1 2013-11-01 00:00:01 <groglogic> berndj: agreed that kind of scenario is one risk; words on paper (laws) can make anything happen, especially if they have men with guns to back them up; thus, again, the greatest risk is a major legacy system backed by a powerful military like the US; either directly (US gov itself), or, indirectly via courts, patent systems, corporate IP, etc
   2 2013-11-01 00:00:05 Fistful_of_LTC has quit (Ping timeout: 245 seconds)
   3 2013-11-01 00:01:13 Application has joined
   4 2013-11-01 00:01:18 <groglogic> MC1984: agreed and in recent years especially there seem to be cases where a common man would say the USG is violating the Constitution; if that's allowed, anything becomes fair game; slippery slope
   5 2013-11-01 00:01:21 <MC1984> someone did try to trademark bitcoin once or something
   6 2013-11-01 00:01:57 dgolds has joined
   7 2013-11-01 00:02:10 <groglogic> #arduino Chat8923: i clicked on that link you said and it appears to have given me root on your box? what?!
   8 2013-11-01 00:02:16 <groglogic> </sorry>
   9 2013-11-01 00:02:26 <MC1984> wut
  10 2013-11-01 00:04:06 groglogic has quit (Remote host closed the connection)
  11 2013-11-01 00:04:58 Muis_ has joined
  12 2013-11-01 00:05:33 da2ce7 has joined
  13 2013-11-01 00:07:00 Bog4r7 has quit (Ping timeout: 272 seconds)
  14 2013-11-01 00:07:55 Muis has quit (Ping timeout: 240 seconds)
  15 2013-11-01 00:14:45 apurplehorse has quit ()
  16 2013-11-01 00:15:08 reneg has quit (Ping timeout: 272 seconds)
  17 2013-11-01 00:15:13 Gabralkhan has joined
  18 2013-11-01 00:18:57 Anduckkk has quit (Ping timeout: 246 seconds)
  19 2013-11-01 00:19:02 Bog4r7 has joined
  20 2013-11-01 00:19:40 Gabralkhan_ has quit (Ping timeout: 272 seconds)
  21 2013-11-01 00:28:04 macboz has joined
  22 2013-11-01 00:32:05 groglogic has joined
  23 2013-11-01 00:34:37 diki has quit ()
  24 2013-11-01 00:36:30 toffoo has joined
  25 2013-11-01 00:36:40 one_zero has joined
  26 2013-11-01 00:36:40 Gabralkhan has quit (Ping timeout: 272 seconds)
  27 2013-11-01 00:38:50 Gabralkhan has joined
  28 2013-11-01 00:42:37 JimJones has quit (Ping timeout: 246 seconds)
  29 2013-11-01 00:47:29 testicon is now known as testicon^away
  30 2013-11-01 00:48:16 owowo has quit (Ping timeout: 240 seconds)
  31 2013-11-01 00:48:56 wiretapped has quit (Ping timeout: 240 seconds)
  32 2013-11-01 00:49:06 bbrian has joined
  33 2013-11-01 00:50:27 owowo has joined
  34 2013-11-01 00:53:57 Gabralkhan has quit (Ping timeout: 246 seconds)
  35 2013-11-01 00:58:30 Gabralkhan has joined
  36 2013-11-01 01:00:51 canoon has joined
  37 2013-11-01 01:01:29 wiretapped has joined
  38 2013-11-01 01:06:17 soheil has quit (Remote host closed the connection)
  39 2013-11-01 01:07:52 toffoo has quit ()
  40 2013-11-01 01:10:05 saivann has quit (Quit: Ex-Chat)
  41 2013-11-01 01:12:41 ewoly has quit ()
  42 2013-11-01 01:13:30 Gabralkhan has quit (Ping timeout: 272 seconds)
  43 2013-11-01 01:19:34 groglogic has quit (Remote host closed the connection)
  44 2013-11-01 01:20:22 Application has quit (Ping timeout: 240 seconds)
  45 2013-11-01 01:23:47 yano has joined
  46 2013-11-01 01:24:30 Gabralkhan has joined
  47 2013-11-01 01:24:55 askmike has joined
  48 2013-11-01 01:26:48 bbrian has quit (Ping timeout: 272 seconds)
  49 2013-11-01 01:29:26 askmike has quit (Ping timeout: 260 seconds)
  50 2013-11-01 01:33:26 saivann has joined
  51 2013-11-01 01:36:16 Application has joined
  52 2013-11-01 01:36:40 Zarutian has quit (Quit: Zarutian)
  53 2013-11-01 01:37:25 soheil has joined
  54 2013-11-01 01:40:33 skinnkavaj has quit (Ping timeout: 265 seconds)
  55 2013-11-01 01:40:56 skinnkavaj has joined
  56 2013-11-01 01:40:56 skinnkavaj has quit (Changing host)
  57 2013-11-01 01:40:56 skinnkavaj has joined
  58 2013-11-01 01:42:00 soheil has quit (Ping timeout: 265 seconds)
  59 2013-11-01 01:42:54 pwnobs has quit (Remote host closed the connection)
  60 2013-11-01 01:44:30 Bog4r7 has quit (Remote host closed the connection)
  61 2013-11-01 01:45:41 saivann has quit (Quit: Ex-Chat)
  62 2013-11-01 01:50:55 catcowllama_ has quit (Read error: Connection reset by peer)
  63 2013-11-01 01:50:55 catcow has quit (Read error: Connection reset by peer)
  64 2013-11-01 01:52:03 crass has quit (Read error: Connection reset by peer)
  65 2013-11-01 01:52:15 robocoin_ has joined
  66 2013-11-01 01:52:39 crass has joined
  67 2013-11-01 01:53:03 bitspill has quit (Read error: Operation timed out)
  68 2013-11-01 01:55:08 X-Scale has quit (Ping timeout: 245 seconds)
  69 2013-11-01 01:55:15 robocoin has quit (Ping timeout: 240 seconds)
  70 2013-11-01 01:56:30 Tantadruj has quit (Quit: DoubleRecall Turns Paywalls Into Advertising Dollars - NYTimes.com http://nyti.ms/odHOgy)
  71 2013-11-01 01:59:30 justusranvier has joined
  72 2013-11-01 01:59:34 X-Scale has joined
  73 2013-11-01 02:01:29 groglogic has joined
  74 2013-11-01 02:04:10 zer0def has quit (Read error: Operation timed out)
  75 2013-11-01 02:06:48 <BlueMatt> should I pull-request a commit to make -logtimestamps appear on -printtoconsole ?
  76 2013-11-01 02:07:02 catcowllama has joined
  77 2013-11-01 02:07:02 catcow has joined
  78 2013-11-01 02:07:02 catcowllama has quit (Changing host)
  79 2013-11-01 02:07:02 catcowllama has joined
  80 2013-11-01 02:07:02 catcow has quit (Changing host)
  81 2013-11-01 02:07:02 catcow has joined
  82 2013-11-01 02:10:34 <_alp_> lol I got banned on #bitcoin for clowning mastercoins
  83 2013-11-01 02:12:21 moleccc has joined
  84 2013-11-01 02:13:34 Gabralkhan has quit (Ping timeout: 272 seconds)
  85 2013-11-01 02:13:51 maaku has quit (Remote host closed the connection)
  86 2013-11-01 02:15:15 molec has quit (Ping timeout: 240 seconds)
  87 2013-11-01 02:15:21 maaku has joined
  88 2013-11-01 02:15:41 maaku is now known as Guest3800
  89 2013-11-01 02:19:16 bitspill has joined
  90 2013-11-01 02:20:26 Guest3800 has quit (Quit: No Ping reply in 180 seconds.)
  91 2013-11-01 02:22:12 Wild_Bill_ has joined
  92 2013-11-01 02:24:20 Tantadruj has joined
  93 2013-11-01 02:24:24 Wild_Bill has quit (Ping timeout: 248 seconds)
  94 2013-11-01 02:29:43 groglogic has quit (Remote host closed the connection)
  95 2013-11-01 02:30:58 Angus has joined
  96 2013-11-01 02:33:00 X-Scale has left ()
  97 2013-11-01 02:38:13 soheil_ has joined
  98 2013-11-01 02:42:24 soheil_ has quit (Ping timeout: 240 seconds)
  99 2013-11-01 02:45:46 wacol has joined
 100 2013-11-01 02:50:35 eristisk has joined
 101 2013-11-01 02:50:48 wacol has quit (Quit: Lost terminal)
 102 2013-11-01 02:57:07 da2ce7 has quit (Quit: Textual IRC Client: www.textualapp.com)
 103 2013-11-01 02:59:25 roconnor has quit (Ping timeout: 240 seconds)
 104 2013-11-01 03:04:17 jevin has quit (Read error: Operation timed out)
 105 2013-11-01 03:05:48 crass has quit (Read error: Connection reset by peer)
 106 2013-11-01 03:07:14 eoss has joined
 107 2013-11-01 03:07:38 jevin has joined
 108 2013-11-01 03:12:04 crass has joined
 109 2013-11-01 03:13:49 Raziel has quit (Ping timeout: 246 seconds)
 110 2013-11-01 03:17:08 da2ce7 has joined
 111 2013-11-01 03:18:27 elevatioN has joined
 112 2013-11-01 03:19:36 eristisk has quit (Ping timeout: 240 seconds)
 113 2013-11-01 03:20:32 da2ce7 has quit (Client Quit)
 114 2013-11-01 03:22:15 Subo1977_ has joined
 115 2013-11-01 03:24:16 Subo1977 has quit (Ping timeout: 240 seconds)
 116 2013-11-01 03:31:57 eristisk has joined
 117 2013-11-01 03:34:10 eristisk has quit (Excess Flood)
 118 2013-11-01 03:34:38 eristisk has joined
 119 2013-11-01 03:35:43 <warren> Luke-Jr: where does gentoo store source tarballs?
 120 2013-11-01 03:36:50 Namworld has quit ()
 121 2013-11-01 03:39:03 soheil_ has joined
 122 2013-11-01 03:39:31 <lianj> gentoo? get some arch juice
 123 2013-11-01 03:40:50 saivann has joined
 124 2013-11-01 03:41:56 <warren> lianj: you know where they keep theirs?  I'm comparing tarballs for integrity
 125 2013-11-01 03:43:05 <Luke-Jr> warren: mirrors?
 126 2013-11-01 03:43:23 <Luke-Jr> warren: if they don't match the hash, they'll fetch from the original site
 127 2013-11-01 03:43:42 soheil_ has quit (Ping timeout: 251 seconds)
 128 2013-11-01 03:43:44 roconnor has joined
 129 2013-11-01 03:43:46 <lianj> ls /var/cache/pacman/pkg/ but every PKGBULD must require the sha/md5 for the compressed sources file anyway
 130 2013-11-01 03:44:30 <warren> I dont' run these distros
 131 2013-11-01 03:44:43 <warren> I'm comparing the tarball sources of different repos
 132 2013-11-01 03:45:40 <Luke-Jr> warren: Gentoo just uses upstreams' tarballs, but automates mirroring them so as to not overload weak upstream servers
 133 2013-11-01 03:46:15 <Luke-Jr> is everyone else getting spammed by Google too?
 134 2013-11-01 03:46:22 <warren> Luke-Jr: yes, I'm asking how to find gentoo's mirror
 135 2013-11-01 03:46:44 <lianj> Luke-Jr: google spam?
 136 2013-11-01 03:46:52 <Luke-Jr> warren: distfiles.gentoo.org
 137 2013-11-01 03:47:09 <Luke-Jr> lianj: at least it claims to be: This message was sent to you by FeedBurner (feedburner.google.com)
 138 2013-11-01 03:47:18 <Apocalyptic> nope Luke
 139 2013-11-01 03:47:30 <lianj> Luke-Jr: nope
 140 2013-11-01 03:47:43 <Apocalyptic> but then i don't use this FeedBurner
 141 2013-11-01 03:47:47 <Luke-Jr> neither do I
 142 2013-11-01 03:47:50 <Luke-Jr> never heard of it before today
 143 2013-11-01 03:47:55 <lianj> hah
 144 2013-11-01 03:48:06 <lianj> wat, never even heard of it?
 145 2013-11-01 03:48:10 <Luke-Jr> when it showed up at about 5 different email aliases, plus 2 of BFGMiner's mailing list addresses
 146 2013-11-01 03:48:49 <lianj> i get spammed alot but never a feedburner mail
 147 2013-11-01 03:49:13 <lianj> you must be some kind of unicorn
 148 2013-11-01 03:50:12 <BlueMatt> Luke-Jr: none of my pseudo-honeypot addresses have gotten anything
 149 2013-11-01 03:53:00 <Luke-Jr> it's promoting a domain similar to http://www.ykashvault.com
 150 2013-11-01 03:53:19 saivann has quit (Read error: No route to host)
 151 2013-11-01 03:53:51 malaimo has quit (Ping timeout: 246 seconds)
 152 2013-11-01 03:54:30 saivann has joined
 153 2013-11-01 03:55:03 arioBarzan has joined
 154 2013-11-01 03:55:24 bitspill has quit (Ping timeout: 265 seconds)
 155 2013-11-01 03:55:55 Mobius_ has joined
 156 2013-11-01 03:55:56 malaimo has joined
 157 2013-11-01 03:56:39 Mobius_ is now known as MobiusL
 158 2013-11-01 03:57:13 mapppum has joined
 159 2013-11-01 04:00:42 mappum has quit (Ping timeout: 272 seconds)
 160 2013-11-01 04:03:14 hsmiths has quit (Read error: No route to host)
 161 2013-11-01 04:04:29 hsmiths has joined
 162 2013-11-01 04:04:52 apurplehorse has joined
 163 2013-11-01 04:05:48 mapppum has quit (Read error: Connection reset by peer)
 164 2013-11-01 04:06:14 mapppum has joined
 165 2013-11-01 04:07:26 roconnor has quit (Quit: Konversation terminated!)
 166 2013-11-01 04:09:05 arioBarzan has quit (Remote host closed the connection)
 167 2013-11-01 04:11:16 sensorii has quit (Ping timeout: 240 seconds)
 168 2013-11-01 04:13:31 sensorii has joined
 169 2013-11-01 04:17:58 transisto has joined
 170 2013-11-01 04:19:47 GNULinuxGuy has quit (Ping timeout: 260 seconds)
 171 2013-11-01 04:20:09 GNULinuxGuy has joined
 172 2013-11-01 04:20:13 apurplehorse has quit ()
 173 2013-11-01 04:22:19 arioBarzan has joined
 174 2013-11-01 04:25:40 elevatioN has quit (Ping timeout: 268 seconds)
 175 2013-11-01 04:26:43 Gabralkhan has joined
 176 2013-11-01 04:27:11 rdponticelli has quit (Quit: No Ping reply in 180 seconds.)
 177 2013-11-01 04:29:48 rdponticelli has joined
 178 2013-11-01 04:29:50 rck109d has joined
 179 2013-11-01 04:33:03 Eiii has quit (Ping timeout: 246 seconds)
 180 2013-11-01 04:35:00 Eiii has joined
 181 2013-11-01 04:39:46 soheil_ has joined
 182 2013-11-01 04:44:24 soheil_ has quit (Ping timeout: 272 seconds)
 183 2013-11-01 04:45:22 TheSeven has joined
 184 2013-11-01 04:45:23 [7] has quit (Disconnected by services)
 185 2013-11-01 04:47:54 owowo has quit (Quit: °!°                    ~~ S4n1tY 1S Fut1l3 ~~                        °!°)
 186 2013-11-01 04:54:15 molec has joined
 187 2013-11-01 04:57:33 molecular has quit (Ping timeout: 246 seconds)
 188 2013-11-01 04:57:35 MobiusL has quit (Quit: Ex-Chat)
 189 2013-11-01 05:00:36 rdponticelli has quit (Ping timeout: 240 seconds)
 190 2013-11-01 05:00:59 gingpark has joined
 191 2013-11-01 05:04:23 dgolds has quit (Remote host closed the connection)
 192 2013-11-01 05:04:59 dgolds has joined
 193 2013-11-01 05:07:29 Polyatomic has joined
 194 2013-11-01 05:07:29 Polyatomic has quit (Changing host)
 195 2013-11-01 05:07:29 Polyatomic has joined
 196 2013-11-01 05:09:43 dgolds has quit (Ping timeout: 260 seconds)
 197 2013-11-01 05:23:41 arioBarzan has quit (Read error: Connection reset by peer)
 198 2013-11-01 05:24:07 arioBarzan has joined
 199 2013-11-01 05:30:50 arioBarzan has quit (Read error: Connection reset by peer)
 200 2013-11-01 05:31:04 arioBarzan has joined
 201 2013-11-01 05:33:00 arioBarzan has quit (Read error: Connection reset by peer)
 202 2013-11-01 05:33:27 linq has quit (Ping timeout: 272 seconds)
 203 2013-11-01 05:35:27 dgolds has joined
 204 2013-11-01 05:39:09 gingpark has quit (Ping timeout: 240 seconds)
 205 2013-11-01 05:39:27 eoss has quit (Remote host closed the connection)
 206 2013-11-01 05:40:18 dgolds has quit (Ping timeout: 265 seconds)
 207 2013-11-01 05:40:34 soheil_ has joined
 208 2013-11-01 05:42:05 soheil_ has quit (Read error: Connection reset by peer)
 209 2013-11-01 05:42:14 soheil_ has joined
 210 2013-11-01 05:46:52 Bkil has joined
 211 2013-11-01 05:47:03 soheil_ has quit (Ping timeout: 260 seconds)
 212 2013-11-01 05:52:12 Bkil has quit (Quit: Bye)
 213 2013-11-01 06:08:18 AusBitBank has quit (Ping timeout: 245 seconds)
 214 2013-11-01 06:14:24 mapppum has quit (Ping timeout: 252 seconds)
 215 2013-11-01 06:17:52 <tiyoT> finally my local news mention bitcoin
 216 2013-11-01 06:26:26 jeewee has joined
 217 2013-11-01 06:32:53 Coincidental has joined
 218 2013-11-01 06:36:13 dgolds has joined
 219 2013-11-01 06:37:36 sensorii has quit (Ping timeout: 240 seconds)
 220 2013-11-01 06:37:38 maaku has joined
 221 2013-11-01 06:37:53 atian_ has quit (Ping timeout: 246 seconds)
 222 2013-11-01 06:38:11 sensorii has joined
 223 2013-11-01 06:39:18 atian has joined
 224 2013-11-01 06:40:27 dgolds has quit (Ping timeout: 246 seconds)
 225 2013-11-01 06:43:03 soheil_ has joined
 226 2013-11-01 06:44:01 ralphtheninja has joined
 227 2013-11-01 06:44:24 zorro35 has joined
 228 2013-11-01 06:47:11 soheil_ has quit (Ping timeout: 240 seconds)
 229 2013-11-01 06:48:04 atian has quit (Read error: Operation timed out)
 230 2013-11-01 06:51:49 atian has joined
 231 2013-11-01 06:53:02 linq has joined
 232 2013-11-01 06:58:09 reneg has joined
 233 2013-11-01 06:59:03 paracyst has quit ()
 234 2013-11-01 07:00:25 warren has quit (Quit: ZNC - http://znc.in)
 235 2013-11-01 07:00:51 reneg1 has joined
 236 2013-11-01 07:01:35 warren has joined
 237 2013-11-01 07:01:45 atian has quit (Ping timeout: 272 seconds)
 238 2013-11-01 07:03:01 reneg has quit (Ping timeout: 272 seconds)
 239 2013-11-01 07:03:36 Ashaman has joined
 240 2013-11-01 07:03:54 atian has joined
 241 2013-11-01 07:06:55 zorro35 has left ()
 242 2013-11-01 07:09:53 MobPhone has quit (Ping timeout: 240 seconds)
 243 2013-11-01 07:10:45 <HaltingState> I am doing a high security physically secure hardware wallet; for storing large amounts of bitcoins safely, even if your computer is compromised; i have a dongle for 2 factor auth 8 bit aurdino dongle, a program, and a 32 bit arm processor server for key storage/signing and physical security of keys
 244 2013-11-01 07:10:49 warren has quit (Quit: ZNC - http://znc.in)
 245 2013-11-01 07:10:54 Ashaman has quit (Remote host closed the connection)
 246 2013-11-01 07:11:38 <HaltingState> is there support in wallets for signing keys if the private key is not local? like doing RPC to remote server to get signature?
