1 2013-02-01 00:03:57 dvide has quit ()
   2 2013-02-01 00:04:29 novusordo has joined
   3 2013-02-01 00:08:46 novusordo is now known as Guest44214
   4 2013-02-01 00:10:11 agricocb has joined
   5 2013-02-01 00:11:10 num1 has joined
   6 2013-02-01 00:14:38 tomboy64 has joined
   7 2013-02-01 00:16:44 Lyspooner has joined
   8 2013-02-01 00:17:18 MC1984 has quit (Ping timeout: 248 seconds)
   9 2013-02-01 00:18:32 tcatm has quit (Quit: No Ping reply in 180 seconds.)
  10 2013-02-01 00:18:49 tcatm has joined
  11 2013-02-01 00:18:49 tcatm has quit (Changing host)
  12 2013-02-01 00:18:49 tcatm has joined
  13 2013-02-01 00:21:31 RBecker is now known as rbecker
  14 2013-02-01 00:21:55 Maged has joined
  15 2013-02-01 00:22:14 <rbecker> slush: is your pool down? :)
  16 2013-02-01 00:22:32 <rbecker> [2013-01-31 19:04:06] Switching pool 0 http://mining.bitcoin.cz:8332 to stratum+tcp://stratum.bitcoin.cz:3333
  17 2013-02-01 00:22:32 <rbecker> [2013-01-31 19:04:06] JSON stratum auth failed: (unknown reason)
  18 2013-02-01 00:22:33 <rbecker> [2013-01-31 19:04:06] Unable to get work from pool 0 http://mining.bitcoin.cz:8332
  19 2013-02-01 00:22:47 <sipa> it's been overasiced? :(
  20 2013-02-01 00:23:01 <slush> rbecker: no, but you obviously use bad credentials :)
  21 2013-02-01 00:23:17 <slush> oh
  22 2013-02-01 00:23:19 <rbecker> i didn't, they're the same creds that've been in cgminer all along
  23 2013-02-01 00:23:30 <slush> rbecker: mining.bitcoin.cz is not a mining address
  24 2013-02-01 00:23:40 <rbecker> it's always worked before
  25 2013-02-01 00:23:45 <rbecker> and it did say it was doing a stratum switch
  26 2013-02-01 00:23:50 <slush> rbecker: no, it doesn't work for many months
  27 2013-02-01 00:23:57 <rbecker> been working fine for me recently
  28 2013-02-01 00:24:01 <rbecker> anyways, i gotta go
  29 2013-02-01 00:24:28 <slush> rbecker: check the credentials, that "unknown reason" appears when people use bad credentials
  30 2013-02-01 00:24:54 <slush> rbecker: and really change your URL to api.bitcoin.cz:8332
  31 2013-02-01 00:26:24 spenvo has joined
  32 2013-02-01 00:26:34 MC1984 has joined
  33 2013-02-01 00:30:08 spenvo has quit (Client Quit)
  34 2013-02-01 00:30:10 zooko` is now known as zooko
  35 2013-02-01 00:30:51 spenvo has joined
  36 2013-02-01 00:31:42 gargar has quit (Ping timeout: 245 seconds)
  37 2013-02-01 00:37:56 <agath> I hope to not be boring asking a newbie question but.. does anyone know where I can find an example source code (C? Ruby?) of a software mining client?
  38 2013-02-01 00:38:15 <agath> or a document that explains the mining algorhythm in detail?
  39 2013-02-01 00:38:56 dvide has joined
  40 2013-02-01 00:39:18 Hasimir- has joined
  41 2013-02-01 00:40:09 <andytoshi> i've never heard of a miner without source code
  42 2013-02-01 00:40:12 knotwork has joined
  43 2013-02-01 00:40:20 <andytoshi> the default client has a CPU miner, which is probably the simplest thing you'll find
  44 2013-02-01 00:40:40 <andytoshi> git://github.com/bitcoin/bitcoin.git
  45 2013-02-01 00:40:46 Hasimir has quit (Ping timeout: 248 seconds)
  46 2013-02-01 00:42:38 RainbowDashh has joined
  47 2013-02-01 00:42:52 D34TH_ has joined
  48 2013-02-01 00:43:28 D34TH has quit (Read error: Connection reset by peer)
  49 2013-02-01 00:44:02 D34TH_ is now known as D34TH
  50 2013-02-01 00:44:09 D34TH has quit (Changing host)
  51 2013-02-01 00:44:09 D34TH has joined
  52 2013-02-01 00:45:08 spenvo has quit (Quit: This computer has gone to sleep)
  53 2013-02-01 00:46:43 CodeShark has joined
  54 2013-02-01 00:48:14 <andytoshi> heya CodeShark, any luck with autotools?
  55 2013-02-01 00:50:18 tucenaber has joined
  56 2013-02-01 00:50:43 <CodeShark> andytoshi: made some progress but still having a few minor issues
  57 2013-02-01 00:52:41 andytoshi has quit (Quit: WeeChat 0.3.9.2)
  58 2013-02-01 00:52:46 <CodeShark> I actually rather like my hand-rolled configure...but maintenance will be an issue. bash doesn't exactly encourage the cleanest of programming styles :)
  59 2013-02-01 00:55:31 <HM> gah autotools. filth
  60 2013-02-01 00:55:39 <Luke-Jr> agath: it's literally just SHA256d..
  61 2013-02-01 00:56:39 rbecker is now known as RBecker
  62 2013-02-01 00:57:52 andytoshi has joined
  63 2013-02-01 00:58:51 dlb76 has joined
  64 2013-02-01 01:04:13 <agath> yes I was aware that it's SHA256, but I was wondering about the data involved, I'm reading some stuff... thanks for the suggestions :)
  65 2013-02-01 01:07:11 <doublec> agath: maybe https://github.com/jgarzik/pyminer
  66 2013-02-01 01:08:42 zooko has quit (Ping timeout: 276 seconds)
  67 2013-02-01 01:08:48 <agath> ty
  68 2013-02-01 01:11:01 spenvo has joined
  69 2013-02-01 01:11:22 mappum has joined
  70 2013-02-01 01:12:35 <Luke-Jr> agath: https://en.bitcoin.it/wiki/Block_hashing_algorithm
  71 2013-02-01 01:24:44 <num1> Say I were a developer who wanted to start contributing to bitcoin and learn more about the network, where would be a good place to start?
  72 2013-02-01 01:26:03 <CodeShark> here
  73 2013-02-01 01:26:39 <CodeShark> first master this document: https://en.bitcoin.it/wiki/Protocol_specification
  74 2013-02-01 01:27:42 <copumpkin> in the avalon review, he says that the network will become far more secure
  75 2013-02-01 01:27:46 <copumpkin> I'm not sure I buy that
  76 2013-02-01 01:29:12 <Scrat> copumpkin: if ASICs are the last stage of the evolution then the network will be more secure if everyone has access to them
  77 2013-02-01 01:29:47 <Scrat> whereas previously only a well funded attacker could have done that
  78 2013-02-01 01:29:53 <Scrat> thats the way I see it
  79 2013-02-01 01:29:59 <sipa> i agree
  80 2013-02-01 01:29:59 <copumpkin> well, my point is that a well funded attacker can just buy a fuckton of avalon ascis now
  81 2013-02-01 01:30:03 <num1> As far as code goes there's not a big body to learn about right? You have the official client which talks on the network to gossip about blocks and transactions, mines, and verifies blocks.
  82 2013-02-01 01:30:15 <copumpkin> so the determining factor is the distribution of the commonly available units
  83 2013-02-01 01:30:20 <copumpkin> and not their power
  84 2013-02-01 01:30:26 <copumpkin> or is that wrong?
  85 2013-02-01 01:30:45 <num1> Are there any other big things I'm missing or popular alternative implementations?
  86 2013-02-01 01:30:47 <sipa> num1: the code of the reference client is not that much, but it's not easy to understand all details (and it's not written very elegantly)
  87 2013-02-01 01:31:37 <Scrat> their supply is limited
  88 2013-02-01 01:31:39 TD has quit (Quit: TD)
  89 2013-02-01 01:31:48 <sipa> copumpkin: agree, it's the distribution that is relevant
  90 2013-02-01 01:32:01 D34TH has quit (Quit: Leaving)
  91 2013-02-01 01:32:12 <copumpkin> Scrat: but they're going to turn it up if they get a huge order. It might be lagged, but that's how markets work
  92 2013-02-01 01:32:23 <sipa> copumpkin: but consider this: 1) the evolution to specialized chips was inevitable  2) the alternative was a cartel of miners investing in chips that would only be available to them
  93 2013-02-01 01:32:37 <copumpkin> yeah
  94 2013-02-01 01:32:43 <copumpkin> so it is improving things
  95 2013-02-01 01:33:01 <sipa> you may argue whether it's an improvement versus the current situation
  96 2013-02-01 01:33:03 <copumpkin> because we're getting closer to state of the art and are likely unable to squeeze major increments out past this point
  97 2013-02-01 01:33:12 <Scrat> it will be just like gpus/fpgs. the difference is that there wont be a huge performance jump after that, only incremental increases
  98 2013-02-01 01:33:13 <sipa> but it's certainly the lesser evil of two possible futures
  99 2013-02-01 01:34:05 TD has joined
 100 2013-02-01 01:34:12 TD has quit (Client Quit)
 101 2013-02-01 01:34:37 <Scrat> although one might argue that going from 110 nm to 3x nm is more than a generational increase
 102 2013-02-01 01:34:50 <sipa> BFL's chips are (allegedly) 65nm, iirc
 103 2013-02-01 01:40:20 drizzt_ has quit (Quit: Konversation terminated!)
 104 2013-02-01 01:41:19 tradefortress has joined
 105 2013-02-01 01:41:33 <tradefortress> is there any way to find out which pool relayed a block? it seems like Deepbit & slush are relaying from the same IP
 106 2013-02-01 01:41:47 <sipa> you know what 'relay' means?
 107 2013-02-01 01:42:05 <sipa> it's very unlikely that they do not both relay it at some point
 108 2013-02-01 01:43:24 <sipa> if you want to know who produced it: no way of being certain
 109 2013-02-01 01:43:51 <sipa> blockchain.info uses the IP they first saw a block being relayed from as a way for determining the source
 110 2013-02-01 01:44:07 <sipa> this is however known to be frequently wrong
 111 2013-02-01 01:44:44 <tradefortress> I don't think it determines it with relayed by
 112 2013-02-01 01:44:56 <tradefortress> because a "Deepbit" block and a "Slush" block had the same IP
 113 2013-02-01 01:45:00 <tradefortress> yet it says they're from different pools.
 114 2013-02-01 01:45:19 <sipa> ah, maybe it now retrieves information about blocks from the pools themself
 115 2013-02-01 01:45:36 <doublec> or from information in the coinbase
 116 2013-02-01 01:45:45 <slush> tradefortress: that IP is just an IP of node who firstly relayed it
 117 2013-02-01 01:46:06 <CodeShark> no, it's the IP address of the first computer that blockchain.info got it from :)
 118 2013-02-01 01:46:12 <CodeShark> doesn't mean it was the first to relay it
 119 2013-02-01 01:46:46 <gmaxwell> it does multiple things ... which just makes its failure modes more confusing.
 120 2013-02-01 01:47:43 RainbowDashh has quit (Remote host closed the connection)
 121 2013-02-01 01:47:50 zooko has joined
 122 2013-02-01 01:47:59 RainbowDashh has joined
 123 2013-02-01 01:48:14 PhantomSpark has joined
 124 2013-02-01 01:48:22 <slush> CodeShark: you're right, I mean "..who firstly relayed it to blockchain.info"
 125 2013-02-01 01:48:31 PhantomSpark has quit (Client Quit)
 126 2013-02-01 01:49:14 <sipa> someone should build a faster relayer that is connected to everyone, and relay everything to blockchain.info first
 127 2013-02-01 01:49:24 <sipa> just to screw up the statistics :p
 128 2013-02-01 01:49:42 <tradefortress> kay, so I guess I can't find out which pool mined a block :/
 129 2013-02-01 01:50:03 <sipa> tradefortress: just go to the websites of the largest pools and see whether they mention the block?
 130 2013-02-01 01:50:17 <doublec> tradefortress: try http://blockorigin.pfoe.be/blocklist.php
 131 2013-02-01 01:50:25 <doublec> tradefortress: which does what sipa suggests
 132 2013-02-01 01:50:29 <doublec> tradefortress: and aggregates it
 133 2013-02-01 01:50:31 <gmaxwell> tradefortress: depends on what you need to know it for...
 134 2013-02-01 01:50:47 <tradefortress> ah cool, what about deepbit delaying it by 1 hour?
 135 2013-02-01 01:50:57 <gmaxwell> if e.g. you need to know it in a way which is robust against pools _lying_ then there is no way to know for sure.
 136 2013-02-01 01:51:35 <doublec> tradefortress: the pools that fdelay show as unknown on that site until the pool publishes the data
 137 2013-02-01 01:51:51 <moore> sipa if you could build a faster relayer you could just "cheat" at satoshidice.com
 138 2013-02-01 01:52:01 <moore> that would be even more fun
 139 2013-02-01 01:52:25 spenvo has quit (Quit: This computer has gone to sleep)
 140 2013-02-01 01:54:23 spenvo has joined
 141 2013-02-01 01:55:39 <HM>  <copumpkin> well, my point is that a well funded attacker can just buy a fuckton of avalon ascis now
 142 2013-02-01 01:56:06 PhantomSpark has joined
 143 2013-02-01 01:56:42 <HM> copumpkin: the difficulty will rise if that happens making it unviable except for anyone intent on destroying bitcoin regardless of cost
 144 2013-02-01 01:57:00 <HM> copumpkin: and there are easier ways to destroy bitcoin if someone powerful enough wanted to
 145 2013-02-01 01:57:20 <HM> or at least set it bacl
 146 2013-02-01 01:57:22 <HM> back
 147 2013-02-01 01:57:55 <HM> besides Avalon have no incentive to flood the market and risk an upset if they want their business to flourish in the medium to long term
 148 2013-02-01 01:59:03 <num1> moore, really? I guess I don't know enough about how satoshi dice work, but why would that let you cheat?
 149 2013-02-01 01:59:35 <moore> well it might not work but
 150 2013-02-01 01:59:48 <moore> they way they do stuff is you send them a trasaction
 151 2013-02-01 02:00:06 <moore> then they send you one back using your output as a input
 152 2013-02-01 02:00:28 <moore> then if you don't like the way the dice came out
 153 2013-02-01 02:01:00 <moore> and you can flood the a new trasaction to the network that spends the same output used in the inital trasaction you sent them
 154 2013-02-01 02:01:13 <moore> your bet with them will get "rolled back"
 155 2013-02-01 02:01:14 <CodeShark> the only problem is that the first transaction you sent has already been propagated
 156 2013-02-01 02:01:28 <CodeShark> the only way that could work is by chaining a bunch of transactions
 157 2013-02-01 02:01:30 <moore> how do they verify that?
 158 2013-02-01 02:01:55 <CodeShark> by the time satoshi dice sends their transaction to the network, your transaction to them has already propagated
 159 2013-02-01 02:01:56 <moore> if all the nodes they relay too are slow you could still win
 160 2013-02-01 02:02:13 <sipa> well ideally you send the first transaction directly to them, and you get the reply back from them directly as well
 161 2013-02-01 02:02:30 <CodeShark> but presumably they will have already relayed your transaction to them to a bunch of other peers, sipa
 162 2013-02-01 02:02:35 <sipa> during that time, they can make sure they have informed a large number of peers
 163 2013-02-01 02:02:41 <moore> there seems to be people doing/trying this now
 164 2013-02-01 02:02:43 num1 has quit (Quit: Ex-Chat)
 165 2013-02-01 02:02:51 <moore> http://blockchain.info/double-spends
 166 2013-02-01 02:03:02 <sipa> actually, they could intentionally delay the response, in order to give it more chance to propagate
 167 2013-02-01 02:03:03 zooko` has joined
 168 2013-02-01 02:03:08 <sipa> moore: there are several ways to help
 169 2013-02-01 02:03:18 <moore> ya I agree
 170 2013-02-01 02:03:22 <moore> but are they now?
 171 2013-02-01 02:03:36 <HM> satoshi can fix that easier than you can exploit it though
 172 2013-02-01 02:03:38 <sipa> no idea, but a few seconds would do a lot already
 173 2013-02-01 02:03:45 <HM> err satoshidice that is
 174 2013-02-01 02:03:47 zooko has quit (Ping timeout: 264 seconds)
 175 2013-02-01 02:04:08 <moore> if you look at that list of double spends they are overwhelmingly SatoshiDICE
 176 2013-02-01 02:04:31 <HM> all transactions are overwhelmingly SatoshiDICE :/
 177 2013-02-01 02:04:53 <CodeShark> so maybe they haven't patched up this hole
 178 2013-02-01 02:04:57 <moore> also one of the big pools could game them too
 179 2013-02-01 02:05:14 <gmaxwell> Kinda telling(?) that when people complain to SD about SD burdening the network they get a "so? you better fix it" but when someone attacks sd and sd doesn't even care people are looking out for ways for them to fix themselves.
 180 2013-02-01 02:05:30 <moore> just refuse loosing transactions
 181 2013-02-01 02:05:41 EPiSKiNG- has joined
 182 2013-02-01 02:05:43 <tradefortress> http://blockorigin.pfoe.be/blocklist.php is outdated?
 183 2013-02-01 02:05:49 <sipa> moore: or block them altogether :)
 184 2013-02-01 02:05:51 <gmaxwell> moore: lots of miners already refuse to mine SD transactions— wouldn't be an enormous patch to only mine when SD loses.... 0_o
 185 2013-02-01 02:06:01 <doublec> tradefortress: delayed I think
 186 2013-02-01 02:06:23 <CodeShark> that would be hillarious, gmaxwell :)
 187 2013-02-01 02:06:27 Zarutian has quit (Quit: Zarutian)
 188 2013-02-01 02:06:31 <moore> gmaxwell that would be a good way to make them go away
 189 2013-02-01 02:07:04 <gmaxwell> moore: or to stop announcing their losses as transactions, which is half of what some people irritated with them want. Those very tiny outputs are .. ugh.
 190 2013-02-01 02:07:06 <doublec> is there a patch for refusing SD transactions?
 191 2013-02-01 02:08:02 <HM> SD could increase the fees to compensate miners, or do a deal with one of the mining groups
 192 2013-02-01 02:08:07 <gmaxwell> doublec: yes. I'm on slow internet now so I can't go find one of the public ones, but IIRC luke has a branch for example.
 193 2013-02-01 02:08:08 <HM> make them bid for it
 194 2013-02-01 02:08:38 <jaakkos> moore: interesting attack on satoshidice. perhaps if the attacker aggressively connects to a huge number of nodes, they could poison the network with the second transaction easily because it starts propagating faster than the original one?
 195 2013-02-01 02:08:43 <gmaxwell> HM: no the point is that if you don't mine their txn— even if other miners do— those delays make double spends easier.
 196 2013-02-01 02:09:22 <HM> you mean if you want to be malicious?
 197 2013-02-01 02:09:27 <gmaxwell> every block mined by a miner which blocks X is a block which might instead contain a later doublespend against X.
 198 2013-02-01 02:09:40 <gmaxwell> I don't even mean malicious— it's just the natural effect of blocking something
 199 2013-02-01 02:09:44 <doublec> gmaxwell: thanks, found Luke-Jr's branch
 200 2013-02-01 02:10:00 <gmaxwell> If you blcok something that improves the odds of a not-something doublespend marginally.
 201 2013-02-01 02:10:31 <HM> right, you mean if a spend isn't seen by the full network hash power then there is an increased time window for a double spend
 202 2013-02-01 02:10:52 <gmaxwell> HM: and take care with malice, SD's responses to complaints have been 'well, if its a burden you should go fix that'—  if their spammy txn getting denied is a problem then perhaps an inkind respond is due.
 203 2013-02-01 02:11:05 <moore> the issue hear is that the SatoshiDICE trick to be fast is actually not really safe
 204 2013-02-01 02:11:06 <gmaxwell> s/respond/response/
 205 2013-02-01 02:11:24 <HM> i don't think malice is an issue, i think the free market should handle it
 206 2013-02-01 02:11:28 <HM> somehow
 207 2013-02-01 02:11:32 <gmaxwell> moore: yea, and also has a lot of costs on the network.
 208 2013-02-01 02:11:40 <moore> you can't be sure of the ordering of transactions until they are in a block
 209 2013-02-01 02:12:01 <moore> well if the network wants them to stop it is clear how to do it
 210 2013-02-01 02:12:19 <gmaxwell> so you get unsafe and very greatly increased txn volume...  it also has misled a bunch of people about the safty of unconfirmed txn, but I guess it might ultimately have the reverse effect.
 211 2013-02-01 02:12:24 <moore> well I guess they could just be more tricky but it would hurt there UX
 212 2013-02-01 02:12:38 <gmaxwell> moore: network wide collusion opens up other hard questions.
 213 2013-02-01 02:12:40 EPiSKiNG- has quit (Read error: Connection reset by peer)
 214 2013-02-01 02:12:53 <gmaxwell> Though special cases that make themselves flaming identifyable .. well thats kinda special.
 215 2013-02-01 02:12:58 <moore> gmaxwell, yes
 216 2013-02-01 02:13:08 <HM> interesting question
 217 2013-02-01 02:13:18 <HM> what's the biggest spender on the network?
 218 2013-02-01 02:13:23 <gmaxwell> moore: did you see that they wouldn't even change to compressed public keys because they were (apparently?) so determined not to change addresses. Weird.
 219 2013-02-01 02:13:24 <HM> 2nd biggest*
 220 2013-02-01 02:13:37 paraipan has quit (Quit: Saliendo)
 221 2013-02-01 02:13:48 <moore> hu
 222 2013-02-01 02:13:49 <moore> no
 223 2013-02-01 02:13:56 <moore> I gotta take off
 224 2013-02-01 02:13:58 <gmaxwell> HM: obviously you can't identify single users on the network (usually)... so what do you really mean by spender?
 225 2013-02-01 02:14:00 <moore> nice chatting
 226 2013-02-01 02:14:05 <gmaxwell> Indeed.
 227 2013-02-01 02:14:06 dvide has quit (Ping timeout: 264 seconds)
 228 2013-02-01 02:14:37 <HM> well SatoshiDICE is well tracked
 229 2013-02-01 02:15:01 <HM> are there any other large known groups pushing out anything comparable?
 230 2013-02-01 02:15:02 <gmaxwell> maybe. Again what do you mean by spender?
 231 2013-02-01 02:15:14 <HM> number of transactions
 232 2013-02-01 02:15:27 <jaakkos> do miners always accept the transaction that they see first, if multiple alternatives are available?
 233 2013-02-01 02:15:29 <gmaxwell> If I just send to myself a million times am I the biggest "spender"?
 234 2013-02-01 02:15:34 <sipa> why do you classify by? IP? computer? wallet?
 235 2013-02-01 02:15:36 EPiSKiNG- has joined
 236 2013-02-01 02:15:47 <gmaxwell> jaakkos: they reject doublespends against txn they've already accepted.
 237 2013-02-01 02:15:56 <Scrat> HM: probably mtgox
 238 2013-02-01 02:16:02 <HM> gmaxwell: if you put a million transactions on to the network, then yes?
 239 2013-02-01 02:16:07 <jaakkos> gmaxwell: so yes
 240 2013-02-01 02:16:28 <gmaxwell> HM: in any case if you're asking about txn count by address, I imagine that deepbit is still #2.
 241 2013-02-01 02:16:44 <Scrat> and the other 2 letter website which shall remain unnamed
 242 2013-02-01 02:16:50 <gmaxwell> But mtgox may in fact be much more— but they don't usually generate totally obvious non-private txn.
 243 2013-02-01 02:17:33 <CodeShark> mtgox keeps much of their internal txs off the block chain
 244 2013-02-01 02:17:51 <HM> exchanges don't seem like a huge candidate to me. i think a lot of people use them to trade/speculate
 245 2013-02-01 02:17:59 <HM> scalp etc
 246 2013-02-01 02:18:35 <gmaxwell> HM: they also do fairly large volumes of I/O.
 247 2013-02-01 02:18:54 <gmaxwell> ys, most of their own volume is internal— but that volume is a significant multiple of the whole blockchain at times.
 248 2013-02-01 02:21:17 geb has quit (Ping timeout: 248 seconds)
 249 2013-02-01 02:22:22 <HM> the block hashing algorithm is surprisingly simple
 250 2013-02-01 02:22:23 petertod1 is now known as petertodd
 251 2013-02-01 02:22:35 stalled has joined
 252 2013-02-01 02:25:26 <petertodd> HM: Too simple... should have been merge mining from the start; currently because the coinbase is a normal tx, you always have relatively lengthy proof chains to the merkle root for any merge mined chains.
 253 2013-02-01 02:26:41 zooko` has quit (Ping timeout: 245 seconds)
 254 2013-02-01 02:26:52 <HM> i don;t yet understand the merkle root
 255 2013-02-01 02:26:53 geb has joined
 256 2013-02-01 02:27:02 topace has quit (Read error: Connection reset by peer)
 257 2013-02-01 02:27:24 <Luke-Jr> gmaxwell: lol
 258 2013-02-01 02:27:30 topace has joined
 259 2013-02-01 02:27:32 <HM> i follow the concept of a merkle tree but haven't put together how the merkle root hash works in the block header
 260 2013-02-01 02:27:44 dvide_ has joined
 261 2013-02-01 02:27:45 <HM> i haven't looked at any of the databade/blockchain stuff
 262 2013-02-01 02:27:54 freewil has quit (Read error: Connection reset by peer)
 263 2013-02-01 02:27:54 topace is now known as Guest82692
 264 2013-02-01 02:28:42 <CodeShark> HM, the idea is twofold: 1) you only need to hash the block header regardless of the number of transactions. 2) you can get a subset of transactions and verify they haven't been tampered with without needing all the transactions
 265 2013-02-01 02:29:06 freewil has joined
 266 2013-02-01 02:29:30 <petertodd> HM: 3) You can prove the existence of a transaction with a short merkle path to the block header.
 267 2013-02-01 02:29:47 quijibo has joined
 268 2013-02-01 02:29:48 <CodeShark> oh, yes - also (3) :)
 269 2013-02-01 02:30:46 <CodeShark> (1) means mining power required does not depend on the size of the block
 270 2013-02-01 02:30:58 denisx has quit (Remote host closed the connection)
 271 2013-02-01 02:31:09 <HM> i don't get it. isn't the merkle root hash a hash of all the tx hashes?
 272 2013-02-01 02:31:10 denisx has joined
 273 2013-02-01 02:31:12 dvide_ has quit (Client Quit)
 274 2013-02-01 02:31:21 <HM> how can you verify a subset and trust the entire block header?
 275 2013-02-01 02:31:22 denisx_ has joined
 276 2013-02-01 02:31:41 <CodeShark> it's a binary tree - each node is a hash of concatenating the hashes of the two children
 277 2013-02-01 02:31:56 Zarutian has joined
 278 2013-02-01 02:32:10 <sipa> HM: for each node, you either specify the two children, or its hash
 279 2013-02-01 02:32:17 <sipa> for the children you do the same thing
 280 2013-02-01 02:33:01 <sipa> so if you have a 10-level deep tree (1024 transactions), you can prove a transaction belongs to the block, by providing the transaction + 10 hashes along the part to the root
 281 2013-02-01 02:33:20 <petertodd> http://en.wikipedia.org/wiki/File:Hash_Tree.svg is a nice drawing
 282 2013-02-01 02:33:26 <HM> doesn't that mean a block has to contain a power of 2 tx's?
 283 2013-02-01 02:33:47 <sipa> HM: there are special rules when a level has an odd number of entries
 284 2013-02-01 02:33:54 owowo has quit (Quit: sayonara)
 285 2013-02-01 02:34:00 <sipa> (an unfortunately chosen rule, by the way...)
 286 2013-02-01 02:34:11 <HM> well of course
 287 2013-02-01 02:34:28 <sipa> in practice that changes little
 288 2013-02-01 02:35:19 denisx has quit (Ping timeout: 246 seconds)
 289 2013-02-01 02:35:19 denisx_ is now known as denisx
 290 2013-02-01 02:35:35 <HM> right so the subset you verify has to be one of the branches
 291 2013-02-01 02:35:55 <sipa> you can do it for arbitrary subsets, but it becomes a bit more complex
 292 2013-02-01 02:36:03 <HM> and because you have the root, you can traverse to the leaf of that branch and verify the leaves have the correct hashes
 293 2013-02-01 02:37:12 <fiesh> on my freebsd machine here, bitcoind has been running for several months now and consumes 416M RES memory (not that I'm super happy with that, but ok), on a debian box I started it last night and it is already at 776M... both 0.7.2.  Seems quite unreasonable to me, any recommendations?
 294 2013-02-01 02:38:16 <doublec> fiesh: number of connections?
 295 2013-02-01 02:38:52 <HM> i have a similar situation fiesh
 296 2013-02-01 02:40:07 <Scrat> maybe it has a different interpretation of res in ps?
 297 2013-02-01 02:40:08 <doublec> or 64bit vs 32bit binaries?
 298 2013-02-01 02:40:14 <Scrat> certainly is the case for free
 299 2013-02-01 02:40:32 <fiesh> 8 vs 22
 300 2013-02-01 02:40:53 <fiesh> 64bit, statically linked (which shouldn't account for much)
 301 2013-02-01 02:40:56 techlife has quit (Ping timeout: 255 seconds)
 302 2013-02-01 02:41:25 <HM> http://pastebin.com/WUnuhMVH
 303 2013-02-01 02:41:27 <HM> here's mine
 304 2013-02-01 02:41:37 dvide_ has joined
 305 2013-02-01 02:41:43 <HM> doing nothing but keeping the blockchain up to date
 306 2013-02-01 02:42:08 <fiesh> yyes
 307 2013-02-01 02:42:17 <HM> virtual memory is closer to 1.3 GiB for me
 308 2013-02-01 02:42:35 <HM> not sure how you account for swap but that's up to 600 MiB so that must be the difference
 309 2013-02-01 02:42:40 <HM> (box is doing nothing else)
 310 2013-02-01 02:45:10 randy-waterhouse has joined
 311 2013-02-01 02:47:12 <fiesh> I suppose bitcoind has been compiled against valgrind or something similar and verified that it doesn't leak?!
 312 2013-02-01 02:47:31 techlife has joined
 313 2013-02-01 02:47:48 <HM> i was thing boehm
 314 2013-02-01 02:47:51 <HM> thinking*
 315 2013-02-01 02:47:54 <CodeShark> valgrind actually gives me a lot of memory errors for bitcoind
 316 2013-02-01 02:48:39 <sipa> CodeShark: really?
 317 2013-02-01 02:48:45 <CodeShark> yer
 318 2013-02-01 02:48:53 <CodeShark> doesn't give you any errors?
