1 2011-11-15 00:00:08 tower has quit (Disconnected by services)
   2 2011-11-15 00:00:11 towerr has joined
   3 2011-11-15 00:03:24 erus` has quit (Quit: ChatZilla 0.9.87 [Firefox 8.0/20111104165243])
   4 2011-11-15 00:04:16 bernie has joined
   5 2011-11-15 00:04:16 bernie has quit (Changing host)
   6 2011-11-15 00:04:16 bernie has joined
   7 2011-11-15 00:05:19 magn3ts has joined
   8 2011-11-15 00:11:15 Shaded has joined
   9 2011-11-15 00:11:59 MimeNarrator has joined
  10 2011-11-15 00:13:34 ale_ has quit (Ping timeout: 265 seconds)
  11 2011-11-15 00:14:12 btc_novice has left ()
  12 2011-11-15 00:15:04 darkmethod has quit (Quit: Leaving...)
  13 2011-11-15 00:15:54 towerr has quit (Ping timeout: 248 seconds)
  14 2011-11-15 00:17:21 [eval] has quit (Ping timeout: 255 seconds)
  15 2011-11-15 00:18:32 Matt_von_Mises has quit (Quit: Leaving.)
  16 2011-11-15 00:19:23 Kolky has quit (Quit: Bye bye!)
  17 2011-11-15 00:20:12 Snapman[afkers] is now known as Snapman
  18 2011-11-15 00:20:42 AAA_awright has quit (Read error: Connection reset by peer)
  19 2011-11-15 00:21:01 AAA_awright has joined
  20 2011-11-15 00:23:42 CaptainDDL has quit (Ping timeout: 244 seconds)
  21 2011-11-15 00:24:49 copumpkin has joined
  22 2011-11-15 00:25:44 andywork has joined
  23 2011-11-15 00:31:22 amiller has quit (Ping timeout: 248 seconds)
  24 2011-11-15 00:32:38 tower has joined
  25 2011-11-15 00:33:28 <andywork> I read that the bitcoin client operates in a peer-to-peer network with no central authority. Does this mean that each client have to scan through the whole internet to locate other peers, where noone acts as supernodes?
  26 2011-11-15 00:34:21 <cocktopus> andywork: there are methods to seed a list of initial peers
  27 2011-11-15 00:34:36 <cocktopus> some hardcoded, some DNS based, and some based on IRC
  28 2011-11-15 00:34:45 <cocktopus> for "bootstrapping"
  29 2011-11-15 00:35:18 <cocktopus> once running, it can collect info about additional peers from other peers
  30 2011-11-15 00:35:19 <iocor> what's the irc channel details?
  31 2011-11-15 00:35:32 wasabi2 has joined
  32 2011-11-15 00:35:35 <cocktopus> look at the debug logs for the client
  33 2011-11-15 00:36:31 wasabi has quit (Ping timeout: 256 seconds)
  34 2011-11-15 00:36:54 <andywork> so each client shares their found peers with other peers?
  35 2011-11-15 00:37:41 <cocktopus> yes, as i understand it
  36 2011-11-15 00:38:51 minimoose has joined
  37 2011-11-15 00:40:24 iocor has quit (Quit: Computer has gone to sleep.)
  38 2011-11-15 00:42:21 Shaded has quit (Quit: Shaded)
  39 2011-11-15 00:43:46 <andywork> cool :)
  40 2011-11-15 00:46:55 osmosis has joined
  41 2011-11-15 00:51:46 cronopio has quit (Quit: leaving)
  42 2011-11-15 00:54:20 Shaded has joined
  43 2011-11-15 00:54:59 TD_ has quit (Quit: TD_)
  44 2011-11-15 00:57:10 eoss has joined
  45 2011-11-15 00:57:10 eoss has quit (Changing host)
  46 2011-11-15 00:57:10 eoss has joined
  47 2011-11-15 01:01:19 wolfspraul has joined
  48 2011-11-15 01:05:19 <roconnor> have people thought of how OP_EVAL interacts with OP_IF?
  49 2011-11-15 01:05:53 <roconnor> OP_IF OP_PUSHDATA <OP_ENDIF> OP_EVAL
  50 2011-11-15 01:07:25 <roconnor> ugh, OP_IF statements cannot cross scriptSig scriptPubKey boundaries
  51 2011-11-15 01:07:47 * roconnor regrets looking at OP_IF
  52 2011-11-15 01:11:39 kdible has joined
  53 2011-11-15 01:13:44 andywork has quit (Quit: Lämnar)
  54 2011-11-15 01:15:32 <roconnor> intersting, you can have multipel OP_ELSE's in one OP_IF statement.
  55 2011-11-15 01:16:09 <cjdelisle> which gets eval'd? first, last or all?
  56 2011-11-15 01:16:51 <luke-jr> or it could be more interesting
  57 2011-11-15 01:17:03 <luke-jr> iff true, it could be every odd block
  58 2011-11-15 01:17:07 <luke-jr> and iff false, every even block
  59 2011-11-15 01:17:15 <roconnor> luke-jr is right
  60 2011-11-15 01:17:21 <luke-jr> ie, the 2nd else could negate the 1st
  61 2011-11-15 01:17:58 <luke-jr> that could be actually useful behaviour, potentially… maybe
  62 2011-11-15 01:18:08 <luke-jr> can you do an ELSE after the ENDIF?
  63 2011-11-15 01:18:22 <roconnor> luke-jr: Not really
  64 2011-11-15 01:18:27 <luke-jr> IF <A-code> ELSE <B-code> ENDIF <common> ELSE <A-code> ELSE <B-code> ENDIF
  65 2011-11-15 01:18:29 <luke-jr> aww
  66 2011-11-15 01:18:50 <roconnor> maybe I should break down and make a wiki account
  67 2011-11-15 01:18:50 <Mqrius> avast antivirus detects a threat in cgminer... is this normal behaviour?
  68 2011-11-15 01:20:34 CaptainDDL has joined
  69 2011-11-15 01:22:33 <roconnor> actually the wiki is wrong about OP_IF ... if the top stack value is NOT 0, the statements are exectued ...
  70 2011-11-15 01:23:38 <roconnor> acutally mistaking true for 1 is problematic throughout the article
  71 2011-11-15 01:24:20 <gavinandresen> roconnor: nice catch.  Did you fix it already?
  72 2011-11-15 01:24:34 <roconnor> nope; I don't have a wiki account
  73 2011-11-15 01:24:46 <gavinandresen> I'll pay for it.
  74 2011-11-15 01:24:52 <gavinandresen> ... just send me the bill.
  75 2011-11-15 01:25:27 slush has joined
  76 2011-11-15 01:25:41 Turingi has joined
  77 2011-11-15 01:26:31 crazy_imp has quit (Ping timeout: 276 seconds)
  78 2011-11-15 01:27:55 crazy_imp has joined
  79 2011-11-15 01:28:12 <phantomcircuit> gavinandresen, http://instantrimshot.com/
  80 2011-11-15 01:28:50 <roconnor> WTF
  81 2011-11-15 01:29:01 <roconnor> it says it as no loops or nesting ''if'' statements
  82 2011-11-15 01:29:07 <roconnor> ... but it does have nesting if statements
  83 2011-11-15 01:29:47 wolfspraul has quit (Quit: leaving)
  84 2011-11-15 01:30:02 wolfspraul has joined
  85 2011-11-15 01:30:32 Shaded has quit (Quit: Shaded)
  86 2011-11-15 01:31:01 <roconnor> I'm pretty sure it has nesting if statements
  87 2011-11-15 01:31:21 <roconnor> otherwise my life would be simpler :)
  88 2011-11-15 01:32:26 backwardation25 has joined
  89 2011-11-15 01:32:31 backwardation25 has quit (Remote host closed the connection)
  90 2011-11-15 01:33:23 <gavinandresen> yes, it definitely has nesting ifs
  91 2011-11-15 01:34:13 <etotheipi_> this is why I haven't tried implementing OP_IF in my scripting engine yet... just not worth the effort atm
  92 2011-11-15 01:35:24 <roconnor> block 00000000009c304ccfb10c84f26bea6b3b245c146c16bb8d22a9e3848ee866a0 in testnet has an OP_IF :(
  93 2011-11-15 01:35:37 <etotheipi_> what??? I don't remember that
  94 2011-11-15 01:35:51 <roconnor> :O
  95 2011-11-15 01:36:09 <etotheipi_> oh yeah
  96 2011-11-15 01:36:12 <etotheipi_> there are 3 such scripts
  97 2011-11-15 01:37:27 <roconnor> luke-jr: alternating OP_ELSE's could have been more useful if IF statements could span across siganture-pubkey script boundaries.
  98 2011-11-15 01:37:46 <roconnor> alas that cannot happen
  99 2011-11-15 01:39:22 <gavinandresen> Why is OP_IF such a pain for you to implement?  It's less than 20 lines of SatoshiCode...
 100 2011-11-15 01:40:01 <etotheipi_> gavinandresen, I can write a neural-network back propagation routine in 20 lines of code (I've done it)
 101 2011-11-15 01:40:11 theorbtwo has joined
 102 2011-11-15 01:40:48 <gavinandresen> Of all the things you have to implement to get an alternative bitcoin implementation right, the scripting seems like something that is reasonably straightforward.
 103 2011-11-15 01:41:01 <etotheipi_> I do agree with that, besides OP_CHECKSIG
 104 2011-11-15 01:41:09 <roconnor> I'm pretty sure neural-network back propagation is easier than OP_IF :D
 105 2011-11-15 01:41:17 <gavinandresen> Yes, besides CHECKSIG...
 106 2011-11-15 01:41:30 <etotheipi_> OP_IF may be simpler than I originally thought... I was just confused looking at the satoshi code
 107 2011-11-15 01:41:41 <etotheipi_> maybe I should've spent more than 5 minutes trying to decode it
 108 2011-11-15 01:41:55 <roconnor> gavinandresen: the fact that the alternate stack is cleared between script calls and if statements cannot span script calls caught me off guard
 109 2011-11-15 01:42:45 <roconnor> there are about 1000 implicit design decisions in Satoshi's script code that you have to get exactly right
 110 2011-11-15 01:42:54 <gavinandresen> I asked Satoshi why allow anything besides "Push data" in the scriptSig, but never got an answer out of him.
 111 2011-11-15 01:43:11 <roconnor> Hostestly reimplementing Bitcoin is probably 10x or 100x harder than implementing bitcoin.
 112 2011-11-15 01:43:14 <gavinandresen> ... if it were up to me....
 113 2011-11-15 01:43:20 <etotheipi_> this makes me very happy I'm not aiming for full scripting/blockchain verification
 114 2011-11-15 01:43:29 <roconnor> because when implementing bitcoin the arbitrary choices can be made arbitrarily
 115 2011-11-15 01:43:53 <roconnor> but when reimplementing bitcoin the arbitrary choices must be made exactly the same way
 116 2011-11-15 01:44:19 <gavinandresen> roconnor: I dunno, Satoshi said he worked on it for a couple of years before the first release.  He had to think through a ton of details
 117 2011-11-15 01:44:50 <gavinandresen> roconnor: ... so 10x harder would mean it'll take you 20 years to re-implement
 118 2011-11-15 01:44:57 <roconnor> so you cannot choose to disallow improperly formed DER encoded ECDSA signatures (to give an example that Satoshi gave exactly 0 thought to).
 119 2011-11-15 01:45:31 <sipa> satoshi was no crypto expert - he seems to have used off-the-shelf encodings and routines
 120 2011-11-15 01:45:42 <gavinandresen> roconnor: it is what it is.  It would have been astonishing if he'd managed to get everything 100% right the first time.
 121 2011-11-15 01:46:00 <roconnor> gavinandresen: Intersting question: how long will it take for my code to be fully compatible with bitcoin?  probably never.
 122 2011-11-15 01:46:06 <roconnor> well, I worry about it at least
 123 2011-11-15 01:46:20 <sipa> roconnor: are you that far off?
 124 2011-11-15 01:46:27 <etotheipi_> but I totally understand about reimplementing:  that's basically my real job:  intermediary between big gov't contractor ... and the contractor:  we have to implement their code/design and tell the gov't whether it works
 125 2011-11-15 01:46:32 Snapman is now known as Snapman[afkers]
 126 2011-11-15 01:46:35 <theorbtwo> roconnor: If I understand your objection correctly, then you could always disallow such signatures in the future, or on a rolling basis.
 127 2011-11-15 01:46:50 <Eliel> gavinandresen: it took years to design bitcoin, but the time spent implementing the design can't be that big :P
 128 2011-11-15 01:46:52 <gavinandresen> I don't want you to stop worrying, I worry too... (which is why I plan on getting back to the cross-implementation testing project when I can)
 129 2011-11-15 01:46:53 <etotheipi_> (*** between contractor and gov't)  my life is reimplementing other peoples' designs
 130 2011-11-15 01:46:56 Cusipzzz has joined
 131 2011-11-15 01:47:32 <etotheipi_> fortunately, the result is not as "picky" about details as the Bitcoin protocol is
 132 2011-11-15 01:47:39 <gavinandresen> etotheipi_: there's plenty of design left to "build out" bitcoin infrastructure
 133 2011-11-15 01:47:40 <roconnor> sipa: It is really hard to tell; I'm constantly surprised by the things I get wrong
 134 2011-11-15 01:47:47 <roconnor> sipa: like this DER encoding issue.
 135 2011-11-15 01:47:59 <sipa> true; it's the details that bite
 136 2011-11-15 01:48:10 denisx has joined
 137 2011-11-15 01:48:11 <roconnor> and there are sooo many of them.
 138 2011-11-15 01:48:20 <etotheipi_> gavinandresen, I agree with you, except there's quite a bit of initial hurdle to get over
 139 2011-11-15 01:48:20 eueueue has joined
 140 2011-11-15 01:48:28 <gavinandresen> ... like berkeley db not deleting data when you ask it to....
 141 2011-11-15 01:48:42 <roconnor> gavinandresen: :D
 142 2011-11-15 01:48:47 <etotheipi_> 4 months later and I'm still trying to get the base for a light client working
 143 2011-11-15 01:49:11 <etotheipi_> I can tell you what I'm *not* using to store ECDSA keys :)
 144 2011-11-15 01:49:50 <roconnor> It's quite pleasing that, TD, etotheipi_ and I are really working in reimplementing different aspects of bitcoin
 145 2011-11-15 01:50:14 <roconnor> I think
 146 2011-11-15 01:50:30 wboy1 has quit (Ping timeout: 244 seconds)
 147 2011-11-15 01:50:34 <sipa> however, none of them will ever be combined
 148 2011-11-15 01:50:46 <sipa> as they are in 3 different languages
 149 2011-11-15 01:50:53 <roconnor> :D
 150 2011-11-15 01:50:57 <etotheipi_> what is TD doing it in?
 151 2011-11-15 01:51:02 <roconnor> etotheipi_: Java
 152 2011-11-15 01:51:09 <sipa> Java, he's the author of BitcoinJ
 153 2011-11-15 01:51:19 <etotheipi_> ahh
 154 2011-11-15 01:51:20 <sipa> and genjix's libbitcoin in C++ is a fourth
 155 2011-11-15 01:51:30 <roconnor> BitcoinJ doesn't implement scripting
 156 2011-11-15 01:51:31 <etotheipi_> well to be fair... I have a a fairly expansive C++ code base
 157 2011-11-15 01:51:34 <cocktopus> HASKELL 4 EVAR
 158 2011-11-15 01:51:50 <roconnor> which is fine; it isn't designed to
 159 2011-11-15 01:51:59 <roconnor> I don't have any wallet anything
 160 2011-11-15 01:52:25 <etotheipi_> I don't think there's any other way to do the bulk processing of the blockchain in python
 161 2011-11-15 01:52:32 <etotheipi_> or maybe I'm just not very good at python
 162 2011-11-15 01:52:36 <etotheipi_> :)
 163 2011-11-15 01:53:22 <sipa> it was a discussion on #haskell-blah, almost a year ago, that brought me here :)
 164 2011-11-15 01:54:27 <roconnor> sipa: copumpkin is a bitcoiner
 165 2011-11-15 01:54:43 <copumpkin> probably wasn't me
 166 2011-11-15 01:54:54 <copumpkin> I only started bitcoining in juneish
 167 2011-11-15 01:55:11 <sipa> i believe EvanR was here as well back then
 168 2011-11-15 01:56:16 <cjdelisle> 20:40 < sipa> satoshi was no crypto expert - he seems to have used off-the-shelf encodings and routines <-- using well reviewed well understood routines is the mark of a professional
 169 2011-11-15 01:56:47 <sipa> cjdelisle: absolutely
 170 2011-11-15 01:57:08 <roconnor> cjdelisle: and that is why we don't have key recovery :(
 171 2011-11-15 01:57:22 <sipa> but he wouldn't have chosen DER encoding for signatures, which is 72 bytes instead of 64
 172 2011-11-15 01:57:57 <sipa> and we wouldn't been using 279-byte serialized private keys, when 32 bytes suffice
 173 2011-11-15 01:58:04 <sipa> and indeed, key recovery
 174 2011-11-15 01:58:06 <cjdelisle> indeed, that does sound like a quickly made decision
 175 2011-11-15 01:58:08 <roconnor> sipa: apparently we can start droping leading 0x00's now
 176 2011-11-15 01:58:34 <roconnor> save a byte or two on signatures :)
 177 2011-11-15 01:58:35 <sipa> roconnor: great; 72-1/255 bytes
 178 2011-11-15 02:00:05 <roconnor> cjdelisle: as of yet, OP_CHECKMULTISIG ... makes no sense to me.  Then again neither does CODESEPARATOR.
 179 2011-11-15 02:00:11 wboy1 has joined
 180 2011-11-15 02:00:21 <sipa> nobody understands CODESEPARATOR, it seems
 181 2011-11-15 02:00:42 <cjdelisle> is it used in the chain?
 182 2011-11-15 02:00:44 <sipa> it seems i joined #bitcoin-discussion on november 17th, 2010
 183 2011-11-15 02:00:50 <roconnor> oh wait, I just understood OP_CHECKMULTISIG
 184 2011-11-15 02:00:52 <sipa> can't believe it's been a year
 185 2011-11-15 02:01:05 <cjdelisle> happy birthday
 186 2011-11-15 02:01:10 <sipa> not yet
 187 2011-11-15 02:01:21 <cjdelisle> ahh well you can open your present anyway
 188 2011-11-15 02:02:24 <cjdelisle> If you're interested in light clients, I have written a proposal which hashes out gmaxwell's "unspent transaction tree" plan http://btc.pastebay.com/144544
 189 2011-11-15 02:02:31 <luke-jr> I joined #bitcoin-dev on Jan 1, 2011
 190 2011-11-15 02:02:33 <luke-jr> :p
 191 2011-11-15 02:03:48 <gavinandresen> CODESEPARATOR I can't figure out, either.  CHECKMULTISIG isn't hard-- just <m sigs> m <n pubkeys> n OP_CHECKMULTISIG, the signatures must be in the same order as the public keys, there have to be exactly m of them.  And there's a bug, so you have to put something extra on the stack before everything if you want it to work...
 192 2011-11-15 02:04:30 <gavinandresen> ... oh, and all m of the signatures must validate.
 193 2011-11-15 02:04:42 <roconnor> gavinandresen: ya, I just figured out CHECKMULTISIG.  :D
 194 2011-11-15 02:05:00 <roconnor> the "must be in the same order" was missing in my head
 195 2011-11-15 02:05:05 <sipa> ow, it seems it was december 17, not november; excuse me!
 196 2011-11-15 02:05:16 <gavinandresen> let me know if you figure out what CODESEPARATOR is good for.  (I know how it works... but can't see how it could be useful)
 197 2011-11-15 02:06:26 <cjdelisle> can't call satoshi on the secret phone and ask "what the hell was that for?"?
 198 2011-11-15 02:06:37 <sipa> nobody will answer
 199 2011-11-15 02:06:41 Shaded has joined
 200 2011-11-15 02:07:27 <gavinandresen> secret phone seems to have a permanent busy signal
 201 2011-11-15 02:07:32 <cjdelisle> I had always assumed he was aroundish, kind of guiding development from a distance, able to warn about dangerous ideas erc..
 202 2011-11-15 02:07:56 <roconnor> cjdelisle: If he was he'd be warning about OP_EVAL :D
 203 2011-11-15 02:08:23 <sipa> he disappeared somewhere in march?
 204 2011-11-15 02:08:31 <cjdelisle> Yea, I kind of have second thoughts about op_eval given how stupifyingly complicated the scripting system really is.
 205 2011-11-15 02:09:04 <gavinandresen> ... that's like having an objection to #include because C++ is complicated...
 206 2011-11-15 02:09:24 <sipa> OP_EVAL does have some very nice use cases
 207 2011-11-15 02:09:47 <cjdelisle> The only thing is this is running untrusted code so any kind of a buffer overflow or a crashbug -> ruin
 208 2011-11-15 02:10:10 <sipa> we're already running untrusted code
 209 2011-11-15 02:10:16 <gavinandresen> yup
 210 2011-11-15 02:10:30 <gavinandresen> if OP_EVAL will crash you, then you will probably crash without it, too
 211 2011-11-15 02:10:45 <cjdelisle> yup, that's why adding anything to the programming language is scary...
 212 2011-11-15 02:11:14 <cjdelisle> but then as a hanger on, I am in no position to bitch about it.
 213 2011-11-15 02:11:32 <iddo> 279 bytes for private keys is the format of wallet.dat? why not change it to 32 bytes if it's private data that doesn't affect the protocol?
 214 2011-11-15 02:11:34 <gavinandresen> removing stuff or leaving it exactly the way it is... is scary, too....
 215 2011-11-15 02:11:55 Snapman[afkers] is now known as Snapman
 216 2011-11-15 02:11:56 <sipa> iddo: in encrypted wallets it is changed
 217 2011-11-15 02:12:06 <iddo> cool
 218 2011-11-15 02:12:20 <iddo> what was the advantage of 279 byte choice?
 219 2011-11-15 02:12:46 <sipa> using openssl's default serialization function
 220 2011-11-15 02:12:46 <denisx> luke-jr: you added the prev-hash code to pushpool, right?
 221 2011-11-15 02:13:01 roconnor has quit (Ping timeout: 260 seconds)
 222 2011-11-15 02:13:06 <iddo> ok
 223 2011-11-15 02:13:22 <denisx> luke-jr: do have also a changed version of poclbm which gets fresh work when it gets that message?
 224 2011-11-15 02:13:45 <luke-jr> denisx: I recently hacked my local copy to get new work when it gets any error
 225 2011-11-15 02:14:03 OneFixt has quit (Ping timeout: 240 seconds)
 226 2011-11-15 02:16:32 OneFixt has joined
 227 2011-11-15 02:16:33 OneFixt has quit (Changing host)
 228 2011-11-15 02:16:33 OneFixt has joined
 229 2011-11-15 02:19:00 Shaded has quit (Quit: Shaded)
 230 2011-11-15 02:20:43 eastender has quit (Ping timeout: 248 seconds)
 231 2011-11-15 02:24:14 gavinandresen has quit (Quit: gavinandresen)
 232 2011-11-15 02:25:32 eastender has joined
 233 2011-11-15 02:26:15 eueueue has quit (Quit: Page closed)
 234 2011-11-15 02:27:41 <denisx> luke-jr: smart move
 235 2011-11-15 02:27:43 Shaded has joined
 236 2011-11-15 02:29:54 <denisx> you should push that change
 237 2011-11-15 02:31:41 Turingi has quit (Read error: Connection reset by peer)
 238 2011-11-15 02:31:42 Shaded has quit (Client Quit)
 239 2011-11-15 02:39:05 <luke-jr> denisx: m0mchil doesn't seem to accept contributions
 240 2011-11-15 02:42:25 <denisx> luke-jr: some month ago he was very open to me
 241 2011-11-15 02:43:59 <denisx> I asked for "greater precision for rejected shares percentage" and he did it
 242 2011-11-15 02:44:45 <denisx> ok, thats 4 month ago
 243 2011-11-15 02:53:27 <luke-jr> yeah, but he wrote it too :p
 244 2011-11-15 02:56:55 marf_away has quit (Ping timeout: 258 seconds)
 245 2011-11-15 02:57:36 tower has quit (Ping timeout: 240 seconds)
 246 2011-11-15 03:13:47 eoss has quit (Remote host closed the connection)
 247 2011-11-15 03:24:13 eastender has quit (Ping timeout: 245 seconds)
 248 2011-11-15 03:28:09 nmat has joined
 249 2011-11-15 03:32:50 pnicholson has left ()
 250 2011-11-15 03:35:53 Shaded has joined
 251 2011-11-15 03:41:19 agricocb has quit (Remote host closed the connection)
 252 2011-11-15 03:41:48 agricocb has joined
 253 2011-11-15 03:42:49 ej_ has joined
 254 2011-11-15 03:47:15 agricocb has quit (Remote host closed the connection)
 255 2011-11-15 03:47:32 ej_ has quit (Ping timeout: 252 seconds)
 256 2011-11-15 03:48:08 agricocb has joined
 257 2011-11-15 03:48:17 NickelBot has joined
 258 2011-11-15 03:49:15 <Mqrius> !seen ThomasV
 259 2011-11-15 03:49:15 <gribble> ThomasV was last seen in #bitcoin-dev 2 days, 18 hours, 21 minutes, and 31 seconds ago: <ThomasV> Mqrius: you here?
 260 2011-11-15 03:49:16 <spaola> ThomasV (~ThomasV@unaffiliated/thomasv) was last seen quitting from #mtgox 4 hours, 17 minutes ago stating (Ping timeout: 244 seconds).
 261 2011-11-15 03:49:40 eastender has joined
 262 2011-11-15 03:53:43 alanp_ has joined
 263 2011-11-15 03:53:47 dvide has quit ()
 264 2011-11-15 03:54:39 wolfspraul has quit (Quit: Lost terminal)
 265 2011-11-15 03:55:05 alanp has quit (Ping timeout: 260 seconds)
 266 2011-11-15 03:55:23 CaptDDL has joined
 267 2011-11-15 03:56:41 CaptainDDL has quit (Ping timeout: 276 seconds)
 268 2011-11-15 03:56:47 tower has joined
 269 2011-11-15 04:07:00 TheSeven has quit (Disconnected by services)
 270 2011-11-15 04:07:27 [7] has joined
 271 2011-11-15 04:09:53 roconnor has joined
 272 2011-11-15 04:10:31 * roconnor wonders how CODESEPARATOR interacts with IF
 273 2011-11-15 04:11:21 SomeoneWeirdzzzz is now known as SomeoneWeird
 274 2011-11-15 04:11:38 <roconnor> it looks like a CODESEPARATOR that isn't executed in an IF block isn't activated
 275 2011-11-15 04:14:41 <roconnor> ooh, the wiki gets this right :D
 276 2011-11-15 04:15:41 Lolcust has quit (Read error: Connection reset by peer)
 277 2011-11-15 04:21:22 knotwork has joined
 278 2011-11-15 04:21:22 knotwork has quit (Changing host)
 279 2011-11-15 04:21:22 knotwork has joined
 280 2011-11-15 04:29:04 <etotheipi_> maybe I'll try to find ways to use OP_CODESEPARATOR on testnet....