 247 2013-11-01 07:11:42 warren has joined
 248 2013-11-01 07:11:46 <HaltingState> or no standard yet?
 249 2013-11-01 07:14:06 <HaltingState> is you have device that is storing private keys and have transaction you want to sign, then the wallet (on the potentially compromised computer) needs to query out to the server with the private keys to get signature for hash; then there needs to be dongle or 2 factor auth to prevent the compromised computer from approving transactions on its own
 250 2013-11-01 07:14:12 Thepok has joined
 251 2013-11-01 07:14:53 <HaltingState> the only solution so far seems to be a dongle and inputing a pin or shared secret with the private key storage server
 252 2013-11-01 07:14:59 sustrik has joined
 253 2013-11-01 07:17:43 dgolds has joined
 254 2013-11-01 07:21:56 Ashaman has joined
 255 2013-11-01 07:22:15 dgolds has quit (Ping timeout: 260 seconds)
 256 2013-11-01 07:28:43 <Thepok> ;;genrate 333
 257 2013-11-01 07:28:46 <gribble> The expected generation output, at 333.0 Mhps, given difficulty of 390928787.638, is 0.00042838459399 BTC per day and 1.78493580829e-05 BTC per hour.
 258 2013-11-01 07:29:19 <michagogo> cloud!uid14316@wikia/Michagogo|HaltingState: sounds like you're describing the Trezor
 259 2013-11-01 07:29:20 ralphtheninja has quit (Read error: Operation timed out)
 260 2013-11-01 07:31:15 paybitcoin has quit (Ping timeout: 240 seconds)
 261 2013-11-01 07:31:47 paybitcoin has joined
 262 2013-11-01 07:32:23 <BlueMatt> HaltingState: its not really implemented yet, though hardware wallets are being designed, so its coming soon(ish)
 263 2013-11-01 07:33:00 Tantadruj has quit (Quit: DoubleRecall Turns Paywalls Into Advertising Dollars - NYTimes.com http://nyti.ms/odHOgy)
 264 2013-11-01 07:34:33 gingpark has joined
 265 2013-11-01 07:36:08 AusBitBank has joined
 266 2013-11-01 07:37:02 <HaltingState> michagogo|cloud, no; if trezor is stolen, you lose the bitcoin
 267 2013-11-01 07:37:06 <HaltingState> or lose access to them
 268 2013-11-01 07:37:26 <HaltingState> trezor protects you from having bitcoin stolen if your machine is compromised, so its massive improvement
 269 2013-11-01 07:38:34 <michagogo> cloud!uid14316@wikia/Michagogo|HaltingState: unless I'm misremembering, trezor has a backup mechanism, and is (optionally? Idk) password-protected
 270 2013-11-01 07:38:37 <HaltingState> this is a 3 server proticol, where as trevor is 2 server
 271 2013-11-01 07:41:29 <HaltingState> I ordered hardware and 32 bit ARM board and will play with this; i was reading about BIOS rootkits and rootkits in ethernet card firmaware and realizing how insecure any computer connected to internet is; you cant trust your home computer with the private keys
 272 2013-11-01 07:43:44 <HaltingState> many of the attacks now persist between operating system installs and even physically switching the hard disc out; and infect bootable installations such as tails
 273 2013-11-01 07:43:44 soheil_ has joined
 274 2013-11-01 07:44:49 <HaltingState> so someone installs a rootkit on your firmware or bios and you reinstall and think your safe but they can steal your wallet; you run off bootable usb and think you are safe, but they can still steal your wallet; you swap out hard drives and reinstall os, you are still rooted
 275 2013-11-01 07:47:21 <HaltingState> i looked at security enhanced linux and running everything as userspace linux from readonly file and its still not safe against many of these attacks; the thing storing the private keys and doing signing has to be dedicated hardware and the key cant leave the device and you have to assume the computer running wallet is compromised
 276 2013-11-01 07:48:23 soheil_ has quit (Ping timeout: 265 seconds)
 277 2013-11-01 07:48:27 Nesetalis has quit (Ping timeout: 252 seconds)
 278 2013-11-01 07:48:50 <HaltingState> for a normal user, security enhanced linux with policy restricting while processes can access wallet  folder + password on wallet, is probably enough; but for exchanges and people with more bitcoin...
 279 2013-11-01 07:49:37 <HaltingState> security enhanced linux can log policy violations; allow processes to list the wallet folder but log whenever anything tries to read from it; so its good for a honey pot to see what is trying to steal or access wallets
 280 2013-11-01 07:54:36 zer0def has joined
 281 2013-11-01 07:57:35 MobPhone has joined
 282 2013-11-01 08:08:56 midnightmagic has quit (Ping timeout: 260 seconds)
 283 2013-11-01 08:11:01 midnightmagic has joined
 284 2013-11-01 08:11:42 OrP has joined
 285 2013-11-01 08:12:36 linq has quit (Ping timeout: 268 seconds)
 286 2013-11-01 08:13:47 Arnavion has quit (Quit: Arnavion)
 287 2013-11-01 08:14:11 fanquake has joined
 288 2013-11-01 08:14:27 Arnavion has joined
 289 2013-11-01 08:14:27 asa1024 has quit (Quit: asa1024)
 290 2013-11-01 08:14:56 xiangfu has joined
 291 2013-11-01 08:15:08 Alina-malina has joined
 292 2013-11-01 08:18:36 tmsk has joined
 293 2013-11-01 08:20:46 rphlx has joined
 294 2013-11-01 08:23:05 OrP has quit (Ping timeout: 268 seconds)
 295 2013-11-01 08:24:47 Thepok has quit (Ping timeout: 260 seconds)
 296 2013-11-01 08:25:45 jtimon has joined
 297 2013-11-01 08:32:25 OrP has joined
 298 2013-11-01 08:32:28 RoboTeddy has joined
 299 2013-11-01 08:32:35 jeewee has quit (Quit: Leaving.)
 300 2013-11-01 08:33:49 <RoboTeddy> if we wanted to change the POW function (e.g. to prevent centralization due to the nature of ASIC manufacturing), would we be able to? or would any proposed change just be out-voted by people who have already invested in specialized hardware and have an incentive to keep it valuable?
 301 2013-11-01 08:36:41 <swulf--> You can always change the POW function and try to get people to use it.  It's a game-theory/economic question on whether or not people will jump on board.
 302 2013-11-01 08:37:23 <RoboTeddy> swulf--: right -- has anyone theorized about what might happen?
 303 2013-11-01 08:38:23 ovidiusoft has joined
 304 2013-11-01 08:39:36 <swulf--> I imagine most people would take the most economically beneficial decision. That is, they'd want to avoid a fork of the blockchain.
 305 2013-11-01 08:44:33 soheil_ has joined
 306 2013-11-01 08:45:27 jeewee has joined
 307 2013-11-01 08:48:57 soheil_ has quit (Ping timeout: 252 seconds)
 308 2013-11-01 08:50:21 Eiii has quit ()
 309 2013-11-01 08:50:29 a_meteor has joined
 310 2013-11-01 08:52:47 loltu- has joined
 311 2013-11-01 08:53:26 Nopik has joined
 312 2013-11-01 08:53:36 mpt has joined
 313 2013-11-01 08:53:53 canoon has quit (Remote host closed the connection)
 314 2013-11-01 08:54:43 zeiris_ has joined
 315 2013-11-01 08:55:50 Tyklol has joined
 316 2013-11-01 08:55:51 pjorrit_ has joined
 317 2013-11-01 08:56:11 bladesilent has joined
 318 2013-11-01 08:56:40 xnyhps_ has joined
 319 2013-11-01 08:56:53 agentbob_ has joined
 320 2013-11-01 08:56:56 mikey__ has joined
 321 2013-11-01 08:56:58 dugo_ has joined
 322 2013-11-01 08:57:28 lianj_ has joined
 323 2013-11-01 08:58:55 askmike has joined
 324 2013-11-01 08:59:42 jtimon has quit (Ping timeout: 264 seconds)
 325 2013-11-01 08:59:51 edcba has quit (Read error: Connection reset by peer)
 326 2013-11-01 09:00:17 edcba has joined
 327 2013-11-01 09:01:07 bitnumus_ has joined
 328 2013-11-01 09:01:29 bitnumus_ is now known as Guest25808
 329 2013-11-01 09:02:32 johnsoft has joined
 330 2013-11-01 09:02:32 MCM-Mike has joined
 331 2013-11-01 09:02:32 stonecoldpat has joined
 332 2013-11-01 09:03:01 xnyhps_ is now known as xnyhps
 333 2013-11-01 09:03:34 t7 has joined
 334 2013-11-01 09:05:55 Coincidental has quit (Remote host closed the connection)
 335 2013-11-01 09:06:38 Coincidental has joined
 336 2013-11-01 09:07:48 nanotube has joined
 337 2013-11-01 09:07:56 OrP has quit (Ping timeout: 240 seconds)
 338 2013-11-01 09:08:30 jgarzik has joined
 339 2013-11-01 09:11:02 Coincidental has quit (Ping timeout: 265 seconds)
 340 2013-11-01 09:16:17 jtimon has joined
 341 2013-11-01 09:19:48 dgolds has joined
 342 2013-11-01 09:23:18 ThomasV has joined
 343 2013-11-01 09:23:32 askmike has quit (Remote host closed the connection)
 344 2013-11-01 09:23:57 askmike has joined
 345 2013-11-01 09:24:07 dgolds has quit (Ping timeout: 245 seconds)
 346 2013-11-01 09:27:09 askmike_ has joined
 347 2013-11-01 09:28:06 askmike has quit (Ping timeout: 246 seconds)
 348 2013-11-01 09:29:00 tiyoT has quit (Ping timeout: 250 seconds)
 349 2013-11-01 09:29:24 gingpark has quit (Ping timeout: 265 seconds)
 350 2013-11-01 09:33:16 Ashaman has quit (Ping timeout: 265 seconds)
 351 2013-11-01 09:34:00 gingpark has joined
 352 2013-11-01 09:34:57 sensorii has quit (Ping timeout: 240 seconds)
 353 2013-11-01 09:34:57 justusranvier has quit (Ping timeout: 240 seconds)
 354 2013-11-01 09:35:37 RoboTeddy has quit (Remote host closed the connection)
 355 2013-11-01 09:42:25 Anduck has joined
 356 2013-11-01 09:44:19 sensorii has joined
 357 2013-11-01 09:45:25 soheil has joined
 358 2013-11-01 09:48:17 aupiff has quit (Ping timeout: 246 seconds)
 359 2013-11-01 09:49:21 aupiff has joined
 360 2013-11-01 09:50:00 soheil has quit (Ping timeout: 252 seconds)
 361 2013-11-01 09:54:39 gingpark has quit (Remote host closed the connection)
 362 2013-11-01 09:54:56 gingpark has joined
 363 2013-11-01 09:56:29 justusranvier has joined
 364 2013-11-01 09:58:40 macboz has quit (Quit: This computer has gone to sleep)
 365 2013-11-01 10:02:28 justusranvier has quit (Remote host closed the connection)
 366 2013-11-01 10:04:57 justusranvier has joined
 367 2013-11-01 10:07:43 rdponticelli has joined
 368 2013-11-01 10:08:02 jtimon has quit (Ping timeout: 272 seconds)
 369 2013-11-01 10:08:30 FabianB_ has joined
 370 2013-11-01 10:09:24 FabianB has quit (Ping timeout: 246 seconds)
 371 2013-11-01 10:10:52 macboz has joined
 372 2013-11-01 10:13:56 rdponticelli has quit (Ping timeout: 240 seconds)
 373 2013-11-01 10:20:25 dgolds has joined
 374 2013-11-01 10:21:20 bladesilent has quit (Ping timeout: 272 seconds)
 375 2013-11-01 10:24:43 dgolds has quit (Ping timeout: 245 seconds)
 376 2013-11-01 10:27:37 i2pRelay has quit (Ping timeout: 240 seconds)
 377 2013-11-01 10:28:13 tmsk has quit (Quit: tmsk)
 378 2013-11-01 10:28:28 gavinandresen has quit (Quit: gavinandresen)
 379 2013-11-01 10:28:39 Anduck has quit (Ping timeout: 246 seconds)
 380 2013-11-01 10:29:07 random_cat_ has quit (Remote host closed the connection)
 381 2013-11-01 10:29:37 i2pRelay has joined
 382 2013-11-01 10:31:30 ovidiusoft has quit (Ping timeout: 264 seconds)
 383 2013-11-01 10:31:57 handle has joined
 384 2013-11-01 10:32:32 random_cat_ has joined
 385 2013-11-01 10:33:23 Anduck has joined
 386 2013-11-01 10:33:23 Anduck has quit (Changing host)
 387 2013-11-01 10:33:23 Anduck has joined
 388 2013-11-01 10:34:04 jeewee has quit (Quit: Leaving.)
 389 2013-11-01 10:36:06 macboz has quit (Quit: This computer has gone to sleep)
 390 2013-11-01 10:36:48 jeewee1 has joined
 391 2013-11-01 10:46:01 ovidiusoft has joined
 392 2013-11-01 10:46:04 soheil has joined
 393 2013-11-01 10:46:15 rdponticelli has joined
 394 2013-11-01 10:46:30 sustrik has quit (Ping timeout: 264 seconds)
 395 2013-11-01 10:49:59 Raziel has joined
 396 2013-11-01 10:50:21 soheil has quit (Ping timeout: 246 seconds)
 397 2013-11-01 10:53:17 rdponticelli has quit (Ping timeout: 240 seconds)
 398 2013-11-01 10:55:40 TD[away] is now known as TD
 399 2013-11-01 10:59:00 sustrik has joined
 400 2013-11-01 10:59:23 Tyklol is now known as Tykling
 401 2013-11-01 10:59:42 bladesilent has joined
 402 2013-11-01 11:07:13 Coincidental has joined
 403 2013-11-01 11:10:00 graingert has quit (Quit: Ex-Chat)
 404 2013-11-01 11:10:12 eristisk has quit (Quit: killall -9 irc)
 405 2013-11-01 11:10:15 graingert has joined
 406 2013-11-01 11:10:39 graingert is now known as Guest6290
 407 2013-11-01 11:12:17 Coincidental has quit (Ping timeout: 272 seconds)
 408 2013-11-01 11:12:18 eristisk has joined
 409 2013-11-01 11:16:37 eristisk has quit (Ping timeout: 240 seconds)
 410 2013-11-01 11:21:13 dgolds has joined
 411 2013-11-01 11:25:51 dgolds has quit (Ping timeout: 260 seconds)
 412 2013-11-01 11:25:56 one_zero has quit ()
 413 2013-11-01 11:26:01 rdymac has quit (Excess Flood)
 414 2013-11-01 11:27:12 rdymac has joined
 415 2013-11-01 11:30:57 eristisk has joined
 416 2013-11-01 11:44:03 askmike_ has quit (Remote host closed the connection)
 417 2013-11-01 11:44:17 eristisk has quit (Ping timeout: 240 seconds)
 418 2013-11-01 11:46:51 soheil has joined
 419 2013-11-01 11:51:54 soheil has quit (Ping timeout: 264 seconds)
 420 2013-11-01 11:53:56 a_meteor has quit (Ping timeout: 265 seconds)
 421 2013-11-01 12:01:37 xiangfu has quit (Remote host closed the connection)
 422 2013-11-01 12:10:38 mappum has joined
 423 2013-11-01 12:10:51 macboz has joined
 424 2013-11-01 12:17:55 ahmedbodi has quit (Ping timeout: 264 seconds)
 425 2013-11-01 12:21:56 dgolds has joined
 426 2013-11-01 12:25:34 ahmedbodi has joined
 427 2013-11-01 12:26:39 dgolds has quit (Ping timeout: 272 seconds)
 428 2013-11-01 12:28:43 theorbtwo has quit (Remote host closed the connection)
 429 2013-11-01 12:30:56 bladesilent has quit (Quit: Leaving)
 430 2013-11-01 12:35:30 CryptoBuck has quit (Ping timeout: 245 seconds)
 431 2013-11-01 12:37:08 Polyatomic has quit (Quit: Polyatomic)
 432 2013-11-01 12:38:06 elevatioN has joined
 433 2013-11-01 12:42:34 rdymac has quit (Excess Flood)
 434 2013-11-01 12:42:42 rdymac has joined
 435 2013-11-01 12:43:07 elevatioN has quit (Ping timeout: 272 seconds)
 436 2013-11-01 12:43:47 Evilmax has joined
 437 2013-11-01 12:43:47 Evilmax has quit (Changing host)
 438 2013-11-01 12:43:47 Evilmax has joined
 439 2013-11-01 12:43:58 elevatioN has joined
 440 2013-11-01 12:44:16 <Evilmax> ;;blocks
 441 2013-11-01 12:44:18 <gribble> 267285
 442 2013-11-01 12:47:41 soheil has joined
 443 2013-11-01 12:49:39 Zarutian has joined
 444 2013-11-01 12:50:49 AusBitBank has quit (Ping timeout: 245 seconds)
 445 2013-11-01 12:51:38 elevatioN has quit ()
 446 2013-11-01 12:52:12 soheil has quit (Ping timeout: 259 seconds)
 447 2013-11-01 12:56:00 Namworld has joined
 448 2013-11-01 12:58:28 roconnor has joined
 449 2013-11-01 13:08:04 Coincidental has joined
 450 2013-11-01 13:08:56 Neozonz has quit (Ping timeout: 264 seconds)
 451 2013-11-01 13:10:54 theorbtwo has joined
 452 2013-11-01 13:11:19 Neozonz has joined
 453 2013-11-01 13:13:15 Coincidental has quit (Ping timeout: 272 seconds)
 454 2013-11-01 13:22:25 <helo> i suppose POW could work with several options for algo/difficulty in a particular block
 455 2013-11-01 13:22:41 dgolds has joined
 456 2013-11-01 13:23:42 <sipa> how do you weigh different types of PoW against eachother?