 319 2013-02-01 02:48:58 <sipa> there are a few standard harmless ones caused by BDB
 320 2013-02-01 02:49:03 <CodeShark> ok
 321 2013-02-01 02:49:08 <sipa> and one about moneyformat at shutdown
 322 2013-02-01 02:49:08 <CodeShark> perhaps it's only that
 323 2013-02-01 02:49:40 <fiesh> so what is the "standard memory footprint" that people normally have?
 324 2013-02-01 02:49:53 <fiesh> can't find any information on the net
 325 2013-02-01 02:50:24 <sipa> having lots of connections increases memory usage a lot
 326 2013-02-01 02:50:42 <fiesh> why is that by the way?
 327 2013-02-01 02:50:58 <fiesh> but anyway, I only have very few connections
 328 2013-02-01 02:50:59 <sipa> big buffers (too big...)
 329 2013-02-01 02:51:12 <fiesh> hmm they must be way too big then I guess ;)
 330 2013-02-01 02:51:13 <HM> hmmm
 331 2013-02-01 02:51:23 <HM> boehm can be loaded as a LD_PRELOAD library
 332 2013-02-01 02:51:49 <HM> overrides malloc etc. does bitcoind do any fancy/obscure memory allocation?
 333 2013-02-01 02:52:15 <sipa> HM: i expect boehm to work, would be an interesting experiment
 334 2013-02-01 02:53:32 <gmaxwell> CodeShark: you need to recompile your openssl with a non-braindamaged one.
 335 2013-02-01 02:53:38 <fiesh> hmm bitcoind must be using the most ginormous buffers in the history of mankind...
 336 2013-02-01 02:53:56 <CodeShark> gmaxwell: ?
 337 2013-02-01 02:53:56 Guest82692 has quit (Read error: No route to host)
 338 2013-02-01 02:54:00 <sipa> fiesh: a few MiB per connection
 339 2013-02-01 02:54:05 <gmaxwell> Stock openssl copies uninitilized memory into its randomness pool and then makes valgrind complain about *.
 340 2013-02-01 02:54:15 Guest82692 has joined
 341 2013-02-01 02:54:29 <HM> sipa is hat a malloc()'s recv() buffer or a TCP/socket buffer?
 342 2013-02-01 02:54:33 <fiesh> sipa: then I don't understand I guess, what's being buffered that is so huge?
 343 2013-02-01 02:54:51 <CodeShark> how do I disable that when compiling openssl, gmaxwell?
 344 2013-02-01 02:54:55 <gmaxwell> It's a throughly dumb feature that seldom has any gain, and the harm of busting one of the most valuable pro-security debugging instrumentations avilable couldn't possible be worth it.
 345 2013-02-01 02:54:57 <sipa> fiesh: blocks
 346 2013-02-01 02:55:03 <gmaxwell> CodeShark: -DPURIFY IIRC.
 347 2013-02-01 02:55:27 <andytoshi> thx gmaxwell, i didn't know that either
 348 2013-02-01 02:55:38 <sipa> fiesh: to get some reasonable throughput, you want to be able to be sending/receiving them while processing
 349 2013-02-01 02:55:46 copumpkin has quit (Ping timeout: 252 seconds)
 350 2013-02-01 02:55:59 <gmaxwell> andytoshi: if you're using RPMs based on my specs.. I compile with -DPURIFY (or whatever the knob is)
 351 2013-02-01 02:56:02 <fiesh> sipa: but for sending you surely don't need to buffer each connection individually?!
 352 2013-02-01 02:56:13 <sipa> fiesh: yeah, there are a lot of improvements possible
 353 2013-02-01 02:56:24 <fiesh> sipa: and for receiving, once the blockchain is up to date, you'd only want to receive one at a time anyway, right?
 354 2013-02-01 02:56:27 <andytoshi> gmaxwell: nope, i compile my own
 355 2013-02-01 02:56:29 zooko` has joined
 356 2013-02-01 02:56:48 <sipa> fiesh: the send buffer is only 1 MiB these days, it seems
 357 2013-02-01 02:56:56 <sipa> it used to be 10 MiB, that was really overkill :)
 358 2013-02-01 02:57:04 <fiesh> sipa: if that's per connection, then I'd say it's still brutally huge ;)
 359 2013-02-01 02:57:14 <sipa> fiesh: meh
 360 2013-02-01 02:57:30 <HM> fortunately ram is cheap
 361 2013-02-01 02:57:52 <sipa> jeff wrote a nice improvement for the receiver part, but it occassionally caused segfaults, and i'm not sure anyone ever found them
 362 2013-02-01 02:58:29 <HM> you could share buffers across sockets, surely
 363 2013-02-01 02:59:27 <sipa> patches welcome :)
 364 2013-02-01 02:59:38 <sipa> (but not right now)
 365 2013-02-01 03:00:01 <HM> frozen for 0.8-rc1?
 366 2013-02-01 03:00:16 <sipa> we don't have an RC yet, but it won't be long
 367 2013-02-01 03:00:54 <fiesh> hmm, after a restart it's now at a res of 246M
 368 2013-02-01 03:01:33 <fiesh> so I think there was something seriously fishy going on, maybe not a memory leak but just keeping memory that is unnecessary
 369 2013-02-01 03:04:35 <HM> bitcointalk still down
 370 2013-02-01 03:04:37 <HM> sheesh
 371 2013-02-01 03:06:45 freewil has quit (Quit: Leaving)
 372 2013-02-01 03:07:00 RBecker is now known as rbecker
 373 2013-02-01 03:07:04 <petertodd> Too many people trying to talk about $20.50/BTC? It was running slow earlier...
 374 2013-02-01 03:08:02 <randy-waterhouse> asic threads just went nutz alos
 375 2013-02-01 03:09:32 <HM> i notice the bitcoin foundations box has 4 ASICs to Jeffs 3
 376 2013-02-01 03:09:51 <HM> pushing 89 GH/s
 377 2013-02-01 03:11:04 <jgarzik> woo!
 378 2013-02-01 03:11:09 geb has quit (Ping timeout: 260 seconds)
 379 2013-02-01 03:11:11 <jgarzik> slush's UI claims my ASIC found a block
 380 2013-02-01 03:11:28 <petertodd> Awesome!
 381 2013-02-01 03:11:37 <randy-waterhouse> cool first ASIC block .... well officially
 382 2013-02-01 03:12:13 <randy-waterhouse> jgarzik: what stratum  difficulty you using for the asic?
 383 2013-02-01 03:13:11 <andytoshi> congrats jgarzik, very cool!
 384 2013-02-01 03:14:34 geb has joined
 385 2013-02-01 03:14:34 geb has quit (Excess Flood)
 386 2013-02-01 03:18:34 geb has joined
 387 2013-02-01 03:18:34 geb has quit (Excess Flood)
 388 2013-02-01 03:18:54 geb has joined
 389 2013-02-01 03:18:54 geb has quit (Excess Flood)
 390 2013-02-01 03:18:58 <jgarzik> The first ASIC-found block: http://blockchain.info/block/00000000000001291f56c477342d39f0c9b31803e55c7b52ce2eab7c91eb2c0c?site=slush
 391 2013-02-01 03:19:34 geb has joined
 392 2013-02-01 03:19:53 <abracadabra> pffft
 393 2013-02-01 03:19:59 <abracadabra> only 0.3262 BTC in fees!
 394 2013-02-01 03:20:01 <abracadabra> :)
 395 2013-02-01 03:22:05 <petertodd> Reminds me, any idea what's going on with 1JmQN8NvX3XXWWrJW3rEEcKQMQd5DUgkH3? 9f2fd3fb459cfdb60f1b1d261d44cbb536e0c58a2ee3ec137e029bf962490910 has a 3BTC fee, and 7faff9ccc2962aa9ca3c291c7737410f432c596bf12a503dc19ae9fdbb9a8f7c 4.5BTC, maybe a bug in someone's custom code?
 396 2013-02-01 03:23:02 <petertodd> (and others at 3BTC, 1.2BTC etc.)
 397 2013-02-01 03:23:32 Tykling has quit (Ping timeout: 252 seconds)
 398 2013-02-01 03:23:41 <gavinandresen> sendrawtransaction should probably have an idiot switch turned on by default...
 399 2013-02-01 03:23:54 loltu has quit (Ping timeout: 252 seconds)
 400 2013-02-01 03:24:37 <gavinandresen> …although part of me thinks that losing a few BTC might teach people to be more careful when playing around dangerous equipment
 401 2013-02-01 03:24:57 <andytoshi> i think so, it'll be good to get some urban myths going about sendrawtransaction
 402 2013-02-01 03:25:05 <petertodd> The idiot switch for sendrawtransaction is to disable it... but yeah, we need better tools to review tx's.
 403 2013-02-01 03:25:16 Tykling has joined
 404 2013-02-01 03:25:48 <petertodd> You need to be able to present a list of pubkeys and do an independent check that the tx really can be spent by them for instance.
 405 2013-02-01 03:25:50 <gavinandresen> simple idiot switch would be a setting "don't sendrawtransaction if fee > 1% of input value"
 406 2013-02-01 03:26:35 <petertodd> That'd be a good start. Someone in gmaxwell's "taint" thread did that by accident - 1BTC fee.
 407 2013-02-01 03:26:41 <randy-waterhouse> a big red switch is the standard for idiot-proofing
 408 2013-02-01 03:26:47 <gavinandresen> well, if the pub keys don't pass the IsStandard() check then sendrawtransaction won't send them
 409 2013-02-01 03:27:19 <petertodd> Sure, but you need to be able to validate separately, perhaps even on a different machine for multisig.
 410 2013-02-01 03:27:48 <petertodd> Might not be worth it in RPC proper, but someone should write it somewhere.
 411 2013-02-01 03:28:09 <HM> the idiot switch would be to explicitly specify the fee in the API
 412 2013-02-01 03:28:09 Tykling has quit (Excess Flood)
 413 2013-02-01 03:28:24 <HM> then bark at them if the numbers don't add up
 414 2013-02-01 03:28:30 sgornick has quit (Ping timeout: 264 seconds)
 415 2013-02-01 03:28:32 Zarutian has quit (Quit: Zarutian)
 416 2013-02-01 03:28:44 Tykling has joined
 417 2013-02-01 03:29:45 PhantomSpark has quit (Ping timeout: 276 seconds)
 418 2013-02-01 03:30:42 Tykling has quit (Excess Flood)
 419 2013-02-01 03:31:10 <gavinandresen> BlueMatt: by the way… I failed to build a mingw-w64 openssl:  a_strex.c:574:1: internal compiler error: in inline_call, at ipa-inline-transform.c:269
 420 2013-02-01 03:32:08 <BlueMatt> gavinandresen: wonderful...
 421 2013-02-01 03:32:20 tomboy65 has joined
 422 2013-02-01 03:32:22 tomboy64 has quit (Remote host closed the connection)
 423 2013-02-01 03:32:30 tomboy65 is now known as tomboy64
 424 2013-02-01 03:32:58 <BlueMatt> gavinandresen: Im probably gonna have to make a second chroot on jenkins for precise and use the current deps builds from the old mingw
 425 2013-02-01 03:33:16 <gavinandresen> BlueMatt: ok.  Feel free to clone the whole VM if you need to.
 426 2013-02-01 03:33:44 Tykling has joined
 427 2013-02-01 03:33:46 <BlueMatt> gavinandresen: yea, Ill see what I can do, it may be until the weekend though, as I cant just install w64 in the current vm (would require a libc upgrade.....)
 428 2013-02-01 03:33:58 <gavinandresen> wonderful....
 429 2013-02-01 03:34:14 loltu has joined
 430 2013-02-01 03:34:28 asa1024 has joined
 431 2013-02-01 03:35:15 Hasimir- is now known as Hasimir
 432 2013-02-01 03:35:16 Hasimir has quit (Changing host)
 433 2013-02-01 03:35:16 Hasimir has joined
 434 2013-02-01 03:35:46 <BlueMatt> luckily its easily reproduceable, though bisect is giving me different commit causes depending on the order of the bisect...
 435 2013-02-01 03:36:54 freakazoid has quit (Ping timeout: 252 seconds)
 436 2013-02-01 03:38:32 sgornick has joined
 437 2013-02-01 03:42:56 testnode9 has quit (Ping timeout: 245 seconds)
 438 2013-02-01 03:43:08 Guest44214 is now known as novusordo
 439 2013-02-01 03:43:15 novusordo has quit (Changing host)
 440 2013-02-01 03:43:15 novusordo has joined
 441 2013-02-01 03:44:58 testnode9 has joined
 442 2013-02-01 03:47:23 Garr255 has quit (Read error: Connection reset by peer)
 443 2013-02-01 03:47:50 Garr255 has joined
 444 2013-02-01 03:47:55 Garr255 has quit (Changing host)
 445 2013-02-01 03:47:55 Garr255 has joined
 446 2013-02-01 03:50:37 b4epoche has quit (Read error: Operation timed out)
 447 2013-02-01 03:50:56 <jgarzik> hrm
 448 2013-02-01 03:51:10 <jgarzik> can anyone verify that bitcointalk.org did not change IP addresses in the last N hours?
 449 2013-02-01 03:51:45 fiesh has quit (Ping timeout: 256 seconds)
 450 2013-02-01 03:53:04 fiesh has joined
 451 2013-02-01 03:54:25 <petertodd> www.webboar.com says in mar 2012 it had 209.44.108.236, not useful I know
 452 2013-02-01 03:54:51 b4epoche has joined
 453 2013-02-01 03:55:26 Goonie has quit (Ping timeout: 248 seconds)
 454 2013-02-01 03:59:58 <petertodd> there you go: http://toolbar.netcraft.com/site_report?url=http://bitcointalk.org unchanged
 455 2013-02-01 04:00:15 <jgarzik> cool, that backs up what #bitcoin is saying
 456 2013-02-01 04:00:28 <petertodd> was someone suspicious?
 457 2013-02-01 04:00:38 <petertodd> ssl cert looks ok too
 458 2013-02-01 04:00:40 HM has quit (Ping timeout: 252 seconds)
 459 2013-02-01 04:01:00 <jgarzik> I was suspicious
 460 2013-02-01 04:01:11 <jgarzik> it suddenly asked for a password again, when it came back up
 461 2013-02-01 04:02:10 <petertodd> ah, I got that too
 462 2013-02-01 04:02:31 <petertodd> still seeing template parse errors
 463 2013-02-01 04:06:38 zooko` has quit (Ping timeout: 248 seconds)
 464 2013-02-01 04:06:59 HM has joined
 465 2013-02-01 04:11:55 <doublec> yeah I'm seeing it as well
 466 2013-02-01 04:14:33 Tykling has quit (Excess Flood)
 467 2013-02-01 04:16:23 <Luke-Jr> looking for more eyes on http://codepad.org/5rhxeTAh - this is to test master block acceptance vs 0.6.0
 468 2013-02-01 04:16:44 Tykling has joined
 469 2013-02-01 04:17:03 Tykling has quit (Excess Flood)
 470 2013-02-01 04:21:44 Tykling has joined
 471 2013-02-01 04:24:21 RainbowDashh has quit (Ping timeout: 276 seconds)
 472 2013-02-01 04:24:59 freakazoid has joined
 473 2013-02-01 04:27:04 [7] has quit (Disconnected by services)
 474 2013-02-01 04:27:11 TheSeven has joined
 475 2013-02-01 04:31:56 dvide_ has quit ()
 476 2013-02-01 04:33:34 twobitcoins_ has joined
 477 2013-02-01 04:33:41 mappum has quit (Remote host closed the connection)
 478 2013-02-01 04:34:20 mappum has joined
 479 2013-02-01 04:34:50 ShaTwo has joined
 480 2013-02-01 04:35:01 dust-otc has joined
 481 2013-02-01 04:35:22 <ShaTwo> Hello… I wanted to show you a weir think that just happened to me...
 482 2013-02-01 04:35:24 <ShaTwo> Check http://blockchain.info/address/1Q8nFaxzPUmgKzD2wGLTy8i3ogBNUgfHoo
 483 2013-02-01 04:36:28 <ShaTwo> If you check the 0.01710561 transaction… that was added by the client.
 484 2013-02-01 04:36:35 luke-jr_ has joined
 485 2013-02-01 04:36:50 nus has quit (Read error: Connection reset by peer)
 486 2013-02-01 04:36:50 nus has joined
 487 2013-02-01 04:36:50 Luke-Jr has quit (Excess Flood)
 488 2013-02-01 04:36:50 stalled has quit (Ping timeout: 276 seconds)
 489 2013-02-01 04:36:50 Belkaar has quit (Ping timeout: 276 seconds)
 490 2013-02-01 04:36:50 inlikeflynn has quit (Ping timeout: 276 seconds)
 491 2013-02-01 04:36:50 Arnavion has quit (Remote host closed the connection)
 492 2013-02-01 04:36:50 nus has quit (Changing host)
 493 2013-02-01 04:36:50 nus has joined
 494 2013-02-01 04:37:03 twobitcoins has quit (Ping timeout: 276 seconds)
 495 2013-02-01 04:37:10 Arnavion has joined
 496 2013-02-01 04:37:16 inlikeflynn has joined
 497 2013-02-01 04:37:16 Belkaar has joined
 498 2013-02-01 04:37:32 luke-jr_ is now known as Luke-Jr
 499 2013-02-01 04:38:35 inlikeflynn has quit (Changing host)
 500 2013-02-01 04:38:35 inlikeflynn has joined
 501 2013-02-01 04:38:56 Maged has quit (Read error: No route to host)
 502 2013-02-01 04:40:25 Maged has joined
 503 2013-02-01 04:40:36 stalled has joined
 504 2013-02-01 04:43:48 loltu has quit (Excess Flood)
 505 2013-02-01 04:46:14 loltu has joined
 506 2013-02-01 04:51:20 denisx has quit (Remote host closed the connection)
 507 2013-02-01 04:51:31 denisx has joined
 508 2013-02-01 04:55:16 sgornick has quit (Quit: Ex-Chat)
 509 2013-02-01 04:55:20 spenvo has quit (Quit: This computer has gone to sleep)
 510 2013-02-01 04:57:28 spenvo has joined
 511 2013-02-01 04:58:23 brwyatt is now known as Away!~brwyatt@brwyatt.net|brwyatt
 512 2013-02-01 05:04:30 RainbowDashh has joined
 513 2013-02-01 05:07:49 denisx has quit (Remote host closed the connection)
 514 2013-02-01 05:08:00 denisx has joined
 515 2013-02-01 05:12:20 JZavala has joined
 516 2013-02-01 05:24:50 HM has quit (Ping timeout: 252 seconds)
 517 2013-02-01 05:26:44 tcatm has quit (Quit: No Ping reply in 180 seconds.)
 518 2013-02-01 05:27:00 tcatm has joined
 519 2013-02-01 05:27:01 tcatm has quit (Changing host)
 520 2013-02-01 05:27:01 tcatm has joined
 521 2013-02-01 05:27:14 HM has joined
 522 2013-02-01 05:34:18 denisx has quit (Quit: denisx)
 523 2013-02-01 05:36:42 toffoo has joined
 524 2013-02-01 05:37:04 spenvo has quit (Quit: This computer has gone to sleep)
 525 2013-02-01 05:39:59 loltu has quit (Excess Flood)
 526 2013-02-01 05:40:29 tomboy64 has quit (Remote host closed the connection)
 527 2013-02-01 05:40:44 loltu has joined
 528 2013-02-01 05:41:04 edwincheese has joined
 529 2013-02-01 05:41:31 edwincheese has quit (Remote host closed the connection)
 530 2013-02-01 05:41:42 edwincheese has joined
 531 2013-02-01 05:43:07 tomboy64 has joined
 532 2013-02-01 05:43:32 Mad7Scientist has quit (Ping timeout: 255 seconds)
 533 2013-02-01 05:48:48 RainbowDashh is now known as Dii
 534 2013-02-01 05:50:00 Dii is now known as in
 535 2013-02-01 05:50:47 swappermall_ has quit (Ping timeout: 256 seconds)
 536 2013-02-01 05:52:16 techlife has quit (Ping timeout: 272 seconds)
 537 2013-02-01 05:53:40 BTCTrader has joined
 538 2013-02-01 05:55:53 BTCTrader has quit (Client Quit)
 539 2013-02-01 05:55:58 Mad7Scientist has joined
 540 2013-02-01 05:56:02 spenvo has joined
 541 2013-02-01 05:56:22 Mad7Scientist is now known as Guest50360
 542 2013-02-01 05:57:34 Hasimir has quit (Ping timeout: 248 seconds)
 543 2013-02-01 05:57:44 FredEE has quit (Quit: FredEE)
 544 2013-02-01 05:57:45 techlife has joined
 545 2013-02-01 05:59:50 Hasimir has joined
 546 2013-02-01 06:00:07 Hasimir is now known as Guest73685
 547 2013-02-01 06:00:32 Jamesonwa has joined
 548 2013-02-01 06:02:35 Guest73685 has quit (Changing host)
 549 2013-02-01 06:02:35 Guest73685 has joined
 550 2013-02-01 06:03:05 Guest73685 is now known as Hasimir
 551 2013-02-01 06:03:11 BTCTrader has joined
 552 2013-02-01 06:03:43 in is now known as RainbowDashh
 553 2013-02-01 06:12:15 KIDC has quit (Ping timeout: 276 seconds)
 554 2013-02-01 06:12:40 KIDC has joined
 555 2013-02-01 06:12:40 KIDC has quit (Changing host)
 556 2013-02-01 06:12:40 KIDC has joined
 557 2013-02-01 06:24:28 Hasimir has quit (Read error: Connection reset by peer)
 558 2013-02-01 06:25:12 Hasimir- has joined
 559 2013-02-01 06:28:31 Hasimir- is now known as Hasimir
 560 2013-02-01 06:28:32 Hasimir has quit (Changing host)
 561 2013-02-01 06:28:32 Hasimir has joined
 562 2013-02-01 06:28:46 osmosis has joined
 563 2013-02-01 06:28:59 Lyspooner has quit (Remote host closed the connection)
 564 2013-02-01 06:33:27 spenvo has quit (Ping timeout: 252 seconds)
 565 2013-02-01 06:33:46 snakie has joined
 566 2013-02-01 06:34:21 osmosis has quit (Remote host closed the connection)
 567 2013-02-01 06:37:10 edwincheese has quit (Quit: edwincheese)
 568 2013-02-01 06:38:27 yellowhat has joined
 569 2013-02-01 06:38:50 <andytoshi> [5~
 570 2013-02-01 06:41:16 MrTiggr has quit (Ping timeout: 245 seconds)
 571 2013-02-01 06:51:28 edwincheese has joined
 572 2013-02-01 07:04:06 RazielZ has joined
 573 2013-02-01 07:07:30 FredEE has joined
 574 2013-02-01 07:11:16 freakazoid has quit (Ping timeout: 245 seconds)
 575 2013-02-01 07:14:59 grau has joined
 576 2013-02-01 07:21:20 Detritus has quit (Ping timeout: 244 seconds)
 577 2013-02-01 07:24:07 Guest91877 has quit (Ping timeout: 240 seconds)
 578 2013-02-01 07:31:51 Detritus has joined
 579 2013-02-01 07:36:20 Guest91877 has joined
 580 2013-02-01 07:41:35 ovidiusoft has joined
 581 2013-02-01 07:47:17 MrSambal has quit ()
 582 2013-02-01 07:49:07 brwyatt is now known as brwyatt|Away
 583 2013-02-01 07:52:04 toobreev has joined
 584 2013-02-01 07:52:05 Goonie has joined
 585 2013-02-01 07:53:04 toobreevforjoe has joined
 586 2013-02-01 07:56:06 veerboot has quit (Ping timeout: 264 seconds)
 587 2013-02-01 07:57:13 toobreev has quit (Ping timeout: 260 seconds)
 588 2013-02-01 08:01:41 Luke-Jr has quit (Ping timeout: 245 seconds)
 589 2013-02-01 08:02:35 FredEE has quit (Read error: Connection reset by peer)
 590 2013-02-01 08:06:53 sgstair has quit (Read error: Connection reset by peer)
 591 2013-02-01 08:07:19 b4epoche has quit (Ping timeout: 244 seconds)
 592 2013-02-01 08:07:35 sgstair has joined
 593 2013-02-01 08:09:41 b4epoche has joined
 594 2013-02-01 08:10:51 edwincheese has quit (Quit: edwincheese)
 595 2013-02-01 08:12:39 Luke-Jr has joined
 596 2013-02-01 08:13:55 grau has quit (Remote host closed the connection)
 597 2013-02-01 08:15:28 edwincheese_ has joined
 598 2013-02-01 08:16:24 rdymac has joined
 599 2013-02-01 08:18:04 nanotube has quit (Remote host closed the connection)
 600 2013-02-01 08:18:04 gribble has quit (Remote host closed the connection)
 601 2013-02-01 08:20:28 rdymac has quit (Read error: Connection reset by peer)
 602 2013-02-01 08:23:01 luke-jr_ has joined
 603 2013-02-01 08:23:15 gribble has joined
 604 2013-02-01 08:23:21 mappum has quit (Ping timeout: 260 seconds)
 605 2013-02-01 08:23:31 Luke-Jr has quit (Read error: Connection reset by peer)
 606 2013-02-01 08:24:36 nanotube has joined
 607 2013-02-01 08:32:08 altamic has joined
 608 2013-02-01 08:32:08 altamic has quit (Changing host)
 609 2013-02-01 08:32:08 altamic has joined
 610 2013-02-01 08:33:32 CodesInChaos has joined
 611 2013-02-01 08:38:29 edwincheese_ has quit (Quit: edwincheese_)
 612 2013-02-01 08:41:26 rdponticelli has quit (Quit: No Ping reply in 180 seconds.)
 613 2013-02-01 08:43:02 rdponticelli has joined
 614 2013-02-01 08:51:03 RainbowDashh has quit (Quit: Computer has gone to sleep.)
 615 2013-02-01 08:51:10 rdymac has joined
 616 2013-02-01 08:55:29 HM has quit (Ping timeout: 252 seconds)
 617 2013-02-01 08:57:00 HM has joined
 618 2013-02-01 08:59:38 randy-waterhouse has quit (Quit: Out.)
 619 2013-02-01 09:04:04 rdymac has quit (Remote host closed the connection)
 620 2013-02-01 09:06:43 Goonie has left ()
 621 2013-02-01 09:07:27 ShaTwo has quit (Ping timeout: 252 seconds)
 622 2013-02-01 09:11:41 luke-jr_ has quit (Ping timeout: 245 seconds)
 623 2013-02-01 09:12:27 luke-jr_ has joined
 624 2013-02-01 09:15:01 <thermoman> are there any known issues with big wallet.dat files? with big i mean 500 megabytes and more
 625 2013-02-01 09:15:44 <thermoman> besides the fact that starting with such a wallet.dat file the daemon needs a good 10 minutes to come up fully
 626 2013-02-01 09:17:59 <SomeoneWeird> 0_____o
 627 2013-02-01 09:18:06 <SomeoneWeird> how many damn keys do you have
 628 2013-02-01 09:18:06 HM has quit (Read error: Connection reset by peer)
 629 2013-02-01 09:18:26 HM has joined
 630 2013-02-01 09:20:31 luke-jr_ has quit (Excess Flood)
 631 2013-02-01 09:20:54 luke-jr_ has joined
 632 2013-02-01 09:21:42 <thermoman> SomeoneWeird: some :)
 633 2013-02-01 09:22:56 yareyare has joined
 634 2013-02-01 09:24:56 tonikt has joined
 635 2013-02-01 09:29:02 Hasimir- has joined
 636 2013-02-01 09:29:24 alex__ has joined
 637 2013-02-01 09:29:34 Hasimir has quit (Ping timeout: 276 seconds)
 638 2013-02-01 09:30:01 <alex__> yo
 639 2013-02-01 09:30:15 <alex__> any good btc sites that take Gdmp?
 640 2013-02-01 09:30:24 Hasimir- is now known as Hasimir
 641 2013-02-01 09:30:26 Hasimir has quit (Changing host)
 642 2013-02-01 09:30:27 Hasimir has joined
 643 2013-02-01 09:33:27 alex__ has quit (Remote host closed the connection)
 644 2013-02-01 09:33:59 HM has quit (Ping timeout: 252 seconds)
 645 2013-02-01 09:34:34 <CodeShark> if you're using 500MB wallets, chances are you've got fairly poor wallet management policies :)
 646 2013-02-01 09:36:00 <CodeShark> the satoshi client doesn't make good wallet management simple to begin with, though
 647 2013-02-01 09:36:18 <SomeoneWeird> ^^
 648 2013-02-01 09:37:07 HM has joined
 649 2013-02-01 09:37:20 toffoo has quit ()
 650 2013-02-01 09:37:49 tomboy64 has quit (Quit: Uhh ... gotta go.)
 651 2013-02-01 09:42:43 Insu has joined
 652 2013-02-01 09:44:59 MrTiggr has joined
 653 2013-02-01 09:47:49 <MC1984> bitcoins behavior where it empties an input totally and sends the change back to another address you own
 654 2013-02-01 09:47:57 <MC1984> does it have to do that?
