1 2013-02-02 00:00:03 <CodeShark> thta's been known to happen
  2 2013-02-02 00:00:36 <HM> would that still work with split keying
  3 2013-02-02 00:00:50 <HM> or does that eliminate any possibility of poor randomisation
  4 2013-02-02 00:00:54 <HM> it does doesn't it
  5 2013-02-02 00:00:56 <HM> hmm
  6 2013-02-02 00:02:29 <HM> good stuff
  7 2013-02-02 00:38:06 <bsdunx> Is there a preferred way to support debug builds with this qmake generator, i've been hacking the .pro file for support of gdb and disabling stack-protector in debug build
  8 2013-02-02 00:55:50 <bsdunx> HM was that you I was speaking with yesterday?
  9 2013-02-02 01:05:42 <HM> bsdunx: about what?
 10 2013-02-02 01:05:57 <HM> i'm not familiar with qmake, only cmake
 11 2013-02-02 01:06:15 <bsdunx> unit test related patch
 12 2013-02-02 01:06:24 <HM> nope
 13 2013-02-02 01:06:36 <bsdunx> sorry I don't have a log perhaps it was TD
 14 2013-02-02 01:06:49 <HM> that seems like a reasonable assumption
 15 2013-02-02 01:06:50 <bsdunx> some two letter capitalized nick =s
 16 2013-02-02 01:07:19 <Luke-Jr> bsdunx: qmake bitcoin-qt.pro CONFIG+=debug DEBUGFLAGS=-ggdb
 17 2013-02-02 01:07:26 <gmaxwell> bsdunx: why are you disabling the stack protector
 18 2013-02-02 01:07:43 <bsdunx> to not mess with debugger
 19 2013-02-02 01:07:46 <gmaxwell> I don't recall it ever gumming up debugging.
 20 2013-02-02 01:07:57 <gmaxwell> It shouldn't.
 21 2013-02-02 01:25:29 <moore> I wrote up a post on how people are "steeling" from satoshi dice if any one is interested
 22 2013-02-02 01:27:25 <jgarzik> Avalon miner: 66.3 Ghps / 620 Watts / 5.6 Amps
 23 2013-02-02 01:28:44 <gmaxwell> jgarzik: might want to consider running yours on 240v... a lot (all?) of pc power supplies are a couple percent more efficient at 240v.
 24 2013-02-02 01:29:17 <gmaxwell> moore: where?
 25 2013-02-02 01:31:27 <moore> https://plus.google.com/u/0/106313804833283549032/posts/Ccg2VreMTXw
 26 2013-02-02 01:31:33 <moore> it is short
 27 2013-02-02 01:32:38 <HM> hmmm
 28 2013-02-02 01:32:55 <MC1984> didnt avalon quote 400 watts?
 29 2013-02-02 01:33:08 <MC1984> umm over 50% more power usage isnt good
 30 2013-02-02 01:33:23 <moore> I did not want to just drop the link in the channel unless people wanted to see it as don't know if people are intrested it what I have been talking about
 31 2013-02-02 01:33:28 <gmaxwell> MC1984: no they gave a range which included 600..
 32 2013-02-02 01:33:42 <gmaxwell> and at 240v I bet that psu is under 600.
 33 2013-02-02 01:34:04 <HM> moore: i find it interesting. ty
 34 2013-02-02 01:34:08 <MC1984> why is 240 more efficient
 35 2013-02-02 01:35:01 <gmaxwell> makes the switching frequency higher and the little transformer in the switching psu is more efficient at a higher frequency. I think. I never much cared _why_ it was more efficient.
 36 2013-02-02 01:35:24 <Luke-Jr> moore: they should be applauded
 37 2013-02-02 01:35:39 <gmaxwell> moore: well written.
 38 2013-02-02 01:35:41 <MC1984> oh yeah they use little switchmodes
 39 2013-02-02 01:35:57 <moore> who should be applauded?
 40 2013-02-02 01:36:05 <moore> thanks
 41 2013-02-02 01:36:24 <Luke-Jr> moore: anyone who stops Dice from abusing the network
 42 2013-02-02 01:36:30 <gmaxwell> I assume luke means The double spending SD players. :)
 43 2013-02-02 01:36:38 <HM> Luke-Jr: is it abuse really?
 44 2013-02-02 01:36:41 <moore> well I think it is happening
 45 2013-02-02 01:36:44 <Luke-Jr> HM: yes
 46 2013-02-02 01:36:44 <MC1984> it could be 400 watts thermal and 620 at the wall
 47 2013-02-02 01:37:01 <MC1984> i wonder if jgarzik could get the model number of that psu
 48 2013-02-02 01:37:06 <Luke-Jr> moore: I should get Inaba to upgrade EclipseMC. The latest block_dice code is more complex to block inputs as well :D
 49 2013-02-02 01:37:14 <moore> I bet that they will either have to raise there take quite a bit or stop
 50 2013-02-02 01:37:17 <gmaxwell> HM: they use the network very inefficiently, creating externalized costs in the process.  Thats the facts of it... is it abuse is a boring question of definitions.
 51 2013-02-02 01:38:21 <HM> it's hard to say it's malicious
 52 2013-02-02 01:38:33 <gmaxwell> in particular their txout dust loss notification transactions may have a really significant long term burden.
 53 2013-02-02 01:38:34 <Luke-Jr> HM: they know it harms Bitcoin and they still do it. that's malicious IMO.
 54 2013-02-02 01:38:49 <moore> if my post gets any attention I suspect it will not help SD out much
 55 2013-02-02 01:39:21 <HM> modern western living harms the environment, but we still do it. does that make us malicious? because we're wasteful and don't care enough to change?
