05-09 01:15:58 <Corndawg> mind = mined
315 2013-05-09 01:16:04 btcls has quit (Quit: Page closed)
316 2013-05-09 01:16:36 <BlueMatt> its being worked on
317 2013-05-09 01:16:52 <Corndawg> oh so is possible, just taking time?
318 2013-05-09 01:17:27 mollison has left ("PART #bitcoin-mining :QUIT :Leaving.")
319 2013-05-09 01:17:29 <BlueMatt> took it a while to get to the top of priority lists
320 2013-05-09 01:17:36 <Corndawg> maybe a BTC wallet address thats a known 1-800 number or somehting liek that
321 2013-05-09 01:17:54 <BlueMatt> wat?
322 2013-05-09 01:18:16 <Corndawg> I dunno, just thinking of how would the system know to charge the recipiant instead of the sender
323 2013-05-09 01:18:25 <Corndawg> maybe an address thats known
324 2013-05-09 01:18:33 <Corndawg> like 1dice = Satoshi Dice addr
325 2013-05-09 01:18:38 <sipa> Corndawg: it doesn't
326 2013-05-09 01:18:39 <Corndawg> something like that
327 2013-05-09 01:18:56 <BlueMatt> no, the recipient just creates a new txn that depends on the previous with the fee they want
328 2013-05-09 01:18:56 <sipa> Corndawg: the idea is just that the recipient is able to add a fee afterwards
329 2013-05-09 01:19:17 <Corndawg> oic... but doesnt that just clutter the system with more tx's?
330 2013-05-09 01:19:17 <BlueMatt> or you do fancy payment protocol giggling
331 2013-05-09 01:19:34 <Corndawg> like a payback tx right?
332 2013-05-09 01:20:03 <BlueMatt> there are no payback tx
333 2013-05-09 01:20:12 <BlueMatt> because "There is no from address" -topic
334 2013-05-09 01:20:30 andyh2 has joined
335 2013-05-09 01:23:31 dvide has quit (Ping timeout: 252 seconds)
336 2013-05-09 01:23:50 <BlueMatt> also, not really...they should be pruneable
337 2013-05-09 01:25:11 <gonffen> Corndawg: in what you're describing, couldn't the vendor just charge less and implore the sender to use a fee?
338 2013-05-09 01:25:53 <Corndawg> sure but he can't enforce the fee right?
339 2013-05-09 01:26:13 <Corndawg> what a vendor wants most I imagine is for 6 onfirms as fast as possible
340 2013-05-09 01:26:22 <Corndawg> to elim fraud and theft
341 2013-05-09 01:26:29 <Corndawg> mimnimize rather
342 2013-05-09 01:26:32 <Corndawg> not elim totally
343 2013-05-09 01:26:51 <Corndawg> so some way to facilitate that other than "hope something" of the user
344 2013-05-09 01:26:58 <Corndawg> rely on users altruism
345 2013-05-09 01:27:33 <Corndawg> other currencyes have 'chargebacks' to deal with arbitration
346 2013-05-09 01:27:42 <Corndawg> BTC sees chargebacks as negative
347 2013-05-09 01:27:51 Haifisch has joined
348 2013-05-09 01:28:29 <Corndawg> fair enough, but the merchant side needs some protection from his risk of letting you walk away with goods that are technically not yet paid for (will be in the system confirms the tx, but mgiht have a chance of doublespend)
349 2013-05-09 01:28:58 <Corndawg> I guess he could raise prices... but wouldn't that put off some users from paying in BTC?
350 2013-05-09 01:29:26 nova907767 has joined
351 2013-05-09 01:30:12 andyh2 has quit (Ping timeout: 276 seconds)
352 2013-05-09 01:32:11 <sipa> gmaxwell: ok, generating new pruning statistics :)
353 2013-05-09 01:32:34 <sipa> a bit more efficiently than before (it basically just did a full gettxoutsetinfo after every block...)
354 2013-05-09 01:32:49 nova90 has quit (Ping timeout: 268 seconds)
355 2013-05-09 01:33:11 <BlueMatt> Corndawg: theoretically a future extension of the payment protocol could allow receivers to add fees to the same txn, but for now its fairly simple to just create a second tx if its a problem
356 2013-05-09 01:33:57 <BlueMatt> (well, aside from it not currently working)
357 2013-05-09 01:35:39 defunctzombie is now known as defunctzombie_zz
358 2013-05-09 01:35:47 <gmaxwell> sipa: getting size stats too?
359 2013-05-09 01:35:52 <gmaxwell> (txout size I mean)
360 2013-05-09 01:36:41 <sipa> yes
361 2013-05-09 01:37:09 Optimo has quit (Ping timeout: 245 seconds)
362 2013-05-09 01:38:59 <wei_> does anyone have an example of using bitcoinj to monitor transactions on arbitrary addresses? or if not bitcoinj, any other client
363 2013-05-09 01:43:42 MobPhone has quit (Read error: Connection reset by peer)
364 2013-05-09 01:44:06 MobPhone has joined
365 2013-05-09 01:46:20 trang has joined
366 2013-05-09 01:47:48 richcollins has quit (Quit: richcollins)
367 2013-05-09 01:48:45 MobiusL has quit (Remote host closed the connection)
368 2013-05-09 01:49:08 SirDefaced has joined
369 2013-05-09 01:49:48 MobiusL has joined
370 2013-05-09 01:52:19 franl has joined
371 2013-05-09 01:52:34 Skav has joined
372 2013-05-09 01:53:05 xire has joined
373 2013-05-09 01:54:53 <zooko> da2ce7: I didn't see anything wrong with your idea that you then called stupid.
374 2013-05-09 01:55:12 MoALTz has joined
375 2013-05-09 01:56:06 <sipa> gmaxwell: interesting... the average utxo size in bytes per unspent output drops from around 70 initially, to 40 now
376 2013-05-09 01:56:09 <sipa> bytes
377 2013-05-09 01:56:34 <sipa> (not unexpected, the first blocks almost only did pay-to-pubkey)
378 2013-05-09 01:56:35 <gmaxwell> reduction in pay to pubkey outputs?
379 2013-05-09 01:56:52 MobPhone has quit (Ping timeout: 268 seconds)
380 2013-05-09 01:56:55 <MoALTz> i've been reading a lot of technical details about bitcoin the last few days, but what does utxo mean?
381 2013-05-09 01:57:02 <sipa> unspent transaction output
382 2013-05-09 01:57:07 <MoALTz> thanks
383 2013-05-09 01:57:28 <sipa> it's sometimes also just called 'coin'
384 2013-05-09 01:57:49 <xire> If the average time until the next block is solved in about 10 minutes, what is the time window within that period in which it is likely that a transaction will be included in the next block?
385 2013-05-09 01:58:09 <xire> Is there a cut-off time of sorts?
386 2013-05-09 01:58:31 <sipa> no
387 2013-05-09 01:58:33 <gmaxwell> xire: blocks are solved continually the average is just an average.
388 2013-05-09 01:58:58 <franl> xire, you can't predict which block will contain a pending transaction.
389 2013-05-09 01:59:05 <gmaxwell> it's like raindrops. You can observe them falling at a specific average rate... but their arrivials are random.
390 2013-05-09 01:59:06 ssams has joined
391 2013-05-09 01:59:45 <xire> That sounds so beautiful!
392 2013-05-09 01:59:56 ssams has left ()
393 2013-05-09 02:00:02 <Diablo-D3> so gmaxwell is teaching chaos theory 101?
394 2013-05-09 02:00:40 <cjd> http://stackoverflow.com/questions/4690526/can-i-run-googles-native-client-outside-of-the-browser <-- interesting
395 2013-05-09 02:01:03 <cjd> make some of the lower level attacks nonsensical
396 2013-05-09 02:01:10 <Diablo-D3> btw, you can
397 2013-05-09 02:01:19 <Diablo-D3> I dont think anyone has written the equivalent of nodejs yet
398 2013-05-09 02:01:23 resinate has quit (Ping timeout: 255 seconds)
399 2013-05-09 02:01:26 <xire> Hm, so it could be possible for a transaction that is broadcast one second before a block is solved to be included in that block? This is assuming the transaction has a good transaction fee
400 2013-05-09 02:01:40 <sipa> xire: that's unlikely, but possible
401 2013-05-09 02:01:46 <sipa> there are finite propagation delays
402 2013-05-09 02:01:55 <gmaxwell> xire: yes, it's perfectly possibleâ not likely because it takes time to get to miners...
403 2013-05-09 02:02:20 <gmaxwell> but if you were to say .. 30 seconds or a minute... more likely.
404 2013-05-09 02:02:46 <sipa> probably as of ~10 seconds or so it's reasonable
405 2013-05-09 02:03:09 <sipa> actually, that would be interesting to measure
406 2013-05-09 02:03:15 <gmaxwell> sipa: well, most pools don't LP on new transactions ... and stratum basically runs until it gets LPed. :(
407 2013-05-09 02:03:28 <xire> That's what I'm trying to do
408 2013-05-09 02:03:31 <sipa> gmaxwell: really? :o
409 2013-05-09 02:03:36 <gmaxwell> so I'm sure miner latency increases things a lot.
410 2013-05-09 02:03:51 <xire> Just give probability of inclusion of a transaction in the next block
411 2013-05-09 02:04:09 <gmaxwell> xire: oh but that has nothing to do with ten minutes.
412 2013-05-09 02:04:43 <gmaxwell> xire: lets say that right now its been 30 minutes since the last block .... we actually expect the next block to happen in........ 10 minutes.
413 2013-05-09 02:05:08 <gmaxwell> more time passing doesn't change the expected time until the next one.
414 2013-05-09 02:06:19 richcollins has joined
415 2013-05-09 02:06:43 <xire> cool
416 2013-05-09 02:06:56 <xire> I think that gives me what I need!
417 2013-05-09 02:08:24 <zooko> Folks: what's the best document for a programmer (me) to learn about the current controversy about dust and the change to fee structure?
418 2013-05-09 02:08:43 <cjd> the patch
419 2013-05-09 02:08:51 <cjd> it's not super low level tricky code
420 2013-05-09 02:09:22 <cjd> I looked at it and I was like "yeah, ok"
421 2013-05-09 02:09:25 <xire> thanks guys
422 2013-05-09 02:09:34 <zooko> cjd: thanks.
423 2013-05-09 02:09:46 <gmaxwell> zooko: as a programmer I don't think it's possible for you to understand the _controversy_. You can go see the pull which has discussion.
424 2013-05-09 02:10:15 seeingidog__ has quit (Quit: Leaving.)
425 2013-05-09 02:10:22 Skav has quit (Read error: Connection reset by peer)
426 2013-05-09 02:10:28 <zooko> ok, thanks.
427 2013-05-09 02:10:30 <cjd> oh right, the controversy.. maybe look here https://github.com/litecoin-project/litecoin/tree/master/src
428 2013-05-09 02:10:33 MobPhone has joined
429 2013-05-09 02:10:33 * cjd runs
430 2013-05-09 02:10:45 <sipa> the controversy is mostly because some people never thought there would be one, and other disagreed :p
431 2013-05-09 02:11:00 <gmaxwell> yea, the pull mostly looks boring.
432 2013-05-09 02:15:41 px has joined
433 2013-05-09 02:16:24 Skav has joined
434 2013-05-09 02:18:30 MobPhone has quit (Ping timeout: 245 seconds)
435 2013-05-09 02:21:24 Michail1_ is now known as Michail1
436 2013-05-09 02:21:39 etotheipi_ has quit (Quit: Ex-Chat)
437 2013-05-09 02:21:47 <sipa> ;;bc,blocks
438 2013-05-09 02:21:48 <gribble> 235239
439 2013-05-09 02:25:32 LorenzoMoney is now known as LorenzoMoney_AFK
440 2013-05-09 02:25:39 Haifisch has quit (Read error: Connection reset by peer)
441 2013-05-09 02:25:41 <zooko> "Note: I spent several hours trying, and failing, to create unit tests for this patch" â¹
442 2013-05-09 02:25:52 Haifisch has joined
443 2013-05-09 02:25:56 * sipa 's fault
444 2013-05-09 02:26:06 Casimir1904 has quit (Ping timeout: 276 seconds)
445 2013-05-09 02:29:22 franl has quit (Quit: O Elbereth! Gilthoniel! We still remember ...)
446 2013-05-09 02:30:39 <zooko> sipa: why?
447 2013-05-09 02:30:50 systemParanoid has joined
448 2013-05-09 02:31:09 <sipa> zooko: when I wrote CWallet, I never tested whether an instance without backing database file would work
449 2013-05-09 02:31:18 * zooko nods.
450 2013-05-09 02:31:28 <zooko> We could write "unit tests" which provide a backing db file...
451 2013-05-09 02:33:08 <tumak> wei_: i'm using bitcoinjs for that
452 2013-05-09 02:35:19 <wei_> tumak: is there a monitoring example?
453 2013-05-09 02:35:19 <sipa> gmaxwell: have fun: http://bitcoin.sipa.be/pruning.txt
454 2013-05-09 02:35:44 <sipa> (height, timestamp, unspent outputs, total outputs, unspent output size, all outputs sie)
455 2013-05-09 02:35:53 <zooko> Where is the code that determines which candidate transactions get put into a block that a miner is working on?
456 2013-05-09 02:35:54 trang has quit (Read error: Operation timed out)
457 2013-05-09 02:36:01 trang_ has joined
458 2013-05-09 02:36:22 <sipa> zooko: CreateNewBlock, main.cpp
459 2013-05-09 02:36:26 <zooko> Thanks!
460 2013-05-09 02:36:34 <tumak> wei_: yes, for webclient. i guess you want to do that from nodejs though?
461 2013-05-09 02:37:18 <wei_> yes
462 2013-05-09 02:37:37 <wei_> thanks for the tip though
463 2013-05-09 02:37:40 <wei_> looking now
464 2013-05-09 02:37:44 mrkent has quit (Ping timeout: 260 seconds)
465 2013-05-09 02:37:49 <tumak> well, the low level api is getAffectedTransactions
466 2013-05-09 02:38:00 <tumak> i guess there is that kind of api in bitcoinj as well
467 2013-05-09 02:38:28 <tumak> wei_: i'd use bitcoinjs if you want to do socket.io remoting of that, as most of necessary code is already written and its actually user friendly
468 2013-05-09 02:38:35 mrkent has joined
469 2013-05-09 02:38:36 mrkent has quit (Changing host)
470 2013-05-09 02:38:36 mrkent has joined
471 2013-05-09 02:38:58 jaequery has joined
472 2013-05-09 02:39:29 <tumak> wei_: this is how it's done on bitcoinjs thin client https://github.com/katuma/bitcoinjs-gui/blob/master/scripts/exitnode.js#L59
473 2013-05-09 02:39:54 jaequery has quit (Client Quit)
474 2013-05-09 02:42:30 <gmaxwell> zooko: that decision is also greatly influenced by which transactions made it into the mempool in the first place.
475 2013-05-09 02:43:04 <zooko> gmaxwell: where is that determined?
476 2013-05-09 02:43:10 AlbertTuring has joined
477 2013-05-09 02:43:26 XertroV has quit (Quit: Leaving.)
478 2013-05-09 02:43:50 <sipa> gmaxwell: http://bitcoin.sipa.be/pruning.png
479 2013-05-09 02:44:38 <sipa> zooko: accepttomempool something
480 2013-05-09 02:45:10 systemParanoid has quit (Ping timeout: 245 seconds)
481 2013-05-09 02:46:17 <zooko> sipa: thanks!
482 2013-05-09 02:46:26 <sipa> gmaxwell: with a linear scale it looks far more impressive!
483 2013-05-09 02:46:34 <wei_> tumak: thanks that looks similar to what I need
484 2013-05-09 02:48:34 jaequery has joined
485 2013-05-09 02:48:36 <gmaxwell> sipa: that upward bend in the utxo size on the right there is unfortunateâ but hey, can't see it yet in log scale.
486 2013-05-09 02:48:59 [\\\] has quit (Ping timeout: 256 seconds)
487 2013-05-09 02:49:04 <gmaxwell> zooko: most important part being IsStandard() which is where the dust patch changes it.
488 2013-05-09 02:49:36 <zooko> gmaxwell: I see that.
489 2013-05-09 02:50:00 <sipa> somehow gmaxwell manages to give way better answers, by not answering the actual questions...
490 2013-05-09 02:50:29 <gmaxwell> If people knew what they actually needed to be asking, they probably could have answered the question themselves.
491 2013-05-09 02:50:30 <zooko> Well, I'm actually looking for something else besides the IsDust() thing right now...
492 2013-05-09 02:50:48 [\\\] has joined
493 2013-05-09 02:50:53 <zooko> But yes, the part about "what goes into mempool" is highly relevant to what I actually needed to be asking.
494 2013-05-09 02:51:57 resinate has joined
495 2013-05-09 02:52:39 Belxjander has joined
496 2013-05-09 02:53:50 <gmaxwell> IsStandard mostly serves the purpose of making is so that to exploit some nasty bug (dos attack, netsplit, forwards incompatiblity) the requires unusual transactions you have to be able to mine a block (or at least convince someone who can to do it for you).
497 2013-05-09 02:55:06 <zooko> I don't really understand that.
498 2013-05-09 02:55:56 fiesh has quit (Read error: Operation timed out)
499 2013-05-09 02:56:13 fiesh has joined
500 2013-05-09 02:56:19 _caedes has joined
501 2013-05-09 02:56:21 agricocb has joined
502 2013-05-09 02:56:23 <gmaxwell> There have several times been discovered exploitable vulnerabilities in the initial codebase that required mining non-standard scripts. Things like OP_RETURN returning true and letting you spend any coin.
503 2013-05-09 02:57:28 <zooko> Hm.
504 2013-05-09 02:57:59 <zooko> So IsStandard() filters what (standard) miners will accept into blocks, but not what a recipient would accept as a valid transaction?
505 2013-05-09 02:58:01 systemParanoid has joined
506 2013-05-09 02:58:03 <nsh> *features
507 2013-05-09 02:58:26 <gmaxwell> zooko: IsStandard makes everyone (with that code) drop the transactions unless they show up in a block.
508 2013-05-09 02:58:56 * zooko reads IsStandard
509 2013-05-09 02:59:23 <gmaxwell> The forward incompatiblity part is more subtle. Bitcoin was designed to be forwards compatible in specific ways so new things could be added without breaking old nodes. For example, there are several OP_NOPs for this purpose. Existing nodes treat them as NO_OPs but future nodes can use them to trigger more rules in transactions, so long as if a parsing with the new opcodes as noops also works.
510 2013-05-09 02:59:46 caedes has quit (Ping timeout: 268 seconds)
511 2013-05-09 02:59:54 <gmaxwell> The problem is that to deploy one of these new rules safely you need (1) a supermajority of miners imposing it, (2) no one mining transactions incompatible with the new rule.
512 2013-05-09 03:00:29 <gmaxwell> If there is nothing like IsStandard blocking them, however, griefers could be getting older nodes to mine weird looking transactions that violate a new rule.
513 2013-05-09 03:00:40 <zooko> Hm.
514 2013-05-09 03:00:42 <gmaxwell> And then deploying any new feature is risky.
515 2013-05-09 03:01:04 <gmaxwell> But a narrower IsStandard could accomplish that end, and ultimately I believe we'll use a narrower one like that.
516 2013-05-09 03:01:09 <zooko> Subtle.
517 2013-05-09 03:01:24 <cjd> interesting
518 2013-05-09 03:01:40 <nsh> cabbages
519 2013-05-09 03:01:47 <gmaxwell> ... and not just grifers. I mean, there is an economic advantage to miners to get their software-upgrading-lazy minority hashrate competition to mine invalid blocks.
520 2013-05-09 03:02:29 <cjd> It's easy to see IsStandard() as not very useful (at least for OP_RETURN attacks) since the attacker can just go hit up luke to get mined
521 2013-05-09 03:02:45 <gmaxwell> cjd: luke doesn't apply no restrictions, he applies his own.
522 2013-05-09 03:03:01 <cjd> that helps
523 2013-05-09 03:03:08 <cjd> esp. if they are unpublished
524 2013-05-09 03:03:17 <gmaxwell> cjd: and some cases like the lshift bug can't be mined, but instead exploited nodes as they were accepted into memory pools.
525 2013-05-09 03:03:35 <cjd> yeap
526 2013-05-09 03:03:40 <gmaxwell> well s/exploited/crashed/ I'm not aware of it being used to do anything other than crash.
527 2013-05-09 03:04:08 <cjd> I'm not familiar with the lshift one but that was the case with the hash tree thing
528 2013-05-09 03:04:18 * zooko nods.
529 2013-05-09 03:04:19 <gmaxwell> The anti-dos purpose of IsStandard could be accomplished instead by things like ratelimiting and prioritizing and using finite sized pools. But there are some protocol challenges, because if the acceptance behavior is "non-determinstic" you really want a way to find out if your transaction made it anywhere... and we don't have that yet.
530 2013-05-09 03:05:10 <cjd> yeah, it would be really cool if miners could send a +1 for a transaction when they let it in
531 2013-05-09 03:05:26 <gmaxwell> cjd: or if everyone used something like p2pool.
532 2013-05-09 03:05:34 <zooko> gmaxwell: yeah, that's related to what I'm thinking about.
533 2013-05-09 03:05:35 <cjd> indeed
534 2013-05-09 03:05:39 <gmaxwell> P2pool makes concrete proof that hashpower is being applied to particular transactions.
535 2013-05-09 03:05:46 <cjd> otoh p2pool means everything is vanilla
536 2013-05-09 03:06:22 saracen has quit (Ping timeout: 252 seconds)
537 2013-05-09 03:06:33 <gmaxwell> I'm actually now concerned about the dust change because I'm worried that people are going to over-popularize the knobs to adjust them. And then people are going to set random values. There was already multiple people on the forums posting saying that they'd turn them _up_ because they think that the default dust limit isn't agressive enough.
538 2013-05-09 03:07:03 <gmaxwell> cjd: huh? p2pool makes no special restrictions on the transactions you mine today.
539 2013-05-09 03:07:11 <gmaxwell> thats up to each user.
540 2013-05-09 03:07:14 fanquake has joined
541 2013-05-09 03:07:33 RBecker is now known as rbecker
542 2013-05-09 03:07:38 <cjd> yeah but without centralized pools, there will be less interest in tinkering with the rules, less attention paid
543 2013-05-09 03:07:48 <cjd> makes vulnerability a binary state
544 2013-05-09 03:07:50 <gmaxwell> cjd: though I've actually suggested that in the future p2pool like systems might actually refuse to pay participants who replace with-fee transactions in the share chain with doublespends.
545 2013-05-09 03:07:51 Haifisch has quit (Remote host closed the connection)
546 2013-05-09 03:08:15 <cjd> hmm
547 2013-05-09 03:08:32 <gmaxwell> cjd: I dunno about that. It's hard to say relative to hashpower, but several of the larger p2pool users have been running their own rules (I know because many have gotten help from me patching bitcoind to implement them)
548 2013-05-09 03:08:47 <cjd> hmmahh, that's cool
549 2013-05-09 03:08:53 <cjd> s/^hmm//
550 2013-05-09 03:09:19 <gmaxwell> The major pools mostly don't twiddle with the rules. :( I consider this unfortunate to a degree.
551 2013-05-09 03:09:31 <gmaxwell> But at least it means the default stuff is probably not so terrible.
552 2013-05-09 03:09:43 <cjd> yeah, that is unfortunate
553 2013-05-09 03:09:44 saracen has joined
554 2013-05-09 03:10:08 trang_ has quit (Read error: Connection reset by peer)
555 2013-05-09 03:10:27 trang has joined
556 2013-05-09 03:17:05 trang has quit (Quit: trang)
557 2013-05-09 03:17:19 <cjd> re timestamping, suppose I take a hash, point multiply it into a pubkey, hash that and pay to it, then pay from it to another key and reveal the private key... am I still breaking the rules?
558 2013-05-09 03:17:26 <cjd> it's not in the utxo set forevar...
559 2013-05-09 03:20:20 <Luke-Jr> cjd: why don't you want to do it the right way?
560 2013-05-09 03:20:50 XertroV has joined
561 2013-05-09 03:20:54 <cjd> proof size constraint & lazieness
562 2013-05-09 03:22:13 <Luke-Jr> to prove the timestamp you need the blockchain headers. your size constraint is impractical
563 2013-05-09 03:22:37 <cjd> everyone stores headers, that storage is ok
564 2013-05-09 03:22:42 <cjd> the problem is the proof itself
565 2013-05-09 03:23:20 <Luke-Jr> anyhow, I don't see how your size is notably affected by doing it right
566 2013-05-09 03:23:29 <Luke-Jr> either way, you need the merkle leaves
567 2013-05-09 03:23:33 <Luke-Jr> s/leaves/links
568 2013-05-09 03:24:09 <cjd> "right way" == the p2pool thing?
569 2013-05-09 03:24:37 <Luke-Jr> cjd: no, just make a merkle tree of message hashes, and include them directly in the block coinbase
570 2013-05-09 03:24:43 macboz has quit (Ping timeout: 276 seconds)
571 2013-05-09 03:24:57 <Luke-Jr> well, in the merged mining merkletree
572 2013-05-09 03:26:21 <cjd> so pay some small fee to a miner to have the root of my tree included in the merged mining tree?
573 2013-05-09 03:26:26 ngc0202 has quit (Ping timeout: 255 seconds)
574 2013-05-09 03:27:11 <cjd> could work
575 2013-05-09 03:28:21 macboz has joined
576 2013-05-09 03:30:15 CodeShark has quit (Read error: Connection reset by peer)
577 2013-05-09 03:30:23 CodeShark has joined
578 2013-05-09 03:30:33 <gmaxwell> cjd: your proof size is no larger doing it the O(1) merged mining merkletree way than having it in a transaction.
579 2013-05-09 03:30:48 andyh2 has joined
580 2013-05-09 03:31:02 Corndawg has quit (Read error: Connection reset by peer)
581 2013-05-09 03:31:05 <gmaxwell> cjd: who says you pay a fee? if you construct good software for this miners will run it just to keep the timestamp crap out of the blockchain.
582 2013-05-09 03:31:15 Namworld has quit (Read error: Connection reset by peer)
583 2013-05-09 03:31:22 Corndawg has joined
584 2013-05-09 03:31:27 <cjd> hmm
585 2013-05-09 03:31:34 <cjd> actually you make a very good point
586 2013-05-09 03:31:55 digitalmagus2 has quit (Remote host closed the connection)
587 2013-05-09 03:32:00 <cjd> I'd happily pay a fee for a short proof but others would happily tollerate a long proof and should not need to pay a fee
588 2013-05-09 03:32:37 <gmaxwell> cjd: yea, and if the proofs are in a non-binary tree... people could pay for placement.
589 2013-05-09 03:32:42 geb_ has quit (Quit: ZNC - http://znc.sourceforge.net)
590 2013-05-09 03:32:42 digitalmagus has joined
591 2013-05-09 03:33:26 <cjd> but then I am not entirely sure if my usecase is more a colorcoin usecase or a timestamping usecase
592 2013-05-09 03:33:34 <cjd> depends on ambition really
593 2013-05-09 03:33:44 brson has quit (Ping timeout: 260 seconds)
594 2013-05-09 03:33:49 WKNiGHT has quit (Ping timeout: 276 seconds)
595 2013-05-09 03:34:28 malaimo has quit (Ping timeout: 276 seconds)
596 2013-05-09 03:35:09 malaimo has joined
597 2013-05-09 03:35:16 geb has joined
598 2013-05-09 03:36:04 andyh2 has quit (Ping timeout: 260 seconds)
599 2013-05-09 03:36:14 Guest96003 has joined
600 2013-05-09 03:37:57 <nsh> The best lack all conviction, while the worst / Are full of passionate intensity.
601 2013-05-09 03:38:15 <Diablo-D3> what are we talking about here?
602 2013-05-09 03:39:04 <gmaxwell> well, be aware that bitcoin transactions may be moderately expensive in the future, and the networks friendlyness towards non-currency uses can't be guaranteed. (my personal view is that activity that otherwise looks and smells like currency activity and doesn't produce any additional costs like utxo bloat is just ducky)
603 2013-05-09 03:39:41 LorenzoMoney_AFK has quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/)
604 2013-05-09 03:40:10 <MarceFe> I need implement authentication with sql in stratum mining, someone know how to?
605 2013-05-09 03:40:26 <cjd> thing is, for me dropping 2$ on one proof is ok if that's how it has to be, less is better but I'm certainly not constrained to 1 satoshi outputs
606 2013-05-09 03:41:22 AndChat64721 has joined
607 2013-05-09 03:41:44 richcollins has quit (Quit: richcollins)
608 2013-05-09 03:42:13 SkillsToShow_ has quit (Read error: Connection reset by peer)
609 2013-05-09 03:42:16 <cjd> ofc I always have to consider the possibility of future witch hunts against transaction patterns or even addresses which are "associated with bad activity"
610 2013-05-09 03:43:05 <zooko> Witch hunts on whose part?
611 2013-05-09 03:43:26 <nsh> in a p2p system, everyone should have an equal chance to be hunting the witch
612 2013-05-09 03:43:27 AlbertTuring2 has joined
613 2013-05-09 03:44:05 <cjd> but admittedly if there is a miner's conspiracy to not mine valid transactions involving a list of addresses then we're in deep trouble anyway
614 2013-05-09 03:44:28 <helo> there already probably is
615 2013-05-09 03:44:41 * cjd eyesroll
616 2013-05-09 03:44:54 SkillsToShow has joined
617 2013-05-09 03:44:58 <zooko> Would this conspiracy also conspire to refuse to build on blocks of other miners that did include these witchy transactions?