 655 2013-02-01 09:48:08 <MC1984> do inputs have to be emptied
 656 2013-02-01 09:48:21 <Diablo-D3> yes, otherwise the coins are lost
 657 2013-02-01 09:49:10 <CodeShark> an output must either be spent completely or not at all
 658 2013-02-01 09:49:18 <CodeShark> there's no such thing as spending only part of an output in bitcoin
 659 2013-02-01 09:49:33 <CodeShark> so if you don't want to spend it all, you have to send change to another output
 660 2013-02-01 09:49:54 <MC1984> can you send change back to the same address
 661 2013-02-01 09:50:01 <CodeShark> yes
 662 2013-02-01 09:50:05 <CodeShark> although the satoshi client doesn't enable that
 663 2013-02-01 09:50:12 <MC1984> can an output be an input too
 664 2013-02-01 09:50:18 <CodeShark> no
 665 2013-02-01 09:50:34 <CodeShark> inputs and outputs are two different structures and are kept separate in the block chain
 666 2013-02-01 09:51:56 <MC1984> how can i see for EG the wikileaks public address getting some coins taken off it periodicly then
 667 2013-02-01 09:52:18 <CodeShark> not sure what you mean
 668 2013-02-01 09:53:47 <MC1984> http://blockchain.info/address/1HB5XMLmzFVj8ALj6mfBsbifRoD4miY36v
 669 2013-02-01 09:54:01 <MC1984> they cash out small amounts but it stil keeps a significant balance
 670 2013-02-01 09:54:18 <MC1984> it doesnt empty the address
 671 2013-02-01 09:54:29 <CodeShark> balances don't really exist
 672 2013-02-01 09:54:34 PK has joined
 673 2013-02-01 09:54:34 <CodeShark> that's a high-level abstraction
 674 2013-02-01 09:54:44 <CodeShark> at the block chain level, all there are are unspent output
 675 2013-02-01 09:54:46 <CodeShark> at the block chain level, all there are are unspent outputs
 676 2013-02-01 09:54:54 <CodeShark> multiple outputs can be claimable by the same key
 677 2013-02-01 09:55:21 <CodeShark> blockchain.info calculates balance by adding up all the unspent outputs claimable by the key corresponding to that address
 678 2013-02-01 09:55:40 <MC1984> i didnt know that
 679 2013-02-01 09:56:12 <MC1984> i have been conflating address with output all this time then
 680 2013-02-01 09:56:19 <CodeShark> it's a common mistake
 681 2013-02-01 09:56:34 <CodeShark> people are used to thinking in terms of bank accounts and mailboxes
 682 2013-02-01 09:56:42 <CodeShark> but bitcoin doesn't really work like that
 683 2013-02-01 09:58:22 <MC1984> so you can spend an output and send the change back to the same address
 684 2013-02-01 09:58:32 <CodeShark> a more appropriate analogy would be that each output is an entirely separate mailbox...but you can still open up multiple mailboxes with a single key
 685 2013-02-01 09:58:49 <CodeShark> yes you can
 686 2013-02-01 09:58:54 <MC1984> so on a high level it looks like youve just spend a bit of the address
 687 2013-02-01 09:59:02 <CodeShark> yes
 688 2013-02-01 09:59:06 <MC1984> ok
 689 2013-02-01 09:59:46 <MC1984> so bitcoins change transactions are a necessity, not just a rather pointless anonymity kludge
 690 2013-02-01 09:59:46 <CodeShark> and once a mailbox is opened, it can no longer be used
 691 2013-02-01 10:00:03 <CodeShark> yes
 692 2013-02-01 10:00:28 <CodeShark> and if you don't care about anonymity, it's totally fine to send change to the same address
 693 2013-02-01 10:01:06 <CodeShark> perhaps that should be an option in the satoshi client - reduces wallet file kludge :)
 694 2013-02-01 10:01:14 <MC1984> i was gonna suggest they be nixed t save chain space since bitcoin turned out to be far less anon thn anyone realised anyway lol
 695 2013-02-01 10:01:31 <MC1984> but oh well
 696 2013-02-01 10:01:36 <CodeShark> it doesn't save any block chain space (unless you compress the block chain)
 697 2013-02-01 10:01:46 <CodeShark> but it does reduce wallet size
 698 2013-02-01 10:02:08 <CodeShark> unless you're using a deterministic wallet
 699 2013-02-01 10:02:33 Jamesonwa has quit (Remote host closed the connection)
 700 2013-02-01 10:02:38 <MC1984> i dont see how wallet size could be a roblem for anyone
 701 2013-02-01 10:02:52 <MC1984> merchats maybe?
 702 2013-02-01 10:02:52 optimator_ has joined
 703 2013-02-01 10:03:09 <CodeShark> deterministic wallets are probably the way to go - and yes, for merchants, wallet management would be horrendous if they have to be dealing with thousands of daily transactions
 704 2013-02-01 10:03:35 <CodeShark> so bitcoind's wallet is not recommended for high-volume usage :)
 705 2013-02-01 10:03:49 <CodeShark> it's ok for personal usage, as long as you're not making too many transactions
 706 2013-02-01 10:03:51 <MC1984> no
 707 2013-02-01 10:04:07 optimator has quit (Ping timeout: 256 seconds)
 708 2013-02-01 10:04:19 <MC1984> merhcats use armoury or bitpay or something
 709 2013-02-01 10:05:16 <MC1984> so multiple unspent outputs can constitute the "balance" of a bitcoin address, and multiple addresses can constitute a wallet right?
 710 2013-02-01 10:05:48 <CodeShark> you can think of it like that, yes
 711 2013-02-01 10:06:19 <CodeShark> the wallet file in the satoshi client also stores the transactions to and from the wallet
 712 2013-02-01 10:06:33 <CodeShark> but those could be reconstructed from the block chain and mempool
 713 2013-02-01 10:07:07 <MC1984> goddam bitcoin y u so complicated
 714 2013-02-01 10:09:01 <MC1984> are there any privacy implications by associating outputs backwrds and forwards in the chain, like there are by associating addresses
 715 2013-02-01 10:09:29 <CodeShark> not sure what you mean
 716 2013-02-01 10:11:10 <MC1984> the way coins 'move' through the chain via input>output>input etc
 717 2013-02-01 10:11:39 <MC1984> going a layer below addresses, could coins themselves be associated to deanonymise a user
 718 2013-02-01 10:11:52 <CodeShark> that's all traceable 100%. if you want real privacy the only real viable option is to mix coins and use joint transactions
 719 2013-02-01 10:13:15 <CodeShark> if you have 10 inputs and 10 outputs and the 10 inputs belong to 10 different people and the 10 outputs are each in fresh wallets and their order is randomized, it becomes impossible to know which output belongs to whom
 720 2013-02-01 10:13:16 <doublec> MC1984: wallet size does get to be an issue. On one of my sites the wallet is >100MB.
 721 2013-02-01 10:15:04 <CodeShark> the bitcoind wallet is really crappy for merchants
 722 2013-02-01 10:15:34 <CodeShark> there are far better approaches to wallets
 723 2013-02-01 10:15:46 <CodeShark> unfortunately, no complete solution I'm fully satisfied with yet
 724 2013-02-01 10:16:44 <MC1984> hmm blockchain.info doesnt seem to be able to display inputs and outputs level, only addresses
 725 2013-02-01 10:17:30 <CodeShark> it does display inputs and outputs
 726 2013-02-01 10:17:40 <CodeShark> the inputs are on the left of the transaction, the outputs on the right - and you can also view their scripts
 727 2013-02-01 10:18:34 ken` has joined
 728 2013-02-01 10:19:05 <MC1984> oh youre right
 729 2013-02-01 10:21:12 B0g4r7 has quit (Ping timeout: 276 seconds)
 730 2013-02-01 10:26:50 winterblack has joined
 731 2013-02-01 10:27:00 rdymac has joined
 732 2013-02-01 10:28:04 B0g4r7 has joined
 733 2013-02-01 10:30:42 yellowhat has quit (Remote host closed the connection)
 734 2013-02-01 10:30:46 skeledrew has quit (Remote host closed the connection)
 735 2013-02-01 10:31:04 skeledrew has joined
 736 2013-02-01 10:31:23 gribble has quit (Excess Flood)
 737 2013-02-01 10:39:23 ahf has quit (Quit: Reconnecting)
 738 2013-02-01 10:39:23 ahf has joined
 739 2013-02-01 10:39:24 gribble has joined
 740 2013-02-01 10:41:54 <MC1984> holy shit there are 'high priority' tx from jan 11 still waiting for a block?
 741 2013-02-01 10:42:25 <CodeShark> give an example
 742 2013-02-01 10:43:39 <MC1984> http://bitcoincharts.com/bitcoin/txlist/
 743 2013-02-01 10:43:47 <MC1984> thats a large page btw
 744 2013-02-01 10:43:47 <CodeShark> chances are they depend on a transaction that either violates fee rules for block inclusion...or a transaction that has been doublespent
 745 2013-02-01 10:43:59 <MC1984> theres loads of them
 746 2013-02-01 10:44:34 Lolcust has quit (Quit: Nap time)
 747 2013-02-01 10:44:48 Lolcust has joined
 748 2013-02-01 10:46:12 <CodeShark> doublespends
 749 2013-02-01 10:46:44 <MC1984> they havent got unconfirmed inputs
 750 2013-02-01 10:47:10 <CodeShark> I checked 5028fdd720acffffc31118f81372535e97954b604b9981e17f6b17fa8087939f
 751 2013-02-01 10:47:13 <CodeShark> it was doublespent
 752 2013-02-01 10:47:38 <MC1984> how do you check
 753 2013-02-01 10:47:46 <CodeShark> I have a database
 754 2013-02-01 10:48:09 <MC1984> oh
 755 2013-02-01 10:48:20 <MC1984> how much of that list is real txs waiting then
 756 2013-02-01 10:48:39 rdymac has quit (Remote host closed the connection)
 757 2013-02-01 10:49:29 <MC1984> and how long does it take for a dead tx to clear out of the mempool
 758 2013-02-01 10:49:30 one_zero has quit ()
 759 2013-02-01 10:51:40 Insu has quit (Remote host closed the connection)
 760 2013-02-01 10:54:09 <CodeShark> if it's been doublespent, it should get kicked out of the mempool as soon as a transaction conflicting with it is seen...not sure if that's been added to bitcoind
 761 2013-02-01 10:54:30 <CodeShark> I mean, if a block is connected that has a transaction conflicting with it
 762 2013-02-01 10:54:47 <CodeShark> but blockchain.info is not a validation node - it's a database
 763 2013-02-01 10:55:00 <CodeShark> so it keeps even transactions that all validating nodes have rejected
 764 2013-02-01 10:55:59 Insu has joined
 765 2013-02-01 10:59:44 drizztbsd has joined
 766 2013-02-01 11:04:02 <sipa> CodeShark: since the 0.8 code
 767 2013-02-01 11:04:27 <CodeShark> so the current HEAD has that behavior, yes?
 768 2013-02-01 11:04:35 <CodeShark> master HEAD
 769 2013-02-01 11:04:58 word has quit (Read error: Connection reset by peer)
 770 2013-02-01 11:05:20 word has joined
 771 2013-02-01 11:06:40 bitafterbit has joined
 772 2013-02-01 11:06:43 <sipa> indeed
 773 2013-02-01 11:07:18 <sipa> it kicks out txs out of the mempool that depend on inputs that are spent by a confirmed transaction
 774 2013-02-01 11:08:13 <CodeShark> we could also get it to alert the wallets...although I'd like to move more in the direction of separation between validation and wallets
 775 2013-02-01 11:08:26 WolfAlex has joined
 776 2013-02-01 11:09:12 Garr255 has quit (Read error: Connection reset by peer)
 777 2013-02-01 11:09:14 <CodeShark> but that would be a way to support the negative confirmation thing
 778 2013-02-01 11:09:39 Garr255 has joined
 779 2013-02-01 11:09:39 Garr255 has quit (Changing host)
 780 2013-02-01 11:09:39 Garr255 has joined
 781 2013-02-01 11:09:47 <CodeShark> although it would only work if the conflict occurs while the node is running
 782 2013-02-01 11:10:16 <CodeShark> restart the node - and the entire mempool is gone
 783 2013-02-01 11:10:45 valparaiso has quit (Changing host)
 784 2013-02-01 11:10:45 valparaiso has joined
 785 2013-02-01 11:12:09 <sipa> and also only works for full nodes
 786 2013-02-01 11:12:28 tonikt has quit (Read error: Connection reset by peer)
 787 2013-02-01 11:16:42 jenic has quit (Changing host)
 788 2013-02-01 11:16:42 jenic has joined
 789 2013-02-01 11:20:23 <CodeShark> who wrote multibit? and is there some documentation available that's more technical rather than the for-bitcoin-newbie stuff at the multibit site?
 790 2013-02-01 11:22:53 HM has quit (Ping timeout: 252 seconds)
 791 2013-02-01 11:25:09 <sipa> jim burton
 792 2013-02-01 11:27:03 HM has joined
 793 2013-02-01 11:28:49 yareyare has joined
 794 2013-02-01 11:30:29 talso has quit (Ping timeout: 255 seconds)
 795 2013-02-01 11:31:47 <CodeShark> I guess the source code isn't totally illegible...
 796 2013-02-01 11:37:45 talso has joined
 797 2013-02-01 11:42:41 Zarutian has joined
 798 2013-02-01 11:44:01 BTCOxygen has joined
 799 2013-02-01 11:47:32 <CodeShark> ECKey newKey = new ECKey(); perWalletModelData.getWallet().keychain.add(newKey);
 800 2013-02-01 11:47:40 <CodeShark> hmmm - I guess no support for deterministic wallets...
 801 2013-02-01 11:47:47 <sipa> no, indeed
 802 2013-02-01 11:48:03 <sipa> though i believe he said planning to support bip32
 803 2013-02-01 11:49:43 rbecker is now known as RBecker
 804 2013-02-01 11:56:06 <CodeShark> so armory is the only major wallet project right now that supports deterministic wallets?
 805 2013-02-01 11:58:13 <CodeShark> there's also electrum...but that relies on a custom server
 806 2013-02-01 12:01:07 <SomeoneWeird> and armory relies on the satoshi client
 807 2013-02-01 12:01:48 <CodeShark> right - that in and of itself wouldn't bother me too much if it weren't for the fact that it needs to scan the block chain files directly
 808 2013-02-01 12:02:16 <SomeoneWeird> yeah, i'm kinda pissed at how big the blockchain is tbh
 809 2013-02-01 12:02:54 <SomeoneWeird> i'm on a capped net and it's already nearly 10% of my monthly cap
 810 2013-02-01 12:02:56 <CodeShark> space isn't really a major issue for me atm
 811 2013-02-01 12:03:05 <CodeShark> I can easily get servers with a few hundred gigs if necessary
 812 2013-02-01 12:03:29 <CodeShark> but I don't want the wallet client relying on the low-level data format of the satoshi client persistent data
 813 2013-02-01 12:03:38 <SomeoneWeird> oh yeah, space isn't a problem
 814 2013-02-01 12:03:41 <SomeoneWeird> it's the bandwidth needed
 815 2013-02-01 12:04:11 <CodeShark> in fact, I'd be totally fine with block chain storage redundancy if it means the two processes needn't be sharing the same data files
 816 2013-02-01 12:04:55 Prattler has quit (Quit: ZNC - http://znc.in)
 817 2013-02-01 12:04:56 <SomeoneWeird> yeah
 818 2013-02-01 12:05:37 RBecker is now known as rbecker
 819 2013-02-01 12:06:32 <CodeShark> as for bandwidth, I can transfer the entire block chain in a matter of minutes from one server to another. It's building all the tables and indices with slow storage engines that's a serious problem
 820 2013-02-01 12:07:39 <CodeShark> but I suppose a wallet app doesn't need the level of indexing that a historical database app does
 821 2013-02-01 12:07:58 <SomeoneWeird> yeah
 822 2013-02-01 12:09:09 sssssssssss has joined
 823 2013-02-01 12:10:24 <sssssssssss> hi, how do I enable gribble to a new channel? or I need to install it to make it work in a new channel?
 824 2013-02-01 12:11:54 <sipa> poke its owner
 825 2013-02-01 12:12:00 <sipa> aka nanotube
 826 2013-02-01 12:12:04 <sipa> (iirc)
 827 2013-02-01 12:12:47 <sssssssssss> ok, thanks
 828 2013-02-01 12:18:28 Prattler has joined
 829 2013-02-01 12:21:48 b4epoche has quit (Ping timeout: 255 seconds)
 830 2013-02-01 12:22:31 <CodeShark> for a backend wallet solution, the set of requirements are drastically distinct from, say, a mobile wallet
 831 2013-02-01 12:22:58 tradefortress has quit (Ping timeout: 245 seconds)
 832 2013-02-01 12:23:21 <CodeShark> I don't really care about initial sync time, storage space, and bandwidth. At least not nearly as much as I care about security and concurrency
 833 2013-02-01 12:23:46 voodster has joined
 834 2013-02-01 12:24:04 <CodeShark> and availability
 835 2013-02-01 12:24:14 <CodeShark> and ease of maintenance
 836 2013-02-01 12:24:41 <CodeShark> if that means having to store 100GB of data and waiting three days before I can connect a website to it, fine
 837 2013-02-01 12:25:00 b4epoche has joined
 838 2013-02-01 12:25:51 <CodeShark> I want a nice API that supports high concurrency with strong access controls
 839 2013-02-01 12:26:00 rdponticelli has quit (Ping timeout: 276 seconds)
 840 2013-02-01 12:26:22 <CodeShark> so that a bunch of shopping cart instances can be accessing it
 841 2013-02-01 12:26:51 <CodeShark> and I want to be able to completely separate the public key API from the private key signing operations
 842 2013-02-01 12:27:06 <CodeShark> have them run on separate systems entirely
 843 2013-02-01 12:27:10 davout has joined
 844 2013-02-01 12:27:11 davout has quit (Changing host)
 845 2013-02-01 12:27:11 davout has joined
 846 2013-02-01 12:28:24 <CodeShark> the public key API just needs to provide receiving addresses and a simple alerting mechanism for transactions and confirmations
 847 2013-02-01 12:30:04 <CodeShark> I've had to build most of this stuff for projects...but much of it was developed in an ad-hoc manner and would require some serious repackaging to make it easily deployable
 848 2013-02-01 12:31:08 <CodeShark> I would prefer to start a new open source project - or adapt an existing one - to support these kinds of features. with the idea of eventually packaging it nicely as a ready-made solution for web developers
 849 2013-02-01 12:31:57 <CodeShark> most of the wallet focus for bitcoind and the discussions I've seen here have centered around endusers - not really merchants
 850 2013-02-01 12:34:02 <CodeShark> we also need SPV clients and all that...but I don't think it really makes sense to try to support all these things in a single project
 851 2013-02-01 12:35:50 rdymac has joined
 852 2013-02-01 12:45:53 <CodeShark> but it might be nice to have some degree of coordination between multiple projects
 853 2013-02-01 12:47:34 <bsdunx> by way of libBitCoin!?
 854 2013-02-01 12:48:15 <CodeShark> I'm fully open to library standardization if it expedites progress
 855 2013-02-01 12:49:01 <CodeShark> but the type of coordination I speak of goes beyond just source libraries
 856 2013-02-01 12:49:44 <HM> implementation plurality is an important thing
 857 2013-02-01 12:49:56 <CodeShark> as in competition?
 858 2013-02-01 12:49:59 <HM> yes
 859 2013-02-01 12:50:01 <CodeShark> sure
 860 2013-02-01 12:50:29 <HM> the problem is you can't have people changing the protocol willy nilly, the most popular client ultimately decides the fate of the network
 861 2013-02-01 12:50:51 <HM> look at bittorrent...uTorrent introduced uTP, there has been transport encryption
 862 2013-02-01 12:51:06 <CodeShark> right - it might be a good idea to separate the standards body from the satoshi dev core
 863 2013-02-01 12:51:13 <CodeShark> :)
 864 2013-02-01 12:51:32 <CodeShark> perhaps we should have a standards body that only deals with protocol stuff and not implementations
 865 2013-02-01 12:51:50 <bsdunx> do we have an RFC yet? Has that even seen a thought?
 866 2013-02-01 12:51:56 <sipa> bsdunx: we have BIPs
 867 2013-02-01 12:52:32 <HM> bittorrent doesn't have an RFC either, the client devs are either submissive, and implement what their users want, or coordinate with one another
 868 2013-02-01 12:52:34 <bsdunx> Can you link the current authoritative spec?
 869 2013-02-01 12:52:36 <sipa> or do you mean a traditional RFC describing bitcoin as a whole
 870 2013-02-01 12:52:58 <bsdunx> protocol specs
 871 2013-02-01 12:53:07 drizzt_ has joined
 872 2013-02-01 12:53:13 <sipa> bsdunx: the authorative spec is still the reference client code, but changes to the protocol are always written as BIPs these days
 873 2013-02-01 12:53:29 <HM> yeah the BIP system looks very nice
 874 2013-02-01 12:53:30 drizztbsd has quit (Ping timeout: 252 seconds)
 875 2013-02-01 12:53:44 <HM> the wiki is quite good bsdunx
 876 2013-02-01 12:54:13 <sipa> the problem with a distributed consensus protocol is that a format spec doesn't really matter - if every client implements things in the same way, but different from the spec, it is the clients that are correct and the spec that is wrong
 877 2013-02-01 12:54:36 <sipa> at least when dealing with network rule
 878 2013-02-01 12:54:49 <CodeShark> having the spec defined by a reference implementation is a problem, though
 879 2013-02-01 12:54:53 <sipa> agree
 880 2013-02-01 12:55:14 <sipa> it would be nice to have it encoded as a standard document
 881 2013-02-01 12:55:44 <HM> an extensible wire format might be the way to go one day
 882 2013-02-01 12:55:49 <sipa> but my point is that even in that case, the correct rules will be the implemented ones
 883 2013-02-01 12:56:07 <sipa> the P2P protocol is easy to change, that's not the problem
 884 2013-02-01 12:56:23 <CodeShark> not without a hard fork
 885 2013-02-01 12:56:37 <sipa> P2P protocol has nothing to do with network rules
 886 2013-02-01 12:56:51 <CodeShark> but agreed, that's not really the issue
 887 2013-02-01 12:56:59 <sipa> you can invent a new protocol to relay blocks and transactions, and as long as you have gateways, there is not a single problem
 888 2013-02-01 12:57:08 <HM> true
 889 2013-02-01 12:57:32 <sipa> but the validity rules for blocks and transactions themself are always defined by the implementation(s), and a specification cannot be authorative
 890 2013-02-01 12:57:52 <CodeShark> unfortunately that's the case now - but does it have to be such?
 891 2013-02-01 12:58:00 <sipa> yes, it has to be such
 892 2013-02-01 12:58:04 <CodeShark> why?
 893 2013-02-01 12:58:21 <sipa> because the point of a consensus protocol is that everyone implements the exact same rules
 894 2013-02-01 12:58:34 <HM> true sipa, but you can have a packet format that allows new things to be added in a way that doesn't involve going out of channel. look at IP and TCP "options" for example
 895 2013-02-01 12:58:48 <HM> but then maybe lots of smaller channels is better
 896 2013-02-01 12:58:50 <sipa> HM: i am not talking about the P2P protocol
 897 2013-02-01 12:58:57 <sipa> i'm talking about the block/tx validity rules
 898 2013-02-01 12:59:01 <HM> ah
 899 2013-02-01 12:59:11 <sipa> for the P2P protocol, all traditional mechanisms are fine
 900 2013-02-01 12:59:30 <sipa> you can have extensions, versions, proprietary extensions even, gateways, ...
 901 2013-02-01 12:59:30 <CodeShark> block/tx validity rules depend on the p2p protocol only inasmuch as a specific data encoding is required for hashing
 902 2013-02-01 13:00:09 <HM> well if one hash becomes suspect you can always do newhash(oldhash(x))
 903 2013-02-01 13:00:13 <sipa> but the rules that govern which block and txn are valid, are always defined by the network, and not by any text document that tries to describe how it *should* work
 904 2013-02-01 13:00:28 <sipa> HM: which means you get the collision problems from both
 905 2013-02-01 13:01:04 <sipa> and consensus is far more important than "correctness" according to some external standard
 906 2013-02-01 13:01:25 <CodeShark> but sipa, shouldn
 907 2013-02-01 13:01:34 <CodeShark> shouldn't it be possible to describe those rules abstractly?
 908 2013-02-01 13:01:43 <CodeShark> without having to give a specific implementation?
 909 2013-02-01 13:01:49 <CodeShark> can't it just be done through pseudocode?
 910 2013-02-01 13:01:54 <sipa> *describe* yes, but it cannot be authorative
 911 2013-02-01 13:02:07 <CodeShark> why not?
 912 2013-02-01 13:02:10 <sipa> if there is a mismatch between the standard and the implementation, the standard is wrong
 913 2013-02-01 13:02:36 <sipa> don't get me wrong
 914 2013-02-01 13:02:52 <sipa> i'm very much in favor of describing the network rules in a formal way
 915 2013-02-01 13:03:37 <sipa> and hopefully that can be done correctly enough, so that someone only needs the document and not any (other) implementation to base upon
 916 2013-02-01 13:03:38 <CodeShark> I understand there are some major issues right now that present hurdles - for instance, noncannonical OpenSSL encodings
 917 2013-02-01 13:03:43 t7 has joined
 918 2013-02-01 13:03:55 <sipa> that's an excellent example
 919 2013-02-01 13:04:12 <sipa> imagine 2y ago we would have written a standard for the network rules
 920 2013-02-01 13:04:22 <sipa> completely accurate up to everyone's knowledge
 921 2013-02-01 13:04:38 <sipa> and someone would make a new implementation, exactly matching that document
 922 2013-02-01 13:05:01 <CodeShark> yes, and it would break
 923 2013-02-01 13:05:22 <sipa> ... and as soon as someone puts a compressed pubkey in the wild, you get a hard fork in the network
 924 2013-02-01 13:05:22 <CodeShark> but it could be argued that that was an oversight in the original implementation :)
 925 2013-02-01 13:05:23 JDuke128 has joined
 926 2013-02-01 13:05:37 <sipa> you may argue that, but it doesn't change a thing
 927 2013-02-01 13:05:39 <CodeShark> however, it doesn't really matter much now does it
 928 2013-02-01 13:06:09 <sipa> the truth is that consensus is more important than correctness - whatever that last thing means
 929 2013-02-01 13:06:11 JDuke128 has quit (Max SendQ exceeded)
 930 2013-02-01 13:06:35 <HM> maybe instead of a formal body that aims to keep control then, you need a nightwatch sort of organisation that keeps devs up to date with changes in the network
 931 2013-02-01 13:06:50 JDuke128 has joined
 932 2013-02-01 13:07:00 <HM> like a bitcoin news site that doesn't deal with ZOMG the price went up $0.5!
 933 2013-02-01 13:07:09 <HM> just raw technical stuff
 934 2013-02-01 13:07:12 <sipa> just to be clear: there has - up to my knowledge - never been an incompatible change to the network rules
 935 2013-02-01 13:07:22 <CodeShark> but standards bodies DO exist for this kind of stuff...the problem with bitcoin is that even minor changes in the rules can produce a fork
 936 2013-02-01 13:07:46 JDuke128 has quit (Max SendQ exceeded)
 937 2013-02-01 13:07:53 <CodeShark> with other kinds of networks, there could be patches and workarounds
 938 2013-02-01 13:08:07 <sipa> yes, and for the P2P protocol such things are fine
 939 2013-02-01 13:08:24 JDuke128 has joined
 940 2013-02-01 13:09:00 <CodeShark> if a particular HTTP server throws in a nonstandard field in the header, perhaps some browsers will not be able to render the page...but it won't break the whole network
 941 2013-02-01 13:09:13 JDuke128 has quit (Max SendQ exceeded)
 942 2013-02-01 13:09:28 <HM> it'd be insanity to develop a client that didn't ignore unknown headers
 943 2013-02-01 13:09:50 JDuke128 has joined
 944 2013-02-01 13:09:52 <CodeShark> well, perhaps the header is required to render it correctly :)
 945 2013-02-01 13:10:08 <CodeShark> just throwing out an example - perhaps a contrived one
 946 2013-02-01 13:10:30 JDuke128 has quit (Max SendQ exceeded)
 947 2013-02-01 13:10:33 <CodeShark> point is bitcoin doesn't have the luxury of patching up servers and clients after-the-fact
 948 2013-02-01 13:11:08 JDuke128 has joined
 949 2013-02-01 13:11:16 <HM> there could be more risk atm for mostly unused features
 950 2013-02-01 13:11:27 <CodeShark> by the time you realize a significant portion of the deployed codebase is incompatible, it's too late
 951 2013-02-01 13:11:35 <HM> if they're eventually used then alternative implementations might not have bothered
 952 2013-02-01 13:12:00 <HM> i'm thinking sequence numbrs and time locks
 953 2013-02-01 13:12:10 <HM> but maybe it's not a problem
 954 2013-02-01 13:13:57 * HM goes back to punching javascript in the face
 955 2013-02-01 13:14:14 <Scrat> CodeShark: to make a binary protocol more dynamic you have to complicate it a lot
 956 2013-02-01 13:14:27 <Scrat> HTTP is just ascii
 957 2013-02-01 13:14:45 <CodeShark> my point has nothing to do with the specific format of the data, though
 958 2013-02-01 13:15:05 <Scrat> oh sorry
 959 2013-02-01 13:15:11 <CodeShark> same argument would apply to binary protocols
 960 2013-02-01 13:15:47 <CodeShark> if two machines cannot speak to one another because of a difference in rules, for most networks, it might still be possible to patch one or the other or both sides
 961 2013-02-01 13:16:11 <CodeShark> but with bitcoin, by the time you're able to patch it, there could already be a fork in the block chain
 962 2013-02-01 13:16:44 <Scrat> HM: CS makes it bearable. still crappy but bearable :p
 963 2013-02-01 13:16:52 luke-jr_ is now known as Luke-Jr
 964 2013-02-01 13:17:35 <CodeShark> this is actually a very serious problem going forward
 965 2013-02-01 13:18:50 <HM> Is 'Bitcoin' a registered trademark?
 966 2013-02-01 13:19:02 <HM> the thought hadn't occured to me until just now
 967 2013-02-01 13:19:06 <CodeShark> I'm not sure what the solution is - perhaps we'd have to start entertaining the possibility of retroactive reorgs
 968 2013-02-01 13:19:19 <CodeShark> ugh :P
 969 2013-02-01 13:19:54 <Luke-Jr> HM: I think most places, yes
 970 2013-02-01 13:20:05 <HM> So who owns it? :S
 971 2013-02-01 13:20:07 <bsdunx> I sure sparked some fire asking about a spec huh. =s
 972 2013-02-01 13:20:07 <sipa> retroactive reorgs?
 973 2013-02-01 13:20:16 <Luke-Jr> HM: MtGox
 974 2013-02-01 13:20:38 <Luke-Jr> HM: some trademark troll tried to register it a few years ago, so MtGox intercepted and licensed it for free use
 975 2013-02-01 13:21:05 <CodeShark> sipa: if later on it is found that implementations have allowed blocks that are technically invalid according to the official rules, the blocks would have to be invalidated
 976 2013-02-01 13:21:11 <HM> Luke-Jr: nice
 977 2013-02-01 13:23:43 da2ce744 has joined
 978 2013-02-01 13:24:02 <CodeShark> sipa: this isn't really a complete (nor necessarily desirable) solution
 979 2013-02-01 13:24:16 <CodeShark> sipa: the official rules could still be ambiguous unintentionally
 980 2013-02-01 13:24:43 <CodeShark> and when people have money riding on the outcome of deciding whether or not to invalidate blocks, it can get bloody
 981 2013-02-01 13:25:01 <sipa> CodeShark: no way
 982 2013-02-01 13:25:12 <sipa> in that case, the official rules were wrong
 983 2013-02-01 13:25:16 <sipa> and they need to be fixed :)
 984 2013-02-01 13:25:22 <CodeShark> lol - that's how it works right now
 985 2013-02-01 13:25:28 <sipa> that's the only possible way
 986 2013-02-01 13:25:36 <Luke-Jr> sipa: not necessarily
 987 2013-02-01 13:25:44 <HM> **** the rules! is a philosophy i can get behind
 988 2013-02-01 13:25:45 <Luke-Jr> sipa: we've fixed rules before that invalidated long chains
 989 2013-02-01 13:26:04 <sipa> when it was about obviously inacceptable chains
 990 2013-02-01 13:26:15 <Luke-Jr> well, those are the real rules :P
 991 2013-02-01 13:26:41 <sipa> in the end, the bitcoin chain is whatever 'we' (= all users) accept as chain
 992 2013-02-01 13:27:09 <CodeShark> I still would prefer the criteria for acceptability to be spelled out formally and not in implementations, if possible
 993 2013-02-01 13:27:46 <Luke-Jr> sipa: I'd suggest a slightly different reality: the blockchain is whatever 'he' (= the person you want to pay) accepts as chain ;)
 994 2013-02-01 13:27:58 <HM> the stuff that happens at the tip of the blockchain is murky to me too
 995 2013-02-01 13:28:04 <sipa> Luke-Jr: that's pointless
 996 2013-02-01 13:28:07 <Luke-Jr> :P
 997 2013-02-01 13:28:39 <sipa> there is a large difference between a) an inconsistency between the implementation(s) and what the consensus of users wants  and b) an inconsistency between two implementations or a standard and an implementation, where both alternatives are acceptable to the user base
 998 2013-02-01 13:28:56 <sipa> an example of a) is a cryptographic flag
 999 2013-02-01 13:28:58 <sipa> *flaw
1000 2013-02-01 13:29:01 copumpkin has joined
1001 2013-02-01 13:29:31 <sipa> if the alternative is making bitcoin useless because of a trivial exploit, you indeed need something like a retroactive fix
1002 2013-02-01 13:29:50 <sipa> but for b) it can only reduce confidence in the system
1003 2013-02-01 13:30:33 <CodeShark> as long as we rely on a single reference implementation to define the network, we haven't solved decentralization
1004 2013-02-01 13:30:58 agricocb has quit (Quit: Leaving.)