 457 2013-11-01 13:23:42 <sipa> to determine the best chain
 458 2013-11-01 13:23:44 patcon has joined
 459 2013-11-01 13:24:16 roconnor has quit (Quit: Konversation terminated!)
 460 2013-11-01 13:24:51 <helo> separate difficulties for each algo?
 461 2013-11-01 13:27:22 <helo> i guess the algos might have to be chosen via height modulo num_algos
 462 2013-11-01 13:27:32 <helo> for a particular height
 463 2013-11-01 13:27:33 dgolds has quit (Ping timeout: 272 seconds)
 464 2013-11-01 13:27:48 kennyhag has joined
 465 2013-11-01 13:27:48 Thepok has joined
 466 2013-11-01 13:28:09 patcon has quit (Remote host closed the connection)
 467 2013-11-01 13:28:30 hsmiths has quit (Read error: Connection reset by peer)
 468 2013-11-01 13:28:50 kennyhag has quit (Client Quit)
 469 2013-11-01 13:30:29 rdymac has quit (Read error: Connection reset by peer)
 470 2013-11-01 13:30:32 <sipa> helo: you haven't answered my question :)
 471 2013-11-01 13:30:38 hsmiths has joined
 472 2013-11-01 13:30:50 <sipa> what if you have two competing chains, and one has a higher powA, and the other has a higher powB
 473 2013-11-01 13:30:54 <sipa> which is longest?
 474 2013-11-01 13:31:16 <helo> oh right, total sum difficulty :/
 475 2013-11-01 13:33:36 kennyhag has joined
 476 2013-11-01 13:34:42 rdymac has joined
 477 2013-11-01 13:37:27 kennyhag has quit (Client Quit)
 478 2013-11-01 13:42:22 <helo> with independent difficulties for each algo, i was thinking you would still just use total sum difficulty of all algos
 479 2013-11-01 13:42:35 jeewee1 has quit (Quit: Leaving.)
 480 2013-11-01 13:42:49 jeewee has joined
 481 2013-11-01 13:43:16 <sipa> you'd need some conversion factor between them
 482 2013-11-01 13:43:22 <helo> since each algo is forced to be used for just as many blocks as the others
 483 2013-11-01 13:43:24 <sipa> and this factor will probably need to change over time
 484 2013-11-01 13:44:30 <Belxjander> Identify Each algorithm and have ID:SUM as verification strings per-block?
 485 2013-11-01 13:44:46 <Belxjander> allowing for the algorithms to be added to but not removed from a common library?
 486 2013-11-01 13:44:59 <sipa> Belxjander: adding an algorithm is a hard fork
 487 2013-11-01 13:45:11 <Belxjander> sipa: oh... ouch
 488 2013-11-01 13:45:19 <sipa> it's not easier or harder to abstract from it, as you'd need to update every full node in the work anyway
 489 2013-11-01 13:45:34 <sipa> *world
 490 2013-11-01 13:45:35 <helo> yes, def hard
 491 2013-11-01 13:46:36 <helo> it would still boil down to "whoever makes silicon the cheapest wins", it would just require several different fab runs
 492 2013-11-01 13:47:03 <helo> s/cheapest/best/
 493 2013-11-01 13:47:19 jeewee has quit (Ping timeout: 260 seconds)
 494 2013-11-01 13:47:34 <gmaxwell> helo: no it wouldn't... amazingly you can put TWO CIRCUITS on a single chip. You might have heard of this, it's called an INTEGRATED CIRCUIT.
 495 2013-11-01 13:48:14 <gmaxwell> since modern miners seem to be thermally limited, it would even nicely make use of the extra chip area that otherwise you couldn't use because of too much power density.
 496 2013-11-01 13:48:21 soheil has joined
 497 2013-11-01 13:49:17 <sipa> loi
 498 2013-11-01 13:49:39 <helo> gmaxwell: what is this circuit dilly you speak of?
 499 2013-11-01 13:49:59 <petertodd> gmaxwell: speaking of: http://spectrum.ieee.org/semiconductors/devices/the-status-of-moores-law-its-complicated
 500 2013-11-01 13:50:07 <Belxjander> gmaxwell: I was thinking of a combination "ASIC+MCU" arrangement where the ASIC can be changed by using a standard Bus Interface to the MCU and the MCU is what the Host System talks to
 501 2013-11-01 13:51:04 <petertodd> helo: gmaxwell's saying that you can't use all the silicon on a chip due to thermal concerns, so you can assign that silicon totally different uses. (like different PoW algorithms)
 502 2013-11-01 13:51:18 macboz_ has joined
 503 2013-11-01 13:51:30 <petertodd> helo: IE you could actually engineer a cost-effective ASIC that did both sha256^2 and scrypt, but not both at the same time
 504 2013-11-01 13:52:04 <gmaxwell> and certantly you don't need more fab runs.
 505 2013-11-01 13:52:27 <helo> i don't know enough about ASIC design... i assumed it is most efficient to have a more narrow focus
 506 2013-11-01 13:52:31 <gmaxwell> petertodd: not quite accurate, rather you could take advantage of the fact that you don't need to run them both at the same time.
 507 2013-11-01 13:52:42 soheil has quit (Ping timeout: 240 seconds)
 508 2013-11-01 13:53:03 <gmaxwell> helo: only in terms of design and validation, congrats you further shifted the market towards fewer makers.
 509 2013-11-01 13:53:06 macboz has quit (Ping timeout: 264 seconds)
 510 2013-11-01 13:53:08 <helo> so the most efficient double-sha256 chip won't be also the most efficient scrypt chip
 511 2013-11-01 13:53:59 <t7> helo: it will not even be able to compute another functions
 512 2013-11-01 13:54:01 <petertodd> gmaxwell: ? that's what I said I thought
 513 2013-11-01 13:54:03 <t7> -s
 514 2013-11-01 13:54:10 <gmaxwell> helo: no, doesn't matter, it's just mm of silicon, and as I mentioned if you don't need both at the same time you can gain from power gating.
 515 2013-11-01 13:54:14 <helo> but sounds like that is actually the case so... i guess i better get back to my real job ;)
 516 2013-11-01 13:54:34 rdymac has quit (Excess Flood)
 517 2013-11-01 13:55:40 <kjj> the chip is unlikely to beat dedicated chips in either function
 518 2013-11-01 13:56:32 <petertodd> helo: it might be, at some density double-sha256^2 logic will produce more heat than can be removed, so the sha256^2 silicon has to be less dense than it could be anyway. It probably won't be absolutely the most efficient, but it can be very close, and in reality the two uses could create a cheaper overall product.
 519 2013-11-01 13:56:37 hsmiths has quit (Read error: Connection reset by peer)
 520 2013-11-01 13:56:50 <petertodd> (sell one design to two markets basically)
 521 2013-11-01 13:56:56 <gmaxwell> kjj: thats untrue, except in irrelevant analysis. you can get the same performance per mm of silicon used, it doesn't matter how you dice it. Actually you can get superlinear performance from power gating. e.g. each mm works better because of the mm of inactive chip next to it.
 522 2013-11-01 13:57:12 rdymac has joined
 523 2013-11-01 13:57:27 <kjj> gmaxwell: unless the sizes needed just magically happen to work out at the thermal optimum
 524 2013-11-01 13:57:51 <kjj> in reality, you are likely to be better off with just unused mass in those places.
 525 2013-11-01 13:58:09 <gmaxwell> kjj: in which case one chip is no better or worse than two.
 526 2013-11-01 13:58:10 <petertodd> helo: speaking of, was watching a video the other day on the cray 2; did you realize not only was it liquid cooled, but they actually had little spray jets that blasted the surface of the chips constantly?
 527 2013-11-01 13:58:35 <Luke-Jr> kjj: unused mass vs usable-but-off mass is not really different
 528 2013-11-01 13:58:40 <kjj> there may be a market for a chip that gives unusually bad performance at either of the two functions, but I doubt it would be a very large market
 529 2013-11-01 13:59:11 pjorrit_ has quit (Read error: Operation timed out)
 530 2013-11-01 13:59:58 <petertodd> kjj: I wouldn't assume that; sha256^2 especially has very small minimum sizes on the calculation units, so you have a tonne of flexibility on how you physically route and position the silicon
 531 2013-11-01 13:59:59 <kjj> Luke-Jr: I agree.  but I'm saying that the fraction of the die area devoted to each function is unlikely to result in optimal characteristics for either one.  there will be a trade off in one direction or the other, or a balance where they both suck (relatively speaking)
 532 2013-11-01 14:00:16 <gmaxwell> kjj: what are you talking about? helo repeated a perennial bad proposal (I've seen it on the forum at least three times) where every other block has to be a different pow with the mistaken belief that people would be forced into multiple fab runs to mine for it.
 533 2013-11-01 14:00:19 <Luke-Jr> kjj: it's only worse performance when you're doing both concurrently - and measured at power/Gh might actually be better
 534 2013-11-01 14:00:35 <Luke-Jr> most mining ASICs have more efficient power usage at slower speeds and lower heat
 535 2013-11-01 14:00:57 <kjj> Luke-Jr: no, it would likely be worse even when one pipeline was idle
 536 2013-11-01 14:01:00 pjorrit has joined
 537 2013-11-01 14:01:18 <gmaxwell> kjj: nonsense. area is area. Performance per unit area would be the same, likely better because of the power comments, certantly no worse... unless you're just counting development overhead, in which case you're creating a centeralizing force in increasing development costs.
 538 2013-11-01 14:01:21 ThomasV has quit (Ping timeout: 252 seconds)
 539 2013-11-01 14:01:37 elliwilli has joined
 540 2013-11-01 14:01:50 <Luke-Jr> kjj: erm, you might be right IF anyone was trying to meet optimal charictaristics..
 541 2013-11-01 14:01:58 <Belxjander> Luke-Jr: I ran into the same issue with an AMD64X2 processor... factory clocked for 2.6GHz but it operates like crap ... clock it back to 2.2GHz and it can take mid-winter through mid-summer with a constant performance capability... more invariant than adaptive scaling for conditions
 542 2013-11-01 14:01:59 <gmaxwell> kjj: we're not suggesting a single circuit that would perform both functions, you just put two specialized circuits side by side. they're independant otherwise.
 543 2013-11-01 14:02:12 <kjj> ugh
 544 2013-11-01 14:02:16 <kjj> yes, I understand that
 545 2013-11-01 14:02:31 <kjj> each function will requires some fraction of the die area
 546 2013-11-01 14:02:42 <kjj> the complete chip will be a blend of the two functions
 547 2013-11-01 14:03:05 Zarutian has quit (Quit: Zarutian)
 548 2013-11-01 14:03:08 * helo stabs heat in the eyes
 549 2013-11-01 14:03:21 <Luke-Jr> (all this being said, it'd be pretty neat if someone took advantage of the unused space on a mining ASIC for EC algos)
 550 2013-11-01 14:03:24 <kjj> each function needs some unused die space to meet thermal requirements
 551 2013-11-01 14:03:38 patcon has joined
 552 2013-11-01 14:03:47 <Luke-Jr> kjj: it doesn't matter if it's unused or simply off
 553 2013-11-01 14:03:49 <kjj> it is unlikely that the fraction of space devoted to the two functions will come anywhere near the optimal fraction for that function
 554 2013-11-01 14:04:19 <gmaxwell> kjj: sure, you can consider that part of the size of the function's implementation, except to the extent that you can _share_ that space when the functions are not running concurrently.
 555 2013-11-01 14:04:21 elliwilli has quit (Client Quit)
 556 2013-11-01 14:04:23 <Luke-Jr> it's unlikely that either function will be optimal at all
 557 2013-11-01 14:04:24 <kjj> this conversation would go a lot easier if you people would read what I'm actually saying instead of answering the arguments that you pretend that I'm making even though I'm not
 558 2013-11-01 14:04:39 hsmiths has joined
 559 2013-11-01 14:04:53 <Luke-Jr> bitfury might be going for optimal design, but I don't think anyone else is
 560 2013-11-01 14:05:01 <helo> Luke-Jr: isn't multi-function ASIC only beneficial (to fabricators) if the different functions happen one-at-a-time?
 561 2013-11-01 14:05:42 <helo> i.e. if your chip is hashing full speed all the time, it would just add additional heat to run EC algos from time to time
 562 2013-11-01 14:05:44 <kjj> at a given density, say that SHA meets thermal limits at 70% of die used
 563 2013-11-01 14:05:48 <Luke-Jr> helo: I'm thinking along the lines of a hardware wallet that can also mine when idle
 564 2013-11-01 14:06:12 <helo> kjj: i think the problem is that in the long run, heat dissipation becomes the biggest constraint
 565 2013-11-01 14:06:16 <kjj> that leaves 30% for the other function, which is likely to mean that the other function is not operating anywhere near as fast as it could
 566 2013-11-01 14:06:22 <gmaxwell> kjj: ...
 567 2013-11-01 14:06:25 <Luke-Jr> helo: eg, the primary function might be the other algorithm
 568 2013-11-01 14:06:42 <kjj> and yes, I know there can be some overlap.  the other function might get 40%
 569 2013-11-01 14:08:10 <gmaxwell> kjj: take two logical dies. one of type a, optimal for its design. one of type b optimal for its design. stick one next to the other.  Tada. You will get no worse than either alone, now realize that since they don't run concurrently you can share some of their dissipation area. You can now do more of a+b per area than you could do a+b per area seperated. This can only improve things.
 570 2013-11-01 14:08:20 <helo> Luke-Jr: ahh, yeah i wouldn't mind if my CPU mined while it was idle :)
 571 2013-11-01 14:08:21 <kjj> by blending the two functions, you are unlikely to end up with anywhere near an optimal ratio.  your chip will not perform as well as the competition's chip.
 572 2013-11-01 14:08:43 patcon has quit (Ping timeout: 272 seconds)
 573 2013-11-01 14:09:03 <Belxjander> kjj: the use of "blending" is equivalent to merging the two functions into sharing partial functionality which may be a poor choice of words
 574 2013-11-01 14:09:14 <gmaxwell> kjj: it will perform superior at performing _mining_ in this fool proposal (sorry helo), which was the goal.
 575 2013-11-01 14:09:59 <kjj> gmaxwell: it will not be superior to two actual chips that are actually optimized.
 576 2013-11-01 14:10:01 <helo> i've learned enough from my various fool proposals that i think it outweighs the ego hit ;)
 577 2013-11-01 14:10:07 <gmaxwell> kjj: if for some reason sharing space doesn't help, you can always not do that and still produce a single part with a single mask set.
 578 2013-11-01 14:10:11 dvide has joined
 579 2013-11-01 14:10:44 <gmaxwell> kjj: Sorry, you're incorrect, unless you're adopting an irrelevant defintion of superior which is unrelated to the application.
 580 2013-11-01 14:10:50 <kjj> because the optimal dissipation area for a is not a+b, nor is that optimal for b
 581 2013-11-01 14:11:48 <sipa> since you're not doing a and b at the same time, you can put more a+b on the same dissipation area than if it only needs to do a or b
 582 2013-11-01 14:11:58 <gmaxwell> kjj: ...  you aren't forced to share dissapation area. If a need space x including its extra space for dissipation and b needs space y, then at most you need space x+y, probably less, but no worse than space x+y.
 583 2013-11-01 14:13:04 <sipa> and indeed, if the ratio you're aiming at is 3/7 on a 1 cm^2 chip, take 0.3 cm^2 of optimal-a die and 0.7 cm^2 of optimal-b die, and put them next to eachother
 584 2013-11-01 14:13:04 <sipa> at worst
 585 2013-11-01 14:13:09 <sipa> probably you can put it all on 0.9 cm^2 :)
 586 2013-11-01 14:13:15 <kjj> if you have area for (a+b), you should be using a die of size a', not a
 587 2013-11-01 14:13:21 agricocb has joined
 588 2013-11-01 14:13:34 <Belxjander> gmaxwell: if both circuits are on-chip and set for exclusion locking where one function is active at any given time with a cooling off between changeover... then the requirement would be the difference of X and Y with the larger value preferred,   is how I would read that... is that wrong?
 589 2013-11-01 14:14:19 <gmaxwell> so even if you have no savings from dissapation sharing (which is crazy, but whatever), you can still happily share masks and still do no worse than having them seperate for the join task of running both functions.
 590 2013-11-01 14:14:38 <kjj> sipa: my point is that whatever split you choose, your chip will underperform in at least one of the two functions, compared to a single purpose chip
 591 2013-11-01 14:14:43 <gmaxwell> Belxjander: whatever sharing you could get would be highly task and geometry specific.
 592 2013-11-01 14:14:59 <gmaxwell> kjj: it won't under perform for mining, which requires both of the functions.
 593 2013-11-01 14:15:01 <sipa> kjj: yes but the point is that you don't need both functions at the same time
 594 2013-11-01 14:15:07 <kjj> gahhh
 595 2013-11-01 14:15:12 <kjj> I fucking know that you don't need both at the same time
 596 2013-11-01 14:15:39 <kjj> I'm saying that even when idle, the area devoted to the other function is not likely to be the area that would be simply unused in a single purpose chip
 597 2013-11-01 14:16:01 <helo> kjj: for the parts that would underperform if interleaved, you just don't interleave
 598 2013-11-01 14:16:05 <gmaxwell> even regardless of that point, you can do no worse.  If you need 1000 sha256 units and 1000 sha3 units,  you will not be worse off with 1000 chips that do both sha2 and sha3.