618 2013-05-09 03:45:12 <nsh> yay unto the seventh generation
619 2013-05-09 03:45:14 <Luke-Jr> cjd: it's miners' job to filter out spam. if spammers make it easy by using known addresses, why not? ;)
620 2013-05-09 03:45:24 AlbertTuring has quit (Ping timeout: 276 seconds)
621 2013-05-09 03:46:03 <cjd> because these kinds of rules go south fast
622 2013-05-09 03:47:54 <cjd> not unlike the irc ridiculousness where they kline people for saying dcc send... thus you can use the rule to kill anyone's bot as long as it will echo a message for you
623 2013-05-09 03:48:16 <cjd> you get more vulnerability, not less
624 2013-05-09 03:48:32 <nsh> +1
625 2013-05-09 03:48:35 <gmaxwell> zooko: as nsh says...
626 2013-05-09 03:48:39 Namworld has joined
627 2013-05-09 03:48:56 <gmaxwell> I recently closed and rejected a pull request to add a GUI for blacklisting addresses.
628 2013-05-09 03:49:01 <Luke-Jr> cjd: but they're necessary 'rules'.
629 2013-05-09 03:49:15 <cjd> gmaxwell: +1
630 2013-05-09 03:49:17 <Luke-Jr> cjd: if you don't filter spam, the system cannot function
631 2013-05-09 03:49:19 <gmaxwell> (and there has been a moderate amount of "the developers are evil scumm" for that)
632 2013-05-09 03:49:22 <cjd> that stuff is scary
633 2013-05-09 03:49:43 <gmaxwell> (of course, accepting it would have resulted in even more)
634 2013-05-09 03:49:52 <gmaxwell> (plus been bad for bitcoin)
635 2013-05-09 03:49:55 <cjd> it sounds so good
636 2013-05-09 03:50:02 [7] has quit (Disconnected by services)
637 2013-05-09 03:50:08 <cjd> until you start reasoning it out D:
638 2013-05-09 03:50:10 TheSeven has joined
639 2013-05-09 03:50:29 <gmaxwell> cjd: well, fortunately "correct" use of bitcoin makes filtering by addresses pretty ineffective.
640 2013-05-09 03:50:52 <cjd> /nod
641 2013-05-09 03:51:02 _caedes has quit (Ping timeout: 256 seconds)
642 2013-05-09 03:51:23 Thepok has joined
643 2013-05-09 03:51:32 <cjd> and as things grow the pay pattern will tend toward immitating a hash function
644 2013-05-09 03:51:39 <cjd> :)
645 2013-05-09 03:52:03 <nsh> cjd, ?
646 2013-05-09 03:52:12 <cjd> nothin :)
647 2013-05-09 03:52:16 <nsh> l
648 2013-05-09 03:52:21 <nsh> *k
649 2013-05-09 03:53:21 fanquake_ has joined
650 2013-05-09 03:55:08 <nsh> hmm
651 2013-05-09 03:55:29 paybitcoin has joined
652 2013-05-09 03:55:34 fanquake has quit (Ping timeout: 256 seconds)
653 2013-05-09 03:55:35 fanquake_ is now known as fanquake
654 2013-05-09 03:55:56 jspilman has joined
655 2013-05-09 03:56:12 molecular has joined
656 2013-05-09 03:56:29 paybitcoin1 has quit (Ping timeout: 252 seconds)
657 2013-05-09 03:57:13 <nsh> Virtual particles are also excitations of the underlying fields, but are "temporary" in the sense that they appear in calculations of interactions, but never as asymptotic states or indices to the scattering matrix. As such the accuracy and use of virtual particles in calculations is firmly established, but their "reality" or existence is a question of philosophy rather than science.
658 2013-05-09 03:59:35 molec has quit (Ping timeout: 268 seconds)
659 2013-05-09 03:59:47 jaequery has quit (Quit: Textual IRC Client: www.textualapp.com)
660 2013-05-09 04:03:51 gfinn has joined
661 2013-05-09 04:04:11 tumak has quit (Read error: Connection reset by peer)
662 2013-05-09 04:04:28 _caedes has joined
663 2013-05-09 04:05:14 systemParanoid has quit (Quit: Leaving)
664 2013-05-09 04:13:05 SkillsToShow has quit (Read error: Connection reset by peer)
665 2013-05-09 04:14:56 richcollins has joined
666 2013-05-09 04:15:11 SkillsToShow has joined
667 2013-05-09 04:15:16 <Raccoon> Everyone please help block access to that trojan/malware site Bitcoin-Alarm by reporting it to google, so Firefox and Chrome block access.
668 2013-05-09 04:15:20 <Raccoon> --->> https://www.google.com/safebrowsing/report_badware/?url=http%3A%2F%2Fwww%2EBitcoin-Alarm%2Ecom%2F
669 2013-05-09 04:16:03 <cjd> ;;t
670 2013-05-09 04:16:04 <gribble> Google Safe Browsing: Report a Malware Page
671 2013-05-09 04:16:11 <cjd> what are they running?
672 2013-05-09 04:16:18 alexwaters has joined
673 2013-05-09 04:16:42 FabianB_ has quit (Read error: Operation timed out)
674 2013-05-09 04:16:53 <cjd> wallet stealer?
675 2013-05-09 04:17:02 <alexwaters> anyone going to the conference?
676 2013-05-09 04:17:05 FabianB has joined
677 2013-05-09 04:17:05 FabianB has quit (Changing host)
678 2013-05-09 04:17:05 FabianB has joined
679 2013-05-09 04:17:21 <alexwaters> i'm hoping to actually meet Luke Dash, main reason I'm going
680 2013-05-09 04:18:13 zoinky has quit (Quit: Leaving.)
681 2013-05-09 04:18:50 <Raccoon> cjd: yes, wallet stealer
682 2013-05-09 04:18:55 <Raccoon> trojan
683 2013-05-09 04:19:16 <Raccoon> cjd: they've been hammering people in #bitcoin with PM spam.
684 2013-05-09 04:19:24 <Raccoon> [11:09] <ywau8_14638> www.Bitcoin-Alarm.com - You will like it!
685 2013-05-09 04:19:42 <cjd> right
686 2013-05-09 04:21:12 * cjd sends link to ircerr for his collection
687 2013-05-09 04:25:31 RazielZ has quit (Ping timeout: 246 seconds)
688 2013-05-09 04:25:53 dlb76 has quit (Ping timeout: 245 seconds)
689 2013-05-09 04:27:52 <Luke-Jr> alexwaters: O.o
690 2013-05-09 04:28:18 Elmf has quit (Ping timeout: 276 seconds)
691 2013-05-09 04:28:38 <jgarzik> Off-topic: IBM's homomorphic encryption lib: https://github.com/shaih/HElib
692 2013-05-09 04:28:50 ageis has quit (Quit: http://ageispolis.net)
693 2013-05-09 04:28:53 Thepok has quit (Ping timeout: 252 seconds)
694 2013-05-09 04:29:09 <cjd> Someone told me that HE is really slow, like something that takes AES seconds takes months with HE
695 2013-05-09 04:29:30 zoinky has joined
696 2013-05-09 04:30:19 <zooko> cjd: yes.
697 2013-05-09 04:30:40 macboz_ has joined
698 2013-05-09 04:30:45 ageis has joined
699 2013-05-09 04:31:00 <Raccoon> jgarzik, zooko: hit that link please.
700 2013-05-09 04:32:22 macboz_ has quit (Max SendQ exceeded)
701 2013-05-09 04:32:47 <jgarzik> Here is a free link to LWN's coverage on IBM homomorphic encryption: http://lwn.net/SubscriberLink/549665/0f64b379d75a1c0a/
702 2013-05-09 04:32:56 <jgarzik> good overview article on HE
703 2013-05-09 04:33:07 <jgarzik> it jibes with what cjd just said
704 2013-05-09 04:33:17 macboz_ has joined
705 2013-05-09 04:33:51 macboz has quit (Ping timeout: 264 seconds)
706 2013-05-09 04:34:49 rushed has joined
707 2013-05-09 04:34:51 <gmaxwell> cjd: yea, it's really slow. And no IO ... but .. if you just need it toâ sayâ tally a vote.. no big deal, and the computation can be justified.
708 2013-05-09 04:35:03 macboz__ has joined
709 2013-05-09 04:35:29 <cjd> if you want secure voting, you can just use a blockchain
710 2013-05-09 04:35:30 <alexwaters> jgarzik: can I think of it as essentially a black box that can be passed to a untrusted party? the party can make calls to the black box to receive information, but not see what's in the box? struggling with understanding how the party gets any use out of the ciphertext
711 2013-05-09 04:35:30 r0sc0e has joined
712 2013-05-09 04:35:35 <cjd> in some random way
713 2013-05-09 04:35:46 <cjd> maybe put the blockchain next to the voting booth, then it will be secure
714 2013-05-09 04:35:54 <cjd> or dust payments, that will surely help
715 2013-05-09 04:36:33 <gmaxwell> cjd: you could put it on your hat and then you would be secure too.
716 2013-05-09 04:36:38 <alexwaters> i remember google mentioning using it for search, so that even they would not know what a user searched for - but my understanding of how that works is scant
717 2013-05-09 04:36:39 <gmaxwell> wear it like a tierra.
718 2013-05-09 04:36:43 <cjd> :D
719 2013-05-09 04:37:48 <cjd> poly1305 is somewhat of a similar idea, you multiply modulo 2**130-5 each block of data and then add a secret to the result
720 2013-05-09 04:38:10 macboz_ has quit (Ping timeout: 276 seconds)
721 2013-05-09 04:38:18 <cjd> with the secret, the other end can repeat the process to authenticate the data
722 2013-05-09 04:38:51 <cjd> similar to the general idea of adding numbers w/o knowing their meaning that is
723 2013-05-09 04:40:24 AndChat has joined
724 2013-05-09 04:40:53 MacMiner741 has joined
725 2013-05-09 04:41:41 * midnightmagic backs slowly away from the huge group of drunk people.
726 2013-05-09 04:42:35 BTCOxygen has joined
727 2013-05-09 04:42:39 <alexwaters> or maybe it was that certain parts of their system would not know what the user searched for, still cool
728 2013-05-09 04:42:59 mollison has joined
729 2013-05-09 04:43:05 * Raccoon wants to play pinball whenenver he sees midnightmagic talk
730 2013-05-09 04:43:17 lolcookie has joined
731 2013-05-09 04:43:29 AlbertTuring has joined
732 2013-05-09 04:43:30 AndChat64721 has quit (Ping timeout: 245 seconds)
733 2013-05-09 04:43:50 macboz__ has quit (Ping timeout: 255 seconds)
734 2013-05-09 04:44:49 phish- has left ("Leaving")
735 2013-05-09 04:44:50 BTCOxygen has quit (Ping timeout: 252 seconds)
736 2013-05-09 04:45:12 <alexwaters> https://github.com/alexwaters/bitcointesting.org anyone care to help? i won't be able to work on it for a month
737 2013-05-09 04:45:18 ThomasV has joined
738 2013-05-09 04:46:46 macboz__ has joined
739 2013-05-09 04:47:08 AlbertTuring2 has quit (Ping timeout: 256 seconds)
740 2013-05-09 04:47:40 AtashiCon has quit (Quit: AtashiCon)
741 2013-05-09 04:48:29 <MarceFe> I need implement authentication with sql in stratum mining, someone know how to? Si habla castellano mejor! jaja
742 2013-05-09 04:51:17 digitalmagus has quit ()
743 2013-05-09 04:51:54 AtashiCon has joined
744 2013-05-09 04:53:13 grau_ has joined
745 2013-05-09 04:54:38 XertroV has quit (Ping timeout: 256 seconds)
746 2013-05-09 04:55:27 grau has quit (Ping timeout: 264 seconds)
747 2013-05-09 04:55:46 SteveDekorte has joined
748 2013-05-09 04:56:31 macboz_ has joined
749 2013-05-09 04:57:46 <SteveDekorte> is it a standard for a peer to send the first year worth of blocks following a completed handshake?
750 2013-05-09 04:58:51 zooko has quit (Remote host closed the connection)
751 2013-05-09 04:59:25 <cjd> not if you didn't ask for them
752 2013-05-09 04:59:41 macboz__ has quit (Ping timeout: 252 seconds)
753 2013-05-09 05:02:45 vigilyn2 has quit (Ping timeout: 276 seconds)
754 2013-05-09 05:04:04 <SteveDekorte> cjd: ok, thanks - I just didn't see that packet sent but maybe I'm not logging it
755 2013-05-09 05:05:36 <cjd> yeah, satoshi nodes do ask for blocks if it's not up to date
756 2013-05-09 05:09:26 vigilyn has joined
757 2013-05-09 05:10:26 SteveDekorte has quit (Quit: SteveDekorte)
758 2013-05-09 05:12:00 nidefawl has joined
759 2013-05-09 05:12:14 fanquake_ has joined
760 2013-05-09 05:12:55 SteveDekorte has joined
761 2013-05-09 05:13:59 fanquake has quit (Ping timeout: 252 seconds)
762 2013-05-09 05:14:00 fanquake_ is now known as fanquake
763 2013-05-09 05:16:27 ThomasV has quit (Ping timeout: 264 seconds)
764 2013-05-09 05:19:18 andyh2 has joined
765 2013-05-09 05:22:55 tumak has joined
766 2013-05-09 05:26:50 xire has quit (Ping timeout: 245 seconds)
767 2013-05-09 05:27:21 asd34sd has joined
768 2013-05-09 05:27:32 XertroV has joined
769 2013-05-09 05:28:16 sdfsdf has quit (Ping timeout: 264 seconds)
770 2013-05-09 05:31:35 macboz_ has quit (Ping timeout: 252 seconds)
771 2013-05-09 05:32:32 r0sc0e has quit ()
772 2013-05-09 05:34:14 * MacMiner741 has been synchronizing for many hours.. something wrong?
773 2013-05-09 05:34:37 <cjd> is the # of weeks decreasing?
774 2013-05-09 05:37:05 <MarceFe> cjd satoshi tommi?
775 2013-05-09 05:37:21 <cjd> o_O
776 2013-05-09 05:37:39 <MarceFe> hahaaha
777 2013-05-09 05:43:16 AndChat has quit (64721!~AndChat64@101.173.33.154|Ping timeout: 264 seconds)
778 2013-05-09 05:43:35 AlbertTuring2 has joined
779 2013-05-09 05:43:57 FredEE has quit (Quit: FredEE)
780 2013-05-09 05:45:39 AlbertTuring has quit (Ping timeout: 276 seconds)
781 2013-05-09 05:50:41 xire has joined
782 2013-05-09 05:50:41 xire has quit (Changing host)
783 2013-05-09 05:50:41 xire has joined
784 2013-05-09 05:55:31 wei_ has quit (Quit: wei_)
785 2013-05-09 06:00:00 fanquake_ has joined
786 2013-05-09 06:00:14 owowo has quit (Quit: sayonara)
787 2013-05-09 06:02:01 fanquake has quit (Ping timeout: 276 seconds)
788 2013-05-09 06:02:01 fanquake_ is now known as fanquake
789 2013-05-09 06:03:17 FredEE has joined
790 2013-05-09 06:10:45 stainrat has joined
791 2013-05-09 06:11:33 GordonG3kko has joined
792 2013-05-09 06:13:00 sud3n has quit (Ping timeout: 245 seconds)
793 2013-05-09 06:17:22 mollison has left ("PART #bitcoin-mining :QUIT :Leaving.")
794 2013-05-09 06:19:44 gt_ has joined
795 2013-05-09 06:20:56 gt_ has quit (Client Quit)
796 2013-05-09 06:23:26 safra has joined
797 2013-05-09 06:24:22 <SteveDekorte> is an inv block packet supposed to be broadcast automatically to peers when a valid new block is seen?
798 2013-05-09 06:27:00 paybitcoin has quit (Ping timeout: 256 seconds)
799 2013-05-09 06:27:06 paybitcoin1 has joined
800 2013-05-09 06:28:19 <jgarzik> SteveDekorte: probably, presuming you want to serve a block to others
801 2013-05-09 06:29:00 <SteveDekorte> is there a doc that describes these details - I'm not seeing them in https://en.bitcoin.it/wiki/Protocol_specification but maybe I missed something
802 2013-05-09 06:29:30 <gmaxwell> No.
803 2013-05-09 06:29:41 <gmaxwell> And that page is considerably incomplete.
804 2013-05-09 06:29:46 <SteveDekorte> I'm not seeing other nodes broadcast new blocks atm but I thought I was before
805 2013-05-09 06:30:13 <SteveDekorte> gmaxwell: is there a better source?
806 2013-05-09 06:30:19 <jgarzik> The source
807 2013-05-09 06:31:00 jeewee has joined
808 2013-05-09 06:32:00 <SteveDekorte> I was afraid someone would say that :)
809 2013-05-09 06:32:04 MarceFe has quit ()
810 2013-05-09 06:33:32 andyh2 has quit (Quit: Leaving...)
811 2013-05-09 06:33:33 zoinky has quit (Quit: Leaving.)
812 2013-05-09 06:37:54 _caedes has quit (Remote host closed the connection)
813 2013-05-09 06:42:09 stainrat has quit (Remote host closed the connection)
814 2013-05-09 06:43:31 <zooko_> Hey SteveDekorte.
815 2013-05-09 06:44:05 richcollins has quit (Quit: richcollins)
816 2013-05-09 06:44:29 Nash has joined
817 2013-05-09 06:46:07 duSn has joined
818 2013-05-09 06:46:56 Nash has quit (Read error: Operation timed out)
819 2013-05-09 06:52:27 <midnightmagic> +1 Raccoon for the pinball reference. BTW, good alternative is zen pinball on ps3
820 2013-05-09 06:52:43 rushed has quit (Quit: rushed)
821 2013-05-09 06:53:14 <Raccoon> midnightmagic: I have this feeling we talked about pinball like 5 years ago
822 2013-05-09 06:53:48 caput has quit (Ping timeout: 245 seconds)
823 2013-05-09 06:54:20 nus has quit (Read error: Connection reset by peer)
824 2013-05-09 06:54:21 <midnightmagic> Raccoon: It's possible. Maybe more like 3-4.
825 2013-05-09 06:54:41 caput has joined
826 2013-05-09 06:54:49 nus has joined
827 2013-05-09 06:55:04 <Raccoon> when you can remember an atari game like midnightmagic, 3..4..5 become a blur :p
828 2013-05-09 06:55:46 zz_highPriestLOL is now known as highPriestLOL
829 2013-05-09 06:56:33 nus has quit (Read error: Connection reset by peer)
830 2013-05-09 06:57:11 nus has joined
831 2013-05-09 06:57:57 <midnightmagic> I played it on an Apple. I never did find the release for the C=64 though..
832 2013-05-09 06:58:02 AndChat64721 has joined
833 2013-05-09 06:58:11 <MacMiner741> do people handle litecoin questions the same as bitcoin questions?
834 2013-05-09 06:59:02 wei_ has joined
835 2013-05-09 06:59:22 AndChat has joined
836 2013-05-09 07:00:06 wei_ has quit (Client Quit)
837 2013-05-09 07:00:38 <SteveDekorte> hi zooko_
838 2013-05-09 07:02:26 AndChat64721 has quit (Ping timeout: 255 seconds)
839 2013-05-09 07:04:51 Nash has joined
840 2013-05-09 07:04:54 <Graet> MacMiner741, rarely, #litecoin-dev might be better ;)
841 2013-05-09 07:04:55 wei_ has joined
842 2013-05-09 07:05:17 <jordandotdev> hey all I have a question about running testnet
843 2013-05-09 07:05:19 <MacMiner741> ty
844 2013-05-09 07:05:29 <Graet> :)
845 2013-05-09 07:05:39 <jordandotdev> umm so should I create a 2nd directory and start with the -testnet and -datadir flag pointing to this new directory?
846 2013-05-09 07:05:55 <jordandotdev> if I've already downloaded the blockchain once - to save time later?
847 2013-05-09 07:06:06 <midnightmagic> jordandotdev: No it'll automatically use a subdirectory in your .bitcoin directory, called "testnet3"
848 2013-05-09 07:06:23 <jordandotdev> ok, do I need to modify my bitcoin.conf when I use the -testnet flag?
849 2013-05-09 07:06:37 <midnightmagic> jordandotdev: Just use -testnet, and make sure you have it aliased or something so yo don't accidentally spent your real bitcoin.
850 2013-05-09 07:06:50 <midnightmagic> jordandotdev: Better yet, shut down you rmain bitcoind while you're monkeying around with testnet.
851 2013-05-09 07:06:55 <jordandotdev> there isn't any real btc on there right now
852 2013-05-09 07:07:04 <midnightmagic> jordandotdev: No, you don't have to.
853 2013-05-09 07:07:10 <jordandotdev> just testing my web-ui
854 2013-05-09 07:07:18 <midnightmagic> want some testnet coins?
855 2013-05-09 07:07:27 <jordandotdev> not yet ... haven't started up yet !
856 2013-05-09 07:07:51 <midnightmagic> Well. I have a whole bunch, so let me know, I'll send some.
857 2013-05-09 07:08:05 <jordandotdev> thanx!
858 2013-05-09 07:08:13 <midnightmagic> don't thank me yet. :)
859 2013-05-09 07:08:35 fanquake has left ()
860 2013-05-09 07:09:07 <SteveDekorte> zooko_: you used to hang out on the io channel right?
861 2013-05-09 07:10:45 <jordandotdev> midnightmagic - one last question can I use the same 'accounts' when running -testnet
862 2013-05-09 07:11:08 <jordandotdev> and they'll have testnet addresses?
863 2013-05-09 07:12:11 <midnightmagic> jordandotdev: The databases are in a separate location in a subdirectory under your .bitcoin directory. There is no overlap. You'll have to create the other accounts all over again.
864 2013-05-09 07:12:21 <jordandotdev> got it no problem
865 2013-05-09 07:13:11 <jordandotdev> I'm gonna move over to armory I think for security reasons
866 2013-05-09 07:13:20 <jordandotdev> but this is a good start -
867 2013-05-09 07:14:27 <jordandotdev> does anyone recommend armory versus bitcoind for security?
868 2013-05-09 07:18:18 Evilmax has joined
869 2013-05-09 07:18:38 mrkent has quit (Read error: Connection reset by peer)
870 2013-05-09 07:18:38 <wumpus> eh you still needs bitcoind with armory
871 2013-05-09 07:18:47 Evilmax has left ()
872 2013-05-09 07:18:49 alexwaters has quit (Quit: Leaving.)
873 2013-05-09 07:18:51 <jordandotdev> ok I get it
874 2013-05-09 07:18:56 <jordandotdev> it runs on top of it?
875 2013-05-09 07:19:00 <wumpus> yes
876 2013-05-09 07:19:11 <wumpus> it uses it for the block chain but not the wallet
877 2013-05-09 07:19:12 alexwaters has joined
878 2013-05-09 07:19:12 mrkent has joined
879 2013-05-09 07:19:48 digitalmagus has joined
880 2013-05-09 07:22:43 rushed has joined
881 2013-05-09 07:23:49 alexwaters has quit (Ping timeout: 256 seconds)
882 2013-05-09 07:24:13 Prattler has joined
883 2013-05-09 07:25:04 Quetzalcoatl_ has quit (Ping timeout: 246 seconds)
884 2013-05-09 07:25:23 whiterabbit has joined
885 2013-05-09 07:27:02 grau_ has quit (Remote host closed the connection)
886 2013-05-09 07:27:03 wrabbit has quit (Ping timeout: 276 seconds)
887 2013-05-09 07:27:08 whiterabbit is now known as wrabbit
888 2013-05-09 07:27:35 grau has joined
889 2013-05-09 07:30:40 LainZ has quit (Ping timeout: 264 seconds)
890 2013-05-09 07:34:00 <MacMiner741> is there any kind of local encryption protection, to encrypt keystrokes into your bitcoin wallet?
891 2013-05-09 07:34:45 twobitcoins_ has joined
892 2013-05-09 07:34:48 vigilyn2 has joined
893 2013-05-09 07:35:09 vigilyn has quit (Read error: Connection reset by peer)
894 2013-05-09 07:36:16 <kadoban> MacMiner741: you can encrypt your wallet with a passphrase if that's what you mean. otherwise, probably not
895 2013-05-09 07:37:59 twobitcoins has quit (Ping timeout: 255 seconds)
896 2013-05-09 07:38:44 ProfMac has joined
897 2013-05-09 07:38:45 cypher has quit (Ping timeout: 276 seconds)
898 2013-05-09 07:41:23 o-p has quit (Ping timeout: 252 seconds)
899 2013-05-09 07:43:28 paybitcoin has joined
900 2013-05-09 07:45:22 o-p has joined
901 2013-05-09 07:46:33 paybitcoin1 has quit (Ping timeout: 276 seconds)
902 2013-05-09 07:47:14 cypher has joined
903 2013-05-09 07:48:31 brwyatt is now known as brwyatt|Away
904 2013-05-09 07:52:05 SteveDekorte has quit (Quit: SteveDekorte)
905 2013-05-09 07:54:37 toffoo has quit ()
906 2013-05-09 07:55:39 gfinn has quit (Ping timeout: 276 seconds)
907 2013-05-09 07:56:03 <CodeShark> you can use dedicated hardware
908 2013-05-09 08:05:08 ovidiusoft has joined
909 2013-05-09 08:05:09 eoss has joined
910 2013-05-09 08:05:09 eoss has quit (Changing host)
911 2013-05-09 08:05:09 eoss has joined
912 2013-05-09 08:07:45 nizeguy has joined
913 2013-05-09 08:08:05 xire has quit (Ping timeout: 245 seconds)
914 2013-05-09 08:11:11 Nash has quit (Ping timeout: 268 seconds)
915 2013-05-09 08:16:21 <nsh> it is possible to get similar results to fully homomorphic encryption with an amortization speed-up using pre-computations with a secret-sharing algorithm
916 2013-05-09 08:16:26 <nsh> which is kinda neat
917 2013-05-09 08:16:39 <nsh> iirc there was a good google talk about it
918 2013-05-09 08:17:31 <nsh> sorry, multi-party computation
919 2013-05-09 08:18:10 <nsh> this, possibly: http://www.youtube.com/watch?v=LRAN_w1_qmw Multi-Party Computation: From Theory to Practice (54:29) 2,761 views, â
â
â
â
â by 35 raters
920 2013-05-09 08:18:29 viperhr1 has quit (Ping timeout: 255 seconds)
921 2013-05-09 08:21:22 wei_ has quit (Quit: wei_)
922 2013-05-09 08:22:17 phpwn has quit (Ping timeout: 268 seconds)
923 2013-05-09 08:22:41 wei_ has joined
924 2013-05-09 08:23:30 iddo has quit (Ping timeout: 245 seconds)
925 2013-05-09 08:23:49 iddo has joined
926 2013-05-09 08:24:14 CaptainBlaze has joined
927 2013-05-09 08:24:57 xire has joined
928 2013-05-09 08:24:57 xire has quit (Changing host)
929 2013-05-09 08:24:57 xire has joined
930 2013-05-09 08:26:48 FredEE has quit (Quit: FredEE)
931 2013-05-09 08:29:14 ovidiusoft has quit (Ping timeout: 252 seconds)
932 2013-05-09 08:29:33 andrew_scorpil1 has joined
933 2013-05-09 08:30:20 <andrew_scorpil1> hi guys. Anyone using mtgox vanilla WebSocket streaming API? Does it _ever_ work?
934 2013-05-09 08:30:32 ovidiusoft has joined
935 2013-05-09 08:30:56 <weex> andrew_scorpil1: i hear there's some ddos action now but you can try #mtgox
936 2013-05-09 08:31:12 <andrew_scorpil1> thanks
937 2013-05-09 08:32:40 xire has quit (Ping timeout: 245 seconds)
938 2013-05-09 08:37:43 MobiusL has quit (Quit: Ex-Chat)
939 2013-05-09 08:38:36 Thepok has joined
940 2013-05-09 08:38:49 caput has quit (Ping timeout: 245 seconds)
941 2013-05-09 08:38:49 gritball has quit (Ping timeout: 245 seconds)
942 2013-05-09 08:39:54 egis has joined
943 2013-05-09 08:44:17 lolcookie has quit (Ping timeout: 256 seconds)
944 2013-05-09 08:44:38 macboz_ has joined
945 2013-05-09 08:46:51 fanquake has joined
946 2013-05-09 08:48:46 r0sc0e has joined
947 2013-05-09 08:49:05 ovidiusoft has quit (Ping timeout: 255 seconds)
948 2013-05-09 08:49:47 caput has joined
949 2013-05-09 08:49:47 xenesis has joined
950 2013-05-09 08:50:22 ovidiusoft has joined
951 2013-05-09 08:51:24 Muis has quit (Read error: Connection reset by peer)
952 2013-05-09 08:52:12 Jasmin68k has joined
953 2013-05-09 08:52:34 daktak has quit (Read error: Connection reset by peer)
954 2013-05-09 08:52:38 xenesis_ has joined
955 2013-05-09 08:53:07 Mobius_ has joined
956 2013-05-09 08:53:09 daktak has joined
957 2013-05-09 08:54:48 xenesis has quit (Ping timeout: 276 seconds)
958 2013-05-09 08:54:48 xenesis_ is now known as xenesis
959 2013-05-09 08:55:04 caput has quit (Ping timeout: 245 seconds)
960 2013-05-09 08:55:19 icellan has joined
961 2013-05-09 08:55:19 matas has joined
962 2013-05-09 08:55:27 XertroV has quit (Ping timeout: 276 seconds)
963 2013-05-09 08:56:01 Mobius_ is now known as MobiusL
964 2013-05-09 08:56:01 MobiusL has quit (Excess Flood)
965 2013-05-09 08:56:18 xenesis_ has joined
966 2013-05-09 08:56:26 Mobius_ has joined
967 2013-05-09 08:56:33 t7 has joined
968 2013-05-09 08:56:48 Mobius_ is now known as MobiusL
969 2013-05-09 09:00:00 xenesis has quit (Ping timeout: 276 seconds)
970 2013-05-09 09:00:00 xenesis_ is now known as xenesis
971 2013-05-09 09:02:30 caput has joined
972 2013-05-09 09:03:08 Lolcust has quit (Quit: Nap time)
973 2013-05-09 09:06:23 Lolcust has joined
974 2013-05-09 09:11:41 xire has joined
975 2013-05-09 09:11:41 xire has quit (Changing host)
976 2013-05-09 09:11:41 xire has joined
977 2013-05-09 09:11:43 <t7> the go btc node is pretty nice
978 2013-05-09 09:13:20 <t7> well its not a node
979 2013-05-09 09:13:33 <t7> but has the network protocol done already
980 2013-05-09 09:13:54 icellan has quit (Quit: icellan)
981 2013-05-09 09:14:24 icellan has joined
982 2013-05-09 09:14:48 <nsh> t7, what size board?