1005 2013-02-01 13:31:19 <CodeShark> this could very well be bitcoin's own undoing
1006 2013-02-01 13:31:32 <Luke-Jr> CodeShark: hence my work on Eloipool recently, that nobody seems interested in reviewing <.<
1007 2013-02-01 13:31:41 <HM> i think there a bunch of implementations at this point
1008 2013-02-01 13:31:50 <sipa> CodeShark: i'm not talking about a reference implementation or not
1009 2013-02-01 13:32:10 <CodeShark> sipa: then what are you talking about?
1010 2013-02-01 13:32:39 <sipa> i'm just saying that the *correct* rules are always defined by the network (whatever implementationS it consists of), and not by a standards document
1011 2013-02-01 13:32:45 <Luke-Jr> CodeShark: any interest? :P http://codepad.org/58dWbqWm - goal is to verify proposed blocks with N implementations before mining them
1012 2013-02-01 13:33:36 <CodeShark> that's an interesting possibility, Luke-Jr
1013 2013-02-01 13:33:36 <sipa> and right now the only fully-validation codebase with widespread deployment on the network is the reference client
1014 2013-02-01 13:33:44 <sipa> i hate that possibility
1015 2013-02-01 13:33:57 <sipa> but it may be the only choice at some point in the future
1016 2013-02-01 13:34:01 <Luke-Jr> sipa: ? :o
1017 2013-02-01 13:34:05 JZavala has quit (Ping timeout: 260 seconds)
1018 2013-02-01 13:34:18 <sipa> let me rephrase: i hate the idea that such a measure is necessary
1019 2013-02-01 13:34:24 <sipa> but it may be inevitable
1020 2013-02-01 13:34:35 <Luke-Jr> sipa: it's a good way to test things, in any case
1021 2013-02-01 13:34:40 <sipa> certainly
1022 2013-02-01 13:34:49 <CodeShark> yes - and it might be the only way to really prevent forks
1023 2013-02-01 13:35:07 <Luke-Jr> that's why my code up there will write a file with details for any template rejected by one or more implementations
1024 2013-02-01 13:35:58 <Luke-Jr> (so that should there be an incompatibility, we can investigate why)
1025 2013-02-01 13:36:21 <sipa> yes, as a way for comparing implementations with eachother, it is very nice
1026 2013-02-01 13:37:12 <CodeShark> such a tool should be running on the network regardless of whether it is widely used by miners, IMHO
1027 2013-02-01 13:37:25 datagutt has joined
1028 2013-02-01 13:37:42 <Luke-Jr> actually, I wonder if running a fuzzer through Proposals would make sense
1029 2013-02-01 13:37:49 <HM> like a watchdog
1030 2013-02-01 13:38:12 <gmaxwell> another fun thing to do would be to have a collection of nodes mining 'diff 0.00001' forks at every height.
1031 2013-02-01 13:38:37 <Luke-Jr> O.o
1032 2013-02-01 13:38:48 Prattler has quit (Quit: ZNC - http://znc.in)
1033 2013-02-01 13:39:05 <gmaxwell> e.g. take N nodes, put them behind a gateway that doesn't allow block to exit, and rig their diff logic so they produce blocks every few seconds. (but prefer the real chain).
1034 2013-02-01 13:40:05 <gmaxwell> so at every height you get a bunch of forks, and can tell if they go divergent. So you'd learn that there is a txn in mempool that forces divergence even before the network mines it.
1035 2013-02-01 13:40:30 <CodeShark> you can separate proof-of-work verification from everything else
1036 2013-02-01 13:40:46 <CodeShark> so you can compare blocks without regard for the nonce
1037 2013-02-01 13:41:06 <kuzetsa> oh wow... that's actually kinda smart
1038 2013-02-01 13:41:37 <Luke-Jr> CodeShark: that's what Proposals do
1039 2013-02-01 13:41:51 <Luke-Jr> https://github.com/bitcoin/bitcoin/pull/1816
1040 2013-02-01 13:41:52 <CodeShark> ok, I might be reinventing stuff
1041 2013-02-01 13:42:01 <CodeShark> to be completely honest, I haven't even touched mining code
1042 2013-02-01 13:42:20 <gmaxwell> Building a fork is even better than just comparing proposals.
1043 2013-02-01 13:43:06 <gmaxwell> ... e.g. what if a bad txout leaves your coin data in a inconsistent shape and the divergence comes later?
1044 2013-02-01 13:43:38 * kuzetsa swallows nervously
1045 2013-02-01 13:43:49 <kuzetsa> gmaxwell: theoretically, that may have already happened and we just haven't noticed.
1046 2013-02-01 13:44:32 <gmaxwell> kuzetsa: harder for a single piece of software to diverge with itself.
1047 2013-02-01 13:44:47 zooko` has joined
1048 2013-02-01 13:45:22 <kuzetsa> I'm not sure I understand "inconsistent shape" or "... software to diverge with itself"
1049 2013-02-01 13:45:28 sssssssssss has quit (Ping timeout: 245 seconds)
1050 2013-02-01 13:45:45 <kuzetsa> like... the individual words, sure, but I'm not 100% sure I understand this context / usage case for the concepts
1051 2013-02-01 13:45:45 <sipa> BIP30 was an example of a fix for the code to become inconsistent with other instances of itself
1052 2013-02-01 13:46:48 <HM> CodeShark: most of the voodoo seems to be in the mining/blockchain rules
1053 2013-02-01 13:47:26 <CodeShark> HM: agreement on what constitutes a valid block is the very essence
1054 2013-02-01 13:49:29 Goonie has joined
1055 2013-02-01 13:50:37 Prattler has joined
1056 2013-02-01 13:51:25 Prattler has quit (Client Quit)
1057 2013-02-01 13:51:31 <kuzetsa> sipa: thanks. I just tracked down the BIP30 code and confirmed presence of the BIP30 code in 0.7.2 linux source tarball thinger :)
1058 2013-02-01 13:51:51 <kuzetsa> and git blame says it's yours :)
1059 2013-02-01 13:52:19 <gmaxwell> and?
1060 2013-02-01 13:52:32 <sipa> kuzetsa: yes, i know that - but do you understand what it is about?
1061 2013-02-01 13:52:33 <kuzetsa> gmaxwell: it doesn't seem to address the problem currently under discussion, so ... I unno
1062 2013-02-01 13:52:48 <kuzetsa> oh, the BIP30?
1063 2013-02-01 13:52:56 <gmaxwell> Nothing can address the problem currently under discussion.
1064 2013-02-01 13:52:59 <kuzetsa> yeah, it seems related to an attack I read about a few months ago
1065 2013-02-01 13:53:20 <gmaxwell> as sipa says "BIP30 was an example of a fix for the code to become inconsistent with other instances of itself"
1066 2013-02-01 13:53:48 t7 has quit (Read error: Connection reset by peer)
1067 2013-02-01 13:54:01 <gmaxwell> Which was said becase I said its "harder for a single piece of software to diverge with itself" and Sipa then used bip30 of an example of it not being impossible.
1068 2013-02-01 13:54:02 <sipa> kuzetsa: BIP30 was about fixing a problem, where if two nodes (running the same code) are on the network, and one sees a (very specific) block that gets reorganized, and the other doesn't see the block at all, both nodes end up with a different database state, where one would except a specific future transactions and the other wouldn't
1069 2013-02-01 13:54:12 t7 has joined
1070 2013-02-01 13:54:15 <kuzetsa> gmaxwell / sipa: huh? what? I thought BIP30 only addresses ... yeah, that
1071 2013-02-01 13:54:18 <sipa> this was very subtle, and very specific
1072 2013-02-01 13:54:28 <gmaxwell> s/would except a/would accept a/
1073 2013-02-01 13:54:45 <sipa> but it shows it is possible to have bugs that causes nodes running the same code to become inconsistent with eachother
1074 2013-02-01 13:54:55 rdymac has quit (Quit: This computer has gone to sleep)
1075 2013-02-01 13:54:59 <kuzetsa> yeah, fair enough.
1076 2013-02-01 13:55:26 <sipa> just changing any network rule or constant or pretty much everything, would just result in nodes running that code to be inconsistent with other nodes _not_ running the same code
1077 2013-02-01 13:56:11 * kuzetsa nods
1078 2013-02-01 13:57:59 <CodeShark> the fragility of bitcoin worries me
1079 2013-02-01 13:58:25 <gmaxwell> sipa: or 'changing' any rule can easily cause old nodes to become inconsistent with new bootstrapped copies of the same software
1080 2013-02-01 13:58:49 <gmaxwell> CodeShark: it's no fault of bitcoin-the-software ... it's the fundimental nature of distibuted consensus.
1081 2013-02-01 13:59:01 <kuzetsa> I'm kinda wondering if something similar to the TS protocol (IRC) could be added to the bitcoin network for the purpose of transitioning to newer versions of the protocol / changing constants / etc.
1082 2013-02-01 13:59:48 <HM> CodeShark: meh, bitcoin will always be as vulnerable as the coders that write it and the network that carries it
1083 2013-02-01 13:59:48 <CodeShark> if only there were a relatively smooth way to merge forks that did not require centralization
1084 2013-02-01 14:00:02 <kuzetsa> http://en.wikipedia.org/wiki/Internet_Relay_Chat#Timestamping <--- seems reasonably well weitten, by wikipedia standards.
1085 2013-02-01 14:00:10 <kuzetsa> oops typo *written
1086 2013-02-01 14:00:20 <Scrat> kuzetsa: hmm what does the TS protocol have to do with versioning or transitioning
1087 2013-02-01 14:00:34 <kuzetsa> "so and so is opped, or a new nick, etc."
1088 2013-02-01 14:00:51 <sipa> CodeShark: that's impossible, the way i see it
1089 2013-02-01 14:00:54 <kuzetsa> and netsplits / inconsistent states, deciding which nicks, op states, bans, etc. get merged.
1090 2013-02-01 14:01:04 <HM> CodeShark: nobody seems to be running around screaming about the fragility of the Internet, yet practically every core protocol is a mess
1091 2013-02-01 14:01:16 <sipa> CodeShark: as the entire reason for the consensus protocol is deciding about the order of transactions
1092 2013-02-01 14:01:30 <gmaxwell> CodeShark: all you can do is further defer consensus.
1093 2013-02-01 14:01:38 <gmaxwell> and if you're going to do that, no real need to merge.
1094 2013-02-01 14:04:20 <HM> kuzetsa: the irc netsplit timestamp thing assumes all the irc servers are trusted (which makes sense but it's a different thing entirely)
1095 2013-02-01 14:05:02 <kuzetsa> HM: yeah, but I was looking more at things such as implementing a protocol change or a network rule to change a constant
1096 2013-02-01 14:05:40 <gmaxwell> kuzetsa: you won't find any— they'll all depend on some authority to approve the constant.
1097 2013-02-01 14:05:51 <kuzetsa> there are already some protocol changes which were done by "voting" (after a certain percent of the bitcoin network supports <x> feature)
1098 2013-02-01 14:06:46 <gmaxwell> Not quite.
1099 2013-02-01 14:07:04 <kuzetsa> well not changing the constants defined by protocol so far, but for SOMETHING I remember this has happened
1100 2013-02-01 14:07:14 <kuzetsa> the specific thing(s) don't come to mind though
1101 2013-02-01 14:07:41 <gmaxwell> And we have no way to measure 'a certain percent of the bitcoin network supports <x> feature'.. we can measure mining, which is fine for but only acceptable for things which simply need a hashpower supermajority to work.
1102 2013-02-01 14:07:46 <gmaxwell> e.g. P2SH.
1103 2013-02-01 14:08:26 <gmaxwell> (e.g. things which only involve rejecting otherwise valid transactions)
1104 2013-02-01 14:09:26 <kuzetsa> actually, p2sh sounds right... I'm pretty sure it was like something in the signatures method(s)
1105 2013-02-01 14:09:48 <kuzetsa> yeah, that was it --> http://blockchain.info/p2sh
1106 2013-02-01 14:10:50 yareyare has quit (Quit: zzz)
1107 2013-02-01 14:13:35 <kuzetsa> I had a bit of trouble remembering since I hadn't heard of bitcoin before june 2012 (so it wasn't something I experienced personally)
1108 2013-02-01 14:15:01 <HM> wow "generations can't be spent for 100 blocks"
1109 2013-02-01 14:15:08 <HM> TIL
1110 2013-02-01 14:15:19 <kuzetsa> HM: really? I thought it was 120 blocks
1111 2013-02-01 14:15:32 <HM> well the wiki vocab page needs updating then :P
1112 2013-02-01 14:16:08 <sipa> the protocol requires 100 blocks (=101 confirmations)
1113 2013-02-01 14:16:21 <sipa> the client enforces 119 blocks (=120 confirmations)
1114 2013-02-01 14:16:30 occulta has joined
1115 2013-02-01 14:16:56 <kuzetsa> oh
1116 2013-02-01 14:17:20 <HM> sipa: why the difference?
1117 2013-02-01 14:17:24 <gmaxwell> s/enforces/won't spend before/  for clarity (otherwise it sounds like it holds a higher protocol rule)
1118 2013-02-01 14:17:36 <davout> ohai
1119 2013-02-01 14:17:47 <davout> who manages the "trusted users" list on the bitcoin wiki
1120 2013-02-01 14:18:42 <gmaxwell> "trusted users" list?
1121 2013-02-01 14:19:21 <davout> yeah
1122 2013-02-01 14:19:35 <davout> i kinda choked discovering this
1123 2013-02-01 14:21:25 <sipa> link?
1124 2013-02-01 14:21:59 WolfAlex_ has joined
1125 2013-02-01 14:22:20 <gmaxwell> search on the wiki turns up nothing.
1126 2013-02-01 14:22:29 * gmaxwell takes away davout's crackpipe
1127 2013-02-01 14:22:31 PhantomSpark has joined
1128 2013-02-01 14:23:03 jgarzik has quit (Read error: Operation timed out)
1129 2013-02-01 14:23:09 <kuzetsa> perhaps davout is trying to edit a page marked "protected"
1130 2013-02-01 14:23:21 jgarzik has joined
1131 2013-02-01 14:23:40 <davout> https://dl.dropbox.com/u/52452135/Capture%20d%E2%80%99%C3%A9cran%202013-02-01%20%C3%A0%2015.05.31.png
1132 2013-02-01 14:23:45 jgarzik is now known as Guest65828
1133 2013-02-01 14:23:51 <gmaxwell> (or he's bonced off BitcoinPayment?)
1134 2013-02-01 14:24:03 <Eliel> davout: 404
1135 2013-02-01 14:24:07 <davout> well, there seem to be a bunch of pages that were marked "protected" recently
1136 2013-02-01 14:24:19 <kuzetsa> https://en.bitcoin.it/w/index.php?title=Main_Page&action=edit ---> You do not have permission to edit this page, for the following reasons: The action you have requested is limited to users in the group: Trusted users.
1137 2013-02-01 14:24:32 <kuzetsa> (I'm neither logged in, nor in the "Trusted users" group that I'm aware)
1138 2013-02-01 14:24:50 <gmaxwell> oh, thats what the name of the group is for BitcoinPayment apparently
1139 2013-02-01 14:25:00 WolfAlex has quit (Ping timeout: 244 seconds)
1140 2013-02-01 14:25:01 <gmaxwell> See the big bright orange notice on the login screen.
1141 2013-02-01 14:25:41 ThomasV has joined
1142 2013-02-01 14:26:29 <davout> hmm going to look at it, i'm permanently logged in to the wiki so i didn't see this thingy
1143 2013-02-01 14:27:22 * kuzetsa hijack-steals davout's cookies, and trolls everyone using stolen login perma-loggedin credentials
1144 2013-02-01 14:27:51 <davout> good luck hijacking my stuff :)
1145 2013-02-01 14:28:50 <davout> that's kinda lame
1146 2013-02-01 14:28:52 <kuzetsa> if I had that sort of time & cyber resources on my hands I'd be hijacking proper bank fiat money rather than bothering to mess with bitcoin-related ... well anything bitcoin really
1147 2013-02-01 14:28:53 Prattler has joined
1148 2013-02-01 14:29:22 <kuzetsa> appologies for the sarcastic dead-pan humor, I wasn't serious & didn't mean to sound as such.
1149 2013-02-01 14:31:05 zooko` has quit (Ping timeout: 260 seconds)
1150 2013-02-01 14:31:11 <davout> what ?
1151 2013-02-01 14:31:25 <davout> talking about the paying users become trusted users
1152 2013-02-01 14:31:28 <kuzetsa> davout: I have no interest in your cookies, it was a (bad) joke.
1153 2013-02-01 14:31:37 <kuzetsa> "paying users" ?
1154 2013-02-01 14:31:41 <davout> yeah
1155 2013-02-01 14:31:47 <kuzetsa> I don't understand
1156 2013-02-01 14:31:56 <davout> well paying .05 BTC to get edit privileges on the wiki...
1157 2013-02-01 14:32:12 <davout> my cookies are delicious, you should desire them
1158 2013-02-01 14:32:12 <kuzetsa> they charge money for edit privs?
1159 2013-02-01 14:32:17 <kuzetsa> ... when did that start?
1160 2013-02-01 14:32:22 <davout> i don't know
1161 2013-02-01 14:32:37 <kuzetsa> hmm... isn't the bitcoin wiki hosted by mtgox?
1162 2013-02-01 14:32:45 <kuzetsa> ... damn non-non-profits :(
1163 2013-02-01 14:33:04 <davout> yes it is
1164 2013-02-01 14:33:11 <davout> which is also lame if you ask me
1165 2013-02-01 14:33:20 <davout> anyway
1166 2013-02-01 14:33:36 <davout> guess the wiki will remain unedited
1167 2013-02-01 14:35:08 <kuzetsa> https://en.bitcoin.it/wiki/BitcoinPayment#Questions_or_comments.3F ---> If you have questions or would like help with the process, drop by on #bitcoin-wiki on freenode IRC <--- any reason you're not asking in there instead of here?
1168 2013-02-01 14:35:55 <davout> i'm not very much of an irc kind of person, didn't even know there was a channel for the wiki
1169 2013-02-01 14:35:55 <gmaxwell> kuzetsa: it's not a profit thing— you don't contribute to the discourse by trolling.
1170 2013-02-01 14:36:16 <gmaxwell> IIRC it was nanotube's recommendation.
1171 2013-02-01 14:37:06 <davout> gmaxwell: doesn't really make much sense to me, discourages honest editors, i think a whitelist would have been more efficient
1172 2013-02-01 14:37:27 Guest65828 has quit (Changing host)
1173 2013-02-01 14:37:27 Guest65828 has joined
1174 2013-02-01 14:37:32 Guest65828 is now known as jgarzik_
1175 2013-02-01 14:38:01 dvide has joined
1176 2013-02-01 14:39:51 <jgarzik_> I think adding a whitelist would be good
1177 2013-02-01 14:40:00 <jgarzik_> let a few trusted community members whitelist
1178 2013-02-01 14:40:08 <jgarzik_> but keep the payment stuff.
1179 2013-02-01 14:40:32 occulta has quit (Remote host closed the connection)
1180 2013-02-01 14:42:51 agricocb has joined
1181 2013-02-01 14:44:02 <davout> i think a manually whitelist by itself is the right way to go, with the payment system you can automate rogue edits at anytime for only .05 BTC with a much higher return possibility
1182 2013-02-01 14:44:56 <kuzetsa> ;;calc 0.05 * [ticker --avg]
1183 2013-02-01 14:44:56 <gribble> 1.042772
1184 2013-02-01 14:45:05 BTCOxygen has quit (Ping timeout: 246 seconds)
1185 2013-02-01 14:45:31 <kuzetsa> yeah, but at a dollar per sockpuppet... I dunno
1186 2013-02-01 14:46:02 <kuzetsa> davout: glad you got it sorted :)
1187 2013-02-01 14:47:23 nus has quit (Quit: Leaving)
1188 2013-02-01 14:48:15 <davout> paying one dollar for the opportunity to scam people out of much more money isn't a bad deal
1189 2013-02-01 14:49:21 <gmaxwell> davout: hey, if someone wants to blow 0.05 btc per easily reverted junk edit, .. fine with me... "we've found a way to sustainably fund the bitcoin foundation!"
1190 2013-02-01 14:50:43 Grishnakh has quit (Ping timeout: 264 seconds)
1191 2013-02-01 14:50:54 <HM> bI'm a Bit disappointed more Bitcoin sites don't use BrowserID.
1192 2013-02-01 14:50:58 <davout> gmaxwell: yeah well... i just think it's kinda weird and that it'd be better to have a list instead
1193 2013-02-01 14:51:38 rdponticelli has joined
1194 2013-02-01 14:51:39 <HM> I wonder if it's possible to use BrowserID using wallet keypairs
1195 2013-02-01 14:53:13 tonikt has joined
1196 2013-02-01 14:54:40 Zarutian has quit (Quit: Zarutian)
1197 2013-02-01 14:55:40 <HM> i think it uses the wrong curve
1198 2013-02-01 14:55:45 <HM> P-256
1199 2013-02-01 14:57:33 Guest82692 is now known as topace
1200 2013-02-01 14:57:39 topace has quit (Changing host)
1201 2013-02-01 14:57:39 topace has joined
1202 2013-02-01 14:58:24 <HM> why can't these people name these curves more creatively
1203 2013-02-01 14:58:37 <HM> SquigglyOneWithTheHump2
1204 2013-02-01 14:59:44 setkeh- has joined
1205 2013-02-01 14:59:57 paraipan has joined
1206 2013-02-01 15:02:23 <gmaxwell> the browserid auths should be shortlived in any case.
1207 2013-02-01 15:04:16 setkeh- has quit (Quit: setkeh.com)
1208 2013-02-01 15:06:21 setkeh has joined
1209 2013-02-01 15:10:00 occulta has joined
1210 2013-02-01 15:10:45 leotreasure has left ()
1211 2013-02-01 15:10:54 JDuke128 has quit (Quit: ["Textual IRC Client: www.textualapp.com"])
1212 2013-02-01 15:13:08 occulta has quit (Remote host closed the connection)
1213 2013-02-01 15:15:59 nnasm has joined
1214 2013-02-01 15:22:17 rdymac has joined
1215 2013-02-01 15:25:38 t7` has joined
1216 2013-02-01 15:29:00 t7 has quit (Ping timeout: 255 seconds)
1217 2013-02-01 15:31:36 pisto has joined
1218 2013-02-01 15:32:02 <pisto> is there a C api or something to access the blocks raw data?
1219 2013-02-01 15:32:34 <kjj> the current file structure is very simple.  you can parse it out if you want it
1220 2013-02-01 15:32:45 <gmaxwell> ::sigh:: it looks like the code that added redeemScript broke signrawtransaction for the regular signing for an input not in the wallet case.
1221 2013-02-01 15:33:11 <gmaxwell> kjj: and then hose yourself becase the files contain orphans. :P
1222 2013-02-01 15:33:50 <gmaxwell> can someone grab d127a741660be02c01855c679ff8de7755bb6c2b2ceaa4848e02b14f4f0aae59  vout 8's scriptPubKey for me?
1223 2013-02-01 15:34:11 <pisto> kjj, how?
1224 2013-02-01 15:34:40 <pisto> is it just a stream of the basic block format?
1225 2013-02-01 15:34:55 <sipa> yes, with 4-byte magic values and 4-byte length records in between
1226 2013-02-01 15:35:03 <kjj> the block format is basically just the raw network protocol "block: message
1227 2013-02-01 15:35:06 nnasm has quit (Quit: Page closed)
1228 2013-02-01 15:35:53 <kjj> as gmaxwell pointed out, not all blocks in the file are part of the main chain.  you have to be careful
1229 2013-02-01 15:36:09 <sipa> pisto: the reference client also exposes raw block data using the getblock RPC
1230 2013-02-01 15:37:28 <jgarzik_> ASIC now mining on BTCGuild - variable difficulty!
1231 2013-02-01 15:38:02 rdymac has quit (Quit: This computer has gone to sleep)
1232 2013-02-01 15:38:09 <CodeShark> having fun with your new toy, jgarzik? :)
1233 2013-02-01 15:38:35 <gmaxwell> CodeShark: he's just having fun giving away a couple percent of his income to a multitude of parties. :P
1234 2013-02-01 15:38:52 <CodeShark> I'll take some, too :p
1235 2013-02-01 15:39:31 andytoshi has quit (Quit: WeeChat 0.3.9.2)
1236 2013-02-01 15:40:36 <epscy> why did they send it jgarzik_, surely gavin would be the more logical choice
1237 2013-02-01 15:41:06 <gmaxwell> epscy: jgarzik bought one
1238 2013-02-01 15:41:15 <CodeShark> lol - good reason
1239 2013-02-01 15:41:19 <gavinandresen> I don't know nuthin about hardware
1240 2013-02-01 15:42:00 <gmaxwell> (so did I, though it's good that they didn't send the first one to me— all my stuff is in a moving van right now)
1241 2013-02-01 15:42:03 <sipa> jgarzik_: what was the problem with p2pool?
1242 2013-02-01 15:43:29 <gmaxwell> sipa: what I got out of the conversation was that it was restarting the controller for every longpoll... but it sounded fixable.
1243 2013-02-01 15:44:22 <t7`> satoshi should have chosen bcrypt or something
1244 2013-02-01 15:44:31 <t7`> then i could run a big botnet :3
1245 2013-02-01 15:44:35 t7` is now known as t7
1246 2013-02-01 15:45:45 <gmaxwell> t7: so you're a theif who steals other people's computing resouces and power?
1247 2013-02-01 15:46:08 <gmaxwell> (—cause thats what it sounds like you're saying!)
1248 2013-02-01 15:48:12 <epscy> oh i thought it was free
1249 2013-02-01 15:48:15 <epscy> fair enough
1250 2013-02-01 15:49:34 <HM> <gmaxwell> the browserid auths should be shortlived in any case.
1251 2013-02-01 15:50:39 <HM> presumably you could do ephemeral-static D-H with the static side being a bitcoin wallet keypair
1252 2013-02-01 15:50:47 <gmaxwell> epscy: do you have bitcoin running right now?  please getrawtransaction d127a741660be02c01855c679ff8de7755bb6c2b2ceaa4848e02b14f4f0aae59  and pastebin me the reslts
1253 2013-02-01 15:50:56 <HM> if the objective was to authenticate against a wallet rather than a trusted third party like Mozilla Persona
1254 2013-02-01 15:51:16 rdponticelli has quit (Quit: No Ping reply in 180 seconds.)
1255 2013-02-01 15:51:20 <gmaxwell> HM: You should go read how browserid works.
1256 2013-02-01 15:51:38 <HM> i thought i had :S
1257 2013-02-01 15:52:44 <gmaxwell> HM: the idea of browser ID is that you generate an ephemeral keypair with a short lifetime and have some identity provider sign it.  (the whole thing also mandates that the identy being confirmed is an email address)
1258 2013-02-01 15:52:47 MrTiggr has quit (Ping timeout: 248 seconds)
1259 2013-02-01 15:52:56 <HM> yes
1260 2013-02-01 15:55:36 <gavinandresen> gmaxwell: signrawtransaction is broken?  open an issue when you have time...
1261 2013-02-01 15:55:49 comepradz has joined
1262 2013-02-01 15:56:03 comepradz has left ()
1263 2013-02-01 15:56:57 <HM> gmaxwell: atm on bitcoin wiki you login using an email/pass. you have no sure way to know that the person who paid the donation is the person logging in
1264 2013-02-01 15:57:09 <gmaxwell> gavinandresen: I'll submit a pull request once I can test my fix. But it seems that no one here will give me the scriptpubkey for d127a741660be02c01855c679ff8de7755bb6c2b2ceaa4848e02b14f4f0aae59:8 and it will probably be two weeks before I have a node with the blockchain again....
1265 2013-02-01 15:57:26 <gmaxwell> HM: sure. And? What does that even mean?
1266 2013-02-01 15:57:42 <gmaxwell> HM: If I get a loan from you to pay it am I no longer the person paying it?
1267 2013-02-01 15:57:44 <sipa> gmaxwell: see PM
1268 2013-02-01 15:57:57 <HM> my musing was that you should be able to have a variant of browserid that utilised wallet keypairs to authenticate that the person logging in is actually the person who made the donation
1269 2013-02-01 15:58:07 <pisto> when is the time field of the block header used, or is it just informative? are there other time related fields?
1270 2013-02-01 15:58:09 <HM> or at least, now controls the wallet
1271 2013-02-01 15:58:31 <sipa> pisto: it is used to calculate difficulty changes
1272 2013-02-01 15:58:35 <HM> i thought it could be useful for more anonymous membership based schemes
1273 2013-02-01 15:58:37 <gavinandresen> gmaxwell: I'd be happy to do it, but am out the door to daughter's orthodontist appointment right now…
1274 2013-02-01 15:58:51 <gmaxwell> sipa got me. :)
1275 2013-02-01 15:59:10 <epscy> yeah sorry i can't do that right now
1276 2013-02-01 15:59:10 <pisto> sipa, what kind of sanity checks are run on it? I can't say I generated this block in 2299 right?
1277 2013-02-01 15:59:18 <jgarzik_> sipa: what gmaxwell said
1278 2013-02-01 15:59:25 jgarzik_ is now known as jgarzik
1279 2013-02-01 15:59:56 <gavinandresen> bug is it needs all of the scriptPubKeys or none?  If yes, you can call signrawtransaction twice....
1280 2013-02-01 16:00:01 <gmaxwell> HM: your goal doesn't make sense to me.  You can never tell who really paid for something.