 599 2013-11-01 14:16:05 <Belxjander> kjj: you keep repeating "blending"... which indicates a mixing of some kind for what I am reading... is this your intended meaning or not?
 600 2013-11-01 14:16:07 * sipa shall withdraw from this discussion
 601 2013-11-01 14:16:29 <gmaxwell> (and likely better, but no worse)
 602 2013-11-01 14:16:41 <helo> kjj: but there are going to be some that can, and that is an efficiency gain
 603 2013-11-01 14:17:03 <kjj> Belxjander: blending could be either putting them on the same die, or putting two dies in the same package.  it is an abstract term here that means sharing
 604 2013-11-01 14:17:35 <gmaxwell> kjj: "the area devoted to the other function is not likely to be the area that would be simply unused in a single purpose chip" we agree, and yet this is irrelevant.
 605 2013-11-01 14:17:49 <kjj> gmaxwell: it is relevant
 606 2013-11-01 14:17:55 <gmaxwell> kjj: say there is no area needed for dissapation. OKAY. Then you don't get any dissapation sharing gains.
 607 2013-11-01 14:18:07 <gmaxwell> You're still no worse of combining them.
 608 2013-11-01 14:18:12 <gmaxwell> s/of/off/
 609 2013-11-01 14:18:22 <kjj> but you do need thermal management
 610 2013-11-01 14:19:15 hsmiths has quit (Quit: Nettalk6 - www.ntalk.de)
 611 2013-11-01 14:20:35 <helo> gmaxwell: is there a kind of computation that would benefit from 3d lithography particularly more than sha256?
 612 2013-11-01 14:21:22 hsmiths has joined
 613 2013-11-01 14:21:59 mintyFresh has joined
 614 2013-11-01 14:23:46 dgolds has joined
 615 2013-11-01 14:23:54 Anduck has quit (Ping timeout: 252 seconds)
 616 2013-11-01 14:24:59 <helo> people whine about POW being wasteful, but if the experiment continues succeed, at some point it may provide incentives for better fab techniques (that could in turn benefit other industries)
 617 2013-11-01 14:26:26 <jgarzik> I don't like this new gavin fee code
 618 2013-11-01 14:26:40 macboz_ has quit (Ping timeout: 265 seconds)
 619 2013-11-01 14:26:41 <jgarzik> My fees are now non-zero, for the first time in my bitcoin history
 620 2013-11-01 14:26:48 <jgarzik> ;p
 621 2013-11-01 14:28:01 ahmedbodi is now known as ahmedbodz-shops
 622 2013-11-01 14:28:16 <sipa> meh, i still think we first need a wallet that can deal with non-confirming transactions and cancelling and respending with different fees
 623 2013-11-01 14:28:23 <sipa> before we touch the fee code to somethingt hat is inevitably more fragile
 624 2013-11-01 14:28:48 Anduck has joined
 625 2013-11-01 14:28:53 dgolds has quit (Ping timeout: 272 seconds)
 626 2013-11-01 14:29:26 <kjj> wow.  the oldest transaction in my main wallet is only 114721 confirmations.  I could have sworn that was 200000+ not long ago
 627 2013-11-01 14:29:29 patcon has joined
 628 2013-11-01 14:30:26 <Luke-Jr> kjj: My ASIC's so fast it finishes an infinite loop in 5 minutes.
 629 2013-11-01 14:30:45 <sipa> Luke-Jr: it runs linux?
 630 2013-11-01 14:30:47 <kjj> a gift from Chuck Norris?
 631 2013-11-01 14:31:49 <jgarzik> heh
 632 2013-11-01 14:32:04 <Luke-Jr> sipa: only if you strip out the gotos
 633 2013-11-01 14:32:22 <kjj> what would be the implications of a signed cancellation request message type?
 634 2013-11-01 14:32:36 <Luke-Jr> kjj: pointless
 635 2013-11-01 14:33:10 <Luke-Jr> respending just needs some sensible relay logic
 636 2013-11-01 14:33:23 <sipa> very hard to guarantee it reaches all nodes, and miners
 637 2013-11-01 14:33:26 <kjj> well, semi-pointless.  you couldn't count on it, but it would be useful if you are getting ready to replace something
 638 2013-11-01 14:33:41 <sipa> just send the replacement instead :)
 639 2013-11-01 14:34:06 <kjj> the default node behavior in the face of replacements would need to be improved
 640 2013-11-01 14:34:24 <kjj> or was it?  I haven't been keeping up on things for a few months
 641 2013-11-01 14:34:44 <sipa> replacement is disable whatsoever
 642 2013-11-01 14:35:15 <Luke-Jr> ideally it should probably just allow any replacemnet as long as the outputs are >= the same value
 643 2013-11-01 14:35:21 <kjj> yeah, so you have to wait for the old transaction to expire out of mempools before you can replace now
 644 2013-11-01 14:35:44 <gmaxwell> there is no expire in the running network. nodes just get restarted from time to time.
 645 2013-11-01 14:35:58 <Luke-Jr> gmaxwell: the mempool can overflow too
 646 2013-11-01 14:36:16 <helo> so is isStandard() going to be modified so to permit relaying of double spends if the longest chain has grown more than N blocks since the first was seen?
 647 2013-11-01 14:36:26 <Luke-Jr> also, if the fee is extremely-too-low, it won't have gotten relayed much in the first place
 648 2013-11-01 14:36:43 <Luke-Jr> helo: IsStandard has nothing to do with that
 649 2013-11-01 14:37:27 <helo> or strictly relay double spends that have strictly higher fee and no reduced outputs
 650 2013-11-01 14:38:41 <helo> Luke-Jr: sorry, i lazily think of isStandard() as "relayed transactions"
 651 2013-11-01 14:39:00 bbrian has joined
 652 2013-11-01 14:39:31 <petertodd> Luke-Jr: I've got code to do that BTW, just need to clean up some stuff in it and give it some more testing
 653 2013-11-01 14:40:05 <petertodd> Luke-Jr: I also never found any problems with my replace-by-fee patch, ran it for months on a few nodes
 654 2013-11-01 14:40:41 <Luke-Jr> petertodd: I had to revert it because I kept getting assert failures
 655 2013-11-01 14:40:47 <petertodd> helo: IsStandard() has nothing to do with double-spends
 656 2013-11-01 14:41:08 <petertodd> Luke-Jr: you're thinking the mempool stuff, I'm talking about the dead simple replace-by-fee code I wrote first
 657 2013-11-01 14:41:13 <Luke-Jr> oh
 658 2013-11-01 14:41:23 <helo> right... isstandard is just one requirement for relaying, not /the/ requirement
 659 2013-11-01 14:41:26 <Luke-Jr> but that didn't handle dependencies, right?
 660 2013-11-01 14:41:35 <petertodd> Luke-Jr: correct, which is why it was so simple!
 661 2013-11-01 14:41:43 <Luke-Jr> :p
 662 2013-11-01 14:41:59 <petertodd> Luke-Jr: but we don't have any way to relay dependencies either (IE scorched earth case)
 663 2013-11-01 14:42:30 <petertodd> Luke-Jr: would you be willing to run zeroconf-safe replace-for-fee?
 664 2013-11-01 14:42:44 <Luke-Jr> petertodd: not with dependency resolution
 665 2013-11-01 14:42:46 <Luke-Jr> without*
 666 2013-11-01 14:43:04 <petertodd> Luke-Jr: no, I'm talking about the one that only allows replacement if existing outputs are left untouched (and unspent)
 667 2013-11-01 14:43:19 <petertodd> *and are unspent
 668 2013-11-01 14:43:52 <Luke-Jr> that sounds fine for relaying
 669 2013-11-01 14:44:06 <Luke-Jr> I suppose mining might not need any additional changes right now
 670 2013-11-01 14:44:14 <Luke-Jr> (besides that of relaying)
 671 2013-11-01 14:44:35 <petertodd> Luke-Jr: yeah, I don't think it should affect your mining at all; your fee dependency code is completely in CreateNewBlock() right?
 672 2013-11-01 14:44:42 <petertodd> (it's stateless too right?)
 673 2013-11-01 14:45:32 <Luke-Jr> right
 674 2013-11-01 14:46:15 <petertodd> Luke-Jr: perfect, I'll see if I can get time to get that done this weekend. (jdillon's 4BTC reward for that code is a fair bit of cash frankly)
 675 2013-11-01 14:48:05 <Luke-Jr> I guess I need to find time to debug why your dust-b-gone stuff isn't going through too
 676 2013-11-01 14:49:06 <petertodd> Luke-Jr: yeah, it gets mined with git master FWIW - only a matter of time before a non-eligius miner finally mines a block with a op-return tx.
 677 2013-11-01 14:49:09 soheil has joined
 678 2013-11-01 14:50:10 damethos has joined
 679 2013-11-01 14:50:39 <kjj> Luke-Jr: the problem with output-based replacement is that one of the outputs is change, and it needs to come down, or the replacement needs another input.
 680 2013-11-01 14:51:05 <Luke-Jr> petertodd: you're assuming miners are going to adopt the OP_RETURN whitelisting
 681 2013-11-01 14:51:38 <Luke-Jr> kjj: I think it's a given the replacement needs another input, yes
 682 2013-11-01 14:51:49 <Luke-Jr> kjj: there's also the dependency-based fee for the other case
 683 2013-11-01 14:51:52 <petertodd> kjj: yup, which is why I did a replace-by-fee patch first. But that's controversial as all hell, and zeroconf safe replacement will get a lot of other infrastructure issues fixed so I'm happy to work on it first.
 684 2013-11-01 14:52:27 <kjj> petertodd: we talked about simple fee replacement at the conference in May
 685 2013-11-01 14:52:29 <petertodd> Luke-Jr: it's in git master, so it's a matter of time before some % does. (heck, I've adopted that myself)
 686 2013-11-01 14:53:08 <Luke-Jr> kjj: were you there?
 687 2013-11-01 14:53:15 <petertodd> kjj: oh in person?
 688 2013-11-01 14:53:19 <Luke-Jr> did I meet you? XD
 689 2013-11-01 14:53:19 <kjj> I think we both agreed that since some fraction of the network will be antisocial, we should make the default behavior antisocial so that people understand when things are unsafe
 690 2013-11-01 14:53:32 <petertodd> kjj: agreed
 691 2013-11-01 14:53:37 soheil has quit (Ping timeout: 246 seconds)
 692 2013-11-01 14:53:37 * Luke-Jr remembers having that discussion with petertodd and gmaxwell privately.. O.o
 693 2013-11-01 14:53:52 <petertodd> kjj: And since then I've found so many useful things to do with replace-by-fee that I no longer consider it even antisocial!
 694 2013-11-01 14:54:01 <kjj> haha, yes.  we walked all over San Jose on the way back from the restaurant
 695 2013-11-01 14:54:17 <petertodd> kjj: and jdillon's scorched earth thing is brilliant
 696 2013-11-01 14:54:36 <Luke-Jr> kjj: IRC nicknames don't translate well to the real world for me
 697 2013-11-01 14:54:42 <petertodd> kjj: sheesh, I wasn't even drinking much... :P
 698 2013-11-01 14:54:47 <petertodd> yeah, me neither
 699 2013-11-01 14:54:57 <petertodd> "Who the hell is Luke-Jr?"
 700 2013-11-01 14:55:12 <kjj> when I signed up, I didn't know that the name on the registration form was exactly what would be printed on the name tag
 701 2013-11-01 14:55:22 <Luke-Jr> petertodd: really? I'd think I'm one of the more easily recognisable
 702 2013-11-01 14:55:27 <kjj> so the name tag had my full name instead of my initials, and no one knew who I was
 703 2013-11-01 14:55:30 <Luke-Jr> considering my name tag said Luke-Jr and all
 704 2013-11-01 14:55:46 <Luke-Jr> kjj: scribble on it :p
 705 2013-11-01 14:55:48 <petertodd> Luke-Jr: lol
 706 2013-11-01 14:56:28 * petertodd keeps meaning to change his bitcointalk username
 707 2013-11-01 14:56:29 <Luke-Jr> although Voorhees didn't realise who I was
 708 2013-11-01 14:56:31 <Luke-Jr> lol
 709 2013-11-01 14:56:49 <kjj> as usual when I travel, half of the people I met were from Minnesota, most within 50 miles of me
 710 2013-11-01 14:57:46 <kjj> if it helps, I was the bald one
 711 2013-11-01 14:57:55 <petertodd> Luke-Jr: ha. I had kinda the opposite experience at a conference in Toronto the other week though: "Oh, your retep on bitcointalk! Oh wow I read your stuff all the time!"
 712 2013-11-01 14:58:19 <petertodd> kjj: bald with dark clothing?
 713 2013-11-01 14:58:24 <kjj> yeah
 714 2013-11-01 14:58:30 <kjj> we talked a lot on the way to and from the indian place
 715 2013-11-01 14:58:55 <sipa> kjj: you were at the indian place? :o
 716 2013-11-01 14:59:03 <Luke-Jr> lol
 717 2013-11-01 14:59:12 <sipa> where did you sit?
 718 2013-11-01 14:59:16 <kjj> also, at that bar we were talking to that guy about homomorphic encryption
 719 2013-11-01 14:59:39 <kjj> I was on the side by the wall, I think two or three seats from the end away from the door (near Gavin)
 720 2013-11-01 15:00:15 ahmedbodz-shops is now known as ahmedbodi
 721 2013-11-01 15:00:30 <kjj> or I think I might have moved down once things were thinning out.  I may have started closer to the end
 722 2013-11-01 15:01:04 <petertodd> sheesh, I wasn't even drinking that night :P
 723 2013-11-01 15:01:15 <Luke-Jr> it was nearly 6 months ago
 724 2013-11-01 15:01:27 <sipa> i was sitting opposite gavin, iirc
 725 2013-11-01 15:01:28 <Luke-Jr> dunno about you guys, but my memory of details like this isn't that good
 726 2013-11-01 15:01:41 <sipa> i have no clue who kjj may have been, though
 727 2013-11-01 15:01:42 <Luke-Jr> I think I was sitting oppposite maaku
 728 2013-11-01 15:01:49 <sipa> maaku was there too? :o
 729 2013-11-01 15:01:52 <Luke-Jr> ..yes
 730 2013-11-01 15:01:55 <sipa> crap
 731 2013-11-01 15:01:58 <petertodd> ha, I totally didn't realize that
 732 2013-11-01 15:02:04 <Luke-Jr> >_<
 733 2013-11-01 15:02:16 <Luke-Jr> he was all formal looking IIRC
 734 2013-11-01 15:02:23 <kjj> we need special tags for our name badges next year
 735 2013-11-01 15:02:23 <Luke-Jr> suit & tie
 736 2013-11-01 15:02:26 <sipa> kjj, maaku: any pictures of you on the interwebz? :p
 737 2013-11-01 15:03:07 <kjj> sipa: sec
 738 2013-11-01 15:03:15 <sipa> yeah, i did meet some people where i only after a while realized that i knew them undera nickname
 739 2013-11-01 15:03:45 MobPhone has quit (Read error: Connection reset by peer)
 740 2013-11-01 15:04:07 MobPhone has joined
 741 2013-11-01 15:05:17 <Luke-Jr> maaku = https://1.gravatar.com/avatar/6c8c7467704c98e3db450f2cdccd4bc5?d=https%3A%2F%2Fa248.e.akamai.net%2Fassets.github.com%2Fimages%2Fgravatars%2Fgravatar-user-420.png&r=x&s=140
 742 2013-11-01 15:05:51 <sipa> hmm, doesn't ring a bell
 743 2013-11-01 15:06:20 <Luke-Jr> his nametag would have said Mark, and maybe mentioned Freicoin
 744 2013-11-01 15:06:56 OrP has joined
 745 2013-11-01 15:07:23 rdymac has quit (Read error: Connection reset by peer)
 746 2013-11-01 15:07:55 <kjj> petertodd: it was Adam Back that we were talking to in the bar
 747 2013-11-01 15:08:33 <petertodd> kjj: ah, yeah, now I remember
 748 2013-11-01 15:08:52 <petertodd> kjj: you were sitting south of me
 749 2013-11-01 15:08:53 Coincidental has joined
 750 2013-11-01 15:11:12 rdymac has joined
 751 2013-11-01 15:11:17 <kjj> heh.  are the lifetime members on the foundation page listed in any particular order?
 752 2013-11-01 15:11:52 Eiii has joined
 753 2013-11-01 15:12:20 <gmaxwell> they're probably listed by member ID order.
 754 2013-11-01 15:13:09 Coincidental has quit (Read error: Operation timed out)
 755 2013-11-01 15:13:37 <petertodd> Probably true of annual members too.
 756 2013-11-01 15:13:57 <petertodd> I'm #17
 757 2013-11-01 15:14:28 <kjj> I'm 4th on the lifetime list.  had no idea I was so fast signing up
 758 2013-11-01 15:15:07 <gmaxwell> it's across then down.
 759 2013-11-01 15:15:22 <kjj> aww.  that's not as good
 760 2013-11-01 15:15:25 <gmaxwell> e.g. gavin peter mike pieter me jeff Wladimir
 761 2013-11-01 15:16:20 <Luke-Jr> problem with lifetime membership is that you can't really leave if it goes sour
 762 2013-11-01 15:16:49 <Luke-Jr> eg, if Trace had won the election
 763 2013-11-01 15:17:04 <petertodd> Luke-Jr: well, you can leave, but you can't make a financial difference
 764 2013-11-01 15:18:11 <petertodd> ha, funny to see Mastercoin on the silver list
 765 2013-11-01 15:18:47 testicon^away is now known as testicon^work
 766 2013-11-01 15:24:03 _ingsoc has joined
 767 2013-11-01 15:24:20 dgolds has joined
 768 2013-11-01 15:25:37 Anduck has quit (Ping timeout: 272 seconds)
 769 2013-11-01 15:25:50 Andrevan has quit (Quit: WeeChat 0.4.2-rc1)
 770 2013-11-01 15:26:03 JimJones has joined
 771 2013-11-01 15:26:35 charlie2 has joined
 772 2013-11-01 15:27:42 swulf-- has quit (Ping timeout: 252 seconds)
 773 2013-11-01 15:29:06 dgolds has quit (Ping timeout: 264 seconds)
 774 2013-11-01 15:29:20 <kjj> the share difficulty on p2pool is now higher than the bitcoin difficulty was when I started mining
 775 2013-11-01 15:30:16 bibbybob has joined
 776 2013-11-01 15:30:31 <gmaxwell> kjj: share diff is different per user.