983 2013-05-09 09:17:24 <t7> wut
984 2013-05-09 09:17:38 <t7> not the board game... :3
985 2013-05-09 09:18:03 <nsh> :)
986 2013-05-09 09:18:06 <SomeoneWeird> wat
987 2013-05-09 09:18:08 rushed has quit (Quit: rushed)
988 2013-05-09 09:18:09 <SomeoneWeird> oh
989 2013-05-09 09:18:10 <SomeoneWeird> lol
990 2013-05-09 09:18:44 <nsh> not impressed then. if someone implements a slim node using actual go board and pebble manipulations, well, that dude(tte) would get a pizza
991 2013-05-09 09:19:07 icellan has quit (Client Quit)
992 2013-05-09 09:20:47 xenesis_ has joined
993 2013-05-09 09:20:57 <jgm> Their proof of concept site is okay but I wish that people would start using units when quoting bitcoin values. 0.00081754 is just too ugly to look at (let alone 5.43e-05)
994 2013-05-09 09:21:35 <jgm> Oh, and they seem to have lost all the inputs in the transaction view
995 2013-05-09 09:22:49 eoss has quit (Ping timeout: 256 seconds)
996 2013-05-09 09:22:54 <duSn> if you think 0.00081754 is 5.43e-05 then i think we should do some bidness
997 2013-05-09 09:23:00 Casimir1904 has joined
998 2013-05-09 09:23:33 <jgm> duSn: Separate entries on one of their pages
999 2013-05-09 09:24:03 xenesis has quit (Ping timeout: 276 seconds)
1000 2013-05-09 09:24:03 xenesis_ is now known as xenesis
1001 2013-05-09 09:25:24 Jasmin68k has quit (Quit: Leaving.)
1002 2013-05-09 09:25:51 Jasmin68k has joined
1003 2013-05-09 09:27:41 saulimus has joined
1004 2013-05-09 09:29:19 XertroV has joined
1005 2013-05-09 09:29:23 TD has joined
1006 2013-05-09 09:30:05 muhoo has quit (Ping timeout: 245 seconds)
1007 2013-05-09 09:30:47 hazrd has joined
1008 2013-05-09 09:30:50 <hazrd> hey all
1009 2013-05-09 09:32:05 r0sc0e has quit ()
1010 2013-05-09 09:32:53 <hazrd> is there a way to get the current block from bitcoind? 'getinfo' only shows "blocks" : 234833, for example, which isn't current - would be nice to see how many blocks I would still need to go to be up to date
1011 2013-05-09 09:35:10 Thepok has quit (Ping timeout: 245 seconds)
1012 2013-05-09 09:37:18 muhoo has joined
1013 2013-05-09 09:37:58 Tykling has quit (Excess Flood)
1014 2013-05-09 09:38:01 <t7> hazrd, it must be in the client somewhere because it shows a progress bar in the gui
1015 2013-05-09 09:38:17 <t7> not sure its exposed by bitcoind though
1016 2013-05-09 09:39:27 <hazrd> t7, hmmm - does the client maybe connect to blockchain.info (or something else) to get this info? or potentially a peer...?
1017 2013-05-09 09:39:46 CodesInChaos has joined
1018 2013-05-09 09:41:03 <t7> i doubt it
1019 2013-05-09 09:41:16 <t7> probably asks other peers
1020 2013-05-09 09:41:21 <hazrd> ok - will go sniff through the source some more
1021 2013-05-09 09:41:23 <hazrd> thanks ;)
1022 2013-05-09 09:46:19 viperhr1 has joined
1023 2013-05-09 09:47:06 <sipa> harthere are two heuristics for guessing the total numer of blocks
1024 2013-05-09 09:47:15 <sipa> and they are both just guesses
1025 2013-05-09 09:47:47 <sipa> one is using the median of the number of blocks claimed by your peers (who can lie about it)
1026 2013-05-09 09:48:06 <sipa> another is looking at the timestamp of the last block
1027 2013-05-09 09:49:15 <hazrd> sipa, so extrapolate current block timestamp with potential blocks to current timestamp
1028 2013-05-09 09:49:17 <hazrd> ?
1029 2013-05-09 09:49:41 <hazrd> (not that i don't trust other peers... :) )
1030 2013-05-09 09:49:43 loltu has quit (Quit: ZNC - http://znc.sourceforge.net)
1031 2013-05-09 09:50:00 <hazrd> s/with/from/
1032 2013-05-09 09:50:45 Muis has joined
1033 2013-05-09 09:51:03 <sipa> indeed
1034 2013-05-09 09:51:04 xenesis_ has joined
1035 2013-05-09 09:51:24 LainZ has joined
1036 2013-05-09 09:51:24 <hazrd> actually, that makes sense - since i have timestamps per block, it wouldn't be hard to "guess" which block i should hit at a specific timestamp
1037 2013-05-09 09:51:32 <hazrd> sipa, perfect, thanks!
1038 2013-05-09 09:51:46 <sipa> it's mostly used guessing "am i done, or not?", not for extrapolating the total number of blocks
1039 2013-05-09 09:51:59 <hazrd> understood ;)
1040 2013-05-09 09:52:52 Tykling has joined
1041 2013-05-09 09:54:26 loltu has joined
1042 2013-05-09 09:54:36 xenesis has quit (Ping timeout: 276 seconds)
1043 2013-05-09 09:54:36 xenesis_ is now known as xenesis
1044 2013-05-09 09:55:35 iwilcox_ has joined
1045 2013-05-09 09:55:36 iwilcox_ has quit (Changing host)
1046 2013-05-09 09:55:36 iwilcox_ has joined
1047 2013-05-09 09:56:35 kicek has joined
1048 2013-05-09 09:57:35 hnz has quit (Ping timeout: 245 seconds)
1049 2013-05-09 09:57:52 iwilcox has quit (Ping timeout: 268 seconds)
1050 2013-05-09 09:58:10 drizztbsd has quit (Remote host closed the connection)
1051 2013-05-09 09:59:35 Anduck has joined
1052 2013-05-09 09:59:39 Anduck has quit (Changing host)
1053 2013-05-09 09:59:39 Anduck has joined
1054 2013-05-09 10:01:39 jrra has quit (Ping timeout: 252 seconds)
1055 2013-05-09 10:02:43 hnz has joined
1056 2013-05-09 10:04:21 iwilcox_ is now known as iwilcox
1057 2013-05-09 10:07:15 Casimir1904 has quit (Ping timeout: 245 seconds)
1058 2013-05-09 10:08:07 MacMiner741 has quit (Read error: Connection reset by peer)
1059 2013-05-09 10:08:33 Jasmin68k has quit (Quit: Leaving.)
1060 2013-05-09 10:10:54 caput has quit (Ping timeout: 245 seconds)
1061 2013-05-09 10:12:14 icellan has joined
1062 2013-05-09 10:12:39 xire has quit (Ping timeout: 252 seconds)
1063 2013-05-09 10:15:09 o-p has quit (Ping timeout: 268 seconds)
1064 2013-05-09 10:16:24 Bjander has joined
1065 2013-05-09 10:16:27 viperhr1 has quit (Ping timeout: 264 seconds)
1066 2013-05-09 10:17:03 matas has quit (Ping timeout: 264 seconds)
1067 2013-05-09 10:18:11 o-p has joined
1068 2013-05-09 10:21:55 CaptainBlaze has quit (Quit: CaptainBlaze)
1069 2013-05-09 10:23:08 Bjander has quit (Remote host closed the connection)
1070 2013-05-09 10:23:45 Mr_G has joined
1071 2013-05-09 10:27:09 caput has joined
1072 2013-05-09 10:28:05 n5 has quit (Ping timeout: 268 seconds)
1073 2013-05-09 10:28:42 px has quit (Ping timeout: 268 seconds)
1074 2013-05-09 10:32:19 daktak has quit (Read error: Connection reset by peer)
1075 2013-05-09 10:32:43 daktak has joined
1076 2013-05-09 10:32:49 mappum has quit (Ping timeout: 256 seconds)
1077 2013-05-09 10:33:39 px has joined
1078 2013-05-09 10:33:45 px has quit (Max SendQ exceeded)
1079 2013-05-09 10:37:57 px has joined
1080 2013-05-09 10:38:42 swulf--1 has joined
1081 2013-05-09 10:41:21 pagoda has joined
1082 2013-05-09 10:41:39 swulf-- has quit (Ping timeout: 264 seconds)
1083 2013-05-09 10:44:45 icellan has quit (Quit: icellan)
1084 2013-05-09 10:44:49 Jasmin68k has joined
1085 2013-05-09 10:46:00 Jasmin68k has quit (Client Quit)
1086 2013-05-09 10:46:07 icellan has joined
1087 2013-05-09 10:46:41 highPriestLOL is now known as zz_highPriestLOL
1088 2013-05-09 10:46:46 Jasmin68k has joined
1089 2013-05-09 10:46:57 icellan has quit (Client Quit)
1090 2013-05-09 10:48:25 egis has quit (Quit: Leaving)
1091 2013-05-09 10:49:25 kicek has quit (Remote host closed the connection)
1092 2013-05-09 10:52:28 Ogig has joined
1093 2013-05-09 10:55:00 Bjander has joined
1094 2013-05-09 10:55:50 rbecker is now known as RBecker
1095 2013-05-09 10:59:38 Xexe has joined
1096 2013-05-09 11:00:44 icellan has joined
1097 2013-05-09 11:00:48 <Xexe> hi folks, can somebody point me out to the place in the sourcecode where the math function limiting the emission is defined?
1098 2013-05-09 11:02:38 TD has quit (Quit: TD)
1099 2013-05-09 11:05:23 Belxjander has quit (Ping timeout: 256 seconds)
1100 2013-05-09 11:06:15 * jaromil rotfl
1101 2013-05-09 11:06:33 <jaromil> sry but there is no #define MAX_BITCOINS 21000000000
1102 2013-05-09 11:06:56 <jaromil> its not a single math function
1103 2013-05-09 11:08:01 <jaromil> your best chance is looking at this and figuring out changes across codebases https://bitcointalk.org/index.php?topic=193025.0;topicseen
1104 2013-05-09 11:08:08 Bjander has quit (Remote host closed the connection)
1105 2013-05-09 11:08:21 Bjander has joined
1106 2013-05-09 11:11:05 <sipa> see GetBlockValue in main.cpp
1107 2013-05-09 11:11:21 macboz_ has quit (Quit: This computer has gone to sleep)
1108 2013-05-09 11:11:39 chmod755 has joined
1109 2013-05-09 11:14:30 Tykling has quit (Excess Flood)
1110 2013-05-09 11:14:41 XertroV has quit (Quit: Leaving.)
1111 2013-05-09 11:14:52 gritball has joined
1112 2013-05-09 11:15:13 paracyst has quit ()
1113 2013-05-09 11:15:26 one_zero has quit ()
1114 2013-05-09 11:16:06 reizuki__ has joined
1115 2013-05-09 11:17:39 Bjander has quit (Read error: Connection reset by peer)
1116 2013-05-09 11:19:24 Tykling has joined
1117 2013-05-09 11:19:50 SirDefaced has quit ()
1118 2013-05-09 11:20:06 drizztbsd has joined
1119 2013-05-09 11:21:10 <pagoda> Hi, I am currently programming my own miner. Can any1 maybe let me know if I am on the right path here:
1120 2013-05-09 11:21:48 <SomeoneWeird> uh
1121 2013-05-09 11:21:59 <pagoda> If I want to submit a "getwork" share; I should submit the header I used to generate the hash, not the hash it self?
1122 2013-05-09 11:22:51 wamatt has quit (Quit: wamatt)
1123 2013-05-09 11:22:58 <pagoda> If so, the header should be a in Hex Little Endian format?
1124 2013-05-09 11:23:02 <matjeh> you need to submit the header and the hash, so it can be validated
1125 2013-05-09 11:24:22 <pagoda> Header & hash? Should they be concatenated? Or as separate parameters?
1126 2013-05-09 11:24:46 <SomeoneWeird> no offense, but if you can't figure out it yourself, there's no point in writing a miner
1127 2013-05-09 11:25:32 testnode9 has joined
1128 2013-05-09 11:25:52 <Xexe> so it's a math curve defined in the code by cutting the Subsidy var by half each time interval?
1129 2013-05-09 11:26:00 <Xexe> so it must be a reverse of y=x^2 right? but the base strangely not e but 2 instead, hmmm why is that?
1130 2013-05-09 11:26:02 nomailin1 has joined
1131 2013-05-09 11:26:17 <Xexe> so what math fucntion precisely is that anyway?
1132 2013-05-09 11:27:05 <Xexe> ohh not y=x^2, but y=2^x..
1133 2013-05-09 11:28:35 BurtyB has quit (Read error: Connection reset by peer)
1134 2013-05-09 11:29:10 BurtyB has joined
1135 2013-05-09 11:29:13 <sipa> Xexe: the subsidy is 50 BTC the first 210000 blocks, then 25 BTC the next 210000 blocks, ...
1136 2013-05-09 11:29:18 <sipa> 210000 is ~ 4 years
1137 2013-05-09 11:29:29 BurtyB has quit (Read error: Connection reset by peer)
1138 2013-05-09 11:29:35 <sipa> (50+25+12.5+6.25+3.125+...) * 2100000 = 21M
1139 2013-05-09 11:29:50 <Xexe> jaromil, how come it's not a single math curve? hmmm it gotta be..
1140 2013-05-09 11:29:58 <Xexe> sipa, yes that's correct..
1141 2013-05-09 11:30:15 <skinnkavaj> Oh Mike Hearn love that guy, he wrote this in mailing list:
1142 2013-05-09 11:30:20 <skinnkavaj> . It's hard to imagine an organisation as
1143 2013-05-09 11:30:23 <skinnkavaj> a big as a mobile carrier engaging in financial scamming (roaming fees
1144 2013-05-09 11:30:26 <skinnkavaj> excepted).
1145 2013-05-09 11:30:32 <skinnkavaj> I've said this before, but I think it's worth repeating. The
1146 2013-05-09 11:30:35 <skinnkavaj> double-spend protection the block chain gives you has a sweet spot
1147 2013-05-09 11:30:38 <skinnkavaj> where it's really, really valuable (essential even) and then there are
1148 2013-05-09 11:30:40 <Xexe> it means to me that the func is just defined that way, kinda empirically..
1149 2013-05-09 11:30:41 <skinnkavaj> lots of kinds of transactions on either side of that sweet spot that
1150 2013-05-09 11:30:44 <skinnkavaj> don't really benefit from it.
1151 2013-05-09 11:30:50 <skinnkavaj> Obvious/trivial case where you don't need a block chain - Facebook
1152 2013-05-09 11:30:53 <skinnkavaj> buys Instagram for a gajillion coins. The legal system is plenty good
1153 2013-05-09 11:30:56 <skinnkavaj> enough to ensure the payments are honoured. Another example, when my
1154 2013-05-09 11:30:59 skinnkavaj has joined
1155 2013-05-09 11:31:15 <sipa> Xexe: yeah
1156 2013-05-09 11:31:18 <sipa> Xexe: gut feeling
1157 2013-05-09 11:31:42 <BlueMatt> skinnkavaj: we've mostly all read it already, no need to paste 20 lines of an email in irc...
1158 2013-05-09 11:33:43 <Xexe> i just realised that the base of the func is not the magical e number as i had assumed before, but 2 instead.
1159 2013-05-09 11:34:30 pagoda has quit (Quit: Page closed)
1160 2013-05-09 11:35:09 <Xexe> looks not that natural to me at the moment, but it can be..
1161 2013-05-09 11:36:40 <sipa> Xexe: my guess: implementation simplicity
1162 2013-05-09 11:36:47 <sipa> and easy-to-explain
1163 2013-05-09 11:36:59 <sipa> "the reward halves every 4 years"
1164 2013-05-09 11:37:19 <Xexe> sipa, yes it well maybe
1165 2013-05-09 11:37:25 reizuki__ has quit (Quit: Konversation terminated!)
1166 2013-05-09 11:38:22 xenesis_ has joined
1167 2013-05-09 11:41:17 BurtyB has joined
1168 2013-05-09 11:41:44 xenesis has quit (Ping timeout: 252 seconds)
1169 2013-05-09 11:42:06 px has quit (Ping timeout: 268 seconds)
1170 2013-05-09 11:42:52 pizzacat has joined
1171 2013-05-09 11:43:45 peddamat has quit (Quit: peddamat)
1172 2013-05-09 11:45:06 xenesis_ has quit (Ping timeout: 276 seconds)
1173 2013-05-09 11:46:59 Jackneill has joined
1174 2013-05-09 11:49:04 triciam has quit (Quit: TTFN)
1175 2013-05-09 11:53:33 <sipa> ;;nethash
1176 2013-05-09 11:53:34 <gribble> 86004.7581216
1177 2013-05-09 11:54:16 <sipa> in two days, we'll likely cross 2^70 hashes as PoW in the main chain
1178 2013-05-09 12:01:08 parus_ is now known as parus
1179 2013-05-09 12:01:25 wei_ has quit (Quit: wei_)
1180 2013-05-09 12:03:06 AlbertTuring2 has quit (Quit: Nettalk6 - www.ntalk.de)
1181 2013-05-09 12:05:10 davout has joined
1182 2013-05-09 12:08:07 fanquake_ has joined
1183 2013-05-09 12:08:13 icellan has quit (Quit: icellan)
1184 2013-05-09 12:09:09 safra has quit (Ping timeout: 276 seconds)
1185 2013-05-09 12:09:15 fanquake has quit (Ping timeout: 264 seconds)
1186 2013-05-09 12:09:16 fanquake_ is now known as fanquake
1187 2013-05-09 12:11:09 tucenaber has quit (Read error: Connection reset by peer)
1188 2013-05-09 12:11:45 BTCOxygen has quit (1!~BTCOxygen@unaffiliated/btcoxygen|Ping timeout: 276 seconds)
1189 2013-05-09 12:12:21 neo2 has joined
1190 2013-05-09 12:12:43 icellan has joined
1191 2013-05-09 12:15:37 XertroV has joined
1192 2013-05-09 12:16:05 duSn has quit (Quit: leaving)
1193 2013-05-09 12:16:28 matas has joined
1194 2013-05-09 12:18:56 AndChat has quit (64721!~AndChat64@1.153.236.148|Ping timeout: 245 seconds)
1195 2013-05-09 12:19:09 kicek has joined
1196 2013-05-09 12:19:29 paraipan has joined
1197 2013-05-09 12:20:47 safra has joined
1198 2013-05-09 12:20:55 safra has quit (Remote host closed the connection)
1199 2013-05-09 12:21:35 rdymac has joined
1200 2013-05-09 12:22:05 daybyter has joined
1201 2013-05-09 12:22:27 LainZ has quit (Ping timeout: 264 seconds)
1202 2013-05-09 12:22:54 ahf has quit (Quit: Reconnecting)
1203 2013-05-09 12:23:21 ahf has joined
1204 2013-05-09 12:25:11 takeyourhatoff has quit (Ping timeout: 245 seconds)
1205 2013-05-09 12:26:59 BTCOxygen has joined
1206 2013-05-09 12:28:29 BTCOxygen has quit (1!~BTCOxygen@unaffiliated/btcoxygen|Max SendQ exceeded)
1207 2013-05-09 12:28:34 tucenaber has joined
1208 2013-05-09 12:28:34 tucenaber has quit (Changing host)
1209 2013-05-09 12:28:34 tucenaber has joined
1210 2013-05-09 12:29:21 BTCOxygen has joined
1211 2013-05-09 12:30:49 BTCOxygen has quit (1!~BTCOxygen@unaffiliated/btcoxygen|Max SendQ exceeded)
1212 2013-05-09 12:31:43 BTCOxygen has joined
1213 2013-05-09 12:36:11 fanquake_ has joined
1214 2013-05-09 12:37:49 fanquake has quit (Ping timeout: 256 seconds)
1215 2013-05-09 12:37:50 fanquake_ is now known as fanquake
1216 2013-05-09 12:37:55 XertroV has quit (Quit: Leaving.)
1217 2013-05-09 12:38:01 MobPhone has joined
1218 2013-05-09 12:39:46 Skav has quit (Ping timeout: 245 seconds)
1219 2013-05-09 12:40:14 Skav has joined
1220 2013-05-09 12:42:48 MobPhone has quit (Ping timeout: 252 seconds)
1221 2013-05-09 12:45:20 agricocb has quit (Quit: Leaving.)
1222 2013-05-09 12:48:51 MoALTz has quit (Ping timeout: 252 seconds)
1223 2013-05-09 12:49:58 icellan has quit (Quit: icellan)
1224 2013-05-09 12:51:08 MoALTz has joined
1225 2013-05-09 12:53:58 MobPhone has joined
1226 2013-05-09 12:54:35 andreas has joined
1227 2013-05-09 12:55:27 daybyter has quit (Ping timeout: 252 seconds)
1228 2013-05-09 12:55:45 Jackneill has quit (Remote host closed the connection)
1229 2013-05-09 12:56:09 wamatt has joined
1230 2013-05-09 12:56:45 Skav has quit (Ping timeout: 256 seconds)
1231 2013-05-09 13:00:44 icellan has joined
1232 2013-05-09 13:00:44 icellan has quit (Client Quit)
1233 2013-05-09 13:01:04 gjs278 has quit (Remote host closed the connection)
1234 2013-05-09 13:01:27 gjs278 has joined
1235 2013-05-09 13:03:10 XertroV has joined
1236 2013-05-09 13:03:49 caput has quit (Ping timeout: 245 seconds)
1237 2013-05-09 13:07:58 Skav has joined
1238 2013-05-09 13:08:51 RazielZ has joined
1239 2013-05-09 13:09:04 RazielZ has quit (Read error: Connection reset by peer)
1240 2013-05-09 13:11:03 MobPhone has quit (Ping timeout: 264 seconds)
1241 2013-05-09 13:12:15 RazielZ has joined
1242 2013-05-09 13:12:30 Namworld has quit ()
1243 2013-05-09 13:12:33 agricocb has joined
1244 2013-05-09 13:21:40 tonikt has joined
1245 2013-05-09 13:22:01 agricocb has quit (Remote host closed the connection)
1246 2013-05-09 13:22:53 agricocb has joined
1247 2013-05-09 13:27:33 cc_8 has joined
1248 2013-05-09 13:27:33 cc_8 is now known as alphaguru
1249 2013-05-09 13:27:41 Jasmin68k has quit (Quit: Leaving.)
1250 2013-05-09 13:33:21 agricocb has quit (Remote host closed the connection)
1251 2013-05-09 13:34:26 agricocb has joined
1252 2013-05-09 13:38:49 skinnkavaj has quit (Ping timeout: 256 seconds)
1253 2013-05-09 13:38:58 Mr_G has quit (Ping timeout: 256 seconds)
1254 2013-05-09 13:39:39 agricocb has quit (Remote host closed the connection)
1255 2013-05-09 13:39:42 skinnkavaj has joined
1256 2013-05-09 13:39:43 skinnkavaj has quit (Changing host)
1257 2013-05-09 13:39:43 skinnkavaj has joined
1258 2013-05-09 13:40:08 agricocb has joined
1259 2013-05-09 13:44:29 XertroV has quit (Quit: Leaving.)
1260 2013-05-09 13:50:41 suporte85 has quit (Quit: Saindo)
1261 2013-05-09 13:51:20 rdponticelli has quit (Ping timeout: 276 seconds)
1262 2013-05-09 13:51:45 tonikt has quit (Quit: Leaving)
1263 2013-05-09 13:52:34 andreas has quit (Read error: Connection reset by peer)
1264 2013-05-09 13:53:34 grau_ has joined
1265 2013-05-09 13:54:00 CaptainBlaze has joined
1266 2013-05-09 13:54:01 savetheinternet has joined
1267 2013-05-09 13:54:44 CaptainBlaze has left ()
1268 2013-05-09 13:55:10 rdponticelli has joined
1269 2013-05-09 13:56:59 CaptainBlaze has joined
1270 2013-05-09 13:57:06 grau has quit (Ping timeout: 256 seconds)
1271 2013-05-09 14:02:23 Muis_ has joined
1272 2013-05-09 14:06:10 Muis has quit (Ping timeout: 256 seconds)
1273 2013-05-09 14:09:08 grau_ has quit (Remote host closed the connection)
1274 2013-05-09 14:09:41 grau has joined
1275 2013-05-09 14:11:02 Guest96003 has quit (Changing host)
1276 2013-05-09 14:11:02 Guest96003 has joined
1277 2013-05-09 14:11:08 Guest96003 is now known as WKNiGHT
1278 2013-05-09 14:11:58 GordonG3kko has quit (K-Lined)
1279 2013-05-09 14:12:51 altamic has joined
1280 2013-05-09 14:15:27 Jere_Jones has quit ()
1281 2013-05-09 14:15:44 malaimo has quit (Ping timeout: 252 seconds)
1282 2013-05-09 14:16:04 Jere_Jones has joined
1283 2013-05-09 14:16:35 <hazrd> can someone explain to me how the values after line 36 of checkpoints.cpp is calculated?
1284 2013-05-09 14:17:23 malaimo has joined
1285 2013-05-09 14:18:06 Anduck has quit (Ping timeout: 245 seconds)
1286 2013-05-09 14:19:23 taha has joined
1287 2013-05-09 14:20:22 BTCOxygen is now known as 1!~BTCOxygen@unaffiliated/btcoxygen|BTCOxygen
1288 2013-05-09 14:20:52 XertroV has joined
1289 2013-05-09 14:21:00 <sipa> hazrd: i probably can
1290 2013-05-09 14:21:11 cads has quit (Quit: Leaving)
1291 2013-05-09 14:21:31 <sipa> oh, line 36... those are just the checkpoints themself
1292 2013-05-09 14:21:32 <hazrd> sipa, i understand that each check needs to be surrounded by blocks with "good" timestamps - but that's it
1293 2013-05-09 14:21:43 <sipa> the hashes of the blocks that are checksums
1294 2013-05-09 14:21:56 <hazrd> ok, so those are chosen at random?
1295 2013-05-09 14:22:07 <hazrd> (the hard-coded values, i mean)
1296 2013-05-09 14:22:15 <sipa> when there's a new release, and it has been long enough since the previous checkpoint, a new one is added
1297 2013-05-09 14:23:16 <hazrd> i've been wracking my brain to find some correlation between value 1, 2, 1500 and 4032
1298 2013-05-09 14:23:35 agricocb has quit (Quit: Leaving.)
1299 2013-05-09 14:23:40 <hazrd> so, in short, there is none, it's simply a arb number based on "it's been a long enough time"?
1300 2013-05-09 14:24:07 <edcba> i still think those blocks should only locked according to a formula
1301 2013-05-09 14:24:20 <edcba> ie no maintenance for releases
1302 2013-05-09 14:24:38 <hazrd> edcba, that's what i thought and for the life of me, I couldn't figure out the pattern
1303 2013-05-09 14:24:52 <hazrd> (i'm reasonably good with that, so it was very frustrating :P )
1304 2013-05-09 14:24:53 <edcba> ie if chain longer than 32 then lock 3/4 of the chain
1305 2013-05-09 14:25:14 <hazrd> at least it makes sense now - thank you :)
1306 2013-05-09 14:25:21 <sipa> 1, 2, 1500, 4032?
1307 2013-05-09 14:25:33 <devurandom> May a transaction be confirmed in two blocks? Or may it always only appear in one?
1308 2013-05-09 14:25:45 <sipa> devurandom: the second would be a double spend
1309 2013-05-09 14:25:46 <rdponticelli> hazrd: One chosen because it had a lot of elevens, gavin's favourite number... :p
1310 2013-05-09 14:26:03 eqolo has joined
1311 2013-05-09 14:26:06 <devurandom> sipa: But you could easily verify that the same transaction appears in two blocks, not?
1312 2013-05-09 14:26:12 <sipa> devurandom: no
1313 2013-05-09 14:26:14 <eqolo> sipa: i have a problem bro
1314 2013-05-09 14:26:23 <devurandom> sipa: Why not?
1315 2013-05-09 14:26:29 <rdponticelli> One of the last ones... ^
1316 2013-05-09 14:26:31 <eqolo> you know me since some days ago, iam new here right?
