00:34:23 | jae: | jae is now known as Guest67735 |
01:03:43 | lnovy: | lnovy is now known as zz_lnovy |
01:37:05 | jae: | jae is now known as Guest94104 |
03:09:02 | triazo: | triazo is now known as triazo_ |
03:09:11 | triazo_: | triazo_ is now known as triazo |
03:09:24 | triazo: | triazo is now known as triazo_ |
03:09:57 | triazo_: | triazo_ is now known as triazo |
03:11:53 | triazo: | triazo is now known as triazo_ |
03:13:27 | triazo_: | triazo_ is now known as triazo |
03:24:47 | triazo: | triazo is now known as triazo_ |
03:24:58 | triazo_: | triazo_ is now known as triazo |
03:32:23 | triazo: | triazo is now known as triazo_ |
03:32:29 | triazo_: | triazo_ is now known as triazo |
03:33:11 | zz_lnovy: | zz_lnovy is now known as lnovy |
04:14:01 | Taek: | Assuming transactions were ~500 bytes each, launch a DoS on the transaction pool to push transactions up to 1c each would cost $312,000 per day. |
04:14:47 | Taek: | If I were to instead spend that money on bandwidth and try to do a traditional ddos on the mempool, would I get farther? |
04:27:46 | Luke-Jr: | Taek: you mean 1c ea to get confirmed within … how long? |
04:28:54 | Taek: | confirmed before the DoS attack ends. If you're dumping $312,000 per day into fees, you can continuously keep the mempool full of 1c transactions |
04:31:55 | Taek: | s/launch/launching |
05:27:42 | bosma_: | bosma_ is now known as bosma |
05:34:43 | lnovy: | lnovy is now known as zz_lnovy |
06:38:17 | zz_lnovy: | zz_lnovy is now known as lnovy |
07:28:43 | gwillen: | gwillen is now known as Guest53442 |
08:05:13 | sinisalo.freenode.net: | topic is: This channel is not about short-term Bitcoin development | http://bitcoin.ninja/ | This channel is logged. | For logs and more information, visit http://bitcoin.ninja |
08:05:13 | sinisalo.freenode.net: | Users on #bitcoin-wizards: andy-logbot Relos Guyver2 adam3us Guest53442 b_lumenkraft gielbier ThomasV hktud0 Mably sl01 sparetire_ DrWat thrasher` felipelalli priidu copumpkin pollux-bts bosma TheSeven hulkhogan_ akstunt600 SubCreative v7wb96g GAit Dr-G AirBreather d1ggy_ NewLiberty DougieBot5000 superobserver spinza belcher mm_1 mkarrer [d__d] Madars Tiraspol kyuupichan gavinandresen nickler EasyAt hashtag Iriez dc17523be3 ttttemp mountaingoat Transisto2 jmcn koshii |
08:05:13 | sinisalo.freenode.net: | Users on #bitcoin-wizards: gmaxwell PRab go1111111 wizkid057 justanotheruser Luke-Jr akrmn cdecker harrow dansmith_btc face metamarc Keefe jrayhawk K1773R Guest95228 luny Emcy melvster tromp_ prodatalab stonecoldpat theymos waxwing Apocalyptic LeMiner ebfull prosodyContext ggreer tromp Hunger- PaulCapestany c0rw|zZz Pan0ram1x jcorgan isis Cory cryptowest_ mengine sneak HM grandmaster Starduster bsm117532 lnovy harrigan andytoshi scoria brand0 larraboj bedeho2 OneFixt |
08:05:13 | sinisalo.freenode.net: | Users on #bitcoin-wizards: vonzipper mappum MoALTz Jouke weex gnusha nsh triazo jonasschnelli CryptoGoon berndj leakypat platinuum Oizopower jbenet dasource btcdrak tucenaber helo Taek iddo GreenIsMyPepper amiller epscy adams__ wiz michagogo mikolalysenko artifexd lmacken yrashk cfields Krellan coryfields catlasshrugged Alanius null_radix kanzure bliljerk_ azariah warptangent sparetire davout comboy TD-Linux yorick crescend1 veox Zouppen huseby _whitelogger wumpus |