 777 2013-11-01 15:30:52 <gmaxwell> (or at least for fast users)
 778 2013-11-01 15:31:33 <JimJones> gmaxwell,
 779 2013-11-01 15:31:34 <kjj> I'm a slow miner, less than one share per block
 780 2013-11-01 15:31:38 <JimJones> https://blockchain.info/tx/c07d47a5e1d023d0fd2c47c600a4ca178c7a2954754c88f3fb7f729bafb94539
 781 2013-11-01 15:31:45 <JimJones> still unconfirmed...
 782 2013-11-01 15:32:47 Guest25808 has left ()
 783 2013-11-01 15:33:16 <gmaxwell> JimJones: what software did you create that txn with?
 784 2013-11-01 15:33:25 <JimJones> blockchain.info
 785 2013-11-01 15:33:34 <gmaxwell> I really wish you could get a raw txn out of bc.i.
 786 2013-11-01 15:33:47 <gmaxwell> Anyone have that transaction?
 787 2013-11-01 15:33:49 <JimJones> asked to adjust fees, and i adjusted according to blockchain.info
 788 2013-11-01 15:34:05 <gmaxwell> ^ what does that mean?
 789 2013-11-01 15:34:57 <JimJones> when seending btc's via blockchain.info it suggests fees.. and i accepted the suggested fees according to blockchain.info
 790 2013-11-01 15:35:00 bitspill has joined
 791 2013-11-01 15:35:10 <JimJones> a raw txn?
 792 2013-11-01 15:35:32 <JimJones> sending*
 793 2013-11-01 15:35:36 <gmaxwell> JimJones: the actual transaction data. My local nodes don't have it.
 794 2013-11-01 15:35:41 <kjj> not showing up in my nodes
 795 2013-11-01 15:35:47 <JimJones> o.O
 796 2013-11-01 15:36:02 <JimJones> https://bitcointalk.org/index.php?topic=322194
 797 2013-11-01 15:36:03 <gmaxwell> don't even seem to have ever been offered it.
 798 2013-11-01 15:36:29 <kjj> from my logs, "not enough fees"
 799 2013-11-01 15:36:30 <JimJones> opened a thread here piuk said he re-boardcasted the transaction
 800 2013-11-01 15:36:46 mpt has quit (Disconnected by services)
 801 2013-11-01 15:36:55 mpr has joined
 802 2013-11-01 15:37:03 elevatioN has joined
 803 2013-11-01 15:37:10 <gmaxwell> okay, one of my nodes got offered it at least.
 804 2013-11-01 15:37:19 mpr is now known as Guest99141
 805 2013-11-01 15:37:25 <gmaxwell> but that one has the minfee cranked up, so it didn't take it.
 806 2013-11-01 15:37:57 <JimJones> https://blockchain.info/pushtx
 807 2013-11-01 15:38:31 <petertodd> Looks like that tx has slightly too low fees by the default rules.
 808 2013-11-01 15:38:43 <JimJones> hm
 809 2013-11-01 15:39:05 damethos has quit (Remote host closed the connection)
 810 2013-11-01 15:39:05 <kjj> ERROR: CTxMemPool::accept() : not enough fees c07d47a5e1d023d0fd2c47c600a4ca178c7a2954754c88f3fb7f729bafb94539, 340000 < 350000
 811 2013-11-01 15:39:06 <petertodd> Fees should be 0.0034235, but it's just 0.0034
 812 2013-11-01 15:39:10 <petertodd> yup
 813 2013-11-01 15:39:16 * gmaxwell facepalms
 814 2013-11-01 15:39:26 <JimJones> lol
 815 2013-11-01 15:39:42 <petertodd> gmaxwell: this is why I write all my code using 0.00011 BTC/KB...
 816 2013-11-01 15:39:45 <gmaxwell> JimJones: welp I'm sorry for telling you it was hunky dory a day ago.
 817 2013-11-01 15:39:54 <JimJones> np =P
 818 2013-11-01 15:40:31 <JimJones> what happens now?
 819 2013-11-01 15:40:42 <JimJones> this kinda never happened to me
 820 2013-11-01 15:40:44 <gmaxwell> JimJones: I ass-u-med that with it being "34k"  and a fee of 0.0034 on bc.i that the number was rounded up and that the fee was right, since it was so close to right. :)
 821 2013-11-01 15:40:52 <kjj> the good news is that since most nodes won't relay, resending is easy-ish
 822 2013-11-01 15:41:03 <gmaxwell> I think eligius would take that, except we can't get the raw txn.
 823 2013-11-01 15:41:09 <Apocalyptic> JimJones, I also don't have that tx
 824 2013-11-01 15:41:12 <gmaxwell> kjj: I don't think replacing is easy on bc.i at all.
 825 2013-11-01 15:41:24 <kjj> don't they expire things at 30 hours?
 826 2013-11-01 15:41:24 Zarutian has joined
 827 2013-11-01 15:41:49 <gmaxwell> If someone actually had the tx we could hand it directly to eligius and it would get mined, I'm pretty sure.
 828 2013-11-01 15:42:10 * petertodd thinks we really need a completely independent tx broadcast mechanism that validates nothing and rate-limits based on something else.
 829 2013-11-01 15:42:13 <gmaxwell> someone should ask piuk to please round _up_ the fees instead of truncating.
 830 2013-11-01 15:42:14 <kjj> you know what would come in handy right now?  a utility that connected to a specific node, pretended to be a full node, and fetched the one raw transaction you want
 831 2013-11-01 15:42:50 <gmaxwell> kjj: you're assuming its in anyone's mempools at all. e.g. it was never offered to my laptop.
 832 2013-11-01 15:42:52 <petertodd> kjj: ooh, and make it have a mode where it could connect to the whole network to find the tx
 833 2013-11-01 15:43:06 <gmaxwell> I suppose this one probably is.
 834 2013-11-01 15:43:07 <kjj> gmaxwell: blockchain.info /should/ have it in theirs
 835 2013-11-01 15:43:20 ThomasV has joined
 836 2013-11-01 15:43:49 <gmaxwell> would be good to run some collectors that save every txn they see.
 837 2013-11-01 15:44:26 <gmaxwell> JimJones: in any case, two possibilities, one you doublespend that transaction with a higher fee, or two we get the actual transaction data and manually submit it to eligius.
 838 2013-11-01 15:45:02 Nesetalis has joined
 839 2013-11-01 15:45:04 bbrian has quit (Ping timeout: 246 seconds)
 840 2013-11-01 15:45:13 <TD> i thought b.i timed out transactions after 24 hours
 841 2013-11-01 15:45:15 <JimJones> yea,  can anyone contact piuk about the transaction data?
 842 2013-11-01 15:45:25 <JimJones> still hasn't been 24h
 843 2013-11-01 15:45:33 <Apocalyptic> TD, where did you get that from ?
 844 2013-11-01 15:45:41 <TD> er
 845 2013-11-01 15:45:45 <TD> i sort of heard it on the grapevine
 846 2013-11-01 15:45:48 * TD doesn't know why he thinks that
 847 2013-11-01 15:46:39 <petertodd> Do bc.i wallets even save the tx's in them? Like, if I somehow create a tx in a bc.i wallet that won't get mined, does the tx go away?
 848 2013-11-01 15:48:28 daybyter has joined
 849 2013-11-01 15:49:30 <JimJones> how does one doublespend ?
 850 2013-11-01 15:49:50 soheil has joined
 851 2013-11-01 15:50:00 <kjj> just make a new transaction that involves at least one of the other inputs and get it into a block
 852 2013-11-01 15:50:02 <JimJones> maybe it would be easier.. but more ppl could be prevented if i contacted piuk about this
 853 2013-11-01 15:50:09 <petertodd> https://blockchain.info/create-double-spend
 854 2013-11-01 15:50:25 linq has joined
 855 2013-11-01 15:50:25 linq has quit (Changing host)
 856 2013-11-01 15:50:25 linq has joined
 857 2013-11-01 15:50:42 Anduck has joined
 858 2013-11-01 15:50:54 <petertodd> That's actually intended to create simultaneous double-spends, but it'll work in your case too.
 859 2013-11-01 15:51:06 <JimJones> before I do this, wanna know if there is any way to fuck this up and loose the 25btcs? lol
 860 2013-11-01 15:51:12 <gmaxwell> JimJones: nah.
 861 2013-11-01 15:51:25 <gmaxwell> Except via the boring ways you'd expect e.g. sending it to the wrong place.
 862 2013-11-01 15:51:29 Centaure has joined
 863 2013-11-01 15:51:54 <JimJones> ye gonna do this now
 864 2013-11-01 15:52:26 <petertodd> JimJones: do the double-spend with the private key from 1FdcBichMwFx6C3zjWqgC96j6FTeFocj9L, that'll make the whole transaction invalid without touching the more valuable inputs
 865 2013-11-01 15:53:08 <petertodd> You *may* find that the double-spend tool won't work due to bc.i not relaying the double-spend BTW.
 866 2013-11-01 15:55:04 bitspill has quit (Remote host closed the connection)
 867 2013-11-01 15:55:07 soheil has quit (Ping timeout: 272 seconds)
 868 2013-11-01 15:57:12 <JimJones> bc.i crashes when clicking on the "paper wallet" tab oh dear
 869 2013-11-01 15:57:48 <petertodd> :/
 870 2013-11-01 15:58:13 <petertodd> JimJones: I've noticed it crash when clicking on the "Bitcoin-QT" version of that; maybe the bug is even worse?
 871 2013-11-01 15:59:59 <JimJones> is this a problem with bc.i or with me(local client)
 872 2013-11-01 16:00:39 <petertodd> I've seen it too
 873 2013-11-01 16:02:37 _ingsoc has quit (Ping timeout: 272 seconds)
 874 2013-11-01 16:03:17 Subo1977_ has quit (Ping timeout: 240 seconds)
 875 2013-11-01 16:03:25 Subo1977 has joined
 876 2013-11-01 16:04:00 _ingsoc has joined
 877 2013-11-01 16:04:06 <JimJones> in long term, will the transaction revert or will it stay pending if I don't do anything?
 878 2013-11-01 16:06:44 bbrian has joined
 879 2013-11-01 16:08:37 <jouke> bc.info should "forget" about the transaction in 24hours, but that mechanism has been broken before and I find bc.info quite slow today, so I guess they are having troubles of some sort.
 880 2013-11-01 16:08:52 arioBarzan has joined
 881 2013-11-01 16:10:11 dgolds has joined
 882 2013-11-01 16:10:18 <maaku> sipa: I remember seeing you but I'm not sure we talked at the conference
 883 2013-11-01 16:10:35 OrP has quit (Ping timeout: 260 seconds)
 884 2013-11-01 16:11:11 TD is now known as TD[away]
 885 2013-11-01 16:11:23 <maaku> sipa: middle guy here : http://i1.ytimg.com/vi/4Iir3_dYmK8/hqdefault.jpg
 886 2013-11-01 16:11:44 <JimJones> jouke, what u mean? the "forget" transaction thing or the sized blocks mechanism?
 887 2013-11-01 16:12:17 neep3r has joined
 888 2013-11-01 16:12:53 rdymac has quit (Excess Flood)
 889 2013-11-01 16:13:12 rdymac has joined
 890 2013-11-01 16:14:28 <jouke> bc.info wil "revert" the transaction.
 891 2013-11-01 16:15:13 roconnor has joined
 892 2013-11-01 16:15:19 <JimJones> hopefully :p
 893 2013-11-01 16:15:23 <arioBarzan> should coinbase tx's get 120 confirmations before its coins become spendable?
 894 2013-11-01 16:15:26 Centaure has left ()
 895 2013-11-01 16:15:52 <maaku> arioBarzan: 100 is all that's technically required
 896 2013-11-01 16:16:01 <maaku> to show up on the chain
 897 2013-11-01 16:16:09 <maaku> 120 for relay, iirc
 898 2013-11-01 16:16:24 <arioBarzan> https://coinbase.com/network/transactions/a84d76342539ab3249a212a96c9e1411fbb1bc79a03b3e93dc1a16cf42e8b1e3
 899 2013-11-01 16:16:32 <arioBarzan> it has just 93 conf's
 900 2013-11-01 16:17:00 <arioBarzan> some of coins are already spent http://blockchain.info/tx/a84d76342539ab3249a212a96c9e1411fbb1bc79a03b3e93dc1a16cf42e8b1e3
 901 2013-11-01 16:17:42 <maaku> ... in unconfirmed transactions
 902 2013-11-01 16:18:48 <arioBarzan> maaku: oh, I see. So the would remain unconfirmed until 7 more blocks, right?
 903 2013-11-01 16:18:54 <maaku> yes
 904 2013-11-01 16:18:55 <arioBarzan> *they
 905 2013-11-01 16:20:59 bbrian has quit (Ping timeout: 245 seconds)
 906 2013-11-01 16:22:37 elevatioN has quit (Ping timeout: 272 seconds)
 907 2013-11-01 16:24:34 <gmaxwell> a bit annoying that bc.i is displaying invalid transactions.
 908 2013-11-01 16:24:59 OrP has joined
 909 2013-11-01 16:25:05 <gmaxwell> as that greatly complicated figuring out mtgox was doing something wrong.
 910 2013-11-01 16:25:23 <gmaxwell> arioBarzan: are you asking about coinbase, as in freshly created coins, or coinbase the company?
 911 2013-11-01 16:25:47 <arioBarzan> gmaxwell: freshly created coins
 912 2013-11-01 16:25:55 neep3r has quit ()
 913 2013-11-01 16:25:58 <gmaxwell> Okay, had to ask because of the coinbase link. :P
 914 2013-11-01 16:27:07 <JimJones> gmaxwell, jouke, petertodd, kjj, i shall wait more than 24h and come back for feedback
 915 2013-11-01 16:28:49 eoss has joined
 916 2013-11-01 16:30:53 ielo has joined
 917 2013-11-01 16:34:29 twmz has quit (Read error: Connection reset by peer)
 918 2013-11-01 16:40:57 Alina-malina has quit (Ping timeout: 248 seconds)
 919 2013-11-01 16:41:13 Alina-malina has joined
 920 2013-11-01 16:45:12 patcon has quit (Remote host closed the connection)
 921 2013-11-01 16:45:39 patcon has joined
 922 2013-11-01 16:46:26 gamzabaw has joined
 923 2013-11-01 16:47:41 patcon_ has joined
 924 2013-11-01 16:50:25 bbrian has joined
 925 2013-11-01 16:50:41 soheil has joined
 926 2013-11-01 16:50:51 patcon has quit (Ping timeout: 272 seconds)
 927 2013-11-01 16:50:53 markus__ has joined
 928 2013-11-01 16:51:36 <maaku> gmaxwell: well as a service, it's nice that they show *all* transactions, for debugging purposes, but they should display prominent warnings for transactions which cannot be included in a block...
 929 2013-11-01 16:54:21 <ThomasV> hi maaku ; any progress with your utxo tree?
 930 2013-11-01 16:54:34 <arioBarzan> I have seen Gavin's http://github.com/gavinandresen/bitcoin-git/tree/smartfee . Do Bitcoin core-devs are going to recommend tx fees in near future? Is there a range for tx fees for which users could expect their tx get confirmed in following couple of months?
 931 2013-11-01 16:54:50 soheil has quit (Ping timeout: 240 seconds)
 932 2013-11-01 16:55:20 <maaku> ThomasV: yes, working on it right now actually
 933 2013-11-01 16:55:29 <maaku> adding unit tests, writing a bip
 934 2013-11-01 16:55:31 meLon has quit (Ping timeout: 245 seconds)
 935 2013-11-01 16:55:50 <ThomasV> where is your repo?
 936 2013-11-01 16:55:57 <maaku> it's not pushed yet
 937 2013-11-01 16:59:56 ralphtheninja has joined
 938 2013-11-01 17:02:06 t7 has quit (Quit: weekend)
 939 2013-11-01 17:04:39 Tantadruj has joined
 940 2013-11-01 17:05:19 patcon has joined
 941 2013-11-01 17:08:44 patcon_ has quit (Ping timeout: 245 seconds)
 942 2013-11-01 17:09:37 Coincidental has joined
 943 2013-11-01 17:10:10 patcon has quit (Ping timeout: 240 seconds)
 944 2013-11-01 17:14:33 Coincidental has quit (Ping timeout: 272 seconds)
 945 2013-11-01 17:17:07 patcon has joined
 946 2013-11-01 17:17:27 ovidiusoft has quit (Ping timeout: 272 seconds)
 947 2013-11-01 17:19:33 elevatioN has joined
 948 2013-11-01 17:19:55 OrP has quit (Ping timeout: 246 seconds)
 949 2013-11-01 17:21:21 melvster has quit (Quit: Leaving)
 950 2013-11-01 17:22:27 jeewee has joined
 951 2013-11-01 17:25:19 Anduck has quit (Ping timeout: 272 seconds)
 952 2013-11-01 17:26:01 arioBarzan has quit (Quit: Leaving...)
 953 2013-11-01 17:29:59 Beef has quit (Read error: Connection reset by peer)
 954 2013-11-01 17:34:54 melvster has joined
 955 2013-11-01 17:35:48 devrandom has quit (Ping timeout: 245 seconds)
 956 2013-11-01 17:36:50 devrandom has joined
 957 2013-11-01 17:38:49 Relicka has joined
 958 2013-11-01 17:38:50 Anduck has joined
 959 2013-11-01 17:41:06 charlie2 has left ()
 960 2013-11-01 17:41:19 swulf-- has joined
 961 2013-11-01 17:41:24 <swulf--> Does the genesis block have nHeight == 0 or 1?