1317 2013-05-09 14:26:39 <hazrd> sipa, yes, the checkpoints chosen ;)
1318 2013-05-09 14:26:39 <sipa> devurandom: because the database only remembers unspent transaction outputs
1319 2013-05-09 14:26:57 <devurandom> ?
1320 2013-05-09 14:26:59 <sipa> hazrd: i don't see those numbers anywhere
1321 2013-05-09 14:27:00 <edcba> hmm
1322 2013-05-09 14:27:14 <edcba> so you can't have leftovers for 1 address ?
1323 2013-05-09 14:27:17 XertroV has quit (Ping timeout: 252 seconds)
1324 2013-05-09 14:27:19 <sipa> devurandom: so if the transaction was already spent, you can't detect the duplicate anymore as a duplicate
1325 2013-05-09 14:27:31 <devurandom> What I assumed is that the transactionid is unique and if it appears twice, I can throw away one of them?
1326 2013-05-09 14:27:36 <sipa> devurandom: it's just some transaction that spends an input that doesn't exist, which is invalid, exactly as it would be
1327 2013-05-09 14:27:43 mrkent has quit (Ping timeout: 276 seconds)
1328 2013-05-09 14:27:46 <devurandom> "throw away" i.e. not bother about it, because I've seen it already.
1329 2013-05-09 14:28:00 <hazrd> sipa, apologies - wrong code - the numbers in the official bitcoin source are 11111, 33333, 74000 etc.
1330 2013-05-09 14:28:01 <devurandom> sipa: So the transaction has no unique id?
1331 2013-05-09 14:28:05 <edcba> can i make some tx with (A,B) sending to (C,D,E,A) ?
1332 2013-05-09 14:28:08 <sipa> devurandom: yes, it has
1333 2013-05-09 14:28:11 <sipa> devurandom: its hash
1334 2013-05-09 14:28:28 <devurandom> Ok, the hash becomes unique, because a transaction may only appear in one block.
1335 2013-05-09 14:28:30 <chmod755> edcba, A = Alice, B = Bob?
1336 2013-05-09 14:28:33 <devurandom> Then it makes sence.
1337 2013-05-09 14:28:37 <devurandom> sense
1338 2013-05-09 14:28:44 <edcba> chmod755: some bitcoin addrs
1339 2013-05-09 14:28:46 <devurandom> sipa: Thanks :)
1340 2013-05-09 14:28:52 <sipa> devurandom: but you seem to think there's a special rule forbidding duplicate transactions
1341 2013-05-09 14:28:59 systemParanoid has joined
1342 2013-05-09 14:29:04 <sipa> devurandom: there is no such thing: it's simply invalid because it spends something that was already spent
1343 2013-05-09 14:29:11 CaptainBlaze has quit (Quit: CaptainBlaze)
1344 2013-05-09 14:29:26 <devurandom> sipa: So the same transaction id (i.e. hash) may appear in multiple blocks)?
1345 2013-05-09 14:29:30 <sipa> devurandom: no!
1346 2013-05-09 14:29:47 <sipa> devurandom: the second transaction will be consuming the same inputs as the first, so it is invalid
1347 2013-05-09 14:29:56 <devurandom> So there is a rule that forbids duplicate transactions?
1348 2013-05-09 14:29:59 <sipa> NO!
1349 2013-05-09 14:30:09 <sipa> there is a rule that forbids spending the same coin twice
1350 2013-05-09 14:30:09 <devurandom> But you just named the rule...?
1351 2013-05-09 14:30:21 <chmod755> edcba, you mean like this: https://blockchain.info/tx/c0a758e4dbe0a10519e24c2dcea94307f978aca6adc2ebfedb8ab6e933fbfd1a ?
1352 2013-05-09 14:30:23 <sipa> which is pretty much the most basic rule in bitcoin
1353 2013-05-09 14:30:31 <sipa> no double spending allowed
1354 2013-05-09 14:30:38 <chmod755> edcba, multiple inputs / outputs?
1355 2013-05-09 14:30:47 <devurandom> But in the end its the same. Whether I forbid that a transaction hash appears twice, or I forbid the transaction to appear twice - same...
1356 2013-05-09 14:30:58 <sipa> devurandom: NO
1357 2013-05-09 14:31:12 <sipa> devurandom: it is not about forbidding the same transaction hash appearing twice
1358 2013-05-09 14:31:19 <edcba> chmod755: yes but having 1 of inputs = 1 of outputs
1359 2013-05-09 14:31:23 <cjd> devurandom: I could craft 2 completely different transactions which target the same money, that would be invalid
1360 2013-05-09 14:31:23 <sipa> it is forbidding spending an input that is already spent
1361 2013-05-09 14:31:37 gavinandresen has quit (Quit: gavinandresen)
1362 2013-05-09 14:31:48 <devurandom> And "already spent" means: appears in a lower block?
1363 2013-05-09 14:31:54 <sipa> yes
1364 2013-05-09 14:32:06 <t7> err, already spent in a lower block
1365 2013-05-09 14:32:12 <sipa> or in the same block
1366 2013-05-09 14:32:20 <devurandom> Well, has a transaction in a lower block that draws from this output.
1367 2013-05-09 14:32:20 <eqolo> sipa: I feel very badly treated, please take a look: http://pastebin.com/Q7vwdeiR
1368 2013-05-09 14:32:31 <cjd> devurandom: also it will not accept transactions into it's pool if it knows about one already which targets the same money
1369 2013-05-09 14:32:35 n5 has joined
1370 2013-05-09 14:32:38 <helo> eqolo: i addressed you in #bitcoin
1371 2013-05-09 14:32:42 <cjd> that's what the patch on the list is all about
1372 2013-05-09 14:32:43 <devurandom> Well, if it appears in the same block, the block itself is garbage...
1373 2013-05-09 14:32:52 <edcba> so an address is either spent or has a constant amount of btc for its life ??
1374 2013-05-09 14:32:52 <eqolo> sipa: Is this fair? What have I violated the rules, to get banned in this way?
1375 2013-05-09 14:33:03 <sipa> eqolo: i have no authority in #bitcoin-otc
1376 2013-05-09 14:33:14 <t7> eqolo, you sent 4 messages of the same thing
1377 2013-05-09 14:33:17 <eqolo> sipa: but please take a look. and see what you think about it
1378 2013-05-09 14:33:25 <edcba> helo: it's not #bitcoin complain service...
1379 2013-05-09 14:33:31 Diapolis has joined
1380 2013-05-09 14:33:35 <helo> edcba: better than in here
1381 2013-05-09 14:33:39 <SomeoneWeird> eqolo, if you were banned you were banned for a reason, go discuss it with ne0futur
1382 2013-05-09 14:33:49 <cjd> eqolo: it's always fair, how dare you question it (unless it was vaginaroda, in which case it is always unfair)
1383 2013-05-09 14:34:05 Diapolis_ has joined
1384 2013-05-09 14:34:05 <edcba> discuss it in private
1385 2013-05-09 14:34:07 <sipa> devurandom: not more or less garbage than if the transaction already existed in a former block
1386 2013-05-09 14:34:13 <eqolo> SomeoneWeird: <ne0futur> i wont answer then ne0futur You must log in with services to message this user
1387 2013-05-09 14:34:22 <sipa> devurandom: or if the transaction would consume a coin already consumed by a different transaction
1388 2013-05-09 14:34:36 <eqolo> i feel very bad about this. i didnt do anything
1389 2013-05-09 14:34:39 <devurandom> ok, thanks
1390 2013-05-09 14:34:40 <devurandom> And computing power controls who is able to double-spend? I.e. if the majority of the hashrate is in evil hands, they can create such transactions that draw from already spend outputs?
1391 2013-05-09 14:34:52 <sipa> devurandom: within one chain you _cannot_ double spend
1392 2013-05-09 14:35:04 <cjd> ^
1393 2013-05-09 14:35:04 CaptainBlaze has joined
1394 2013-05-09 14:35:07 <edcba> if the majority decides something it's the rule...
1395 2013-05-09 14:35:08 <sipa> devurandom: double spending always results from making someone believe one chain is valid, and than switching to another
1396 2013-05-09 14:35:14 <sipa> edcba: NO NO NO
1397 2013-05-09 14:35:17 <edcba> :)
1398 2013-05-09 14:35:18 Irencus has joined
1399 2013-05-09 14:35:25 <cjd> hehe
1400 2013-05-09 14:35:28 <devurandom> Ok, what's so bad about someone controling the majority of the hashrate?
1401 2013-05-09 14:35:47 <cjd> they can roll back blocks by not mining against them
1402 2013-05-09 14:35:53 <sipa> devurandom: because that means they can make a new longer chain, and have others switch to it
1403 2013-05-09 14:35:54 <eqolo> cjd: take a look. ofc i dare question that, if i get banned for nothing
1404 2013-05-09 14:35:57 <cjd> so you believe this is the chain and then suddenly it changes
1405 2013-05-09 14:36:00 <edcba> i mean it's a futile exercise to imagine what happens when someone has >50%
1406 2013-05-09 14:36:10 <SomeoneWeird> eqolo, this is very offtopic here, take it somewhere else
1407 2013-05-09 14:36:13 <sipa> devurandom: double spending is a bad term, it should be called 'transaction cancelling'
1408 2013-05-09 14:36:17 <devurandom> sipa: So double-spending is actually giving someone btc that dont actually exist to make the human happy, but he will later figure out that the money he has is actually forged?
1409 2013-05-09 14:36:45 <edcba> ghost spending !
1410 2013-05-09 14:36:50 <sipa> devurandom: bitcoin is a consensus system, designed to make the world agree as fast as possible about the order of transactions in history
1411 2013-05-09 14:36:57 <sipa> devurandom: but it's not perfect
1412 2013-05-09 14:37:22 <sipa> and if someone controls a majority of the hashpower, they get to cheat the system
1413 2013-05-09 14:37:40 <sipa> but whatever happens, within _one_ chain, double spending is never possible
1414 2013-05-09 14:38:02 <sipa> it's just because there are always chains competing with eachother, with typically only 1 or 2 blocks disagreement
1415 2013-05-09 14:38:11 Diapolis has quit (Ping timeout: 256 seconds)
1416 2013-05-09 14:38:29 <cjd> Also ofc controlling a majority of the hashpower allows them to change the basic rules of the client and force nodes to autoupdate
1417 2013-05-09 14:38:40 <sipa> cjd: ...
1418 2013-05-09 14:38:41 <sipa> wut?
1419 2013-05-09 14:38:44 <cjd> =)
1420 2013-05-09 14:38:46 <devurandom> Ok, so as long as MrEvil builds on his wrong chain, he can put transactions in there and draw real inputs and pay people with it. But then at any moment he can build from an earlier block a new chain with completely new transactions that draw from the same inputs, and the people he formerly paid suddenly have invalid money?
1421 2013-05-09 14:39:04 <sipa> devurandom: basically, yes
1422 2013-05-09 14:39:24 <devurandom> cjd: autoupdate???
1423 2013-05-09 14:39:38 <sipa> devurandom: he was kidding (i hope)
1424 2013-05-09 14:39:44 <devurandom> Is the algorithm encoded within the blocks themselves?
1425 2013-05-09 14:39:45 <cjd> guize cmon
1426 2013-05-09 14:39:50 <sipa> devurandom: no!
1427 2013-05-09 14:39:50 <devurandom> ok...
1428 2013-05-09 14:39:53 <cjd> critical thinking !
1429 2013-05-09 14:39:54 <sipa> devurandom: absolutely not
1430 2013-05-09 14:40:06 <devurandom> Well, that would make autoupdate possible...
1431 2013-05-09 14:40:12 <devurandom> A great feature :P
1432 2013-05-09 14:40:19 <devurandom> Ok, thanks for explaining all this!
1433 2013-05-09 14:40:30 <sipa> devurandom: it would also completely defeat (well, call it 'change') the basic security principles
1434 2013-05-09 14:40:42 <devurandom> yeah
1435 2013-05-09 14:40:45 <sipa> as it means the majority of hashpower gets to decide pretty much everything
1436 2013-05-09 14:40:57 <sipa> they could vote to give themself a higher mining subsidy for example
1437 2013-05-09 14:41:23 <sipa> which is pretty much the greatest thing about bitcoin: those creating money need to follow strict rules that can be verified by everyone
1438 2013-05-09 14:41:32 <cjd> I think autoupdate would be a nice idea for a part of the code, namely the network facing stuff, certainly not hinging on the majority hashrate though and the autoupdate code would have to be sandboxed (IE: nacl)
1439 2013-05-09 14:41:34 <devurandom> So, if I am mining and find a new block being published, I should check whether I believe it and then discard the transactions that are already contained in there, and then start hashing the remaining transactions?
1440 2013-05-09 14:41:46 <sipa> devurandom: indeed
1441 2013-05-09 14:41:56 <sipa> but you restart continuously anyway
1442 2013-05-09 14:42:02 <devurandom> Why?
1443 2013-05-09 14:42:13 SirDefaced has joined
1444 2013-05-09 14:42:16 <sipa> because it's just random hashing
1445 2013-05-09 14:42:24 <sipa> there is no progress towards finding a block
1446 2013-05-09 14:42:26 <devurandom> But on the other hand this means, that not all transactions will end up in blocks. Some will just never appear in any chain...?
1447 2013-05-09 14:42:35 <sipa> why not?
1448 2013-05-09 14:42:44 toffoo has joined
1449 2013-05-09 14:42:53 <sipa> which ones won't?
1450 2013-05-09 14:43:23 <devurandom> re: random hashing: I have to try all nonces to be confident that this block is not hashable, right? I.e. before I havent tried all of them, I cannot say that it will never go below the target.
1451 2013-05-09 14:44:25 <sipa> no, you don't need to try all nonces
1452 2013-05-09 14:44:40 <devurandom> Well, if there is a new block every 10 minutes and the transactions in it are part of its hash, and I have to throw away my progress when a new block appears, that means that I have to backlog these transactions. So whether a transaction appears within a block is random.
1453 2013-05-09 14:44:53 <sipa> there is no such thing as 'progress'
1454 2013-05-09 14:45:00 <sipa> you just hash, and check, and hash and check
1455 2013-05-09 14:45:09 <sipa> and every attempt has exactly the same chance as any other
1456 2013-05-09 14:45:15 <cjd> 'progress' is an interesting falacy
1457 2013-05-09 14:45:17 <sipa> whether they come from the same set of transactions or not
1458 2013-05-09 14:45:25 <cjd> a lot of people think that it is real
1459 2013-05-09 14:45:29 taha_ has joined
1460 2013-05-09 14:45:56 <hazrd> sipa, you oddball genius!!
1461 2013-05-09 14:46:06 <devurandom> Ok... So miners will stop hashing this set of transactions at some point, when they believe it becomes unrealistic that there is a nonce that can hash it below the target?
1462 2013-05-09 14:46:15 Belxjander has joined
1463 2013-05-09 14:46:17 <sipa> devurandom: no
1464 2013-05-09 14:46:22 <hazrd> i finally understand the underlying concept
1465 2013-05-09 14:46:33 <cjd> hazrd: nope
1466 2013-05-09 14:46:39 <cjd> bitcoin == russian doll
1467 2013-05-09 14:46:41 <SirDefaced> i wonder if a progress based coin would even work.
1468 2013-05-09 14:46:43 <cjd> you never understand it
1469 2013-05-09 14:46:51 taha has quit (Ping timeout: 255 seconds)
1470 2013-05-09 14:47:01 <devurandom> Blind men and an elephant?
1471 2013-05-09 14:47:03 <sipa> devurandom: every few seconds or so they look at their memory pool, pick the transactions they like from it, build a block from it, and then try to hash for a while to find a nonce that solves it
1472 2013-05-09 14:47:19 <devurandom> ok
1473 2013-05-09 14:47:26 <sipa> devurandom: they don't throw away anything by switching to a new set of transactions a few seconds later
1474 2013-05-09 14:47:30 <SirDefaced> how is nonce pronounced?
1475 2013-05-09 14:47:37 <sipa> nons
1476 2013-05-09 14:47:47 <SirDefaced> in my mind it was
1477 2013-05-09 14:47:53 <SirDefaced> newonce idky
1478 2013-05-09 14:47:59 <hazrd> cjd, but you can... the concept is that if you look deep enough, you will finally find the matroiska that isn't empty.... it's the source for all the others
1479 2013-05-09 14:48:03 <devurandom> So they dont really care how many nonces they tried - at random times they just switch to a new block?
1480 2013-05-09 14:48:10 <sipa> devurandom: yes
1481 2013-05-09 14:48:29 <sipa> as long as you are constantly trying new block+nonce combinations, you can't do better
1482 2013-05-09 14:48:38 <sipa> it doesn't matter whether they come from the same block or not
1483 2013-05-09 14:48:59 <devurandom> Ok, and when a new block is published, they have to remove the transactions that are now spent from their pool?
1484 2013-05-09 14:49:01 <hazrd> cjd, i'm being full of shit btw :P
1485 2013-05-09 14:49:06 <sipa> devurandom: indeed
1486 2013-05-09 14:49:12 <devurandom> sipa: "come from the same block"?
1487 2013-05-09 14:49:14 <devurandom> Who?
1488 2013-05-09 14:49:50 neo2 has quit (Remote host closed the connection)
1489 2013-05-09 14:49:53 <sipa> devurandom: it doesn't matter whether you try all 4 billion nonce with one particular block candidate, or try 2 billion nonces from two block candidates each
1490 2013-05-09 14:49:57 <sipa> devurandom: the chances are the same
1491 2013-05-09 14:50:26 <devurandom> So every once in a while they build a new block from the current transactions and hope they find a nonce that fits? It does not matter at all how long they have been searching and so on. They just need to make sure not to mine the same block+nonce combinations twice?
1492 2013-05-09 14:50:38 <sipa> devurandom: bing
1493 2013-05-09 14:50:40 <sipa> devurandom: bingo
1494 2013-05-09 14:50:47 <devurandom> Thanks a lot! :)
1495 2013-05-09 14:51:03 <sipa> yw
1496 2013-05-09 14:51:30 <devurandom> ymmd
1497 2013-05-09 14:52:19 PhantomSpark has quit (Ping timeout: 276 seconds)
1498 2013-05-09 14:53:12 altamic has quit (Quit: altamic)
1499 2013-05-09 14:54:00 Diapolis_ has quit (Remote host closed the connection)
1500 2013-05-09 14:55:11 n5 has quit (Ping timeout: 256 seconds)
1501 2013-05-09 14:56:11 m00p has joined
1502 2013-05-09 14:57:21 jeewee has quit (Quit: Leaving.)
1503 2013-05-09 14:58:53 XertroV has joined
1504 2013-05-09 14:58:59 XertroV has quit (Client Quit)
1505 2013-05-09 14:59:32 t7 has quit (Quit: Leaving)
1506 2013-05-09 15:02:09 TD has joined
1507 2013-05-09 15:02:16 Michail1 is now known as Michail1_
1508 2013-05-09 15:02:26 fanquake has quit (Quit: fanquake)
1509 2013-05-09 15:04:47 taha_ has quit (Ping timeout: 276 seconds)
1510 2013-05-09 15:07:49 t7 has joined
1511 2013-05-09 15:08:41 cc_8 has joined
1512 2013-05-09 15:08:41 cc_8 has quit (Client Quit)
1513 2013-05-09 15:12:12 Cryo has joined
1514 2013-05-09 15:13:04 eqolo has quit ()
1515 2013-05-09 15:15:05 Xqr has joined
1516 2013-05-09 15:16:15 owowo has joined
1517 2013-05-09 15:17:52 agricocb has joined
1518 2013-05-09 15:20:45 jevin has quit (Quit: Textual IRC Client: www.textualapp.com)
1519 2013-05-09 15:21:45 r0sc0e has joined
1520 2013-05-09 15:23:22 Rothgar has quit (Quit: Page closed)
1521 2013-05-09 15:25:56 Michail1_ is now known as Michail1
1522 2013-05-09 15:26:19 jevin has joined
1523 2013-05-09 15:29:21 XertroV has joined
1524 2013-05-09 15:33:02 <jgarzik> http://bitcoinmagazine.com/btcd-a-full-bitcoin-alternative-written-in-go/
1525 2013-05-09 15:33:34 richcollins has joined
1526 2013-05-09 15:33:43 XertroV has quit (Ping timeout: 256 seconds)
1527 2013-05-09 15:33:49 wei_ has joined
1528 2013-05-09 15:34:31 <jgarzik> another incomplete full implementation ;p
1529 2013-05-09 15:34:39 wei_ has quit (Client Quit)
1530 2013-05-09 15:34:45 <BlueMatt> heh
1531 2013-05-09 15:36:18 <BlueMatt> I have to say I'm amazed how much press they are generating for doing yet another full implementation
1532 2013-05-09 15:36:26 <BlueMatt> and not having even gotten that far afawk
1533 2013-05-09 15:37:12 <sipa> jgarzik: i don't call a serialization library "an implementation"
1534 2013-05-09 15:38:09 mrkent has joined
1535 2013-05-09 15:38:09 mrkent has quit (Changing host)
1536 2013-05-09 15:38:09 mrkent has joined
1537 2013-05-09 15:38:26 <jgarzik> I don't either, but Bitcoin Magazine apparently does ;p
1538 2013-05-09 15:38:40 <BlueMatt> " Importantly, there are no alternative implementations of Bitcoin that are currently being used to any significant degree by miners." uhhhhh...I hope not, and I sure as hell hope btcd doesnt get any mining uptake either
1539 2013-05-09 15:39:32 zz_highPriestLOL is now known as zz_zz_highPriest
1540 2013-05-09 15:39:58 richcollins has quit (Quit: richcollins)
1541 2013-05-09 15:43:34 <SomeoneWeird> so lame
1542 2013-05-09 15:45:06 alexwaters1 has joined
1543 2013-05-09 15:45:44 <jaromil> hype always brings lameness
1544 2013-05-09 15:46:15 AndChat64721 has joined
1545 2013-05-09 15:47:59 <jaromil> anyone here knows of people implementing again something like witcoin? I loved that so much that I might need to rewrite it, mpfh.
1546 2013-05-09 15:50:59 <TD> jaromil: i guess the reddit tipbot is the closest
1547 2013-05-09 15:52:32 triciam has joined
1548 2013-05-09 15:52:34 mrkent has quit (Remote host closed the connection)
1549 2013-05-09 15:52:49 mrkent has joined
1550 2013-05-09 15:53:28 <jgarzik> hah. "If Java had true garbage collection, most programs would delete themselves upon execution."
1551 2013-05-09 15:54:25 <TD> lol
1552 2013-05-09 15:54:42 <BlueMatt> so very true
1553 2013-05-09 15:55:03 MoALTz_ has joined
1554 2013-05-09 15:55:05 <sipa> ha
1555 2013-05-09 15:55:11 <The_Fly> lol
1556 2013-05-09 15:55:49 <cuqa> can someone give me the difficulty as displayed in bitcoin client pls? im trying to figure out how to calculate the target, but getting different numbers than on wiki
1557 2013-05-09 15:55:49 <jaromil> :^D - TD 10x looking at it
1558 2013-05-09 15:56:48 JTF195 has joined
1559 2013-05-09 15:56:48 <sipa> ;;diff
1560 2013-05-09 15:56:49 <gribble> 1.0076292883418716E7
1561 2013-05-09 15:56:53 <sipa> cuqa: ^
1562 2013-05-09 15:56:57 MoALTz has quit (Ping timeout: 256 seconds)
1563 2013-05-09 15:57:31 <jaromil> TD mmm tipbot is just for voluntary tips, not even sure it counts for rating up
1564 2013-05-09 15:57:52 n5 has joined
1565 2013-05-09 15:59:26 <cuqa> thanks
1566 2013-05-09 16:00:21 ovidiusoft has quit (Quit: BEREEE)
1567 2013-05-09 16:00:24 <jgarzik> quite a significant number of "non-standard transaction type" when I login
1568 2013-05-09 16:00:35 <jgarzik> 2013-05-09 15:37:05 ERROR: CTxMemPool::accept() : nonstandard transaction type
1569 2013-05-09 16:00:38 quaz0r has quit (Ping timeout: 256 seconds)
1570 2013-05-09 16:00:53 <jgarzik> (login == when my bitcoind automatically starts after login, locally)
1571 2013-05-09 16:01:07 * jgarzik needs to put together a tx-logging patch, stat
1572 2013-05-09 16:01:23 <jgarzik> If I had to guess, I would speculate these are dust
1573 2013-05-09 16:01:41 <jgarzik> obviously higher number of incidents here, anyway, with HEAD
1574 2013-05-09 16:02:09 <sipa> jgarzik: indeed, almost certainly dust
1575 2013-05-09 16:02:14 wei_ has joined
1576 2013-05-09 16:02:38 <sipa> a better error message would be nice
1577 2013-05-09 16:02:47 <jgarzik> yes
1578 2013-05-09 16:02:50 <BlueMatt> jgarzik: or a block explorer that supports grepping debug.log for interesting txn
1579 2013-05-09 16:03:05 <BlueMatt> (and a patch to dump them, obviously)
1580 2013-05-09 16:03:05 <jgarzik> BlueMatt: debug.log doesn't tell you enough for that
1581 2013-05-09 16:03:11 toffoo has quit ()
1582 2013-05-09 16:03:22 <jaromil> jgarzik: is that picocoin?
1583 2013-05-09 16:03:39 <jaromil> oh no you wrote bitcoind sry
1584 2013-05-09 16:03:52 <sipa> jgarzik: seen https://github.com/realazthat/overblock ?
1585 2013-05-09 16:04:02 <jgarzik> As discussed yesterday, minimally invasive would be a -txdebug=1 optional setting, which dumps transactions to a secondary log.
1586 2013-05-09 16:04:22 <jgarzik> i.e. full raw tx (perhaps plus time, IP, client info)
1587 2013-05-09 16:04:32 LorenzoMoney has joined
1588 2013-05-09 16:04:54 <jgarzik> sipa: neat
1589 2013-05-09 16:04:58 <BlueMatt> go for it
1590 2013-05-09 16:05:10 Diapolis has joined
1591 2013-05-09 16:05:44 wei_ has quit (Client Quit)
1592 2013-05-09 16:06:59 <jgarzik> + reason for dump (dust / script / DER / ...)
1593 2013-05-09 16:07:39 rlifchitz has quit (Ping timeout: 256 seconds)
1594 2013-05-09 16:07:48 <sipa> :( dang, my libsecp256k1 unit test fails after running for 36 minutes
1595 2013-05-09 16:08:06 <sipa> well, i guess it's good news that i have unit tests then...
1596 2013-05-09 16:10:23 <BlueMatt> sipa: yay testing!
1597 2013-05-09 16:11:09 <realazthat> sipa: hi
1598 2013-05-09 16:11:53 <helo> what version of ubuntu is used to build for releases?
1599 2013-05-09 16:12:03 <BlueMatt> lucid
1600 2013-05-09 16:12:14 <sipa> realazthat: hi!
1601 2013-05-09 16:14:00 davout has quit (Remote host closed the connection)
1602 2013-05-09 16:14:27 quaz0r has joined
1603 2013-05-09 16:14:52 LainZ has joined
1604 2013-05-09 16:16:26 TD has quit (Quit: TD)
1605 2013-05-09 16:17:22 Diapolis_ has joined
1606 2013-05-09 16:18:18 <sipa> heh, how do i search the issues list on github?
1607 2013-05-09 16:18:30 <sipa> i would have sworn there was a search box somewhere at the top of the page
1608 2013-05-09 16:18:42 <BlueMatt> definitely used to be
1609 2013-05-09 16:18:57 <BlueMatt> yes, very very top of the pate
1610 2013-05-09 16:18:58 TD has joined
1611 2013-05-09 16:18:58 <BlueMatt> ge
1612 2013-05-09 16:19:25 <BlueMatt> or https://github.com/bitcoin/bitcoin/search
1613 2013-05-09 16:19:31 rlifchitz has joined
1614 2013-05-09 16:19:41 <wumpus> there used to be a search box in the issues list itself, but now there is only the general search at the top
1615 2013-05-09 16:19:55 <realazthat> wumpus: hi
1616 2013-05-09 16:19:56 <wumpus> I've already had my huh?! moment with that too
1617 2013-05-09 16:20:03 <wumpus> hello realazthat
1618 2013-05-09 16:20:22 <realazthat> wumpus: you doing anything interesting with bitcoin?
1619 2013-05-09 16:20:28 Diapolis has quit (Ping timeout: 256 seconds)
1620 2013-05-09 16:20:31 <realazthat> you seem .... not idle
1621 2013-05-09 16:20:50 <wumpus> realazthat: well, I hack on it sometimes, usually the gui stuff :)
1622 2013-05-09 16:20:55 <realazthat> ah cool
1623 2013-05-09 16:21:10 <realazthat> well, nice to see you :D
1624 2013-05-09 16:21:16 <realazthat> not idling
1625 2013-05-09 16:21:17 LorenzoMoney has quit (Ping timeout: 246 seconds)
1626 2013-05-09 16:21:38 <wumpus> and you?
1627 2013-05-09 16:21:45 <realazthat> mmm
1628 2013-05-09 16:21:53 <realazthat> I done this: https://github.com/realazthat/overblock
1629 2013-05-09 16:21:59 <realazthat> gonna try and improve it a bit
1630 2013-05-09 16:23:11 <wumpus> looks like a useful concept
1631 2013-05-09 16:23:22 <wumpus> a local block explorer
1632 2013-05-09 16:23:32 <hazrd> why is .gov.us so full of crap with regard to cryptography!?
1633 2013-05-09 16:23:52 <hazrd> i can't get a pre-compiled openssl lib with ECC for centos/redhat
1634 2013-05-09 16:23:56 <hazrd> ARGH!