1281 2013-02-01 16:00:22 <gmaxwell> gavinandresen: no, thats its forcing me to provide a redeemScript for a non-p2sh output.
1282 2013-02-01 16:00:32 <sipa> pisto: not earlier than the median of the past 11 blocks, not later than the current time + 2 hours
1283 2013-02-01 16:00:53 <HM> gmaxwell: if you run a site that requires paid membership you want to be sure the people accessing the service are the same people who paid
1284 2013-02-01 16:01:06 <gmaxwell> HM: thats not possible to do.
1285 2013-02-01 16:01:20 <gmaxwell> Its a kind of deep and fundimental impossiblity.
1286 2013-02-01 16:01:38 <sipa> HM: what if i want to pay for someone else's membership?
1287 2013-02-01 16:01:56 <HM> sipa: then send them the money first
1288 2013-02-01 16:02:02 <HM> let them signup
1289 2013-02-01 16:03:02 <gmaxwell> HM: say I want to share my membership with you. I first transfer the payment amount to a new key, I then 'pay' with that key (getting into the bad and horribly borken assumption that you can determine a 'sender' from a transaction) and that key pays. Then I share that privkey with yo.
1290 2013-02-01 16:03:52 <HM> there's nothing wrong with that use case
1291 2013-02-01 16:03:59 Diapolo has joined
1292 2013-02-01 16:04:10 zipet has joined
1293 2013-02-01 16:04:16 <gmaxwell> HM: huh? it achieved absolutely nothing.
1294 2013-02-01 16:04:24 Detritus has quit (Ping timeout: 264 seconds)
1295 2013-02-01 16:04:25 rdponticelli has joined
1296 2013-02-01 16:04:28 <kjj> hmm.  the network protocol checksum, is that really calculated JUST on the payload like the wiki says?
1297 2013-02-01 16:04:56 <gmaxwell> You know which people have paid in the way its done now by virtue of who you gave the address that got paid to.
1298 2013-02-01 16:05:21 <gmaxwell> And doing anything else can't prevent someone from sharing or handing off an account .
1299 2013-02-01 16:05:54 <HM> i never said preventing sharing was a goal
1300 2013-02-01 16:05:55 <gmaxwell> HM: what am I missing?
1301 2013-02-01 16:06:23 <sipa> kjj: afaik yes
1302 2013-02-01 16:06:33 <gmaxwell> HM: what is the goal then?
1303 2013-02-01 16:07:08 Detritus has joined
1304 2013-02-01 16:07:24 <HM> gmaxwell: imagine i subscribe to an online magazine and i pay with bitcoin. with browserid my authentication method is associated with my email address and my password, and a trusted third party. very weak and potentially vulnerable channels. If those are compromised i lose my subscription. If I can prove i'm the person (actually wallet holder) of the original subscription payment then i avoid that additional guff
1305 2013-02-01 16:08:20 <sipa> so as soon as you have your subscription that you paid 1 BTC for, you go seel it to N>10 people for 0.1 BTC each... ???... profit!
1306 2013-02-01 16:08:26 <HM> browserid is still centralised in the sense that it's a trusted third party, the blockchain is open to everyone and anyone can verify i made a payment to the magazine
1307 2013-02-01 16:08:53 <HM> it's not about preventing abuse of sharing accounts
1308 2013-02-01 16:09:35 <gmaxwell> HM: why not just give them a recovery key when you sign up if the problem you are trying to solve is losing your email account?
1309 2013-02-01 16:09:48 <gmaxwell> I don't see why anything you're talking about has any reason to be bound to a payment.
1310 2013-02-01 16:10:14 <HM> because it's the initial payment that gives you the right to use the service, now ownership of a password/email account etc.
1311 2013-02-01 16:10:19 <HM> browserid is about identity
1312 2013-02-01 16:10:29 <HM> i'm talking about membership
1313 2013-02-01 16:10:43 zipet has quit (Quit: Page closed)
1314 2013-02-01 16:10:46 <HM> you can (try) to prevent account sharing using other means
1315 2013-02-01 16:11:39 <HM> browserid isn't portable either, sites choose to trust mozilla persona and you have to go with it. the blockchain is distributed so anyone can verify you made a membership payment
1316 2013-02-01 16:11:41 <gmaxwell> HM: virtually all sites that have identites need you to create accounts before you make a payment. And what of when it requires periodic payments? is your identity changing each time? really?
1317 2013-02-01 16:12:07 <gmaxwell> HM: thats not true at all. You don't have to trust any particular intermediary.
1318 2013-02-01 16:12:44 <gmaxwell> Sites can choose to trust one to use when email provders don't provide their own ticket signing gateways.
1319 2013-02-01 16:13:45 <HM> intermediary?
1320 2013-02-01 16:16:07 <gmaxwell> HM:  the idea behind browserid is that your email provider will sign your tickets for you.  But because it will take time before that is widely deployed, people with email addresses with non-suppoting providers can have their addresses certified by a third party assuming the recieving parties (the sites you log into ) choose to trust it.
1321 2013-02-01 16:16:09 eckey has joined
1322 2013-02-01 16:16:28 <HM> ah yes, so you can have a chain of trust
1323 2013-02-01 16:16:42 <gmaxwell> 0_o
1324 2013-02-01 16:17:03 Yrouel has joined
1325 2013-02-01 16:17:05 <Yrouel> hi
1326 2013-02-01 16:17:10 Nachtwind has left ("Verlassend")
1327 2013-02-01 16:17:16 <HM> e.g. i can login to someones blog, if they choose BrowserIDVendorFooBar, as long as BrowserIDVendorFooBar trust Mozilla Persona
1328 2013-02-01 16:17:25 <HM> but that's irrelevant to what i'm proposing
1329 2013-02-01 16:17:54 <Yrouel> is possible to recover user and password from a bitcoin stealer malware having the its exe?
1330 2013-02-01 16:18:07 <Yrouel> the one used to send the wallet by mail
1331 2013-02-01 16:18:23 t7 has quit (Read error: Connection reset by peer)
1332 2013-02-01 16:18:25 <Yrouel> (I haven't been victim from it)
1333 2013-02-01 16:18:38 <gmaxwell> HM: no.. thats not what I'm saying.
1334 2013-02-01 16:18:49 t7 has joined
1335 2013-02-01 16:19:21 <sipa> Yrouel: username/password of what?
1336 2013-02-01 16:19:28 zooko` has joined
1337 2013-02-01 16:19:58 <Yrouel> sipa this stealer is programmed by the attacker which choose what to steal and use an email account to receive the stolen goods
1338 2013-02-01 16:20:00 <gmaxwell> HM: browserID is about turning control of your email address(es) into an account that can directly authenticate with sites.
1339 2013-02-01 16:20:13 zooko` is now known as zooko
1340 2013-02-01 16:20:19 <sipa> Yrouel: you haven't answered my question
1341 2013-02-01 16:20:31 <Yrouel> this is done from the command center which based on such settings build the exe that the victim has to run
1342 2013-02-01 16:20:32 <Yrouel> so
1343 2013-02-01 16:20:41 <Yrouel> the exe must know how to send that email
1344 2013-02-01 16:20:46 <gmaxwell> HM: The idea has nothing to do with Mozilla Persona. Mozilla Persona is just a bootstrapping mechenism because expecting all email providers to support it on day one is unreasonable.
1345 2013-02-01 16:21:05 <Yrouel> so my quesion would be possible to recover them from the trojan?
1346 2013-02-01 16:21:15 <Yrouel> sipa the credentials to the attacker mail
1347 2013-02-01 16:21:17 <gmaxwell> Yrouel: becase ... it's not possibl to send someone an email without knowing their password?
1348 2013-02-01 16:21:28 <sipa> you don't need a password to send a mail
1349 2013-02-01 16:22:35 <Yrouel> mh I based what I said on this screenshot http://puu.sh/162vn
1350 2013-02-01 16:23:08 <Yrouel> and I recall that gmail smtp requires a login
1351 2013-02-01 16:23:14 <CodeShark> what's that? a password cracker?
1352 2013-02-01 16:23:25 <Yrouel> (the exe I have seems to contact gmail)
1353 2013-02-01 16:23:29 <HM> gmaxwell: the browserid assumption is that email address identifies your right to access a service, i'm saying really it's payment.
1354 2013-02-01 16:23:36 <Scrat> Yrouel: you can try dumping all the strings in that file with http://technet.microsoft.com/en-us/sysinternals/bb897439.aspx
1355 2013-02-01 16:23:52 <Yrouel> CodeShark it's a multipurpose stealer
1356 2013-02-01 16:23:54 <Scrat> the hard way would be to disassemble the entire thing
1357 2013-02-01 16:24:53 <Yrouel> Scrat I already used strings however I don't know why certain strings that I can clearly see with a simple texteditor didn't appeared with strings
1358 2013-02-01 16:24:56 rdymac has joined
1359 2013-02-01 16:25:24 <gmaxwell> HM: so, have a party provide a key to you, and then tell them which address to pay to make that key authorized to recieve access.
1360 2013-02-01 16:25:54 <Yrouel> like this part
1361 2013-02-01 16:25:56 <Yrouel> *
1362 2013-02-01 16:25:58 <Yrouel> ops
1363 2013-02-01 16:26:16 <Yrouel> *
1364 2013-02-01 16:26:20 <Yrouel> mh
1365 2013-02-01 16:26:46 <HM> gmaxwell: a key for what?
1366 2013-02-01 16:27:24 <Scrat> you can also run it in a VM with no internet access, add gmail smtp 127.0.0.1 to hosts and open up a tcp listening socket, let it connect and see what it's sending
1367 2013-02-01 16:27:29 <gmaxwell> You achieve nothing by binding it to the payment except by making it impossible to uniformly suppot reoccuring payments, preveting pre-payment identification, complicating third party payments, or the use of indirect payment mechenisms (e.g. web wallets)
1368 2013-02-01 16:27:38 <gmaxwell> HM: a key for authentication.
1369 2013-02-01 16:28:14 <gmaxwell> And if you're worried about the provider cheating and saying you haven't paid— thats what the payment protocol ('invoices') stuff solves.
1370 2013-02-01 16:28:30 <HM> when you buy a padlock you buy a lock and a key to go with it. the person who bought that lock and key, and installed it, should have the right to open the damn lock, not the person who found the key in the street. so if you can prove you bought and installed the lock, that should be sufficient
1371 2013-02-01 16:28:47 <HM> and with bitcoin, the fact that you bought the lock is public to all
1372 2013-02-01 16:29:24 <gmaxwell> HM: the meaning of any particlar transaction is only known to it's participants. (thank god, or bitcoin would be so unprivate it would be worthless)
1373 2013-02-01 16:29:46 eckey has quit (Ping timeout: 276 seconds)
1374 2013-02-01 16:30:18 <HM> that's not entirely true. if one person says "we accept membership of 20 BTC to address X" then anyone who sees such payments can say with certainty that the person who sent the funds was buying membership
1375 2013-02-01 16:30:35 <gmaxwell> (The invoices stuff works by— the provider gives you a signed invoice that specifis what address to pay... if you pay and they refuse to provide service you can show others)
1376 2013-02-01 16:31:16 <gmaxwell> ugh. You are suffering from the confusion that bitcoin transactions have a 'from'. They do not.
1377 2013-02-01 16:31:28 <gmaxwell> You can't recieve payments at a single address and know who is paying you.
1378 2013-02-01 16:31:29 <Luke-Jr> gmaxwell: by "address", you mean scriptPubKey I hope? <.<
1379 2013-02-01 16:31:55 <HM> signing key
1380 2013-02-01 16:32:10 <HM> standard transactions
1381 2013-02-01 16:33:16 <Luke-Jr> gmaxwell: this 'from' nonsense makes me want to implement the p2p let's-all-use-a-single-transaction thing
1382 2013-02-01 16:34:29 <kjj> maybe we could put a sticky on the forums.  people read those, right?
1383 2013-02-01 16:34:44 <HM> it doesn't matter what the protocol says, to thousands of users you have transactions from A to B. because unspent inputs "belong" in someones wallet according to standard scripts
1384 2013-02-01 16:34:46 <CodeShark> bitcoin doesn't care the slightest bit who you really are - all it cares about is whether or not your claim of a particular output is cryptographically valid
1385 2013-02-01 16:34:50 tonikt has quit (Read error: Connection reset by peer)
1386 2013-02-01 16:34:51 <Luke-Jr> HM: and when I pay with newly minted coins that have no signature?
1387 2013-02-01 16:35:02 <Luke-Jr> HM: or from MtGox or any other bank-like wallet?
1388 2013-02-01 16:35:07 <gmaxwell> HM: https://blockchain.info/tx-index/47421492/14947302eab0608fb2650a05f13f6f30b27a0a314c41250000f77ed904475dbb  so you now believe I've held 50kBTC?
1389 2013-02-01 16:35:24 davout has quit (Remote host closed the connection)
1390 2013-02-01 16:35:32 <gmaxwell> HM: it's very very common for the prior output to be owned by another party.
1391 2013-02-01 16:35:42 <HM> that page clearly says you put in 1 BTC
1392 2013-02-01 16:35:47 <HM> i don't see the problem
1393 2013-02-01 16:36:05 <HM> you're talking about network and protocol semantics, i'm talking about what users care about making payments
1394 2013-02-01 16:36:08 <Luke-Jr> HM: just because a signature is valid for Output-A-With-PubKey-B does NOT mean it is valid for Output-C-With-PubKey-B
1395 2013-02-01 16:36:15 <gmaxwell> HM: who is that transaction from?
1396 2013-02-01 16:36:57 <HM> right i see your point. if you have many inputs it's problematic
1397 2013-02-01 16:37:03 b4epoche has quit (Ping timeout: 255 seconds)
1398 2013-02-01 16:37:22 <HM> maybe
1399 2013-02-01 16:37:25 <Luke-Jr> HM: even with only 1 input, you don't know who it's from
1400 2013-02-01 16:37:30 <CodeShark> so who is your acquaintance with 40k bitcoind, gmaxwell? :)
1401 2013-02-01 16:37:32 <gmaxwell> it's always problematic.
1402 2013-02-01 16:37:50 <gmaxwell> CodeShark: https://bitcointalk.org/index.php?topic=139581.0
1403 2013-02-01 16:37:54 <kjj> it is problematic because you are mixing metaphors.  transactions come from other transactions, not people and not keys
1404 2013-02-01 16:38:06 meLon has quit (Remote host closed the connection)
1405 2013-02-01 16:38:24 <Luke-Jr> kjj: not even necessarily that
1406 2013-02-01 16:38:38 <gmaxwell> Someone please announce this txn for me: http://pastebin.com/FKdGC5iM
1407 2013-02-01 16:38:38 <kjj> yeah, the generate is the one exception to that
1408 2013-02-01 16:38:43 <Luke-Jr> kjj: just because transaction A is used as input does NOT mean it's inherently related to all the outputs
1409 2013-02-01 16:39:09 <Luke-Jr> gmaxwell: done
1410 2013-02-01 16:39:10 <gmaxwell> Luke-Jr: for the purpose of tracking coins in the system it is, but not for the purpose of determining ownership of anything.
1411 2013-02-01 16:39:14 <gmaxwell> Luke-Jr: thanks.
1412 2013-02-01 16:39:28 <gmaxwell> what was the txid (if you dunno I can figure it out)
1413 2013-02-01 16:39:40 <kjj> a511bea3b5dc09609c4853d817cde909fdcdc06cc9558500f155ca821d0d511b
1414 2013-02-01 16:39:47 meLon has joined
1415 2013-02-01 16:39:49 <Luke-Jr> gmaxwell: a511bea3b5dc09609c4853d817cde909fdcdc06cc9558500f155ca821d0d511b
1416 2013-02-01 16:39:53 <Luke-Jr> bbiab
1417 2013-02-01 16:40:18 b4epoche has joined
1418 2013-02-01 16:40:21 <HM> kjj: if i give someone my public wallet "address" and then tell them i am going to send them 20 BTC they can verify that
1419 2013-02-01 16:40:25 <HM> if someone else is watching this promise
1420 2013-02-01 16:40:28 <HM> they can also verify that
1421 2013-02-01 16:40:38 <HM> oopts, gmaxwel, no kjj
1422 2013-02-01 16:40:44 <kjj> no, they can verify that SOMEONE sent to it
1423 2013-02-01 16:41:20 <gmaxwell> HM: sure, but you can do it even better by making the promise instead be on the reciving address side.
1424 2013-02-01 16:41:51 <gmaxwell> And then to escape all the gonzo inferrance of 'from' and artifically limiting how transactions can be constructed to keep things compatible with the broken 'from' oncept.
1425 2013-02-01 16:42:17 <CodeShark> I'm gonna have to explicitly filter out 1GMaxweLLbo8mdXvnnC19Wt2wigiYUKgEB from my txinset database, gmaxwell :)
1426 2013-02-01 16:42:22 <HM> people like the 'from' concept
1427 2013-02-01 16:42:36 <HM> they like to feel they "own" the coins they have in their "wallet"
1428 2013-02-01 16:42:39 t7` has joined
1429 2013-02-01 16:42:55 <sipa> HM: and payments have a 'from', bitcoin transactions don't
1430 2013-02-01 16:42:55 <kjj> too bad.  no matter how much people like "from", it still does not exist in any meaningful way in bitcoin
1431 2013-02-01 16:43:02 <sipa> HM: that's why we need a payment protocol
1432 2013-02-01 16:43:05 <gmaxwell> HM: they do own the coins in their wallet... but that doesn't imply that transactions contain a from.
1433 2013-02-01 16:43:17 t7` has quit (Read error: Connection reset by peer)
1434 2013-02-01 16:43:22 <HM> you can extrapolate who owns what by looking at standard scripts, that's how we determine wallet balance after all
1435 2013-02-01 16:43:26 <HM> it's a matter of statefulness
1436 2013-02-01 16:44:22 <HM> i do see your point though
1437 2013-02-01 16:44:24 <gmaxwell> HM: in some cases, in some cases not. A significant fraction of transactions have inputs whos prior to address (thats your inferrance from scripts) tells you nothing about who the payer of a transaction is.
1438 2013-02-01 16:44:46 <kjj> meh.  the balance your node displays is just the sum of the unspent TX that it is capable of signing
1439 2013-02-01 16:45:32 t7 has quit (Ping timeout: 244 seconds)
1440 2013-02-01 16:45:40 <kjj> that is as close to ownership as it gets.  if someone else also has those keys, they can spend them too
1441 2013-02-01 16:45:46 <HM> yes and those TXs are identified by a standard form.
1442 2013-02-01 16:46:09 <gmaxwell> E.g. sipa recieves coin at AAAA  I recieve coin at BBBB.  At some point sipa recieves 1 BTC at AAAA and I recieve 2 BTC at BBBB.  Later you get a payment of 1 BTC of coins whos prior destination address was AAAA.  Who paid you?
1443 2013-02-01 16:46:14 <HM> the script determines how a tx can be spent, if it can only be spent by someone who has a "wallet" privkey then it's "in" their wallet
1444 2013-02-01 16:46:23 <HM> that's the metaphor we have adopted for most use atm
1445 2013-02-01 16:46:57 <HM> if you don't like the metaphor then why did you adopt it ;)
1446 2013-02-01 16:47:12 <kjj> you ever hear of metaphor shear?
1447 2013-02-01 16:47:26 <gmaxwell> HM: kjj's explination is correct. The balance just shows what it's able to spend.
1448 2013-02-01 16:47:35 <gmaxwell> HM: going to answer my hypothetical?
1449 2013-02-01 16:48:02 <HM> hold on a sec
1450 2013-02-01 16:48:41 <HM> if sipa receives 1 BTC at AAAA, and you receive coins who's prior destination was AAAA, then you haven't involved BBBB at all
1451 2013-02-01 16:48:45 <HM> did you write that correctly?
1452 2013-02-01 16:48:57 <gmaxwell> Yes. I did.
1453 2013-02-01 16:49:03 <HM> so what does BBBB have to do with it
1454 2013-02-01 16:49:07 <gmaxwell> So who paid you?
1455 2013-02-01 16:49:24 <HM> Sipa did, assuming the script is of a standard form
1456 2013-02-01 16:49:31 <gmaxwell> No— I (the person who recieves funds at BBBB) paid you. But in this example sipa and I use a webwallet with a pooled coin store, and the coin previously paid to sipa was a better fit.
1457 2013-02-01 16:49:51 <gmaxwell> and this is all with standard transactions on widely used webwallets.
1458 2013-02-01 16:50:26 <HM> wallets can be identified not by transactions but by the requirements on their spending. if transactions in to your webwallet share a common set of requirements on spending them then they exist "in" your wallet
1459 2013-02-01 16:50:56 <gmaxwell> Those requirements aren't necessarily visible in the blockchain.
1460 2013-02-01 16:51:38 <HM> and if they aren't standard script your wallet client wont' recognise them either. right?
1461 2013-02-01 16:51:54 <HM> because your "wallet client" only recognises certain transactions as spendable
1462 2013-02-01 16:52:34 <gmaxwell> The inputs are irrelevant. The reference wallet reconizes transactions that it knows how to spend.
1463 2013-02-01 16:52:40 <HM> sure, we kinda need every node to be able to evaluate all script and "from" is problematic, i agree
1464 2013-02-01 16:53:36 <HM> but the script opcodes and standard form sort of establish a standard for the wallet metaphor
1465 2013-02-01 16:54:04 <gmaxwell> I can't tell what you mean.
1466 2013-02-01 16:54:05 <sipa> HM: it is the wallet that adds the wallet metaphor
1467 2013-02-01 16:54:19 <sipa> things that work at the protocol level should not assume a wallet metaphor exists on top of it
1468 2013-02-01 16:54:48 <sipa> that is why you need a payment protocol: it establishes the notion of a payment from an identity to an identify
1469 2013-02-01 16:55:03 <sipa> instead of just the reassignment of authentication keys for coins
1470 2013-02-01 16:55:43 meLon has quit (Remote host closed the connection)
1471 2013-02-01 16:56:20 <pisto> sipa, I don't really care about orphaned blocks, so I can read out the .dat files directly. some questions: should I just concatenate all the blk*.dat files? you mentioned a signature and a length: is it a single siagnature in the beginning, then a block length for each following block?
1472 2013-02-01 16:56:39 <HM> right well perhaps wallet ownership is irrelevant and too vague, but you can certainly devise a transaction that you can later prove was made by you through cryptographic means.
1473 2013-02-01 16:56:58 <sipa> HM: which is exactly what the payment protocol does
1474 2013-02-01 16:58:35 <HM> I'm lost in metaphors
1475 2013-02-01 16:58:44 <HM> Marginally better than being lost in Javascript
1476 2013-02-01 16:59:20 bitnumus_ has joined
1477 2013-02-01 16:59:33 <kjj> HM: that is only true if you have control of the keys
1478 2013-02-01 16:59:42 <CodeShark> setTimeout(wakeup, 1000000000);
1479 2013-02-01 16:59:53 da2ce7_d has joined
1480 2013-02-01 16:59:54 <HM> my original proposition was that you can create a standard for payment and membership negotiation that uses the blockchain to prove someone paid for membership and has the right to access a service.
1481 2013-02-01 16:59:56 <kjj> in the case of a web wallet, you don't have the keys, and you can't prove any transactions in general
1482 2013-02-01 17:00:02 Diapolo has left ()
1483 2013-02-01 17:00:09 twixed has joined
1484 2013-02-01 17:00:27 <HM> I never said it was useful.
1485 2013-02-01 17:01:10 <kjj> have the service provider create a message indicating the service and the credentials and the payment address, and then sign it with GPG.
1486 2013-02-01 17:02:11 Zarutian has joined
1487 2013-02-01 17:02:14 da2ce7 has quit (Ping timeout: 252 seconds)
1488 2013-02-01 17:02:18 <kjj> in a disupute the payor can publish that message and the whole world will be able to see that it was paid (but not by whom), and it will include the service info
1489 2013-02-01 17:02:32 <helo> HM: signmessage would be nice for that, but it's not easy to know what address was used to sign a transaction one has sent
1490 2013-02-01 17:03:53 <HM> there are some use cases i can think of. if you subscribe to A, then B may offer you a discount. but A might not be willing to prove to B than you do in fact subscribe, and you might not want to give B your credentials for A
1491 2013-02-01 17:04:07 <kjj> HM: cookie
1492 2013-02-01 17:04:53 <HM> that's basically what browserid does. Mozilla Persona have no idea who you are proving yourself to
1493 2013-02-01 17:04:58 <kjj> but in general, bitcoin can't help you with these schemes.
1494 2013-02-01 17:05:09 <gmaxwell> HM: thats what what kjj described can allow.
1495 2013-02-01 17:05:30 <kjj> ok, that's like the 4th time I've seen that word go past today.  I'd better go find out what browserid is
1496 2013-02-01 17:06:25 <gmaxwell> kjj: it's just openid done right (e.g. doesn't enable a service provider to spy on you)
1497 2013-02-01 17:06:56 <HM> openid means your openid provider sees everything you're doing
1498 2013-02-01 17:07:02 <kjj> I don't mind my openid provider spying on me, since I run it on my own server
1499 2013-02-01 17:07:21 <kjj> and generally speaking, I already know where I'm using it
1500 2013-02-01 17:07:42 <HM> kjj: right but openid can't prove email address ownership unless the service you're using trusts your openid provider to do so
1501 2013-02-01 17:07:46 <kjj> of course, most openid sites only work if your provider is either google or facebook...
1502 2013-02-01 17:07:59 <gmaxwell> kjj: I suppose you don't use many site then... I have my own openid too, but it almost never works with anything.
1503 2013-02-01 17:08:07 <gmaxwell> right.
1504 2013-02-01 17:08:33 <kjj> for the most part, I just use xyzzy to generate insanely long passwords for each site
1505 2013-02-01 17:08:56 meLon has joined
1506 2013-02-01 17:08:56 meLon has quit (Changing host)
1507 2013-02-01 17:08:56 meLon has joined
1508 2013-02-01 17:09:17 <HM> yeah, and it sucks arse without a browser integrated password store
1509 2013-02-01 17:09:27 <CodeShark> I use my own cipher that encrypts URLs to passwords
1510 2013-02-01 17:09:27 valparaiso is now known as valparaiso_afk
1511 2013-02-01 17:09:32 <CodeShark> which I can do in my head
1512 2013-02-01 17:09:39 eckey has joined
1513 2013-02-01 17:09:57 <kjj> heh.  if you can do it in your head, then every site you use can decrypt it using their own URL
1514 2013-02-01 17:10:01 <pjorrit> and then the url changes ;p
1515 2013-02-01 17:10:03 <HM> CodeShark: you replace http with hteapeepee don't you :P
1516 2013-02-01 17:10:15 <CodeShark> I add some entropy to it, kjj :)
1517 2013-02-01 17:10:22 <CodeShark> random numbers I've memorized :)
1518 2013-02-01 17:10:28 <kjj> HM: I store them in my browser too, i they aren't important
1519 2013-02-01 17:11:07 <CodeShark> lol HM
1520 2013-02-01 17:11:15 <CodeShark> actually, aichteapeepee
1521 2013-02-01 17:11:30 <CodeShark> aichteeteepee
1522 2013-02-01 17:11:37 <HM> oh yeah
1523 2013-02-01 17:11:43 <HM> 2 t's, 1 p
1524 2013-02-01 17:11:59 <HM> that was just to throw you off my scent >_>
1525 2013-02-01 17:12:46 <kjj> interesting.  I'm curious exactly how persona verifies email addresses
1526 2013-02-01 17:13:31 <HM> they send you an email
1527 2013-02-01 17:13:33 <HM> you click a link
1528 2013-02-01 17:13:36 <gmaxwell> kjj: standard challenge response AFAIR.
1529 2013-02-01 17:13:50 <kjj> who challenges?
1530 2013-02-01 17:13:55 <kjj> each site?
1531 2013-02-01 17:15:50 <eckey> 	hello. I have a requirement to sort several bc transactions by arrival time. Is a tx timestamped when it first arrives at a node?
1532 2013-02-01 17:15:57 <HM> kjj: the idea is email vendors provide public keys that authentication services trust, directly or through a chain of authenticating parties
1533 2013-02-01 17:16:20 <kjj> eckey: yes, but you probably shouldn't be doing that
1534 2013-02-01 17:16:25 <HM> then it's challenge response between the service you want to log in to and your email vendor.
1535 2013-02-01 17:16:41 <kjj> my email vendor is also me
1536 2013-02-01 17:17:04 <eckey> kjj: please explain. thx
1537 2013-02-01 17:17:30 <HM> yes, which is why it doesn't really prove that email addresses are receivable / usable or anything
1538 2013-02-01 17:17:47 <kjj> eckey: when a transaction comes in as a transaction (not in a block) it gets a timestamp when it is added to the mempool
1539 2013-02-01 17:18:11 <eckey> does that ts move to the block chain?
1540 2013-02-01 17:18:18 <kjj> but that timestamp is not useful, because transactions happen in a particular order that has nothing at all to do with when you see them
1541 2013-02-01 17:18:27 <kjj> no
1542 2013-02-01 17:19:04 <kjj> when a transaction is included in a block, they all assume the block's timestamp, which has only a rough correlation to any actual time
1543 2013-02-01 17:19:49 <kjj> so, a transaction timestamp is purely local, and not real because the transaction hasn't REALLY happened yet
1544 2013-02-01 17:19:52 <eckey> and it the tx didn't include the fee, it could have been a long while since it first entered the pool...
1545 2013-02-01 17:20:07 <CodeShark> absolute time has no real meaning in bitcoin - the only thing that must be agreed upon is the order
1546 2013-02-01 17:20:35 <eckey> and the order is defined by the block sequence.
1547 2013-02-01 17:20:41 <CodeShark> yup
1548 2013-02-01 17:21:22 <CodeShark> even block times cannot be generally known precisely
1549 2013-02-01 17:21:42 <eckey> if my client has a certain cut-off time to send his bitcoins, how can I verify that?
1550 2013-02-01 17:22:03 <kjj> erm, probably not, but maybe
1551 2013-02-01 17:22:35 <kjj> you could watch incoming transactions, and record when you first see it, and mark it, in your database, as having come in before the deadline
1552 2013-02-01 17:22:46 <CodeShark> you can check to see when you yourself saw the transaction. most of the time your node will receive it within seconds if you are connected to enough nodes
1553 2013-02-01 17:22:47 <kjj> but it could be hours or days or never before it shows up in a block
1554 2013-02-01 17:23:05 <eckey> correct, that doesn't matter
1555 2013-02-01 17:23:05 <CodeShark> unless it was conflicted
1556 2013-02-01 17:23:07 davout has joined
1557 2013-02-01 17:23:07 davout has quit (Changing host)
1558 2013-02-01 17:23:07 davout has joined
1559 2013-02-01 17:23:15 davout has quit (Remote host closed the connection)
1560 2013-02-01 17:24:02 <eckey> would I need to be a "node" with incoming connections to see incoming transactions?