 281 2011-11-15 04:29:09 <etotheipi_> *useful ways
 282 2011-11-15 04:29:59 <roconnor> etotheipi_: be sure to use it with an IF statement :)
 283 2011-11-15 04:30:43 <etotheipi_> roconnor, I made this diagram (http://dl.dropbox.com/u/1139081/BitcoinImg/OpCheckSigDiagram.png) based on various descriptions of how OP_CHECKSIG works... please let me know if I got the OP_CS wrong
 284 2011-11-15 04:31:47 <etotheipi_> though, I can't really illustrate interactions with OP_CS and OP_IF, that would probably muddy the hell out of the diagram
 285 2011-11-15 04:32:04 <roconnor> create a subscript from the last *executed* OP_CODESEPARATOR
 286 2011-11-15 04:32:21 <roconnor> (ya I would bother illustrating it)
 287 2011-11-15 04:32:47 <etotheipi_> so my interpretation is that CHECKSIG would include the opcodes *after* this script chunk, but not before
 288 2011-11-15 04:32:53 <etotheipi_> i.e. script_part4
 289 2011-11-15 04:35:16 <roconnor> you are missing several steps based on the value of the HashType  ... perhaps that is on purpose
 290 2011-11-15 04:37:14 wasabi has joined
 291 2011-11-15 04:37:35 <roconnor> Step 10 makes it sounds like the result of each signature check is pushed onto the stack wheras in reality only one bool is pushed onto the stack indiciating if all the signatures check or if some signature failed.
 292 2011-11-15 04:38:05 <etotheipi_> the title of the diagram is (OP_CHECKSIG (SIGHASH_ALL only)
 293 2011-11-15 04:38:18 <etotheipi_> I was just trying to get the simple, standard case down
 294 2011-11-15 04:38:31 <roconnor> ah okay
 295 2011-11-15 04:38:42 <etotheipi_> oooh, good point about the pushing
 296 2011-11-15 04:38:52 wasabi2 has quit (Ping timeout: 240 seconds)
 297 2011-11-15 04:39:11 <etotheipi_> I actually never quite understood the other hashcodes
 298 2011-11-15 04:39:54 <roconnor> I've implemented the other hashcodes ... but I've never tested it :/
 299 2011-11-15 04:41:00 <etotheipi_> can you give me the gist of what they're supposed to accomplish?
 300 2011-11-15 04:41:17 <etotheipi_> it looks like you can sign an input without locking in the other inputs
 301 2011-11-15 04:41:22 RobinPKR_ has joined
 302 2011-11-15 04:43:18 <roconnor> etotheipi_: it mostly controls weather the outputs are included in the signature or not
 303 2011-11-15 04:43:39 RobinPKR has quit (Ping timeout: 248 seconds)
 304 2011-11-15 04:43:40 RobinPKR_ is now known as RobinPKR
 305 2011-11-15 04:44:05 <roconnor> etotheipi_: so sigHashNone clears all the outputs
 306 2011-11-15 04:44:24 <roconnor> etotheipi_: and sigHashSingle clears all the outputs but one
 307 2011-11-15 04:44:49 <etotheipi_> so sigHashNone is like making a blank check
 308 2011-11-15 04:45:53 <roconnor> I'm not certain, but I think so
 309 2011-11-15 04:46:43 <etotheipi_> but does it require all the inputs to be in place when you sign it?
 310 2011-11-15 04:47:06 <roconnor> everything but the scripts for the inputs
 311 2011-11-15 04:47:24 <etotheipi_> I'm having a tough time thinking of use-cases for this... except maybe manually constructing partial transactions, that you want to give to others to sign and distribute the outputs how they decide to
 312 2011-11-15 04:47:29 <roconnor> AnyoneCanPay means all the other inputs are eliminated
 313 2011-11-15 04:48:09 <roconnor> so I guess SigHashNone plus AnyoneCanPay is like a blank check
 314 2011-11-15 04:50:09 <etotheipi_> oh wait, not a blank check....
 315 2011-11-15 04:50:18 <etotheipi_> output value is fixed, recipient isn't
 316 2011-11-15 04:50:30 <roconnor> bearer cheque
 317 2011-11-15 04:50:34 amiller has joined
 318 2011-11-15 04:51:32 * roconnor needs to rip out his OP_CODESEPARATOR optimization to implement OP_IF
 319 2011-11-15 04:52:08 <cjdelisle> that doesn't make a whole lot of sense since evil miners could replace the output in a sighashnone tx with their own.
 320 2011-11-15 04:52:40 <etotheipi_> roconnor, you are making me really glad I'm not trying to replicate the BTC protocol... just the pieces I need for a functioning client
 321 2011-11-15 04:52:42 <roconnor> cjdelisle: I think you are right
 322 2011-11-15 04:52:48 <etotheipi_> cjdelisle, that's partially true
 323 2011-11-15 04:53:03 <etotheipi_> cjdelisle, it's possible for the miners to do that if ALL the inputs are signed using sighashnone
 324 2011-11-15 04:53:14 <etotheipi_> but it's possible to have multiple inputs with different sighashes
 325 2011-11-15 04:53:40 <cjdelisle> ofc the miner could throw out the input that doesn't use sighashnone
 326 2011-11-15 04:53:41 <roconnor> etotheipi_: sure but an evil miner to drop the transaction entirely and make a new transaction that they output to themselves. ... it seems
 327 2011-11-15 04:53:54 <roconnor> oh wait, no they cannot
 328 2011-11-15 04:54:06 <roconnor> I see
 329 2011-11-15 04:54:19 <cjdelisle> ahh only if you use sighashnone | anyonecanpay
 330 2011-11-15 04:54:23 <roconnor> but if you has AnyoneCanPay and
 331 2011-11-15 04:54:30 <roconnor> cjdelisle: right, exactly
 332 2011-11-15 04:54:30 <etotheipi_> right
 333 2011-11-15 04:55:52 <etotheipi_> sometimes I wonder if I throw out a transaction script that anyone can redeem, or something crazy like this... how long would it take for someone else to find, recognize it, and redeem it
 334 2011-11-15 04:56:20 <roconnor> some people do that for fun
 335 2011-11-15 04:56:32 <roconnor> though usually they announce it on the forum
 336 2011-11-15 04:56:52 <etotheipi_> well, I know I would've found quite a few while debugging my blockchain reader
 337 2011-11-15 04:57:04 <roconnor> :)
 338 2011-11-15 04:57:06 <etotheipi_> any non-standardness got some thorough attention from me
 339 2011-11-15 04:57:23 EPiSKiNG- has quit ()
 340 2011-11-15 04:57:47 <roconnor> etotheipi_: are you using openssl (indirectly)?
 341 2011-11-15 04:58:03 <etotheipi_> haha, it's funny you ask that
 342 2011-11-15 04:58:22 <etotheipi_> I just had an epic battle with Crypto++ and almost switched to openssl... yesterday
 343 2011-11-15 04:58:29 Cusipzzz has quit (Quit: KVIrc 4.1.1 Equilibrium http://www.kvirc.net/)
 344 2011-11-15 04:58:41 <roconnor> does everyone use the same wrong DER decoding as openssl?!
 345 2011-11-15 04:58:58 <roconnor> oh wait, maybe you parse DER yourself?
 346 2011-11-15 04:59:02 <etotheipi_> so I actually have a pure-python implementation, which manually parses the script
 347 2011-11-15 04:59:06 <cjdelisle> do you have a converter for the openssl encoding of bignumbers to something else?
 348 2011-11-15 04:59:08 <etotheipi_> right, I don't use the DER
 349 2011-11-15 04:59:28 <etotheipi_> I spent some time picking apart the DER scripts
 350 2011-11-15 04:59:54 <etotheipi_> since they are so consistent, I was able to figure out how they're typically encoded (which isn't a lot of different ways in BTC)
 351 2011-11-15 05:00:11 devrandom has quit (Ping timeout: 248 seconds)
 352 2011-11-15 05:00:17 <etotheipi_> though, I know it's going to come back and bite me sometime in the future... some strange DER encoding will show up in the blockchain and my code will croak
 353 2011-11-15 05:00:39 <roconnor> etotheipi_: there is no problem
 354 2011-11-15 05:00:46 <roconnor> DER is only used for ECDSA signatures
 355 2011-11-15 05:00:48 <roconnor> that is it
 356 2011-11-15 05:01:09 <cjdelisle> that's kind of strange
 357 2011-11-15 05:01:21 <cjdelisle> I'd have expected the sig to be represented as X . Y
 358 2011-11-15 05:01:23 <roconnor> cjdelisle: it is openssl's fault
 359 2011-11-15 05:01:23 <etotheipi_> cjdelisle, it isn't too difficult to pull out the (r,s) values from the DER
 360 2011-11-15 05:01:36 <etotheipi_> in fact, I illustrated it here:  https://bitcointalk.org/index.php?topic=29416.0
 361 2011-11-15 05:01:48 <etotheipi_> you have to zoom in to see the individual bytes, though
 362 2011-11-15 05:02:21 <etotheipi_> (bottom left of the first image)
 363 2011-11-15 05:02:47 <etotheipi_> there's a boatload of size bytes...
 364 2011-11-15 05:03:00 <cjdelisle> ahh that's very simple
 365 2011-11-15 05:03:07 <cjdelisle> 0x04 . X . Y
 366 2011-11-15 05:03:19 <roconnor> this reminds me; I need to set if compressed public keys can be accepted
 367 2011-11-15 05:03:52 <etotheipi_> roconnor, is that where you put only the signature in the script?
 368 2011-11-15 05:04:15 <roconnor> nope, that is key recovery
 369 2011-11-15 05:04:25 <roconnor> compressed public keys look like:
 370 2011-11-15 05:04:41 <roconnor> 0x02 . X  or 0x03 . X  depending on the parity of Y
 371 2011-11-15 05:04:56 <roconnor> I conjecture that they work fine in bitcoin
 372 2011-11-15 05:04:59 <etotheipi_> oh because there's only 1 or two possible y values for a given X
 373 2011-11-15 05:05:00 <cjdelisle> and that is accepted in the chain?
 374 2011-11-15 05:05:09 <roconnor> cjdelisle: I conjecture so
 375 2011-11-15 05:05:12 <roconnor> but haven't tired
 376 2011-11-15 05:05:14 <cjdelisle> neato
 377 2011-11-15 05:05:28 <cjdelisle> patents be damned, that is cool
 378 2011-11-15 05:05:50 <etotheipi_> I thought the key recovery was cool:  only include the sig, then back-solve for the public key
 379 2011-11-15 05:05:50 zapnap has joined
 380 2011-11-15 05:05:56 <roconnor> I guess etotheipi_ hasn't found any of these in the wild
 381 2011-11-15 05:06:09 <roconnor> ya, this only cuts the key size in half
 382 2011-11-15 05:06:18 <roconnor> key recovery cuts the key size to 2 bits
 383 2011-11-15 05:06:40 <cjdelisle> you can derive a public key from a sig?
 384 2011-11-15 05:07:02 <roconnor> cjdelisle: from a sig and the data being signed
 385 2011-11-15 05:07:08 <etotheipi_> I've been waiting for some day when I get stuck in an airport for a delayed flight... to try to do the key recovery problem.... I refuse to look up the solution because I want to see if I can figure it out
 386 2011-11-15 05:07:15 <etotheipi_> :)
 387 2011-11-15 05:07:18 <cjdelisle> heh
 388 2011-11-15 05:07:20 <roconnor> cjdelisle: well that only narrows it down to at most 4 publick keys
 389 2011-11-15 05:07:58 <roconnor> and almost always at most 2 public keys
 390 2011-11-15 05:08:48 <cjdelisle> so you test each and figure out which works or are they both actually valid? sorry for the dumb question..
 391 2011-11-15 05:08:59 <roconnor> I forget
 392 2011-11-15 05:09:28 <roconnor> Originally I was thinking you would add 2 bits to secify which of the upto 4 keys you want
 393 2011-11-15 05:09:55 <roconnor> but maybe you could just try all 4 and see if any match the hash ... I'm not sure what the security implications of this would be.
 394 2011-11-15 05:10:02 <etotheipi_> well only one of them will match the Hash160 on the TxOut script
 395 2011-11-15 05:10:02 <roconnor> I think it would be okay
 396 2011-11-15 05:10:26 <roconnor> but sipa and I weren't 100% certain this was okay
 397 2011-11-15 05:10:30 <roconnor> maybe only 90% certain
 398 2011-11-15 05:11:02 <cjdelisle> but if there are actually 2 or 4 valid keys then there are 2 or 4 valid keypairs and that means ecdsa256 is really more like 255 or 254
 399 2011-11-15 05:11:26 <roconnor> closer to 255
 400 2011-11-15 05:11:26 <etotheipi_> not really... a public key is 64 bytes
 401 2011-11-15 05:11:42 <etotheipi_> not 32
 402 2011-11-15 05:11:59 <roconnor> but yes, there are often two public keys that would yield the same signature on the same data.
 403 2011-11-15 05:12:05 <roconnor> IIRC
 404 2011-11-15 05:12:55 <etotheipi_> right, reducing it to 510 or 511 bits of public key
 405 2011-11-15 05:13:00 <cjdelisle> I had no idea
 406 2011-11-15 05:13:22 <etotheipi_> there are:  115792089237316195423570985008687907852837564279074904382605163141518161494337
 407 2011-11-15 05:13:31 <etotheipi_> points on the secp256k1 curve
 408 2011-11-15 05:13:43 <etotheipi_> which is "0xfffffffffffffffffffffffffffffffebaaedce6af48a03bbfd25e8cd0364141" in hex
 409 2011-11-15 05:13:58 devrandom has joined
 410 2011-11-15 05:14:00 <etotheipi_> each one of those points is public key
 411 2011-11-15 05:14:12 <etotheipi_> which is really f***ing close to 2^256
 412 2011-11-15 05:14:29 <cjdelisle> well, it's 1/2 of 2^256
 413 2011-11-15 05:14:52 <cjdelisle> if every other key is a collision
 414 2011-11-15 05:15:20 <etotheipi_> oh, now I'm getting the nature of your question
 415 2011-11-15 05:15:21 <iddo> doesn't really matter for bitcoin because there are collisions of different keys for same 160 bit hashed address?
 416 2011-11-15 05:15:35 <roconnor> etotheipi_: I forget but I think identity element isn't allowed to be a public key
 417 2011-11-15 05:15:41 <roconnor> etotheipi_: so subtract one
 418 2011-11-15 05:15:43 <etotheipi_> iddo, you're absolutely correc,t but I"m still interested in the answer to this
 419 2011-11-15 05:15:54 <iddo> ok
 420 2011-11-15 05:16:24 <etotheipi_> so I guess the question then, is, it looks like given a signature, there are sometimes up to 4 points on the curve (4 public keys) which could've produced that signature
 421 2011-11-15 05:17:26 <cjdelisle> so ecc256 is slightly less strong than 255 bits
 422 2011-11-15 05:17:46 <etotheipi_> so you're probably right...
 423 2011-11-15 05:18:13 <etotheipi_> except that if two different people ended up with those two public/private keypairs... they could sign for each others' TxOuts, but they wouldn't realize it
 424 2011-11-15 05:18:46 denisx has quit (Quit: denisx)
 425 2011-11-15 05:18:55 <etotheipi_> actually... that's not *necessarily* true... the fact that four different public keys could've created that sig doesn't mean that all four keys are the *same*
 426 2011-11-15 05:19:06 <etotheipi_> it could be different four keys depending on which random number was used during signing
 427 2011-11-15 05:19:11 <cjdelisle> yea, it's just a theoretical interest, I'm sure *real* cryptographers have looked it over and took it into account.
 428 2011-11-15 05:19:11 redditbtc has joined
 429 2011-11-15 05:19:37 <etotheipi_> damnit, you might've just nerd-sniped me
 430 2011-11-15 05:19:47 <roconnor> :D
 431 2011-11-15 05:19:48 <etotheipi_> (or I sniped myself...)
 432 2011-11-15 05:20:07 * etotheipi_ runs off to look this up
 433 2011-11-15 05:20:54 <iddo> are you saying that given signature where are 4 privkeys, or given a pubkey there are 4 privkeys?
 434 2011-11-15 05:21:45 <roconnor> there is a bijection between private and public keys
 435 2011-11-15 05:21:50 <etotheipi_> I'm not positive:  we know that a signature may have up to four private/public keypairs that signed for it... but I don't know if those are actually "identical" keys... in fact I'm fairly certain they aren't
 436 2011-11-15 05:22:26 <etotheipi_> in fact, I would guess that the four different public keys that you find... would be different if the signer picked a different random number, but that's just a hypothesis
 437 2011-11-15 05:22:27 wolfspraul has joined
 438 2011-11-15 05:22:28 <roconnor> how can different keypairs be "identical"?
 439 2011-11-15 05:22:54 <iddo> roconnor: is it bijection or just 1-1 from sk to pk?
 440 2011-11-15 05:23:04 <roconnor> iddo: bijection
 441 2011-11-15 05:23:54 <iddo> well if it's bijection then you should get 256 bits of security
 442 2011-11-15 05:23:57 <etotheipi_> the question I want to know is:  you look at a signature and find two possible public/private keypairs that created that signature... if you now sign a new message with those two keypairs, are you going to get the same signature?
 443 2011-11-15 05:24:20 <iddo> i.e. given pubkey you have to guess 256 bits if you use brute force
 444 2011-11-15 05:24:20 Beremat has quit (Read error: Connection reset by peer)
 445 2011-11-15 05:24:52 <roconnor> iddo: there are 0xfffffffffffffffffffffffffffffffebaaedce6af48a03bbfd25e8cd0364141 public key / private key pairs
 446 2011-11-15 05:25:11 <etotheipi_> I'm just saying theoretically... if you could magically recover the public-private key pairs of the two keys that could've signed that message... are they actually the same?  I don't think so
 447 2011-11-15 05:25:13 <roconnor> iddo: one less if you don't count the infinity / 0 pair
 448 2011-11-15 05:25:20 kdible has quit (Ping timeout: 265 seconds)
 449 2011-11-15 05:25:56 <roconnor> etotheipi_: of course they wouldn't be the same; that is the definition of having two keys
 450 2011-11-15 05:25:58 <roconnor> :)
 451 2011-11-15 05:26:02 <iddo> ah that's less than 2^256 ? how much is that hex num?
 452 2011-11-15 05:26:32 <roconnor> iddo: what do you mean "how much"?
 453 2011-11-15 05:26:35 Snapman is now known as Snapman[afkers]
 454 2011-11-15 05:26:35 <etotheipi_> Number of keys:  115792089237316195423570985008687907852837564279074904382605163141518161494337  2^256 = 115792089237316195423570985008687907853269984665640564039457584007913129639936
 455 2011-11-15 05:26:49 <iddo> thanks:)
 456 2011-11-15 05:27:01 <roconnor> iddo: is is about 2^256 - 2^128
 457 2011-11-15 05:27:19 <roconnor> little more
 458 2011-11-15 05:27:45 <etotheipi_> yeah, I realize that was a pretty useless message with those numbers
 459 2011-11-15 05:28:15 <cjdelisle> just checked
 460 2011-11-15 05:28:30 <cjdelisle> it works by trying one then trying the other
 461 2011-11-15 05:28:35 <iddo> it's not true that for any signature and any unrelated privkey, you can find plaintext message that if you sign with that privkey you get that signature?
 462 2011-11-15 05:28:42 <cjdelisle> if one doesn't verify it flips the sign
 463 2011-11-15 05:28:48 <roconnor> etotheipi_: I had a computer checked proof of primality drawn up for that number :)
 464 2011-11-15 05:29:30 <roconnor> iddo: I think it is (almost) true
 465 2011-11-15 05:29:37 <etotheipi_> I hope it's prime
 466 2011-11-15 05:29:41 <roconnor> iddo: I think you can do it about 50% of the time
 467 2011-11-15 05:29:46 lfm_ has joined
 468 2011-11-15 05:29:59 <roconnor> etotheipi_: the order of elliptic curves don't have to be prime, but it is quite nice when they are.
 469 2011-11-15 05:30:15 <etotheipi_> It's a G(p) curve... where p is prime
 470 2011-11-15 05:30:30 <iddo> roconnor: so what was the point about having 4 possible keys given a signature? isn't it true that there are about 2^255 keys instead of just 4 ?
 471 2011-11-15 05:31:17 <iddo> or was it given signature+plaintext hmm..
 472 2011-11-15 05:31:35 <roconnor> iddo: for a given signature *and given data* there are at most 4 public/private key pairs that could have made that signature
 473 2011-11-15 05:31:47 <iddo> ok ok
 474 2011-11-15 05:32:17 <roconnor> and you can easily compute the 4 public keys
 475 2011-11-15 05:32:19 Snapman[afkers] is now known as Snapman
 476 2011-11-15 05:32:28 <roconnor> or however many there are (could be 0 keys)
 477 2011-11-15 05:32:32 <lfm_> yes it would be vanishingly unlikly the other 3 correspond to any actual key
 478 2011-11-15 05:32:35 <roconnor> not so easy to compute the private keys :D
 479 2011-11-15 05:32:41 <cjdelisle> roconnor: I checked the document, you can narrow it down by testing them using the verification routine so ecc256 is apparently truely ecc256
 480 2011-11-15 05:33:24 <lfm_> ping lfm
 481 2011-11-15 05:33:28 <etotheipi_> * 2^256-epsilon
 482 2011-11-15 05:34:25 <cjdelisle> But then saving a little space in the chain at the cost of increasing processor cost looks like a bad deal
 483 2011-11-15 05:34:39 <iddo> i don't think it affect the security, if you find one of the other 4 privkeys given the signature then it won't help you sign new messages if the privkey you found doesn't correspond to the right pubkey
 484 2011-11-15 05:35:09 <cjdelisle> better idea is to work on an unspent tx tree so we don't need to cart the chain around with us everywhere
 485 2011-11-15 05:35:11 <roconnor> cjdelisle: I'm moderately sure the key recovery is only slightly more expensive than normal signature verification.
 486 2011-11-15 05:35:51 <etotheipi_> I haven't seen the proof, but any ECC arithmatic operations are pretty expensive
 487 2011-11-15 05:36:12 <etotheipi_> especially if it has a x^-1 in it... which most of them do
 488 2011-11-15 05:36:35 <roconnor> yes, but the key recovery operation is, I think, the same work as verification
 489 2011-11-15 05:36:42 <roconnor> is is basically verification run backwards
 490 2011-11-15 05:36:42 <etotheipi_> oh, I see
 491 2011-11-15 05:36:49 <roconnor> *it is
 492 2011-11-15 05:37:02 <cjdelisle> even if it's the same cost and 1/2 of the time you get the wrong one, it is still expensive
 493 2011-11-15 05:37:04 <etotheipi_> oh yeah, I guess you don't need to verify the pub key if you just derived it from the sig
 494 2011-11-15 05:37:37 <roconnor> all the 2 or 4 keys are easy to compute from each other
 495 2011-11-15 05:37:49 <cjdelisle> mmhmm
 496 2011-11-15 05:38:03 <etotheipi_> gah, I gotta go to sleep...
 497 2011-11-15 05:38:05 <roconnor> first of all we can discount the 4 key problem
 498 2011-11-15 05:38:18 <roconnor> since that isn't likely to come up in this universe's lifetime
 499 2011-11-15 05:38:23 <cjdelisle> well unspent tx tree would allow a client to be running in like 3 minutes from a cold start so that's way way more important IMO
 500 2011-11-15 05:38:44 <etotheipi_> btw, to answer the earlier question, my code has a pure-python ECDSA implementaiton, which is slow as dirt... but I also added a C++ call, via Crypto++ and SWIG... just so I can have a fast, secure method availble if SWIG is around
 501 2011-11-15 05:38:49 <roconnor> and I think the 2 keys are like the negation or each other, or some other trivial operation.
 502 2011-11-15 05:39:04 <roconnor> etotheipi_: and DER?
 503 2011-11-15 05:39:14 <roconnor> etotheipi_: always decoded in python?
 504 2011-11-15 05:39:39 <roconnor> cjdelisle: I'd be more interested in using compresed public keys
 505 2011-11-15 05:39:51 zapnap has quit (Remote host closed the connection)
 506 2011-11-15 05:39:54 <roconnor> which I expect can be used right now with no protocol changes
 507 2011-11-15 05:40:26 <cjdelisle> I have a proposal for unspent tx trees with no breaking protocol changes
 508 2011-11-15 05:40:55 <etotheipi_> crypto++ handles the DER for me
 509 2011-11-15 05:41:03 <cjdelisle> you just link them into the merged mining hashtree and clients which don't support them don't notice
 510 2011-11-15 05:41:13 <cjdelisle> basicly all it affects is the coinbase
 511 2011-11-15 05:41:14 <roconnor> etotheipi_: really, the same way openssl does?
 512 2011-11-15 05:41:18 <etotheipi_> oh crap... no it doesn't
 513 2011-11-15 05:41:19 * roconnor goes looks
 514 2011-11-15 05:41:22 <roconnor> oh
 515 2011-11-15 05:41:23 <roconnor> whew
 516 2011-11-15 05:41:38 <etotheipi_> that's right, I just remembered I had to mangle it to make it work
 517 2011-11-15 05:41:51 <etotheipi_> it has to be 64 bytes, unsigned, big-endian (x,y) pair
 518 2011-11-15 05:42:01 <roconnor> cjdelisle: how do you stop miners from lying about the tree?
 519 2011-11-15 05:42:37 <cjdelisle> each tree references the last tree, if a miner lies you know because there are more transactions in the tree which didn't happen in between
 520 2011-11-15 05:43:53 <cjdelisle> If you hold the entire tree on your desktop, verifying it is trivial because it references the last tree so you just look at the blocks in between the latest and the one before
 521 2011-11-15 05:44:09 <etotheipi_> roconnor, so you're on your own with the DER... I picked apart the bytemap and manually entered the (x,y) pair as large integers, either to a python module, or crypto++
 522 2011-11-15 05:44:11 <cjdelisle> you can decide how many trees back you want to go depending on your paranoia level
 523 2011-11-15 05:44:31 <etotheipi_> (one nice benefit of python is the arbitrary-length integers:  2**2000 returns a valid [very long] result)
 524 2011-11-15 05:44:54 <roconnor> etotheipi_: no problem.  I'm good;  I just wanted to know what others were doing
 525 2011-11-15 05:44:58 <cjdelisle> ahh nice of python to do that
 526 2011-11-15 05:45:23 <etotheipi_> it means you can implement EC math without importing any modules
 527 2011-11-15 05:45:56 <etotheipi_> actually... any of these assymetric encryptions... it's slow, but it's ridiculously portable and no dependencies :)
 528 2011-11-15 05:46:27 <cjdelisle> here's the full unspent tx tree proposal http://btc.pastebay.com/144544
 529 2011-11-15 05:46:48 <cjdelisle> there are a few missing details such as how a tree makes reference to the tree before it.
 530 2011-11-15 05:47:07 <lfm_> so its not full then
 531 2011-11-15 05:47:17 osmosis has quit (Quit: Leaving)
 532 2011-11-15 05:47:58 <cjdelisle> but I went out of my way to make sure it would function without any one node holding the entire unspent tx tree since that could easily grow to gigabytes of size.