1635 2013-05-09 16:24:32 <realazthat> wumpus: if you have any other useful ideas ...
1636 2013-05-09 16:24:43 richcollins has joined
1637 2013-05-09 16:25:36 Diapolis_ has quit (Ping timeout: 245 seconds)
1638 2013-05-09 16:25:42 Diapolis has joined
1639 2013-05-09 16:25:53 <sipa> realazthat: wumpus is the primary author of the Bitcoin-Qt GUI :)
1640 2013-05-09 16:26:43 <sipa> BlueMatt: ok, and how do i search the issues list with that? :S
1641 2013-05-09 16:26:45 <realazthat> sipa: oh cool
1642 2013-05-09 16:26:54 <wumpus> realazthat: well it would indeed be nice to be able to browse the content of blocks (and look up the "balance" for arbitrary addresses) without consulting a centralized service, which also crashes half of the time :)
1643 2013-05-09 16:26:57 <realazthat> sipa: haha, I know him from another channel
1644 2013-05-09 16:27:23 <wumpus> sipa: you can search in the titles of issues with it, but it seems more limited than the old search
1645 2013-05-09 16:27:35 <wumpus> maybe there's some special commands you can enter into the search box though...
1646 2013-05-09 16:28:35 LorenzoMoney has joined
1647 2013-05-09 16:28:52 <BlueMatt> sipa: umm...click the issues button on the side and type something in the search box?
1648 2013-05-09 16:29:01 <BlueMatt> (using the /search link)
1649 2013-05-09 16:30:04 XertroV has joined
1650 2013-05-09 16:31:36 <wumpus> BlueMatt: ah now I get it, thanks https://github.com/bitcoin/bitcoin/search?type=Issues
1651 2013-05-09 16:32:43 <sipa> wumpus: heh, how did you find that?
1652 2013-05-09 16:32:53 gfinn has joined
1653 2013-05-09 16:32:54 * BlueMatt just linked it?
1654 2013-05-09 16:33:22 rdymac has quit (Read error: Connection reset by peer)
1655 2013-05-09 16:33:29 <sipa> ah, there's a link from the generic bitcoin/bitcoin search to only issues
1656 2013-05-09 16:33:43 <BlueMatt> <BlueMatt> sipa: umm...click the issues button on the side and type something in the search box?
1657 2013-05-09 16:33:44 <BlueMatt> <BlueMatt> (using the /search link)
1658 2013-05-09 16:33:52 <BlueMatt> :0
1659 2013-05-09 16:33:54 <BlueMatt> :)
1660 2013-05-09 16:34:16 * sipa needs to learn to read
1661 2013-05-09 16:34:21 XertroV has quit (Ping timeout: 252 seconds)
1662 2013-05-09 16:34:25 nsillik has joined
1663 2013-05-09 16:34:25 <BlueMatt> welcome to the club
1664 2013-05-09 16:34:26 <sipa> hah: http://www.shirtoshi.com/shirt.php?id=6
1665 2013-05-09 16:34:44 <BlueMatt> hah
1666 2013-05-09 16:35:58 maximeb has joined
1667 2013-05-09 16:36:48 <sipa> (J)ohn Lennon, (P)aul McCartney, (G)eorge Harrison, Ringo (S)tarr ?
1668 2013-05-09 16:38:09 <wumpus> hehe
1669 2013-05-09 16:40:29 rdymac has joined
1670 2013-05-09 16:43:28 matas has quit (Ping timeout: 264 seconds)
1671 2013-05-09 16:43:48 michagogo has joined
1672 2013-05-09 16:49:10 <gmaxwell> 08:44 < sipa> :( dang, my libsecp256k1 unit test fails after running for 36 minutes
1673 2013-05-09 16:49:17 <gmaxwell> \O/ Hurray for testing!
1674 2013-05-09 16:49:24 lolcookie has joined
1675 2013-05-09 16:49:36 <BlueMatt> gmaxwell: I had the exact same response
1676 2013-05-09 16:50:36 LainZ has quit (Ping timeout: 255 seconds)
1677 2013-05-09 16:50:48 TD has quit (Quit: TD)
1678 2013-05-09 16:51:36 seeingidog__ has joined
1679 2013-05-09 16:52:11 matas has joined
1680 2013-05-09 16:52:29 SirDefaced has quit (Ping timeout: 256 seconds)
1681 2013-05-09 16:54:26 TD has joined
1682 2013-05-09 16:55:17 phpwn has joined
1683 2013-05-09 16:56:04 TD has quit (Client Quit)
1684 2013-05-09 16:57:01 n5 has quit (Ping timeout: 256 seconds)
1685 2013-05-09 17:00:07 saivann has quit ()
1686 2013-05-09 17:00:23 petertodd has quit (Quit: leaving)
1687 2013-05-09 17:00:50 petertodd has joined
1688 2013-05-09 17:01:00 jrra has joined
1689 2013-05-09 17:02:39 saivann has joined
1690 2013-05-09 17:06:25 <NxTitle> libsecp256k1?
1691 2013-05-09 17:08:01 zz_zz_highPriest is now known as highPriestLOL
1692 2013-05-09 17:08:17 zoinky has joined
1693 2013-05-09 17:09:05 vigilyn2 has quit (Read error: Connection reset by peer)
1694 2013-05-09 17:09:13 vigilyn2 has joined
1695 2013-05-09 17:09:32 m00p has quit (Quit: Leaving)
1696 2013-05-09 17:11:03 Diapolo has joined
1697 2013-05-09 17:14:27 czaanja has quit (Ping timeout: 255 seconds)
1698 2013-05-09 17:16:12 <Edward_Black> A quick question - when bitcoin hashes the block twice, does that actually mean 2 rounds of SHA256, or 2*(recommended number of SHA256 rounds), which would amount to 128 rounds of SHA256 total ?
1699 2013-05-09 17:16:43 <sipa> it's SHA256(SHA256(blockheader))
1700 2013-05-09 17:17:31 <sipa> since the blockheader is 80 bytes, that means the inner SHA256 processes two data blocks (it uses 64 bytes per input data block), and the processing of each input data block needs 64 rounds
1701 2013-05-09 17:17:42 phpwn has quit (Read error: Connection reset by peer)
1702 2013-05-09 17:18:07 <sipa> the ouput is 32 bytes, so the outer SHA256 only processes one input data block
1703 2013-05-09 17:18:14 <sipa> i.e., 192 rounds in total
1704 2013-05-09 17:19:03 <Edward_Black> Thanks
1705 2013-05-09 17:19:44 hazrd has quit (Quit: plik)
1706 2013-05-09 17:19:56 caput has joined
1707 2013-05-09 17:20:03 <sipa> NxTitle: my own efficient implementation of ecdsa/secp256k1
1708 2013-05-09 17:21:25 <NxTitle> ahh, I see
1709 2013-05-09 17:22:07 FredEE has joined
1710 2013-05-09 17:23:39 matas has quit (Read error: Operation timed out)
1711 2013-05-09 17:23:42 nsillik_ has joined
1712 2013-05-09 17:23:59 melvster has joined
1713 2013-05-09 17:24:31 savetheinternet has quit (Quit: Computer has gone to sleep.)
1714 2013-05-09 17:25:49 drizztbsd has quit (Read error: Connection reset by peer)
1715 2013-05-09 17:26:28 SirDefaced has joined
1716 2013-05-09 17:27:26 nsillik has quit (Ping timeout: 268 seconds)
1717 2013-05-09 17:27:26 nsillik_ is now known as nsillik
1718 2013-05-09 17:28:16 rainworm has quit ()
1719 2013-05-09 17:28:24 uyppp has joined
1720 2013-05-09 17:28:58 melvster has quit (Ping timeout: 276 seconds)
1721 2013-05-09 17:29:12 highPriestLOL is now known as zz_highPriestLOL
1722 2013-05-09 17:29:37 paybitcoin has quit (Ping timeout: 276 seconds)
1723 2013-05-09 17:29:54 brson has joined
1724 2013-05-09 17:30:08 savetheinternet has joined
1725 2013-05-09 17:30:12 zz_highPriestLOL is now known as zz_zz_highPriest
1726 2013-05-09 17:30:21 XertroV has joined
1727 2013-05-09 17:30:21 alphaguru has quit ()
1728 2013-05-09 17:32:07 <anarchy5> sipa: isnt the input work block always 512 bits, just padded with constants, plus not all 80 bytes of the header are hashed (just a few of them)
1729 2013-05-09 17:32:31 <skinnkavaj> https://blog.conformal.com/btcd-a-bitcoind-alternative-written-in-go/ what do you think of this?
1730 2013-05-09 17:34:50 nomailin1 has quit (Ping timeout: 268 seconds)
1731 2013-05-09 17:34:51 XertroV has quit (Ping timeout: 252 seconds)
1732 2013-05-09 17:34:58 <sipa> anarchy5: input block is indeed always 512 bits, so the 80 bytes input are padded to 128 bytes before feeding to the block processing function
1733 2013-05-09 17:35:05 <sipa> anarchy5: and all 80 bytes of the header are hashed
1734 2013-05-09 17:35:12 <sipa> skinnkavaj: too early to judge
1735 2013-05-09 17:36:12 <gmaxwell> skinnkavaj: it's currently only deserialization code.
1736 2013-05-09 17:36:23 <nsh> sipa, what's a sane strategy to a/b test new implementations against the reference bitcoind?
1737 2013-05-09 17:36:37 <nsh> is there some protocol test suite, or could one be devised easily?
1738 2013-05-09 17:36:51 <nsh> (et al)
1739 2013-05-09 17:37:01 <gmaxwell> nsh: bluematt created an agreement tester.
1740 2013-05-09 17:37:12 <gmaxwell> It's far from perfect, but it tests a lot.
1741 2013-05-09 17:37:39 <nsh> right on
1742 2013-05-09 17:37:55 HaltingState has quit (Ping timeout: 268 seconds)
1743 2013-05-09 17:38:09 <nsh> would it make sense to abstract the protocol to a BNF specification or something similar?
1744 2013-05-09 17:38:15 <gmaxwell> No.
1745 2013-05-09 17:38:18 <Diablo-D3> [01:09:09] <anarchy5> sipa: isnt the input work block always 512 bits, just padded with constants, plus not all 80 bytes of the header are hashed (just a few of them)
1746 2013-05-09 17:38:21 <Diablo-D3> learn how sha256 works kthx
1747 2013-05-09 17:38:24 <nsh> i've no experience in this department
1748 2013-05-09 17:38:29 nomailing1 has joined
1749 2013-05-09 17:38:38 rdymac has quit (Quit: This computer has gone to sleep)
1750 2013-05-09 17:38:42 <gmaxwell> nsh: the wire protocol itself is pretty uninteresting. It's not essential to bitcoin cat all.
1751 2013-05-09 17:38:44 Diapolo has left ()
1752 2013-05-09 17:38:47 <gmaxwell> s/cat/at/
1753 2013-05-09 17:38:55 <Diablo-D3> BITCOIN CAT
1754 2013-05-09 17:38:56 kicek has quit (Ping timeout: 245 seconds)
1755 2013-05-09 17:40:27 <anarchy5> Diablo-D3: i was talking about mining, not hashing the merkle tree, wrong context
1756 2013-05-09 17:40:29 ThomasV has joined
1757 2013-05-09 17:40:30 <gmaxwell> nsh: the important and hard part is the distributed consensus algorithimâ a pure mathmatical function that when presented with sequences of blocks, determines their validity, and selects the unique best chain.
1758 2013-05-09 17:40:39 <nsh> mmm
1759 2013-05-09 17:40:45 <Diablo-D3> anarchy5: I wrote a miner. Your turn.
1760 2013-05-09 17:40:46 * nsh nods
1761 2013-05-09 17:41:31 <nsh> but there is perhaps some analogous abstract way to specify this that is language-independent which would facilitate reimplementation in new code
1762 2013-05-09 17:42:10 <nsh> hmm, perhaps not one with any particular currency...
1763 2013-05-09 17:42:31 <Diablo-D3> nsh: gmaxwell was right about the wire protocol not being essential
1764 2013-05-09 17:42:37 <Diablo-D3> neither is the on disk format
1765 2013-05-09 17:42:40 <Diablo-D3> its just a set of rules
1766 2013-05-09 17:43:17 <gmaxwell> nsh: you can certantly write it in other ways ... but "would facilitate reimplementation in new code" is less clearâ I mean, it is code.
1767 2013-05-09 17:43:34 toffoo has joined
1768 2013-05-09 17:44:00 <gmaxwell> nsh: so, e.g. you could write an expression of it that could be translated into other languages, but the trainslations must be faithful.
1769 2013-05-09 17:44:15 n5 has joined
1770 2013-05-09 17:44:36 <nsh> ok
1771 2013-05-09 17:45:27 Diapolis has quit (Read error: Connection reset by peer)
1772 2013-05-09 17:45:46 Diapolis has joined
1773 2013-05-09 17:46:27 <nsh> i think i was straining towards something like PlusCal, but that does not appear to be a very mature language/project. it seems to me there ought to be better standardised ways to specify in a language-agnostic fashion the essentials of an algorithm/algorithm-set
1774 2013-05-09 17:46:28 <gmaxwell> and part of the problem is that the behavior of openssl and potentially some behavior of leveldb, potentially some behavior of boost, etc. is incorporated by reference currently.
1775 2013-05-09 17:46:35 savetheinternet has quit (Quit: Textual IRC Client: www.textualapp.com)
1776 2013-05-09 17:46:40 <nsh> that would allow for easy testing
1777 2013-05-09 17:46:54 <nsh> right
1778 2013-05-09 17:47:00 jeewee has joined
1779 2013-05-09 17:47:16 <sipa> so we just need a bitcoin-block-validation Turing machine
1780 2013-05-09 17:47:18 <sipa> simple!
1781 2013-05-09 17:47:21 <gmaxwell> nsh: "easy testing" I mean, fuck testing, it actually has to be right. It's very difficult to prove that two programs are the same function.
1782 2013-05-09 17:48:10 <nsh> gmaxwell, yeah, i think i may have been optimistically ignoring the halting problem. it occasionally happens
1783 2013-05-09 17:48:57 kicek has joined
1784 2013-05-09 17:50:09 HaltingState has joined
1785 2013-05-09 17:50:24 caedes has joined
1786 2013-05-09 17:50:24 caedes has quit (Changing host)
1787 2013-05-09 17:50:24 caedes has joined
1788 2013-05-09 17:50:53 <sipa> gmaxwell, wumpus: hmm, how useful is mlocking() data structures that get allocated on the stack?
1789 2013-05-09 17:52:40 LorenzoMoney has left ("Ciao!")
1790 2013-05-09 17:52:58 <wumpus> why would that be less useful than mlocking structures on the heap?
1791 2013-05-09 17:53:18 * helo gives up trying to build statically
1792 2013-05-09 17:53:25 <gmaxwell> sipa: I don't know any reason that sections of the stack couldn't get swapped out... but it seems kinda pointless, in particular the tlb twiddling is going to kill any performance advantage of having the stuff on the stack.
1793 2013-05-09 17:54:22 <gmaxwell> I'm generally of the view that we ought to just preallocate mlocked heap at start and do all our mlocked stuff out of it... esp since mlock can fail at runtime.
1794 2013-05-09 17:54:39 <wumpus> yes, that'd be better, to have an mlocked memory pool
1795 2013-05-09 17:55:03 <sipa> wumpus: it's certainly useful, but given that the stack is a pretty much constantly-accessed memory area (at least while the given thread is executing), it shouldn't really be swapped out
1796 2013-05-09 17:55:12 n5 has quit (Ping timeout: 252 seconds)
1797 2013-05-09 17:55:39 <wumpus> but boost::pool isn't suitable for it, I tried it back in the day
1798 2013-05-09 17:55:55 <sipa> but under high memory pressure i suppose even the stack can get swapped out
1799 2013-05-09 17:55:58 <wumpus> so we'd need to write an custom allocator or find some other custom malloc library...
1800 2013-05-09 17:56:16 <wumpus> indeed it's not impossible
1801 2013-05-09 17:56:29 <wumpus> especially older pages
1802 2013-05-09 17:56:40 <wumpus> way up the stack
1803 2013-05-09 17:56:54 <sipa> i think most data structures that need locking are a) ~ the same size and b) allocated in a stack-like fashion
1804 2013-05-09 17:57:07 <gmaxwell> wumpus: mlockpool->ptr mlockpool->off
1805 2013-05-09 17:57:08 <sipa> so a very simple memory allocator may suffice
1806 2013-05-09 17:57:13 <gmaxwell> yea, a stack.
1807 2013-05-09 17:57:30 <wumpus> hmm are you sure? what about all the private keys
1808 2013-05-09 17:57:40 <wumpus> I don't think the allocation pattern is a simple stack
1809 2013-05-09 17:57:50 <sipa> it isn't exactly one
1810 2013-05-09 17:58:15 <sipa> but you have a few very-long-lived ones, and a few very-shorted-lived ones, afaik
1811 2013-05-09 17:58:16 <wumpus> but yes, for the stack-allocated structures it obviously is :)
1812 2013-05-09 17:58:31 <wumpus> that's true
1813 2013-05-09 17:58:50 <gmaxwell> wumpus: the question is basically how much overhead you get if you make free only free the tail. At least I don't think we have any interleaved lifetimes.
1814 2013-05-09 17:59:07 resinate has quit (Quit: resinate)
1815 2013-05-09 17:59:28 <wumpus> private keys are never destroyed... so you're probably right
1816 2013-05-09 17:59:33 <sipa> in my key refactor i made CSecret constant-size (a wrapper around unsigned char data[32]), and added some functions to allocators.h to have it lock its data array at construction and unlock at destruction
1817 2013-05-09 18:00:02 zoinky has quit (Quit: Leaving.)
1818 2013-05-09 18:00:48 <sipa> i suppose that it doesn't really have much of a performance issue, as operations with them are not frequent
1819 2013-05-09 18:00:59 <sipa> except when doing a wallet encryption, i guess
1820 2013-05-09 18:01:21 MoALTz_ is now known as MoALTz
1821 2013-05-09 18:02:49 <wumpus> using a fixed-size pool adds an extra tuning knob (how big to make the pool) and also the decision what to do if the pool overflows (ie, allocate a new pool? we don't want it to fail with a lot of keys)
1822 2013-05-09 18:03:11 <wumpus> indeed, let's first be sure that it is a performance issue at all
1823 2013-05-09 18:03:54 mrkent has quit (Remote host closed the connection)
1824 2013-05-09 18:04:04 FredEE has left ()
1825 2013-05-09 18:04:23 <petertodd> "Erik has said that off-the-blockchain betting is coming soon to SDICE too." https://bitcointalk.org/index.php?topic=77870.msg2079045#msg2079045
1826 2013-05-09 18:04:26 <petertodd> Nice!
1827 2013-05-09 18:04:48 <helo> victory
1828 2013-05-09 18:05:06 zoinky has joined
1829 2013-05-09 18:05:14 resinate has joined
1830 2013-05-09 18:05:18 <petertodd> Heh, I wonder how much my replace-by-fee thing encouraged him.
1831 2013-05-09 18:06:39 <helo> maybe some day future devs will name a checkpoint after you
1832 2013-05-09 18:07:57 <nsh> +1
1833 2013-05-09 18:07:58 <petertodd> helo: I've always thought we should auction off checkpoint naming rights, and wear NASCAR style sponsorship suits.
1834 2013-05-09 18:08:11 <wumpus> hehe
1835 2013-05-09 18:08:14 <sipa> _naming_ checkpoints?
1836 2013-05-09 18:08:21 <sipa> i was hoping to just get rid of them...
1837 2013-05-09 18:09:12 <helo> heh
1838 2013-05-09 18:09:13 <wumpus> well we could name the releases...
1839 2013-05-09 18:10:31 <sipa> Bulky Blockchain, Cluttered Code, Dodgy Daemon, ...
1840 2013-05-09 18:10:41 <wumpus> :D
1841 2013-05-09 18:11:32 <nsh> hehe
1842 2013-05-09 18:13:18 da2ce7_d has joined
1843 2013-05-09 18:15:43 da2ce7 has quit (Ping timeout: 256 seconds)
1844 2013-05-09 18:15:43 <wumpus> Economic Error, Faulty Fee, Goofy GUI, ...
1845 2013-05-09 18:15:58 zoinky has quit (Quit: Leaving.)
1846 2013-05-09 18:17:22 <petertodd> sipa: Deflationary Daemon
1847 2013-05-09 18:17:27 * petertodd ducks
1848 2013-05-09 18:17:30 <sipa> Addictive Activity, Hashing Hoard, ...
1849 2013-05-09 18:18:02 <petertodd> Profitable PoW
1850 2013-05-09 18:18:28 <sipa> Meager Mining
1851 2013-05-09 18:18:56 saracen has quit (Ping timeout: 245 seconds)
1852 2013-05-09 18:19:04 <kinlo> uhmz
1853 2013-05-09 18:19:16 <petertodd> kinlo: oh, that's catchy
1854 2013-05-09 18:19:17 <kinlo> I'm pro!
1855 2013-05-09 18:19:27 * sipa feels he may not be fit for a PR job
1856 2013-05-09 18:19:28 <pigeons> Shrinking Subsidy
1857 2013-05-09 18:19:47 <kinlo> well, we could use other cool names
1858 2013-05-09 18:20:13 <sipa> i have no problem with release (code)names, but they have to be a bit original :)
1859 2013-05-09 18:20:17 <petertodd> kinlo: how about "liquid nitrogen"?
1860 2013-05-09 18:20:44 <kinlo> you could do it like the linux kernel does
1861 2013-05-09 18:20:49 zooko_ has quit (Read error: Operation timed out)
1862 2013-05-09 18:21:24 <cjd> Amshel, Biddle, Morgan, Hamelton...
1863 2013-05-09 18:21:30 <petertodd> sipa: broken hash functions? MD5, SHA...
1864 2013-05-09 18:21:37 * petertodd may also not be suitable to do PR
1865 2013-05-09 18:21:37 <sipa> petertodd: haha
1866 2013-05-09 18:22:10 zoinky has joined
1867 2013-05-09 18:22:29 <gmaxwell> no no, you can't name the checkpoints. They can't go away if they have names, it would be like killing kittens.
1868 2013-05-09 18:22:39 <sipa> Killing Kittens.
1869 2013-05-09 18:22:44 <kinlo> mmmz
1870 2013-05-09 18:22:56 <kinlo> Unicycling Gorilla <- anyone can guess which project? :p
1871 2013-05-09 18:22:58 mode80_ has joined
1872 2013-05-09 18:23:03 <sipa> nope
1873 2013-05-09 18:23:13 saracen has joined
1874 2013-05-09 18:23:29 <kinlo> see, you can put such fun in without having to name your entire distribution like that :)
1875 2013-05-09 18:23:41 <kinlo> linux 3.9.0 is named Unicycling Gorilla :)
1876 2013-05-09 18:24:00 <sipa> 3.8, it seems
1877 2013-05-09 18:24:10 <petertodd> actually, you guys are doing it all wrong, see, we need to promote anonymity, which means we should name every release Satoshi
1878 2013-05-09 18:24:20 <petertodd> Satoshi, Satoshi, Satoshi...
1879 2013-05-09 18:24:24 <gmaxwell> hahahahaa
1880 2013-05-09 18:24:39 <wumpus> all releases are created equal?
1881 2013-05-09 18:24:39 <MoALTz> dustware
1882 2013-05-09 18:24:50 <petertodd> wumpus: per unit hashing power yes
1883 2013-05-09 18:24:51 <sipa> perhaps add a varying adjective in front of it
1884 2013-05-09 18:25:02 <gmaxwell> does warner bros sell trademarked satoshi masks for us to wear at the bitcoin conference?
1885 2013-05-09 18:25:14 <petertodd> reminds me, we should all get "Hi! My name is" nametags for the conference and write satoshi on every last one
1886 2013-05-09 18:25:20 <kinlo> http://tech-specialist.info/ubuntu/ :)
1887 2013-05-09 18:25:24 <petertodd> gmaxwell: lol!
1888 2013-05-09 18:25:28 <wumpus> hahaha
1889 2013-05-09 18:25:43 <kinlo> hehe :)
1890 2013-05-09 18:26:05 daybyter has joined
1891 2013-05-09 18:26:38 <gmaxwell> Just don't send peel off stickers through a laser printer....
1892 2013-05-09 18:26:50 <cjd> lol
1893 2013-05-09 18:26:51 <kinlo> :)
1894 2013-05-09 18:26:56 mode80_ has left ()
1895 2013-05-09 18:27:05 <petertodd> gmaxwell: I thought they made special ones for that...
1896 2013-05-09 18:27:17 <gmaxwell> yea, but they're made out of pure lies.
1897 2013-05-09 18:27:28 <petertodd> what's the melting poitn of lies?
1898 2013-05-09 18:27:33 <gmaxwell> "Doesn't come unstuck and stick to the fuser ... nine out of ten times"
1899 2013-05-09 18:27:39 <kinlo> :)
1900 2013-05-09 18:30:52 XertroV has joined
1901 2013-05-09 18:31:14 <MC1984> ever seen a kid put an a4 sheet of sticky back plastic thru a laser printer?
1902 2013-05-09 18:31:16 <MC1984> i have
1903 2013-05-09 18:31:28 <MC1984> it was scrap
1904 2013-05-09 18:32:02 <nsh> ooft
1905 2013-05-09 18:32:47 forrestv has quit (Quit: ZNC - http://znc.sourceforge.net)
1906 2013-05-09 18:33:32 <MC1984> it was the art departments tri colour expensive edition laser printer too
1907 2013-05-09 18:33:35 <MC1984> hilarity ensued
1908 2013-05-09 18:33:47 mollison has joined
1909 2013-05-09 18:35:18 XertroV has quit (Ping timeout: 252 seconds)
1910 2013-05-09 18:36:09 mode80_ has joined
1911 2013-05-09 18:36:14 Elmf has joined
1912 2013-05-09 18:36:28 santoscork has joined
1913 2013-05-09 18:37:02 gavinandresen has joined
1914 2013-05-09 18:37:02 gavinandresen has quit (Changing host)
1915 2013-05-09 18:37:02 gavinandresen has joined
1916 2013-05-09 18:38:31 savetheinternet has joined
1917 2013-05-09 18:40:57 freewil has quit (Read error: Connection reset by peer)
1918 2013-05-09 18:41:14 AndChat64721 has quit (Ping timeout: 276 seconds)
1919 2013-05-09 18:41:28 freewil has joined
1920 2013-05-09 18:42:10 mode80_ has quit (Quit: Page closed)
1921 2013-05-09 18:42:21 fishfish has joined
1922 2013-05-09 18:42:58 mode80 has joined
1923 2013-05-09 18:43:26 ThomasV has quit (Ping timeout: 256 seconds)
1924 2013-05-09 18:44:00 santoscork has quit (Ping timeout: 255 seconds)
1925 2013-05-09 18:44:10 LainZ has joined
1926 2013-05-09 18:44:12 santoscork has joined
1927 2013-05-09 18:45:35 CaptainBlaze has quit (Quit: CaptainBlaze)
1928 2013-05-09 18:46:31 Anduck has joined
1929 2013-05-09 18:46:32 Anduck has quit (Changing host)
1930 2013-05-09 18:46:32 Anduck has joined
1931 2013-05-09 18:47:36 forrestv_ has joined
1932 2013-05-09 18:48:06 forrestv_ is now known as forrestv
1933 2013-05-09 18:48:18 forrestv has quit (Changing host)
1934 2013-05-09 18:48:18 forrestv has joined
1935 2013-05-09 18:49:08 iwilcox_ has joined
1936 2013-05-09 18:49:13 agricocb has quit (Ping timeout: 256 seconds)
1937 2013-05-09 18:49:40 iwilcox has quit (Ping timeout: 256 seconds)
1938 2013-05-09 18:49:42 zoinky has quit (Quit: Leaving.)
1939 2013-05-09 18:51:43 <MoALTz> if you had 2 proof-of-worked items, A and B, where B is inside A, would it be reasonable to add the the estimates of the work done on both individually together to give a total "work done" score? would it be vulnerable to being gamed?
1940 2013-05-09 18:52:06 <MoALTz> i'm worried that such a scheme might degenerate to "fastest cpu wins"
1941 2013-05-09 18:52:25 <MoALTz> (but i suspect that it won't)
1942 2013-05-09 18:52:40 richcollins has quit (Quit: richcollins)
1943 2013-05-09 18:53:49 mrkent has joined
1944 2013-05-09 18:55:13 defunctzombie_zz is now known as defunctzombie
1945 2013-05-09 18:55:31 <MoALTz> my idea is that people who send transactions could do work on them (incrementing a nonce) to get a low transaction hash and then having the required tx fee reduced accordingly. miners could be given an incentive to include such tx by lowering the difficulty of the target that they have to find a solution for
1946 2013-05-09 18:55:42 <MoALTz> resurrecting hashmail again
1947 2013-05-09 18:56:58 <MoALTz> you could go to an extreme: low enough tx hashes giving negative fees (the input account gets paid).
1948 2013-05-09 18:57:06 melvster has joined
1949 2013-05-09 18:57:10 <helo> what is progress=.... in SetBestChain?
1950 2013-05-09 18:57:37 Diapolis has quit (Remote host closed the connection)
1951 2013-05-09 18:58:18 <wumpus> Checkpoints::GuessVerificationProgress(pindexBest)
1952 2013-05-09 18:58:43 <mollison> MoALTz: transaction fees are good because if they resolve which transactions get into the block if there is contention for space, and eventually they will help incentivize miners
1953 2013-05-09 18:58:47 melvster1 has joined
1954 2013-05-09 18:58:54 <mollison> *delete that first if
1955 2013-05-09 18:59:01 <wumpus> it's a guess how far the block chain verification progress is from 0 to 1
1956 2013-05-09 18:59:30 AndChat64721 has joined
1957 2013-05-09 18:59:52 <helo> so it depends on how many transactions have already been seen and accepted?