1561 2013-02-01 17:24:08 <CodeShark> no
1562 2013-02-01 17:24:10 <kjj> pay attention that the transaction could also show up in a block, without you having seen it first
1563 2013-02-01 17:24:27 <kjj> you'd need to be a node, but you don't need incoming connections
1564 2013-02-01 17:24:44 <CodeShark> whenever nodes you're connected to (regardless of who initiated the connection) see a new transaction they accept into their mempool, they will tell you about it
1565 2013-02-01 17:25:22 <CodeShark> or at least that's how bitcoind behaves
1566 2013-02-01 17:25:45 <eckey> ok, that's why I can see my grayed-out transaction in Bitcoin-Qt almost immediately...
1567 2013-02-01 17:26:23 <eckey> (send from another wallet)
1568 2013-02-01 17:26:28 <eckey> *sent
1569 2013-02-01 17:32:13 andytoshi has joined
1570 2013-02-01 17:35:54 owowo has joined
1571 2013-02-01 17:36:35 tonikt has joined
1572 2013-02-01 17:43:19 bsdunx has quit (Read error: Connection reset by peer)
1573 2013-02-01 17:43:38 <jgarzik> hrm
1574 2013-02-01 17:43:44 <jgarzik> is Avalon batch #2 sale open?
1575 2013-02-01 17:43:46 owowo has quit (Quit: sayonara)
1576 2013-02-01 17:44:38 <abracadabra> no
1577 2013-02-01 17:44:42 rdymac has quit (Quit: This computer has gone to sleep)
1578 2013-02-01 17:44:52 Dyaheon- has quit ()
1579 2013-02-01 17:44:54 <abracadabra> buying more jgarzik?
1580 2013-02-01 17:44:55 <abracadabra> ;)
1581 2013-02-01 17:46:54 bsdunx has joined
1582 2013-02-01 17:47:00 <Yrouel> btw I think I've fount those credentials but unfortunately he password has been changed
1583 2013-02-01 17:47:20 <Yrouel> (at least is what google says when I try to log in)
1584 2013-02-01 17:47:56 drizzt_ has quit (Remote host closed the connection)
1585 2013-02-01 17:48:28 bitnumus has quit (Quit: Bye)
1586 2013-02-01 17:49:26 BTCOxygen has joined
1587 2013-02-01 17:49:36 altamic has quit (Quit: altamic)
1588 2013-02-01 17:50:55 owowo has joined
1589 2013-02-01 17:52:54 JDuke128 has joined
1590 2013-02-01 17:52:57 rdymac has joined
1591 2013-02-01 17:53:43 ThomasV_ has joined
1592 2013-02-01 17:54:49 ThomasV has quit (Remote host closed the connection)
1593 2013-02-01 17:54:55 ThomasV_ has quit (Client Quit)
1594 2013-02-01 17:55:20 ThomasV has joined
1595 2013-02-01 17:55:56 grau has joined
1596 2013-02-01 17:56:16 Dyaheon has joined
1597 2013-02-01 17:57:28 t7 has joined
1598 2013-02-01 17:57:47 <bsdunx> Question to core devs, how against CMake are you?
1599 2013-02-01 17:58:10 <bitmarco> cmake *sigh*
1600 2013-02-01 17:58:17 <sipa> not at all, if someone wants to write *and maintain* a cmake config, patches welcome
1601 2013-02-01 17:58:30 <jgarzik> puke
1602 2013-02-01 17:58:41 <CodeShark> scons anyone? :)
1603 2013-02-01 17:59:02 <jgarzik> scons is awesome... in theory
1604 2013-02-01 17:59:18 <jgarzik> theory != widespread practice and deployment and support
1605 2013-02-01 17:59:39 <CodeShark> why did bash have to be such a crappy programming language? :p
1606 2013-02-01 17:59:57 <sipa> CodeShark: because it wasn't (originally) designed a programming language
1607 2013-02-01 18:00:21 <kjj> perl has the same problem
1608 2013-02-01 18:01:10 grau has quit (Remote host closed the connection)
1609 2013-02-01 18:01:18 <CodeShark> I think I'd even take the BASIC interpreter that came with the Apple II over bash :p
1610 2013-02-01 18:01:30 <sipa> CodeShark: depends for what
1611 2013-02-01 18:01:31 <bsdunx> o_O
1612 2013-02-01 18:02:03 <sipa> i think any autoconfig system that is maintained and able to build all our current platform combinations (including Qt, gitian, pulltester, ...), would be a major improvement
1613 2013-02-01 18:02:07 <kjj> bash makes half decent glue, if the parts that you need to stick together aren't too misshapen
1614 2013-02-01 18:03:18 <bsdunx> BM was that you I was speaking to yesterday? If so could you list the requirements for me again WRT unit tests.
1615 2013-02-01 18:05:50 bitnumus has joined
1616 2013-02-01 18:09:23 <bsdunx> has autohell improved much since the late 90's?
1617 2013-02-01 18:10:58 TD has joined
1618 2013-02-01 18:11:16 bitnumus_ has quit (Changing host)
1619 2013-02-01 18:11:16 bitnumus_ has joined
1620 2013-02-01 18:11:16 bitnumus has quit (Changing host)
1621 2013-02-01 18:11:16 bitnumus has joined
1622 2013-02-01 18:13:03 bitnumus_ has quit (Quit: Leaving)
1623 2013-02-01 18:19:49 HM has quit (Remote host closed the connection)
1624 2013-02-01 18:19:50 nus has joined
1625 2013-02-01 18:21:11 <CodeShark> when it works, it seems to work really well. but when it doesn't, there doesn't seem to be much rhyme or reason for it
1626 2013-02-01 18:22:07 rdymac has quit (Quit: This computer has gone to sleep)
1627 2013-02-01 18:23:16 <CodeShark> I'm not a huge fan of script generators generally - I'd much rather have a decent language as part of the standard shell and have modular, well-designed libraries that can be called from it
1628 2013-02-01 18:26:05 <bsdunx> Hmm, I'm looking over the SCons user guide and it doesn't seem too horrible.
1629 2013-02-01 18:26:53 rdymac has joined
1630 2013-02-01 18:27:56 Jamesonwa has joined
1631 2013-02-01 18:28:15 <CodeShark> python is certainly a better thought-out language than bash :p
1632 2013-02-01 18:28:18 <jgarzik> scons is well designed
1633 2013-02-01 18:28:32 <jgarzik> however, ugly autotools is already integrated into multiple OS's and packaging systems
1634 2013-02-01 18:29:19 <CodeShark> but does that really matter on the user end?
1635 2013-02-01 18:29:43 zooko has quit (Ping timeout: 264 seconds)
1636 2013-02-01 18:30:47 dust-otc has quit (Remote host closed the connection)
1637 2013-02-01 18:30:50 <CodeShark> scons can be easily installed on all aforementioned OSes
1638 2013-02-01 18:31:00 zooko has joined
1639 2013-02-01 18:31:27 <midnightmagic> openid *could* be used to prove email ownership if we had vouching and could stamp the openid with a google sig that says, "yes, this guy des have google address blah"
1640 2013-02-01 18:31:50 <sipa> CodeShark: can you trick it into doing a cross-compile using mingw?
1641 2013-02-01 18:31:57 <Diablo-D3> google provides that, but only some openid servers do
1642 2013-02-01 18:32:57 <midnightmagic> Diablo-D3: Cross-provider vouching?
1643 2013-02-01 18:33:11 <CodeShark> good question, sipa - however, if you're not already more an expert in scons than I am, it wouldn't take you long to become one :p
1644 2013-02-01 18:33:22 <Diablo-D3> midnightmagic: learn how openid works =P
1645 2013-02-01 18:33:28 <sipa> CodeShark: i have used it once :p
1646 2013-02-01 18:33:42 <sipa> (used it, not written or even read a config)
1647 2013-02-01 18:33:42 <midnightmagic> Diablo-D3: No, I mean, Google will vouch a non-google openid?
1648 2013-02-01 18:33:52 <Diablo-D3> no?
1649 2013-02-01 18:34:00 <Diablo-D3> google provides an openid auth server
1650 2013-02-01 18:34:13 <Diablo-D3> think of them as something like kerberos domains
1651 2013-02-01 18:34:33 <CodeShark> software packaging isn't my biggest forte
1652 2013-02-01 18:34:42 <CodeShark> which is why I need to do more of it :p
1653 2013-02-01 18:35:56 mappum has joined
1654 2013-02-01 18:36:10 JDuke128 has quit (Ping timeout: 252 seconds)
1655 2013-02-01 18:36:28 <midnightmagic> Yah I know that. I'm talking about people using their own openid but with a vouching mechanism
1656 2013-02-01 18:36:42 <midnightmagic> i don't want random people telling google when I'm using their openid.
1657 2013-02-01 18:36:49 <CodeShark> speaking of which I don't think I'm too far from having the autopuke configure for bitcoind: https://github.com/CodeShark/bitcoin/tree/autotools
1658 2013-02-01 18:37:21 techlife has quit (Ping timeout: 272 seconds)
1659 2013-02-01 18:37:22 <gmaxwell> midnightmagic: better to use browserid which solved that problem from day 1.
1660 2013-02-01 18:38:01 <sipa> by the way: i'll be on vacation coming week, and probably not very internet-connectable
1661 2013-02-01 18:38:03 da2ce7 has joined
1662 2013-02-01 18:38:03 <midnightmagic> gmaxwell: I never liked openid anyway.
1663 2013-02-01 18:38:30 <sipa> so don't wait for me to do an RC :)
1664 2013-02-01 18:38:48 JDuke128 has joined
1665 2013-02-01 18:39:02 rdymac has quit (Remote host closed the connection)
1666 2013-02-01 18:39:44 da2ce744 has quit (Ping timeout: 244 seconds)
1667 2013-02-01 18:40:42 altamic has joined
1668 2013-02-01 18:40:43 altamic has quit (Changing host)
1669 2013-02-01 18:40:43 altamic has joined
1670 2013-02-01 18:43:15 techlife has joined
1671 2013-02-01 18:43:16 techlife has quit (Max SendQ exceeded)
1672 2013-02-01 18:43:18 DingoRabiit has joined
1673 2013-02-01 18:43:23 <DingoRabiit> So is bitcointalk.org down?
1674 2013-02-01 18:44:07 techlife has joined
1675 2013-02-01 18:44:08 techlife has quit (Max SendQ exceeded)
1676 2013-02-01 18:44:42 techlife has joined
1677 2013-02-01 18:44:42 heebo has joined
1678 2013-02-01 18:44:43 techlife has quit (Max SendQ exceeded)
1679 2013-02-01 18:45:29 techlife has joined
1680 2013-02-01 18:45:29 techlife has quit (Max SendQ exceeded)
1681 2013-02-01 18:46:51 techlife has joined
1682 2013-02-01 18:46:52 techlife has quit (Max SendQ exceeded)
1683 2013-02-01 18:47:44 techlife has joined
1684 2013-02-01 18:47:45 techlife has quit (Max SendQ exceeded)
1685 2013-02-01 18:47:56 altamic has quit (Ping timeout: 245 seconds)
1686 2013-02-01 18:48:17 techlife has joined
1687 2013-02-01 18:48:18 techlife has quit (Max SendQ exceeded)
1688 2013-02-01 18:48:52 techlife has joined
1689 2013-02-01 18:48:53 techlife has quit (Max SendQ exceeded)
1690 2013-02-01 18:49:36 toffoo has joined
1691 2013-02-01 18:49:54 techlife has joined
1692 2013-02-01 18:49:55 techlife has quit (Max SendQ exceeded)
1693 2013-02-01 18:50:13 rdymac has joined
1694 2013-02-01 18:50:57 techlife has joined
1695 2013-02-01 19:00:39 twixed has quit (Read error: Connection reset by peer)
1696 2013-02-01 19:00:47 rdymac has quit (Remote host closed the connection)
1697 2013-02-01 19:01:15 twixed has joined
1698 2013-02-01 19:01:49 bakh has joined
1699 2013-02-01 19:05:56 rdponticelli has quit (Quit: No Ping reply in 180 seconds.)
1700 2013-02-01 19:06:15 rdymac has joined
1701 2013-02-01 19:08:24 rdponticelli has joined
1702 2013-02-01 19:08:49 Grishnakh has joined
1703 2013-02-01 19:09:52 rdymac has quit (Remote host closed the connection)
1704 2013-02-01 19:10:56 Joric has joined
1705 2013-02-01 19:12:19 CodeShark has quit (Remote host closed the connection)
1706 2013-02-01 19:15:28 eckey has quit (Remote host closed the connection)
1707 2013-02-01 19:16:10 DingoRabiit has left ()
1708 2013-02-01 19:16:24 HM has joined
1709 2013-02-01 19:17:15 <HM> it sucks having a weak GPU
1710 2013-02-01 19:17:26 <HM> faster than 1 CPU core but slower than 2 in vanitygen :|
1711 2013-02-01 19:17:53 <Eliel> but faster if you use both CPU and GPU :P
1712 2013-02-01 19:18:00 freakazoid has joined
1713 2013-02-01 19:18:17 <Joric> buy a vanity address on vanitypool it's safe
1714 2013-02-01 19:19:10 <HM> Eliel: yeah, but using GPU seems to hang X entirely.
1715 2013-02-01 19:19:40 <Eliel> yeah, vanitypool is a simpler way to get them.
1716 2013-02-01 19:19:56 <Luke-Jr> hmm, I wonder if it would make sense to try to add vanitypool support to BFGMiner :p
1717 2013-02-01 19:20:05 <Joric> gpu mining is about to extinct anyway
1718 2013-02-01 19:20:17 <Luke-Jr> Joric: BFGMiner has always been FPGA+ focussed
1719 2013-02-01 19:20:50 <HM> how is vanity address generation safe? :|
1720 2013-02-01 19:21:05 <Joric> HM, split key cryptography
1721 2013-02-01 19:21:13 <gmaxwell> Luke-Jr: it would be— what would be most interesting is to have a way to set your electricity price and usage.. and have it run the most profitable jobs (including the idle job) at any time.
1722 2013-02-01 19:21:16 <Luke-Jr> HM: I think so
1723 2013-02-01 19:21:24 root2 has joined
1724 2013-02-01 19:21:26 <Luke-Jr> HM: whether Joric's implementation is or not, is over my head
1725 2013-02-01 19:21:45 <Luke-Jr> gmaxwell: indeed
1726 2013-02-01 19:22:49 <Luke-Jr> 2013-02-01 19:04:05,857 JSONRPCServer   INFO    Longpoll woke up 12459 clients in 15.010 seconds
1727 2013-02-01 19:22:57 <Luke-Jr> ^ anyone have a botnet slayer that won't get us DDoS'd? :/
1728 2013-02-01 19:23:10 <Joric> HM, https://vanitypool.appspot.com/faq This website aims to generate vanity addresses without the need to trust a third party (...)
1729 2013-02-01 19:23:32 <HM> except the trust party telling me i don't need to trust him :P
1730 2013-02-01 19:23:38 <HM> third party*
1731 2013-02-01 19:24:29 <Luke-Jr> Joric: hm, this pool doesn't allow mining multiple addresses at once? :o
1732 2013-02-01 19:24:59 <HM> oh modulo magic
1733 2013-02-01 19:25:00 <Luke-Jr> I thought there was a way to vanity-mine unlimited addresses concurrently
1734 2013-02-01 19:26:52 <HM> Joric: ta, read post on bitcointalk that explains it succintly.
1735 2013-02-01 19:27:19 <Joric> someone ordered 1Casascius for 3.2 btc gotta try to crack it )
1736 2013-02-01 19:28:28 nym_ is now known as nym
1737 2013-02-01 19:30:26 Joric has quit ()
1738 2013-02-01 19:30:29 JZavala has joined
1739 2013-02-01 19:34:35 grau has joined
1740 2013-02-01 19:34:38 <nym> are there any other uses for double sha256 hashing?
1741 2013-02-01 19:35:09 <HM> not many, unless it can be modified to do single hashes
1742 2013-02-01 19:35:30 <HM> or PBKDF2 :)
1743 2013-02-01 19:37:11 zooko has quit (Read error: Operation timed out)
1744 2013-02-01 19:40:26 root2 has quit (Quit: Leaving)
1745 2013-02-01 19:42:11 andytoshi has quit (Quit: WeeChat 0.3.9.2)
1746 2013-02-01 19:42:42 root2 has joined
1747 2013-02-01 19:44:40 apexseals has quit ()
1748 2013-02-01 19:45:34 Jamesonwa has quit (Remote host closed the connection)
1749 2013-02-01 19:47:20 Jamesonwa has joined
1750 2013-02-01 19:47:42 Jamesonwa has quit (Remote host closed the connection)
1751 2013-02-01 19:49:27 mikeyve is now known as Mikej0h
1752 2013-02-01 19:50:05 Mikej0h has quit (Changing host)
1753 2013-02-01 19:50:05 Mikej0h has joined
1754 2013-02-01 19:51:31 D34TH has joined
1755 2013-02-01 19:53:07 ShaTwo has joined
1756 2013-02-01 19:54:12 darkskiez has quit (Ping timeout: 245 seconds)
1757 2013-02-01 19:56:51 darkskiez has joined
1758 2013-02-01 19:59:02 yellowhat has joined
1759 2013-02-01 20:00:48 zooko has joined
1760 2013-02-01 20:00:52 Hasimir has quit (Read error: Operation timed out)
1761 2013-02-01 20:03:29 qwebirc9076 has joined
1762 2013-02-01 20:03:43 <midnightmagic> Luke-Jr: There is. And apparently you can even mine vanity addresses for other people.
1763 2013-02-01 20:04:14 <Luke-Jr> midnightmagic: well, I meant in a pool setup yes
1764 2013-02-01 20:04:23 Hasimir has joined
1765 2013-02-01 20:04:38 Hasimir is now known as Guest66276
1766 2013-02-01 20:08:31 Diapolo has joined
1767 2013-02-01 20:09:25 <Eliel> no-one appears to have replied to the person who asks about integrating MinorFS2 support into bitcoin. That sounds like a good idea to me.
1768 2013-02-01 20:09:36 valparaiso_afk is now known as valparaiso
1769 2013-02-01 20:10:27 <Eliel> (-dev mailing list message)
1770 2013-02-01 20:11:00 twixed has quit (Quit: Leaving)
1771 2013-02-01 20:11:49 twixed has joined
1772 2013-02-01 20:11:50 FredEE has joined
1773 2013-02-01 20:11:55 <Eliel> http://minorfs.polacanthus.net/wiki/Goal_and_use_cases
1774 2013-02-01 20:12:05 <HM> anyone here know why i don't have "ec.h" in openssl-devel?
1775 2013-02-01 20:12:14 <HM> this is a Amazon AWS linux
1776 2013-02-01 20:12:19 EasyAt_ is now known as EasyAt
1777 2013-02-01 20:12:22 <HM> i'm not used to rpm distros
1778 2013-02-01 20:12:32 EasyAt has quit (Changing host)
1779 2013-02-01 20:12:32 EasyAt has joined
1780 2013-02-01 20:12:45 <kjj> CentOS ships an OpenSSL that doesn't include EC math, making it useless to bitcoin
1781 2013-02-01 20:13:01 <HM> brilliant.. :|
1782 2013-02-01 20:13:09 <HM> is there a quick solution ?
1783 2013-02-01 20:13:27 andytoshi has joined
1784 2013-02-01 20:13:41 <sipa> on centos/redhat/fedora you need to compile openssl yourself
1785 2013-02-01 20:14:12 <kjj> as far as I know, my script stil works.
1786 2013-02-01 20:14:22 topace has quit (Read error: No route to host)
1787 2013-02-01 20:14:23 <kjj> but I haven't tried it recently
1788 2013-02-01 20:14:25 <kjj> https://bitcointalk.org/index.php?topic=65818.0;all
1789 2013-02-01 20:14:33 <andytoshi> gmaxwell has openssl rpm's on his site
1790 2013-02-01 20:14:58 <HM> andy, not familiar with his site, got a url?
1791 2013-02-01 20:15:20 <andytoshi> one moment..
1792 2013-02-01 20:15:30 <mappum> will the amount of reorgs go down in the future? i remember someone saying eventually everything will just use 2 confirms
1793 2013-02-01 20:15:48 <andytoshi> is anyone here familiar with handling x509 certs on fedora?
1794 2013-02-01 20:16:04 <kjj> mappum: seems unlikely.  the reorg rate has to do with network latency, which can't be reduced much
1795 2013-02-01 20:17:06 toffoo has quit ()
1796 2013-02-01 20:17:44 <andytoshi> can't find link, will  check my irc logs..
1797 2013-02-01 20:17:49 Jamesonwa has joined
1798 2013-02-01 20:18:14 <andytoshi> http://people.xiph.org/~greg/openssl/
1799 2013-02-01 20:19:06 Jamesonwa has quit (Remote host closed the connection)
1800 2013-02-01 20:19:27 <HM> hmm
1801 2013-02-01 20:19:39 <HM> what are the build requirements for openssl, is it a pain to build
1802 2013-02-01 20:20:05 saispe has joined
1803 2013-02-01 20:20:06 <Scrat> MH: nothing
1804 2013-02-01 20:20:08 <andytoshi> HM: very few, i build it myself no problem
1805 2013-02-01 20:20:14 Goonie has quit (Ping timeout: 248 seconds)
1806 2013-02-01 20:20:15 <andytoshi> takes half an hour or so though
1807 2013-02-01 20:20:30 Guest50360 is now known as Mad7Scientist
1808 2013-02-01 20:20:47 <andytoshi> gmaxwell mentioned that you want to add -DPURIFY or something like that, because otherwise openssl uses unallocated memory in its entropy pool
1809 2013-02-01 20:20:49 <Scrat> yeah if you run make test :p
1810 2013-02-01 20:20:53 <andytoshi> this doesn't affect security, but screws up valgrind badly
1811 2013-02-01 20:21:25 Goonie has joined
1812 2013-02-01 20:21:28 <gavinandresen> BDB needs a tweak to be valgrind-friendly, too
1813 2013-02-01 20:22:04 PK has quit ()
1814 2013-02-01 20:22:48 <Luke-Jr> andytoshi: I'd think it'd be better to use a valgrind suppression IMO
1815 2013-02-01 20:23:34 <andytoshi> haven't actually tried it luke, not sure if the errors are simple enough to be suppressed
1816 2013-02-01 20:23:55 andytoshi has quit (Quit: trying to configure my irc client, will be off and on)
1817 2013-02-01 20:25:30 <TD> argh
1818 2013-02-01 20:25:44 <TD> i have a horrible sense of deja vu. do NOT fuck with openssl to make valgrind happy.
1819 2013-02-01 20:25:51 <TD> oh, he's gone
1820 2013-02-01 20:26:07 BTCOxygen has quit (Ping timeout: 264 seconds)
1821 2013-02-01 20:26:18 <gavinandresen> TD: really?  I run with a -DPURIFY openssl and haven't had any problems...
1822 2013-02-01 20:27:10 <sipa> i don't run with a -DPURIFY openssl (unless ubuntu's default is), and haven't had any problems
1823 2013-02-01 20:27:20 <sipa> with valgrind
1824 2013-02-01 20:27:29 <TD> suppressions are better. i just dislike seeing people talk about that, given the debian disaster
1825 2013-02-01 20:27:34 <TD> the potential for messes is large
1826 2013-02-01 20:27:40 rdponticelli has quit (Ping timeout: 276 seconds)
1827 2013-02-01 20:28:06 rbecker is now known as RBecker
1828 2013-02-01 20:29:55 <Luke-Jr> gavinandresen: for a number of years, I was using a password generator that basically had 16 bits of entropy, and never had any problems..
1829 2013-02-01 20:30:00 <Luke-Jr> doesn't mean it's a good idea ;)
1830 2013-02-01 20:30:40 topace has joined
1831 2013-02-01 20:31:04 topace is now known as Guest7347
1832 2013-02-01 20:31:47 qwebirc9076 has quit (Quit: Page closed)
1833 2013-02-01 20:32:54 tonikt has quit (Read error: Connection reset by peer)
1834 2013-02-01 20:34:43 CodeShark has joined
1835 2013-02-01 20:38:15 <Luke-Jr> ok, Eligius is now running every block it considers making through master, and logging any failures it (hopefully won't) hits
1836 2013-02-01 20:38:33 JDuke128 has quit (Quit: ["Textual IRC Client: www.textualapp.com"])
1837 2013-02-01 20:43:27 rdponticelli has joined
1838 2013-02-01 20:48:09 andytoshi has joined
1839 2013-02-01 20:51:56 rng29a has joined
1840 2013-02-01 20:51:56 rng29a has quit (Changing host)
1841 2013-02-01 20:51:56 rng29a has joined
1842 2013-02-01 20:52:32 b4epoche has quit (Ping timeout: 245 seconds)
1843 2013-02-01 20:55:12 b4epoche has joined
1844 2013-02-01 20:58:37 JZavala has quit (Remote host closed the connection)
1845 2013-02-01 20:59:34 JZavala has joined
1846 2013-02-01 21:02:19 <bsdunx> I'm not really familiar with qt, what is the process for updating translations once a string is changed in source?
1847 2013-02-01 21:03:45 <sipa> bsdunx: the translations file is updated every once in a while, and certainly rught before release candidates
1848 2013-02-01 21:03:59 <sipa> that's pushed to transifex (or it updates automatically, dunno)
1849 2013-02-01 21:03:59 <bsdunx> alright
1850 2013-02-01 21:04:04 rdponticelli has quit (Ping timeout: 276 seconds)
1851 2013-02-01 21:04:12 <sipa> where people do the actual transactions to other languages
1852 2013-02-01 21:04:20 <sipa> and those are merged before release
1853 2013-02-01 21:04:57 iddo has quit (Changing host)
1854 2013-02-01 21:04:57 iddo has joined
1855 2013-02-01 21:06:58 RedEmerald has quit (Quit: ZNC - http://znc.in)
1856 2013-02-01 21:07:07 <helo> Luke-Jr: what password generator was that?
1857 2013-02-01 21:07:16 <helo> (to be sure i'm not using that one)
1858 2013-02-01 21:07:36 <HM> nice
1859 2013-02-01 21:07:46 <HM> pushing 17 MH/s on Amazon AWS for vanitygen
1860 2013-02-01 21:08:16 <HM> one GPU :|
1861 2013-02-01 21:08:21 <sipa> Maddr/s you mean?
1862 2013-02-01 21:08:29 <HM> it says Mkeys/s
1863 2013-02-01 21:08:34 <HM> it's still hashing isn't it :S
1864 2013-02-01 21:08:34 <sipa> ok, same thing
1865 2013-02-01 21:08:38 <sipa> no
1866 2013-02-01 21:08:48 <HM> well hashes are involved
1867 2013-02-01 21:08:48 <sipa> hashing is part of it, but it's far from the most expensive operation
1868 2013-02-01 21:09:01 cardpuncher has joined
1869 2013-02-01 21:09:44 Diapolo has quit (Ping timeout: 245 seconds)
1870 2013-02-01 21:09:59 occulta has joined
1871 2013-02-01 21:10:27 <HM> sipa, what bit is then?
1872 2013-02-01 21:10:41 <CodeShark> elliptic curve point multiplication is the most expensive part?
1873 2013-02-01 21:10:55 <Diablo-D3> yes
1874 2013-02-01 21:10:57 <HM> I thought EC was pretty damn fast
1875 2013-02-01 21:11:16 <cardpuncher> Hi, a little question regarding translations: In "Failed to read block info", are we talking about all the blocks or of a single block?
1876 2013-02-01 21:12:15 <sipa> cardpuncher: one block
1877 2013-02-01 21:12:28 <cardpuncher> Thanks.
1878 2013-02-01 21:12:46 Hashdog has joined
1879 2013-02-01 21:13:08 <Scrat> it's the fastest of the bunch for going from privkey to pubkey by far
1880 2013-02-01 21:13:11 <sipa> HM: try it on your CPU and you'll see how fast it is :p
1881 2013-02-01 21:13:14 <Scrat> but the opposite is quite slow compared to rsa
1882 2013-02-01 21:13:21 <Scrat> for comparable key sizes
1883 2013-02-01 21:13:42 <sipa> HM: actually, vanitygen doesn't do EC multiplication, only EC addition
1884 2013-02-01 21:13:53 <sipa> but even that is slower than SHA256 afaik
1885 2013-02-01 21:14:09 <sipa> EC multiplication is 2-3 orders of magnitude slower still
1886 2013-02-01 21:14:14 <HM> sipa, my CPU keeps up with my GPU
1887 2013-02-01 21:14:21 <sipa> ?
1888 2013-02-01 21:14:23 <HM> because i have a mobile nvidia chip
1889 2013-02-01 21:14:24 <Scrat> wouldnt RSA be a nice fit for bitcoin if the keys werent so damn big?
1890 2013-02-01 21:14:26 <HM> it's appalling
1891 2013-02-01 21:14:34 Diapolo has joined
1892 2013-02-01 21:15:10 <sipa> Scrat: yes, but that's like saying that helicopters would be a nice fit for replacing cars if they weren't so expensive
1893 2013-02-01 21:15:12 <Scrat> not like you need maximum address generation speed, but you do need maximum verification speed
1894 2013-02-01 21:15:18 <Scrat> lol :p
1895 2013-02-01 21:15:35 <sipa> RSA needs 3000 bits keys for the same security level
1896 2013-02-01 21:15:53 <Diapolo> bsdunx: currently I'm trying to keep the translations stuff current, any more question on this?
1897 2013-02-01 21:15:58 <TD> well, we might lose some of the nice things you can do with ecc also
1898 2013-02-01 21:16:32 <HM> i have both Fermi GPUs running now. 35 MKey/s combined
1899 2013-02-01 21:16:42 <HM> my laptop GPU was only pushing 330KKey/s
1900 2013-02-01 21:16:46 <HM> my CPU could do that :S
1901 2013-02-01 21:16:51 <Scrat> HM: are they M1020something?
1902 2013-02-01 21:16:55 <HM> M2050
1903 2013-02-01 21:17:13 <Scrat> still sucky compared to K10/20
1904 2013-02-01 21:17:24 <Scrat> amazon bought thousands of them but they still haven't deployed them
1905 2013-02-01 21:17:36 <HM> they'll probably be extra large?
1906 2013-02-01 21:18:09 <Scrat> each card is like 5 times faster in SP performance than the last gen
1907 2013-02-01 21:19:02 <HM> when people talk about GPGPU they always say "oh man, everyone has a CUDA/OpenCL GPU in their laptops these days"....yeah they do, but they mostly suck
1908 2013-02-01 21:19:07 <sipa> HM: anyway, ECDSA verification is essentially sort of an extended version of an EC multiplication (plus a few preparation steps that are fast), and the fastest i've seen is around 12000 verifications per second (on a 6-core 3.2 GHz modern high-end CPU)
1909 2013-02-01 21:20:08 <sipa> (all cores combined)
1910 2013-02-01 21:20:35 <D34TH> hmm
1911 2013-02-01 21:20:45 <D34TH> vanity gen isnt working for me
1912 2013-02-01 21:20:52 <D34TH> llvm errors
1913 2013-02-01 21:21:05 <rng29a> D34TH: are you using the ATI SDK?