 962 2013-11-01 17:41:47 <sipa> 0
 963 2013-11-01 17:41:56 PiZZaMaN2K has quit (away!~PiZZaMaN2@unaffiliated/pizzaman2k|Remote host closed the connection)
 964 2013-11-01 17:42:07 PiZZaMaN2K has joined
 965 2013-11-01 17:43:39 <swulf--> cool, thanks:)
 966 2013-11-01 17:44:55 PiZZaMaN2 has joined
 967 2013-11-01 17:45:02 PiZZaMaN2 has quit (Client Quit)
 968 2013-11-01 17:45:12 Diablo-D3 has quit (Quit: This computer has gone to sleep)
 969 2013-11-01 17:45:19 _ingsoc has quit (Ping timeout: 265 seconds)
 970 2013-11-01 17:47:01 _ingsoc has joined
 971 2013-11-01 17:48:18 PiZZaMaN2K has quit (Read error: Operation timed out)
 972 2013-11-01 17:48:49 t7 has joined
 973 2013-11-01 17:49:49 soheil_ has joined
 974 2013-11-01 17:50:00 a_meteor has joined
 975 2013-11-01 17:54:18 a_meteor has quit (Ping timeout: 240 seconds)
 976 2013-11-01 17:56:35 jeewee has quit (Quit: Leaving.)
 977 2013-11-01 17:56:49 jeewee has joined
 978 2013-11-01 18:00:42 elevatioN has quit ()
 979 2013-11-01 18:01:11 jeewee has quit (Ping timeout: 260 seconds)
 980 2013-11-01 18:01:38 debiantoruser has quit (Ping timeout: 245 seconds)
 981 2013-11-01 18:04:08 Beef has joined
 982 2013-11-01 18:04:10 debiantoruser has joined
 983 2013-11-01 18:12:51 <petertodd> arioBarzan: tx fees are based on supply and demand, so no, there's no way of knowing for sure what fees will be in the future.
 984 2013-11-01 18:13:07 melvster has quit (Remote host closed the connection)
 985 2013-11-01 18:15:18 <petertodd> maaku: how are you going to prevent your utxo tree from giving incentives to use the utxo set as a database?
 986 2013-11-01 18:15:45 <maaku> petertodd: not following
 987 2013-11-01 18:15:59 <maaku> (using the utxo as a database is rather the point)
 988 2013-11-01 18:16:12 <petertodd> maaku: e.g. if I can query it securely, then you've just make it easy to make a better namecoin on top of bitcoin, bloating the utxo set, than namecoin itself
 989 2013-11-01 18:16:39 <petertodd> maaku: and you get something which is actually secure, rather than namecoin's vulnerability to 51% attack
 990 2013-11-01 18:17:10 <maaku> you can query securly based on indexed attributes only
 991 2013-11-01 18:17:21 <maaku> for for the validation index, it's txid:n
 992 2013-11-01 18:17:42 <petertodd> maaku: ok, so the utxo tree isn't going to be indexed by scriptPubKey anymore?
 993 2013-11-01 18:17:45 <maaku> for the wallet index, it's scriptPubKey(:txid:n)
 994 2013-11-01 18:17:56 <maaku> two separate indices
 995 2013-11-01 18:18:07 <maaku> we'll see if they both get accepted
 996 2013-11-01 18:18:18 <petertodd> what do you mean by wallet index exactly? that's committed by miners still though right?
 997 2013-11-01 18:19:01 <maaku> if the distributed-update version of the tree is used, it makes more sense to index by hash(scriptPubKey)
 998 2013-11-01 18:19:20 <maaku> in which case I'm not sure how useful that would be for namecoin-like applicaitons
 999 2013-11-01 18:19:44 <maaku> petertodd: yes, it's the index that's useful for lightweight wallet applications
1000 2013-11-01 18:19:54 <petertodd> hash(scriptPubKey) helps a little bit, but there's lots of database applications where that's still useful, e.g. just pay to scriptPubKey H(name)
1001 2013-11-01 18:19:57 <maaku> vs the txid index which is useful for validating nodes
1002 2013-11-01 18:20:20 <petertodd> yeah, I strongly think only a txid index should be implemented
1003 2013-11-01 18:21:55 <petertodd> it's interesting how, for example, mastercoin has defined that all mastercoin transactions pay to the exodus address - eevn now this makes it easy to write a mastercoin client that uses SPV bloom filters to get the data it needs. (albeit not totally securely)
1004 2013-11-01 18:22:12 rdymac has quit (Read error: Connection reset by peer)
1005 2013-11-01 18:22:53 agricocb has quit (Remote host closed the connection)
1006 2013-11-01 18:23:05 <maaku> i think it's a strange inversion of priorities to kill lightweight, secure wallets because of fear over blockchain bloating parasitic apps
1007 2013-11-01 18:23:16 agricocb has joined
1008 2013-11-01 18:23:25 <maaku> assuming the performance issues of running two indices can be solved anyway
1009 2013-11-01 18:23:45 <maaku> but long term, Merkle mountain ranges is the solution
1010 2013-11-01 18:24:41 <petertodd> I think it's a strange inversion of priorities to make it convenient to write wallets that don't contribute back to the network in any way at all, when we can at least get them to get enough data to do some validation. (and fraud proofing) at linear cost
1011 2013-11-01 18:24:56 <petertodd> I worry there won't get a second chance at getting this right.
1012 2013-11-01 18:26:22 <maaku> petertodd: what i'm saying is that the correct tool for combatting parasitic apps is fees, for example
1013 2013-11-01 18:27:04 eristisk has joined
1014 2013-11-01 18:27:16 <petertodd> I mean, the amount of bandwidth required without scriptPubKey-indexed utxo trees is similar to what bloom filters require
1015 2013-11-01 18:27:30 <maaku> besides, if the wallet index is not adopted, it will be merged mined and people will use it anyway
1016 2013-11-01 18:28:07 <maaku> boom filters and scriptPubKey-indexed auth trees don't have the same security properties
1017 2013-11-01 18:28:08 <petertodd> maaku: right, but the problem is long-term UTXO growth, which fees don't do a great job on
1018 2013-11-01 18:28:44 <petertodd> for instance, if someone implemented dns names on bitcoin, the value for a given name is way above the reasonable lower-bound value for a given wallet transaction.
1019 2013-11-01 18:30:34 <petertodd> maaku: no, I'm saying in any scheme where you get a per-block list of transaction outputs sorted by scriptPubKey, a completely secure wallet would query each block, and ask for proof that the scriptPubKey prefixes they requested either did or did not exist in that block; it's about as much bandwidth as bloom filters use today. (a bit more, but not by a large amount)
1020 2013-11-01 18:32:21 <maaku> yes, i'm not arguing about bandwidth, i'm saying that boom filters requires some trust in the peers you are connecting to, vs auth trees which do not
1021 2013-11-01 18:33:38 <petertodd> oh sure, but I'm saying you can get just as much security for wallets as full UTXO auth trees, with only per-block indexes, and that lets you either expire UTXO's, or more likely do txo commitments
1022 2013-11-01 18:33:59 reneg1 has quit (Ping timeout: 272 seconds)
1023 2013-11-01 18:34:28 <petertodd> those options make UTXO bloating applications unattractive again and/or reduce their harm
1024 2013-11-01 18:36:03 <maaku> petertodd: how are hash(scriptPubKey) indices useful for parasitic apps, as they wouldn't allow transfer of ownership?
1025 2013-11-01 18:36:12 Coincidental has joined
1026 2013-11-01 18:37:37 rdymac has joined
1027 2013-11-01 18:38:42 <petertodd> maaku: you use the hash(scriptPubKey) as a means to ensure you've seen all ownership related transactions, not as the means to define ownership directly
1028 2013-11-01 18:39:38 tmsk has joined
1029 2013-11-01 18:40:01 <petertodd> maaku: e.g. you get a set of transactions related to the asset, and use the hash(scriptPubKey) table to be sure your original set of transactions was complete and you aren't missing any
1030 2013-11-01 18:40:34 <petertodd> maaku: this works even if the UTXO database provides no information what-so-ever beyond the scriptPubKey has and value out
1031 2013-11-01 18:41:04 <petertodd> s/has/hash/
1032 2013-11-01 18:42:28 chris_l has joined
1033 2013-11-01 18:43:22 patcon has quit (Read error: Connection reset by peer)
1034 2013-11-01 18:43:35 patcon has joined
1035 2013-11-01 18:44:05 <maaku> ok i'm not sure i follow that explanation, but i can think of at least one other tricky way to do it so that's enough
1036 2013-11-01 18:44:35 <maaku> back to the point, I think it's critical that transactions pay for the blockchain storage that they actually use
1037 2013-11-01 18:44:37 <petertodd> heh
1038 2013-11-01 18:45:10 <maaku> if you have that, and if parasitic apps still dominate in fees ... well maybe you just discovered something more valuable then bitcoins as payment?
1039 2013-11-01 18:45:36 <petertodd> yes, but there's no way to change that re: the UTXO set sadly, other than get rid of the UTXO set. sure, txo commitments does that mostly, but it's something that should be implemented next, rather than as yet another follow up measure
1040 2013-11-01 18:46:41 gavinandresen has joined
1041 2013-11-01 18:46:45 <petertodd> maaku: well... that worries me a fair bit in many ways. I mean, timestamping is a perfect example where we might have a unlimited supply of transactions that are work a cent or two, which is higher than what many people assume we can keep transaction fees too. (e.g. without a blocksize limit)
1042 2013-11-01 18:46:47 <maaku> well, not necessarily. you could charge exorbitant fees for adding to the UTXO, then "credit" a portion of those for cleaning it up. i've suggested that in this channel before
1043 2013-11-01 18:47:30 <petertodd> maaku: right, but the problem is that can't be enforced other than a rather invasive soft-fork that changes the economics
1044 2013-11-01 18:48:02 <petertodd> (e.g. you'd force miners to include a anyone-can-spend output paying some portion of fees to the miner 100 blocks from now)
1045 2013-11-01 18:51:14 GingerGeek[Away] is now known as GingerGeek
1046 2013-11-01 18:52:29 _ingsoc has quit (Ping timeout: 245 seconds)
1047 2013-11-01 18:53:41 <maaku> well i would rather have utxo indices + fee-related soft fork than the current status quo
1048 2013-11-01 18:54:03 patcon has quit (Remote host closed the connection)
1049 2013-11-01 18:54:34 patcon has joined
1050 2013-11-01 18:54:43 <maaku> fees aren't something i've had the time to properly look into though, so i'd voice these concerns to gavinandresen & others who are working on that
1051 2013-11-01 18:57:09 <gavinandresen> voice what concerns?  bitcoinstats logging seems broken....
1052 2013-11-01 18:57:36 <Eliel> how about extending the p2p protocol so it's possible to transfer the input transactions along with the new transaction. Then you can add an option to just simply drop txouts older than X blocks from the general utxo set. Would open a business model for long term archival nodes that'd give out older data for a fee.
1053 2013-11-01 18:58:04 <petertodd> I think you're going to find the political issues of trying to make utxo's costly with a soft-fork to be pretty much impossible frankly. I'd give it 50:50 that (U)TXO indexes of any kind will actually happen. :(
1054 2013-11-01 18:58:41 <petertodd> Eliel: yup, I proposed that the other day with something called TXO commitments
1055 2013-11-01 18:58:44 patcon has quit (Ping timeout: 245 seconds)
1056 2013-11-01 18:58:46 <gavinandresen> peter is worrying about what might possibly happen in 5 or 10 years again?  yeesh….
1057 2013-11-01 18:59:27 <gavinandresen> bitcoin would never have happened if Satoshi waited for it to be perfect before releasing it
1058 2013-11-01 18:59:29 <petertodd> gavinandresen: I'm an engineer, not a programmer at a web 2.0 startup.
1059 2013-11-01 18:59:31 _ingsoc has joined
1060 2013-11-01 18:59:44 <ThomasV> lol
1061 2013-11-01 18:59:47 <petertodd> yes, worse is better
1062 2013-11-01 18:59:50 <gavinandresen> petertodd: okey dokey.  So do some more engineering and less theorizing.
1063 2013-11-01 18:59:58 <Belxjander> maybe dumb... but would it be viable to have a "blockchain storage" where blocks of the blockchain for N blocks were saved as a file and then the next N blocks were saved into the next using numeric iterated filenames ?
1064 2013-11-01 19:00:14 <gavinandresen> (and stop being so pig-headed in pull requests, please, you're annoying the hell out of me)
1065 2013-11-01 19:00:39 <petertodd> gavinandresen: standard practice at my day job is 6 months of theory and 1 month of implementation, and the implementations work perfectly alarmingly often.
1066 2013-11-01 19:00:45 <Belxjander> and then dealing with a "virtual blockchain filesystem" treating the blockchain as a continuous filesystem of numeric consecutively named files?
1067 2013-11-01 19:01:09 <petertodd> gavinandresen: meh, write more robust code without obvious security flaws
1068 2013-11-01 19:01:58 <gavinandresen> petertodd: a patch to rate-limit LogPrint would be nifty, if you are worried about fill-up-debug.log
1069 2013-11-01 19:02:14 <petertodd> gavinandresen: I'm not, I'm worried about attackers creating fake log entries.
1070 2013-11-01 19:02:22 <gavinandresen> I am not worried about that, because I don't run -debug=net
1071 2013-11-01 19:02:35 cads has joined
1072 2013-11-01 19:02:37 <ThomasV> Belxjander: are you proposing that as a solution  to utxo bloating?
1073 2013-11-01 19:03:02 <petertodd> gavinandresen: you don't, lots of people do. just escape that text string.
1074 2013-11-01 19:03:21 <gavinandresen> petertodd: "patches welcome"
1075 2013-11-01 19:03:25 RoboTeddy has joined
1076 2013-11-01 19:03:41 <petertodd> gavinandresen: as I say, it's ridiculous to add obvious security flaws that take a line of code or two to fix
1077 2013-11-01 19:03:45 <gavinandresen> This is the kind of pig-headedness that will drive me crazy.
1078 2013-11-01 19:03:53 <Belxjander> ThomasV: I'm considering writing an AmigaOS client and other than the blockchain and wallet data structures don't really have much comprehension of what things are or how they properly interconnect... and also finding a serious lack of familiarity with C++ makes it hard to read for me
1079 2013-11-01 19:03:57 <gavinandresen> Instead of berating me, WRITE SOME FRICKING CODE
1080 2013-11-01 19:03:59 <petertodd> gavinandresen: lol, good.
1081 2013-11-01 19:04:47 <kinlo> 01|20:01:29 < gavinandresen> Instead of berating me, WRITE SOME FRICKING CODE
1082 2013-11-01 19:04:50 rdymac has quit (Excess Flood)
1083 2013-11-01 19:04:50 <kinlo> eek
1084 2013-11-01 19:04:55 <kinlo> sorry, mouse, ignore me
1085 2013-11-01 19:04:56 <Belxjander> ThomasV: so I would like to get a basic comprehension of what I should be looking for and then work on code for that one step at a time
1086 2013-11-01 19:05:00 <petertodd> gavinandresen: I have just as much respect for people who audit, review and test as I do the people who write the code in the first place.
1087 2013-11-01 19:05:12 rdymac has joined
1088 2013-11-01 19:05:17 <ThomasV> not sure if this has already been discussed, but I wonder if it will make sense, for verifying nodes, to use a data structure that makes verifying recent utxos cheaper than older ones. this would create a 'natural' incentive to move old utxos before they  require more fees
1089 2013-11-01 19:05:43 <petertodd> ThomasV: yup, and I think I have a solution in the form of something I call txo commitments
1090 2013-11-01 19:05:54 <ThomasV> which is?
1091 2013-11-01 19:06:02 groglogic has joined
1092 2013-11-01 19:06:44 <gavinandresen> ThomasV: doesn't make sense right now, the UTXO set is small.  It may never make sense, we don't know how fast the UTXO set will grow or how fast memory density will grow.
1093 2013-11-01 19:06:49 <midnightmagic> just hug it out guys, it works better
1094 2013-11-01 19:07:10 <gavinandresen> ThomasV: it is very possible that even at global scale the UTXO set will fit in your iPhone 11's memory.
1095 2013-11-01 19:07:15 MagicFab_ has joined
1096 2013-11-01 19:07:48 <ThomasV> gavinandresen: we were talking about parasite applications that bloat it
1097 2013-11-01 19:08:07 <petertodd> ThomasV: https://bitcointalk.org/index.php?topic=314467.msg3371194#msg3371194 <- gmaxwell's written it up here for me; I'm still working on a paper on it
1098 2013-11-01 19:08:17 <gavinandresen> ThomasV: which parasite applications?  We just pulled OP_RETURN so they are less parasitic....
1099 2013-11-01 19:08:45 <ThomasV> OP_RETURN might be a double edged sword..
1100 2013-11-01 19:08:52 <petertodd> gavinandresen: yes, which is why I was talking to maaku about how UTXO commitments give incentives again because it lets you easily look them up securely
1101 2013-11-01 19:09:04 <petertodd> ThomasV: well, OP_RETURN is strictly better than the alternative :/
1102 2013-11-01 19:09:12 <gavinandresen> OP_RETURN never makes it into the UTXO set, which is the subject at hand, right?
1103 2013-11-01 19:09:22 <ThomasV> sure
1104 2013-11-01 19:09:24 gamzabaw has quit (Quit: gamzabaw)
1105 2013-11-01 19:09:27 <petertodd> ThomasV: (modulo the fact that we allow zero-value OP_RETURN txos)
1106 2013-11-01 19:09:54 <gmaxwell> petertodd: its a very interesting idea, but I'm still not sure that its something bitcoin could do in the future. It's easy to describe it as too big of a change. E.g. if you speak about what it makes it possible vs whats necessary.
1107 2013-11-01 19:12:30 <petertodd> gmaxwell: sure, the long-term version of it where you're providing proofs. OTOH in the short term when you don't take advantage of it I'd argue it's just as good as other commitments, and works very similarly.
1108 2013-11-01 19:13:02 <gmaxwell> petertodd: yea, I think thats probably true. But it wouldn't stop people from spazzing out about it enabling things they don't think they'd like.
1109 2013-11-01 19:13:25 <petertodd> gmaxwell: what do you mean by enabling?
1110 2013-11-01 19:13:37 MagicFab_ has quit (Ping timeout: 265 seconds)
1111 2013-11-01 19:16:01 <petertodd> gmaxwell: after all, you only need to take advantage of the full capability of TXO commitments if UTXO growth is an issue, if it isn't, just use it like the txid:vout commitment that we want for other reasons.
1112 2013-11-01 19:16:02 <gmaxwell> petertodd: e.g. one of the things that stuff permits is a universe where wallets have to track their own transactions. I don't believe that universe would ever exist, — you'd just get tracking nodes that do it for you. But someone will, no doubt, fixate on that idea.
1113 2013-11-01 19:16:50 <gmaxwell> e.g. it enables more optimization than people would prudently use, and since we don't have good models today to compensate people for running more expensive things some may worry that enabling that will result in that.