1958 2013-05-09 18:59:54 <MoALTz> having tx being able to contribute to proof-of-work would mean that ordinary nodes that do it will help keep the hash rate done generating block high (because they'll be doing it too to lower their tx fee)
1959 2013-05-09 19:00:09 * MoALTz will look at this idea more in his own time
1960 2013-05-09 19:00:22 <mollison> MoALTz: i suspect you could not use a proof of work scheme to prevent flooding the bitcoin network because the amount of work needed to make that effective would make it hard for regular users to send transactions
1961 2013-05-09 19:00:31 <helo> wumpus: it is sometimes > 1 0_o
1962 2013-05-09 19:00:56 richcollins has joined
1963 2013-05-09 19:01:14 <wumpus> then you're very up to date, or it's a causality violation and you're actually in the future
1964 2013-05-09 19:01:24 melvster has quit (Ping timeout: 256 seconds)
1965 2013-05-09 19:04:09 <MoALTz> mollison: the scheme i imagine is: you want to send a payment you include a fee but a certain (likely small, but possibly larger if you're lucky) amount gets refunded to the sender account; you generate a transaction body (including the fee), then do a nonce increment run on it for a user-defined amount of time. select the nonce with the lowest hash, wrap up that transaction, sign and send it to the network. the miner then
1966 2013-05-09 19:04:09 <MoALTz> gets to choose using their own reasoning whether they want high transaction fees, or tranactions with (luckily) suitable hashes to reduce their effective block target
1967 2013-05-09 19:04:23 <MoALTz> sorry. i'll write up a proposal when i've looked at it more
1968 2013-05-09 19:05:19 <mollison> MoALTz: basically you are proposing a scheme where people can "help" the miners
1969 2013-05-09 19:05:37 BTCOxygen has quit (Ping timeout: 252 seconds)
1970 2013-05-09 19:05:41 <MoALTz> yes. in leiu of fees
1971 2013-05-09 19:05:48 <MoALTz> *instead of
1972 2013-05-09 19:05:51 <mollison> MoALTz: BTW, miners would always choose to "take the help" because the miners are basically all racing each other
1973 2013-05-09 19:06:18 <MoALTz> it depends on the calculation of the contribution that the transaction work gives
1974 2013-05-09 19:06:20 agricocb has joined
1975 2013-05-09 19:06:37 <mollison> i don't really see the point of doing something like this... and it probably causes problems anyway
1976 2013-05-09 19:06:45 <rdponticelli> MoALTz: Feels like a nightmare...
1977 2013-05-09 19:06:45 <MoALTz> you want to make it so that it is always more profitable for a transaction sender to just mine than to try and get a partial refund
1978 2013-05-09 19:07:16 <mollison> the amount of work a transaction sender does relative to the overall difficulty will be truly negligible
1979 2013-05-09 19:07:33 <mollison> unless the transaction sender has invested in mining hardware / is actually a miner
1980 2013-05-09 19:07:40 <rdponticelli> Probably something so overcomplicated can be full of bugs
1981 2013-05-09 19:08:02 CaptainBlaze has joined
1982 2013-05-09 19:08:02 richcollins has quit (Quit: richcollins)
1983 2013-05-09 19:08:19 <mollison> so basically it reduces to "pay a miner to mine against my transaction to make my transaction cheaper" which equals out to naught in the end
1984 2013-05-09 19:08:41 <mollison> because you'd have to pay the miner the same amount that you save by doing a proof of work as part of the transaction
1985 2013-05-09 19:09:28 <helo> MoALTz: so the difficulty required could go down based on the work performed in transactions?
1986 2013-05-09 19:09:37 ngc0202 has joined
1987 2013-05-09 19:09:44 <mollison> helo: yes that is what he is saying
1988 2013-05-09 19:09:51 <MoALTz> helo: yes
1989 2013-05-09 19:10:31 resinate has quit (Read error: Connection reset by peer)
1990 2013-05-09 19:10:38 <helo> seems pretty convoluted... what problem does this solve?
1991 2013-05-09 19:10:53 <mollison> MoALTz: the user could get the equivalent financial benefit by running mining pool software on their system
1992 2013-05-09 19:11:01 mode80 has quit (Quit: Textual IRC Client: www.textualapp.com)
1993 2013-05-09 19:11:11 <MoALTz> poorer wallet owners trying to send with fewer fees (run your computer overnight on a transaction you want to send)
1994 2013-05-09 19:11:22 resinate has joined
1995 2013-05-09 19:11:49 <mollison> MoALTz: that is horrible user experience. If we ever get to a point where fees are that expensive, poor users need to be using bitcoin banks, not the blockchain
1996 2013-05-09 19:11:51 <cjd> random thought
1997 2013-05-09 19:12:00 <cjd> Bitcoin messages should all be signed
1998 2013-05-09 19:12:11 <mollison> MoALTz: e.g. bank of mollison... i can atomically transfer bitcoins between two accounts :)
1999 2013-05-09 19:12:17 <cjd> and should all contain the sender's IP address (which is told to them by the recipient)
2000 2013-05-09 19:12:43 JDuke128 has joined
2001 2013-05-09 19:12:45 <cjd> because then an exploit message can be relayed to EG: dns seeds
2002 2013-05-09 19:13:06 <helo> cjd: what if nodoby knows the ip address?
2003 2013-05-09 19:13:13 skinnkavaj has quit (Read error: Connection reset by peer)
2004 2013-05-09 19:13:15 <cjd> recipient knows it
2005 2013-05-09 19:13:22 <helo> i.e. it was through tor
2006 2013-05-09 19:13:40 <chmod755> helo, backtrace it!
2007 2013-05-09 19:13:42 <cjd> if under attack, ban tor
2008 2013-05-09 19:13:55 skinnkavaj has joined
2009 2013-05-09 19:13:55 skinnkavaj has quit (Changing host)
2010 2013-05-09 19:13:55 skinnkavaj has joined
2011 2013-05-09 19:13:59 <helo> i thought it was if under attack, use tor :(
2012 2013-05-09 19:14:01 <cjd> if not under attack ban tor \:D/
2013 2013-05-09 19:14:20 <chmod755> if under attack destroy tor
2014 2013-05-09 19:14:39 <CodeShark> if under tor destroy backtrace
2015 2013-05-09 19:15:01 yano has quit (Read error: Connection reset by peer)
2016 2013-05-09 19:15:20 <cjd> meh
2017 2013-05-09 19:15:32 <cjd> it requires pretty heavy changes to the protocol
2018 2013-05-09 19:15:52 <cjd> because you can't have someone use one key when talking to some nodes and another key when talking to others
2019 2013-05-09 19:16:58 yano has joined
2020 2013-05-09 19:17:22 <cjd> and actually ip is unimportant, just the key is enough because anyone who could make use of a "proof of evil" already knows their IP<->key mapping
2021 2013-05-09 19:17:27 rdymac has joined
2022 2013-05-09 19:19:30 SteveDekorte has joined
2023 2013-05-09 19:21:59 zoinky has joined
2024 2013-05-09 19:23:09 <gavinandresen> It is alert time; anybody see anything wrong with this alert message: "Action required: see http://bitcoin.org/may15.html for more information"
2025 2013-05-09 19:23:24 treetrunk has quit (Remote host closed the connection)
2026 2013-05-09 19:25:29 <gmaxwell> gavinandresen: Of the alert? that sounds okay. Luke wanted to update the may15.html page to also link to the backport thread. I've seen that he's done a lot of work testing those against this issue and I think they're likely a viable option for anyone who has dependencies keeping them from updating... but I don't actually know that they're worth mentioning on the page since the workaround there is ~sufficient.
2027 2013-05-09 19:26:01 treetrunk has joined
2028 2013-05-09 19:26:08 <gavinandresen> gmaxwell: mmm. I'm not a fan of backports, I don't like supporting lots of versions.
2029 2013-05-09 19:26:29 <jgarzik> gavinandresen: ACK alert time
2030 2013-05-09 19:26:32 <gavinandresen> gmaxwell: I'm having second thoughts on alert expiration....
2031 2013-05-09 19:26:57 <jgarzik> gavinandresen: you now prefer an expire time, versus infinite?
2032 2013-05-09 19:27:00 <gavinandresen> I was thinking expires-after-a-year, but now I'm thinking expires-after-â¦. a month? two months? might be better
2033 2013-05-09 19:27:11 nsillik has quit (Quit: nsillik)
2034 2013-05-09 19:27:37 <gavinandresen> My thinking is: we're very likely to get a block that triggers the problem in the next month or two.
2035 2013-05-09 19:27:43 <gmaxwell> just make it a month, and in a month we can consider sending another one if we think it matters.
2036 2013-05-09 19:27:46 <gavinandresen> ⦠so if you have not upgraded ,you'll be forked anyway.
2037 2013-05-09 19:28:00 <gavinandresen> And the "you're forked" message is higher priority than the alert, so that's what you'll see
2038 2013-05-09 19:28:10 <gmaxwell> That other one should probably say something like "If your height is less than X Your node is probably stuck."
2039 2013-05-09 19:28:13 <gavinandresen> Unless you workaround/patchâ¦. in which case you don't want to see the alert
2040 2013-05-09 19:28:16 <gmaxwell> ah.
2041 2013-05-09 19:28:21 <gmaxwell> indeed!
2042 2013-05-09 19:28:55 BlackPrapor has joined
2043 2013-05-09 19:29:04 <gavinandresen> Ok. So I think a month is the right answer.
2044 2013-05-09 19:29:07 Chuky has joined
2045 2013-05-09 19:29:10 n5 has joined
2046 2013-05-09 19:29:48 swulf-- has joined
2047 2013-05-09 19:30:32 <gmaxwell> One problem with assuming a big block will fork nodes is that the bustage isn't determinstic, so even with a big block a number of unfixed nodes will likely survive.
2048 2013-05-09 19:30:43 <jgarzik> I think we'll only get a block that triggers when a $BigPoolOp upgrades
2049 2013-05-09 19:30:58 <jgarzik> could be a day, after that. could be months, otherwise.
2050 2013-05-09 19:31:13 BlackPrapor has quit (Client Quit)
2051 2013-05-09 19:31:21 XertroV has joined
2052 2013-05-09 19:31:23 <jgarzik> there needs to be pool guidance
2053 2013-05-09 19:31:30 BlackPrapor has joined
2054 2013-05-09 19:31:33 <jgarzik> "this is what we think and why"
2055 2013-05-09 19:31:51 swulf--1 has quit (Ping timeout: 245 seconds)
2056 2013-05-09 19:31:54 iwilcox_ is now known as iwilcox
2057 2013-05-09 19:32:01 <gmaxwell> jgarzik: hm? all major miners should be on 0.8.1 already, thats whats actually preventing another forking event right now
2058 2013-05-09 19:33:03 <jgarzik> gmaxwell: soft limits are all set in a cautionary mode
2059 2013-05-09 19:33:08 <jgarzik> afaik
2060 2013-05-09 19:33:08 rdymac has quit (Ping timeout: 256 seconds)
2061 2013-05-09 19:33:13 <gmaxwell> But even if you create a very large block, what happens is that some unfixed <0.8 nodes get stuck and some don't. I'm not even sure if there is any single block that can reliably stick all of them. I think reliably sticking them takes a reorg.
2062 2013-05-09 19:33:42 resinate has quit (Ping timeout: 260 seconds)
2063 2013-05-09 19:33:45 <gmaxwell> (and of course, they won't all see a reorg even if you create one)
2064 2013-05-09 19:34:08 ThomasV has joined
2065 2013-05-09 19:34:13 mappum has joined
2066 2013-05-09 19:34:23 JDuke128 has quit (Quit: Computer has gone to sleep.)
2067 2013-05-09 19:34:45 <jgarzik> gmaxwell: Thus, the trigger in 402f19b64530775a7e4ded025c80d8c16a55e454 will not immediately cause increased blocks
2068 2013-05-09 19:35:00 Ferroh_ has joined
2069 2013-05-09 19:35:26 OneFixt has quit (Read error: Connection reset by peer)
2070 2013-05-09 19:35:30 rdymac has joined
2071 2013-05-09 19:35:46 BitCoroner has quit (Read error: Operation timed out)
2072 2013-05-09 19:35:47 OneFixt has joined
2073 2013-05-09 19:35:52 XertroV has quit (Ping timeout: 252 seconds)
2074 2013-05-09 19:36:00 <gmaxwell> jgarzik: fair enough. I think slush was running with the softlimit set to the maximum when the initial incident happened. I was implicitly assuming he still had it there, but you're right, thats not a good assumption.
2075 2013-05-09 19:36:03 BitCoroner has joined
2076 2013-05-09 19:36:05 <nsh> what problem is likely to be soon triggered by a block?
2077 2013-05-09 19:36:13 vazakl has quit (Read error: Connection reset by peer)
2078 2013-05-09 19:36:29 BTCOxygen has joined
2079 2013-05-09 19:36:30 qbasicer has quit (Read error: Operation timed out)
2080 2013-05-09 19:36:36 vazakl has joined
2081 2013-05-09 19:36:41 <jgarzik> nsh: http://bitcoin.org/may15.html
2082 2013-05-09 19:36:49 <nsh> ty
2083 2013-05-09 19:37:25 <nsh> oh ok
2084 2013-05-09 19:37:32 Ferroh has quit (Ping timeout: 256 seconds)
2085 2013-05-09 19:37:40 <nsh> thought there was a new issue
2086 2013-05-09 19:37:44 qbasicer has joined
2087 2013-05-09 19:37:54 <gavinandresen> 4am in Tokyo⦠I think I'll wait until it is daylight there before broadcasting the alert (first may15-related alert caused them some issues)
2088 2013-05-09 19:40:58 <edcba> what happens with that ?
2089 2013-05-09 19:41:05 icellan has joined
2090 2013-05-09 19:41:25 <edcba> some settings not working with blockchain length ?
2091 2013-05-09 19:41:40 <petertodd> are the DNS seeds preferably returning >=0.8 nodes?
2092 2013-05-09 19:41:55 <chmod755> petertodd, good question....
2093 2013-05-09 19:42:45 <kinlo> there isn't an alert sent already?
2094 2013-05-09 19:44:00 <jgarzik> petertodd: good point
2095 2013-05-09 19:44:39 ThomasV has quit (Ping timeout: 272 seconds)
2096 2013-05-09 19:45:03 <gavinandresen> kinlo: there have already been two alerts sent about the May 15 deadline, this will be the third and final
2097 2013-05-09 19:45:11 <kinlo> ic
2098 2013-05-09 19:46:07 rdymac has quit (Read error: Connection reset by peer)
2099 2013-05-09 19:46:33 resinate has joined
2100 2013-05-09 19:46:42 <gmaxwell> gavinandresen: ha. Do alterts trigger mtgox to hold deposits or something?
2101 2013-05-09 19:46:43 icellan has quit (Quit: icellan)
2102 2013-05-09 19:46:47 wei_ has joined
2103 2013-05-09 19:47:13 <kinlo> doesn't it trigger many clients?
2104 2013-05-09 19:47:17 icellan has joined
2105 2013-05-09 19:47:19 <helo> just noticed that pqosrh6wfaucet32.onion has a banscore of 40 from 'vin empty'
2106 2013-05-09 19:47:53 <kinlo> it makes sense to disable automated processes on any alert
2107 2013-05-09 19:48:09 <gavinandresen> gmaxwell: something like that
2108 2013-05-09 19:48:27 <petertodd> jgarzik>: It's not a major improvement, but it would help ensure nodes at least see the longer (hopefully) 0.8-only chain.
2109 2013-05-09 19:49:00 <gmaxwell> petertodd: at the moment the test should actually be !0.8.0 :P after may 15th, >=0.8.0
2110 2013-05-09 19:49:16 <gmaxwell> petertodd: but I don't think it really matters now.
2111 2013-05-09 19:50:21 <jgarzik> petertodd: It helps with bitcoinj clients, which are sadly, annoyingly over-reliant on DNS seeds
2112 2013-05-09 19:50:31 <jgarzik> which covers a big end-user case
2113 2013-05-09 19:51:38 rdymac has joined
2114 2013-05-09 19:54:08 <petertodd> jgarzik: yup, I was thinking SPV too, particularly the way they are constanting turned on and off
2115 2013-05-09 19:54:55 <petertodd> If anyway has a patch to Sipa's seed I can apply it now to my testnet seed to do a bit of testing.
2116 2013-05-09 19:55:45 uyppp has quit (Quit: uyppp)
2117 2013-05-09 19:56:31 digitalmagus has quit (Remote host closed the connection)
2118 2013-05-09 19:56:48 digitalmagus has joined
2119 2013-05-09 19:57:30 <gmaxwell> I thought sipa's seeder already had a version test?
2120 2013-05-09 19:58:06 <SteveDekorte> is the list of the hard coded seeds maintained anywhere outside of the bitcoin-qt source?
2121 2013-05-09 19:58:19 <michagogo> SteveDekorte: on the wiki somewhere
2122 2013-05-09 19:58:27 <gmaxwell> michagogo: it is? I don't think so.
2123 2013-05-09 19:58:37 <michagogo> I thought I saw it at some point
2124 2013-05-09 19:58:41 <gmaxwell> There is an alternative list at least.
2125 2013-05-09 19:59:01 <SteveDekorte> michagogo: I've been poking around there but haven't found one
2126 2013-05-09 19:59:07 <gmaxwell> SteveDekorte: those nodes aren't special, they're just randomly selected high uptime nodes found on the public internet.
2127 2013-05-09 19:59:30 <SteveDekorte> gmaxwell: yeah, I just don't want to pick one and trust it's getaddr
2128 2013-05-09 19:59:31 SirDefaced has quit ()
2129 2013-05-09 19:59:36 <michagogo> https://en.bitcoin.it/wiki/Satoshi_Client_Node_Discovery
2130 2013-05-09 20:00:01 <michagogo> Oh, you mean the seed *nodes*?
2131 2013-05-09 20:00:09 <michagogo> I was thinking the seed DNS servers
2132 2013-05-09 20:00:10 <SteveDekorte> michagogo: yeah, thanks :)
2133 2013-05-09 20:00:14 <michagogo> nvm
2134 2013-05-09 20:00:38 <SteveDekorte> bitcoin-ruby calls them dns seeds for some reason
2135 2013-05-09 20:01:07 systemParanoid has quit (Quit: Leaving)
2136 2013-05-09 20:01:11 <michagogo> SteveDekorte: It calls the seed *nodes* that?
2137 2013-05-09 20:01:18 <michagogo> Because there *are* dns seeds
2138 2013-05-09 20:01:32 Eremes has joined
2139 2013-05-09 20:02:04 <Eremes> guys i sent 1 BTC to an address, its been 1hour ++ and not even 1 confirmation, could someone tell me why is it possible ?
2140 2013-05-09 20:02:40 MobPhone has joined
2141 2013-05-09 20:02:42 <michagogo> Eremes: What's the transaction ID?
2142 2013-05-09 20:02:48 <gmaxwell> ugh. yet another bitcoin validation implementation I was unaware of.
2143 2013-05-09 20:02:53 <gmaxwell> Is the author of bitcoin-ruby in here?
2144 2013-05-09 20:02:54 <petertodd> gmaxwell: it's only a pre 0.3 test if I understand correctly, so making it 0.8 specific isn't completely trivial
2145 2013-05-09 20:02:57 ThomasV has joined
2146 2013-05-09 20:02:58 <Eremes> 320dbeda826aeefe0bf7c0622222b9cad6619b4885b63c70b38d0a0a0ba95a10
2147 2013-05-09 20:03:08 <Eremes> very annoying
2148 2013-05-09 20:03:13 <michagogo> My node's seen it
2149 2013-05-09 20:03:15 <michagogo> ;;tslb
2150 2013-05-09 20:03:17 <gribble> Time since last block: 7 minutes and 50 seconds
2151 2013-05-09 20:03:17 <SteveDekorte> michagogo: it looked like it but I'm just learning this stuff as a go so I might have misunderstood it :)
2152 2013-05-09 20:03:40 <michagogo> bc.i has also seen it
2153 2013-05-09 20:04:15 <Eremes> michagogo: any ideas why its not confirmed yet ?
2154 2013-05-09 20:04:36 <michagogo> Hmm
2155 2013-05-09 20:04:45 <michagogo> It spends an output that already has 25 confirms
2156 2013-05-09 20:04:54 santoscork has quit (Quit: Quiet while I make like a cat)
2157 2013-05-09 20:05:14 <Eremes> my client said Im a scammer .lol
2158 2013-05-09 20:05:33 <michagogo> It's got a fee, though one less than the suggested.
2159 2013-05-09 20:05:36 <michagogo> Eremes: Hmm?
2160 2013-05-09 20:05:45 <cjd> screenshot?
2161 2013-05-09 20:05:49 <Eremes> because not even 1 confirmation for an hour
2162 2013-05-09 20:05:54 <Eremes> already gave him one
2163 2013-05-09 20:05:59 <wumpus> well the transaction is even visible on blockchain.info
2164 2013-05-09 20:06:00 Skav has quit (Ping timeout: 256 seconds)
2165 2013-05-09 20:06:05 <Eremes> it is visible
2166 2013-05-09 20:06:26 <gmaxwell> Eremes: it has fees below the base fee threshold, so it's treated as a free transaction and it has quite low priority.
2167 2013-05-09 20:06:27 <Eremes> but he wont believe me until 1 confirm
2168 2013-05-09 20:06:35 <michagogo> Oh, your client == the other person
2169 2013-05-09 20:06:45 <cjd> oh lol
2170 2013-05-09 20:06:49 <michagogo> I understood that as whatever bitcoin software you're using
2171 2013-05-09 20:06:57 <cjd> I thought bitcoind was like "you're a scammer"
2172 2013-05-09 20:06:59 <Eremes> gmaxwell: i paid the electrum suggested fee for 0.00002
2173 2013-05-09 20:07:08 <michagogo> Eremes: one less 0 there
2174 2013-05-09 20:07:16 <michagogo> 0.0002
2175 2013-05-09 20:07:19 <gmaxwell> Eremes: well, the electrum suggestion was dumb.
2176 2013-05-09 20:07:19 <Eremes> 0.0002
2177 2013-05-09 20:07:20 <Eremes> yes
2178 2013-05-09 20:07:23 <Eremes> i paid that one
2179 2013-05-09 20:07:29 <Eremes> 0.0002 -> i paid
2180 2013-05-09 20:07:30 <michagogo> gmaxwell: Oh, really? Too-low fee is the same as no fee?
2181 2013-05-09 20:07:53 <gmaxwell> michagogo: yes. Otherwise someone could just produce a flood of txn with 1e-8 fees and get ahead of all the free transactions.
2182 2013-05-09 20:08:08 <chmod755> cjd, haha
2183 2013-05-09 20:08:13 <Eremes> ok
2184 2013-05-09 20:08:16 <Eremes> 1 confirmation finally !
2185 2013-05-09 20:08:20 <gmaxwell> Eremes: in any case, I'd expect this to get mined in a couple more blocks.
2186 2013-05-09 20:08:27 <gmaxwell> ah, there you go.
2187 2013-05-09 20:08:28 n5 has quit (Ping timeout: 256 seconds)
2188 2013-05-09 20:08:31 Mr_G has joined
2189 2013-05-09 20:08:46 <gmaxwell> (my own node wasn't even trying to put it in the next block when you asked)
2190 2013-05-09 20:08:55 <Eremes> lol
2191 2013-05-09 20:09:00 <cjd> gmaxwell: arguably that is a good thing since it promotes a market for fees rather than a fixed price
2192 2013-05-09 20:09:14 <cjd> not so good for those who pay no fee ofc
2193 2013-05-09 20:09:17 n5 has joined
2194 2013-05-09 20:09:30 Dragone2 has joined
2195 2013-05-09 20:09:34 <gmaxwell> cjd: 1e-8 btc is not a fee, come on now. The marginal cost to the miner of including that txn is almost certantly higher than 1e-8.
2196 2013-05-09 20:09:41 <ThomasV> gmaxwell: 0.0002 is not below the base fee threshold
2197 2013-05-09 20:09:51 <cjd> bidding gotta start somewhere xD
2198 2013-05-09 20:09:58 <gmaxwell> ThomasV: it is. The base fee threshold for mining in 0.8.1 is 0.0005 BTC.
2199 2013-05-09 20:10:15 <gmaxwell> ThomasV: we'll relay at 0.0001 but mining happenes at 0.0005 BTC/kb.
2200 2013-05-09 20:10:21 zooko_ has joined
2201 2013-05-09 20:10:31 <ThomasV> gmaxwell: miners policy isnt decided by bitcoind
2202 2013-05-09 20:10:33 <gmaxwell> This is lowered in git, but thats not released and thus not widely used.
2203 2013-05-09 20:10:43 <gmaxwell> ThomasV: _most_ miners are on the defaults.
2204 2013-05-09 20:11:06 <gmaxwell> (and, in fact it _is_ determined by the miner's bitcoind)
2205 2013-05-09 20:11:35 Dragone2 has quit (Quit: Ciao a tutti, e' stato un piacere parlare con voi. A presto! - Dragone2 Script (5.2 Alpha) - Scaricabile da http://www.risposteinformatiche.it/IRC/index.html)
2206 2013-05-09 20:11:37 <gmaxwell> (as GBT doesn't return anything that bitcoind doesn't think should be included)
2207 2013-05-09 20:11:50 <ThomasV> _most_ as in most hashing power?
2208 2013-05-09 20:12:04 <gmaxwell> ThomasV: yes, as in most hashing power.
2209 2013-05-09 20:12:18 <ThomasV> how do you know?
2210 2013-05-09 20:12:31 <gmaxwell> Because I went around asking people when I went to lower it to 0.0001 in git.
2211 2013-05-09 20:13:47 Mr_G has quit (Ping timeout: 272 seconds)
2212 2013-05-09 20:14:46 caedes has quit (Remote host closed the connection)
2213 2013-05-09 20:14:49 JDuke128 has joined
2214 2013-05-09 20:15:44 <gmaxwell> ThomasV: I note that the txn in question didn't get confirmed until it would have met the free transaction test.
2215 2013-05-09 20:18:57 <ThomasV> gmaxwell: should I raise electrum's default fee? why are miners not maximizing the fees they collect?
2216 2013-05-09 20:19:47 <jgarzik> ThomasV: fees are economically invisible, compared to 25 BTC reward
2217 2013-05-09 20:19:49 <ThomasV> it feels like the "market for fees" isn't really a market
2218 2013-05-09 20:19:55 <jgarzik> ThomasV: correct :)
2219 2013-05-09 20:20:05 <ThomasV> jgarzik: for now
2220 2013-05-09 20:20:10 <jgarzik> ThomasV: correct x2 :)
2221 2013-05-09 20:20:38 <gmaxwell> ThomasV: 0.0002 isn't economically interesting right nowâ it might even be more than the marginal increase in orphaning risk from including the transaction (I doubt it, but its hard to be sure)
2222 2013-05-09 20:20:46 maximeb has quit (Quit: maximeb)
2223 2013-05-09 20:21:14 <jgarzik> The cost of "mucking with bitcoind and a fee market" is higher for pool ops than just leaving it be
2224 2013-05-09 20:21:16 <Ry4an> That 4000+BTC feel that floated through a few weeks ago surely made some pool's day.
2225 2013-05-09 20:21:24 <gmaxwell> A point here is that 0.0005/0.0001 per kb is a minimum, to actually be market competative you need much higher fees.
2226 2013-05-09 20:21:24 <jgarzik> There is a distinct cost to patching and maintaining your own bitcoind variant
2227 2013-05-09 20:21:25 <Ry4an> s/feel/fee/
2228 2013-05-09 20:21:39 <michagogo> Who's a sysop on the wiki?
2229 2013-05-09 20:21:43 <gmaxwell> Ry4an: I missed that.
2230 2013-05-09 20:21:47 <gmaxwell> michagogo: why?
2231 2013-05-09 20:21:47 <jgarzik> What's a sysop?
2232 2013-05-09 20:21:49 * jgarzik runs
2233 2013-05-09 20:22:35 <michagogo> gmaxwell: jgarzik: Whichever MediaWiki: page controls the message you see when you register (or maybe it's log in?) needs to be modified to remove the reference to being logged in on all wikimedia foundation sites
2234 2013-05-09 20:22:35 ali1234 has joined
2235 2013-05-09 20:22:39 <gmaxwell> ThomasV: most bitcoinj based things are just putting a static 0.001 BTC fee on everything. This is a small fee on large transactions, but when there is one on a tiny transaction its enormous.
2236 2013-05-09 20:23:15 <michagogo> Ry4an: 4000+ fee? o_O
2237 2013-05-09 20:23:20 <michagogo> Got a txid?
2238 2013-05-09 20:23:31 <Ry4an> Lemme did it up. It was funny stuff.
2239 2013-05-09 20:23:35 <gmaxwell> michagogo: I thought that was static message which is part of the centeral login extension.
2240 2013-05-09 20:23:37 <petertodd> jgarzik: you're pissing on my childhood
2241 2013-05-09 20:23:45 <jgarzik> petertodd: mine too
2242 2013-05-09 20:23:58 <jgarzik> petertodd: hence the running :) I'm an old WWiV SysOp
2243 2013-05-09 20:24:08 <michagogo> gmaxwell: Erm, are there bitcoin wikis?