 533 2011-11-15 05:48:28 WakiMiko_ has quit (Ping timeout: 240 seconds)
 534 2011-11-15 05:49:04 RazielZ has joined
 535 2011-11-15 05:50:23 WakiMiko has joined
 536 2011-11-15 05:54:04 BurtyB has quit (Ping timeout: 248 seconds)
 537 2011-11-15 05:55:02 minimoose has quit (Quit: minimoose)
 538 2011-11-15 05:56:47 BurtyB has joined
 539 2011-11-15 06:16:28 BurtyB has quit (Ping timeout: 248 seconds)
 540 2011-11-15 06:16:57 amiller has quit (Remote host closed the connection)
 541 2011-11-15 06:17:02 BurtyB has joined
 542 2011-11-15 06:18:15 amiller has joined
 543 2011-11-15 06:20:29 OneFixt_ has joined
 544 2011-11-15 06:22:52 OneFixt has quit (Ping timeout: 248 seconds)
 545 2011-11-15 06:37:30 alanp_ has quit (Ping timeout: 256 seconds)
 546 2011-11-15 06:39:28 wasabi2 has joined
 547 2011-11-15 06:40:03 wasabi has quit (Ping timeout: 245 seconds)
 548 2011-11-15 06:40:21 AStove has joined
 549 2011-11-15 06:46:40 Lolcust_Backup has quit (Quit: Nap time)
 550 2011-11-15 06:47:42 urstroyer has joined
 551 2011-11-15 06:49:29 larsivi has quit (Read error: Operation timed out)
 552 2011-11-15 06:51:42 Lolcust_Backup has joined
 553 2011-11-15 06:55:03 eastender has quit (Ping timeout: 245 seconds)
 554 2011-11-15 06:58:19 RobinPKR_ has joined
 555 2011-11-15 06:58:22 RobinPKR has quit (Read error: Connection reset by peer)
 556 2011-11-15 06:58:22 RobinPKR_ is now known as RobinPKR
 557 2011-11-15 07:15:46 CaptDDL is now known as CaptainDDL
 558 2011-11-15 07:17:36 ThomasV has joined
 559 2011-11-15 07:23:23 agath has quit (Ping timeout: 244 seconds)
 560 2011-11-15 07:27:24 devrandom has quit (Ping timeout: 248 seconds)
 561 2011-11-15 07:32:12 molecular has quit (Ping timeout: 248 seconds)
 562 2011-11-15 07:32:33 molecular has joined
 563 2011-11-15 07:33:44 Lolcust has joined
 564 2011-11-15 07:34:27 devrandom has joined
 565 2011-11-15 07:37:47 Cory has quit (Read error: Connection reset by peer)
 566 2011-11-15 07:38:46 wasabi has joined
 567 2011-11-15 07:40:42 wasabi2 has quit (Ping timeout: 260 seconds)
 568 2011-11-15 07:40:44 DaQatz has quit (Ping timeout: 248 seconds)
 569 2011-11-15 07:41:50 redditbtc has left ()
 570 2011-11-15 07:53:02 Cory has joined
 571 2011-11-15 07:53:02 Cory has quit (Changing host)
 572 2011-11-15 07:53:02 Cory has joined
 573 2011-11-15 07:53:04 Cory has quit (Excess Flood)
 574 2011-11-15 07:53:24 Cory has joined
 575 2011-11-15 07:53:28 Cory is now known as Guest22306
 576 2011-11-15 07:53:51 Guest22306 is now known as Cory
 577 2011-11-15 07:53:57 Cory has quit (Changing host)
 578 2011-11-15 07:53:57 Cory has joined
 579 2011-11-15 07:54:56 larsivi has joined
 580 2011-11-15 07:56:17 Nesetalis has quit (Read error: Connection reset by peer)
 581 2011-11-15 07:56:31 Nesetalis has joined
 582 2011-11-15 07:57:36 nmat has quit (Ping timeout: 265 seconds)
 583 2011-11-15 07:58:41 gronager has joined
 584 2011-11-15 07:59:50 agath has joined
 585 2011-11-15 08:03:05 HaltingState has quit (Quit: Leaving)
 586 2011-11-15 08:04:44 HaltingState has joined
 587 2011-11-15 08:05:04 HaltingState has quit (Client Quit)
 588 2011-11-15 08:05:28 HaltingState has joined
 589 2011-11-15 08:05:36 larsivi has quit (Read error: No route to host)
 590 2011-11-15 08:07:09 HaltingState has quit (Client Quit)
 591 2011-11-15 08:09:45 larsivi has joined
 592 2011-11-15 08:15:02 DaQatz has joined
 593 2011-11-15 08:15:39 HaltingState has joined
 594 2011-11-15 08:25:36 RazielZ has quit (Ping timeout: 252 seconds)
 595 2011-11-15 08:27:00 toggles has quit (Ping timeout: 258 seconds)
 596 2011-11-15 08:27:21 toggles has joined
 597 2011-11-15 08:30:34 AStove has quit ()
 598 2011-11-15 08:32:35 erus` has joined
 599 2011-11-15 08:46:17 magn3ts has quit (Quit: Leaving)
 600 2011-11-15 08:49:14 sytse has quit (Ping timeout: 240 seconds)
 601 2011-11-15 08:50:08 sytse has joined
 602 2011-11-15 09:02:04 chrisb__ has joined
 603 2011-11-15 09:06:32 ThomasV has quit (Ping timeout: 256 seconds)
 604 2011-11-15 09:13:57 abragin has joined
 605 2011-11-15 09:20:54 darrob has joined
 606 2011-11-15 09:21:01 chrisb__ has quit (Quit: Ex-Chat)
 607 2011-11-15 09:21:13 sgstair has quit (Read error: Connection reset by peer)
 608 2011-11-15 09:21:41 sgstair has joined
 609 2011-11-15 09:22:13 Rav3nSw0rd has quit (Read error: Connection reset by peer)
 610 2011-11-15 09:22:17 maqr has quit (Read error: Connection reset by peer)
 611 2011-11-15 09:22:19 Rav3nSw0rd has joined
 612 2011-11-15 09:22:38 maqr has joined
 613 2011-11-15 09:25:38 copumpkin has quit (Ping timeout: 240 seconds)
 614 2011-11-15 09:26:15 copumpkin has joined
 615 2011-11-15 09:29:31 iocor has joined
 616 2011-11-15 09:35:00 rcorreia has quit (Quit: No Ping reply in 180 seconds.)
 617 2011-11-15 09:35:13 rcorreia has joined
 618 2011-11-15 09:36:12 abragin has quit (Ping timeout: 260 seconds)
 619 2011-11-15 09:38:49 plutonic has joined
 620 2011-11-15 09:41:06 abragin has joined
 621 2011-11-15 09:41:06 abragin has quit (Changing host)
 622 2011-11-15 09:41:06 abragin has joined
 623 2011-11-15 09:44:23 Nexus7 has joined
 624 2011-11-15 09:46:07 Nexus7 has left ()
 625 2011-11-15 09:55:20 karnac has joined
 626 2011-11-15 09:55:22 karnac has quit (Remote host closed the connection)
 627 2011-11-15 09:55:25 iocor has quit (Quit: Computer has gone to sleep.)
 628 2011-11-15 09:58:21 pickett has quit (Ping timeout: 248 seconds)
 629 2011-11-15 09:59:07 pickett has joined
 630 2011-11-15 10:01:21 ThomasV has joined
 631 2011-11-15 10:11:16 pickett has quit (Remote host closed the connection)
 632 2011-11-15 10:12:02 copumpkin has quit (Ping timeout: 258 seconds)
 633 2011-11-15 10:12:29 copumpkin has joined
 634 2011-11-15 10:13:04 aga has joined
 635 2011-11-15 10:16:50 agath has quit (Ping timeout: 240 seconds)
 636 2011-11-15 10:21:54 wboy1 has quit (Ping timeout: 260 seconds)
 637 2011-11-15 10:22:28 wboy1 has joined
 638 2011-11-15 10:28:19 plutonic has quit (Remote host closed the connection)
 639 2011-11-15 10:28:37 plutonic has joined
 640 2011-11-15 10:40:08 larsivi has quit (Ping timeout: 252 seconds)
 641 2011-11-15 10:40:17 wasabi2 has joined
 642 2011-11-15 10:42:11 <gmaxwell> oh my god
 643 2011-11-15 10:42:18 wasabi has quit (Ping timeout: 256 seconds)
 644 2011-11-15 10:42:25 <gmaxwell> slashdot headline on the msft red ballons paper
 645 2011-11-15 10:42:26 <gmaxwell> "Researchers Locate Flaw In Bitcoin Protocol"
 646 2011-11-15 10:43:57 <coderrr> lol
 647 2011-11-15 10:44:01 <coderrr> shit
 648 2011-11-15 10:44:09 <coderrr> should i add a note to my post
 649 2011-11-15 10:55:25 devrandom has quit (Ping timeout: 248 seconds)
 650 2011-11-15 10:57:12 <edcba> what ?!
 651 2011-11-15 11:00:07 <edcba> potential flaw in summary
 652 2011-11-15 11:00:23 <edcba> i guess when i read the paper it will be some stupid assumption
 653 2011-11-15 11:00:37 <edcba> "The flaw pointed out in the paper is that there is a negative incentive for miners to forward Bitcoin transactions"
 654 2011-11-15 11:00:40 <edcba> ok
 655 2011-11-15 11:01:37 <gmaxwell> edcba: thats not something new in the paper though, it's been rehashed many times.
 656 2011-11-15 11:01:42 <edcba> solution no fee
 657 2011-11-15 11:01:55 <edcba> of course it still costs bw to miners
 658 2011-11-15 11:02:13 <edcba> but p2p part of bitcoin still sux
 659 2011-11-15 11:02:18 <gmaxwell> The people who wrote the paper didn't claim that was new.
 660 2011-11-15 11:02:41 <coderrr> edcba, let me know if you want me to change someting
 661 2011-11-15 11:02:49 devrandom has joined
 662 2011-11-15 11:03:04 <edcba> i didn't read the whole thing yet
 663 2011-11-15 11:03:19 <edcba> but now it's on /. i guess it's a bit late to redact many thing :)
 664 2011-11-15 11:03:54 MobiusL has quit (Quit: Leaving)
 665 2011-11-15 11:05:27 <edcba> i don't yet understand the proposed solution :)
 666 2011-11-15 11:05:50 <Edward_Black> edcba actually, the fee mechanic of BTC is due to long-term economical plans for subsidy removal, and forcing miners to subsist on fees
 667 2011-11-15 11:06:15 <edcba> yes but i still don't know if it's a good idea
 668 2011-11-15 11:06:18 <gmaxwell> They hardly describe it— what they're talking about is how you should figure out the rewards in a system which incentivizes forwarding by paying forwarders. They don't talk about how to construct such a system.
 669 2011-11-15 11:06:24 <Edward_Black> Nobody knows edcba
 670 2011-11-15 11:06:26 <edcba> running the system should be enough of a sensitive
 671 2011-11-15 11:06:33 <Edward_Black> Bitcoin is experiment in economics
 672 2011-11-15 11:06:40 <edcba> you have money in it you want it to subsist
 673 2011-11-15 11:06:50 <Edward_Black> I am a buy and send user
 674 2011-11-15 11:07:38 <edcba> having a chain of forwarder quite restrict the anonimity
 675 2011-11-15 11:07:52 * edcba should read the pdf
 676 2011-11-15 11:08:24 da2ce7 has quit (Ping timeout: 245 seconds)
 677 2011-11-15 11:08:37 <Edward_Black> Actually, edcba, there are alt-chains where this "problem" is not prominent at all ;~) but they have other...issues
 678 2011-11-15 11:08:41 <gmaxwell> edcba: if you cared about anonymity you'd use something like tor to get it to the first hop regardless.
 679 2011-11-15 11:08:48 MobiusL has joined
 680 2011-11-15 11:10:32 <Edward_Black> Well, given that only miners are disincentivized from forwarding tx, and given that supermajority of IRL btc miners are, like, 8 pools, as long as some of your contacts are not pool nodes (and AFAIK non-pool nodes vastly outnumber pool nodes, so it is virtually guaranteed)  the issue presented in the paper is completely hypothetical
 681 2011-11-15 11:10:46 <edcba> that pdf is unreadable :)
 682 2011-11-15 11:11:03 <coderrr> edcba, reason for the summary :P
 683 2011-11-15 11:11:08 <Edward_Black> It would be a severe issue for a "spherical bitcoin in a vacuum operating as Satoshi intended"
 684 2011-11-15 11:11:11 <edcba> yes thank you :)
 685 2011-11-15 11:11:14 Shaded has quit (Quit: Shaded)
 686 2011-11-15 11:11:37 <gmaxwell> Edward_Black: right. And if there really were a great many mining nodes.. you can solve the problem by just connecting to a bunch to give them your transaction directly.
 687 2011-11-15 11:11:41 da2ce7 has joined
 688 2011-11-15 11:11:44 <gmaxwell> Or even paying a hub service, I suppose.
 689 2011-11-15 11:11:50 <Edward_Black> yeah
 690 2011-11-15 11:11:55 <Edward_Black> pretty much a non-issue
 691 2011-11-15 11:12:01 <coderrr> maybe i should have used their words, "incentive problem" rather than "flaw"
 692 2011-11-15 11:12:19 <gmaxwell> You think?
 693 2011-11-15 11:12:27 <MacRohard> perhaps if there were incentives to distribute transactions it would incentivise someone to setup transaction distribution systems outside of the p2p network
 694 2011-11-15 11:12:32 <edcba> miners should pay to have tx forwarded to them
 695 2011-11-15 11:12:34 <edcba> that's all
 696 2011-11-15 11:12:42 <edcba> no need for it to be in bitcoin protocol
 697 2011-11-15 11:12:44 <Edward_Black> well, if most nodes are miners, you don't care that much about transaction distribution
 698 2011-11-15 11:12:55 <gmaxwell> MacRohard: you don't have to bake incentivies into the protocol to have them. baking them in just makes them less flexible.
 699 2011-11-15 11:13:01 <Edward_Black> If minority of nodes are miners, then incentive problem is moot]
 700 2011-11-15 11:13:10 <gmaxwell> MacRohard: miners could already pay people who give them fee bearing txns.
 701 2011-11-15 11:13:49 <Edward_Black> Hehe, gmaxwell, that would virtually guarantee extinction of non-fee bearing transactions
 702 2011-11-15 11:14:01 <gmaxwell> In order to bake in the rewards like they describes you must do a very expensive chain of signatures on transactions as part of forwarding them.
 703 2011-11-15 11:14:13 <gmaxwell> Edward_Black: ha, nah. they'd only give a fraction of the fee.
 704 2011-11-15 11:14:22 <MacRohard> gmaxwell, hmm.. yeah.
 705 2011-11-15 11:14:23 <Edward_Black> Miners are notoriously greedy
 706 2011-11-15 11:14:37 <MacRohard> miners could offer volume discounts for large customers
 707 2011-11-15 11:15:04 <gmaxwell> MacRohard: eligius gets datacenter space from mtgox in exchange for doing free txn processing for them. ::shrugs::
 708 2011-11-15 11:15:14 <Edward_Black> (that is not intended as any anti-miner escapade. I too would be greedy if I was operating with a profit margin that thin)
 709 2011-11-15 11:15:43 <Edward_Black> good thing my CPU-coin mining setup costs me 0 in equipment and 0 in electricity
 710 2011-11-15 11:15:48 <gmaxwell> Edward_Black: you're projecting. I've seen basically no evidence of miners being especially greedy wrt the network.
 711 2011-11-15 11:15:59 <MacRohard> gmaxwell, yea.. i think you've made a good point.. this doesn't need to be done in protocol at all.
 712 2011-11-15 11:16:18 amiller has quit (Remote host closed the connection)
 713 2011-11-15 11:17:00 da2ce7 has quit (Remote host closed the connection)
 714 2011-11-15 11:17:05 <iddo> mtgox relies on eligius generating the block? how would that guarantee 10min ?
 715 2011-11-15 11:17:22 <gmaxwell> iddo: you're really great at putting words in my mouth!
 716 2011-11-15 11:17:53 <MacRohard> iddo, it doesn't.. but it allows them to make 'free' (nofee) transactions i guess.
 717 2011-11-15 11:17:54 amiller has joined
 718 2011-11-15 11:18:14 <gmaxwell> iddo: mtgox can pick transactions for eligius to process w/ priority for free. That doesn't magically deny them the ability to get mined by everyone else too.
 719 2011-11-15 11:18:19 <MacRohard> by paying for them out of band with data hosting instead of inband with fees
 720 2011-11-15 11:18:28 <gmaxwell> MacRohard: more importantly, it allows them to unstick txn being paid _to_ them.
 721 2011-11-15 11:18:57 <MacRohard> gmaxwell, oh? so this mining group knows all the incoming mtgox addrs?
 722 2011-11-15 11:19:17 <iddo> so the point is that if other miners don't process the no-fee txns then eligius will eventually process them?
 723 2011-11-15 11:19:18 <gmaxwell> MacRohard: they give an API to mtgox that mtgox can use to pick txn to approve.
 724 2011-11-15 11:19:30 <Edward_Black> Well, there is relatively few things miners can do about being greedy - as long as there is no code to support behaviors such as incentivizing people who forward "paid" transactions the few things miners can do is, on average (barring exclusive agreements with large market agents ;~) ) 1) sell 2) engage in mild manipulations with timestamps which IIRC were documented in "btc main" (but did little
 725 2011-11-15 11:19:39 da2ce7 has joined
 726 2011-11-15 11:19:39 da2ce7 has quit (Changing host)
 727 2011-11-15 11:19:39 da2ce7 has joined
 728 2011-11-15 11:19:44 <MacRohard> gmaxwell, interesting
 729 2011-11-15 11:20:14 <gmaxwell> iddo: right. I suppose. I'm not a party to the arrangement.
 730 2011-11-15 11:20:31 <gmaxwell> I'm only aware of it because its the reason why eligus mined those mtgox money burning txn.
 731 2011-11-15 11:20:47 <gmaxwell> (which otherwise wouldn't have been permitted by the eligius forwarding policy)
 732 2011-11-15 11:21:27 <gmaxwell> It sounds like a reasonable and fair thing to me— and I'm glad that some of the pools care enough to bother creating such an arrangement. :)
 733 2011-11-15 11:21:41 <edcba> in fact that problem looks like same than peering on internet
 734 2011-11-15 11:22:01 <gmaxwell> edcba: yes, though I wouldn't brag about that.
 735 2011-11-15 11:22:09 <edcba> lol
 736 2011-11-15 11:22:10 <gmaxwell> "I'll only peer with miners the same size as me!"
 737 2011-11-15 11:22:21 plutonic has quit (Quit: plutonic)
 738 2011-11-15 11:22:41 <Edward_Black> Well, peering has the distinct difference that it is far less opaque and far more regulated.
 739 2011-11-15 11:22:42 * edcba will peer only with free peering policy :)
 740 2011-11-15 11:23:40 <Edward_Black> Arrangements between large miners and large market players, as well as between large miners and large miners, are completely unregulated and have potential to become completely opaque
 741 2011-11-15 11:23:56 <gmaxwell> are already opaque
 742 2011-11-15 11:24:24 <Edward_Black> Well, I think pool ops might bother to talk to us simple mortals from time to time now.
 743 2011-11-15 11:24:39 <gmaxwell> Edward_Black: luke does, he's basically the only one that does.
 744 2011-11-15 11:24:57 <Edward_Black> If BTC "takes off hard", large pools will be as open and friendly as Microsoft, if not less so
 745 2011-11-15 11:25:26 <gmaxwell> Thats rubbish.
 746 2011-11-15 11:25:31 <gmaxwell> pools don't have any fantastic lock in.
 747 2011-11-15 11:25:36 <edcba> mtgox allowing dark pools anyway...
 748 2011-11-15 11:25:43 <gmaxwell> 0_o
 749 2011-11-15 11:25:53 <Edward_Black> I thought they dropped the dark pool thing in favor of iceberg mechanic
 750 2011-11-15 11:25:54 <edcba> err dark dunno remember
 751 2011-11-15 11:25:56 RazielZ has joined
 752 2011-11-15 11:26:01 <edcba> hmm
 753 2011-11-15 11:26:44 <gmaxwell> Edward_Black: which of the altchains are you high muckymucky for?
 754 2011-11-15 11:27:06 <Edward_Black> Well, I am the unofficial econ adviser to Tenebrix
 755 2011-11-15 11:27:35 * edcba hungry
 756 2011-11-15 11:27:43 <Edward_Black> I also mine it with my humble work severs (currently at less than 60% perfomance due to issues of massive infrastructure overhaul)
 757 2011-11-15 11:28:14 <iddo> you advised to have constant monetary inflation?
 758 2011-11-15 11:28:25 <Graet> Edward_Black gmaxwell quite a few pool ops are in irc and talk to ppl
 759 2011-11-15 11:28:32 <Edward_Black> I advised to attempt "alternative deflation mechanics" iddo
 760 2011-11-15 11:28:56 <Edward_Black> which aren't currently implemented due to ArtForz being busy and few people being interested to take stab at relevant bounty
 761 2011-11-15 11:29:24 <iddo> how could you implement without risking chain fork?
 762 2011-11-15 11:29:30 <Edward_Black> With no offence intended to pools (please don't shoot my kneecaps guys!) the position is a cartel paradise, with the only thing standing between them and "gilded age 2.0"  being BTC's relatively meh market perfomance and relatively nice personal character of pool ops
 763 2011-11-15 11:29:48 <Edward_Black> iddo the same way you would implement any potentially chain-forking change
 764 2011-11-15 11:29:54 <Edward_Black> I don't recall the right name
 765 2011-11-15 11:30:15 <Edward_Black> But basically change is activated at a given block count
 766 2011-11-15 11:30:30 <iddo> Edward_Black: yeah but monetary policy changes are controversial... unlike technical improvements... you might not get consensus
 767 2011-11-15 11:30:53 <Edward_Black> iddo well, so far the change has not raised a single complaint from any person interested in TBX
 768 2011-11-15 11:31:06 <Edward_Black> And the proposal was quite openly posted
 769 2011-11-15 11:31:18 <iddo> what is the proposal?
 770 2011-11-15 11:31:27 <Edward_Black> Let me find you the link
 771 2011-11-15 11:31:38 <Edward_Black> Please stand by, phone internet is sucky
 772 2011-11-15 11:32:27 <iddo> i thought tenebrix plans to function as money laundering services, by using the millions of pre-mined coins?
 773 2011-11-15 11:32:58 <iddo> so it'd be a centralized laundering service
 774 2011-11-15 11:33:00 <Edward_Black> iddo https://bitcointalk.org/index.php?topic=48114
 775 2011-11-15 11:33:18 <iddo> thanks
 776 2011-11-15 11:33:37 <Edward_Black> iddo the Strong Transaction Decorrelator is just one of the "built in" services planned, and the only one that is centralized
 777 2011-11-15 11:34:05 <Edward_Black> But I suspect that discussion of TBX intra-blockchain services is somewhat offtopic in the bitcoin-dev channel
 778 2011-11-15 11:34:53 larsivi has joined
 779 2011-11-15 11:35:19 <Edward_Black> iddo feel free to ask about other services proposed (and the Strong Transaction Decorrelator lockdown mechanic, that will ensure that laundry funds can't be stolen / sold off) in #Tenebrix if you are curious, though :~)
 780 2011-11-15 11:36:32 <iddo> so how does deflation work? the coinbase is constant forever as it it, you gonna replace it with geometric decreasing coinbase like bitcoin?
 781 2011-11-15 11:37:11 <Edward_Black> iddo umm, I propose we move to #Tenebrix so as not to draw the ire of residents with our offtop :~)
 782 2011-11-15 11:38:14 <iddo> you don't wish to spread the tenebrix gospel to others?
 783 2011-11-15 11:39:04 <Edward_Black> Tenebrix is not a religion, and I am just afraid to cause discomfort to bitcoin-dev residents who might prefer alt-coin dev discussions to be taken to relevant channels
 784 2011-11-15 11:39:48 <gmaxwell> I'd much prefer direct discussion than things like: 03:01 < Edward_Black> good thing my CPU-coin mining setup costs me 0 in equipment and 0 in electricity (which caused me extreme eyeroll)
 785 2011-11-15 11:40:01 <Edward_Black> lolz ;~)
 786 2011-11-15 11:40:07 <iddo> it's quiet right now, if gavin starts to give some updates on encrypted wallet bug then i will shut up immediately:)
 787 2011-11-15 11:41:27 <Edward_Black> iddo okay, the   TLDR version of the reasoning behind the proposal is thus: TBX subsidy remains static in absolute terms and is constantly shrinking relative to coin mass. However, mandatory transaction fees (which are needed anyways) and intra-blockchain service fees scale with adoption and, to smaller degree, coinbase size
 788 2011-11-15 11:41:54 <iddo> maybe i missed it but forum post only talked about fees, nothing about the new coins constantly generated for each block?
 789 2011-11-15 11:42:28 <Edward_Black> Um, the static subsidy is part of original TBX as conceved prior to my arrival at the project
 790 2011-11-15 11:42:29 <iddo> shrinking relative to coin mass = ?
 791 2011-11-15 11:42:37 <Edward_Black> Um, consider this
 792 2011-11-15 11:42:43 <Edward_Black> You pay out 25 units, forver
 793 2011-11-15 11:43:43 <Edward_Black> so, when total coin mass (number of coins available for sale) is, say, 50 units (hypothetically), the next payout will be whopping 50% relative to the entire coin mass
 794 2011-11-15 11:45:05 <iddo> definition of coin mass is the amount of coins that people selling? selling in exchange for what? dollars? merchandise? services?
 795 2011-11-15 11:45:29 <Edward_Black> no, the amount of coins in circulation, that is, available to miners, users, and whomever else
 796 2011-11-15 11:45:51 <iddo> ok, so what shrinks ?
 797 2011-11-15 11:46:12 <Edward_Black> When subsidy is nominally static (always X coins), it is inevitably shrinking relative to ammount of coins previously generated
 798 2011-11-15 11:46:23 <Edward_Black> in percentile expression
 799 2011-11-15 11:47:23 <iddo> ah by shrink you mean the ratio between latest 25 new coins generated, and total of all coins ever generated, shrinks?
 800 2011-11-15 11:47:29 <Edward_Black> yep
 801 2011-11-15 11:48:04 <Edward_Black> Now factor in that per new mechanic, intra-blockchain service fees and mandatory part of any transaction fee are destroyed, and factor in coin loss
 802 2011-11-15 11:48:06 <iddo> by that definition the federal reserves is shrinking the dollar
 803 2011-11-15 11:48:26 <iddo> s/reserves/reserve
 804 2011-11-15 11:48:31 <Edward_Black> Well, technically, you can't even apply definition of inflation to cryptocoins since they aren't a currency
 805 2011-11-15 11:48:45 <Edward_Black> So any fedreserve analogy is kinda moot
 806 2011-11-15 11:49:30 <Edward_Black> Also, inflation/deflation is usually interpreted as price change, not monetary mass change, and in case of coin the econ of service / good payment is kinda screwed up
 807 2011-11-15 11:50:28 <Edward_Black> Technically, the rate at which gold, silver and IIRC platinum enter the market is increasing in nominal terms, but we don't call them inflatory, do we ? ;~)
 808 2011-11-15 11:50:29 <iddo> all i can see is that if someone wanna sell product for x tenebrix then he know the x he receives get diluted with each new block generated, constantly forever...
 809 2011-11-15 11:50:44 <gmaxwell> iddo: key point there is that he knows it.
 810 2011-11-15 11:51:00 <gmaxwell> which is still better than the deal you get with politically controlled fia.