1914 2013-02-01 21:21:08 <sipa> though if we're just talking about a multiplication, and you'd push all cores to 100%, you can probably do up to 3000-4000 per core
1915 2013-02-01 21:21:12 <D34TH> rng29a: yes
1916 2013-02-01 21:21:14 <rng29a> try downloading an older version
1917 2013-02-01 21:21:20 <Scrat> sipa: multithreaded verification is in 0.8?
1918 2013-02-01 21:21:24 <sipa> Scrat: yes
1919 2013-02-01 21:21:24 <rng29a> that worked for me
1920 2013-02-01 21:21:35 <D34TH> i have cat 13
1921 2013-02-01 21:21:44 <Diapolo> it seems my provider switched me to DS-Lite, so I'm now using native IPv6, wow future is coming
1922 2013-02-01 21:21:51 <D34TH> thats probably why
1923 2013-02-01 21:22:55 B0g4r7 has quit (Ping timeout: 276 seconds)
1924 2013-02-01 21:24:59 <Dagger2> DS-Lite, really? or do you just mean dual-stack?
1925 2013-02-01 21:26:46 rdponticelli has joined
1926 2013-02-01 21:27:26 <Diapolo> my router is telling me it's DS-Lite
1927 2013-02-01 21:27:41 <Diapolo> and I can't change my DNSv6 server addresses... firmware is behing it seems
1928 2013-02-01 21:27:51 Cylta has joined
1929 2013-02-01 21:28:03 <rng29a> D34TH: is the OpenCL SDK included in the driver? SDK 2.7 or 2.6 worked for me on linux
1930 2013-02-01 21:28:25 <Cylta> hi. can I ask here about gpg, or it's bad idea?
1931 2013-02-01 21:28:55 B0g4r7 has joined
1932 2013-02-01 21:29:28 <kjj> depends.  what's the question?
1933 2013-02-01 21:29:39 <D34TH> rng29a: oclruntime isnt compatible
1934 2013-02-01 21:30:37 <Dagger2> Diablo-D3: I'm kinda surprised, it seemed like dual stack would be easier than DS-lite
1935 2013-02-01 21:30:50 <Cylta> does clearsign contain publick key? kan I send a encrypted message to a person, without knowing his publick key, but with randomtext with clearsign.
1936 2013-02-01 21:30:53 <Diablo-D3> tabfail
1937 2013-02-01 21:30:57 <Dagger2> (DS-lite = your v4 is tunneled over a native v6 connection, so you end up with a lower MTU for v4)
1938 2013-02-01 21:31:12 <Diapolo> Dagger2: I'm Diapolo ^^
1939 2013-02-01 21:31:24 <Dagger2> but I guess they're just skipping straight ahead to the future rather than the 10 years ago
1940 2013-02-01 21:31:28 <Diapolo> I verified it, it's DS-Lite
1941 2013-02-01 21:31:32 <kjj> clearsign hashes your message, and signs that hash with YOUR private key.  if the recipient has YOUR public key, he can verify that it came from you
1942 2013-02-01 21:31:48 <Dagger2> blarg. you people need to get nicks with unique initial 3 characters :p
1943 2013-02-01 21:32:34 <Cylta> kjj so, I can even verify clearsign correctly without pubkey... thank you for explanation.
1944 2013-02-01 21:32:40 <moore> so I have done a little more thinking about  satoshidice and I have convinced my self that if you can reverse ( by doubble spending ) more then 1.2% of the time you have better odds then the house
1945 2013-02-01 21:33:09 <Luke-Jr> helo: "mkpasswd" - apparently part of the "whois" package and not intended to actually make passwords
1946 2013-02-01 21:33:29 <Luke-Jr> helo: http://dashjr.org/blog/2013/01/12/mkpasswd-does-not-make-passwords
1947 2013-02-01 21:33:47 <rng29a> D34TH: I am not sure what you mean. I tried it 2 weeks ago and after switching to an older SDK version vanity gen worked fine under Ubuntu 12.04
1948 2013-02-01 21:34:11 <Diapolo> Dagger2: sorry, but no :) we keep our names they are unique ^^
1949 2013-02-01 21:34:14 <moore> this means that if more then 1.2% of the miners reject transactions with satoshidice as a output they loose
1950 2013-02-01 21:34:20 <kjj> Cylta: in gpg, you sign with your private key, and others verify with your public key
1951 2013-02-01 21:35:02 <Diapolo> TD: any info, if the google IPv6-DNS is free and uncensored or is Google doing anything evil with it? ^^
1952 2013-02-01 21:35:34 <kjj> not counting logging your every query?
1953 2013-02-01 21:35:44 <moore> this is worse if there are minors that prefer transactions with larger transaction fees as you can for a cost push out satoshidice transactions
1954 2013-02-01 21:35:54 <Dagger2> I've not heard of them doing anything evil with it, but I use ordns.he.net myself
1955 2013-02-01 21:36:09 <Dagger2> ... for those rare times when I'm not running a recursive server myself
1956 2013-02-01 21:36:18 <Cylta> kjj yes, I know, thank you. by some reason I though that clearsign should contain pubkey, just in case...
1957 2013-02-01 21:36:57 <kjj> Cylta: it probably has a fingerprint.  but what good would it to do package the key with the message?  why would you believe it?
1958 2013-02-01 21:37:31 <Cylta> can I send encrypted message to a fingerprint?
1959 2013-02-01 21:37:32 <Diapolo> Dagger2: as I said I can't change it anymore, since my IPS activated DS-Lite :-/
1960 2013-02-01 21:38:00 <Diapolo> Can anyone try to reach my testnet node via 2a02:908:da19:a300:ac1c:eff0:8641:62ae
1961 2013-02-01 21:38:14 <CodeShark> moore: how do we get miners to just block transactions with satoshi dice outputs?
1962 2013-02-01 21:38:24 <rng29a> Cylta: no, but you can lookup the public key with the fingerprint if it is published
1963 2013-02-01 21:38:24 Cylta has quit (Read error: Connection reset by peer)
1964 2013-02-01 21:38:28 <moore> many do allready
1965 2013-02-01 21:38:40 <CodeShark> satoshi dice inputs should be fine, though :)
1966 2013-02-01 21:38:59 <Dagger2> From 2a02:908:da00:1:64ba:f4a5:f412:c5f icmp_seq=1 Destination unreachable: Administratively prohibited
1967 2013-02-01 21:39:01 <kinlo> anyone going to fosdem tomorrow?
1968 2013-02-01 21:39:57 <moore> and any way that just makes it easer
1969 2013-02-01 21:40:09 cardpuncher has quit (Quit: ChatZilla 0.9.89 [SeaMonkey 2.15.1/20130118191211])
1970 2013-02-01 21:40:09 <Luke-Jr> CodeShark: my block_dice branch handles both inputs and outputs
1971 2013-02-01 21:40:19 <Diapolo> Dagger2: thanks ... seems the FW is eating your request ...
1972 2013-02-01 21:40:19 <CodeShark> separately?
1973 2013-02-01 21:40:49 Cylta has joined
1974 2013-02-01 21:41:14 <moore> Luke-Jr, so if 1.3% of miners run your code satoshidice is broken
1975 2013-02-01 21:41:18 <Luke-Jr> CodeShark: it adds outputs to the memorypool with a blacklist flag set, so it can detect inputs
1976 2013-02-01 21:41:44 <Luke-Jr> moore: nah, pretty sure Eligius is over 1.3% alone
1977 2013-02-01 21:41:56 <TD> Diapolo: yes, we attempt to screw over all our users as thoroughly as possible. i don't recommend it.
1978 2013-02-01 21:41:59 <Luke-Jr> moore: and other pools also run it
1979 2013-02-01 21:42:05 <Luke-Jr> I'm pretty sure EclipseMC does, for example
1980 2013-02-01 21:42:15 <TD> Diapolo: there's no way we'd want to support a crap idea like ipv6 ;)
1981 2013-02-01 21:42:34 <Cylta> kjj sorry, was disconnected. if you answered anything, could you repeat it? thank you.
1982 2013-02-01 21:42:52 <Diapolo> TD: :-P I'm the german who has privacy concerns, don't answer that way ^^
1983 2013-02-01 21:42:55 <TD> haha
1984 2013-02-01 21:42:56 <moore> well then then it is the case that it is easy to get odds better then 1%
1985 2013-02-01 21:42:57 <TD> it's fine
1986 2013-02-01 21:42:59 <CodeShark> Cylta: while you were gone we figured out the meaning of everything...sorry, you missed it
1987 2013-02-01 21:43:03 <TD> you mean google public dns rihgt?
1988 2013-02-01 21:43:08 <TD> 8.8.8.8 etc?
1989 2013-02-01 21:43:10 <Luke-Jr> Diapolo: Google is evil. Why would you trust their word? :P
1990 2013-02-01 21:43:38 <Cylta> codeshark ohh, no. I want to know everything too :p
1991 2013-02-01 21:43:44 <kjj> Cylta: 03:20PM <rng29a> Cylta: no, but you can lookup the public key with the fingerprint if it is published
1992 2013-02-01 21:43:51 <moore> Luke-Jr, is there any reference to who is running you branch
1993 2013-02-01 21:43:57 <moore> I want to write up this attack
1994 2013-02-01 21:44:02 <Cylta> kjj okay, thank you.
1995 2013-02-01 21:44:11 alex__ has joined
1996 2013-02-01 21:44:11 <Luke-Jr> moore: attack? you mean defense
1997 2013-02-01 21:44:16 <moore> :)
1998 2013-02-01 21:44:44 <Luke-Jr> moore: SatoshiDice is an attack
1999 2013-02-01 21:44:50 <moore> attack aginst satoshidice
2000 2013-02-01 21:45:09 <moore> I think people are doing this now by the way
2001 2013-02-01 21:45:10 <Scrat> moore: but how easy is it to double spend consistently
2002 2013-02-01 21:45:27 ThomasV has quit (Ping timeout: 245 seconds)
2003 2013-02-01 21:45:32 <Luke-Jr> moore: defence from the attack of SatoshiDice*
2004 2013-02-01 21:46:09 <Diapolo> TD: yeah that one and the other common DNSv6 one :)
2005 2013-02-01 21:46:10 <moore> you win if more then 1.3 % of miners are blocking transactions with satoshidice as a output
2006 2013-02-01 21:46:22 <sipa> why?
2007 2013-02-01 21:46:27 <TD> i haven't used google public dns actually
2008 2013-02-01 21:46:30 <alex__> how come i cant find a place to get btc that checks out on sos?>
2009 2013-02-01 21:46:31 <Cylta> 1.2001% :)
2010 2013-02-01 21:46:32 <TD> it can cause occasional issues with load balancing
2011 2013-02-01 21:46:37 <TD> i know that. but it's a technical issue
2012 2013-02-01 21:46:39 <sipa> where does the numver 1.3% come from?
2013 2013-02-01 21:46:40 <Diapolo> 2001:4860:4860::8888 and 2001:4860:4860::8844
2014 2013-02-01 21:46:48 <moore> sipa, why what the dice thing?
2015 2013-02-01 21:46:53 <sipa> yes
2016 2013-02-01 21:47:02 <moore> ok so hear is how it works
2017 2013-02-01 21:47:05 <Cylta> sipa house comission + small number
2018 2013-02-01 21:47:20 <sipa> imho you need 100% of miners to block it, to stop it completely
2019 2013-02-01 21:47:41 <moore> if you bet on the 50% odds you expect to loose 1.9% of the coins you bet
2020 2013-02-01 21:47:41 <Diapolo> Dagger2: can you try once more?
2021 2013-02-01 21:47:54 <Cylta> sipa they don't want to stop it. they justwant to rob it... (
2022 2013-02-01 21:47:59 <sipa> oh
2023 2013-02-01 21:48:07 <sipa> i don't care about that
2024 2013-02-01 21:48:22 <moore> well robing it will stop it
2025 2013-02-01 21:48:26 <moore> I don't want to do it
2026 2013-02-01 21:48:28 <Luke-Jr> Cylta: nobody wants to rob Dice, just stop their attack
2027 2013-02-01 21:48:45 <sipa> pleade stop calling it an attack
2028 2013-02-01 21:48:45 <Cylta> sipa lol. you do not care that somebody could get robbed?
2029 2013-02-01 21:48:55 rdponticelli has quit (Ping timeout: 276 seconds)
2030 2013-02-01 21:48:59 <Dagger2> Diapolo: same. just to check, is 2a02:908:da00:1:64ba:f4a5:f412:c5f actually a machine you own?
2031 2013-02-01 21:49:00 <moore> I just want to show that one can and that there approach is strictly unsafe
2032 2013-02-01 21:49:06 <Luke-Jr> sipa: when they stop attacking, I will
2033 2013-02-01 21:49:09 <sipa> Cylta: i don't care about that site or those who gamble their money on it
2034 2013-02-01 21:49:31 <Diapolo> Dagger2: my PC I'm writing from
2035 2013-02-01 21:49:36 <moore> which I suspect will put them out of work
2036 2013-02-01 21:49:45 <CodeShark> SD will find very little sympathy from this channel generally, methinks
2037 2013-02-01 21:49:46 <Dagger2> isn't that 2a02:908:da19:a300:ac1c:eff0:8641:62ae?
2038 2013-02-01 21:50:18 <Cylta> codeshark nobody likes SD in general.
2039 2013-02-01 21:50:20 <sipa> Cylta: but doing double spendimg attacks on it sounds morally wrong
2040 2013-02-01 21:50:32 <Diapolo> Dagger2: 2a02:908:da19:a300:ac1c:eff0:8641:62ae yes, sorry I didn't look what you were posting
2041 2013-02-01 21:50:44 <Cylta> sipa at least you said it :-)
2042 2013-02-01 21:50:48 <Luke-Jr> CodeShark: personally, I'd love to do a SD-like service without the DDoS - but it's illegal :<
2043 2013-02-01 21:50:57 <sipa> blocking their transactions, in order to promote other usage, is something else imho
2044 2013-02-01 21:50:59 <fiesh> a big pool could also crush satoshi dice
2045 2013-02-01 21:51:10 <CodeShark> Luke-Jr: nobody has to know :p
2046 2013-02-01 21:51:15 <fiesh> once you have more than roughly 5% of the computing power, you can beat their system
2047 2013-02-01 21:51:15 <moore> fiesh I don't think so
2048 2013-02-01 21:51:24 <Dagger2> Diapolo: so the "administratively prohibited" responses are coming from another machine then
2049 2013-02-01 21:51:27 <Diapolo> sipa: a quick help, do I need -externalip if I want to be reachable via IPv6 or is -bind sufficient?
2050 2013-02-01 21:51:28 eckey has joined
2051 2013-02-01 21:51:29 <Luke-Jr> CodeShark: doesn't make it legal
2052 2013-02-01 21:51:37 <moore> unless you are using it to reverse bets
2053 2013-02-01 21:51:44 <fiesh> they tell you immediately if you have won or not -- so you only find a block that puts in the winning transactions
2054 2013-02-01 21:51:47 <sipa> Diapolo: externalip sets the ip your node announces
2055 2013-02-01 21:51:52 <fiesh> and leave out the losing ones
2056 2013-02-01 21:51:55 <Diapolo> sipa: thanks
2057 2013-02-01 21:51:57 <sipa> Diapolo: binds sets what you listen on
2058 2013-02-01 21:52:02 <moore> fiesh, yes
2059 2013-02-01 21:52:03 <sipa> you may need neither or both
2060 2013-02-01 21:52:09 rdponticelli has joined
2061 2013-02-01 21:52:09 <moore> and it is happening now
2062 2013-02-01 21:52:14 <kinlo> yeah, SD isn't legal for our laws, are they legal in their own country?
2063 2013-02-01 21:52:20 <fiesh> moore: it's happening now?
2064 2013-02-01 21:52:21 <Luke-Jr> kinlo: SD is New York
2065 2013-02-01 21:52:34 <Luke-Jr> kinlo: New York is big on anti-gambling..
2066 2013-02-01 21:52:34 <sipa> ireland, i read
2067 2013-02-01 21:52:37 <moore> fiesh http://blockchain.info/double-spends
2068 2013-02-01 21:52:42 <Luke-Jr> sipa: the server is in Ireland, but Erik is NY
2069 2013-02-01 21:52:52 <kinlo> Luke-Jr: so it is probably illegal
2070 2013-02-01 21:52:52 <HM> i love games of chance
2071 2013-02-01 21:52:57 <TD> the creator isn't actually erik
2072 2013-02-01 21:52:59 <fiesh> oh, somebody is doing it, I always wondered why nobody was doing it, thought about this 5 months ago
2073 2013-02-01 21:53:01 alex__ has quit (Ping timeout: 264 seconds)
2074 2013-02-01 21:53:01 <kinlo> yeah, international law, you gotta love it :)
2075 2013-02-01 21:53:03 <TD> i don't really understand what vorhees has to do with it, to be honest
2076 2013-02-01 21:53:04 <HM> just generated 2 vanities in the first 0.5% of probability
2077 2013-02-01 21:53:07 <TD> he doesn't actually write the code or run it
2078 2013-02-01 21:53:11 <Luke-Jr> TD: I doubt that's relevant.
2079 2013-02-01 21:53:15 <Luke-Jr> he does actually run it, AFAIK
2080 2013-02-01 21:53:34 <TD> i wonder if it even has a legal owner. or if it's even really a company at all.
2081 2013-02-01 21:53:59 <TD> i'd be surprised if fireduck had sold it, but who knows
2082 2013-02-01 21:54:00 <Luke-Jr> TD: the courts don't care about technicalities trying to evade the law
2083 2013-02-01 21:54:00 <kinlo> bitcoin is pretty easy to be made with no real owner
2084 2013-02-01 21:54:27 <CodeShark> well, he's not in prison and he seems to be making money
2085 2013-02-01 21:54:28 <TD> if vorhees doesn't actually do anything with it beyond chest-thumping, what would they charge him with? i'm not saying i think he's on solid ground though
2086 2013-02-01 21:54:33 <Luke-Jr> CodeShark: yet.
2087 2013-02-01 21:54:46 <TD> i mean there was that guy who was writing software for foreign companies who got charged, but that did seem to be an out of control prosecutor
2088 2013-02-01 21:54:58 <Luke-Jr> TD: IANAL, but I hear even promoting gambling is illegal in NY
2089 2013-02-01 21:55:04 <TD> huh
2090 2013-02-01 21:55:10 <fiesh> I don't see what's wrong with SD -- if people want to use their bitcoin for it, it seems to be a good thing
2091 2013-02-01 21:55:24 <CodeShark> busting gambling software people is often about stopping money laundering rackets more than stopping gambling
2092 2013-02-01 21:55:27 <fiesh> I personally have no understanding why anyone would want to play it, but then tastes are different
2093 2013-02-01 21:55:27 <Luke-Jr> fiesh: it's a DDoS attack against Bitcoin; the gambling, I don't care about
2094 2013-02-01 21:55:44 <Cylta> what's the daily turnover of SD?
2095 2013-02-01 21:55:57 <sipa> fiesh: what people do with their money! is theur business and i don't care about it. but technically it can be done in a way that's much less stressful on the network
2096 2013-02-01 21:55:58 <fiesh> Luke-Jr: well, only in the sense that people voluntarily participate in it.  it's quite different from a ddos using a botnet or whatever
2097 2013-02-01 21:56:07 <Cylta> 1k? 33k? 1mln btc?
2098 2013-02-01 21:56:12 <CodeShark> as satoshi dice can't really be used for money laundering, it probably isn't a huge target yet
2099 2013-02-01 21:56:28 <fiesh> sipa: how would it be less stressful?  haven't thought about that
2100 2013-02-01 21:56:38 <gavinandresen> yes, lets get 0.8 out the door and get the payment protocol implemented so SD doesn't have to use nasty hacks to do what their users want
2101 2013-02-01 21:56:54 <Luke-Jr> fiesh: not abusing transactions as a messaging transport
2102 2013-02-01 21:57:05 <fiesh> oh I see
2103 2013-02-01 21:57:32 <Luke-Jr> fiesh: so, you'd send a deposit, bet as much as you want on the site, then when you stop for ~5 mins or whatever, it sends the balance back
2104 2013-02-01 21:57:32 <gavinandresen> speaking of which… sipa, are the 0.8-tagged issues the only blockers for getting 0.8 out?
2105 2013-02-01 21:57:49 <sipa> gavinandresen: i'll have a look later this evening
2106 2013-02-01 21:58:02 <sipa> gavinandresen: btw, tonorrow i'll be on vacation for a week
2107 2013-02-01 21:58:28 <eckey> Luke: FYI http://www.wired.com/threatlevel/2013/01/coder-charged-for-gambling-software/all/
2108 2013-02-01 21:58:33 vessenes has joined
2109 2013-02-01 21:58:33 <fiesh> Luke-Jr: yeah, but I guess that's part of what people like about it, it's totally account-free
2110 2013-02-01 21:58:41 <Diapolo> Dagger2: very strange since that ISP change I can't get a single incoming connection... that sucks
2111 2013-02-01 21:58:42 <fiesh> Luke-Jr: of course that wouldn't require an account, but it's similar in spirit
2112 2013-02-01 21:58:46 <gavinandresen> sipa: no, I forbid it.  (kidding-- going someplace nice?)
2113 2013-02-01 21:58:48 <Luke-Jr> gavinandresen: 2253 2243
2114 2013-02-01 21:59:02 <sipa> gavinandresen: i hear it's cold an snowy over there
2115 2013-02-01 21:59:04 <phantomcircuit> eckey, that article is totally wrong, he was running the site and later tried to claim he was just a programmer
2116 2013-02-01 21:59:28 <sipa> gavinandresen: thankfully i intend to use that snow ib helping me glide down a mountain
2117 2013-02-01 21:59:39 <fiesh> also, does the current code have something that penalizes high activity addresses that don't pay fees?  would that be an option?
2118 2013-02-01 21:59:50 <gavinandresen> 2243 isn't a showstopper. I already tagged 2253 for 0.8
2119 2013-02-01 22:00:13 <Luke-Jr> hmm, github is showing it as untagged for me; weird
2120 2013-02-01 22:00:34 <Cylta> fiesh there is. transaction fithout fees normaly taked much more time to confirm. I think that's enogh.
2121 2013-02-01 22:00:39 <Luke-Jr> I don't see how 2243 isn't a showstopper, since it prevents building bitcoind without bitcoin-qt
2122 2013-02-01 22:00:45 ThomasV has joined
2123 2013-02-01 22:00:53 <sipa> Luke-Jr: i have no such problem
2124 2013-02-01 22:00:53 <Dagger2> Diapolo: "for your security", no doubt
2125 2013-02-01 22:01:07 <sipa> Luke-Jr: i never build qt
2126 2013-02-01 22:01:08 <fiesh> Cylta: but all transactions without fees are treated the same, no?  no matter if the address creates a horrendous amount of them or just one
2127 2013-02-01 22:01:13 <Luke-Jr> sipa: does your system have a CXXFLAGS defined?
2128 2013-02-01 22:01:14 <Dagger2> Diapolo: mind if I nmap you? maybe you have some ports I can reach, even if a ping doesn't work
2129 2013-02-01 22:01:19 <sipa> Luke-Jr: no
2130 2013-02-01 22:01:25 <Diapolo> Dagger2: sure
2131 2013-02-01 22:01:39 <Luke-Jr> sipa: that's probably why
2132 2013-02-01 22:01:46 <sipa> what flag do you need that breaks thigs?
2133 2013-02-01 22:01:53 <Diapolo> 8333 and 18333 should be forwarded :-/
2134 2013-02-01 22:01:55 <Cylta> fiesh yes. do you want to see 1mln new addresses from sd? I don't :-)
2135 2013-02-01 22:02:00 <Luke-Jr> sipa: any. even a null CXXFLAGS breaks it
2136 2013-02-01 22:02:11 <Luke-Jr> or at least ' '
2137 2013-02-01 22:02:23 <Luke-Jr> not sure if make specially checks for the null string
2138 2013-02-01 22:02:27 <sipa> well then don't set that
2139 2013-02-01 22:02:33 <Dagger2> Diapolo: perhaps it was the obvious port to try, but 8333 is open and I can connect to it
2140 2013-02-01 22:02:51 <sipa> i agree that compatibility is nicer, but claiming it breaks builds is an exaggeration
2141 2013-02-01 22:02:55 <Luke-Jr> sipa: it's pretty standard practice to set CXXFLAGS..
2142 2013-02-01 22:03:11 <Dagger2> Diapolo: pings are still being blocked though
2143 2013-02-01 22:03:32 <fiesh> Cylta: I don't think they can easily do that, can they?  Without ruining their customers experience who just want to send to one fixed address in order to gamble
2144 2013-02-01 22:03:51 rng29a has quit (Quit: Leaving)
2145 2013-02-01 22:04:25 <Diapolo> Dagger2: do you reach it via IPv4 or v6?
2146 2013-02-01 22:04:35 <Cylta> fiesh id prefer to not to tuch them. they are able to create a lot of problems by bypassing any restrictions.
2147 2013-02-01 22:05:15 <fiesh> hmm possible
2148 2013-02-01 22:05:39 <Dagger2> # telnet 2a02:908:da19:a300:ac1c:eff0:8641:62ae 8333
2149 2013-02-01 22:05:39 <Dagger2> Connected to 2a02:908:da19:a300:ac1c:eff0:8641:62ae.
2150 2013-02-01 22:06:05 <Luke-Jr> fiesh: when you define customer experience as abusing/attacking bitcoin, there is no compromise possible.
2151 2013-02-01 22:06:44 <Diapolo> Dagger2: that should be the bitcoin node
2152 2013-02-01 22:07:51 <sipa> Luke-Jr: in any case, i'm fine with your pull (even for 0.8) if it doesn't break pulltester or gitian
2153 2013-02-01 22:08:03 <Dagger2> Diapolo: I don't have a bitcoin client to test with, but I can connect, which is the important part
2154 2013-02-01 22:08:04 <Luke-Jr> pulltester is broken period it seems <.<
2155 2013-02-01 22:08:25 <sipa> /poke BlueMatt
2156 2013-02-01 22:08:36 ThomasV has quit (Ping timeout: 252 seconds)
2157 2013-02-01 22:08:58 <BlueMatt> pong
2158 2013-02-01 22:09:27 <Cylta> bluematt wrong answer, should be "ouch!" :)
2159 2013-02-01 22:09:35 Cylta has quit (Read error: Connection reset by peer)
2160 2013-02-01 22:09:39 <BlueMatt> sorry, sipa: ouch!
2161 2013-02-01 22:09:51 <Diapolo> Dagger2: indeed :) have you any knowledge, if DS-Lite has problems with incoming IPv4 connections as they use v6 for encapsulation it could be problematic?
2162 2013-02-01 22:10:07 Cylta has joined
2163 2013-02-01 22:10:25 <eckey> I'm looking at my transaction--I sent .001BTC.  The tx shows one outpoint, a "to" address for the .001, and then my originating address with the balance-.001.  Is this correct?
2164 2013-02-01 22:10:45 <eckey> Sending the balance back to me?
2165 2013-02-01 22:10:53 <Cylta> yes
2166 2013-02-01 22:11:05 <Diapolo> Dagger2: can you nmap 37.24.148.96 also?
2167 2013-02-01 22:11:09 <eckey> seems like a lot of extra work...
2168 2013-02-01 22:11:21 <BlueMatt> sipa: yea, sorry, pulltester/jenkins are on hiatus until I can figure out why the latest tests are failing on windows
2169 2013-02-01 22:11:47 <Cylta> eckey it will be send do one of 100(default) if addresses in your's wallet's pool
2170 2013-02-01 22:12:07 <CodeShark> sipa's going on vacation, pulltester isn't working...it's going to be one scary week!
2171 2013-02-01 22:12:07 <sipa> BlueMatt: where do you expect the problem is?
2172 2013-02-01 22:12:16 <BlueMatt> sipa: mingw
2173 2013-02-01 22:12:34 <Cylta> eckey I think it has been done like that because therd is no other way to split coins in particular ratio.
2174 2013-02-01 22:12:45 <BlueMatt> sipa: in other words, Im not sure :)
2175 2013-02-01 22:12:56 <BlueMatt> dont know whats up? blame the compiler!
2176 2013-02-01 22:13:00 <eckey> So the outpoint takes then entire amount?
2177 2013-02-01 22:13:24 <eckey> then gives back the remainder?
2178 2013-02-01 22:13:26 <Dagger2> Diapolo: well... it's just a regular tunnel. it should act the same as a native connection, except with a slightly reduced MTU
2179 2013-02-01 22:13:44 Cylta has quit (Read error: Connection reset by peer)
2180 2013-02-01 22:13:47 <Dagger2> Diapolo: the issue is that your ISP might be using NAT and not giving you a public v4 address
2181 2013-02-01 22:13:52 Cylta has joined
2182 2013-02-01 22:14:05 <Dagger2> which would of course prevent any incoming connections
2183 2013-02-01 22:14:08 <Cylta> eckey you can't send only part of piece of bitcoin, you should specify to what pieces you want to split it. I think so
2184 2013-02-01 22:14:18 <Dagger2> nmap gives "All 1000 scanned ports on aftr-37-24-148-96.unity-media.net (37.24.148.96) are filtered"
2185 2013-02-01 22:14:53 <Diapolo> Dagger2: that could be the case, so I'm reachable via IPv6 only currently, as I tried to scan 8333 and 18333 for that IP, which didn't work
2186 2013-02-01 22:15:05 <Diapolo> Dagger2: thanks for your help  :)
2187 2013-02-01 22:15:10 <gavinandresen> sipa BlueMatt : are you convinced that the block acceptance discrepancy is NOT a code change, and is a compiler bug of some kind?
2188 2013-02-01 22:15:30 <CodeShark> eckey: addresses do not have balances. outputs can be claimed by whoever has a key that can sign against the address used in that output - but each output is like a separate mailbox. you can open different mailboxes with the same key, but once it's been opened it cannot be used again
2189 2013-02-01 22:15:48 <CodeShark> so you need to create a new mailbox for the change
2190 2013-02-01 22:15:51 <BlueMatt> gavinandresen: at this point Im entirely unconvinced the test isnt bad (its a pretty strange test)
2191 2013-02-01 22:16:05 <BlueMatt> but it does pass on 0.7 and on master on linux
2192 2013-02-01 22:16:10 <Dagger2> Diapolo: if you can get your router to cough up its WAN address, you can see if it matches 37.24.148.96
2193 2013-02-01 22:16:18 <CodeShark> eckey: in other words, outputs cannot be partially spent
2194 2013-02-01 22:16:33 <gavinandresen> BlueMatt: mmm.  Can you gitian-build a windows binary and see if THAT passes?