2244 2013-05-09 20:24:15 <petertodd> jgarzik: nice! yeah I was in iLink for years
2245 2013-05-09 20:24:19 <gmaxwell> michagogo: there are multiple ones, yesâ different languages.
2246 2013-05-09 20:24:23 <michagogo> Ah
2247 2013-05-09 20:24:33 <michagogo> And there's no way to change that? Seems weird.
2248 2013-05-09 20:25:03 <gmaxwell> jgarzik: down with WWIV, wildcat! forever (not reallyâ¦)
2249 2013-05-09 20:25:08 <jgarzik> ;p
2250 2013-05-09 20:25:15 TD has joined
2251 2013-05-09 20:25:25 <jgarzik> WWIV taught me Pascal, then WWIV taught me C.
2252 2013-05-09 20:25:47 <Ry4an> I blindly applied sooo many WWIV .mod files before I understood what I was changing.
2253 2013-05-09 20:26:07 richcollins has joined
2254 2013-05-09 20:26:14 porquilho has joined
2255 2013-05-09 20:26:16 <Ry4an> (1) hand apply .diff, (2) compile, (3) "dammit!"
2256 2013-05-09 20:26:27 <jgarzik> what was that msdos multi-tasking gadget that bbs people used?
2257 2013-05-09 20:26:31 <jgarzik> desqview?
2258 2013-05-09 20:26:39 <gmaxwell> I wrote multiuser chat systems for two different BBS systems. This was where I learned the truth of "Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it."
2259 2013-05-09 20:26:41 <petertodd> jgarzik: yup
2260 2013-05-09 20:26:44 <Ry4an> yeah. Towartd the end 4dos has something similar.
2261 2013-05-09 20:26:53 <gmaxwell> jgarzik: qemm+desqview <3
2262 2013-05-09 20:27:02 <jgarzik> there ya go
2263 2013-05-09 20:27:07 <cjd> gmaxwell: +1
2264 2013-05-09 20:27:17 <petertodd> whew, maybe it's safe to admit I did some visual basic programming on windows 3.1
2265 2013-05-09 20:27:45 <gmaxwell> One of the BBS systems I coded for was written in microsoft quickbasic plus assembly.
2266 2013-05-09 20:27:46 <cjd> gmaxwell: fortunately if you write a clever programand talk a lot, sooner or later some physics grad students show up to hold your hand xD
2267 2013-05-09 20:28:06 <cjd> >why cjdns is not total crap anymore
2268 2013-05-09 20:28:11 <Eremes> gtg guys
2269 2013-05-09 20:28:15 <Eremes> thanks for everything :)
2270 2013-05-09 20:28:30 Eremes has quit ()
2271 2013-05-09 20:28:38 <jgarzik> some of the older BBS's were in *BASIC, yeah <shiver> with a little peek-ing and poke-ing when that failed
2272 2013-05-09 20:29:04 JDuke128 has quit (Quit: Computer has gone to sleep.)
2273 2013-05-09 20:29:10 <cjd> nodejs for dos
2274 2013-05-09 20:29:17 <jgarzik> hah
2275 2013-05-09 20:29:22 <gmaxwell> man, I can't remember the name of that board software. I created the most over designed multiuser chat system for that.
2276 2013-05-09 20:29:41 <jgarzik> in the days when compuserve was big
2277 2013-05-09 20:29:52 <jgarzik> *poof*
2278 2013-05-09 20:30:32 <ProfMac> I wrote a terminal program for a bbs on a 6809 in assembly. It waited until whitespace or 5 seconds, then it presented a complete token. This was much easier to read than a character at a time.
2279 2013-05-09 20:30:40 * nsh commits gmaxwell's wisdom to memory
2280 2013-05-09 20:30:53 <nsh> (re code/debug cleverness)
2281 2013-05-09 20:31:30 <gmaxwell> nsh: it's Kernighan's wisdom, I'm just borrowing it
2282 2013-05-09 20:31:32 serp has quit (Ping timeout: 246 seconds)
2283 2013-05-09 20:31:55 XertroV has joined
2284 2013-05-09 20:32:25 <michagogo> gmaxwell: I looked into it, with the help of uselang=qqx
2285 2013-05-09 20:32:30 <gmaxwell> (it's from https://en.wikipedia.org/wiki/The_Elements_of_Programming_Style )
2286 2013-05-09 20:32:40 <michagogo> It's nested in a few MediaWiki: pages deep...
2287 2013-05-09 20:32:43 <michagogo> gmaxwell: https://en.bitcoin.it/w/index.php?title=Special%3AAllMessages&prefix=Centralauth-group&filter=all&lang=en&limit=5000
2288 2013-05-09 20:32:46 <nsh> gmaxwell, ah, well, i appreciate the conveyance as much as the creation
2289 2013-05-09 20:32:57 <gmaxwell> michagogo: thanks!
2290 2013-05-09 20:33:13 guruvan has quit (Remote host closed the connection)
2291 2013-05-09 20:33:14 <michagogo> That's used in the items at the bottom of https://en.bitcoin.it/w/index.php?title=Special%3AAllMessages&prefix=Centralauth-log&filter=all&lang=en&limit=5000
2292 2013-05-09 20:33:15 <gmaxwell> michagogo: will get fixed in a moment.
2293 2013-05-09 20:33:21 <michagogo> gmaxwell: :-)
2294 2013-05-09 20:33:33 <michagogo> Also, qqx is awesome
2295 2013-05-09 20:33:56 guruvan has joined
2296 2013-05-09 20:34:49 <gmaxwell> michagogo: yea, I didn't think to try it because ... I really though I looked before and found it was hardcoded.
2297 2013-05-09 20:34:57 eklass has quit (Quit: Nettalk6 - www.ntalk.de)
2298 2013-05-09 20:36:17 XertroV has quit (Ping timeout: 252 seconds)
2299 2013-05-09 20:36:19 <michagogo> For anyone who doesn't know, appending ?uselang=qqx (or &uselang=qqx if there's already a ?) to any mediawiki URL will show, for any messages definable in MediaWiki: pages, the name of the message rather than the message itself
2300 2013-05-09 20:41:42 PhantomSpark has joined
2301 2013-05-09 20:42:22 eklass has joined
2302 2013-05-09 20:42:41 rdponticelli has quit (Ping timeout: 245 seconds)
2303 2013-05-09 20:43:49 paracyst has joined
2304 2013-05-09 20:45:57 <michagogo> gmaxwell: How come the bitcoin wiki payment thing waits for 6 confirmations?
2305 2013-05-09 20:46:29 <gmaxwell> michagogo: AFAIK because its an mtgox account and thats how the api there works. If you make the deposit from another mtgox account its instant.
2306 2013-05-09 20:46:30 maximeb has joined
2307 2013-05-09 20:46:39 <michagogo> That seems like overkill to me, a confirmation or two should by plenty
2308 2013-05-09 20:46:40 <michagogo> Ahh
2309 2013-05-09 20:46:41 rdymac has quit (Quit: This computer has gone to sleep)
2310 2013-05-09 20:46:53 Hasimir has joined
2311 2013-05-09 20:47:15 <gmaxwell> zero should be fine, so long as it bans you and autoreverts your edits if the transaction is conflicted. :P
2312 2013-05-09 20:47:17 <michagogo> So as far as mtgox's API returns are concerned, the transaction doesn't exist until 6 confirms?
2313 2013-05-09 20:47:37 <michagogo> I'd make it 1 or 2
2314 2013-05-09 20:48:00 <michagogo> (not for the API, in a theoretical rewrite of the payment thing)
2315 2013-05-09 20:48:15 <gmaxwell> I believe mtgox uses an even more sophicated criteria, 6 beyond the longest chain fork that they're aware of.
2316 2013-05-09 20:49:05 nsillik has joined
2317 2013-05-09 20:49:07 rdymac has joined
2318 2013-05-09 20:49:36 maximeb has quit (Client Quit)
2319 2013-05-09 20:49:59 <sipa> 21:39:50 <@gmaxwell> ugh. yet another bitcoin validation implementation I was unaware of. --> ?
2320 2013-05-09 20:50:40 saulimus has quit (Ping timeout: 252 seconds)
2321 2013-05-09 20:51:40 BTCOxygen is now known as BTCNitrogen
2322 2013-05-09 20:51:53 TD has quit (Quit: TD)
2323 2013-05-09 20:53:22 <gmaxwell> sipa: "bitcoin-ruby"
2324 2013-05-09 20:53:34 <sipa> ah
2325 2013-05-09 20:53:48 <gmaxwell> it's lian's but I didn't know it was public.
2326 2013-05-09 20:53:58 <gmaxwell> It, of course, doesn't appear to be correct.
2327 2013-05-09 20:53:59 <sipa> i misread... i thought you said there was a validation rule you weren't aware of
2328 2013-05-09 20:54:08 rdponticelli has joined
2329 2013-05-09 20:54:32 <gmaxwell> sipa: ah, well, I'm sure there are more of those too. :)
2330 2013-05-09 20:54:54 rdymac has quit (Quit: This computer has gone to sleep)
2331 2013-05-09 20:55:12 <hpprinter100> Hi, what ufw rules do i need to allow bitcoind to work? I am using the following config http://pastebay.net/1218862
2332 2013-05-09 20:56:05 <sipa> hpprinter100: testnet?
2333 2013-05-09 20:56:23 <hpprinter100> bitcoin testnet
2334 2013-05-09 20:56:49 qwertyoruiop has joined
2335 2013-05-09 20:57:21 <tumak> hmm, bitcoin-ruby looks nice
2336 2013-05-09 20:57:22 <hpprinter100> i think it maybe because bitcoind is showing as using tcp6 and i have disabled ipv6 in the firewall
2337 2013-05-09 20:57:37 <hpprinter100> so can i specify the bind interface for bitcoind?
2338 2013-05-09 20:57:46 <tumak> concise code
2339 2013-05-09 20:57:52 BTCNitrogen is now known as BTCOxygen
2340 2013-05-09 20:57:53 <sipa> hpprinter100: yes, -bind=IP
2341 2013-05-09 20:57:58 <tumak> gmaxwell: whats wrong with https://github.com/lian/bitcoin-ruby/blob/master/lib/bitcoin/validation.rb ?
2342 2013-05-09 20:58:34 rdponticelli has quit (Ping timeout: 264 seconds)
2343 2013-05-09 21:01:15 <gmaxwell> tumak: It doesn't come close to validating all the rules (not there, not elsewhere in the code as far as I can see)
2344 2013-05-09 21:02:12 grau has quit (Remote host closed the connection)
2345 2013-05-09 21:02:27 imsaguy is now known as pirateat40
2346 2013-05-09 21:02:35 pirateat40 is now known as imsaguy
2347 2013-05-09 21:03:19 tyn has joined
2348 2013-05-09 21:03:33 <tumak> hmm
2349 2013-05-09 21:03:34 <hpprinter100> sipa still cannot get it to connect but it's binding to the ipv4 now ;)
2350 2013-05-09 21:03:55 <sipa> why does it not connect?
2351 2013-05-09 21:04:04 <gmaxwell> tumak: I really don't want to play the "your implementation is wrong" game. I am not a correctness oracle, "keep posting implementations until gmaxwell stops complaining" doesn't scale.
2352 2013-05-09 21:04:25 <hpprinter100> the firewall is buggering it up, il paste my config
2353 2013-05-09 21:06:24 Prattler has quit (Quit: ZNC - http://znc.in)
2354 2013-05-09 21:07:15 <cjd> lol
2355 2013-05-09 21:07:29 <tumak> gmaxwell: indeed, i was just curious which bip in there is missing :/
2356 2013-05-09 21:07:45 <hpprinter100> my coutput is here http://pastebay.net/1218913
2357 2013-05-09 21:07:46 <gmaxwell> Who says whats missing is a BIP?
2358 2013-05-09 21:07:57 <sipa> tumak: i think BIPs are the things people will most likely do right, exactly because they are well-documented
2359 2013-05-09 21:08:01 SirDefaced has joined
2360 2013-05-09 21:08:07 <tumak> gmaxwell: ah
2361 2013-05-09 21:08:10 <tumak> sipa: yep :/
2362 2013-05-09 21:08:13 <sipa> a lot of validation behaviour is not well documented at all
2363 2013-05-09 21:08:26 <sipa> or even not _known_ at all
2364 2013-05-09 21:08:34 <gmaxwell> Like, not known to _anyone_.
2365 2013-05-09 21:08:56 serp has joined
2366 2013-05-09 21:08:59 <gmaxwell> who the @#$*@#(@* knew that BDB imposed a limit on the number of transactions updated in one block?!
2367 2013-05-09 21:09:04 <tumak> re: if you cant store the block in a database, obviously it is not valid :)
2368 2013-05-09 21:09:25 <michagogo> Did that alert go out yet?
2369 2013-05-09 21:09:26 <sipa> gmaxwell: i did
2370 2013-05-09 21:09:55 <tumak> bdb is innocent though
2371 2013-05-09 21:10:04 <sipa> we were setting the limit, by the way
2372 2013-05-09 21:10:04 <tumak> i thought the bug was that any kind of db error was treated as block error
2373 2013-05-09 21:10:18 <hpprinter100> anyone good at ufw please help me getting it to work with bitcoind? my config http://pastebay.net/1218913
2374 2013-05-09 21:10:19 <sipa> tumak: indeed, that too
2375 2013-05-09 21:10:39 <gmaxwell> sipa: yea, but we set it to 10,000
2376 2013-05-09 21:10:55 rdponticelli has joined
2377 2013-05-09 21:11:02 <gmaxwell> okay, who the @#$*@(* knew that the BDB could use 2 locks for ever transaction updated. :P
2378 2013-05-09 21:11:07 <sipa> yes, i was not aware of the exact semantics, and i assumed the number set was chosen to be sufficiently high
2379 2013-05-09 21:11:32 <sipa> but it's not like we can blame BDB for it; it is well-documented behaviour (though insane to use)
2380 2013-05-09 21:12:03 <cjd> C++ problems
2381 2013-05-09 21:12:05 Prattler has joined
2382 2013-05-09 21:12:15 <tumak> yea
2383 2013-05-09 21:12:20 <sipa> cjd: BDB is C
2384 2013-05-09 21:12:33 <tumak> C languages are horrible to define any kind of standard
2385 2013-05-09 21:12:38 <tumak> yet bitcoin does exactly that :(
2386 2013-05-09 21:12:55 <gmaxwell> tumak: huh? C actually has very straight forward, very well defined behavior.
2387 2013-05-09 21:12:56 <cjd> problem is not C++ the language, it's the wonderful delicious abstractions which make everything so easy
2388 2013-05-09 21:13:05 <nsh> sipa: sufficiently high ~= 640k
2389 2013-05-09 21:13:08 <nsh> ?
2390 2013-05-09 21:13:08 <cjd> and push the actual definition of the standard off on libraries
2391 2013-05-09 21:13:21 <tumak> gmaxwell: i mean it in the BNF vs recursive descent parser in C sense
2392 2013-05-09 21:13:26 <tumak> gmaxwell: which one is more conscise? :)
2393 2013-05-09 21:13:33 <sipa> cjd: which is something that's very valuable for development :)
2394 2013-05-09 21:13:40 <sipa> cjd: but not for all types of software
2395 2013-05-09 21:13:52 <cjd> it gets software shipped
2396 2013-05-09 21:14:12 <cjd> but like borrowing from the bank of technical debt, you must pay it all back plus interest
2397 2013-05-09 21:14:59 <cjd> which is learned through these "openssl encoding is a leaky abstraction" and "bdb storage is a leaky abstraction" lessons
2398 2013-05-09 21:15:00 <tumak> cjd: why cant we have just wall of text to standardize bitcoin protocol? :)
2399 2013-05-09 21:15:02 <tumak> i mean
2400 2013-05-09 21:15:08 <tumak> most internet protocols have that
2401 2013-05-09 21:15:26 <cjd> gotta read the code :)
2402 2013-05-09 21:15:27 <tumak> even some stupidly complex ones like ipsec
2403 2013-05-09 21:15:45 <tumak> cjd: not everyone is good at spotting UBs
2404 2013-05-09 21:15:51 <sipa> tumak: though writing it down will help others that are reimplementing it, a document of text cannot ever be the standard for a consensus system like bitcoin
2405 2013-05-09 21:15:51 rdponticelli has quit (Ping timeout: 248 seconds)
2406 2013-05-09 21:16:10 <tumak> C specs usually include UBs which are not pointed out :/
2407 2013-05-09 21:16:21 <cjd> is that haskell bitcoin impl still being developed?
2408 2013-05-09 21:16:45 <sipa> tumak: if the standard would define some sane behaviour, and a significant part of the network would not follow it exactly, those clients would have a _bug_ , but it would be the spec that would be _wrong_
2409 2013-05-09 21:16:46 <tumak> sipa: the consensus is whatever bitcoin-qt does
2410 2013-05-09 21:16:56 <sipa> no, the consensus is whatever the network does
2411 2013-05-09 21:16:59 <tumak> sipa: so that document would be 'a tale of bitcoin-qt internals'
2412 2013-05-09 21:17:00 <tumak> :)
2413 2013-05-09 21:17:03 agricocb has quit (Quit: Leaving.)
2414 2013-05-09 21:17:18 <tumak> sipa: in theory yes, but practice is bitcoin-qt
2415 2013-05-09 21:17:23 <tumak> (that could change in future indeed though)
2416 2013-05-09 21:18:01 <cjd> 16:53 < sipa> no, the consensus is whatever the network does <-- We agree to fork :D
2417 2013-05-09 21:18:26 <sipa> i dislike calling the reference implementation "Bitcoin-Qt" - to me, that has always just been the name of the GUI
2418 2013-05-09 21:18:50 <nsh> +1
2419 2013-05-09 21:18:55 <tumak> let's call it satoshian client!
2420 2013-05-09 21:20:43 Mapleh has quit ()
2421 2013-05-09 21:21:48 chmod755 has quit (Quit: chmod755)
2422 2013-05-09 21:22:42 daybyter has quit (Quit: Konversation terminated!)
2423 2013-05-09 21:23:13 <cjd> would be nice to have better block/tx fuzzing tools
2424 2013-05-09 21:23:37 <cjd> thus you could run your implementation over night against the satoshi client at the same time and hunt for disagreements
2425 2013-05-09 21:23:53 A has joined
2426 2013-05-09 21:24:28 <nsh> hmmm
2427 2013-05-09 21:24:32 <gmaxwell> cjd: there is a good tool for this.
2428 2013-05-09 21:24:48 <nsh> i wonder if you could do a network replay (in fast forward, minus the boring bits) to test new clients
2429 2013-05-09 21:24:51 <gmaxwell> (bluematt's tool)
2430 2013-05-09 21:25:19 <gmaxwell> nsh: thats what bluematt's tool does, it replays a sequence of blocks, including reorgs and invalid blocks.
2431 2013-05-09 21:25:21 <sipa> nsh: that's the most basic test you can do: import the main chain
2432 2013-05-09 21:25:33 <nsh> right
2433 2013-05-09 21:25:36 <cjd> oh nice
2434 2013-05-09 21:25:40 <cjd> I can haz?
2435 2013-05-09 21:25:44 <sipa> nsh: doing testnet is more interesting, as it contains much more weird stuff
2436 2013-05-09 21:25:49 * nsh nods
2437 2013-05-09 21:25:51 <gmaxwell> And indeed, just a regular replay is import.. and that doesn't catch failures to reject invalid stuff, which are the things the programmers get wrong the most often.
2438 2013-05-09 21:25:51 <tumak> http://jenkins.bluematt.me/pull-tester/files/FullBlockTestGenerator.java
2439 2013-05-09 21:30:22 <cjd> https://github.com/TheBlueMatt/test-scripts/blob/master/FullBlockTestGenerator.java
2440 2013-05-09 21:30:25 <cjd> interesting
2441 2013-05-09 21:31:06 savetheinternet has quit (Quit: Computer has gone to sleep.)
2442 2013-05-09 21:31:15 <gmaxwell> The downside of bluematt's excellent test is that its not complete and can never be complete.
2443 2013-05-09 21:31:31 chora is now known as chorao
2444 2013-05-09 21:31:31 chorao has quit (Changing host)
2445 2013-05-09 21:31:31 chorao has joined
2446 2013-05-09 21:31:50 <cjd> yeah
2447 2013-05-09 21:31:57 <gmaxwell> E.g. there may be some sequence of blocks that makes some feasable broken code reject a valid chain, which it'll never find.
2448 2013-05-09 21:32:18 <cjd> it's an avenue of research which gets way too little attention imo
2449 2013-05-09 21:32:26 XertroV has joined
2450 2013-05-09 21:32:38 <tumak> and until bitcoin implementations will be formally verifiable
2451 2013-05-09 21:32:42 <tumak> it will stay that way
2452 2013-05-09 21:33:44 <cjd> the haskell impl is interesting since haskell lends itself to verification moreso than C++
2453 2013-05-09 21:33:45 <tumak> cjd: i think there are even more complex systems than bitcoin which are proven to be correct
2454 2013-05-09 21:34:21 <gmaxwell> tumak: we don't need "proven to be correct" we need proven to be consistent. This is actually much harder for many definitions of "correct".
2455 2013-05-09 21:34:52 ThomasV has quit (Read error: Operation timed out)
2456 2013-05-09 21:36:52 XertroV has quit (Ping timeout: 252 seconds)
2457 2013-05-09 21:37:15 whiterabbit has joined
2458 2013-05-09 21:38:09 <gmaxwell> Much of the time the properties proven correct for non-trivial software are themselves pretty trivial. Not aware of any proofs of exact output consistency for anything more complex than sorting. The closest is probably compcert, and doesn't produce a unique solution.
2459 2013-05-09 21:38:15 <tumak> gmaxwell: sure. what i mean is to write actual proof, beyond mere fuzzy testing.
2460 2013-05-09 21:38:31 <gmaxwell> tumak: I know what you mean.
2461 2013-05-09 21:38:51 jeewee has quit (Quit: Leaving.)
2462 2013-05-09 21:39:16 savetheinternet has joined
2463 2013-05-09 21:39:17 <cjd> there are some proven compilers IIRC
2464 2013-05-09 21:39:22 <gmaxwell> tumak: and even if you achieve that, unless the proof is extractable and people actually run the extracted program, it's just a dead letter.
2465 2013-05-09 21:39:25 <cjd> COQ?
2466 2013-05-09 21:39:36 <gmaxwell> cjd: there is one, compcert.
2467 2013-05-09 21:39:37 <tumak> gmaxwell: yeah
2468 2013-05-09 21:39:44 <cjd> oh indeed, that's it
2469 2013-05-09 21:39:49 graingert has joined
2470 2013-05-09 21:39:50 graingert has quit (Changing host)
2471 2013-05-09 21:39:50 graingert has joined
2472 2013-05-09 21:40:08 <gmaxwell> This is part of the reason that I'm not bouncing for joy excited about writing a spec. Because a spec is just a fantasty unless its somehow executable.
2473 2013-05-09 21:40:17 <tumak> the problem is that proofs are always tied to particular implementation, its hard to do model check across implementations using same proofs
2474 2013-05-09 21:40:20 <tumak> hm
2475 2013-05-09 21:40:31 wrabbit has quit (Ping timeout: 276 seconds)
2476 2013-05-09 21:40:32 whiterabbit is now known as wrabbit
2477 2013-05-09 21:40:38 rdponticelli has joined
2478 2013-05-09 21:40:46 <gmaxwell> If anything a spec that doesn't come along with a way of extracting a praticle implementation from it is actually a risk, because it encourage people to create their own implementations which are not correct and consistent, even if the spec itself is flawless.
2479 2013-05-09 21:41:00 <tumak> gmaxwell: once again, what about some kind of generator?
2480 2013-05-09 21:41:17 <nsh> tumak, how would that work?
2481 2013-05-09 21:41:22 <tumak> ie have bitcoin spec in some neutral DSL
2482 2013-05-09 21:41:24 <gmaxwell> tumak: thats what I was trying to say before. I'm not aware of any work proving that two very distinct complex functions produce the same output.
2483 2013-05-09 21:41:39 <gmaxwell> tumak: this is hard because performance is important.
2484 2013-05-09 21:41:43 <tumak> and translate that DSL into target implementations (using proven translators :)
2485 2013-05-09 21:41:46 <tumak> gmaxwell: yep :/
2486 2013-05-09 21:41:51 <cjd> nah
2487 2013-05-09 21:42:00 <cjd> you want to convert the spec into a set of tests
2488 2013-05-09 21:42:02 <petertodd> tumak: remember that bad performance can be a forking bug in itself
2489 2013-05-09 21:42:02 <nsh> there have been bugs found in proved programs
2490 2013-05-09 21:42:03 <gmaxwell> Otherwise we could just write the code as program for some simple turing machine, and then just make people prove their implementation of the turing machine is corret.
2491 2013-05-09 21:42:06 <gmaxwell> er correct.
2492 2013-05-09 21:42:09 <nsh> and they were undoubtedly much simpler than bitcoind
2493 2013-05-09 21:42:16 <tumak> cjd: basically, extend idea of BNF and parser generators to bitcoin
2494 2013-05-09 21:42:24 <gmaxwell> cjd: the input to the bitcoin function is infinite.
2495 2013-05-09 21:42:27 tyn has quit (Read error: Operation timed out)
2496 2013-05-09 21:42:31 <nsh> ^^
2497 2013-05-09 21:42:35 <tumak> not that things like peer discovery, wire protocol etc are irrelevant
2498 2013-05-09 21:42:40 <tumak> *note
2499 2013-05-09 21:42:50 <nsh> however, it may not be necessary to have complete confidence
2500 2013-05-09 21:42:58 <cjd> indeed but the work retarget function can have only 2**96 inputs
2501 2013-05-09 21:43:02 <nsh> probabilistic "correctness" may be better than no metric at all
2502 2013-05-09 21:43:05 <tumak> gmaxwell: i thought the crux of bitcoin protocol is simply to decide whether block is valid or not
2503 2013-05-09 21:43:19 <cjd> and you can pair it down further by looking for edge cases and writing a tester for it specifically
2504 2013-05-09 21:43:20 <gmaxwell> yes, you can test parts of it exaustively. (well 2^96 is a bit nuts, but that space can be reliably reduced)
2505 2013-05-09 21:43:26 <jgarzik> it's to arrive at a consensus on a timeline of transactions
2506 2013-05-09 21:43:55 <nsh> that's another strong nontriviality
2507 2013-05-09 21:44:00 <tumak> jgarzik: yep
2508 2013-05-09 21:44:10 <nsh> the correct operation of bitcoin is by necessity a funciton of multiple parties
2509 2013-05-09 21:44:11 <gmaxwell> cjd: you can use a linear relaxation to fully test the retarget without 2^96 operations.
2510 2013-05-09 21:44:14 <tumak> so validation and reorg code must be 100% exact same across implementations
2511 2013-05-09 21:44:27 <cjd> yeap
2512 2013-05-09 21:44:30 <jgarzik> no
2513 2013-05-09 21:44:35 <jgarzik> the result matters, not the code
2514 2013-05-09 21:44:40 <cjd> and then you have 9000 more little similar functions :|
2515 2013-05-09 21:44:51 <tumak> jgarzik: we want to be 100% sure the result is the same
2516 2013-05-09 21:44:59 <tumak> jgarzik: that can be achieved only using same code :(
2517 2013-05-09 21:45:11 rdponticelli has quit (Ping timeout: 248 seconds)
2518 2013-05-09 21:45:13 <jgarzik> same rules != same code
2519 2013-05-09 21:45:18 <gmaxwell> tumak: one of the funny things is that some failures are worse than others. And the badness depends on how common a node is.
2520 2013-05-09 21:45:43 <tumak> jgarzik: the rules are *defined* by the code
2521 2013-05-09 21:45:45 <gmaxwell> Yea, and as jeff says, we care about the function it evaluates not the implementation. some code transformations are harmless.... I mean, we don't mind the compiler's optimizations. :)
2522 2013-05-09 21:45:47 <tumak> they are not written on the wall
2523 2013-05-09 21:46:01 <tumak> or better said, defined by that everyone uses satoshi code
2524 2013-05-09 21:46:19 <jgarzik> Bitcoin Wallet users do not use satoshi's code
2525 2013-05-09 21:46:33 <gmaxwell> tumak: f(x)=1+x+2 == f(x)=2+x+1 ... (uh, so long as this is C, and x is an unsigned integer)
2526 2013-05-09 21:46:42 resinate has quit (Quit: resinate)
2527 2013-05-09 21:46:52 <jgarzik> anyway
2528 2013-05-09 21:46:53 guest354 has joined
2529 2013-05-09 21:46:54 <sipa> gmaxwell: it can be signed integer too, no? :)
2530 2013-05-09 21:46:58 <jgarzik> time for a major workstation upgrade
2531 2013-05-09 21:47:01 jgarzik has quit (Quit: poof)
2532 2013-05-09 21:47:20 <gmaxwell> sipa: signed overflow is undefined in C, so perhaps 1+x ends the universe wihle 2+x launches rockets
2533 2013-05-09 21:47:28 <sipa> Great.
2534 2013-05-09 21:47:44 <tumak> gmaxwell: yeah. now do that for something as non-linear as infinite state automata like bitcoin.
2535 2013-05-09 21:47:51 <Luke-Jr> gavinandresen: note that newer libboost on Linux wants the chronos lib
2536 2013-05-09 21:48:20 <gmaxwell> tumak: I suppose there are some ways to represent the state that make it finite. But intractably large, alas.
2537 2013-05-09 21:48:20 <gavinandresen> Luke-Jr: ah, thanks. I couldn't tell if it was newer or older boost that needs it....