 811 2011-11-15 11:51:03 <gmaxwell> er fiat.
 812 2011-11-15 11:51:38 <iddo> also still means the value money he receives tends towards zero
 813 2011-11-15 11:51:52 <iddo> s/value/value of
 814 2011-11-15 11:51:54 <Edward_Black> iddo um, no, since 1) minimum case scenario is that at some point, number of coins lost to fee burning and vandalism (which all scale with adoption and "overall coin mass" per month is equal to that generated by miners per month (which is static forever)
 815 2011-11-15 11:52:26 <Edward_Black> and iddo please stop thinking of "crypto-currencies" as money. They are very much speculative goods with unusual consumption patterns
 816 2011-11-15 11:53:19 <Edward_Black> now, from the moment the N lost to fee furnace, intra bc service fees and vandalism / carelessness = N generated and onwards
 817 2011-11-15 11:53:24 <iddo> do you also have some convincing reasons why constant inflation is supposed to be good? compared to gold or bitcoin etc.
 818 2011-11-15 11:53:35 <gmaxwell> Edward_Black: ah, your altchain burns fees?
 819 2011-11-15 11:54:01 <Edward_Black> gmaxwell technically it is not mine (I merely am an advisory voice) and it will start burning them, yes
 820 2011-11-15 11:54:08 <iddo> burning fee is new proposal for fork that wasn't implemented yet?
 821 2011-11-15 11:54:12 <Edward_Black> yeah
 822 2011-11-15 11:54:41 <gmaxwell> as a miner why would I give higher priority to a txn that burns more fees? whats in it for me?
 823 2011-11-15 11:54:49 <Edward_Black> iddo for starters, new gold "output" is steadily, if slowly, increasing  at an accelerating rate
 824 2011-11-15 11:55:22 <iddo> ok so consider bitcoin instead of gold
 825 2011-11-15 11:55:23 <Edward_Black> gmaxwell well, the proposal linked above deals with how to implement miner "tipping" for priority while still having fee burning
 826 2011-11-15 11:55:55 <gmaxwell> ah, I'd missed a bunch of the conversation in the scrollback. Okay.
 827 2011-11-15 11:57:19 <Edward_Black> iddo I sorta like bitcoin's general idea of "let's manipulate supply to create a percieved value increase", but I am mildly concerned about the raw subsidy cut mechanic, because it does not account for 1) possibility of market failing to properly adjust to new "miner-base" price causing a miner "die off" and a spiral of doom 2) immense potential for encouraging cartelization even in a best-case
 828 2011-11-15 11:57:27 <Edward_Black> ]"miner can subsist on fee" scenario
 829 2011-11-15 11:57:42 <Edward_Black> 3) I generally do not believe in miners subsisting on fees
 830 2011-11-15 11:57:58 <Edward_Black> Bitcoin's experiment in economics is interesting and important
 831 2011-11-15 11:57:59 <iddo> cartelization why?
 832 2011-11-15 11:58:27 <Edward_Black> cartelization because you have a vital interest in driving fees as high as market would support
 833 2011-11-15 11:58:37 <Edward_Black> Which is exactly what cartels were invented for
 834 2011-11-15 11:58:45 <gmaxwell> Edward_Black: one trade off there is that the system doesn't tie the reward for mining with the valuation of the currency units, they can each find their own efficient prices.
 835 2011-11-15 11:58:59 <gmaxwell> vs. tenebrix where they're married for better or worse.
 836 2011-11-15 11:59:09 <gmaxwell> But it'll be interesting to see how it goes.
 837 2011-11-15 12:00:06 <iddo> yes i don't claim it isn't interesting... but i think constant inflation is a bad idea, we'll see
 838 2011-11-15 12:00:09 <Edward_Black> The thing is, the succesful establishment of an efficient price that does not cause miner die-off is not guaranteed, and sadly miner die-off is something we don't want because less "hardcore" miners = more vulnerable chain (and a bunch of pissed d00ds with some serious hashrate)
 839 2011-11-15 12:01:24 <Edward_Black> iddo 1) nothing that happens in x-coins is proper inflation 2) constant increase in coin mass is offset by more aggressive coin loss 3) factors that cause coin loss scale up with adoption while subsidy remains static per nominal assesment, and scales down per relative assesment
 840 2011-11-15 12:01:27 <gmaxwell> Edward_Black: yes thats true. The market can decide to prevent that though... but they don't have to.
 841 2011-11-15 12:02:27 <iddo> maybe i missed something, what causes agressive coin loss? just the burning fees?
 842 2011-11-15 12:02:30 <gmaxwell> It's not clear that tenebrix inflate model becomes that.. the value of the currency units could still go low enough where 50 is not a meaningful support
 843 2011-11-15 12:03:58 <Edward_Black> iddo mandatory part of transaction fees won't be the only fees that are burned. All intra-blockchain services (destination-agnostic communication conceived by ArtForz specifically) will run on burned coins too
 844 2011-11-15 12:04:44 <iddo> hmm ok so i did miss something(s)...
 845 2011-11-15 12:05:01 <Edward_Black> iddo what you seem not to get that even perfectly natural coin loss (vandalism / negligence / owners dying etc.) INEVITABLY reaches a point where it balances out coins added by static-subsidy mining
 846 2011-11-15 12:05:08 iocor has joined
 847 2011-11-15 12:05:11 iocor has quit (Changing host)
 848 2011-11-15 12:05:11 iocor has joined
 849 2011-11-15 12:05:18 <Edward_Black> fee-burning just makes it happen FASTER and at a lower "coin mass" value
 850 2011-11-15 12:05:37 <iddo> i wonder why you choose to do this economic experiment with chain that appears to be a scam because of millions of pre-mined coins controlled by a single person
 851 2011-11-15 12:05:39 <gmaxwell> "destination-agnostic communication" is also something that namecoin provides but of a very specialized type.
 852 2011-11-15 12:06:26 <Edward_Black> gmaxwell the thing Art came up with is far more frugal in regards to the amount  of data put into blockchain :~)
 853 2011-11-15 12:06:50 slush has quit (Quit: Leaving.)
 854 2011-11-15 12:07:01 <Edward_Black> iddo because I am deeply profoundly in love with the "laundry with million coins as a buffer" idea
 855 2011-11-15 12:07:34 <Edward_Black> iddo so much in love that I bothered to procure Lolcust's email so I could offer my advice in most direct manner possible
 856 2011-11-15 12:07:47 <iddo> then you should think of how to implement decentralized laundry? instead of giving total control to single person?
 857 2011-11-15 12:09:08 <iddo> sorry i have to go... l8r
 858 2011-11-15 12:09:21 <Edward_Black> iddo well, I am not that 1337 a cryptographer (I am merely a lucky sysop ;~) ) and apparent lack of a decentralized laundry proposal that would be practical suggests that it is truly a tast for the most awesome pros (alternatively, that it is an untenable task like squaring the circle)
 859 2011-11-15 12:09:45 <epscy> what do you guys make of the microsoft research about miners incentive to relay nodes?
 860 2011-11-15 12:10:03 <epscy> relay transactions sorry
 861 2011-11-15 12:13:05 <Edward_Black> iddo I am not as allergic to centralization as some of respected locals - in fact, I do think that some, mild, centralization is nice if the benefits provided (laundry that does not need a website, w00t!) outweigh the risks (with laundry-lockdown code in place, the only risk is lolcust's laundry being found despite all the anonymization he and Art have planned, which would disable the service)
 862 2011-11-15 12:13:31 <Edward_Black> After all, there is a reason why even insects have their nervous system somewhat cetralized into ganglions
 863 2011-11-15 12:14:27 <Edward_Black> So trading a bit of decentralization for ability to carry out far more anonymous transactions with a punch of a button seems like a decent bargain to me
 864 2011-11-15 12:15:30 <Edward_Black> After all, Tenebrix is just another experiment  - in both economics and intra-blockchain service provision. So its success or lack thereoff isn't gonna take any skin of anyone's nose
 865 2011-11-15 12:17:23 <Edward_Black> frankly, epscy, I do not think it is a major concern in current bitcoin arrangement since supermajority of nodes aren't miners and thus aren't disincentivized from propagating transactions
 866 2011-11-15 12:21:50 <sipa> gmaxwell: what timezone are you in; i seem to see you here almost any time of the day? :)
 867 2011-11-15 12:24:54 danbri has joined
 868 2011-11-15 12:26:42 Beremat has joined
 869 2011-11-15 12:27:50 Acciaio has joined
 870 2011-11-15 12:28:05 <Acciaio> I all how can I start to use bitcoin?
 871 2011-11-15 12:28:32 <Acciaio> I have downloaded the client from bitcoin.org
 872 2011-11-15 12:29:12 <Acciaio> but when I start binary in bitcoin-0.4.0/bin/64/bitcoin it doesn't do anything
 873 2011-11-15 12:29:17 <gmaxwell> sipa: it changes now and again.
 874 2011-11-15 12:29:31 marf_away has joined
 875 2011-11-15 12:29:39 <Edward_Black> Acciaio, nothing at all, as  in no window showing up or so ?
 876 2011-11-15 12:30:09 <Acciaio> no window showed and cannot connect to server
 877 2011-11-15 12:30:25 <Acciaio> it create a .bitcoin folder in my home
 878 2011-11-15 12:31:28 <Edward_Black> Hmmmmmm... I suggest that you paste your debug log (check that .bitcoin folder) to pastebin and link here
 879 2011-11-15 12:31:32 <Acciaio> I have set rpcpassword in bitcoin.conf but the answer is the same can't connect to server
 880 2011-11-15 12:32:17 <Edward_Black> w w wait, you are using bitcoin daemon or the gui client ? And what linux distro are you running ?
 881 2011-11-15 12:32:42 <Edward_Black> also, did you try the 32 bit version ?
 882 2011-11-15 12:33:49 <Acciaio> http://pastebin.com/6dWP6eSH
 883 2011-11-15 12:34:21 <Acciaio> Edward_Black, I have tried both daemond and gui client
 884 2011-11-15 12:34:43 <Acciaio> also tried to start gui client with --server option
 885 2011-11-15 12:34:47 <Edward_Black> Um, what about 32 bit version, did you try it ?
 886 2011-11-15 12:34:58 <Acciaio> I'm using 64bnit version and ubuntu 10.04
 887 2011-11-15 12:35:49 <Edward_Black> try 32 bit one just in case
 888 2011-11-15 12:36:07 <Acciaio> another thing the only way to kill the process is with kill -9
 889 2011-11-15 12:36:18 Beremat has quit (Read error: Connection reset by peer)
 890 2011-11-15 12:36:28 wboy1 has quit (Ping timeout: 260 seconds)
 891 2011-11-15 12:38:04 <Edward_Black> hm. Try 32 bit thing., if no help, check your network (I generally assume you aren't running some massively restrictive firelwall config that meddles with outgoing connections )
 892 2011-11-15 12:38:36 <gmaxwell> I can't see a network issue causing the client not to show up.
 893 2011-11-15 12:38:53 <gmaxwell> Acciaio: is this a new install?
 894 2011-11-15 12:39:16 <gmaxwell> Personally I'd just blow away the .bitcoin directory (saving the wallet.dat unless there was nothing to save)
 895 2011-11-15 12:39:29 <gmaxwell> if its really hanging up there.
 896 2011-11-15 12:39:30 <Acciaio> gmaxwell, yes is a new install
 897 2011-11-15 12:40:29 <Acciaio> I have no outgoing policy
 898 2011-11-15 12:40:34 <Acciaio> in my firewall
 899 2011-11-15 12:40:50 <sipa> that debug.log looks fine
 900 2011-11-15 12:41:06 <sipa> there were issues with certain versions of ubuntu being not compatible with the wx gui
 901 2011-11-15 12:41:28 <sipa> (though not 10.04, afaik(
 902 2011-11-15 12:42:11 <Acciaio> but bitcoin-0.3.23 was working well
 903 2011-11-15 12:42:32 <Acciaio> gui and daemon
 904 2011-11-15 12:43:24 <Edward_Black> Isn't 0.4 + QT-based ?
 905 2011-11-15 12:43:38 <Edward_Black> (re: possible wx woes ^^)
 906 2011-11-15 12:44:29 <sipa> Edward_Black: 0.5 is
 907 2011-11-15 12:44:34 <Edward_Black> Ah, my bad
 908 2011-11-15 12:45:01 <Edward_Black> Acciaio try deleting the .bitcoin folder and running 32 bit version, if that fails, try moving to 0.5 bitcoin :~)
 909 2011-11-15 12:45:55 wboy1 has joined
 910 2011-11-15 12:46:48 <sipa> 0.5 hasn't been released yet :)
 911 2011-11-15 12:47:14 <Acciaio> where can I download 0.5 version?
 912 2011-11-15 12:48:47 <gmaxwell> Acciaio: careful about deleting the folder if you have any coins. be sure to save the wallet.dat if you do.
 913 2011-11-15 12:49:39 <sipa> you could try 0.5rc3, but it's only a release candidate, and there are some known issues with it
 914 2011-11-15 12:49:47 <Acciaio> gmaxwell, I know
 915 2011-11-15 12:50:14 <Acciaio> just backupped my wallet.dat in an usb storage
 916 2011-11-15 12:50:15 <Acciaio> :-D
 917 2011-11-15 12:51:54 <sipa> Acciaio: http://sourceforge.net/projects/bitcoin/files/Bitcoin/bitcoin-0.5.0/test/
 918 2011-11-15 12:53:24 <Acciaio> I'm pastebinning strace of last the lastest execution after deleting .bitcoin folder and with 32bit version
 919 2011-11-15 12:56:12 <ThomasV> Mqrius: thanks for the excutable. I decided to link to the forum thread
 920 2011-11-15 12:56:59 <Acciaio> http://pastebin.com/zxAQXKHc after that there is no update in strace output
 921 2011-11-15 12:57:31 <gmaxwell> need to strace -f because there are multiple threads.
 922 2011-11-15 12:57:42 <Acciaio> ups sorry
 923 2011-11-15 12:58:20 <gmaxwell> but .. hm.. is it getting stuck on dnsseed lookups?
 924 2011-11-15 13:00:26 ahbritto__ has joined
 925 2011-11-15 13:00:59 SomeoneWeird is now known as SomeoneWeirdAFK
 926 2011-11-15 13:02:03 ahbritto_ has quit (Ping timeout: 240 seconds)
 927 2011-11-15 13:02:24 ahbritto has quit (Ping timeout: 255 seconds)
 928 2011-11-15 13:02:26 ahbritto_ has joined
 929 2011-11-15 13:03:08 ahbritto__ has quit (Read error: Connection reset by peer)
 930 2011-11-15 13:03:23 <Acciaio> ?
 931 2011-11-15 13:03:44 <Acciaio> I use opendns server 208.67.222.222
 932 2011-11-15 13:04:03 <Acciaio> http://pastebin.com/Jqw2JmzA
 933 2011-11-15 13:05:01 <gmaxwell> Acciaio: run "dig bitseed.xf2.org"
 934 2011-11-15 13:05:09 <gmaxwell> does it give you a bunch of IPs or?
 935 2011-11-15 13:05:51 <Acciaio> yes
 936 2011-11-15 13:06:51 ahbritto_ has quit (Ping timeout: 244 seconds)
 937 2011-11-15 13:06:52 Sedra has joined
 938 2011-11-15 13:07:30 <Acciaio> http://pastebin.com/cAYNZMHd
 939 2011-11-15 13:09:27 Sedra- has quit (Ping timeout: 244 seconds)
 940 2011-11-15 13:10:31 <gmaxwell> Acciaio: try dig @192.168.0.1 bitseed.xf2.org
 941 2011-11-15 13:12:00 Sedra- has joined
 942 2011-11-15 13:12:07 <Acciaio> http://pastebin.com/8i7HPTZz
 943 2011-11-15 13:13:21 <gmaxwell> odd. well, if I'm not misreading your strace output it looks like its hanging up trying to do a dnsquery on that address.
 944 2011-11-15 13:13:24 <gmaxwell> Acciaio: start with -nodnsseed
 945 2011-11-15 13:13:36 <Mqrius> ThomasV: Okay, that works. If I had a post early in that thread, I'd just update it there, but I don't, really :P
 946 2011-11-15 13:13:49 Sedra has quit (Ping timeout: 252 seconds)
 947 2011-11-15 13:14:30 <ThomasV> well you could make a new thread if you want me to point there
 948 2011-11-15 13:15:16 Sedra has joined
 949 2011-11-15 13:16:00 <Acciaio> ok gmaxwell great it start well now
 950 2011-11-15 13:16:01 <Acciaio> :-D
 951 2011-11-15 13:16:07 <Acciaio> thank u very much
 952 2011-11-15 13:17:50 <gmaxwell> sipa: ^
 953 2011-11-15 13:18:21 Sedra- has quit (Ping timeout: 248 seconds)
 954 2011-11-15 13:18:24 <gmaxwell> Acciaio: Sounds like we have a bug where we get stuck when dnsseed doesn't work right. I don't know _why_ it doesn't work right for you...
 955 2011-11-15 13:18:35 <gmaxwell> probably because resolvers are always screwed up with static binaries...
 956 2011-11-15 13:18:44 ahbritto_ has joined
 957 2011-11-15 13:18:45 <gmaxwell> but the software shouldn't get stuck when it happens, regardless.
 958 2011-11-15 13:20:02 ahbritto__ has joined
 959 2011-11-15 13:20:14 <Acciaio> I can't see this option in --help thank you for your help
 960 2011-11-15 13:21:13 <gmaxwell> Acciaio: it's hidden, presumably because it should never be needed.
 961 2011-11-15 13:21:36 Sedra- has joined
 962 2011-11-15 13:21:48 <gmaxwell> BlueMattBot: tell bluematt that dnsseed appears to have a bug where it can block the client loading, see channel logs
 963 2011-11-15 13:21:48 <BlueMattBot> gmaxwell did you mean me? Unknown command 'tell'
 964 2011-11-15 13:21:48 <BlueMattBot> Use 'BlueMattBot: DONT SPAM the Chan help' to get help!
 965 2011-11-15 13:21:57 <gmaxwell> hah
 966 2011-11-15 13:22:09 <gmaxwell> gribble: tell bluematt that dnsseed appears to have a bug where it can block the client loading, see channel logs
 967 2011-11-15 13:22:16 <gmaxwell> stupid unhelpful bots
 968 2011-11-15 13:22:46 <gmaxwell> Acciaio: can I get an email address from you in order to ask you to test a fix, since you're the only one who has reported this and I don't yet know if I can reproduce it?
 969 2011-11-15 13:22:51 slush has joined
 970 2011-11-15 13:22:58 Sedra has quit (Ping timeout: 260 seconds)
 971 2011-11-15 13:23:20 <Acciaio> yes steal82@gmail.com
 972 2011-11-15 13:24:07 Sedra has joined
 973 2011-11-15 13:24:32 <gmaxwell> Thanks.
 974 2011-11-15 13:26:16 <Acciaio> bye
 975 2011-11-15 13:26:51 Sedra- has quit (Ping timeout: 240 seconds)
 976 2011-11-15 13:27:31 gronager has quit (Quit: Leaving.)
 977 2011-11-15 13:27:59 Sedra- has joined
 978 2011-11-15 13:28:21 wboy1 has quit (Ping timeout: 256 seconds)
 979 2011-11-15 13:28:59 <Acciaio> but now I get 0 connection is ok(I only have to wait) or what can I do?
 980 2011-11-15 13:29:28 <gmaxwell> Acciaio: without dnsseed it can take a bit to get connections..
 981 2011-11-15 13:29:30 <sipa> wait a bit, the IRC seeding should get you some connections soon
 982 2011-11-15 13:29:47 <gmaxwell> unless your IRC seeding isn't connecting either! :)
 983 2011-11-15 13:30:30 Sedra has quit (Ping timeout: 240 seconds)
 984 2011-11-15 13:30:51 <sipa> right... irc seeding also does a dns lookup
 985 2011-11-15 13:31:23 <gmaxwell> yea, so that might get stuck too (but lack the bug that makes it wedge the whole client)
 986 2011-11-15 13:31:51 <gmaxwell> Acciaio: if it's still not connected start up with a couple of -addnode <ipfromthatdigcommand>
 987 2011-11-15 13:32:06 <gmaxwell> e.g. -addnode 190.101.85.148 -addnode 69.164.218.197 -addnode 178.18.90.41
 988 2011-11-15 13:32:12 <gmaxwell> it'll learn more once connected.
 989 2011-11-15 13:32:22 <gmaxwell> unless it can't even connect _at all_ and then I don't know what to say.
 990 2011-11-15 13:32:31 <gmaxwell> other than "fix your computer"
 991 2011-11-15 13:32:33 <gmaxwell> ;)
 992 2011-11-15 13:34:56 wboy1 has joined
 993 2011-11-15 13:35:47 darrob has quit (Disconnected by services)
 994 2011-11-15 13:35:53 darrob has joined
 995 2011-11-15 13:41:31 wasabi has joined
 996 2011-11-15 13:43:25 darrob has quit (Ping timeout: 248 seconds)
 997 2011-11-15 13:43:33 wasabi2 has quit (Ping timeout: 260 seconds)
 998 2011-11-15 13:43:59 <wboy1> Hey Guys,any javascript dev's that are interested in joining a bitcoin related funded startup,drop me a message,Thanks!
 999 2011-11-15 13:44:41 minimoose has joined
1000 2011-11-15 13:48:14 devrandom has quit (Ping timeout: 248 seconds)
1001 2011-11-15 13:48:24 darrob has joined
1002 2011-11-15 13:50:40 wboy1 has quit (Ping timeout: 252 seconds)
1003 2011-11-15 13:52:19 wboy1 has joined
1004 2011-11-15 13:55:16 devrandom has joined
1005 2011-11-15 13:58:21 MC1984 has quit (Read error: Connection reset by peer)
1006 2011-11-15 13:58:43 MC1984 has joined
1007 2011-11-15 13:59:17 copumpkin has quit (Ping timeout: 252 seconds)
1008 2011-11-15 13:59:46 copumpkin has joined
1009 2011-11-15 14:03:00 Mqrius has quit ()
1010 2011-11-15 14:04:43 bitcoinbulletin has quit (Remote host closed the connection)
1011 2011-11-15 14:07:21 MC1984 has quit (Read error: Connection reset by peer)
1012 2011-11-15 14:08:26 MC1984 has joined
1013 2011-11-15 14:11:07 MC1984 has quit (Read error: Connection reset by peer)
1014 2011-11-15 14:12:01 MC1984 has joined
1015 2011-11-15 14:15:01 bitcoinbulletin has joined
1016 2011-11-15 14:18:48 MC1984 has quit (Read error: Connection reset by peer)
1017 2011-11-15 14:19:06 MC1984 has joined
1018 2011-11-15 14:25:13 gavinandresen has joined
1019 2011-11-15 14:32:55 gp5st has joined
1020 2011-11-15 14:35:23 eastender has joined
1021 2011-11-15 14:36:03 <dikidera> omfg at next diff
1022 2011-11-15 14:36:12 <dikidera> 1 mill and 500k
1023 2011-11-15 14:36:20 <dikidera> and it just dropped...
1024 2011-11-15 14:36:54 <dikidera> or not...seems it changed recently
1025 2011-11-15 14:40:57 <Eliel> too radical change is likely a fluke of some kind. Those are estimates, afterall :)
1026 2011-11-15 14:41:12 <UukGoblin> early estimates are always off
1027 2011-11-15 14:41:37 <UukGoblin> but with the current price... I'd expect the diff to go below 1M, actually.
1028 2011-11-15 14:42:01 <UukGoblin> unless perhaps people are fpga'ing heavily
1029 2011-11-15 14:42:17 Turingi has joined
1030 2011-11-15 14:42:17 Turingi has quit (Changing host)
1031 2011-11-15 14:42:17 Turingi has joined
1032 2011-11-15 14:43:10 wasabi2 has joined
1033 2011-11-15 14:43:31 <Eliel> or optimistic that the price will go up soon
1034 2011-11-15 14:43:42 wasabi has quit (Ping timeout: 248 seconds)
1035 2011-11-15 14:44:20 [eval] has joined
1036 2011-11-15 14:45:33 agricocb has quit (Remote host closed the connection)
1037 2011-11-15 14:45:57 <Eliel> the low at $2.04 from the previous crash hasn't been broken so far. If it stays this way, I think it's actually a quite positive sign.
1038 2011-11-15 14:47:50 <gavinandresen> Last chance to comment on the encryption bug PULL-- I'm planning on pulling then tagging a 0.5rc4 and creating binaries for testing.
1039 2011-11-15 14:47:58 <gavinandresen> https://github.com/bitcoin/bitcoin/pull/635
1040 2011-11-15 14:51:42 eastender has quit (Ping timeout: 240 seconds)
1041 2011-11-15 14:53:28 <CIA-89> bitcoin: Pieter Wuille master * r9e9869d / (src/db.cpp src/db.h src/wallet.cpp): Resilvering - http://git.io/o8IJOQ
1042 2011-11-15 14:53:28 <CIA-89> bitcoin: Gavin Andresen master * rd764d91 / (6 files): Obsolete keypool and make sure database removes log files on shutdown. - http://git.io/uKibRw
1043 2011-11-15 14:53:29 <CIA-89> bitcoin: Wladimir J. van der Laan master * r4585f7e / src/qt/askpassphrasedialog.cpp : add message about restarting bitcoin after encrypting wallet succesfully - http://git.io/bGHo-Q
1044 2011-11-15 14:53:29 <CIA-89> bitcoin: Gavin Andresen master * rb6d11a3 / (7 files in 2 dirs):
1045 2011-11-15 14:53:29 <CIA-89> bitcoin: Merge pull request #635 from gavinandresen/encryptionbug
1046 2011-11-15 14:53:29 <CIA-89> bitcoin: Prevent unencrypted private keys from being written to wallet.dat - http://git.io/PL8zOw
1047 2011-11-15 14:59:23 <sipa> gavinandresen:
1048 2011-11-15 14:59:24 <sipa> src/db.cpp:50:42: error: ‘boost::filesystem::basic_path<std::basic_string<char>, boost::filesystem::path_traits>::string_type’ has no member named ‘generic_string’
1049 2011-11-15 14:59:49 <gavinandresen> what?  I just fixed that bug....
1050 2011-11-15 15:00:31 <gavinandresen> sipa: what version of boost do you have installed?
1051 2011-11-15 15:00:31 <sipa> (current master, after your merge)
1052 2011-11-15 15:00:53 <sipa> 1.42
1053 2011-11-15 15:01:11 <gavinandresen> ok, I've got to reverse the logic in the #if BOOST_FILESYSTEM_VERSION blah blah blah
1054 2011-11-15 15:01:41 <luke-jr> FYI: CVE-2011-4447 has been assigned the the wallet-contains-plaintext-privkeys issue
1055 2011-11-15 15:01:54 zapnap has joined
1056 2011-11-15 15:02:16 <CIA-89> bitcoin: Gavin Andresen master * r709c1b2 / src/db.cpp : Fix boost filesystem incompatibility problem - http://git.io/BOvKFA
1057 2011-11-15 15:03:15 darrob has left ("WeeChat 0.3.5")
1058 2011-11-15 15:03:39 <luke-jr> … gavinandresen
1059 2011-11-15 15:03:43 <luke-jr> that change just BREAKS it
1060 2011-11-15 15:03:45 chrisb__ has joined
1061 2011-11-15 15:03:53 <luke-jr> http://www.boost.org/doc/libs/1_47_0/libs/filesystem/v3/doc/v3.html\
1062 2011-11-15 15:04:37 <gavinandresen> can you be more specific on how it breaks it?