1114 2013-11-01 19:17:26 <petertodd> gmaxwell: yeah, well I was talking to maaku earlier about how provided that we have a per-block TXO index, sorted by scriptPubKey, syncing your wallet needs almost the same amount of bandwidth as bloom filters do today. (and with partial prefix queries the same privacy)
1115 2013-11-01 19:17:38 Application has quit (Remote host closed the connection)
1116 2013-11-01 19:18:06 <gmaxwell> petertodd: assuming that no one DOSes you by choosing to flood addresses which are proximal to yours.
1117 2013-11-01 19:19:25 <petertodd> gmaxwell: sure, but that's true of full UTXO commitments too, and in that case, harder to solve!
1118 2013-11-01 19:19:42 djserge has joined
1119 2013-11-01 19:19:46 justusranvier has quit (Remote host closed the connection)
1120 2013-11-01 19:20:05 patcon_ has joined
1121 2013-11-01 19:23:06 <swulf--> I'm curious...what was the motivation to require the coinbase txin scriptSig to begin with the block height?
1122 2013-11-01 19:23:37 <phantomcircuit> swulf--, it guarantees the coinbase is a unique transaction
1123 2013-11-01 19:24:13 dgolds has quit (Remote host closed the connection)
1124 2013-11-01 19:24:42 <swulf--> Why not just require the coinbase be a unique transaction?
1125 2013-11-01 19:24:49 dgolds has joined
1126 2013-11-01 19:25:15 djserge has left ()
1127 2013-11-01 19:25:17 <petertodd> swulf--: easier to force it to be unique
1128 2013-11-01 19:25:38 <skinnkavaj> 12 hours for a transaction with fees. wtf? https://bitcointalk.org/index.php?topic=322194.0
1129 2013-11-01 19:25:40 <ThomasV> is there a chart of the utxo set size over time?
1130 2013-11-01 19:26:45 <swulf--> petertodd: is it?  There are only 265k+ coinbase txns.  Those could be indexed quite efficiently
1131 2013-11-01 19:26:53 <Eliel> skinnkavaj: if I had to guess, I'd think it's got something to do with the multiple outputs to the same address.
1132 2013-11-01 19:27:17 <swulf--> I suppose you'd have to index every tx?
1133 2013-11-01 19:27:29 <phantomcircuit> skinnkavaj, it's obvious spam
1134 2013-11-01 19:27:30 <petertodd> swulf--: yes, but, forcing them to be unque only took like 5 lines of code.
1135 2013-11-01 19:27:39 <swulf--> petertodd: laziness wins:)
1136 2013-11-01 19:28:05 <phantomcircuit> swulf--, simplicity wins
1137 2013-11-01 19:28:17 <petertodd> swulf--: yup! it's also handy, because it means you can figure out the height of a block based on only the block itself; you previously couldn't do that
1138 2013-11-01 19:28:59 <swulf--> phantom: Er, I don't agree...  "Hacks" are often simpler than a real solution to the problem... and often problematic in the future
1139 2013-11-01 19:29:02 dgolds_ has joined
1140 2013-11-01 19:29:04 dgolds has quit (Read error: Connection reset by peer)
1141 2013-11-01 19:29:24 <phantomcircuit> swulf--, it's not a hack, it's a well thought out long term solution
1142 2013-11-01 19:29:43 <petertodd> swulf--: like, literally 2 years or so worth of thought
1143 2013-11-01 19:29:46 <phantomcircuit> specifically indexing every transaction is not going to be an effective solution long term
1144 2013-11-01 19:29:59 <petertodd> phantomcircuit: every coinbase tx
1145 2013-11-01 19:30:06 MagicFab_ has joined
1146 2013-11-01 19:30:20 <swulf--> phantom: I'm not sure I agree that indexing every transaction isn't feasible long-term.
1147 2013-11-01 19:30:41 <swulf--> Is there discussion somewhere about that very topic?
1148 2013-11-01 19:32:12 ThomasV has quit (Ping timeout: 260 seconds)
1149 2013-11-01 19:32:21 <Eliel> skinnkavaj: read the last messages in the thread, they explain why it didn't go through. Apparently the fee was a slightly under the threshold per kb to be acceptable.
1150 2013-11-01 19:32:55 ThomasV has joined
1151 2013-11-01 19:33:05 <BlueMatt> swulf--: there is a clear advantage to not indexing (you dont have to do the extra work/code and you get the height just from the block without additional info) whereas there is no clear advantage to not introducing the requirement?
1152 2013-11-01 19:33:12 <BlueMatt> swulf--: I see no argument here?
1153 2013-11-01 19:33:45 <skinnkavaj> Eliel: Was it because the same adress were being used several times?
1154 2013-11-01 19:33:54 <swulf--> BlueMatt: there are indeed advantages to indexing the transactions, however
1155 2013-11-01 19:33:54 <BlueMatt> swulf--: also, see BIP30
1156 2013-11-01 19:34:00 <BlueMatt> like?
1157 2013-11-01 19:34:14 <Eliel> skinnkavaj: no, don't think so.
1158 2013-11-01 19:34:31 <Eliel> skinnkavaj: well, other than bloating the transaction to the size it was.
1159 2013-11-01 19:34:44 <swulf--> BlueMatt: quickest that comes to mind is you can check for duplicate transactions quickly
1160 2013-11-01 19:35:02 <Eliel> basically, the cause is a bug in blockchain.info fee calculation. It suggeted a bit too low fee.
1161 2013-11-01 19:35:08 <BlueMatt> swulf--: well now its set up so you cant have duplicate transactions at all
1162 2013-11-01 19:35:21 <BlueMatt> swulf--: also, duplicate transactions are impossibly complicated to deal with on the reorg codepaths
1163 2013-11-01 19:35:25 Anduck has quit (Ping timeout: 272 seconds)
1164 2013-11-01 19:35:45 <BlueMatt> swulf--: trying to get that code right is just asking for someone to find a chain fork bug
1165 2013-11-01 19:35:47 <swulf--> I imagine they are
1166 2013-11-01 19:38:54 <gfawkes_> wtf. why the hell is all the research on massive multilevel distributed object-oriented data systems from the fuckin' 1990s.  anyone got any recommendations on more recent literature?
1167 2013-11-01 19:39:26 <BlueMatt> gfawkes_: what are you even trying to do?
1168 2013-11-01 19:39:41 <BlueMatt> also, yea, lots of the distributed systems literature is from the 90s...
1169 2013-11-01 19:39:51 <sipa> ThomasV: i made a utxo set size graph at some point
1170 2013-11-01 19:39:51 Application has joined
1171 2013-11-01 19:39:59 <sipa> but it's very outdated now
1172 2013-11-01 19:40:04 <gfawkes_> bluematt - trying to figure out a granular, scalable, and managable security model
1173 2013-11-01 19:40:05 <ThomasV> when?
1174 2013-11-01 19:40:18 <BlueMatt> gfawkes_: that may be the most vague project Ive ever heard
1175 2013-11-01 19:40:50 <gfawkes_> bluematt - it's for a obj-oriented framework ive been writing
1176 2013-11-01 19:41:00 <sipa> ThomasV: http://bitcoin.sipa.be/pruning.png
1177 2013-11-01 19:41:11 <ThomasV> thks
1178 2013-11-01 19:41:44 <gfawkes_> bluematt - originally put the security model in at the service provider, but after a few thousand objects this proves to be a bottle neck
1179 2013-11-01 19:42:01 owowo has joined
1180 2013-11-01 19:43:12 <gfawkes_> bluematt - conversely, when i move it to the database, it speeds it up, but becomes impossible to actually do entirely any object's database may not necessarily have access to the "user" object's database on a different server and/or database
1181 2013-11-01 19:43:29 <gfawkes_> its a giant nightmare.
1182 2013-11-01 19:43:44 <BlueMatt> gfawkes_: distributed is hard
1183 2013-11-01 19:44:09 Application has quit (Ping timeout: 245 seconds)
1184 2013-11-01 19:44:18 RoboTeddy has quit (Ping timeout: 272 seconds)
1185 2013-11-01 19:44:31 <gfawkes_> bluematt - that's why i was hoping someone here might have some pointers to more recent research on distributed security models :)
1186 2013-11-01 19:45:03 <midnightmagic> gmaxwell: maybe proof-of-storage could be used to detect and prioritize full nodes.
1187 2013-11-01 19:45:37 <midnightmagic> and thus incentivize people maintaining full nodes
1188 2013-11-01 19:49:11 Application has joined
1189 2013-11-01 19:49:45 MiningBuddy has quit (Remote host closed the connection)
1190 2013-11-01 19:51:26 <ThomasV> http://en.wikipedia.org/wiki/Antifragile:_Things_That_Gain_from_Disorder
1191 2013-11-01 19:51:44 gavinandresen has quit (Quit: gavinandresen)
1192 2013-11-01 19:51:57 justusranvier has joined
1193 2013-11-01 19:52:06 gavinandresen has joined
1194 2013-11-01 19:52:30 linq has quit (Ping timeout: 264 seconds)
1195 2013-11-01 19:55:19 Diablo-D3 has joined
1196 2013-11-01 19:55:27 Anduck has joined
1197 2013-11-01 19:55:28 Anduck has quit (Changing host)
1198 2013-11-01 19:55:28 Anduck has joined
1199 2013-11-01 19:56:26 MiningBuddy has joined
1200 2013-11-01 19:56:26 MiningBuddy has quit (Changing host)
1201 2013-11-01 19:56:26 MiningBuddy has joined
1202 2013-11-01 19:57:33 i2pRelay has quit (Remote host closed the connection)
1203 2013-11-01 19:57:47 i2pRelay has joined
1204 2013-11-01 20:00:13 Musk has quit (Quit: Linkinus - http://linkinus.com)
1205 2013-11-01 20:00:40 Wild_Bill_ has quit (Quit: Beware of programmers who carry screwdrivers.)
1206 2013-11-01 20:01:08 ielo has quit (Ping timeout: 272 seconds)
1207 2013-11-01 20:01:17 RoboTeddy has joined
1208 2013-11-01 20:04:32 Application has quit (Remote host closed the connection)
1209 2013-11-01 20:08:32 <MC1984> oh thats where that came from
1210 2013-11-01 20:10:17 rdymac has quit (Excess Flood)
1211 2013-11-01 20:11:12 rdymac has joined
1212 2013-11-01 20:11:46 paracyst has joined
1213 2013-11-01 20:15:06 <Tril> gfawkes_ PM
1214 2013-11-01 20:16:56 johnsoft has quit (Ping timeout: 240 seconds)
1215 2013-11-01 20:17:09 johnsoft has joined
1216 2013-11-01 20:18:31 chris_l has quit (Remote host closed the connection)
1217 2013-11-01 20:21:24 abueesp has joined
1218 2013-11-01 20:21:48 abueesp is now known as Guest39269
1219 2013-11-01 20:25:05 AusBitBank has joined
1220 2013-11-01 20:28:24 android_ has joined
1221 2013-11-01 20:32:25 JimJones has quit (Ping timeout: 246 seconds)
1222 2013-11-01 20:33:09 Nopik has quit (Quit: My iMac has gone to sleep. ZZZzzz…)
1223 2013-11-01 20:35:35 _ingsoc has quit (Quit: leaving)
1224 2013-11-01 20:36:24 _ingsoc has joined
1225 2013-11-01 20:37:26 toffoo has joined
1226 2013-11-01 20:38:11 dongshengcn has joined
1227 2013-11-01 20:40:48 MobPhone has quit (Ping timeout: 260 seconds)
1228 2013-11-01 20:41:52 johnsoft has quit (Ping timeout: 246 seconds)
1229 2013-11-01 20:42:16 MobPhone has joined
1230 2013-11-01 20:44:50 ThomasV has quit (Ping timeout: 272 seconds)
1231 2013-11-01 20:44:53 Starduster_ has joined
1232 2013-11-01 20:45:17 johnsoft has joined
1233 2013-11-01 20:46:11 rdymac has quit (Excess Flood)
1234 2013-11-01 20:47:13 rdymac has joined
1235 2013-11-01 20:48:50 Guest39269 has quit (Read error: Connection reset by peer)
1236 2013-11-01 20:54:46 <maaku> gavinandresen: petertodd was saying that I shouldn't be working on scriptPubKey-indexed committed UTXO trees because parasitic apps can use it to store data on the block chain
1237 2013-11-01 20:54:52 <maaku> my response was that we should change fee policy so people pay proportional to their (ab)use of UTXO storage
1238 2013-11-01 20:54:59 <maaku> your name was mentioned because you're the only one I know of who is actually working on fee reform code
1239 2013-11-01 20:56:34 toffoo has quit (Ping timeout: 246 seconds)
1240 2013-11-01 20:58:43 toffoo has joined
1241 2013-11-01 20:59:09 gulli has joined
1242 2013-11-01 20:59:43 Application has joined
1243 2013-11-01 21:00:20 groglogic has quit (Remote host closed the connection)
1244 2013-11-01 21:00:50 <gulli> is it necessary to include "bitcoin:" when you generate a qr code for an address?
1245 2013-11-01 21:01:01 <sipa> maaku: i think we're talking about two generations of approaches
1246 2013-11-01 21:01:07 <gulli> I mean, I know its not necessary, but why have it there to begin with
1247 2013-11-01 21:01:14 <sipa> gulli: it' s URI
1248 2013-11-01 21:01:31 <sipa> gulli: so it can be clickable, browsers can associate it with an application, ...
1249 2013-11-01 21:01:41 <gulli> I did generate a qr code just with the address and my android phone easily transfered to that address
1250 2013-11-01 21:01:48 <sipa> gulli: of course
1251 2013-11-01 21:02:04 <sipa> maaku: UTXO commitments much more closely resemble how bitcoin works today, IMHO
1252 2013-11-01 21:02:04 <gulli> ok
1253 2013-11-01 21:02:29 <sipa> maaku: but the TXO MMR idea is a much more fundamental solution, making all storage the responsabiliy of clients
1254 2013-11-01 21:02:38 <maaku> sipa: UTXO commitments meaning petertodd's new MMR approach?
1255 2013-11-01 21:03:22 <sipa> no, UTXO commitsments are what you are doing
1256 2013-11-01 21:03:24 <sipa> you commit to a particularly structured form of the UTXO set
1257 2013-11-01 21:03:54 <sipa> petertodd's MMR idea makes blocks commit to an ever-growing structured TXO set, which they don't maintain themself
1258 2013-11-01 21:04:30 <maaku> petertodd: it would be helpful if you referred to your achitecture as MMR or something similarly unique. pretty much all the proposed alternatives are "txo commitments"
1259 2013-11-01 21:04:56 <sipa> txo commitments are very different from utxo commitments :)
1260 2013-11-01 21:05:03 <dobry-den> What's the simplest way to start a node in isolation for testing?
1261 2013-11-01 21:05:18 <petertodd> maaku: oh sorry, I thought everyone was calling it UTXO commitments
1262 2013-11-01 21:05:24 <maaku> sipa: yes, that's my thinking as well. committing a tree index is a direct continuation of bitcoin's current model. MMR has some real advantages in certain future scenarios, but we should make that choice as we get closer to needing it
1263 2013-11-01 21:06:03 <dobry-den> In particular, I just want debug.log to show things relevant to my client.
1264 2013-11-01 21:06:31 <maaku> since it fundamentally changes how wallets and miners operate
1265 2013-11-01 21:06:50 <sipa> maaku: i'm very uncertain whether either will make it into bitcoin-as-we-know-it
1266 2013-11-01 21:06:50 <sipa> maaku: though i'm eager to see what you're reached
1267 2013-11-01 21:07:47 <petertodd> maaku: the thing is, it only changes how miners and wallets operate *if* we need to; as long as the UTXO set is sufficiently small we can use TXO commitments exactly as UTXO commitments would be used.
1268 2013-11-01 21:07:57 <sipa> indeed, UTXO commitments only change how full nodes work (and enable a few novel modes of operation)
1269 2013-11-01 21:07:57 <sipa> TXO commitments (in their full form) require different wallets
1270 2013-11-01 21:08:16 <maaku> sipa: not even the txid:n index? I thought the scriptPubKey index was the only controversial change
1271 2013-11-01 21:08:21 <petertodd> Basically, having the UTXO set, a full UTXO set, means you don't need TXO proofs at all and can ignore the fact that you can make them.
1272 2013-11-01 21:08:32 <sipa> maaku: ?
1273 2013-11-01 21:08:34 <petertodd> maaku: txid:n index is a full UTXO set though.
1274 2013-11-01 21:08:39 <maaku> ah i misread you
1275 2013-11-01 21:08:52 <maaku> thought you said neither one would make it into bitcoin-core
1276 2013-11-01 21:09:01 <sipa> i said that
1277 2013-11-01 21:09:10 <sipa> even UTXO commitments put a very significant extra load on full nodes
1278 2013-11-01 21:09:21 <warren> is there a minimum protobuf version requirement?
1279 2013-11-01 21:09:23 <petertodd> maaku: look at how much of a fight there was over P2SH, and it was maybe 25 lines of consensus related code :(
1280 2013-11-01 21:09:29 <sipa> and we really need to know how it scales before enforcing it
1281 2013-11-01 21:11:05 <petertodd> maaku: I'd quite seriously bet you $1000 that we won't see (U)TXO commitments of any kind in Bitcoin within 3 years. Heck, I'd make the same no-change bet about the blocksize.
1282 2013-11-01 21:11:27 <petertodd> maaku: (as a enforced rule, optional merge-mined is different)
1283 2013-11-01 21:11:41 <sipa> oh, before you misunderstand me: i'm a large proponent of both ideas, and i really want to get that sort of evolution to hapen... i'm just suspicious of whether there will be enough consensus about it
1284 2013-11-01 21:12:12 <petertodd> maaku: I'd suggest you propose it to Litecoin too FWIW.
1285 2013-11-01 21:12:15 <maaku> petertodd: well, i lack the commit bit so it's up to the prejudices of those who have it
1286 2013-11-01 21:12:24 <sipa> commit bit?
1287 2013-11-01 21:12:48 <petertodd> No-body has a commit bit for this stuff...
1288 2013-11-01 21:13:21 <sipa> ah, you mean write access to bitcoin/bitcoin?
1289 2013-11-01 21:13:21 <sipa> meh, this is a choice that has to be made by everyone
1290 2013-11-01 21:13:31 <petertodd> With the noise Amir and the Dark Wallet people are making it might not be long before the so-called "core devs" don't even have much political sway.