2538 2013-05-09 21:48:40 <sipa> gavinandresen, Luke-Jr: does old boost maybe just lack that library?
2539 2013-05-09 21:49:03 <gavinandresen> maybeâ¦. or maybe it wasn't built by default
2540 2013-05-09 21:49:24 <sipa> chrono exists since boost 1.47
2541 2013-05-09 21:49:34 <gmaxwell> tumak: e.g. bitcoin(new_block,coins_database, sum_diff) = {new_coins_database, new_sum_diff} there can only be so many coins....
2542 2013-05-09 21:50:04 alexwaters1 has quit (Ping timeout: 264 seconds)
2543 2013-05-09 21:50:10 <sipa> is there an easy way to test what boost we're linking against from makefiles?
2544 2013-05-09 21:50:29 <tumak> gmaxwell: not disputing that
2545 2013-05-09 21:51:07 Xqr has quit (Read error: Connection reset by peer)
2546 2013-05-09 21:51:12 alexwaters has joined
2547 2013-05-09 21:51:25 rdponticelli has joined
2548 2013-05-09 21:52:03 <tumak> gmaxwell: however to previde 100% certainty two implementations will not fork one has to 1) use shared DSL to define all hardforking rules or 2) somehow check two different codebases are 100% equal
2549 2013-05-09 21:52:20 <cjd> ^^dependencies
2550 2013-05-09 21:52:21 <tumak> kinda reminds of the mess and compiler quirks regarding programming languages grammar :)
2551 2013-05-09 21:52:26 <cjd> BOOST_FOREACH()
2552 2013-05-09 21:52:33 <gmaxwell> tumak: in any case, there is no reason that the fact that our task is provably impossible should discourage us from continuing to be successful at it. :)
2553 2013-05-09 21:53:06 <tumak> gmaxwell: depends
2554 2013-05-09 21:53:09 <gmaxwell> tumak: 100% functionally equivalent. Right. And creating a provably correct compiler is basically at the current boundary of human capability right now.
2555 2013-05-09 21:53:10 xire has joined
2556 2013-05-09 21:53:25 <tumak> that bdb fork was kinda like early warning
2557 2013-05-09 21:53:50 <cjd> bdb caused a fork, openssl caused surprises
2558 2013-05-09 21:54:03 * cjd nominates boost as the next thing to cause headaches
2559 2013-05-09 21:54:10 <tumak> gmaxwell: most compilers use same grammar
2560 2013-05-09 21:54:20 <sipa> at least for openssl we know where it causes potential issues
2561 2013-05-09 21:54:20 <gmaxwell> tumak: it's not the first fork the network has, and we were aware of the risk. The 0.8 rewrite created several forking bugs. The difference is that we caught the other ones.
2562 2013-05-09 21:54:25 <sipa> at least, we think we know
2563 2013-05-09 21:54:26 <tumak> gmaxwell: if its sufficiently simple to be represented by BNF
2564 2013-05-09 21:54:42 <sipa> tumak: BNF doesn't cover semantics
2565 2013-05-09 21:54:47 <tumak> indeed
2566 2013-05-09 21:55:04 <tumak> (using it as oversimplification of DSL used to assure cooperation across implementations)
2567 2013-05-09 21:56:13 <cjd> I suspect the drive for performance will eventually push all logic out of the libraries and make bitcoin far more approachable
2568 2013-05-09 21:56:27 <Ry4an> is Andreas of bitcoin wallet (for android) fame in here?
2569 2013-05-09 21:56:32 <Ry4an> I've got a backup restore question.
2570 2013-05-09 21:56:35 <cjd> using sipasecp256k1 rather than openssl is a big step
2571 2013-05-09 21:56:39 <sipa> Ry4an: yes, nickname Goonie
2572 2013-05-09 21:56:47 <Ry4an> thanks, I couldn't remember the nick
2573 2013-05-09 21:56:53 <sipa> cjd: except my library still uses OpenSSL :p
2574 2013-05-09 21:56:57 <sipa> cjd: (or GMP)
2575 2013-05-09 21:57:03 <cjd> for which part?
2576 2013-05-09 21:57:06 <tumak> BN
2577 2013-05-09 21:57:13 <cjd> I thought it was all done manaully
2578 2013-05-09 21:57:17 <cjd> hmm
2579 2013-05-09 21:57:18 <gmaxwell> mostly the modular inverse.
2580 2013-05-09 21:57:33 <tumak> doing efficient portable BN by hand is kinda hard
2581 2013-05-09 21:57:38 <sipa> yeah, a fast builtin modular inverse would be nice
2582 2013-05-09 21:57:47 <tumak> sure its few hundreds of loc, but performace will suck
2583 2013-05-09 21:57:57 <tumak> because openssl can do stuff vectorized if you tell it to do so
2584 2013-05-09 21:57:57 <sipa> CodeShark has done some attempts, but they were all very slow compared to GMP
2585 2013-05-09 21:58:18 <cjd> you can't beat GMP/openssl for generic bignum
2586 2013-05-09 21:58:21 <sipa> (and OpenSSL is already slow compared to GMP)
2587 2013-05-09 21:58:26 <cjd> but for targetted stuff you can win
2588 2013-05-09 21:58:31 <sipa> as in: order of magnitude difference
2589 2013-05-09 21:58:52 <tumak> yea :(
2590 2013-05-09 21:58:57 <sipa> we only do one modular inverse for a verification
2591 2013-05-09 21:58:58 dvide has joined
2592 2013-05-09 21:59:02 <sipa> but if that inverse takes 20us
2593 2013-05-09 21:59:11 <CodeShark> GMP uses lehmer - I never tried implementing lehmer's algorithm
2594 2013-05-09 21:59:12 <sipa> en de rest of the validation is 90us
2595 2013-05-09 21:59:19 <sipa> then that is relevant :)
2596 2013-05-09 21:59:32 <tumak> writing custom sse3 inv is pointless because you have to do it again for neon arm
2597 2013-05-09 21:59:37 <tumak> and again for whatever architecture :(
2598 2013-05-09 21:59:38 <sipa> (20us is how long OpenSSL takes; GMP does it in 3us or so)
2599 2013-05-09 21:59:47 <CodeShark> I can beat openssl :)
2600 2013-05-09 21:59:48 <CodeShark> but not GMP
2601 2013-05-09 21:59:56 <sipa> i'm fine with 20us
2602 2013-05-09 22:00:04 <sipa> if that means we can drop a dependency
2603 2013-05-09 22:00:14 <Scrat> so, how hard is it to ditch openssl for DER stuff?
2604 2013-05-09 22:00:37 <cjd> even so, farming out a small job like that is much better than sending over a big script and saying "here validate it"
2605 2013-05-09 22:00:37 <sipa> Scrat: as long as DER compliance isn't enforced by the network: very hard
2606 2013-05-09 22:00:47 <tumak> yeah
2607 2013-05-09 22:00:50 <CodeShark> in principle it should be possible to beat GMP since we're using fixed width integers
2608 2013-05-09 22:00:59 <sipa> CodeShark: GMP does too
2609 2013-05-09 22:01:01 <cjd> s/script/sig/
2610 2013-05-09 22:01:07 <tumak> Scrat: you need full blown asn/ber openssl code until then
2611 2013-05-09 22:01:13 <sipa> CodeShark: mpn is fixed-size integers
2612 2013-05-09 22:01:15 <CodeShark> sipa: there's no dynamic allocation at all?
2613 2013-05-09 22:01:33 <tumak> CodeShark: take a look at vanitygen opencl fastinv
2614 2013-05-09 22:01:43 <sipa> CodeShark: i think mpn_gcdext does allocation internally
2615 2013-05-09 22:01:49 <sipa> CodeShark: but apart from that, nothing
2616 2013-05-09 22:01:56 <tumak> CodeShark: the code is really fast with 256 bit ints
2617 2013-05-09 22:02:03 <tumak> problem is, you cant portably parallelize it :(
2618 2013-05-09 22:02:05 <CodeShark> but gcdext is the meat of the inverse algorithm :)
2619 2013-05-09 22:02:26 <sipa> CodeShark: sure, you may be able to optimize that out still
2620 2013-05-09 22:02:44 <sipa> but one single allocation per 100us validation is tolerable and not noticable
2621 2013-05-09 22:02:46 <gmaxwell> sipa: did you figure out what your test failure was?
2622 2013-05-09 22:02:54 <sipa> gmaxwell: no, not yet
2623 2013-05-09 22:03:03 <sipa> but i assume it's a bug in the test
2624 2013-05-09 22:03:11 <sipa> looks like it, at least
2625 2013-05-09 22:03:13 <gmaxwell> CodeShark: that allocation probably ends up being two well predicted branches.
2626 2013-05-09 22:04:03 <CodeShark> GMP is hard to beat :)
2627 2013-05-09 22:04:41 michagogo has quit (Quit: goodnight)
2628 2013-05-09 22:04:55 <sipa> cjd: just to clarify: all field/group code is builtin and very optimized, but scalar operations are done using a bignum lib
2629 2013-05-09 22:05:17 <cjd> I see
2630 2013-05-09 22:06:30 <cjd> anyway each step which raises code up from out of the libraries is a step forward IMO
2631 2013-05-09 22:07:00 <CodeShark> we still need openssl for signing (as long as the wallet isn't separated from the verification/relay engine)
2632 2013-05-09 22:07:07 <cjd> because the people writing libraries are not thinking about consensus the way bitcoin is forced to
2633 2013-05-09 22:07:13 <CodeShark> because we need a sidechannel resistant implementation
2634 2013-05-09 22:07:33 <cjd> ideally the wallet would be seperate
2635 2013-05-09 22:07:40 <CodeShark> yes, indeed
2636 2013-05-09 22:07:41 <tumak> CodeShark: side channel is as easy as using correct point multiplication implementation for that
2637 2013-05-09 22:07:50 <sipa> i'm not convinced OpenSSL is more timing-attack resistant than my implementation...
2638 2013-05-09 22:08:00 <CodeShark> hehe, sipa :)
2639 2013-05-09 22:08:05 <sipa> (not saying the opossite either)
2640 2013-05-09 22:08:29 <CodeShark> you need constant time and constant cache access patterns
2641 2013-05-09 22:08:45 <cjd> this stuff starts to get a bit absurd
2642 2013-05-09 22:09:03 <CodeShark> it's entirely possible to write a nonbranching implementation of point multiplication ;
2643 2013-05-09 22:09:05 <CodeShark> :)
2644 2013-05-09 22:09:36 <cjd> when you start thinking about side channel attacks you soon lose all perspective and are imagining worlds where everybody in the world is an attacker and there is one honest person and he is inside of the computer and the attackers all want his key.......
2645 2013-05-09 22:09:39 lolcookie has quit (Ping timeout: 255 seconds)
2646 2013-05-09 22:09:54 resinate has joined
2647 2013-05-09 22:10:01 <CodeShark> and you can even do a nonbranching multiplicative inverse by using modular exponentiation (although it's slower than lehmer's by a couple orders)
2648 2013-05-09 22:10:08 <cjd> if you're building DRM, that's real, but if you're building DRM, you will lose always.
2649 2013-05-09 22:10:43 <ezdiy> lol @ timing attacks
2650 2013-05-09 22:10:47 <sipa> CodeShark: as far as i can see, OpenSSL uses wNAF for point multiplication, even for signing
2651 2013-05-09 22:11:03 <sipa> CodeShark: which is variable-time, as it depends on the number of non-zero wNAF elements
2652 2013-05-09 22:11:04 <CodeShark> wNAF isn't constant time, is it?
2653 2013-05-09 22:11:05 <CodeShark> right
2654 2013-05-09 22:11:12 <ezdiy> CodeShark: i mean, there will be always timing attacks against super-scalar architectures like x86 so whats the point
2655 2013-05-09 22:11:22 <sipa> CodeShark: and i don't use wNAF there
2656 2013-05-09 22:11:31 <ezdiy> if you collapse a branch, you'll hog one more integer unit or barrel shifter
2657 2013-05-09 22:11:48 <CodeShark> if you're really paranoid you should probably using a dedicated processor for signing transactions anyhow
2658 2013-05-09 22:11:57 <CodeShark> *probably BE using
2659 2013-05-09 22:12:37 agricocb has joined
2660 2013-05-09 22:12:39 <CodeShark> but it makes sense to at least up the difficulty of attacks wherever the cost to do so isn't too hard
2661 2013-05-09 22:12:44 seeingidog__ has quit (Quit: Leaving.)
2662 2013-05-09 22:12:46 <CodeShark> *isn't too high
2663 2013-05-09 22:13:19 <ezdiy> you mean to not fuck it up enough to make practical attacks ... well, uh, practical
2664 2013-05-09 22:13:30 <ezdiy> re: debian getpid() prng
2665 2013-05-09 22:13:42 <cjd> xD
2666 2013-05-09 22:14:07 <cjd> keep an eye out for the easy stuff
2667 2013-05-09 22:14:25 <CodeShark> the most typical attack on a user's priv keys will have little to do with timing :)
2668 2013-05-09 22:14:28 <cjd> wallet stealers and keystroke loggers don't care much about side channel attackability
2669 2013-05-09 22:14:32 <cjd> yeap
2670 2013-05-09 22:16:08 <sipa> CodeShark: i'm pretty sure that my code's potential timing attack vectors are a strict subset of OpenSSL'
2671 2013-05-09 22:16:12 <sipa> '
2672 2013-05-09 22:16:12 <sipa> 's
2673 2013-05-09 22:16:13 imnotamouse has joined
2674 2013-05-09 22:16:55 <CodeShark> I think the only semisensible argument in favor of something like OpenSSL is that it has been tried and tested and used in the field extensively
2675 2013-05-09 22:17:06 <sipa> sure, that's a very good reason
2676 2013-05-09 22:17:11 <CodeShark> other than that I'm 100% in favor of ditching it
2677 2013-05-09 22:17:29 <cjd> openssl is scary but I'd be more afraid of boost
2678 2013-05-09 22:18:20 <sipa> we don't use boost for crypto stuff
2679 2013-05-09 22:18:23 <sipa> THANK GOD
2680 2013-05-09 22:18:29 ngc0202 has quit (Read error: Connection reset by peer)
2681 2013-05-09 22:18:30 <cjd> boost_foreach is like a 1000 line header file and there's so much of that kind of thing that it makes security review on bitcoin practically impossible
2682 2013-05-09 22:19:00 <cjd> if there's a fork bug, we'll know when the chain forks...
2683 2013-05-09 22:19:24 ngc0202 has joined
2684 2013-05-09 22:19:25 imnotamouse has left ()
2685 2013-05-09 22:19:49 qwertyoruiop has quit (Ping timeout: 272 seconds)
2686 2013-05-09 22:20:41 <tumak> hmm, ideally, one we'd end up with code like djb's ed25519
2687 2013-05-09 22:20:54 <sipa> ed25519 is nice, indeed
2688 2013-05-09 22:21:14 <tumak> not much of reusable code though :(
2689 2013-05-09 22:21:47 <sipa> i should benchmark ed25519 implementations again, but i don't think it's that much faster than my ecdsa secp256k1 code
2690 2013-05-09 22:21:57 <sipa> (like, maybe still a factor 1.5 or 2)
2691 2013-05-09 22:21:58 qwertyoruiop has joined
2692 2013-05-09 22:22:01 <cjd> sipa: have you considered sending him your secp256k impl for inclusion in supercop?
2693 2013-05-09 22:22:10 <sipa> cjd: hmm, not really
2694 2013-05-09 22:22:31 <cjd> it would be cool and might be a starting point for working on different/faster implementations
2695 2013-05-09 22:24:47 nsillik has quit (Quit: nsillik)
2696 2013-05-09 22:25:18 <ezdiy> cjd: despite its volumicity, boost is rarely source of security bugs if used correctly
2697 2013-05-09 22:25:26 <ezdiy> if used correctly is the important part though :)
2698 2013-05-09 22:32:17 nsillik has joined
2699 2013-05-09 22:32:55 XertroV has joined
2700 2013-05-09 22:33:51 capuchin has quit (Ping timeout: 245 seconds)
2701 2013-05-09 22:34:00 <gmaxwell> cjd: yea, its ... troubling looking at lcov branch coverage output.. and there will be a line with some really trivial looking boost something on it... and next to it [+][-][-][-][-][-][-][-][-][-][-][-][-][-] like a dozen untested (hopefully unreachable?!) branches.
2702 2013-05-09 22:34:14 sud3n has joined
2703 2013-05-09 22:34:28 gruvfunk has quit (Ping timeout: 276 seconds)
2704 2013-05-09 22:35:53 <ezdiy> thats just templates being almighty
2705 2013-05-09 22:36:28 <ezdiy> m$ atl aint any better
2706 2013-05-09 22:37:17 XertroV has quit (Ping timeout: 252 seconds)
2707 2013-05-09 22:37:43 richcollins has quit (Quit: richcollins)
2708 2013-05-09 22:38:18 savetheinternet has quit (Quit: Computer has gone to sleep.)
2709 2013-05-09 22:39:26 andyh2 has joined
2710 2013-05-09 22:40:07 <cjd> wild idea: we should relay invalid transactions and blocks and (optionally) store them, also fork blocks should be optionally stored.
2711 2013-05-09 22:40:13 savetheinternet has joined
2712 2013-05-09 22:40:28 savetheinternet has quit (Client Quit)
2713 2013-05-09 22:40:46 <Luke-Jr> cjd: we don't even relay all valid transactions
2714 2013-05-09 22:40:51 capuchin has joined
2715 2013-05-09 22:40:59 <cjd> then every new impl can be run against the whole history of attacks to see if it passes anything which is otherwise failed
2716 2013-05-09 22:41:17 <cjd> yeah, the hard thing is not opening up new DoS attacks
2717 2013-05-09 22:42:14 andyh2 has quit (Client Quit)
2718 2013-05-09 22:42:59 richcollins has joined
2719 2013-05-09 22:43:04 seeingidog__ has joined
2720 2013-05-09 22:44:12 XertroV has joined
2721 2013-05-09 22:44:17 <tumak> cjd: how'd be that beneficial?
2722 2013-05-09 22:44:37 andyh2 has joined
2723 2013-05-09 22:44:48 <cjd> when you're writing your client, you can run it against the blockchain and the testnet chain to see a whole bunch of wacky stuff that it should allow
2724 2013-05-09 22:44:51 <tumak> i mean keeping orphans and invalid txes in the db
2725 2013-05-09 22:45:06 <cjd> but you can't run it against a big .dat file full of things which you should deny
2726 2013-05-09 22:45:07 <tumak> cjd: sure, just log the network for that purpose
2727 2013-05-09 22:45:18 <tumak> i cant see why regular clients should be bothered with that though
2728 2013-05-09 22:45:42 <cjd> I wonder if blockchain.info stores that stuff
2729 2013-05-09 22:45:48 <tumak> log the network = run a well connected node so youre likely to catch something strnage
2730 2013-05-09 22:45:53 <tumak> like blockchain.info does :)
2731 2013-05-09 22:46:31 <cjd> the other thing is there's an implication... "if you try to trick me, I'm going to store it and your 0day will be exposed"
2732 2013-05-09 22:46:37 <tumak> yeah, piuk's dumps would be interesting
2733 2013-05-09 22:47:06 brwyatt is now known as Away!~brwyatt@brwyatt.net|brwyatt
2734 2013-05-09 22:47:07 <tumak> cjd: oh, if you mean client based exploits thats different
2735 2013-05-09 22:47:14 <tumak> those things propagate well
2736 2013-05-09 22:47:15 <tumak> :)
2737 2013-05-09 22:47:17 Diapolis has joined
2738 2013-05-09 22:47:36 <cjd> well
2739 2013-05-09 22:47:40 <tumak> cjd: but propagating anything against the rules is kinda pointless
2740 2013-05-09 22:47:43 <tumak> since its harmless
2741 2013-05-09 22:47:57 <cjd> but is it against *your* rules?
2742 2013-05-09 22:48:03 <tumak> only makes sense for double spends in cases like a website which tries to point them out
2743 2013-05-09 22:48:10 <tumak> which is exactly what piuk does
2744 2013-05-09 22:48:52 <cjd> one idea is to have a history of all blockchain schenanigans ever and be able to play them through your alt-impl
2745 2013-05-09 22:48:53 <tumak> cjd: well, lets say the 5000 satoshi min tx
2746 2013-05-09 22:48:58 <tumak> suppose there is client which forwards it
2747 2013-05-09 22:49:00 <cjd> and make sure you are compliant
2748 2013-05-09 22:49:06 <tumak> which is against "rules" imposed by new clients
2749 2013-05-09 22:49:09 <tumak> how is that evil?
2750 2013-05-09 22:49:14 Silverion has quit (Ping timeout: 260 seconds)
2751 2013-05-09 22:49:18 <cjd> it's not, that's not a hard rule
2752 2013-05-09 22:49:21 <tumak> its just an outdated client, or just using a setting with lower limit
2753 2013-05-09 22:49:36 <tumak> cjd: so you mean block validation rules, not just mempool rule
2754 2013-05-09 22:49:45 seeingidog__ has quit (Ping timeout: 252 seconds)
2755 2013-05-09 22:49:54 <tumak> block validation will always cause hard fork if you have the hash power to do it, as simple as that
2756 2013-05-09 22:49:58 rdymac has joined
2757 2013-05-09 22:50:04 <cjd> one (truely) invalid tx makes the whole block invalid
2758 2013-05-09 22:50:13 <tumak> yup
2759 2013-05-09 22:50:20 <cjd> those are the ones I care about
2760 2013-05-09 22:50:31 <tumak> for example, input and output amounts not matching
2761 2013-05-09 22:50:37 <tumak> or fees and coinbase not matching
2762 2013-05-09 22:50:44 <tumak> so lets say you propagate those txes
2763 2013-05-09 22:50:47 <tumak> and then what?
2764 2013-05-09 22:50:59 <cjd> tag them as invalid and propigate
2765 2013-05-09 22:51:10 <cjd> anyone who wants to store them can
2766 2013-05-09 22:51:12 Namworld has joined
2767 2013-05-09 22:51:42 <tumak> too much open to dos imo :(
2768 2013-05-09 22:51:48 mrkent has quit (Quit: Leaving)
2769 2013-05-09 22:51:52 <tumak> since propagation is directed by mempool rules exactly for that reason
2770 2013-05-09 22:52:17 <tumak> you'd simply bypass mempool just by doing your "hey look, i found something strange"
2771 2013-05-09 22:52:37 <rdymac> Hello, Does anyone knows the email of timoncc?
2772 2013-05-09 22:52:37 <tumak> soon everyone would be only propagating strange txes for the sake of it :/
2773 2013-05-09 22:53:21 <cjd> the problem is if everybody forgets about these things, there's no way to build a client without copying and unit testing code line by line
2774 2013-05-09 22:56:40 nomailing1 has quit (Quit: nomailing1)
2775 2013-05-09 22:57:25 <tumak> cjd: well, testnet accepts non-standard txes
2776 2013-05-09 22:57:43 <tumak> and there is indeed lots of nonstandard blocks in there
2777 2013-05-09 22:57:46 emryss has joined
2778 2013-05-09 22:58:13 andyh2 has quit (Quit: Leaving...)
2779 2013-05-09 22:58:20 <tumak> but isStandard is only soft rule, there is indeed nothing for hard rules
2780 2013-05-09 22:58:28 <tumak> other than doing it on your own via well connected node
2781 2013-05-09 22:59:26 <cjd> right so the tempting thing to do is write your validating node, sync testnet and then be like awesome, it works!
2782 2013-05-09 22:59:55 <cjd> paying no attention to all of the possible things that you should be throwing out
2783 2013-05-09 23:00:55 SteveDekorte has quit (Quit: SteveDekorte)
2784 2013-05-09 23:01:00 <tumak> well, the testnet is step 1
2785 2013-05-09 23:01:11 <tumak> step 2 is bluematts block fuzzer
2786 2013-05-09 23:01:32 <tumak> step 3 is something fucks up and users scream bloody murder :/
2787 2013-05-09 23:01:40 <cjd> right
2788 2013-05-09 23:01:53 <cjd> the fuzzer is 1441 LoC, it's awesome but it's also very incomplete
2789 2013-05-09 23:02:12 <tumak> if you can think of things to add, please do so
2790 2013-05-09 23:02:23 <tumak> but its mostly used to test reorg code more than anything
2791 2013-05-09 23:02:41 <cjd> we need something as wacky as testnet except containing only transactions and blocks which should be thrown out
2792 2013-05-09 23:02:50 <tumak> hm
2793 2013-05-09 23:02:57 qbasicer has quit (Remote host closed the connection)
2794 2013-05-09 23:03:13 <tumak> maybe mod bitcoin-qt to NOT throw weird things out when running on testnet
2795 2013-05-09 23:03:19 <cjd> hehe
2796 2013-05-09 23:03:20 <tumak> and instead, randomly re-propagate those
2797 2013-05-09 23:03:40 <cjd> no, it needs to mark them as invalid and then add them to a bad-bin
2798 2013-05-09 23:03:51 <cjd> and then the bad-bin is used to test new impls
2799 2013-05-09 23:04:02 <tumak> well, there is no such thing as bad bin
2800 2013-05-09 23:04:10 <cjd> re proposal
2801 2013-05-09 23:04:11 <cjd> :)
2802 2013-05-09 23:04:14 <tumak> but simply the fact that branches which lost are pruned
2803 2013-05-09 23:04:30 <tumak> so its as simple as disabling pruning of branches which lost
2804 2013-05-09 23:04:39 <tumak> be it due to orphan, double spend or something else
2805 2013-05-09 23:04:51 <tumak> then, when somebody connects to you and ask for block range
2806 2013-05-09 23:05:02 <tumak> you will feed him some of those lost branches at random
2807 2013-05-09 23:05:04 rdymac has quit (Ping timeout: 264 seconds)
2808 2013-05-09 23:05:20 qbasicer has joined
2809 2013-05-09 23:05:23 one_zero has joined
2810 2013-05-09 23:05:36 <cjd> "accept conservitively, liberally emit to make sure the other guy is doing the same ;)"
2811 2013-05-09 23:05:41 <tumak> that way, testnet will soon explode in entropy death of forks
2812 2013-05-09 23:06:34 <tumak> cjd: but i think its interesting idea
2813 2013-05-09 23:06:44 <tumak> there should be some % of how much brokeness is kept/propagated
2814 2013-05-09 23:06:51 <cjd> mhm
2815 2013-05-09 23:06:58 <tumak> to rate limit smart-asses flooding the network with crap
2816 2013-05-09 23:07:14 <tumak> (to the point of saturating bandwith)
2817 2013-05-09 23:07:16 iwilcox is now known as iwiIcox
2818 2013-05-09 23:07:23 iwiIcox is now known as iwilcox
2819 2013-05-09 23:09:26 <tumak> cjd: such a testnet setup would then be a huge maze of broken chains which you must navigate through to sync the correct state
2820 2013-05-09 23:10:29 <tumak> txes would be reversed often as they'd often be propagated only into dead branch etc
2821 2013-05-09 23:10:47 <tumak> kinda continuous bdb forks occuring at random :)
2822 2013-05-09 23:11:21 DoctorBTC has joined
2823 2013-05-09 23:13:06 iwilcox is now known as gribb|e
2824 2013-05-09 23:13:15 gribb is now known as e!~iwilcox@unaffiliated/iwilcox|iwilcox
2825 2013-05-09 23:19:10 <cjd> tumak: actually I'd just shovel all invalid crap at each block height into a pile
2826 2013-05-09 23:20:09 <cjd> but yes, you'd have to tree trace down each fork and back
2827 2013-05-09 23:20:28 tyn has joined
2828 2013-05-09 23:24:04 alexwaters has quit (Quit: Leaving.)
2829 2013-05-09 23:24:49 <sipa> if you just want to test invalid transactions, add them to src/tests/data/tx_invalid.json
2830 2013-05-09 23:24:55 Michail1 is now known as Michail1_
2831 2013-05-09 23:25:34 zooko_ has quit (Ping timeout: 264 seconds)
2832 2013-05-09 23:26:25 nsillik_ has joined
2833 2013-05-09 23:26:58 zooko_ has joined
2834 2013-05-09 23:27:01 <tumak> well, the idea is to have some kind of living system which continuosly keeps certain error rate
2835 2013-05-09 23:27:17 <tumak> synthetic tests are imo limited compared to that
2836 2013-05-09 23:27:30 nsillik has quit (Read error: Operation timed out)
2837 2013-05-09 23:27:30 nsillik_ is now known as nsillik
2838 2013-05-09 23:27:38 <tumak> maybe even fault injection might be interesting
2839 2013-05-09 23:27:44 GordonG3kko has joined
2840 2013-05-09 23:31:01 CodesInChaos has quit (Ping timeout: 276 seconds)
2841 2013-05-09 23:32:28 Luke-Jr has quit (Read error: Connection reset by peer)
2842 2013-05-09 23:34:54 Luke-Jr has joined
2843 2013-05-09 23:39:48 defunctzombie is now known as defunctzombie_zz
2844 2013-05-09 23:47:43 seeingidog__ has joined
2845 2013-05-09 23:49:25 chorao has quit (Quit: ja volto)
2846 2013-05-09 23:52:04 seeingidog__ has quit (Ping timeout: 252 seconds)
2847 2013-05-09 23:52:40 rdymac has joined
2848 2013-05-09 23:54:51 btcven has joined
2849 2013-05-09 23:55:33 rdymac has quit (Remote host closed the connection)
2850 2013-05-09 23:55:40 btcven has quit (Client Quit)
2851 2013-05-09 23:56:01 rdymac has joined
2852 2013-05-09 23:57:53 seongyupyoo has joined