1063 2011-11-15 15:05:02 <gavinandresen> In my test tree, I'm compiling bitcoind with -DBOOST_FILESYSTEM=2 and bitcoin-qt with =3 ....
1064 2011-11-15 15:05:04 dvide has joined
1065 2011-11-15 15:05:32 <gavinandresen> ... so it should compile either way.
1066 2011-11-15 15:05:42 <da2ce7> Diablo-D3, hey, can you please help me re-factor out this bit of java code: https://github.com/da2ce7/Moneychanger/blob/master/src/com/wrapper/ui/Load.java#L54-59
1067 2011-11-15 15:05:47 <luke-jr> gavinandresen: you're not really supposed to define it
1068 2011-11-15 15:06:05 <luke-jr> gavinandresen: BOOST_FILE_SYSTEM=3 is the new API used by the top part
1069 2011-11-15 15:06:11 <luke-jr> gavinandresen: 2 is the old API used by the bottom
1070 2011-11-15 15:06:19 <luke-jr> Boost headers define the correct one
1071 2011-11-15 15:06:21 <gavinandresen> ... right.... I'm only defining it so I can test without having two version of boost installed....
1072 2011-11-15 15:06:41 <gavinandresen> Did I accidently check in a makefile where I define it?
1073 2011-11-15 15:07:02 <luke-jr> no, you changed the #if to check for the wrong version :P
1074 2011-11-15 15:07:15 <luke-jr> oh wait, I see what you did… you switched the sides
1075 2011-11-15 15:07:44 <luke-jr> I'm not sure all the oldest boost versions define it at all
1076 2011-11-15 15:07:54 <gavinandresen> right....
1077 2011-11-15 15:07:55 <luke-jr> whereas all the newer ones supporting v3 *should*
1078 2011-11-15 15:08:01 <gavinandresen> yes....
1079 2011-11-15 15:09:20 <gavinandresen> bitcoin-qt.pro does define BOOST_FILESYSTEM if on macx....  I'll fix that after I get the rc4 builds going
1080 2011-11-15 15:10:43 <luke-jr> my point is just that the #if should check for BOOST_FILESYSTEM_VERSION >= 3, not check for it == 2
1081 2011-11-15 15:10:51 darksk1ez has quit (Ping timeout: 240 seconds)
1082 2011-11-15 15:11:27 <gavinandresen> It checks for == 3.  I'll change it to >= after I get the rc4 builds going.
1083 2011-11-15 15:11:48 darksk1ez has joined
1084 2011-11-15 15:14:35 * luke-jr should wake up before reading patches x.x
1085 2011-11-15 15:14:38 roconnor has quit (Remote host closed the connection)
1086 2011-11-15 15:14:40 roconnor_ has joined
1087 2011-11-15 15:14:58 <luke-jr> gavinandresen: how goes wallet fix?
1088 2011-11-15 15:15:35 <luke-jr> gavinandresen: also, don't forget to have the Qt guy relicense his icons :p
1089 2011-11-15 15:15:42 <dikidera> gavinandresen:i also "skimmed" at that microsoft paper
1090 2011-11-15 15:15:45 <gavinandresen> luke-jr: done, as far as I can tell-- I'm going to spin rc4 binaries to try to encourage more testing
1091 2011-11-15 15:15:49 <dikidera> and wow, the math in there is beyond what i know
1092 2011-11-15 15:15:54 copumpkin has quit (Quit: Computer has gone to sleep.)
1093 2011-11-15 15:16:09 <luke-jr> gavinandresen: done, including a backport, or should I look into that?
1094 2011-11-15 15:17:05 <gavinandresen> luke-jr: backport should be easy....  you should look into that.
1095 2011-11-15 15:17:13 <gavinandresen> (except for the wxwidgets issue)
1096 2011-11-15 15:17:32 * luke-jr grumbles about these commits not being daggy-fixes :p
1097 2011-11-15 15:18:23 agricocb has joined
1098 2011-11-15 15:18:50 <sipa> daggy-fixes?
1099 2011-11-15 15:19:29 <luke-jr> sipa: aka good practice writing bugfixes
1100 2011-11-15 15:19:31 <gavinandresen> Fixing earlier in the directed-acyclic-graph of changes... it is a good idea
1101 2011-11-15 15:19:50 <luke-jr> re "wallet encrypted; bitcoin server stopping, restart to run with encrypted wallet"-- why not just re-exec?
1102 2011-11-15 15:19:53 <gavinandresen> Like brushing your teeth after every meal is a good idea....
1103 2011-11-15 15:20:11 <luke-jr> sipa: http://mercurial.selenic.com/wiki/DaggyFixes
1104 2011-11-15 15:21:43 <sipa> luke-jr: interesting
1105 2011-11-15 15:21:58 <gavinandresen> luke-jr: why not re-exec:  "do the simplest possible thing that works"  -- and it took long enough to get this fix ready, I didn't want to test all of the possible, weird things that could go wrong if we tried to re-exec
1106 2011-11-15 15:22:44 <luke-jr> aka nothing?
1107 2011-11-15 15:22:46 <gavinandresen> (if you want to test re-execing qt apps on Windows....)
1108 2011-11-15 15:25:52 <Diablo-D3> da2ce7: blergh.
1109 2011-11-15 15:26:08 yebyen has joined
1110 2011-11-15 15:26:20 <da2ce7> :P I know, but you have the experence with java; I don't.
1111 2011-11-15 15:26:20 <gavinandresen> you mean nothing weird could possibly happen if we re-exec?  Possible weird thing number 1:  boost::filesystem::lock might still have a handle to the datadir lockfile if we re-exec before the process exits....
1112 2011-11-15 15:26:51 <Diablo-D3> da2ce7: the part highlighted? thats ugly code.
1113 2011-11-15 15:27:19 EPiSKiNG- has joined
1114 2011-11-15 15:27:28 <da2ce7> I've refactored everything arround it... however that code, I want to re-write it with intefaces and anon methods...
1115 2011-11-15 15:27:38 <da2ce7> however I need to first work out what it dose.
1116 2011-11-15 15:27:41 <Diablo-D3> da2ce7: OTCaller should have a constructor that has a OTCallback argument that does the set
1117 2011-11-15 15:28:16 <Diablo-D3> da2ce7: and dear lord is that code ugly.
1118 2011-11-15 15:28:30 <luke-jr> gavinandresen: exec() closes all fds by default
1119 2011-11-15 15:28:44 * da2ce7 know... have a look at the git history... OMG.
1120 2011-11-15 15:28:58 <Diablo-D3> I think whatever is trying to be done is wrong
1121 2011-11-15 15:29:43 <gavinandresen> luke-jr: mmm.  boost::filesystem::lock works by writing a .lock file to the directory.  I have no idea when it removes that .lock file, it might be done in a global destructor... which would happen after the re-exec.  And it would probably seem to work just fine until the race condition is lost on a few people's systems......
1122 2011-11-15 15:29:45 <Diablo-D3> it almost sounds like otapi's class should subclass otcaller and the otcaller code genericsized a bit so a subclass of otcaller can call it on any callback it needs (assuming it cant already)
1123 2011-11-15 15:30:03 <gavinandresen> anyway, my point is there-be-dragons, and I think it is important to get a fix out sooner if possible.
1124 2011-11-15 15:30:05 <Diablo-D3> da2ce7: and whoever made that Utility class needs to be shot
1125 2011-11-15 15:30:13 AStove has joined
1126 2011-11-15 15:30:14 d4de has quit (Ping timeout: 252 seconds)
1127 2011-11-15 15:30:35 <Diablo-D3> da2ce7: it looks like those two methods fiddle with globals, bad bad bad bad bad
1128 2011-11-15 15:31:01 <luke-jr> gavinandresen: could always setup a global destructor to fork-and-exec after things clean up too :p
1129 2011-11-15 15:31:13 <sipa> gavinandresen, luke-jr: if you'd go for re-execing, i'd argue it is better to just put the entire execution of the application in a while loop, that normalle exits, but with a flag reruns
1130 2011-11-15 15:31:26 wboy1 has quit (Read error: Operation timed out)
1131 2011-11-15 15:31:40 <da2ce7> ok... I guess the there is some re-writing of bad code to be done... :S
1132 2011-11-15 15:32:22 <gavinandresen> bitcoin's internal shutdown mechanism needs to be rewritten at some point, it is just plain ugly and wrong....
1133 2011-11-15 15:32:35 danbri has quit (Remote host closed the connection)
1134 2011-11-15 15:33:19 <gavinandresen> oh, and slow, too.
1135 2011-11-15 15:33:20 <Diablo-D3> da2ce7: a crapload. I dont even wanna look at the rest
1136 2011-11-15 15:34:58 <da2ce7> Diablo-D3,  yeah... so far I've only been working on the loader; the loader class didn't even exist before... it was all sitting on events in the setting dialogue.
1137 2011-11-15 15:35:04 <da2ce7> it was a fucking mess.
1138 2011-11-15 15:35:32 <da2ce7> I've got to these 6 lines of code, and have got stuck.
1139 2011-11-15 15:36:19 <luke-jr> http://luke.dashjr.org/tmp/code/0001-add-message-about-restarting-bitcoin-after-encryptin.patch <-- someone please apply to latest 0.4.x and test
1140 2011-11-15 15:39:29 <gavinandresen> luke-jr: is there a trusted build process for 0.4-stable?  I'd be happy to pull a gitorious branch and gitian-build it (after the 0.5rc4 builds are done)
1141 2011-11-15 15:40:46 wboy1 has joined
1142 2011-11-15 15:41:54 <luke-jr> gavinandresen: whatever it was for 0.4.0 I guess
1143 2011-11-15 15:41:56 <CIA-89> bitcoin: Pieter Wuille 0.4.x * r00eae584a262 bitcoind-stable/src/ (db.cpp db.h wallet.cpp): Resilvering
1144 2011-11-15 15:41:57 <CIA-89> bitcoin: Gavin Andresen 0.4.x * r2744ea8c1fbc bitcoind-stable/src/ (db.cpp db.h init.cpp rpc.cpp serialize.h wallet.cpp): Obsolete keypool and make sure database removes log files on shutdown.
1145 2011-11-15 15:41:59 <CIA-89> bitcoin: Gavin Andresen 0.4.x * r0143c024af1e bitcoind-stable/src/db.cpp: Fix boost filesystem incompatibility problem
1146 2011-11-15 15:42:24 <BlueMattBot> Project Bitcoin-Test build #83: FAILURE in 1 min 4 sec: http://jenkins.bluematt.me/job/Bitcoin-Test/83/
1147 2011-11-15 15:42:35 d4de has joined
1148 2011-11-15 15:43:01 <gavinandresen> luke-jr: I think you need to change a URL in contrib/gitian-something if the stable repo is at gitorious...
1149 2011-11-15 15:45:53 wboy1 has quit (Ping timeout: 276 seconds)
1150 2011-11-15 15:47:28 kiba has quit (Read error: Connection reset by peer)
1151 2011-11-15 15:49:02 <luke-jr> gavinandresen: thanks for the heads up
1152 2011-11-15 15:49:25 <luke-jr> gavinandresen: does that wx port look sane?
1153 2011-11-15 15:50:00 <gavinandresen> luke-jr: I don't know nuthin about wxwidgets programming....
1154 2011-11-15 15:50:06 <luke-jr> >_<
1155 2011-11-15 15:50:06 wboy1 has joined
1156 2011-11-15 15:50:34 <gavinandresen> ... but looks reasonable....
1157 2011-11-15 15:50:35 <CIA-89> bitcoin: Luke Dashjr 0.4.x * r1179f6373dff bitcoind-stable/contrib/gitian-descriptors/ (gitian-win32.yml gitian.yml): Update gitian descriptors to point at stable git repo
1158 2011-11-15 15:51:45 <sipa> the wallet encryption issue should probably be mentioned at https://en.bitcoin.it/wiki/Incidents
1159 2011-11-15 15:53:37 <luke-jr> sipa: I agree. :p
1160 2011-11-15 15:53:42 <gavinandresen> sipa: good idea.  As luke said, it is CVE-2011-4447
1161 2011-11-15 15:53:49 <gmaxwell> sipa: but what about the MICROSOFT FLAW! ;)
1162 2011-11-15 15:54:33 <gavinandresen> (wish we'd reported it earlier, could've been 2011-4444 ...)
1163 2011-11-15 15:54:47 <sipa> luke-jr: is there anyway to look that up yet?
1164 2011-11-15 15:54:59 <sipa> on the CBE site searching for 4447 or bitcoin doesn't find anything
1165 2011-11-15 15:55:08 <luke-jr> sipa: not until gavin announces it formally
1166 2011-11-15 15:55:19 <luke-jr> or adds it to the one he already did I guess
1167 2011-11-15 15:55:22 <sipa> i see, so it is just a registered number for now
1168 2011-11-15 15:56:25 andywork has joined
1169 2011-11-15 15:57:20 <gavinandresen> I'll post again when 0.5rc4 and 0.4.1 builds are ready.  Halfway done with the rc4 gitian builds right now...
1170 2011-11-15 15:57:42 <luke-jr> sipa: can you test that wx patch? :P
1171 2011-11-15 15:58:12 * luke-jr uninstalled wx as soon as Bitcoin-Qt was merged :p
1172 2011-11-15 16:02:32 Shaded has joined
1173 2011-11-15 16:02:48 copumpkin has joined
1174 2011-11-15 16:06:15 <k9quaint> I am going to forward all transactions just to irritate Microsoft
1175 2011-11-15 16:07:25 <gmaxwell> k9quaint: don't hate on microsoft research. Their behavior was good here. It wasn't their team that advertised this as a flaw in bitcoin.
1176 2011-11-15 16:09:35 <luke-jr> k9quaint: cool. I have patches for that in my repo
1177 2011-11-15 16:10:48 <cocktopus> fuck /. for calling it a flaw
1178 2011-11-15 16:11:30 <coderrr> i think i acutally used the word flaw first, but in a much less sensationalist way
1179 2011-11-15 16:11:43 <gmaxwell> coderrr: yea, you're naughty too, but it's less bad.
1180 2011-11-15 16:13:24 datagutt has joined
1181 2011-11-15 16:16:35 <luke-jr> http://luke.dashjr.org/tmp/screenshots/snapshot71.png <-- coderrr
1182 2011-11-15 16:16:58 <gmaxwell> Is it a picture of chicken little?
1183 2011-11-15 16:17:03 <luke-jr> Bluebeard
1184 2011-11-15 16:21:58 Zarutian has joined
1185 2011-11-15 16:31:05 Shaded has quit (Quit: Shaded)
1186 2011-11-15 16:33:49 tower has quit (Ping timeout: 260 seconds)
1187 2011-11-15 16:39:11 tower has joined
1188 2011-11-15 16:39:35 Shaded has joined
1189 2011-11-15 16:39:56 `2Fast2BCn_ has joined
1190 2011-11-15 16:40:14 `2Fast2BCn has quit (Ping timeout: 260 seconds)
1191 2011-11-15 16:40:14 `2Fast2BCn_ is now known as `2Fast2BCn
1192 2011-11-15 16:42:56 wasabi has joined
1193 2011-11-15 16:45:33 wasabi2 has quit (Ping timeout: 256 seconds)
1194 2011-11-15 16:45:48 ThomasV has quit (Quit: Leaving)
1195 2011-11-15 16:50:57 Turingi has quit (Read error: Connection reset by peer)
1196 2011-11-15 16:51:05 BlueMatt has joined
1197 2011-11-15 16:51:34 Turingi has joined
1198 2011-11-15 16:53:40 <BlueMatt> gavinandresen: were you given the m$ paper on tx relay before it came out?
1199 2011-11-15 16:53:53 <gavinandresen> BlueMatt: nope
1200 2011-11-15 16:54:01 <BlueMatt> wtf?
1201 2011-11-15 16:54:38 <gavinandresen> meh.  I don't know nuthin about highfalutin-game-theoretic-math-sybil-resistance stuff.....
1202 2011-11-15 16:55:30 <gavinandresen> 0.5rc4 builds are available for sanity testing:  https://sourceforge.net/projects/bitcoin/files/Bitcoin/bitcoin-0.5.0/test/
1203 2011-11-15 16:55:32 <BlueMatt> gavinandresen: it is bs as far as Im concerned: the issue with their attack is that for non-miners, the incentive is for you to relay as much as possible...though miners might have an incentive to not relay txes, the rest of the network does not
1204 2011-11-15 16:55:49 <BlueMatt> and most of the network are not miners
1205 2011-11-15 16:56:07 <gavinandresen> why would I relay if I can save a couple of pennies a month on bandwidth costs if I don't?
1206 2011-11-15 16:56:09 * BlueMatt wonders what TD[gone] makes of this
1207 2011-11-15 16:56:24 <BlueMatt> gavinandresen: well bandwidth aside as most people have unmetered bw
1208 2011-11-15 16:56:26 <gavinandresen> (I think they're correct, there is a small incentive problem)
1209 2011-11-15 16:56:32 <BlueMatt> (or dont come close to hitting on their limit)
1210 2011-11-15 16:56:49 <BlueMatt> gavinandresen: there is, but I dont think it can make a real-world difference even if all miners stop relaying txes
1211 2011-11-15 16:57:07 <coderrr> BlueMatt, what if the network tx rate was orders of magnitude higher than it is today ?
1212 2011-11-15 16:57:33 <BlueMatt> coderrr: the question becomes what the network looks like, not how many txes you get
1213 2011-11-15 16:57:48 <BlueMatt> coderrr: however, in that case, I would see the network as a bunch of deals between tx-makers and miners
1214 2011-11-15 16:57:54 <BlueMatt> eg the one between mtgox and eligius
1215 2011-11-15 16:58:03 <coderrr> yea, the issue would only manifest so far in the future that no one knows waht things will look like
1216 2011-11-15 16:58:04 <BlueMatt> where eligius gives mtgox free txes
1217 2011-11-15 16:58:33 <BlueMatt> in that case, as a miner, you are going to take the txes you get from your contracts, but likely not any others anyway
1218 2011-11-15 16:58:36 <BlueMatt> (unless they have a fee)
1219 2011-11-15 16:58:39 <Eliel> what if we end up having a workable system so people don't need to pool up anymore?
1220 2011-11-15 16:59:01 <Eliel> or rather, automatic pooling
1221 2011-11-15 16:59:27 <luke-jr> BlueMatt: can you test wxBitcoin 0.4.1 patch plz?
1222 2011-11-15 16:59:34 <Eliel> I think, there's no way around that in such a system it makes sense to share the Txs because you're unlikely to find blocks all by yourself.
1223 2011-11-15 16:59:57 <luke-jr> http://luke.dashjr.org/tmp/code/0001-add-message-about-restarting-bitcoin-after-encryptin.patch
1224 2011-11-15 17:00:00 <BlueMatt> in that case, assuming you get a "bitcoin backbone" with supernodes and leafnodes ala gnutella, the issue could manifest itself, however as long as leaf nodes make connections to multiple supernodes (which they will have to) they will get enough miners with their txes for me to not see an issue
1225 2011-11-15 17:00:15 eueueue has joined
1226 2011-11-15 17:00:51 <upb> whats an 'm$ paper' ?
1227 2011-11-15 17:01:01 <BlueMatt> upb: Eliel: out of scope in my opinion: in the case of a p2pool-style system, you can argue no one will take
1228 2011-11-15 17:01:08 <BlueMatt> upb: http://research.microsoft.com/pubs/156072/bitcoin.pdf
1229 2011-11-15 17:01:08 <Eliel> upb: paper published by microsoft research a few days ago.
1230 2011-11-15 17:01:10 <luke-jr> upb: Microsoft wrote a paper outlining a "flaw" in the Bitcoin protocol
1231 2011-11-15 17:01:12 <upb> oh, microsoft
1232 2011-11-15 17:01:32 <BlueMatt> Eliel: out of scope in my opinion: in the case of a p2pool-style system, you can argue no one will take txes or everyone will, so its important the system forces people to take them
1233 2011-11-15 17:01:47 <k9quaint> gmaxwell: I know, bringing sunshine to flaws is always good, but I just like to irritate M$ :)
1234 2011-11-15 17:02:04 <BlueMatt> Eliel: though, for the record, I still think there will be large-scale miners instead of one huge p2pool
1235 2011-11-15 17:02:19 <BlueMatt> (if bitcoin ever takes off)
1236 2011-11-15 17:02:41 * BlueMatt also wonders what gmaxwell and sipa think of the paper
1237 2011-11-15 17:02:51 <k9quaint> BlueMatt: I picture merchant "guilds" or collectives for mining and fast transaction confirmation
1238 2011-11-15 17:02:53 <luke-jr> gavinandresen: I guess since nobody wants to test that wx patch, I'll just put it in 0.4.1rc and someone can report a bug if there is one
1239 2011-11-15 17:03:17 <BlueMatt> k9quaint: more like large-scale miners (and/or pools) which offer contracts for free txes to merchants or other interested parties
1240 2011-11-15 17:03:26 <BlueMatt> luke-jr: give me a minute
1241 2011-11-15 17:03:47 <Eliel> BlueMatt: I'd think there will be large scale miners. But even they might end up using the p2pool-style system.
1242 2011-11-15 17:03:48 andywork has left ("Lämnar")
1243 2011-11-15 17:03:55 andywork has joined
1244 2011-11-15 17:04:04 <luke-jr> Eliel: p2pool can't scale.
1245 2011-11-15 17:04:11 <BlueMatt> Eliel: then we are using the term large-scale miners differently
1246 2011-11-15 17:04:12 traviscj has quit (Remote host closed the connection)
1247 2011-11-15 17:04:13 <gavinandresen> How do I tell qmake to create a -g instead of a -O2 Makefile?
1248 2011-11-15 17:04:34 <BlueMatt> gavinandresen: win32: nfc, linux: iirc you make -f Makefile.debug instead
1249 2011-11-15 17:04:39 <luke-jr> gavinandresen: I'd think "make CXXFLAGS='-ggdb'" afterward
1250 2011-11-15 17:04:41 <BlueMatt> (or maybe thats for win32 instead of linux...)
1251 2011-11-15 17:05:02 <Eliel> luke-jr: yes, p2pool doesn't. Hence I'm talking about p2pool-like system.
1252 2011-11-15 17:05:08 <luke-jr> BlueMatt: btw, that patch goes on top of the latest 0.4.x branch HEAD
1253 2011-11-15 17:05:16 <luke-jr> BlueMatt: or rather, I just applied it there
1254 2011-11-15 17:05:26 <BlueMatt> Eliel: so in other words you think we will have large-scale p2pool and NOT large-scale miners
1255 2011-11-15 17:05:30 <BlueMatt> individual*(
1256 2011-11-15 17:05:36 <CIA-89> bitcoin: Luke Dashjr 0.4.x * r586ea168c2e2 bitcoind-stable/src/ui.cpp: add message about restarting bitcoin after encrypting wallet succesfully
1257 2011-11-15 17:06:11 <Eliel> BlueMatt: why are those mutually exclusive? I think we'll have the latter, whatever the case and I really hope we get the former working.
1258 2011-11-15 17:06:22 * wboy1 slaps upb around a bit with a large trout
1259 2011-11-15 17:06:44 <gavinandresen> qmake CONFIG=debug seems to do the right thing
1260 2011-11-15 17:07:12 <BlueMatt> Eliel: You said the large-scale individual miners will be using a p2pool-like system, which makes them not large-scale individuals
1261 2011-11-15 17:07:16 james has joined
1262 2011-11-15 17:07:19 urstroyer has left ()
1263 2011-11-15 17:07:42 james is now known as Guest73681
1264 2011-11-15 17:07:49 <luke-jr> git://gitorious.org/+bitcoin-stable-developers/bitcoin/bitcoind-stable.git
1265 2011-11-15 17:07:53 <luke-jr> for reference
1266 2011-11-15 17:08:24 <Eliel> BlueMatt: I don't see why it makes them not individual if they choose to reduce their variability with a p2pool-like system.
1267 2011-11-15 17:08:57 <Eliel> but I don't see it as important if they'll use a p2pool like system or not.
1268 2011-11-15 17:09:00 <BlueMatt> Eliel: because then they arent able to act as an individual
1269 2011-11-15 17:09:14 <BlueMatt> Eliel: (assuming the p2pool system is implemented right)
1270 2011-11-15 17:09:46 <luke-jr> BlueMatt: uh, no?
1271 2011-11-15 17:09:51 <Eliel> BlueMatt: implemented right, as in, not allowing miners to pick and choose what transactions to accept?
1272 2011-11-15 17:10:02 <BlueMatt> Eliel: yea
1273 2011-11-15 17:10:17 <luke-jr> that's implemented wrong :P
1274 2011-11-15 17:10:18 <Eliel> that completely messes up the idea of fees set by the market.
1275 2011-11-15 17:11:05 <BlueMatt> fees are fine, but you have to incentivize people to a. forward txes and b. accept *some* txes
1276 2011-11-15 17:11:09 <BlueMatt> even if not all
1277 2011-11-15 17:11:26 devrandom has quit (Ping timeout: 248 seconds)
1278 2011-11-15 17:11:52 <Eliel> if it's some kind of democratic decision, then it doesn't mess up the fees but...
1279 2011-11-15 17:12:20 <luke-jr> ideally, miners would do p2pool around the generation, and keep all fees to themselves
1280 2011-11-15 17:12:25 <luke-jr> then accept fees on their own policies
1281 2011-11-15 17:12:37 <luke-jr> and pools like Eligius would become subscription models
1282 2011-11-15 17:12:41 erus` has quit (Remote host closed the connection)
1283 2011-11-15 17:12:52 <BlueMatt> you mean like eligius contracts to merchants
1284 2011-11-15 17:12:52 <BlueMatt> ?
1285 2011-11-15 17:13:02 <luke-jr> ie, businesses buy transactions in bulk from us, and we tell miners to accept them
1286 2011-11-15 17:13:13 <luke-jr> and miners participating get some kind of guaranteed income
1287 2011-11-15 17:13:16 <BlueMatt> well you tell your pool's miners
1288 2011-11-15 17:13:22 <BlueMatt> but yea
1289 2011-11-15 17:13:49 <BlueMatt> luke-jr: you did already merge the actual encryption fix right?
1290 2011-11-15 17:13:56 <luke-jr> BlueMatt: yes
1291 2011-11-15 17:14:07 <luke-jr> that went in cleanly
1292 2011-11-15 17:14:14 <luke-jr> I had to write something cusotm for wx tho
1293 2011-11-15 17:14:19 <luke-jr> which needs testing
1294 2011-11-15 17:14:31 <BlueMatt> luke-jr: oh, well that is a pretty simple patch, not sure how it could go wrong...
1295 2011-11-15 17:14:37 <BlueMatt> (famous last words)
1296 2011-11-15 17:14:43 <luke-jr> me either, but would be nice to have someone try it :P
1297 2011-11-15 17:16:43 <gavinandresen> Try running it against a previously-encrypted wallet; I'm getting an "Upgrade complete, restart please" and then a crash when I poke OK.
1298 2011-11-15 17:16:48 <gavinandresen> (on osx)
1299 2011-11-15 17:17:01 <gavinandresen> (debugging now....)