2195 2013-02-01 22:16:34 <CodeShark> eckey: moreover, they can only be spent once
2196 2013-02-01 22:16:51 <Cylta> codesharks I already told that :-P
2197 2013-02-01 22:16:56 <Luke-Jr> gavinandresen: has anyone done a block-fuzzer with proof-of-work disabled yet, or would 0.8 benefit from me doing such?
2198 2013-02-01 22:17:07 <BlueMatt> gavinandresen: yes, working on it
2199 2013-02-01 22:17:14 <gavinandresen> BlueMatt: awesome, thanks
2200 2013-02-01 22:17:33 <gavinandresen> Luke-Jr: fuzz-testing always welcome.
2201 2013-02-01 22:17:45 <eckey> Cylta CodeShark: thanks!
2202 2013-02-01 22:17:55 <HM> EC2 was surprisingly easy to use
2203 2013-02-01 22:17:56 <Luke-Jr> gavinandresen: ok, I just wanted to be sure someone hadn't already done it
2204 2013-02-01 22:18:20 <gavinandresen> Luke-Jr: I don't think so, but i'm good at forgetting things like that
2205 2013-02-01 22:18:41 <Diapolo> Dagger2: it's just telling me as IPv4 that it's using a DS-Lite-Tunnel AFTR-Gateway:2a02:908::4065
2206 2013-02-01 22:19:25 ThomasV has joined
2207 2013-02-01 22:19:35 <Diapolo> Luke-jr: could you come up with a fuzz-tester for command-line checks ^^?
2208 2013-02-01 22:19:53 <sipa> gavinandresen: nonetheless, even a compiler bug that causes network rule tests to fail is oretty scary
2209 2013-02-01 22:20:09 <Luke-Jr> Diapolo: I'm no fuzzing expert; I just plan to throw random data at submitblock without the POW check..
2210 2013-02-01 22:20:54 <BlueMatt> sipa: to be fair, its testing a block which was encoded invalidly to make it > MAX_BLOCK_SIZE when the block isnt if encoded properly
2211 2013-02-01 22:21:14 <BlueMatt> sipa: the proper response to that case is debatable in any case
2212 2013-02-01 22:21:50 <sipa> hmmm
2213 2013-02-01 22:21:55 <sipa> nice check :)
2214 2013-02-01 22:22:10 <gavinandresen> BlueMatt: yes, bravo, nice test case!
2215 2013-02-01 22:22:18 <sipa> but indeed, debatable what the outcome should be
2216 2013-02-01 22:22:24 * gavinandresen didn't know the network protocol allowed different encodings
2217 2013-02-01 22:22:33 <BlueMatt> varints
2218 2013-02-01 22:22:35 <sipa> gavinandresen: varint encodings, i suppsoe
2219 2013-02-01 22:22:45 <BlueMatt> the strange thing is the error message
2220 2013-02-01 22:22:53 <sipa> being?
2221 2013-02-01 22:23:00 <Dagger2> Diapolo: the manual for AFTR makes it sound like it does NAT on the internal v4 side. so yeah, you're likely kinda stuffed for incoming v4 connections
2222 2013-02-01 22:23:03 <BlueMatt> 1 wec
2223 2013-02-01 22:23:04 <BlueMatt> sec
2224 2013-02-01 22:23:15 <BlueMatt> ERROR: ConnectBlock() : tried to overwrite transaction
2225 2013-02-01 22:24:35 <Dagger2> Diapolo: (that's not how things were supposed to go; you were supposed to get IPv6 without affecting your v4 connection 5-10 years ago, and only after a transition period would you lose your public v4 address. kinda sucks that they aren't giving you that transition period, but at least they're giving you v6 and not leaving you completely stuffed)
2226 2013-02-01 22:24:38 Cylta has quit (Read error: Connection reset by peer)
2227 2013-02-01 22:24:42 <sipa> BlueMatt: errr what
2228 2013-02-01 22:24:55 <sipa> i suppose printing the txid would help investigate that
2229 2013-02-01 22:25:04 <BlueMatt> sipa: yes, I find it incredibly strange as well
2230 2013-02-01 22:25:09 Cylta has joined
2231 2013-02-01 22:25:36 <HM> Diapolo: count yourself lucky, here ISPs are talking about CGNAT
2232 2013-02-01 22:26:00 <BlueMatt> strangest part is, it started happening recently, like past few weeks
2233 2013-02-01 22:26:12 <sipa> can you do a sort of bisect?
2234 2013-02-01 22:26:24 <BlueMatt> tried, got different results
2235 2013-02-01 22:26:37 <BlueMatt> (both times in code that is entirely unrelated....)
2236 2013-02-01 22:26:48 <sipa> ewww
2237 2013-02-01 22:28:19 <Diapolo> Dagger2: HM: as long as I'm able to accept IPv6 incoming and the Windows privacy extensions are working, then it's okay for me
2238 2013-02-01 22:30:11 <Diapolo> HM: that CGNAT is what is happening to me IPv4 wise, if I'm not misstaken
2239 2013-02-01 22:30:37 <Diapolo> see: https://en.wikipedia.org/wiki/IPv6_transition_mechanisms#Dual-Stack_Lite_.28DS-Lite.29
2240 2013-02-01 22:31:53 <HM> yeah but that's not quite as bad as regular CGNAT
2241 2013-02-01 22:32:18 <Diapolo> luke-jr: too bad :) I guess that could raise some interesting combinations causing some interresting issues ^^
2242 2013-02-01 22:32:56 <Diapolo> HM: not sure about the difference, you are talking about exclusive CGNAT without native IPv6?
2243 2013-02-01 22:33:14 <HM> yes exactly.
2244 2013-02-01 22:33:52 <Diapolo> HM: oh wow that sucks rather hard
2245 2013-02-01 22:34:06 <Diapolo> HM: I would kick such an ISP in the ass ... they should give you money ^^
2246 2013-02-01 22:35:24 <HM> My ISP already force NAT on me
2247 2013-02-01 22:35:37 <HM> their cable modem has a built in NAT that cannot be disabled
2248 2013-02-01 22:36:13 <HM> you have a dedicated external IP of course, but ...when they can't even get the basics right
2249 2013-02-01 22:36:44 <HM> Sad panda stuff
2250 2013-02-01 22:36:52 <Diapolo> HM: which country or ISP is that?
2251 2013-02-01 22:37:01 <HM> UK
2252 2013-02-01 22:37:03 <HM> Virgin
2253 2013-02-01 22:37:07 <eckey> comcast
2254 2013-02-01 22:37:24 TD has quit (Read error: Connection reset by peer)
2255 2013-02-01 22:37:34 TD has joined
2256 2013-02-01 22:38:06 <HM> the FTTC stuff British Telecom have rolled out over the last few years is good
2257 2013-02-01 22:38:11 <Diapolo> HM: how much does that cost and for which bandwith? just interrested
2258 2013-02-01 22:38:13 <eckey> T-mobile
2259 2013-02-01 22:38:53 <HM> Diapolo: save you some time http://store.virginmedia.com/broadband.html
2260 2013-02-01 22:39:03 <Diapolo> ^^ thanks
2261 2013-02-01 22:39:12 Jamesz has joined
2262 2013-02-01 22:39:36 <Dagger2> HM: hm. you should be able to put your VM modem/router thingy into bridge mode to turn its router off
2263 2013-02-01 22:39:44 <HM> basically $55 USD /mo for 100Mbps
2264 2013-02-01 22:39:47 <HM> but there are usage caps
2265 2013-02-01 22:39:53 <HM> time of day things, etc
2266 2013-02-01 22:40:33 <Dagger2> though there was a period of a year after its introduction where the SuperHub had no option in its web UI to turn that on, which they didn't seem to think was much of a problem
2267 2013-02-01 22:40:33 <HM> I'd rather the Mbps arms race ended and ISPs ditched usage caps and throttling
2268 2013-02-01 22:40:54 ShaTwo has quit (Ping timeout: 256 seconds)
2269 2013-02-01 22:40:55 <HM> Dagger2: i don't have a superhub, i have one of their older pieces of shit and they want me to pay to change it
2270 2013-02-01 22:40:57 <Diapolo> HM: i switched to a low-end contract, which has just 10MBit/s (real flat), telephone flat for 20 €/month
2271 2013-02-01 22:41:07 JZavala has quit (Ping timeout: 264 seconds)
2272 2013-02-01 22:42:16 CodesInChaos has quit (Read error: Connection reset by peer)
2273 2013-02-01 22:42:31 CodesInChaos has joined
2274 2013-02-01 22:42:38 <HM> Dagger2: i was dead miffed. i moved house and poked the engineer to try to get the old modem i had from the old place working on the new line
2275 2013-02-01 22:42:50 <HM> for some reason he couldn't get it working so unboxed this POS with forced NAT
2276 2013-02-01 22:42:53 setkeh has quit (Read error: Connection reset by peer)
2277 2013-02-01 22:43:09 <HM> i rang a few months after discovering this crap and complained and they said the superhub only comes with the high end packages.
2278 2013-02-01 22:43:26 <HM> i threatened to ditch them but at the time Infinity wasn't available and the alternative was fairly dire
2279 2013-02-01 22:43:39 <HM> they wouldn't budge anyway
2280 2013-02-01 22:43:58 holorga has left ("...")
2281 2013-02-01 22:44:59 <Diapolo> HM: I really hate that ISP/Router/modem bundling, so that you can't chose the devices YOU want
2282 2013-02-01 22:44:59 <Dagger2> HM: that sucks
2283 2013-02-01 22:45:30 <HM> Diapolo: and service bundling...
2284 2013-02-01 22:45:51 setkeh has joined
2285 2013-02-01 22:46:02 <HM> once upon a time landline calls were cheap
2286 2013-02-01 22:46:15 <HM> now we pay for landlines and VoIP and mobile is cheap
2287 2013-02-01 22:46:34 <Luke-Jr> ok, fuzzing even this is more complicated than I was thinking :/
2288 2013-02-01 22:47:04 <sipa> Luke-Jr: how so?
2289 2013-02-01 22:48:56 <eckey> Maybe it's time to rethink the Internet.  I just purchased an Arduino, XBee shield, and WiFi shield.  Mesh networking in a *very* small package.  If enough people had this kit, we could drive these arrogant ISPs to rethink their offerings.
2290 2013-02-01 22:49:09 <Luke-Jr> sipa: 99.9% of proposals are rejected due to various decode errors, and prevblock/txids not found etc
2291 2013-02-01 22:49:23 <Luke-Jr> eckey: mesh can't compete
2292 2013-02-01 22:49:37 <sipa> Luke-Jr: right, that's expected
2293 2013-02-01 22:49:43 <Luke-Jr> eckey: it'd be slower than tor, and people have incentives (battery life) to be antisocial
2294 2013-02-01 22:49:54 <eckey> if dense enough?
2295 2013-02-01 22:49:59 <Luke-Jr> sipa: yes, I wasn't thinking :P
2296 2013-02-01 22:50:07 ovidiusoft has quit (Ping timeout: 264 seconds)
2297 2013-02-01 22:50:08 rdymac has joined
2298 2013-02-01 22:50:58 <Luke-Jr> I suppose a more advanced fuzzer is possible, but blarg
2299 2013-02-01 22:51:05 Diapolo has left ()
2300 2013-02-01 22:52:22 <eckey> Ever wonder how much it would cost to decommission the last nuclear plant?
2301 2013-02-01 22:52:48 <sipa> ?
2302 2013-02-01 22:52:48 brwyatt is now known as Away!~brwyatt@brwyatt.net|brwyatt
2303 2013-02-01 22:53:01 <Luke-Jr> eckey: why would you want to do that?
2304 2013-02-01 22:53:07 <Luke-Jr> eckey: if anything, we need to build more
2305 2013-02-01 22:53:12 <eckey> responding to HM about the cost of landlines
2306 2013-02-01 22:53:31 <sipa> i don't see the link
2307 2013-02-01 22:53:45 <eckey> I *know* that.  just responding to HM
2308 2013-02-01 22:54:12 twixed has quit (Quit: Leaving)
2309 2013-02-01 22:54:30 <sipa> no, i don't see the link between paying for landlines and decomissioning nuclear plants
2310 2013-02-01 22:55:11 <eckey> As disruptive tech drives products the grave, the price increases.
2311 2013-02-01 22:55:39 <eckey> nevermind
2312 2013-02-01 22:55:55 Cylta has quit (Read error: Connection reset by peer)
2313 2013-02-01 22:56:15 <HM> it's basically DSL uses old copper lines
2314 2013-02-01 22:56:17 <HM> cable doesn't
2315 2013-02-01 22:56:24 <HM> but because it's the same market they can bundle anyway
2316 2013-02-01 22:56:49 <HM> i don't think it's even true that most people want landlines. mobiles seem to dominate my age demographic
2317 2013-02-01 22:57:54 WolfAlex has joined
2318 2013-02-01 22:58:08 <Dagger2> cable still uses old copper lines; it just uses a different protocol and the lines are often shorter
2319 2013-02-01 22:58:12 WolfAlex_ has quit (Ping timeout: 255 seconds)
2320 2013-02-01 22:58:38 <sipa> Dagger2: ?
2321 2013-02-01 22:59:00 <sipa> it uses coax (at least the part between the fiber connection point and the home)
2322 2013-02-01 22:59:20 <Dagger2> which is just copper
2323 2013-02-01 22:59:30 <Dagger2> despite what VM would have you believe from their advertising :(
2324 2013-02-01 22:59:48 <sipa> well, ok, it's copper, but i wouldn't it call it "just copper"
2325 2013-02-01 23:01:04 <HM> RF physics was never my strong point
2326 2013-02-01 23:01:16 <sipa> Dagger2: or do you consider Cat5 also just copper?
2327 2013-02-01 23:01:42 <jrmithdobbs> how about twinax? ;p
2328 2013-02-01 23:01:56 <Dagger2> compared to fibre, sure
2329 2013-02-01 23:02:17 <jrmithdobbs> twinax is only a few ms more latent than fiber
2330 2013-02-01 23:02:25 <jrmithdobbs> (over long hauls)
2331 2013-02-01 23:02:33 <jrmithdobbs> (and it's longest haul is not very long)
2332 2013-02-01 23:03:04 WolfAlex_ has joined
2333 2013-02-01 23:03:58 WolfAlex has quit (Ping timeout: 260 seconds)
2334 2013-02-01 23:07:43 <HM> yes well
2335 2013-02-01 23:08:16 <HM> the way broadband is going we'll all have 10 Gbps connections here in the UK, but nobody will be able to sync their blockchain because we'll all have 1 GB/day data caps
2336 2013-02-01 23:09:04 <HM> so best work on block compression :P
2337 2013-02-01 23:09:35 <sipa> that can only gain you a relatively small constant factor
2338 2013-02-01 23:10:39 drizztbsd has joined
2339 2013-02-01 23:12:24 CodesInChaos has quit (Read error: Connection reset by peer)
2340 2013-02-01 23:12:34 CodesInChaos has joined
2341 2013-02-01 23:12:50 <CodeShark> how much would the dice transactions compress?
2342 2013-02-01 23:13:01 <jrmithdobbs> hardly at all
2343 2013-02-01 23:14:33 toffoo has joined
2344 2013-02-01 23:15:19 <CodeShark> even noting that each instance of OP_DUP OP_HASH160 795d36a28f4eb088572065bc46c05e0827289ec0 OP_EQUALVERIFY OP_CHECKSIG could be reduced to, say, three characters?
2345 2013-02-01 23:16:30 <sipa> if you do that, you're essentially encouraging address reuse
2346 2013-02-01 23:16:43 ken` has quit (Quit: leaving)
2347 2013-02-01 23:17:17 <CodeShark> I've had multiple instances where reusing an address made perfect sense
2348 2013-02-01 23:17:26 <sipa> i know none
2349 2013-02-01 23:18:13 <CodeShark> allowing users to keep a deposit address they use repeatedly, donation addresses, and sending change back to the same address to make wallet management simpler
2350 2013-02-01 23:18:27 <sipa> (except for the fact that no protocol now exists to conveniently negotiate addresses)
2351 2013-02-01 23:18:54 <sipa> wallet management shoundn't be about micromanaging per-address balances
2352 2013-02-01 23:18:57 <CodeShark> say, for instance, you want to have a service where you automatically sell bitcoins at market when people send to a particular address
2353 2013-02-01 23:19:01 <jrmithdobbs> CodeShark: so your use cases are things you shouldn't be doing, ok
2354 2013-02-01 23:19:11 <CodeShark> shouldn't?
2355 2013-02-01 23:19:15 <jrmithdobbs> aye
2356 2013-02-01 23:19:16 <CodeShark> according to whom? :)
2357 2013-02-01 23:19:23 <jrmithdobbs> the security model
2358 2013-02-01 23:19:40 <CodeShark> the security model has little to do with whether you use the same address or not
2359 2013-02-01 23:19:43 <sipa> CodeShark: that is a perfect candidate for a payment uri
2360 2013-02-01 23:19:55 <jrmithdobbs> it has a lot to do with it, actually
2361 2013-02-01 23:20:01 <CodeShark> no, sipa - payment URI is pull. I want push
2362 2013-02-01 23:20:22 <sipa> i don't see the difference
2363 2013-02-01 23:20:22 <CodeShark> I want the sender to be the one who initiates the transaction
2364 2013-02-01 23:20:31 <sipa> yes
2365 2013-02-01 23:20:38 <sipa> of course?
2366 2013-02-01 23:20:41 <CodeShark> and I don't want the sender to have to have a direct connection to me
2367 2013-02-01 23:20:49 <sipa> why not?
2368 2013-02-01 23:21:14 <jrmithdobbs> i want a pony but cannot stand the smell of ponies, suggestions?
2369 2013-02-01 23:22:11 <CodeShark> because I don't want to require that the sender have some special client software - I want them to be able to use existing wallet apps, say
2370 2013-02-01 23:22:54 <CodeShark> anyhow, I think it's really cool to be able to do full push transactions
2371 2013-02-01 23:23:12 Hashdog has quit (Remote host closed the connection)
2372 2013-02-01 23:23:44 <CodeShark> furthermore, as far as change addresses, oftentimes I don't care if it's the same as a sending address because I'm not trying to hide
2373 2013-02-01 23:24:00 <CodeShark> perhaps the sending address is publicly known to belong to me
2374 2013-02-01 23:24:11 <CodeShark> and having to use new change addresses each time just complicates wallet management
2375 2013-02-01 23:24:12 <sipa> CodeShark: im my perfect world, the perfect client app only speaks a payment protocol
2376 2013-02-01 23:24:20 <sipa> and has no P2P support at all
2377 2013-02-01 23:24:42 vessenes has quit (Remote host closed the connection)
2378 2013-02-01 23:24:46 <CodeShark> but that does not exist right now
2379 2013-02-01 23:24:52 <sipa> sure, i'm not talking about now
2380 2013-02-01 23:24:54 <CodeShark> if that existed perhaps I'd reconsider what I said above
2381 2013-02-01 23:25:27 <sipa> right now, i also have a fixed donation address on my website and forum profile
2382 2013-02-01 23:26:14 <CodeShark> in my perfect world, endusers would never even see addresses
2383 2013-02-01 23:26:20 <sipa> absolutely
2384 2013-02-01 23:26:21 <CodeShark> they'd just create contacts
2385 2013-02-01 23:26:48 <sipa> with the contact storing a payment URI
2386 2013-02-01 23:27:51 <sipa> (except for anonymous donations, perhaps)
2387 2013-02-01 23:29:19 <HM> seeing addresses isn't problematic
2388 2013-02-01 23:29:23 <HM> if they're encoded in a friendly way
2389 2013-02-01 23:29:46 <sipa> the biggest problem with addresses is that they are called addresses
2390 2013-02-01 23:29:52 <CodeShark> heh
2391 2013-02-01 23:29:59 <sipa> which implies that they are sort of an identifier for a destination
2392 2013-02-01 23:30:07 <sipa> but they are just key id's
2393 2013-02-01 23:30:36 <sipa> when you decode the transaction output, it is shown as an address
2394 2013-02-01 23:30:47 <sipa> even if nobody intends to accept coins on it anymore
2395 2013-02-01 23:31:02 <HM> well why hash public keys at all.
2396 2013-02-01 23:31:08 <HM> to create addresses
2397 2013-02-01 23:31:16 agricocb has quit (Quit: Leaving.)
2398 2013-02-01 23:31:24 <HM> what was the reason for that
2399 2013-02-01 23:31:28 <sipa> imho it should only be called an address if someone consciously decides to publish/transfer it with the intent of receiving money on it
2400 2013-02-01 23:31:41 <sipa> HM: the reason for that was size of the human-readable string
2401 2013-02-01 23:31:43 <CodeShark> because the full public key is longer than the hash and the hash is secure enough
2402 2013-02-01 23:31:46 <sipa> a very bad reason imho
2403 2013-02-01 23:31:57 <sipa> but it actually adds security as well
2404 2013-02-01 23:31:59 <HM> yeah because a hash is equally hostile to humans
2405 2013-02-01 23:32:15 <sipa> satoshi didn't know public keys could be compressed
2406 2013-02-01 23:32:24 <sipa> so you'd have to transfer 512 bits of data
2407 2013-02-01 23:32:24 grau has quit (Remote host closed the connection)
2408 2013-02-01 23:32:28 <CodeShark> you mean just storing the x coordinate?
2409 2013-02-01 23:32:31 <sipa> yes
2410 2013-02-01 23:32:48 <CodeShark> and then computing sqrt(x^3 + 7)
2411 2013-02-01 23:32:54 <sipa> with pubkey hashes, that was reduced to 160 bits
2412 2013-02-01 23:33:04 <sipa> but this is all beside the point
2413 2013-02-01 23:33:06 <gmaxwell> But the security improvement is nice, if only theoretical. And it layed the cognative way for p2sh which is generally a better way to express scripts.
2414 2013-02-01 23:33:28 <sipa> imho, the basic problem is that we conflate "key identifier" with "address"
2415 2013-02-01 23:33:32 JZavala has joined
2416 2013-02-01 23:33:39 <CodeShark> (x^3 + 7)^(p+1) mod p
2417 2013-02-01 23:33:59 <CodeShark> I guess that's not too much computational work
2418 2013-02-01 23:34:29 * HM nods as if he knows what's going on
2419 2013-02-01 23:35:37 <CodeShark> oh wait, sorry
2420 2013-02-01 23:35:40 <CodeShark> I messed up the exponent :p
2421 2013-02-01 23:36:38 <sipa> that formula could be reduced to (x^3 + 7)^2 mod p
2422 2013-02-01 23:37:29 <CodeShark> a^[(p-1)/2)] mod p = 1, assuming a is a square. if p = 4k+3 (which it is for secp256k1), we have a^[(4k+2)/2] mod p = 1. so a^[2k+2] mod p = a
2423 2013-02-01 23:37:43 <CodeShark> so a^(k+1) mod p = sqrt(a)
2424 2013-02-01 23:38:08 Guest7347 is now known as topace
2425 2013-02-01 23:38:14 topace has quit (Changing host)
2426 2013-02-01 23:38:14 topace has joined
2427 2013-02-01 23:38:46 <HM> https://tools.ietf.org/html/draft-jivsov-ecc-compact-00#section-4.3
2428 2013-02-01 23:38:49 <HM> this seems relevant?
2429 2013-02-01 23:39:18 <CodeShark> yeah
2430 2013-02-01 23:39:48 <sipa> for secp256k1 there are always to solutions, though
2431 2013-02-01 23:39:53 <sipa> *two
2432 2013-02-01 23:40:12 <CodeShark> for any field where 1 has two square roots, no?
2433 2013-02-01 23:40:19 <sipa> yes
2434 2013-02-01 23:40:41 <CodeShark> 1 and p-1
2435 2013-02-01 23:41:03 <CodeShark> are the square roots of 1 for any field Fp
2436 2013-02-01 23:41:12 <CodeShark> or are representatives, I should say :)
2437 2013-02-01 23:41:25 <sipa> representatives?
2438 2013-02-01 23:41:39 <CodeShark> yeah, we could also use -1 or 10p-1
2439 2013-02-01 23:41:49 <sipa> ok, sure
2440 2013-02-01 23:42:11 occulta has quit (Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/)
2441 2013-02-01 23:42:26 <sipa> hmm, sure it's not p-2 ?
2442 2013-02-01 23:42:31 <CodeShark> yes :)
2443 2013-02-01 23:42:43 <CodeShark> (p-1)^2 = p^2 - 2p + 1 == 1 (mod p)
2444 2013-02-01 23:42:57 <sipa> right
2445 2013-02-01 23:43:08 <HM> oh oh! i understood that part :}
2446 2013-02-01 23:43:24 <sipa> i confused elements with exponents
2447 2013-02-01 23:43:33 <sipa> which are elements of Z_{p-1}
2448 2013-02-01 23:43:58 <sipa> anyway
2449 2013-02-01 23:43:59 <CodeShark> right, the multiplicative group
2450 2013-02-01 23:44:25 rdymac has quit (Remote host closed the connection)
2451 2013-02-01 23:44:27 <sipa> but the elements of Z_p referred to by 1 and p-1 are the square roots of the element of Z_p referred to by 1 :)
2452 2013-02-01 23:45:01 <CodeShark> yes, so therefore if y is a solution to y^2 = x^3 + ax + b in Z_p, so is y(p-1)
2453 2013-02-01 23:45:47 <sipa> and y*(p-1) mod p = p-y
2454 2013-02-01 23:45:55 <CodeShark> yep :)
2455 2013-02-01 23:46:31 <CodeShark> hence min(y', p-y') in that link that HM gave
2456 2013-02-01 23:46:38 <HM> modulo prime artihmetic is not at all intuitive to me
2457 2013-02-01 23:47:04 <HM> i spent a few hours trying to figure out how ECs actually worked with regard to pub and private keys and it never did click
2458 2013-02-01 23:47:26 bitafterbit has quit (Read error: Connection reset by peer)
2459 2013-02-01 23:47:30 <sipa> HM: from a high-level point of view, you have points on a curve
2460 2013-02-01 23:47:34 <CodeShark> the modulo arithmetic is just clever tricks to perform what amounts to operations on a cyclic group
2461 2013-02-01 23:47:36 <sipa> each point on the curve has a number
2462 2013-02-01 23:47:48 <sipa> HM: if you have the number, you can calculate the corresponding point
2463 2013-02-01 23:47:54 <sipa> but you can't do the inverse (efficiently)
2464 2013-02-01 23:48:00 <HM> yes
2465 2013-02-01 23:48:02 <CodeShark> the reason for using EC and not just, say, modular addition, is that the latter is very easy to reverse
2466 2013-02-01 23:48:07 <HM> i get the overview, but never understood why
2467 2013-02-01 23:48:23 <HM> points at infinity and all that nonsense
2468 2013-02-01 23:48:28 <HM> gah
2469 2013-02-01 23:48:32 <sipa> that's too low level
2470 2013-02-01 23:48:49 <sipa> well, it's cool to understand the actual computation involved
2471 2013-02-01 23:48:59 <HM> at least with RSA you can sort of say, well calculating X is easy but reversing it is hard
2472 2013-02-01 23:49:02 <HM> and it's semi intuitive
2473 2013-02-01 23:49:10 <CodeShark> it's just a cyclic group. g^n = 1 where g is a generator and n is the order
2474 2013-02-01 23:49:10 da2ce7 has quit (Ping timeout: 276 seconds)
2475 2013-02-01 23:50:14 <CodeShark> and ECs are just a representation for a cyclic group
2476 2013-02-01 23:50:24 <HM> I discovered at uni (EE) that mathematicians are horrible at explaining things to engineers
2477 2013-02-01 23:50:32 <CodeShark> lol
2478 2013-02-01 23:51:20 <CodeShark> ECs have many interesting properties - but for cryptographic purposes, the fact that g^n=1 is the real essence
2479 2013-02-01 23:51:22 <gmaxwell> HM: same dealo, the only reason you're familar with RSA being hard but not ECC is that you have more expirence with factoring than taking discrete logs on cyclic groups. :P
2480 2013-02-01 23:52:08 <CodeShark> factoring actually reduces to taking discrete logs :)
2481 2013-02-01 23:52:34 <CodeShark> or computing order
2482 2013-02-01 23:52:35 <gmaxwell> It's not _obvious_ that factoring should be hard. Telling that a number is even, for example is easy... lots of things that seem apparently hard turn out to have very fast solutions.
2483 2013-02-01 23:52:47 <gmaxwell> (and indeeed, factoring has MUCH faster soltions than seems obvious)
2484 2013-02-01 23:53:01 <gmaxwell> solutions*
2485 2013-02-01 23:54:30 <HM> the problem is people teach this stuff talking about acyclic groups and discrete logarithms, they don't start with a simple case and then say "but what if..." and build up
2486 2013-02-01 23:54:33 rdymac has joined
2487 2013-02-01 23:54:39 TD has quit (Quit: TD)
2488 2013-02-01 23:54:46 <gmaxwell> I've had a lot of fun explaining http://en.wikipedia.org/wiki/Dixon%27s_factorization_method to people. (which is the most powerful factoring method you can explain to $random_technical_people without too much work, it's also the structual basis for the fastest schemes)
2489 2013-02-01 23:55:18 <CodeShark> pedagogically I think it's best to start with special cases and work up to general theory...but once you've gone through that, you're loathe to deal with specific cases anymore :)
2490 2013-02-01 23:55:48 <CodeShark> so what happens is that the more advanced you get in math, the less interesting the special cases seem
2491 2013-02-01 23:56:02 <CodeShark> but the special cases are what motivated the study in the first place :)
2492 2013-02-01 23:56:09 <andytoshi> CodeShark: the problem is that EC groups are just cyclic groups
2493 2013-02-01 23:56:09 <gmaxwell> math is about making no cases special.
2494 2013-02-01 23:56:20 <andytoshi> they -are- simple cases from a mathematical perspective
2495 2013-02-01 23:57:24 <andytoshi> i agree, the group operation feels weird (though my more algebraic friends disagree)
2496 2013-02-01 23:58:07 <sipa> HM: all that matters really, is that there is some operation x->f(x) which transforms an element (from a finite set) to another
2497 2013-02-01 23:58:26 eoss has joined
2498 2013-02-01 23:58:36 DamascusVG has quit (Quit: I Quit - http://www.youtube.com/watch?v=9p97zsQ51Rw)
2499 2013-02-01 23:58:44 <sipa> and there's a way to quickly apply it N times, with very large N
2500 2013-02-01 23:59:07 ralphtheninja has joined
2501 2013-02-01 23:59:41 <HM> that doesn't seem important, what seems important is to understand why going the other way is hard
2502 2013-02-01 23:59:42 <sipa> there's no generic way to find out how many times it was applied
2503 2013-02-01 23:59:55 <CodeShark> a line can only intersect the curve y^2 = x^3 + ax + b at no points, one point, two points, or three points. you then consider the different cases and realize that it's possible to define a group