1291 2013-11-01 21:13:33 <sipa> i'm sure client developers will have an important voice
1292 2013-11-01 21:13:48 <petertodd> (IE, imagine if they start getting some people mining on that code)
1293 2013-11-01 21:14:01 <sipa> but i sure as hell hope "we" don't have the power to force such a change onto the world
1294 2013-11-01 21:14:36 <maaku> sipa: yes, but I don't see how else it could be made aside from the official client bumping a soft-fork version number, and voting a la P2SH
1295 2013-11-01 21:14:55 <warren> Forcing changes on the world is easy when you don't have vendors integrated yet.  https://github.com/litecoin-project/litecoin/pull/80
1296 2013-11-01 21:14:56 <sipa> maaku: that's the final step
1297 2013-11-01 21:14:59 <maaku> in which case, practically speaking, it won't happen unless those with control over the release process want it to
1298 2013-11-01 21:15:18 <sipa> maaku: the step before that, is getting everyone important to agree on it
1299 2013-11-01 21:15:53 <petertodd> maaku: you've got it backwards frankly. The people with control over the release process are the ones deciding if they want to use the releases.
1300 2013-11-01 21:16:05 <sipa> petertodd: in theory
1301 2013-11-01 21:16:08 <maaku> petertodd: that's a nice thought
1302 2013-11-01 21:16:14 <maaku> not how the real world works
1303 2013-11-01 21:16:28 <petertodd> For minor changes, I'm wrong, but for major changes, that's when politics gets into it and you have to actually convince people.
1304 2013-11-01 21:16:58 <gulli> Im going to be making a web application that needs to generate a new address for signed in users, each transactions and so forth. Can anyone point me to the right directions where I should start, that is, some documentation or something? :)
1305 2013-11-01 21:16:59 <petertodd> commitments aren't going to happen without the consent of, at bare minimum, a majority of pool operators.
1306 2013-11-01 21:17:24 <maaku> unless, say, 75% of people just don't care - even if they should - and default to whatever the core developers decide
1307 2013-11-01 21:18:00 <petertodd> What is neat about them, is the implementation can absolutely be to merge-mine them, and like height-in-coinbase, flip the switch to enforce them at the appropriate supermajority.
1308 2013-11-01 21:18:18 GingerGeek is now known as GingerGeek[Away]
1309 2013-11-01 21:18:49 <petertodd> maaku: right, which is true with minor stuff. With major stuff if there is real controversy those people will start wondering what they should do, and the suicide-pact nature of Bitcoin's consensus will make everything uncertain.
1310 2013-11-01 21:20:24 <petertodd> I'd say deployment should probably be a minimum year old trial of merge-mining them without enforcement at all, and then try to get a good sense of what consensus failures crop up. (maybe we can automatically send failures over the p2p network even?)
1311 2013-11-01 21:20:34 <petertodd> *failure reports
1312 2013-11-01 21:22:38 toffoo has quit (Ping timeout: 240 seconds)
1313 2013-11-01 21:22:40 <warren> merge-mining?
1314 2013-11-01 21:22:53 satamusic has quit (Ping timeout: 245 seconds)
1315 2013-11-01 21:23:21 <petertodd> warren: yes, as in the (U)TXO commitment is calculated and included in the block, but correctness isn't enforced
1316 2013-11-01 21:23:38 <maaku> petertodd: that's not merged mining...
1317 2013-11-01 21:23:58 satamusic has joined
1318 2013-11-01 21:24:16 <petertodd> maaku: it's the term you used before, how do you see me using it wrong?
1319 2013-11-01 21:24:50 <maaku> petertodd: the commitments would be on an aux-chain, validated by most-work
1320 2013-11-01 21:25:00 toffoo has joined
1321 2013-11-01 21:25:57 <petertodd> maaku: right, and if you include them in blocks but don't enforce them, they can be validated by most-work and the chain just happens to include some unrelated headers. (make your implementation commit to the previous commitment it considers valid of course)
1322 2013-11-01 21:26:28 <sipa> petertodd: the difference is enforcing that every block has a valid commitment
1323 2013-11-01 21:26:32 <sipa> eh, maaku:
1324 2013-11-01 21:26:47 <sipa> vs it being optional (through an aux chain that you do or do not merge-mine against)
1325 2013-11-01 21:26:47 <maaku> yes, so the most recent committed aux header seen in the bitcoin chain might be wrong (or out of date), since it might not be the most-work aux chain branch
1326 2013-11-01 21:27:03 <sipa> the point should certainly be to have it enforced eventually
1327 2013-11-01 21:27:11 <petertodd> sipa: yes I know, and if you aren't enforcing it, but do commit to what you thought was the previous valid commitment in the chian, you've effectively made a merge-mined chain with difficulty == bitcoin diff.
1328 2013-11-01 21:27:20 <sipa> but merge mining is a nice way of experimenting with it, without burdening every full node
1329 2013-11-01 21:27:22 <maaku> well, aux-chain miners are supposed to strictly validate, but the reader would just to SPV most-work
1330 2013-11-01 21:27:44 markus__ has quit (Read error: Connection reset by peer)
1331 2013-11-01 21:27:59 <petertodd> maaku: right, and any aux chain done totally independently, doesn't give SPV nodes any more ability to validate than a "merge-mined" chain that happens to be the exact same headers.
1332 2013-11-01 21:31:32 <gavinandresen> is there consensus on how to compute the UTXO committment hash yet?
1333 2013-11-01 21:31:58 <petertodd> gavinandresen: there isn't even consensus that we want to compute a UTXO hash instead of a TXO hash
1334 2013-11-01 21:32:03 <phantomcircuit> i like cookies
1335 2013-11-01 21:32:04 <maaku> gavinandresen: consensus, no, I have a prototype for the radix-tree version, and I'm converting that to C++ & writing a series of BIPs
1336 2013-11-01 21:32:17 <petertodd> phantomcircuit: cookie hash?
1337 2013-11-01 21:32:18 <gavinandresen> okey dokey.  I'll go back to ignoring that whole issue until there is consensus.
1338 2013-11-01 21:32:21 <maaku> I know ThomasV and some others have their own implementations
1339 2013-11-01 21:32:43 <phantomcircuit> petertodd, no the baked kind
1340 2013-11-01 21:32:52 <petertodd> maaku: I guess there's a implementation in bitcoind too, I forget if it's a tree though
1341 2013-11-01 21:33:13 <maaku> petertodd: no, it's a serialized dump of the db
1342 2013-11-01 21:33:30 <sipa> phantomcircuit: hash cookies?
1343 2013-11-01 21:33:46 <gavinandresen> somebody should write a whitepaper with pros and cons of the various committment schemes, mabye consensus would emerge
1344 2013-11-01 21:34:03 <sipa> maaku: indeed
1345 2013-11-01 21:34:10 <petertodd> gavinandresen: working on one actually
1346 2013-11-01 21:34:32 <gavinandresen> petertodd: do you have a dog in that fight?
1347 2013-11-01 21:34:34 <petertodd> Also working on a truly insane TXIN commitment scheme...
1348 2013-11-01 21:34:34 <sipa> not useful
1349 2013-11-01 21:34:34 <sipa> (but it's an easy way to check whether two UTXO sets match)
1350 2013-11-01 21:34:47 <petertodd> gavinandresen: yes, I came up with something called MMR TXO commitments
1351 2013-11-01 21:34:58 <gavinandresen> right, bagging peaks....
1352 2013-11-01 21:35:16 <petertodd> gavinandresen: yup, which gives the long-term possibility of not storing the UTXO set at all on full nodes and miners
1353 2013-11-01 21:35:39 jakov has joined
1354 2013-11-01 21:35:44 <sipa> i'm still not entirely convinced it's feasible
1355 2013-11-01 21:35:54 <sipa> but it does sound awesome
1356 2013-11-01 21:36:16 <maaku> there are so many tradeoffs between these designs I don't know if full consensus will ever be reached :\
1357 2013-11-01 21:36:24 <maaku> they're all pick N of M features
1358 2013-11-01 21:36:26 <petertodd> sipa: yeah, fully implemented it's a log(n) bandwidth increase, very roughly speaking, but by the time you actually need it hopefully we all have terabit fiber :p
1359 2013-11-01 21:36:28 <sipa> ETOOMANYSTANDARDS
1360 2013-11-01 21:36:31 <maaku> nothing's got everything
1361 2013-11-01 21:37:15 <petertodd> maaku: heh, the annoying thing about all the stuff I've been working on is how they scale just great to every transaction on the planet, but are x times more bandwidth initially
1362 2013-11-01 21:38:12 <maaku> petertodd: what's bizarre is you working on this stuff at all. I don't know what the need is unless we're increasing the block size...
1363 2013-11-01 21:38:33 <petertodd> maaku: I care about the scalability of Bitcoin, I don't care how we get there.
1364 2013-11-01 21:39:13 <sipa> maaku: my opinion is that the block size should be increased to exactly what the network can tolerate
1365 2013-11-01 21:39:26 <petertodd> maaku: I mean, heck, the TXIN commitments thing I'm also working on, means the blockchain doesn't even have transactions in it at all, and I don't see any reason why it couldn't scale to any transaction volume required.
1366 2013-11-01 21:39:33 <sipa> maaku: if some technology allows scaling an order of magnitude further, that is awesome
1367 2013-11-01 21:41:19 <petertodd> maaku: I'm quite glad that I think I've found ways to make larger blocksizes more auditable, and maybe even as far as make decentralized mining possible at any scale; I like being proven wrong by yet more clever ideas!
1368 2013-11-01 21:41:27 groglogic has joined
1369 2013-11-01 21:41:39 <maaku> well, agreed, it was more of an ironic comment
1370 2013-11-01 21:41:55 eoss has quit (Read error: Connection reset by peer)
1371 2013-11-01 21:42:16 <maaku> I came to this issue from the perspective of wanting 100+MB block sizes, and glad that I can still find some agreement on it
1372 2013-11-01 21:42:35 <maaku> on scalability solutions i mean
1373 2013-11-01 21:43:39 ThomasV has joined
1374 2013-11-01 21:44:03 <petertodd> maaku: just keep in mind that it's easy to increase tx volume, harder to do auditing, harder still to make mining censorship resistance, and then after that you still need to fix the incentive to mine at all problem in the long run as inflation drops
1375 2013-11-01 21:44:06 pecket has quit (Ping timeout: 265 seconds)
1376 2013-11-01 21:44:14 pecket has joined
1377 2013-11-01 21:44:19 <petertodd> fortunately the last one we can wait, ~10+ years away
1378 2013-11-01 21:45:03 rdymac has quit (Excess Flood)
1379 2013-11-01 21:45:21 <maaku> you still need to fix the incentive to mine at all problem in the long run as inflation drops <-- like I did with freicoin?
1380 2013-11-01 21:45:22 * maaku runs
1381 2013-11-01 21:45:35 <petertodd> maaku: lol, well, yes actually!
1382 2013-11-01 21:46:18 <petertodd> maaku: txin commitment scheme I'm writing up would balance inflation to mining reward, so if you contribute back your % of hashing power, basically you're wealth stays constant
1383 2013-11-01 21:47:05 <petertodd> maaku: I think that kind of scheme just can't get political viability until bitcoin gets killed by a 51% attack
1384 2013-11-01 21:47:43 rdymac has joined
1385 2013-11-01 21:49:22 <maaku> maybe that's true in the bitcoin world which has accumulated it's own ideological bias. on the other hand we're seing lots of interest in freicoin/demurrage from the regional currency movement in europe
1386 2013-11-01 21:49:31 <maaku> but ill shutup before i get banned for being off-topic
1387 2013-11-01 21:53:54 Evilmax has quit ()
1388 2013-11-01 21:58:22 MoALTz has quit (Quit: Leaving)
1389 2013-11-01 21:59:15 RoboTeddy has quit (Remote host closed the connection)
1390 2013-11-01 22:00:05 RoboTeddy has joined
1391 2013-11-01 22:01:21 Bog4r7 has joined
1392 2013-11-01 22:02:27 gingpark has quit (Ping timeout: 272 seconds)
1393 2013-11-01 22:04:14 toffoo has quit ()
1394 2013-11-01 22:04:45 <midnightmagic> petertodd: woe be unto the miner who suffers an unpleasant long-term powerout.
1395 2013-11-01 22:06:29 <midnightmagic> or earthquake. or tsunami.. nuclear power accident.. coal burning plant catastrophe.. dam breach.. theft.. sabotage.. vandalism..
1396 2013-11-01 22:09:14 k00shi has joined
1397 2013-11-01 22:09:29 k00shi has left ()
1398 2013-11-01 22:09:47 RoboTeddy has quit (Remote host closed the connection)
1399 2013-11-01 22:10:28 <gulli> Anyone here using bitcoinj?
1400 2013-11-01 22:10:41 RoboTeddy has joined
1401 2013-11-01 22:11:14 cads has quit (Ping timeout: 272 seconds)
1402 2013-11-01 22:13:44 transisto has quit (Read error: Connection reset by peer)
1403 2013-11-01 22:15:55 oPen_syLar has joined
1404 2013-11-01 22:16:02 MagicFab_ has quit (Ping timeout: 272 seconds)
1405 2013-11-01 22:17:20 <Bog4r7> For a while I had to have the wife tell be when she planned to use the oven, and I'd have to shut down some miners.  Sometimes remotely.
1406 2013-11-01 22:19:48 Coincidental has quit (Remote host closed the connection)
1407 2013-11-01 22:20:28 Coincidental has joined
1408 2013-11-01 22:25:10 Coincidental has quit (Ping timeout: 272 seconds)
1409 2013-11-01 22:30:56 viperhr has joined
1410 2013-11-01 22:32:30 Evilmax has joined
1411 2013-11-01 22:35:34 ThomasV has quit (Remote host closed the connection)
1412 2013-11-01 22:36:05 Beef has quit (Read error: Connection reset by peer)
1413 2013-11-01 22:36:40 zer0def has quit (Ping timeout: 246 seconds)
1414 2013-11-01 22:39:01 patcon_ has quit (Remote host closed the connection)
1415 2013-11-01 22:39:33 patcon_ has joined
1416 2013-11-01 22:42:37 tmsk has quit (Quit: tmsk)
1417 2013-11-01 22:42:43 PRab has quit (Remote host closed the connection)
1418 2013-11-01 22:43:11 agricocb has quit (Quit: Leaving.)
1419 2013-11-01 22:44:01 patcon_ has quit (Ping timeout: 246 seconds)
1420 2013-11-01 22:46:07 daybyter has quit (Quit: Konversation terminated!)
1421 2013-11-01 22:47:57 Beef has joined
1422 2013-11-01 22:50:39 <warren> bitcoin master crashes for me
1423 2013-11-01 22:50:43 <warren> [warren@apotheosis bitcoin]$ src/qt/bitcoin-qt
1424 2013-11-01 22:50:43 <warren> bitcoin-qt: key.cpp:135: {anonymous}::CECKey::CECKey(): Assertion `pkey != __null' failed.
1425 2013-11-01 22:50:44 <warren> Aborted (core dumped)
1426 2013-11-01 22:51:21 linq has joined
1427 2013-11-01 22:52:31 patcon_ has joined
1428 2013-11-01 22:54:03 _ingsoc has quit (Quit: leaving)
1429 2013-11-01 22:56:37 sustrik has quit (Ping timeout: 246 seconds)
1430 2013-11-01 22:57:13 dongshengcn is now known as ds|away
1431 2013-11-01 22:58:05 <sipa> warren: isn't that the lack of secp256k1 in your OpenSSL?
1432 2013-11-01 22:58:33 patcon_ has quit (Remote host closed the connection)
1433 2013-11-01 22:58:49 gingpark has joined
1434 2013-11-01 22:59:02 patcon has joined
1435 2013-11-01 23:01:13 transisto has joined
1436 2013-11-01 23:02:07 Tantadruj has quit (Quit: DoubleRecall Turns Paywalls Into Advertising Dollars - NYTimes.com http://nyti.ms/odHOgy)
1437 2013-11-01 23:04:10 patcon has quit (Ping timeout: 272 seconds)
1438 2013-11-01 23:04:59 _Avyd has quit (Ping timeout: 245 seconds)
1439 2013-11-01 23:05:46 ds is now known as away!ds@gateway/shell/cloudant/x-kqsgikodvlabxlrd|dongshengcn
1440 2013-11-01 23:05:49 _Avyd has joined
1441 2013-11-01 23:07:31 dongshengcn is now known as ds|away
1442 2013-11-01 23:12:39 <Luke-Jr> std::bad_alloc
1443 2013-11-01 23:12:40 <Luke-Jr> bitcoin in ThreadMessageHandler()
1444 2013-11-01 23:12:42 <Luke-Jr> O.o
1445 2013-11-01 23:18:57 zer0def has joined
1446 2013-11-01 23:19:09 transisto has quit ()
1447 2013-11-01 23:20:35 transisto has joined
1448 2013-11-01 23:22:43 <warren> ah crap
1449 2013-11-01 23:22:45 <warren> they updated it again
1450 2013-11-01 23:23:01 <warren> sipa: thanks
1451 2013-11-01 23:28:44 Diablo-D3 has quit (Quit: This computer has gone to sleep)
1452 2013-11-01 23:30:07 ralphtheninja has quit (Quit: leaving)
1453 2013-11-01 23:30:49 knotwork_ has joined
1454 2013-11-01 23:30:57 lianj_ is now known as lianj
1455 2013-11-01 23:30:59 knotwork has quit (Read error: Operation timed out)
1456 2013-11-01 23:34:13 Starduster_ has quit (Quit: gotta go)
1457 2013-11-01 23:34:58 groglogi_ has joined
1458 2013-11-01 23:37:27 groglogic has quit (Ping timeout: 272 seconds)
1459 2013-11-01 23:39:42 bbrian has quit (Ping timeout: 240 seconds)
1460 2013-11-01 23:48:40 Diablo-D3 has joined
1461 2013-11-01 23:49:43 rdymac has quit (Excess Flood)
1462 2013-11-01 23:50:07 debiantoruser has quit (Ping timeout: 272 seconds)
1463 2013-11-01 23:51:31 hnz has quit (Ping timeout: 252 seconds)
1464 2013-11-01 23:51:42 debiantoruser has joined
1465 2013-11-01 23:52:13 rdymac has joined
1466 2013-11-01 23:56:22 hnz has joined
1467 2013-11-01 23:58:14 Coincidental has joined
1468 2013-11-01 23:59:50 Polyatomic has joined