1300 2011-11-15 17:19:15 <eueueue> Hi, this message on terminal when closing bitcoin 0.5 is normal:? Application asked to unregister timer 0x5f000005 which is not registered in this thread. Fix application.
1301 2011-11-15 17:19:23 <sipa> gavinandresen: related to this: https://github.com/bitcoin/bitcoin/commit/19197d5e2973785e37b8b0eeefc3d6af0b2bb1fc ?
1302 2011-11-15 17:20:01 <BlueMatt> eueueue: known issue, no known fix, no known negative effects aside from that annoying error message
1303 2011-11-15 17:20:18 traviscj has joined
1304 2011-11-15 17:20:24 <eueueue> Hi, I installed bitcoin 0.5 rc4 and on bitcoin folder I see a file called: wallet.dat.rewrite.
1305 2011-11-15 17:20:38 <eueueue> This file will not disappear?
1306 2011-11-15 17:20:49 <gavinandresen> eueueue: yes, it will.
1307 2011-11-15 17:20:59 <gavinandresen> eueueue: it will become your new wallet.dat file.
1308 2011-11-15 17:21:20 <eueueue> here I already closed and reopened bitcoin and the file is still there
1309 2011-11-15 17:21:25 <eueueue> didn't disappear
1310 2011-11-15 17:22:27 <eueueue> I closed the app but the file didn't disappear.
1311 2011-11-15 17:22:28 <gavinandresen> eueueue: did you close bitcoin before it had a chance to completely rewrite your wallet?
1312 2011-11-15 17:22:37 <eueueue> hum
1313 2011-11-15 17:22:53 <eueueue> let me explaing what I did:
1314 2011-11-15 17:23:12 <luke-jr> gavinandresen: there's code in place to prevent any updates to the stale wallet.dat after it's been rewritten, right?
1315 2011-11-15 17:23:15 <luke-jr> or keep the new one in sync
1316 2011-11-15 17:24:04 <eueueue> I was using bitcoin 0.5 rc3. So I installed bitcoin 0.5 rc4. When I opened 0.5 rc4, a error message appear telling to me close bitcoin
1317 2011-11-15 17:24:14 <eueueue> I closed and reopened it
1318 2011-11-15 17:24:18 <gavinandresen> the code does:   copy everything from wallet.dat to wallet.dat.rewrite.  If copy is successful:  remove wallet.dat.  Rename wallet.dat.rewrite wallet.dat.  Then shutdown and prompt for restart.
1319 2011-11-15 17:24:20 <eueueue> Bitcoin open ok
1320 2011-11-15 17:24:37 danbri has joined
1321 2011-11-15 17:24:37 <eueueue> I stayed with this opened for a tine
1322 2011-11-15 17:24:39 <eueueue> time
1323 2011-11-15 17:24:47 <eueueue> and closed
1324 2011-11-15 17:24:57 <gavinandresen> eueueue: what operating system?
1325 2011-11-15 17:25:12 <eueueue> debian wheezy 64
1326 2011-11-15 17:25:23 <eueueue> will reopen bitcoin to see if the file disappear
1327 2011-11-15 17:25:25 <gavinandresen> Running bitcoind or bitcoin-qt?
1328 2011-11-15 17:25:37 <eueueue> bitcoin qt
1329 2011-11-15 17:26:21 <gavinandresen> sipa, can you tackle this one?  I'm still looking at the osx crash-on-shutdown bug
1330 2011-11-15 17:28:07 graingert has joined
1331 2011-11-15 17:28:18 <eueueue> http://imageshack.us/photo/my-images/9/capturadetelatl.png/
1332 2011-11-15 17:28:52 <graingert> eueueue: what happens if you run another QT app
1333 2011-11-15 17:29:09 <graingert> eueueue: look like an issue with le qtGTK
1334 2011-11-15 17:29:51 <eueueue> But if app works ok, means that my bew wallet is ok right?
1335 2011-11-15 17:29:57 <sipa> gavinandresen: what exactly?
1336 2011-11-15 17:30:19 <eueueue> and I can delete wallet.dat.rewrite
1337 2011-11-15 17:30:20 <eueueue> ?
1338 2011-11-15 17:30:31 <gavinandresen> sipa: figure out why eueueue ended up with a wallet.dat.rewrite
1339 2011-11-15 17:31:19 <eueueue> I'm newbie people. Anything to tell me to do. need be step by step
1340 2011-11-15 17:33:32 ThomasV has joined
1341 2011-11-15 17:37:42 <eueueue> Even with wallet.dat.rewrite still on the folder, can I be sure that my wallet.dat is ok?
1342 2011-11-15 17:37:53 <eueueue> without problems
1343 2011-11-15 17:38:43 eastender has joined
1344 2011-11-15 17:39:15 <sipa> if both exist, you should be able to remove wallet.dat.rewrite
1345 2011-11-15 17:39:19 <sipa> gavinandresen: sorry, not now
1346 2011-11-15 17:40:06 <eueueue> ok
1347 2011-11-15 17:44:25 graingert has quit (Remote host closed the connection)
1348 2011-11-15 17:46:57 Shaded has quit (Quit: Shaded)
1349 2011-11-15 17:48:24 erus` has joined
1350 2011-11-15 17:50:52 localhost has joined
1351 2011-11-15 17:52:38 <BlueMattBot> Yippie, build fixed!
1352 2011-11-15 17:52:38 <BlueMattBot> Project Bitcoin-Test build #84: FIXED in 7 min 42 sec: http://jenkins.bluematt.me/job/Bitcoin-Test/84/
1353 2011-11-15 17:52:39 <BlueMattBot> * gavinandresen: Resilvering
1354 2011-11-15 17:52:39 <BlueMattBot> * gavinandresen: Obsolete keypool and make sure database removes log files on shutdown.
1355 2011-11-15 17:52:40 <BlueMattBot> * gavinandresen: add message about restarting bitcoin after encrypting wallet succesfully
1356 2011-11-15 17:52:40 <BlueMattBot> * gavinandresen: Fix boost filesystem incompatibility problem
1357 2011-11-15 17:52:47 <BlueMatt> gavinandresen: sorry about that...
1358 2011-11-15 17:53:08 <gavinandresen> BlueMatt: no problem.
1359 2011-11-15 17:53:28 <gavinandresen> frickin frackin....
1360 2011-11-15 17:53:48 <cocktopus> fran
1361 2011-11-15 17:54:24 <gavinandresen> osx crash is a call to:  dbenv.close(0);  ... it is throwing a Invalid argument (22) exception...
1362 2011-11-15 17:55:24 <gavinandresen> catching the exception fixes the crash, but what argument is invalid?
1363 2011-11-15 17:58:12 Shaded has joined
1364 2011-11-15 18:02:06 m00p has joined
1365 2011-11-15 18:05:13 Shaded has quit (Quit: Shaded)
1366 2011-11-15 18:06:18 Shaded has joined
1367 2011-11-15 18:06:33 <luke-jr> gavinandresen: docs say "If the cursor is already closed; or if an invalid flag value or parameter was specified."
1368 2011-11-15 18:07:09 <luke-jr> maybe see if it's being closed twice?
1369 2011-11-15 18:07:10 <gavinandresen> luke-jr: I think it might be:  The DbEnv handle should not be closed while any other handle that refers to it is not yet closed;
1370 2011-11-15 18:07:27 <gavinandresen> It is definitely not being closed twice, it is protected by the fDbEnvInit flag
1371 2011-11-15 18:07:51 <gavinandresen> (and my breakpoint is only hitting once)
1372 2011-11-15 18:10:20 cronopio has joined
1373 2011-11-15 18:10:22 <gavinandresen> wtf?  CDB::Close() doesn't call pdb->close() ?
1374 2011-11-15 18:12:03 <gavinandresen> oh, I see, they're closed in void static CloseDb(file...)
1375 2011-11-15 18:13:16 cloudbank has quit (Ping timeout: 240 seconds)
1376 2011-11-15 18:14:20 <gavinandresen> ... so why isn't CDB::Close() calling CloseDb....
1377 2011-11-15 18:14:34 <gavinandresen> (too many fricking layers of abstraction)
1378 2011-11-15 18:15:23 <luke-jr> molecular: you're welcome
1379 2011-11-15 18:15:50 Kolky has joined
1380 2011-11-15 18:16:15 <molecular> luke-jr, ??
1381 2011-11-15 18:16:47 <luke-jr> https://bitcointalk.org/index.php?topic=51604.msg619607#msg619607
1382 2011-11-15 18:16:52 cloudbank has joined
1383 2011-11-15 18:16:53 <luke-jr> isn't that you?
1384 2011-11-15 18:17:23 <molecular> yes, it was just impossible for me to associate this msg here with that post ;)
1385 2011-11-15 18:17:51 <molecular> thanks again for you guys hard work
1386 2011-11-15 18:17:58 <molecular> keep going
1387 2011-11-15 18:18:13 <molecular> btw: did you see the discussion about making it easier for "normal" people to test stuff?
1388 2011-11-15 18:19:03 <molecular> gavinandresen says testing is the bottleneck, you other guys agree on this? if so, I might make an effort getting people to test...
1389 2011-11-15 18:19:19 <luke-jr> molecular: I agree more testing is always better.
1390 2011-11-15 18:19:32 <BlueMatt> molecular: talk to alexwaters
1391 2011-11-15 18:19:39 <luke-jr> molecular: wx 0.4.x HEAD needs testing especially
1392 2011-11-15 18:19:39 <BlueMatt> molecular: and yea, more testing is always better
1393 2011-11-15 18:20:24 <molecular> I'm thinking more along the lines to make a site/page that helps "normal users" conduct orderly test, provide test downloads, lists of what needs testing, etc.
1394 2011-11-15 18:20:33 <molecular> would that be worth the effort?
1395 2011-11-15 18:21:01 <molecular> or would that be redundant? I don't really know what's currently there except github..
1396 2011-11-15 18:21:10 <BlueMatt> molecular: again, talk to alexwaters, jenkins already provides nice git master builds and alex would know best if its worth setting up
1397 2011-11-15 18:21:23 <molecular> k, thanks BlueMatt
1398 2011-11-15 18:21:24 <luke-jr> BlueMatt: how about getting jenkins to do stable builds too?
1399 2011-11-15 18:21:49 <luke-jr> or is 0.4.x too old for that somehow?
1400 2011-11-15 18:22:00 <BlueMatt> luke-jr: oh, you mean 0.4.x builds...
1401 2011-11-15 18:22:11 <BlueMatt> well I would have to go write scripts to do them...
1402 2011-11-15 18:22:15 <luke-jr> and 0.5.x once that's started
1403 2011-11-15 18:22:17 <BlueMatt> (and that would be a pita...)
1404 2011-11-15 18:22:30 <luke-jr> does jenkins use gitian, or no?
1405 2011-11-15 18:22:35 <BlueMatt> no
1406 2011-11-15 18:22:41 <luke-jr> hmm
1407 2011-11-15 18:22:45 <BlueMatt> send me an email as a remainder, and Ill write them sometime in the next couple weeks
1408 2011-11-15 18:22:59 <BlueMatt> it cant use gitian as the jenskins server is hosted on aws
1409 2011-11-15 18:23:18 Backburn has joined
1410 2011-11-15 18:23:25 <BlueMatt> though I have thought about extracting gitian scripts for use on jenkins once or twice...
1411 2011-11-15 18:23:27 <sipa> gavinandresen: bitcoin keeps database files open all the time at bdb level
1412 2011-11-15 18:23:33 <sipa> gavinandresen: only DBFLush actually closes the handles
1413 2011-11-15 18:24:04 <luke-jr> BlueMatt: if I set you up with a KVM-friendly VM for jenkins, would that help?
1414 2011-11-15 18:25:08 <BlueMatt> luke-jr: it could, but I dont see much of an advantage to doing gitian builds on jenkins...writing scripts isnt easy to alway keep them up-to-date but it is more convenient...
1415 2011-11-15 18:25:22 <BlueMatt> luke-jr: plus running non-gitian builds happens much quicker than booting a new vm each time
1416 2011-11-15 18:26:03 <luke-jr> BlueMatt: using gitian would make them deterministic snapshots, plus test the gitian stuff too
1417 2011-11-15 18:26:28 <luke-jr> true
1418 2011-11-15 18:26:31 <BlueMatt> luke-jr: thats true, but jenkins isnt set up to do gitian builds directly and it would miss certain logs
1419 2011-11-15 18:26:41 <luke-jr> oh well
1420 2011-11-15 18:26:55 <BlueMatt> luke-jr: could be done, but setting it up would be more work than its worth imho
1421 2011-11-15 18:27:25 <BlueMatt> (though devrandom did talk about doing a jenkins gitian plugin a while back, so maybe he could make it easier)
1422 2011-11-15 18:27:50 <BlueMatt> anyway, for now Im just gonna add projects for stable branch and see what happens
1423 2011-11-15 18:29:53 traviscj has quit (Remote host closed the connection)
1424 2011-11-15 18:32:07 <gmaxwell> BlueMatt: see the logs from earlier today for the DNSseed hangup that was reported.
1425 2011-11-15 18:36:58 OneFixt_ has quit (Changing host)
1426 2011-11-15 18:36:58 OneFixt_ has joined
1427 2011-11-15 18:37:19 OneFixt_ is now known as OneFixt
1428 2011-11-15 18:38:26 wolfspraul has quit (Ping timeout: 245 seconds)
1429 2011-11-15 18:38:55 traviscj has joined
1430 2011-11-15 18:39:20 BlueMatt_ has joined
1431 2011-11-15 18:39:34 wolfspraul has joined
1432 2011-11-15 18:40:29 BlueMatt has quit (Ping timeout: 260 seconds)
1433 2011-11-15 18:41:42 BlueMatt_ is now known as BlueMatt
1434 2011-11-15 18:42:06 <BlueMatt> gmaxwell: do you remember who or exactly when?
1435 2011-11-15 18:42:16 <BlueMatt> gmaxwell: nvm found it
1436 2011-11-15 18:42:38 <sipa> BlueMatt: Acciaio, about 5-6 hours ago
1437 2011-11-15 18:43:32 <gmaxwell> BlueMatt: sounds like he had breakage which wasn't our fauilt, but DNSseed getting stuck shouldn't prevent startup.
1438 2011-11-15 18:45:33 <BlueMatt> gmaxwell: yea, though vladmir's dnsseed returns more a records than can be fit in a udp packet and thus might cause problems with some obscure broken dns recursors
1439 2011-11-15 18:45:40 <BlueMatt> (as it has to be retried in tcp mode)
1440 2011-11-15 18:46:03 <BlueMatt> I suppose it doesnt matter though as his was removed in https://github.com/bitcoin/bitcoin/commit/02d630c37f6155220ca4265faa7964c689a01be9
1441 2011-11-15 18:46:34 <gmaxwell> looks like that was the one his was querying when it went out to lunch too.
1442 2011-11-15 18:47:28 <BlueMatt> well the call to DNSAddressSeed() should be moved into a thread away from AppInit2
1443 2011-11-15 18:47:34 <BlueMatt> probably in StartNode()
1444 2011-11-15 18:48:48 slush has quit (Ping timeout: 258 seconds)
1445 2011-11-15 18:49:18 <alexwaters> molecular: alright to PM?
1446 2011-11-15 18:49:18 <BlueMatt> should be a simple fix, anyone care to write it?
1447 2011-11-15 18:50:04 <BlueMatt> probably best to just give dns lookup its own thread
1448 2011-11-15 18:50:43 <gavinandresen> more threads.... yay.....
1449 2011-11-15 18:51:05 <BlueMatt> meh, modern threads have so little overhead its not a big deal
1450 2011-11-15 18:51:24 <gavinandresen> it's not performance, it is debug-ability
1451 2011-11-15 18:51:25 <copumpkin> huh?
1452 2011-11-15 18:51:30 <BlueMatt> (hence why I still stand by https://github.com/bitcoin/bitcoin/pull/454)
1453 2011-11-15 18:51:38 <copumpkin> threads are still pretty expensive
1454 2011-11-15 18:51:38 <gavinandresen> and lockind and deadlocks and...
1455 2011-11-15 18:51:54 <BlueMatt> a thread that just adds nodes (like irc thread does anyway)...
1456 2011-11-15 18:52:16 <BlueMatt> copumpkin: not really
1457 2011-11-15 18:52:27 TD_ has joined
1458 2011-11-15 18:53:06 <molecular> alexwaters, sure, pm me
1459 2011-11-15 18:53:29 localhost has quit (Remote host closed the connection)
1460 2011-11-15 18:53:35 agricocb has quit (Quit: Leaving.)
1461 2011-11-15 18:54:41 eueuuee has joined
1462 2011-11-15 18:54:58 eueueue has quit (Ping timeout: 265 seconds)
1463 2011-11-15 18:56:33 <eueuuee> test
1464 2011-11-15 18:56:51 <copumpkin> fail
1465 2011-11-15 18:57:33 <sipa> failing tests are often much more interesting than succeeding ones, though
1466 2011-11-15 18:57:44 <sipa> (students may disagree)
1467 2011-11-15 18:57:56 <Graet> i failed many - and look at me today
1468 2011-11-15 18:58:02 <Graet> :P
1469 2011-11-15 18:58:13 <BlueMatt> sipa: when coding, yea, when taking tests, fu
1470 2011-11-15 18:59:08 <BlueMatt> Graet: I cant, all I see is a name and some text...
1471 2011-11-15 18:59:21 <Graet> hehe
1472 2011-11-15 18:59:29 <roconnor_> sipa: etotheipi_: I've successfully put a compressed public key into testnet
1473 2011-11-15 18:59:37 <roconnor_> http://blockexplorer.com/testnet/tx/48ce0ba13b087f3712cdc354db918436aeeca60596ac7c54572c8c1b9a8b5ba4
1474 2011-11-15 18:59:41 <Graet> better like that BlueMatt ;) tho my pic is in the forums ;)
1475 2011-11-15 18:59:58 <roconnor_> sipa: etotheipi_: notice that the public key is encoded as: 02a32efde012298e69e3601eb94fceb84c900efecdca8abc6a46f20a810acf18b7
1476 2011-11-15 19:00:33 <sipa> wow, and that was just accepted?
1477 2011-11-15 19:00:41 <roconnor_> it appears to be
1478 2011-11-15 19:00:53 <roconnor_> as I conjectured it would
1479 2011-11-15 19:00:54 <sipa> i've done tests earlier, and failed
1480 2011-11-15 19:00:54 <justmoon> sipa: yeah, openssl doesn't care
1481 2011-11-15 19:01:00 <justmoon> I pointed that out a while ago
1482 2011-11-15 19:01:29 <BlueMatt> a. is that a ddos option, and b. if its not should we do that by default?
1483 2011-11-15 19:01:45 <justmoon> roconnor_, props for using pi amount :D
1484 2011-11-15 19:02:05 <sipa> he could have used more decimals!
1485 2011-11-15 19:02:07 <roconnor_> I don't see why it would be ddos; I think it should be by default
1486 2011-11-15 19:02:12 <roconnor_> I should have used more decimals
1487 2011-11-15 19:02:22 <sipa> the cpu power required to verify a compressed pubkey is neglectable imho
1488 2011-11-15 19:02:25 <BlueMatt> roconnor_: well does it take openssl more cpu time to decompress it
1489 2011-11-15 19:02:46 <sipa> maybe a few %
1490 2011-11-15 19:03:01 <BlueMatt> well I would say test and make default
1491 2011-11-15 19:03:04 <roconnor_> BlueMatt: one tricky issues is that there are now two different ways to encode the same public key and these two diffent encodings will have different Hash160s
1492 2011-11-15 19:03:06 <justmoon> roconnor_, do you have the same key in uncompressed hex handy?
1493 2011-11-15 19:03:16 <roconnor_> so some care needs to be done to be aware of this issue
1494 2011-11-15 19:03:42 <roconnor_> justmoon: no, but I think I can compute it ...
1495 2011-11-15 19:03:46 <sipa> if you do a CKey.SetPubKey() of the compressed key, .GetPubKey() will return the full one, i assume
1496 2011-11-15 19:04:03 <roconnor_> sipa is correct
1497 2011-11-15 19:04:13 <sipa> so i hope there are no tests that verify based on the hash of GetPubKey
1498 2011-11-15 19:04:29 <roconnor_> sipa: I didn't notice any
1499 2011-11-15 19:04:46 <roconnor_> I made this by patching bitcoin and just modified GetPubKey()
1500 2011-11-15 19:05:02 <roconnor_> (though I had to run it from scratch)
1501 2011-11-15 19:05:13 <roconnor_> (with an empty .bitcoin directory)
1502 2011-11-15 19:05:15 eastender has quit (Quit: Nettalk6 - www.ntalk.de)
1503 2011-11-15 19:06:02 <sipa> roconnor_: again unconscious decisions that have influenced the standard - what if satoshi had done a Hash160 of the GetPubKey for verifying an address?
1504 2011-11-15 19:06:17 <sipa> that would have meant that you'd have only one address instead of two
1505 2011-11-15 19:06:58 p0s has joined
1506 2011-11-15 19:07:48 TD_ has quit (Quit: TD_)
1507 2011-11-15 19:08:57 <roconnor_> justmoon: 04a32efde012298e69e3601eb94fceb84c900efecdca8abc6a46f20a810acf18b74d6fe0cfc04d7eba0b3340d1c4e50370a65caa2efcc0067d4337c1cca73407a4
1508 2011-11-15 19:09:50 iocor has quit (Quit: Textual IRC Client: http://www.textualapp.com/)
1509 2011-11-15 19:10:30 <justmoon> roconnor_, thanks! do you happen to have the hashForSignature hash corresponding for that transaction as well?
1510 2011-11-15 19:11:05 <roconnor_> not really
1511 2011-11-15 19:11:15 PK has joined
1512 2011-11-15 19:11:24 <roconnor_> I don't quite yet have the tool set build to easily find that
1513 2011-11-15 19:11:33 <justmoon> ok no worries
1514 2011-11-15 19:11:34 <roconnor_> ... I should really build myself some tools for doing that
1515 2011-11-15 19:11:42 <roconnor_> it's a question I've often asked myself
1516 2011-11-15 19:11:46 <justmoon> I'll run the benchmark with a bogus hash first
1517 2011-11-15 19:12:53 <roconnor_> BlueMatt: decompressing a key requires performing a "Square root" operation and a few other arithmetic operations.
1518 2011-11-15 19:13:03 <justmoon> ok, this is somewhat unexpected...
1519 2011-11-15 19:13:11 <justmoon> ah wait
1520 2011-11-15 19:13:18 <justmoon> derp, mixed up the labels on the test cases
1521 2011-11-15 19:13:26 <justmoon> wondered why compressed key was faster ^^
1522 2011-11-15 19:13:31 <BlueMatt> roconnor_: ok, well I would still say it should be benchmarked first...
1523 2011-11-15 19:13:44 <sipa> BlueMatt: key recovery implies doing a key decompression
1524 2011-11-15 19:13:56 <justmoon> compressed key x 1,174 ops/sec ±0.72% (83 runs sampled)
1525 2011-11-15 19:13:57 <justmoon> uncompressed key x 1,236 ops/sec ±1.38% (83 runs sampled)
1526 2011-11-15 19:13:58 <BlueMatt> sipa: oh
1527 2011-11-15 19:13:58 <sipa> and key recovery is only 5% slower than verification
1528 2011-11-15 19:14:07 Litt has quit (Ping timeout: 255 seconds)
1529 2011-11-15 19:14:14 <justmoon> not a huge difference at all
1530 2011-11-15 19:14:24 Litt has joined
1531 2011-11-15 19:14:44 <sipa> 5.2%
1532 2011-11-15 19:15:00 <BlueMatt> +/- 2
1533 2011-11-15 19:15:21 <justmoon> there is a little bit of overhead, so someone should do a test in pure C++ (the above is JavaScript/C++, but BitcoinJS is now faster than bitcoind, so I don't think the JavaScript overhead is too crazy :P)
1534 2011-11-15 19:16:32 <roconnor_> BlueMatt: there is basically no way for miners prevent the use of compressed keys by whitelisting
1535 2011-11-15 19:17:05 <roconnor_> black listing won't work because there is not way to tell what data is a public key and what data just happens to look like a public key
1536 2011-11-15 19:17:09 <BlueMatt> roconnor_: I was under the impression that compressed keys were clearly different (largely due to the size difference thanks to compression)
1537 2011-11-15 19:17:10 Shaded has quit (Quit: Shaded)
1538 2011-11-15 19:17:47 <roconnor_> BlueMatt: right, if miners whitelist acceptable script templates (which they do now?) then you can filter out compressed keys
1539 2011-11-15 19:17:59 <BlueMatt> roconnor_: ok...
1540 2011-11-15 19:18:22 <BlueMatt> roconnor_: there is also the issue of, does bitcoinj and bitcoinjs support compressed keys, or would they need updated
1541 2011-11-15 19:18:32 <BlueMatt> also, how much space is actually saved?
1542 2011-11-15 19:18:44 <sipa> 32 bytes
1543 2011-11-15 19:18:48 <justmoon> bitcoinjs uses openssl, so we support, I can check for bitcoinj, two sec
1544 2011-11-15 19:18:49 <roconnor_> BlueMatt: 33 bytes vs 65 bytes
1545 2011-11-15 19:18:52 <BlueMatt> per tx out?
1546 2011-11-15 19:18:57 <justmoon> per signature
1547 2011-11-15 19:18:58 <roconnor_> per public key
1548 2011-11-15 19:19:02 <justmoon> right
1549 2011-11-15 19:19:03 Litt has quit (Ping timeout: 252 seconds)
1550 2011-11-15 19:19:12 Litt has joined
1551 2011-11-15 19:19:15 <sipa> per txout for spend-to-pubkey, per txin for spend-to-address
1552 2011-11-15 19:19:16 <BlueMatt> ok, so per txin
1553 2011-11-15 19:19:38 <justmoon> well OP_CHECKMULTISIG sounds like it'll get a lot of use
1554 2011-11-15 19:19:40 <roconnor_> BlueMatt: txin for address sens
1555 2011-11-15 19:19:46 <roconnor_> BlueMatt: txout for ip to ip sends
1556 2011-11-15 19:19:54 <BlueMatt> roconnor_: so essentially per txin
1557 2011-11-15 19:19:54 <sipa> roconnor_: what i said :)
1558 2011-11-15 19:20:05 <roconnor_> sipa:  ah sorry
1559 2011-11-15 19:20:24 <sipa> roconnor_: i suspect that it will just work on realnet as well
1560 2011-11-15 19:20:57 <sipa> there is the IsStandard() test, which rules out nonstandard txouts and txins, but it doesn't actually look inside the pubkeu daya
1561 2011-11-15 19:21:25 <sipa> for txins it just checks that it's only data pushes
1562 2011-11-15 19:22:19 <roconnor_> sipa: I'm sure it will work on mainnet as well
1563 2011-11-15 19:22:30 ThomasV has quit (Read error: Operation timed out)
1564 2011-11-15 19:22:47 <BlueMatt> roconnor_: what sipa just said
1565 2011-11-15 19:22:50 <BlueMatt> <sipa> roconnor_: i suspect that it will just work on realnet as well
1566 2011-11-15 19:23:12 <sipa> and he just confirmed my assumption
1567 2011-11-15 19:23:13 <roconnor_> er, I meant that I agree as well
1568 2011-11-15 19:23:16 BlueMatt has quit (Quit: Ex-Chat)
1569 2011-11-15 19:23:16 <roconnor_> :)
1570 2011-11-15 19:23:36 <sipa> we should make sure libbitcoin also supports it
1571 2011-11-15 19:23:46 <roconnor_> what is libbitcoin?