08:05:13 | sinisalo.freenode.net: | Users on #bitcoin-wizards: binaryatrocity heath BananaLotus maaku optimator Eliel narwh4l mr_burdell throughnothing_ elastoma fluffypony Fistful_of_Coins yoleaux Jaamg xabbix mariorz catcow a5m0 smooth dignork runeks Sqt poggy livegnik petertodd richardus nephyrin phedny so afdudley pigeons SwedFTP guruvan ajweiss nanotube forrestv Muis warren Xzibit17 sdaftuar eric roasbeef s1w CryptOprah morcos merlincorey [ace] sturles jaromil Graet indolering ryan-c jessepollak |
08:05:13 | sinisalo.freenode.net: | Users on #bitcoin-wizards: gribble d9b4bef9 starsoccer kumavis otoburb midnightmagic BlueMatt Anduck AdrianG BrainOverfl0w @ChanServ kinlo STRML espes |
08:07:52 | lnovy: | lnovy is now known as zz_lnovy |
08:51:38 | zz_lnovy: | zz_lnovy is now known as lnovy |
09:04:55 | Guest53442: | Guest53442 is now known as gwillen |
10:25:52 | dabura667: | Is there a way to downgrade to pre 0.10.0 without re-downloading the blockchain? |
10:26:09 | dabura667: | My blockchain files are out of order a la headers first. |
10:26:36 | gmaxwell: | no. |
10:26:50 | dabura667: | d'oh |
10:26:50 | gmaxwell: | well okay, you could linearize to create a bootstrap.dat |
10:26:55 | gmaxwell: | and the the old version can use that. |
10:27:01 | gmaxwell: | but uhh.. why are you downgrading? |
10:27:08 | dabura667: | insight |
10:27:13 | gmaxwell: | Seriously vulnerabilties hoy in old versions. |
10:27:32 | dabura667: | Or I could wait until insight supports out of order blocks |
10:27:55 | dabura667: | But instant gratification! (sigh) |
10:28:58 | dabura667: | I'll have to weigh my options... Thanks for the answer. |
10:39:04 | gmaxwell: | dabura667: or look into fixing it yourself? why does it access the block files? |
10:39:14 | gmaxwell: | e.g. if it only does it to speed up the initial sync, perhaps you can just bypass that. |
10:45:41 | maaku: | dabura667: linearize.py and then point insight at the bootstrap.dat? |
10:49:32 | antanst: | dabura667: You can sync insight via RPC, which avoids the blocks-out-of-order problem, albeit slower |
12:00:18 | c0rw|zZz: | c0rw|zZz is now known as c0rw1n |
12:53:45 | DanielBTC: | DanielBTC is now known as Guest91277 |
14:35:29 | Taek: | are these legitimate concerns?: http://www.reddit.com/r/Bitcoin/comments/37tqie/psa_miners_please_use_blockmaxsize1000000_cause/crpwbmj |
15:00:25 | pigeons: | pigeons is now known as Guest53003 |
15:36:19 | kanzure: | better link http://www.reddit.com/r/Bitcoin/comments/37tqie/psa_miners_please_use_blockmaxsize1000000_cause/crpwbmj?context=2 |
15:38:16 | kanzure: | Taek: although not a network propagation detail, it is perhaps worth mentioning that large miners, if there is an easy attack vector, might be compelled by others to actually execute that attack (such as law enforcement) |
15:47:17 | pigeons: | pigeons is now known as Guest25661 |
16:01:43 | jae: | jae is now known as Guest14405 |
16:47:51 | pigeons: | pigeons is now known as Guest78951 |
18:30:14 | jae: | jae is now known as Guest59515 |
20:01:03 | jae: | jae is now known as Guest98664 |
20:06:15 | DanielBTC: | DanielBTC is now known as Guest82874 |
21:20:49 | jae: | jae is now known as Guest25396 |
21:25:51 | bosma_: | bosma_ is now known as bosma |
21:42:51 | jae: | jae is now known as Guest80839 |
22:51:51 | nsh: | .title https://www.youtube.com/watch?v=BwJXiJqG9gI |
22:51:51 | yoleaux: | HotPI Debate: Reversible computing and the validity of Landauer's principle - YouTube |
22:52:27 | jae: | jae is now known as Guest48803 |
23:44:48 | jae: | jae is now known as Guest77004 |
23:55:57 | sausage_factory: | sausage_factory is now known as priidu |