1572 2011-11-15 19:23:53 <sipa> genjix' reimplementation
1573 2011-11-15 19:24:12 <roconnor_> the nice thing about putting this on testnet, is that everyone now has access to this test case :)
1574 2011-11-15 19:24:13 <sipa> (C++, so i assume he also just uses OpenSSL)
1575 2011-11-15 19:24:17 Litt has quit (Ping timeout: 276 seconds)
1576 2011-11-15 19:24:20 <justmoon> sipa: he does
1577 2011-11-15 19:25:13 Litt has joined
1578 2011-11-15 19:25:31 Shaded has joined
1579 2011-11-15 19:26:52 <justmoon> ok, so yeah, bitcoinj does support it (BouncyCastle's ECCurce.decodePoint supports it)
1580 2011-11-15 19:27:49 erle- has joined
1581 2011-11-15 19:28:43 <justmoon> we should probably try and check for patent issues - if there are any then there may be modified versions of openssl floating around that don't have it
1582 2011-11-15 19:28:46 Shaded has quit (Client Quit)
1583 2011-11-15 19:28:48 vsrinivas has joined
1584 2011-11-15 19:29:03 <justmoon> also we need to check what minimum versions of openssl and bouncycastle you need
1585 2011-11-15 19:29:47 <justmoon> although the cat is kind of out of the bag :)
1586 2011-11-15 19:29:50 <sipa> justmoon: afaik the only modified version of openssl throws out EC entirely (RedHat & derivatives)
1587 2011-11-15 19:29:59 <justmoon> sipa: ok, that's good
1588 2011-11-15 19:30:33 <justmoon> http://cr.yp.to/ecdh/patents.html
1589 2011-11-15 19:30:48 <justmoon> "Popular rumor states that point compression is covered by a subsequent Vanstone-Mullin-Agnew patent: US patent 6141420, filed 1994.07.29, granted 2000.10.31. What the patent actually claims are ... [blah blah]"
1590 2011-11-15 19:31:03 Shaded has joined
1591 2011-11-15 19:31:12 <roconnor_> didn't realize he just violated a patent
1592 2011-11-15 19:31:21 <gavinandresen> sipa:  I'm trying to understand some code in CDB::Rewrite... in particular, CDB db(strFile.c_str(), "r");  ... which is never explicitly closed
1593 2011-11-15 19:31:41 <sipa> gavinandresen: its scope ends, no?
1594 2011-11-15 19:31:46 <justmoon> well, if that website is correct it is a *false* rumor that point compression is patented
1595 2011-11-15 19:31:47 <sipa> gavinandresen: so the destructor is called
1596 2011-11-15 19:31:48 <luke-jr> sipa: Gentoo has a USE flag; I don't know if it discards all EC or just parts
1597 2011-11-15 19:32:21 agricocb has joined
1598 2011-11-15 19:32:26 <gavinandresen> yes, and I'm worried the crash I'm seeing on the mac is because it's pdb handle is closed/deleted out from under it
1599 2011-11-15 19:32:27 <sipa> justmoon: that page is what made me decide to use key recovery for message signing (which depends on key decompression)
1600 2011-11-15 19:33:19 <gavinandresen> sipa: is there a problem with calling db.Close() instead of getting the Db* handle from mapDb[strFile] ?
1601 2011-11-15 19:33:30 <justmoon> hmm, actually, there is one client that doesn't support point compression - the browser-based stuff I wrote from scratch: bitcoinjs-lib
1602 2011-11-15 19:33:39 <justmoon> but I can easily port the bouncycastle stuff to add that
1603 2011-11-15 19:33:52 localhost has joined
1604 2011-11-15 19:33:53 <roconnor_> sipa: we should add a new bitcoin function that takes a signature and pushes upto 4 public keys onto the stack (plus pushes the number of keys pushed onto the stack)
1605 2011-11-15 19:34:32 <sipa> roconnor_: read https://gist.github.com/1262449
1606 2011-11-15 19:35:10 <roconnor_> sipa: :D
1607 2011-11-15 19:35:49 <sipa> gavinandresen: wait... there used to be an additional { } around the initial part of that function
1608 2011-11-15 19:35:54 <CIA-89> bitcoinjs/node-bitcoin-p2p: Stefan Thomas master * r7af7815 / lib/db/leveldb/storage.js : Reset spent index on database reset. (LevelDB) - http://git.io/3z9r8g
1609 2011-11-15 19:35:55 <CIA-89> bitcoinjs/node-bitcoin-p2p: Stefan Thomas master * rcf33711 / lib/schema/transaction.js :
1610 2011-11-15 19:35:55 <CIA-89> bitcoinjs/node-bitcoin-p2p: Simplified transaction input loading. Fixes #43.
1611 2011-11-15 19:35:55 <CIA-89> bitcoinjs/node-bitcoin-p2p: The race condition that this code was trying work around no longer
1612 2011-11-15 19:35:55 <CIA-89> bitcoinjs/node-bitcoin-p2p: exists thanks to more sensible block processing logic. - http://git.io/E2n41w
1613 2011-11-15 19:35:55 <CIA-89> bitcoinjs/node-bitcoin-p2p: Stefan Thomas master * r8adfeaf / benchmark/compressedkey.js : Benchmark on ECDSA key compression. - http://git.io/N8n9bw
1614 2011-11-15 19:35:55 <sipa> gavinandresen: not sure what happened to it
1615 2011-11-15 19:35:59 agricocb has quit (Remote host closed the connection)
1616 2011-11-15 19:36:00 larsivi has quit (Read error: Connection reset by peer)
1617 2011-11-15 19:36:40 <sipa> gavinandresen: sounds like you're right
1618 2011-11-15 19:36:44 vorlov has joined
1619 2011-11-15 19:36:56 agricocb has joined
1620 2011-11-15 19:39:02 <justmoon> Posted the benchmark plus results: https://github.com/bitcoinjs/node-bitcoin-p2p/commit/8adfeaf4374846889c567119b5d04ac381d576e0#commitcomment-722664
1621 2011-11-15 19:39:16 <gavinandresen> sipa:  hmmm.   Ran ok, but that isn't the problem....
1622 2011-11-15 19:39:17 Guest73681 is now known as topace__
1623 2011-11-15 19:39:32 <sipa> gavinandresen: should be fixed nonetheless
1624 2011-11-15 19:39:37 <gavinandresen> sipa: yup
1625 2011-11-15 19:39:50 <justmoon> I also tested the overhead - it's on the order of millions of ops/sec, so the results should be very accurate
1626 2011-11-15 19:40:54 Turingi has quit (Quit: Leaving)
1627 2011-11-15 19:41:19 agricocb has quit (Ping timeout: 240 seconds)
1628 2011-11-15 19:42:21 eueuuee has quit (Ping timeout: 265 seconds)
1629 2011-11-15 19:42:25 <jgarzik> neat
1630 2011-11-15 19:42:32 * jgarzik finds the MSFT research paper on bitcoin
1631 2011-11-15 19:42:49 <justmoon> jgarzik, seen the slashdot headline?
1632 2011-11-15 19:43:01 larsivi has joined
1633 2011-11-15 19:43:02 <jgarzik> justmoon: yeah, that's where I found the link
1634 2011-11-15 19:43:05 <justmoon> lol
1635 2011-11-15 19:43:10 <cjdelisle> slashdot is more or less like the register
1636 2011-11-15 19:43:15 <justmoon> ^^
1637 2011-11-15 19:43:22 <sipa> justmoon, roconnor_: to deploy it in bitcoind, the only thing to do is to have newly generated pubkeys be compressed, and use the hash160 of the compressed pubkey
1638 2011-11-15 19:43:30 <sipa> imho
1639 2011-11-15 19:44:23 <sipa> and make sure that CKeyStore's getpubkey also returns the compressed one for those
1640 2011-11-15 19:44:42 <justmoon> sipa: the wallet needs to know which form to use when spending outputs, no?
1641 2011-11-15 19:44:58 <justmoon> i.e. it needs to check for both hashes when scanning
1642 2011-11-15 19:45:08 <sipa> justmoon: not really
1643 2011-11-15 19:45:17 <roconnor_> sipa: that sounds about right
1644 2011-11-15 19:45:27 <sipa> you'll just have an adress->compressedpubkey in your wallet, or an address->normalpubkey
1645 2011-11-15 19:45:34 <sipa> there is never any confusion
1646 2011-11-15 19:45:39 <justmoon> right, got you
1647 2011-11-15 19:45:41 <roconnor_> but I'm not an expert on how wallet in the bitcoin client works
1648 2011-11-15 19:45:45 <jgarzik> reading the paper, sounds like old news
1649 2011-11-15 19:46:05 <sipa> roconnor_: leave that to me :)
1650 2011-11-15 19:46:38 <sipa> but not now - it's again way too interesting in here
1651 2011-11-15 19:46:43 <roconnor_> sipa: it was a big hack just to make the testcase :D
1652 2011-11-15 19:52:07 DaQatz has quit (Ping timeout: 240 seconds)
1653 2011-11-15 19:54:41 Shaded has quit (Quit: Shaded)
1654 2011-11-15 19:56:54 <makomk> sipa: that'd have interesting effects on public key import and export, surely?
1655 2011-11-15 19:57:34 <sipa> makomk: yes, indeed! that's a place were pubkeys are regenerated
1656 2011-11-15 19:58:20 sipa has left ("bbl")
1657 2011-11-15 20:05:48 <justmoon> aw, crap...: if (pubkey.length !== 65) return false;
1658 2011-11-15 20:05:53 <justmoon> gonna have to remove that
1659 2011-11-15 20:06:11 <justmoon> damn, it's so easy to screw up and include a forking difference accidentally
1660 2011-11-15 20:07:28 <CIA-89> bitcoinjs/node-bitcoin-p2p: Stefan Thomas master * r2e88377 / lib/scriptinterpreter.js :
1661 2011-11-15 20:07:28 <CIA-89> bitcoinjs/node-bitcoin-p2p: Removed pubkey length verification.
1662 2011-11-15 20:07:28 <CIA-89> bitcoinjs/node-bitcoin-p2p: Official client doesn't do it, neither should we. - http://git.io/Er5NHQ
1663 2011-11-15 20:07:28 <CIA-89> bitcoinjs/node-bitcoin-p2p: Stefan Thomas master * r74efeab / test/script.js : Test case for compressed key format. - http://git.io/xTqXJQ
1664 2011-11-15 20:07:37 <justmoon> ok, works now
1665 2011-11-15 20:08:23 RazielZ has quit (Quit: Leaving)
1666 2011-11-15 20:10:19 E-sense has joined
1667 2011-11-15 20:10:33 <gavinandresen> sipa: https://github.com/bitcoin/bitcoin/pull/636
1668 2011-11-15 20:11:13 Lexa has quit (Quit: Lexa)
1669 2011-11-15 20:12:01 <gavinandresen> luke-jr: ditto, that pull should be backported to 0.4, too
1670 2011-11-15 20:14:55 DaQatz has joined
1671 2011-11-15 20:15:43 Lexa has joined
1672 2011-11-15 20:21:44 <luke-jr> gavinandresen: looks good. I'll backport once it's in master
1673 2011-11-15 20:26:10 OneFixt has quit (Read error: Connection reset by peer)
1674 2011-11-15 20:26:23 OneFixt has joined
1675 2011-11-15 20:26:49 OneFixt is now known as Guest76535
1676 2011-11-15 20:27:10 vorlov has quit (Quit: vorlov)
1677 2011-11-15 20:29:40 pickett has joined
1678 2011-11-15 20:30:14 Metabank has joined
1679 2011-11-15 20:36:00 alexwaters has quit (Ping timeout: 244 seconds)
1680 2011-11-15 20:36:07 eueueue has joined
1681 2011-11-15 20:37:33 AlexWaters has joined
1682 2011-11-15 20:38:24 cronopio has quit (Ping timeout: 276 seconds)
1683 2011-11-15 20:44:42 wasabi2 has joined
1684 2011-11-15 20:46:55 wasabi has quit (Ping timeout: 248 seconds)
1685 2011-11-15 20:50:30 eueueue has quit (Ping timeout: 265 seconds)
1686 2011-11-15 20:51:43 gfinn has quit (Ping timeout: 248 seconds)
1687 2011-11-15 20:53:12 larsivi has quit (Read error: Connection reset by peer)
1688 2011-11-15 20:55:28 OneFixt_ has joined
1689 2011-11-15 20:57:44 Guest76535 has quit (Ping timeout: 240 seconds)
1690 2011-11-15 21:02:57 larsivi has joined
1691 2011-11-15 21:03:11 Shaded has joined
1692 2011-11-15 21:05:29 b4epoche has quit (Read error: Connection reset by peer)
1693 2011-11-15 21:06:02 b4epoche has joined
1694 2011-11-15 21:06:07 OneFixt_ has quit (Read error: Connection reset by peer)
1695 2011-11-15 21:06:18 OneFixt_ has joined
1696 2011-11-15 21:08:02 <etotheipi_> roconnor_, do you still have the code for recovering the candidate public keys for a signature?
1697 2011-11-15 21:08:24 Beremat has joined
1698 2011-11-15 21:08:49 <etotheipi_> I want to try signing the exact same message twice (with different random numbers) and verify it gives different public keys
1699 2011-11-15 21:08:54 Lolcust_Backup has quit (Quit: Nap time)
1700 2011-11-15 21:09:05 Lolcust_Backup has joined
1701 2011-11-15 21:12:01 denisx has joined
1702 2011-11-15 21:13:29 <gavinandresen> I could use help proofreading the 0.5 draft release notes:   https://sourceforge.net/projects/bitcoin/files/Bitcoin/bitcoin-0.5.0/test/
1703 2011-11-15 21:13:57 <gavinandresen> In particular, are the instructions on recovering from the wallet encryption bug clear?
1704 2011-11-15 21:14:33 <etotheipi_> gavinandresen, I think you shouldn't be so blatant in identifying how someone could exploit the wallet encryption problem
1705 2011-11-15 21:15:00 <denisx> ok, I will read it
1706 2011-11-15 21:15:06 <etotheipi_> I mean, it's enough to say that there was a problem with encryption, but to explicitly state that they are left in the wallet invites people to go looking for them
1707 2011-11-15 21:15:24 <gavinandresen> etotheipi_: I don't think being vague will help at all.  The bad guys will be able to figure it out.
1708 2011-11-15 21:15:30 <gmaxwell> etotheipi_: everyone who matters already knows.
1709 2011-11-15 21:15:31 <etotheipi_> maybe it's arbitrary... but I think you should focus on how to fix the problem, not identifying how someone could go about exploiting it
1710 2011-11-15 21:15:35 <gmaxwell> this channel has public logs.
1711 2011-11-15 21:16:10 <etotheipi_> right, people who want to know will find it...
1712 2011-11-15 21:16:37 PK has quit ()
1713 2011-11-15 21:16:40 <gmaxwell> gavinandresen: the start shut start get address shutdown backup   is clear but seems a little inexplicable. People may skip the seperate step 2 because it seems needless.
1714 2011-11-15 21:16:53 <etotheipi_> okay, I don't need to split hairs here... it just seems unnecessary... but probably irrelevant
1715 2011-11-15 21:17:01 <denisx> in the 3. paragraph bitcoin should have a capital letter
1716 2011-11-15 21:17:05 <gavinandresen> gmaxwell: you can't skip step 2... rewriting automatically shuts down
1717 2011-11-15 21:17:15 <denisx> I mean third listitem
1718 2011-11-15 21:17:37 <denisx> "Shut down bitcoin, then"
1719 2011-11-15 21:17:40 <gmaxwell> gavinandresen: you can let it shutdown, then backup. then startup and get new address.
1720 2011-11-15 21:17:51 <gmaxwell> Is the keypool not rewritten on that first shutdown?
1721 2011-11-15 21:18:16 <gavinandresen> gmaxwell: no, it is not.  unencrypted old private keys leaked into it when I coded it that way
1722 2011-11-15 21:18:31 datagutt has quit (Quit: Computer has gone to sleep.)
1723 2011-11-15 21:19:00 <gmaxwell> ::nods:: Perhaps you should say something why its important to do the backup after instead of first? otherwise thats easy to get wrong.
1724 2011-11-15 21:19:12 <gmaxwell> (e.g. because you read it expecting to not need an extra restart)
1725 2011-11-15 21:19:47 <etotheipi_> isn't it a good idea to transfer funds to the new address?  looks like you only recommend creating the new address
1726 2011-11-15 21:20:02 <gavinandresen> gmaxwell: I'll mention it is important to backup AFTER asking for a new address because asking for a new address triggers creation of new private keys
1727 2011-11-15 21:20:06 <gmaxwell> etotheipi_: If you had a previously encrypted wallet.dat that might have been copied or stolen you should send all of your bitcoins to yourself using a new bitcoin address.
1728 2011-11-15 21:20:11 <gmaxwell> gavinandresen: ::nods::
1729 2011-11-15 21:20:53 <etotheipi_> gmaxwell, I think it is important to mention that user should move all their funds to a newly-created address, if they suspect anyone could've gotten their old wallet file
1730 2011-11-15 21:20:58 <etotheipi_> or just do a blanket recommendation
1731 2011-11-15 21:20:59 wboy1 has quit (Ping timeout: 258 seconds)
1732 2011-11-15 21:21:08 <gavinandresen> etotheipi_:  Maybe I should add a step between 2 and 3 : "
1733 2011-11-15 21:21:23 <gavinandresen> if your wallet may have been compromised send allyour bitcoins to yourself....
1734 2011-11-15 21:21:25 Acciaio has quit (Quit: Ex-Chat)
1735 2011-11-15 21:21:34 <gmaxwell> etotheipi_: I just pasted the text it has on that !
1736 2011-11-15 21:21:54 abragin has quit ()
1737 2011-11-15 21:21:54 <gmaxwell> But yea, making it a step makes sense.
1738 2011-11-15 21:22:00 <AlexWaters> so molecular and I are trying to figure out a way to create a script that i can plug in a pull request number and it will merge that pull request into my local and then push it to the bitcoin testing remote
1739 2011-11-15 21:22:17 <gmaxwell> AlexWaters: oh god please do
1740 2011-11-15 21:22:21 <AlexWaters> how can we find out the referenced branch from a pull request like this https://github.com/bitcoin/bitcoin/pull/636
1741 2011-11-15 21:22:41 <AlexWaters> apparently the little green i only shows up if you have merge access
1742 2011-11-15 21:22:59 abragin has joined
1743 2011-11-15 21:23:02 <AlexWaters> so we need to grab gavin's bitcoin-git.git from that page
1744 2011-11-15 21:23:18 <AlexWaters> any suggestions?
1745 2011-11-15 21:23:49 <etotheipi_> gavinandresen, maybe say "if you ever backed up your wallet to a public location", ... users may not realize what "compromised" means
1746 2011-11-15 21:23:57 <gavinandresen> gmaxwell: god no why not?  A little automation there seems like a good idea to me...
1747 2011-11-15 21:24:17 Zarutian has quit (Ping timeout: 245 seconds)
1748 2011-11-15 21:24:17 <gmaxwell> please do!
1749 2011-11-15 21:24:19 <gmaxwell> not not.
1750 2011-11-15 21:24:22 <molecular> gmaxwell said we _should_ do it
1751 2011-11-15 21:24:43 <gmaxwell> Getting code from pull requests is a pain in the ass that has annoyed me many times. I'd assumed I was github stupid though.
1752 2011-11-15 21:24:52 <gavinandresen> gmaxwell: sorry, eyes getting tired....
1753 2011-11-15 21:25:07 <gmaxwell> ha. Well I'm usually saying "oh god no" to most things. :)
1754 2011-11-15 21:25:51 <luke-jr> gavinandresen: I agree that the exploit vector is too detailed in the release notes, and also I think "steal the encrypted wallet.dat file" is wrong; s/steal/obtain/
1755 2011-11-15 21:26:02 <AlexWaters> the reason being, we ultimately want to make a dynamic "machine" on bitcointesting.org that will allow everyday regular people to create a bitcoin.exe from any pull
1756 2011-11-15 21:26:03 <gavinandresen> AlexWaters: github has APIs for everything, I'm sure there is a way.  I think tcatm did some work to auto-generate the contributors page that might be relevant
1757 2011-11-15 21:26:36 <AlexWaters> where they would plug in a pull request number, and jenkins would build them a binary, and a script would post a link to that binary when complete
1758 2011-11-15 21:26:58 <luke-jr> perhaps: "The wallet encryption feature introduced in Bitcoin version 0.4.0 did not in some cases sufficiently secure the private keys. An attacker who has obtained a copy of the encrypted wallet.dat file might be able to recover some or all of the unencrypted keys and steal the associated coins."
1759 2011-11-15 21:27:10 <AlexWaters> gavinandresen: ok i asked in their channel, but i think it's time i dig into the api a little anyways
1760 2011-11-15 21:27:43 <gavinandresen> luke-jr: I like that
1761 2011-11-15 21:27:56 <etotheipi_> luke-jr, thanks for backing me up!
1762 2011-11-15 21:28:22 <etotheipi_> but yeah... less info on the attack, more info on protecting yourself
1763 2011-11-15 21:29:21 <luke-jr> gavinandresen: I'd also rephrase the 2nd paragraph slightly: "If it is possible that someone may have obtained a copy of your wallet.dat, you should send all of your bitcoins to yourself at a new bitcoin address, and immediately cease using any addresses generated prior to upgrading."
1764 2011-11-15 21:29:42 Zarutian has joined
1765 2011-11-15 21:30:15 <etotheipi_> ...If it is possibel that someone may have obtained a copy of your *encrypted* wallet.dat, you should ..."
1766 2011-11-15 21:30:54 <luke-jr> etotheipi_: it's good advice even if someone obtained a non-encrypted copy :p
1767 2011-11-15 21:31:25 <etotheipi_> luke-jr, while I agree with you... if they copied their unencrypted wallet, they probably already know this
1768 2011-11-15 21:31:55 <etotheipi_> I just envision hearing someone say "oh, I thought you didn't have to do that if your wallet was already encrypted" ...
1769 2011-11-15 21:32:02 <etotheipi_> some people don't read/pay attention well
1770 2011-11-15 21:32:53 sytse has quit (Ping timeout: 240 seconds)
1771 2011-11-15 21:33:13 <luke-jr> true
1772 2011-11-15 21:33:50 sytse has joined
1773 2011-11-15 21:34:49 <molecular> "you should send all of your bitcoins to yourself at a new bitcoin address" <- how does that help if the address comes from the address pool that is also in the copied wallet?
1774 2011-11-15 21:35:10 <gavinandresen> it doesn't come from the old address pool...
1775 2011-11-15 21:35:27 <gavinandresen> when the old wallet is rewritten all of the old pool addresses are 'used up'
1776 2011-11-15 21:35:29 <etotheipi_> all address-pool keys are being marked as used so only new, encrypted-born keys will be generated
1777 2011-11-15 21:35:35 minimoose has quit (Quit: minimoose)
1778 2011-11-15 21:35:57 <luke-jr> molecular: that's why it's AFTER upgrading
1779 2011-11-15 21:36:01 <molecular> I see
1780 2011-11-15 21:36:11 <molecular> sorry, wasn't following before (or the issue).
1781 2011-11-15 21:36:13 <gavinandresen> New README.txt uploaded
1782 2011-11-15 21:36:14 <luke-jr> gavinandresen: did you get paragraph 2 + revision from etotheipi_?
1783 2011-11-15 21:36:24 <luke-jr> "If it is possible that someone may have obtained a copy of your encrypted wallet.dat, you should send all of your bitcoins to yourself at a new bitcoin address, and immediately cease using any addresses generated prior to upgrading."
1784 2011-11-15 21:37:19 wboy1 has joined
1785 2011-11-15 21:37:27 <gavinandresen> How about:  "and stop using old addresses."
1786 2011-11-15 21:37:29 agricocb has joined
1787 2011-11-15 21:38:19 <luke-jr> I'd be more specific about old
1788 2011-11-15 21:38:28 <luke-jr> "If it is possible that someone may have obtained a copy of your encrypted wallet.dat, you should send all of your bitcoins to yourself at a new bitcoin address, and stop using any addresses generated prior to upgrading."
1789 2011-11-15 21:38:33 <luke-jr> also, mention the CVE number :p
1790 2011-11-15 21:39:30 niekie has quit (Read error: Connection reset by peer)
1791 2011-11-15 21:39:57 <luke-jr> gavinandresen: got a draft for the general security announcement?
1792 2011-11-15 21:40:10 <AlexWaters> what about it is possible, though unlikely? how concerned do you want the userbase?
1793 2011-11-15 21:41:25 <luke-jr> AlexWaters: that phrase doesn't make sense where I put "possible"
1794 2011-11-15 21:41:41 wboy1 has quit (Ping timeout: 240 seconds)
1795 2011-11-15 21:41:49 <AlexWaters> "It is possible, though unlikely that someone may have obtained..."
1796 2011-11-15 21:41:53 <gavinandresen> I want the userbase to be extremely concerned, it is very likely they will lose bitcoins if they expose a 0.4.0 encrypted wallet.dat
1797 2011-11-15 21:42:08 <luke-jr> "If it is possible that someone may have obtained a copy of your encrypted wallet.dat (for example, you backed it up to a public location-- NOT recommended even with this fixed!), you should send all of your bitcoins to yourself at a new bitcoin address, and stop using any addresses generated prior to upgrading."
1798 2011-11-15 21:42:21 <gavinandresen> luke-jr: I added the CVE number to the MAJOR BUG FIX heading
1799 2011-11-15 21:42:48 <luke-jr> AlexWaters: I wrote "If" :p
1800 2011-11-15 21:42:57 <luke-jr> without "If", it doesn't make sense
1801 2011-11-15 21:43:04 <tcatm> AlexWaters: have you found the github api for pull requests?
1802 2011-11-15 21:43:09 niekie has joined
1803 2011-11-15 21:43:17 <luke-jr> http://developer.github.com/v3/pulls/
1804 2011-11-15 21:43:21 <luke-jr> cmon, 5 seconds of googlign
1805 2011-11-15 21:43:22 <AlexWaters> luke-jr: ok i'm just being nitpicky anyways. haha
1806 2011-11-15 21:43:59 <AlexWaters> wasn't looking for it yet =P i'm working on my resume / elance portfolio
1807 2011-11-15 21:45:59 wboy1 has joined
1808 2011-11-15 21:46:44 niekie has quit (Read error: Connection reset by peer)
1809 2011-11-15 21:47:21 <gavinandresen> Thanks y'all, I'm going to declare the README good enough for now.
1810 2011-11-15 21:47:45 <gavinandresen> ... and do some sanity testing of the rc4 binaries.. wheeee....
1811 2011-11-15 21:48:28 niekie has joined
1812 2011-11-15 21:49:23 chuckaboo has joined
1813 2011-11-15 21:50:03 zapnap has quit (Remote host closed the connection)
1814 2011-11-15 21:50:45 wboy1 has quit (Ping timeout: 252 seconds)
1815 2011-11-15 21:51:13 Snapman is now known as Snapman[afkers]
1816 2011-11-15 21:51:29 <luke-jr> gavinandresen: what about those fixes to pull? :P
1817 2011-11-15 21:52:25 <gavinandresen> luke-jr: what about them?  I want to see if it is an OSX-specific problem, or if I need to re-spin rc5 with those fixes for linux/windows
1818 2011-11-15 21:52:59 <gavinandresen> luke-jr: and I want to see if I can reproduce the wallet.dat.rewrite problem that was reported
1819 2011-11-15 21:53:13 OneFixt_ has quit (Changing host)
1820 2011-11-15 21:53:13 OneFixt_ has joined
1821 2011-11-15 21:53:19 OneFixt_ is now known as OneFixt
1822 2011-11-15 21:54:39 amiller has quit (Ping timeout: 248 seconds)
1823 2011-11-15 21:56:35 niekie has quit (Read error: Connection reset by peer)
1824 2011-11-15 21:57:13 marf_away2 has joined
1825 2011-11-15 21:57:21 marf_away has quit (Read error: Connection reset by peer)
1826 2011-11-15 21:57:27 wboy1 has joined
1827 2011-11-15 21:58:34 niekie has joined
1828 2011-11-15 22:00:53 niekie has quit (Read error: Connection reset by peer)
1829 2011-11-15 22:01:11 amiller has joined
1830 2011-11-15 22:02:00 <gavinandresen> all righty, not just OSX....
1831 2011-11-15 22:02:55 <CIA-89> bitcoin: Gavin Andresen master * r8d09231 / src/db.cpp : Fix crash-on-osx-on-shutdown bug. And cleanup CDB handling in Rewrite. - http://git.io/EFRiHA
1832 2011-11-15 22:02:56 <CIA-89> bitcoin: Gavin Andresen master * rc4de918 / (bitcoin-qt.pro src/db.cpp): Tweak handling of boost filesystem versions - http://git.io/prOC0Q
1833 2011-11-15 22:02:56 <CIA-89> bitcoin: Gavin Andresen master * r1b93ea0 / (bitcoin-qt.pro src/db.cpp):
1834 2011-11-15 22:02:56 <CIA-89> bitcoin: Merge pull request #636 from gavinandresen/master
1835 2011-11-15 22:02:56 <CIA-89> bitcoin: Fix crash-on-wallet-upgrade bug on OSX - http://git.io/OYcpJQ
1836 2011-11-15 22:03:05 <denisx> I feel discriminiated
1837 2011-11-15 22:03:12 <denisx> I feel discriminated
1838 2011-11-15 22:03:28 Sedra- has quit (Ping timeout: 256 seconds)
1839 2011-11-15 22:04:02 Sedra has joined
1840 2011-11-15 22:04:15 erle- has quit (Quit: erle-)
1841 2011-11-15 22:05:52 larsivi has quit (Read error: Operation timed out)
1842 2011-11-15 22:05:54 AStove has quit ()
1843 2011-11-15 22:06:29 OneFixt has quit (Read error: Connection reset by peer)
1844 2011-11-15 22:06:37 niekie has joined
1845 2011-11-15 22:06:42 OneFixt has joined
1846 2011-11-15 22:07:08 OneFixt is now known as Guest30114
1847 2011-11-15 22:09:42 <denisx> anybody who visits the 28c3?
1848 2011-11-15 22:09:42 <cjdelisle> 17:04 <@rssbawt> [TR] 'Devastating' protocol flaw could paralyze Bitcoin system -
1849 2011-11-15 22:09:46 <cjdelisle> good job TR
1850 2011-11-15 22:10:09 <gmaxwell> how to map people who _want_ to buy bitcoin
1851 2011-11-15 22:10:11 <gavinandresen> what is TR?
1852 2011-11-15 22:10:12 OneFixt_ has joined
1853 2011-11-15 22:10:13 OneFixt_ has quit (Changing host)
1854 2011-11-15 22:10:13 OneFixt_ has joined
1855 2011-11-15 22:10:20 <cjdelisle> The Register
1856 2011-11-15 22:10:31 <gmaxwell> Obviously the register wants to buy bitcoin.
1857 2011-11-15 22:10:32 <cjdelisle> basicly like the slashdot of sensational headlines
1858 2011-11-15 22:10:39 pickett has quit (Ping timeout: 248 seconds)
1859 2011-11-15 22:11:20 Guest30114 has quit (Ping timeout: 240 seconds)
1860 2011-11-15 22:11:33 <gmaxwell> It's kinda ironic that some research dorks right a paper on something they don't even _claim_ is a vulnerability, and they get headlines like that... we fix an actual vulnerablilty, and no one will notice.
1861 2011-11-15 22:12:44 <luke-jr> XD
1862 2011-11-15 22:12:54 <denisx> gmaxwell: what did you fix?
1863 2011-11-15 22:13:03 larsivi has joined
1864 2011-11-15 22:13:06 <gmaxwell> Me? I fixed nothing. That was the royal we.
1865 2011-11-15 22:13:16 <gmaxwell> denisx: the wallet encryption stuff.
1866 2011-11-15 22:13:28 <gmaxwell> It's a real security flaw. Not a serious one for most users, but real.
1867 2011-11-15 22:13:29 <denisx> but that is only an implementation detail
1868 2011-11-15 22:13:34 <justmoon> gmaxwell, the royal we would mean it *was* you alone
1869 2011-11-15 22:13:43 <luke-jr> gavinandresen: maybe put something on bitcoin.org frontpage explaining that Microsoft's "flaw" is not a real issue right now, and easily addressed if it becomes one?
1870 2011-11-15 22:14:00 <gmaxwell> justmoon: hmph. You're right!
1871 2011-11-15 22:14:03 * cjdelisle is thinking of a nice 1-liner
1872 2011-11-15 22:14:13 <justmoon> gmaxwell, :P - I gotta go to bed, good night all!
1873 2011-11-15 22:14:27 <gavinandresen> luke-jr: why do I have to do everything?  ask tcatm....
1874 2011-11-15 22:14:44 <gmaxwell> luke-jr: Bitcoin developers deny deadly flaw, offer excuses.
1875 2011-11-15 22:14:57 <gavinandresen> luke-jr: or, better, submit a pull request to him (I think he's got the bitcoin.org home page setup under source control)
1876 2011-11-15 22:15:09 <gmaxwell> luke-jr: go on the forum, get the microsoft researchers to give you a quote.. use that.
1877 2011-11-15 22:15:21 <gmaxwell> They're being victimized here too.
1878 2011-11-15 22:15:49 <luke-jr> gavinandresen: I don't know who has what access :P
1879 2011-11-15 22:16:02 <gavinandresen> luke-jr: sorry for being grumpy.  I'm grumpy today.
1880 2011-11-15 22:16:12 <gavinandresen> ^today^yesterday
1881 2011-11-15 22:16:18 <luke-jr> gmaxwell: good idea
1882 2011-11-15 22:16:21 <gavinandresen> on second thought: ^today^this week
1883 2011-11-15 22:16:25 <CIA-89> bitcoin: Gavin Andresen 0.4.x * r1aafd7464f67 bitcoind-stable/src/db.cpp: Fix crash-on-osx-on-shutdown bug. And cleanup CDB handling in Rewrite.
1884 2011-11-15 22:16:27 <CIA-89> bitcoin: Gavin Andresen 0.4.x * r831d24a19d66 bitcoind-stable/src/db.cpp: Tweak handling of boost filesystem versions
1885 2011-11-15 22:16:51 <gavinandresen> building rc5 candidates with that fix right now, by the way
1886 2011-11-15 22:18:39 Metabank has quit (Quit: Page closed)
1887 2011-11-15 22:19:41 <parus> Hey! :) I might need some help. I'm using Debian as OS and tried to run the bitcoin software. bitcoin and bitcoind. And it simply doesnt start up. Works perfectly under Arch Linux and Ubuntu!
1888 2011-11-15 22:19:51 <parus> No errors. Simply not responding.
1889 2011-11-15 22:20:01 <parus> And no GUI for 'bitcoin'
1890 2011-11-15 22:20:16 <luke-jr> parus: use the Debian package for bitcoind
1891 2011-11-15 22:20:27 <parus> i have tried it
1892 2011-11-15 22:20:28 <parus> same here
1893 2011-11-15 22:20:29 <luke-jr> parus: 'bitcoin' is deprecated and no longer supported
1894 2011-11-15 22:20:38 <gmaxwell> luke-jr: also the response WRT the microsoft thing should link to or inline this comic: http://www.smbc-comics.com/index.php?db=comics&id=2095
1895 2011-11-15 22:20:45 <luke-jr> parus: post your ~/.bitcoin/debug.log somewhere
1896 2011-11-15 22:21:54 <gmaxwell> (as a way of pointing out that the issue they're discussing is one of economic incentives, not computer security, and that the study of economic incentives isn't an exact science)
1897 2011-11-15 22:22:12 <luke-jr> lol
1898 2011-11-15 22:24:22 <gmaxwell> We can also give the MSFT folks a plug in that iff the situation they're talking about were to actually become a problem, they've contributed a solution which the community of bitcoin users could deploy.
1899 2011-11-15 22:24:33 <cjdelisle> "A Bitcoin transaction is like an extra merital affair, even if the majority keep the secret it still leaks out."
1900 2011-11-15 22:24:46 [eval] is now known as [ev[a]l]
1901 2011-11-15 22:25:24 <luke-jr> gavinandresen: btw, don't forget to be careful with tags. sipa had to merge rc3 after you again tagged off-branch :P
1902 2011-11-15 22:26:33 <gmaxwell> (although their solution has highest operating cost of anything I've heard for this class of problem, not as a fault of their invention but because of the cost of rewarding the relayer chain at all)
1903 2011-11-15 22:26:36 <parus> luke-jr: http://pastebin.com/ctGnYzBk
1904 2011-11-15 22:27:03 <luke-jr> parus: looks like it's still loading
1905 2011-11-15 22:27:20 <gmaxwell> parus: start with -nodnsseed
1906 2011-11-15 22:27:20 copumpkin has quit (Remote host closed the connection)
1907 2011-11-15 22:27:26 graingert1 has joined
1908 2011-11-15 22:27:38 <gmaxwell> oh it's 3.24, nevermind
1909 2011-11-15 22:28:07 <roconnor_> etotheipi_: I still have the code for recovering the candidate public keys for a signature
1910 2011-11-15 22:28:11 <cjdelisle> "A Bitcoin transaction is like a dirty secret, even if the majority keep quiet it still leaks out. Anyway miners, the nodes who are incentivised to keep secrets, make up a tiny fraction of the network." <-- that should explain it in layman's terms
1911 2011-11-15 22:28:54 <gmaxwell> cjdelisle: they aren't very well incentivised either... currently they aren't really at all, but even in the future—
1912 2011-11-15 22:29:14 <cjdelisle> yea, I didn't want to complicate it though
1913 2011-11-15 22:29:21 <cjdelisle> you know, people need a headline ;)
1914 2011-11-15 22:29:23 <gmaxwell> because unless you really think you are the only miner with the txn, it's pretty likely that other people have it, and it's always likely that someone other than you will solve the current block anyways.
1915 2011-11-15 22:30:03 <graingert1> if you want a tx included in a block you are incetivised to tell all the miners
1916 2011-11-15 22:30:11 <parus> gmaxwell: thank you so much
1917 2011-11-15 22:30:16 <parus> works perfectly now
1918 2011-11-15 22:30:19 <gmaxwell> parus: did that fix it?
1919 2011-11-15 22:30:21 <parus> yes
1920 2011-11-15 22:30:25 <gmaxwell> ah yea we turned that on in .24
1921 2011-11-15 22:30:30 <gmaxwell> ha. Two people in one day.
1922 2011-11-15 22:30:34 <gmaxwell> parus: do you use opendns?
1923 2011-11-15 22:30:51 <gmaxwell> and/or do you have a firewall that blocks TCP port 53?
1924 2011-11-15 22:30:59 <cjdelisle> oh
1925 2011-11-15 22:31:06 <cjdelisle> everyone blocks TCP 53
1926 2011-11-15 22:31:14 <cjdelisle> hell verizon blocks TCP:53
1927 2011-11-15 22:31:15 <edcba> is there some list of bitcoin script use. i mean use cases
1928 2011-11-15 22:31:16 <gmaxwell> cjdelisle: nah, not outbound.
1929 2011-11-15 22:31:27 <parus> gmaxwell: not afaik
1930 2011-11-15 22:31:40 <gmaxwell> cjdelisle: without tcp:53 to your dns server you can't get dns replys larger than about 500 bytes.
1931 2011-11-15 22:31:43 <cjdelisle> relying in tcp dns is a bad plan
1932 2011-11-15 22:32:10 <gmaxwell> cjdelisle: yea, well, only one of the dnsseeds gives jumbo responses and its apparently been removed from the code.
1933 2011-11-15 22:32:23 <gmaxwell> I think bluematt was looking into the failure on hangs.
1934 2011-11-15 22:32:38 <gmaxwell> Weird that two people would report it in one day though.
1935 2011-11-15 22:33:54 <parus> gmaxwell: has been buggin me a while however... at least two weaks. but now i really wanted to see that fixed.
1936 2011-11-15 22:34:06 <parus> *weeks
1937 2011-11-15 22:34:30 <gmaxwell> parus: 0.5 probably won't have the problem due to an accidental workaround, but I expect the underlying bug will probably be fixed soon too.
1938 2011-11-15 22:34:48 <parus> thanks :)
1939 2011-11-15 22:34:59 OneFixt_ is now known as OneFixt
1940 2011-11-15 22:35:23 <gmaxwell> parus: actually, if you have a chance could you test the 0.5 release candidate and see if it works for you without the switch?
1941 2011-11-15 22:35:48 <parus> yes, ill try that right now
1942 2011-11-15 22:35:51 <gmaxwell> We might want to mention it in the release notes... god knows there might be a large base of people that didn't update to 3.24/and .4 due to hitting this.
1943 2011-11-15 22:37:52 <gmaxwell> While working on Wikpedia stuff sometimes I'd obvlously break something that (from checking logs I could tell) impacted ohh a good million people or so before I noticed and fixed it, ... and we'd only get 2-3 emailed reports.
1944 2011-11-15 22:39:21 <cjdelisle> oh you work on mediawiki engine?
1945 2011-11-15 22:41:15 zeiris has quit (Ping timeout: 276 seconds)
1946 2011-11-15 22:43:00 <parus> uhm, is the .5 release the one in the official git repo?
1947 2011-11-15 22:44:27 <parus> oh
1948 2011-11-15 22:44:30 <parus> i see m)
1949 2011-11-15 22:44:32 <parus> sry
1950 2011-11-15 22:45:48 <luke-jr> cjdelisle: TCP DNS is basically required now
1951 2011-11-15 22:46:42 <luke-jr> gmaxwell: what commit removed the problematic DNS server?
1952 2011-11-15 22:46:50 <luke-jr> cjdelisle: due to IPv6 and DNSSEC*
1953 2011-11-15 22:47:39 <gmaxwell> https://github.com/bitcoin/bitcoin/commit/02d630c37f6155220ca4265faa7964c689a01be9
1954 2011-11-15 22:47:45 <gmaxwell> luke-jr: ^
1955 2011-11-15 22:48:07 copumpkin has joined
1956 2011-11-15 22:49:31 agricocb has quit (Quit: Leaving.)
1957 2011-11-15 22:50:05 eoss has joined
1958 2011-11-15 22:50:13 eoss has quit (Changing host)
1959 2011-11-15 22:50:13 eoss has joined
1960 2011-11-15 22:50:31 wolfspraul has quit (Quit: leaving)
1961 2011-11-15 22:53:46 <luke-jr> I'd be glad to host one btw
1962 2011-11-15 22:55:07 OneFixt has quit (Read error: Connection reset by peer)
1963 2011-11-15 22:55:19 OneFixt has joined
1964 2011-11-15 22:55:39 <CIA-89> bitcoin: Jeff Garzik 0.4.x * r2bf36b4e7dd0 bitcoind-stable/src/net.cpp: Remove vladimir's DNS seed, at his request.
1965 2011-11-15 22:55:45 OneFixt is now known as Guest45424
1966 2011-11-15 22:57:09 Guest45424 has quit (Changing host)
1967 2011-11-15 22:57:09 Guest45424 has joined
1968 2011-11-15 22:58:32 abragin has quit ()
1969 2011-11-15 22:59:01 Snapman[afkers] is now known as Snapman
1970 2011-11-15 22:59:21 Guest45424 is now known as OneFixt
1971 2011-11-15 23:03:27 tower has quit (Ping timeout: 248 seconds)
1972 2011-11-15 23:05:25 somuchwin has joined
1973 2011-11-15 23:05:34 somuchwin2 has quit (Ping timeout: 252 seconds)
1974 2011-11-15 23:05:48 gp5st has quit (Quit: Leaving.)
1975 2011-11-15 23:07:25 Kolky has quit (Quit: Bye bye!)
1976 2011-11-15 23:08:41 tower has joined
1977 2011-11-15 23:09:09 <hippich> http://faucet.yepcorp.com/ -- open source if anyone interested.
1978 2011-11-15 23:09:38 <luke-jr> hippich: only 0.01 NMC? lame
1979 2011-11-15 23:09:59 <hippich> this is -dev channel? :)
1980 2011-11-15 23:10:06 <luke-jr> hippich: pretty sure when Bitcoin prices were as low as NMC, faucet did 5 BTC
1981 2011-11-15 23:10:18 <hippich> nmc is just for testing.
1982 2011-11-15 23:10:37 <hippich> some one gave me 50 nmc and i put 'em there.
1983 2011-11-15 23:10:39 <hippich> just hour ago.
1984 2011-11-15 23:11:00 ThomasV has joined
1985 2011-11-15 23:11:09 <hippich> this is open source faucet. you can hook it into bitcoin/namecoint/ixcoin/whatevercoin
1986 2011-11-15 23:11:23 <hippich> https://github.com/hippich/Faucet/blob/master/faucet_sample.conf
1987 2011-11-15 23:12:25 <luke-jr> hippich: so you don't plan to run a NMC faucet for real?
1988 2011-11-15 23:12:39 <luke-jr> hippich: btw, ixcoin is a scam, so I wouldn't use it as an example :P
1989 2011-11-15 23:12:41 <hippich> i can keep it running. it do not eat many resources.
1990 2011-11-15 23:12:47 <hippich> ok :)
1991 2011-11-15 23:13:02 <hippich> sample config uses bitcoin as example :)
1992 2011-11-15 23:13:06 <hippich> pretty sane default :)
1993 2011-11-15 23:13:47 <luke-jr> hippich: so if I send 1000 NMC to N1uXWnZHKsV9ks7LKjeusDZ21zGKhiSBmC, you won't cash it out yourself? :p
1994 2011-11-15 23:14:06 <hippich> nope :)
1995 2011-11-15 23:14:15 <luke-jr> will you up the offer to 1 NMC at least? :p
1996 2011-11-15 23:14:22 <hippich> no prob
1997 2011-11-15 23:14:25 <luke-jr> k
1998 2011-11-15 23:14:40 <hippich> balance on site updates once in 5 minutes.
1999 2011-11-15 23:14:41 <luke-jr> sent
2000 2011-11-15 23:16:35 <hippich> updated amount to 1 nmc
2001 2011-11-15 23:16:48 <luke-jr> We give away 1 Namecoins to each unique visitor. <-- plurality bug :P
2002 2011-11-15 23:17:08 <hippich> submit patch :))))
2003 2011-11-15 23:17:09 <hippich> haha :)
2004 2011-11-15 23:17:16 <hippich> sorry, i am bad at english.
2005 2011-11-15 23:17:18 <hippich> how it should be?
2006 2011-11-15 23:17:24 <luke-jr> "1 Namecoin"
2007 2011-11-15 23:17:31 <hippich> ah.
2008 2011-11-15 23:17:33 <hippich> well..
2009 2011-11-15 23:17:38 <hippich> i will add (s)
2010 2011-11-15 23:17:39 <luke-jr> P.S. better Gitorious than GitHub ;)
2011 2011-11-15 23:17:41 <helo> or for that matter, some would argue "5 Namecoin" is correct too
2012 2011-11-15 23:17:45 <luke-jr> pfft, that's cheating XD
2013 2011-11-15 23:17:58 <hippich> :)
2014 2011-11-15 23:18:05 tower has quit (Ping timeout: 240 seconds)
2015 2011-11-15 23:18:11 <hippich> i did it during office hours.. it is already "cheating" :)
2016 2011-11-15 23:18:41 <luke-jr> …
2017 2011-11-15 23:19:01 <luke-jr> that's not just cheating
2018 2011-11-15 23:19:11 <luke-jr> it means someone else owns the code, and you don't have authority to license it
2019 2011-11-15 23:19:47 <luke-jr> so I hope you're "just kidding"
2020 2011-11-15 23:19:47 <hippich> then piece of shit i had to fix over weekend not owned by "someone" :))
2021 2011-11-15 23:19:48 <luke-jr> <.<
2022 2011-11-15 23:20:19 <cjdelisle> 17:40 < luke-jr> cjdelisle: TCP DNS is basically required now <--- hahaha reason #4 why dnssec is a bad joke
2023 2011-11-15 23:21:00 <luke-jr> cjdelisle: it's ok, BSD came up with a stateless TCP for it :P
2024 2011-11-15 23:21:09 <cjdelisle> LOL
2025 2011-11-15 23:21:28 <luke-jr> which I'm planning to use (the concepts from) for my pool software
2026 2011-11-15 23:21:38 <luke-jr> to resist DDoS
2027 2011-11-15 23:21:42 <cjdelisle> hehehe
2028 2011-11-15 23:21:50 <cjdelisle> stateless tcp -> lols
2029 2011-11-15 23:22:22 MimeNarrator has quit (Remote host closed the connection)
2030 2011-11-15 23:22:39 <cjdelisle> Ya know there's a reason why the paper on it is called "How bad ideas turn into crappy protocols"
2031 2011-11-15 23:23:09 graingert1 has quit (Read error: Connection reset by peer)
2032 2011-11-15 23:23:36 copumpkin has quit (Quit: Computer has gone to sleep.)
2033 2011-11-15 23:25:52 justmoon has quit (Ping timeout: 248 seconds)
2034 2011-11-15 23:25:55 <cjdelisle> http://isoc.org/wp/ispcolumn/?p=217
2035 2011-11-15 23:26:10 <cjdelisle> that's the one
2036 2011-11-15 23:26:12 <cjdelisle> "Turning a Bad Idea into a Useless Tool"
2037 2011-11-15 23:26:36 <cjdelisle> like when the paper on it says "Turning a Bad Idea into a Useless Tool" you might want to consider the possibility that it won't work.
2038 2011-11-15 23:26:44 <hippich> Current Namecoin balance: 1050.7485.
2039 2011-11-15 23:26:48 <hippich> here we go
2040 2011-11-15 23:27:20 tower has joined
2041 2011-11-15 23:29:36 sacarlson has quit (Ping timeout: 248 seconds)
2042 2011-11-15 23:29:44 justmoon has joined
2043 2011-11-15 23:32:02 MimeNarrator has joined
2044 2011-11-15 23:33:26 <parus> gmaxwell: i tried to build the most recent candidate, but it keeps breaking all the time, because some dependencies are too new/too old.
2045 2011-11-15 23:34:58 <parus> like miniupnpc, debian wheezy gives me 1.5 and dependency is 1.6.
2046 2011-11-15 23:34:58 <CIA-89> bitcoin: Luke Dashjr * r45ac2d8b684b gentoo/net-p2p/bitcoind/ (Manifest bitcoind-0.4.1_rc4.ebuild): net-p2p/bitcoind: 0.4.1_rc4
2047 2011-11-15 23:34:58 <CIA-89> bitcoin: Luke Dashjr * rc650c820ac16 gentoo/net-p2p/wxbitcoin/ (Manifest wxbitcoin-0.4.1_rc4.ebuild): net-p2p/wxbitcoin: bump to 0.4.1_rc4
2048 2011-11-15 23:35:10 <luke-jr> hmm, I suppose I should have waited for rc5
2049 2011-11-15 23:35:37 <luke-jr> parus: even the last version needed 1.6
2050 2011-11-15 23:35:56 <parus> well, i used the binary.
2051 2011-11-15 23:36:08 <parus> [and it worked]
2052 2011-11-15 23:38:03 <parus> http://pastebin.com/Dgpa7KXu
2053 2011-11-15 23:38:19 <parus> i guess, its the dependency-problem
2054 2011-11-15 23:38:38 <luke-jr> parus: what version are you upgrading from?
2055 2011-11-15 23:39:11 <parus> debian repo provides 1.5-2
2056 2011-11-15 23:39:17 <luke-jr> bitcoind version
2057 2011-11-15 23:39:20 <parus> uh
2058 2011-11-15 23:39:25 agricocb has joined
2059 2011-11-15 23:39:47 tower has quit (Disconnected by services)
2060 2011-11-15 23:40:01 tower has joined
2061 2011-11-15 23:40:20 <parus> hrhr, .3.24 is the one from debian
2062 2011-11-15 23:40:40 <parus> i'll try .4
2063 2011-11-15 23:42:05 TD_ has joined
2064 2011-11-15 23:42:47 m00p has quit (Read error: Connection reset by peer)
2065 2011-11-15 23:43:19 <gmaxwell> parus: make -f Makefile.unix USE_UPNP=    to just disable it.
2066 2011-11-15 23:45:25 sacarlson has joined
2067 2011-11-15 23:52:41 <luke-jr> ^ best option for Debian users right now
2068 2011-11-15 23:53:07 <luke-jr> would be nice if someone restored backward compatibility with 1.5 tho
2069 2011-11-15 23:53:08 <gmaxwell> Though, there aren't 0.5 RC binaries ?
2070 2011-11-15 23:53:15 erus` has quit (Quit: ChatZilla 0.9.87 [Firefox 8.0/20111104165243])
2071 2011-11-15 23:53:25 <luke-jr> gmaxwell: they still require miniupnpc 1.6
2072 2011-11-15 23:54:49 <gmaxwell> wha? they should be static.
2073 2011-11-15 23:55:31 p0s has quit (Remote host closed the connection)
2074 2011-11-15 23:56:26 <luke-jr> gmaxwell: nothing should be static :P
2075 2011-11-15 23:56:29 Matt_von_Mises has joined
2076 2011-11-15 23:56:48 <gjs278> I agree
2077 2011-11-15 23:56:59 <gmaxwell> luke-jr: there is no point in distributing generic binaries for linux that aren't especially of a C++ app.
2078 2011-11-15 23:57:00 <Matt_von_Mises> Hello. If anyone wants to complete a bitcoin survey (optional questions and I'll post the results later) -> http://www.kwiksurveys.com?u=BitcoinSurvey Thanks.
2079 2011-11-15 23:57:39 <parus> okay, it compiled and is running perfectly without -nodnsseed
2080 2011-11-15 23:59:05 <ThomasV> http://www.theregister.co.uk/2011/11/15/bitcoin_flaw/  <--"devastating"
2081 2011-11-15 23:59:07 <ThomasV> lol
2082 2011-11-15 23:59:13 <molecular> Matt_von_Mises, where are you going to publish the results?
2083 2011-11-15 23:59:41 <Matt_von_Mises> Would you like me to add an email field to the survey?
2084 2011-11-15 23:59:56 <Matt_von_Mises> I'll put a link on the bitcoin forums but I could send emails out to people.