 56 2013-02-02 01:39:25 <gmaxwell> HM: they were asked to make some tweaks to improve the efficiency (Even as little as using compressed pubkeys) and the response can best be characterized by 'lol ur problem dudes'.
 57 2013-02-02 01:39:26 <moore> gmaxwell, why long term?
 58 2013-02-02 01:40:14 <HM> i'm not defending them, i just think if  it's malicious then it really is our problem
 59 2013-02-02 01:40:34 <HM> when someone *really* malicious comes along how will we deal if we don't cope with a troll service
 60 2013-02-02 01:40:35 <gmaxwell> moore: because they bloat the set of unspent transactions forever (they're so low value they likely will never be worth the txfees to spend them until they get lost) and the cache locality of the coins database is important to transaction verificiation importance...
 61 2013-02-02 01:40:38 <MC1984> isnt there a way to incentivise more efficient tx structures
 62 2013-02-02 01:40:39 <Luke-Jr> HM: maybe.
 63 2013-02-02 01:40:46 <gmaxwell> and unspent outputs can't be pruned, so more storage too.
 64 2013-02-02 01:40:48 <MC1984> fee policy?
 65 2013-02-02 01:41:05 <Luke-Jr> HM: well, if people are turning their malice against them this way, it seems we have a defence :D
 66 2013-02-02 01:41:08 <gmaxwell> MC1984: /free/ policy, at least... fee policy not so much.
 67 2013-02-02 01:41:19 <moore> ok
 68 2013-02-02 01:41:29 <HM> convince the big mining networks to block them, then game them and put them under
 69 2013-02-02 01:41:30 <MC1984> they pay for all thier txn
 70 2013-02-02 01:41:31 <moore> how well is indexDB working out?
 71 2013-02-02 01:41:38 <MC1984> they dont use free tx
 72 2013-02-02 01:41:40 <Luke-Jr> LevelDB is working nicely I think
 73 2013-02-02 01:41:53 <Luke-Jr> MC1984: 0.0005 BTC is hardly a fee
 74 2013-02-02 01:41:53 <moore> pl
 75 2013-02-02 01:41:55 <moore> cool
 76 2013-02-02 01:41:58 <gmaxwell> at least we can incentivize not bloating the utxo set a bit.. though I don't see how we can incentivize not using 2 transactions when 0.01 transactions would suffie.
 77 2013-02-02 01:42:02 <Luke-Jr> MC1984: the effective cost of Dice transactions are huge
 78 2013-02-02 01:42:03 <HM> i mean anything you do to hinder SD is equally malicious
 79 2013-02-02 01:42:14 <Luke-Jr> HM: more like self defence
 80 2013-02-02 01:42:23 <HM> how many people play SD though?
 81 2013-02-02 01:42:27 <moore> I have been thinking of writing my self a fractal tree index which might be more effechent then the aproach that LevelDB uses
 82 2013-02-02 01:42:31 <HM> do those users feel like you're defending them ? :S
 83 2013-02-02 01:42:34 <gmaxwell> HM: At least their response to problems has been 'you fix it'. ::shrugs::
 84 2013-02-02 01:42:35 <Luke-Jr> HM: far fewer than they want you to think
 85 2013-02-02 01:42:55 <MC1984> youd hope theyd be rational enough to ensure the long tern survival of the netowrk by not being dicks
 86 2013-02-02 01:43:02 <moore> locality wise more efficient
 87 2013-02-02 01:43:14 <HM> moore: 'fractal trees', if they're what i think you mean, are a patent minefield
 88 2013-02-02 01:43:14 <MC1984> thats what i mean about bitcoin assuming a level of rationality that might not exist
 89 2013-02-02 01:43:37 <Luke-Jr> MC1984: their discussions on facebook make me think they might actually not be sane
 90 2013-02-02 01:43:39 <moore> HM, I don't really look at patents much
 91 2013-02-02 01:43:48 <moore> but might matter for bitcoin
 92 2013-02-02 01:44:04 <amiller> moore by fractal tree do you basically mean any of the cache oblivious data structure
 93 2013-02-02 01:44:04 <MC1984> of course that doesnt mean shit if theyre just trying to milk it dry then move onto the next scam somewhere
 94 2013-02-02 01:44:15 <Luke-Jr> MC1984: as in, they look at Bitcoin as a tool to overthrow governments and bring about anarchy worldwide???
 95 2013-02-02 01:44:15 <moore> amiller, ya
 96 2013-02-02 01:44:29 <gmaxwell> MC1984: their their public stock is worth over 600kBTC now at market. The stock the owner owns is worth around 5.4 million BTC at market now.  ... :P short term gains seem to encourage crapping up bitcoin's long term prospets.
 97 2013-02-02 01:44:40 <MC1984> Luke-Jr sounds like punter pleasing rhetoric to me
 98 2013-02-02 01:44:51 <HM> eh
 99 2013-02-02 01:44:59 <HM> how can it be worth 5 million BTC
100 2013-02-02 01:45:00 <moore> I actually have been thinking of a modification that should make lookups cheaper but merges cost more
101 2013-02-02 01:45:07 <HM> that's absurd
102 2013-02-02 01:45:24 <gmaxwell> People are stupid?
103 2013-02-02 01:45:37 <moore> I would do this by using the b-heap layout that varnish uses
104 2013-02-02 01:45:39 <MC1984> gmaxwell all i can say is that bitcoin has to absorb that shit, you cant change human nature
105 2013-02-02 01:45:40 <Luke-Jr> HM: bubble, maybe fraud
106 2013-02-02 01:45:56 <gmaxwell> They sold 10 percent of their shares. People traded the public shares up to a point where they have a market cap of 600kBTC.. ::shrugs::
107 2013-02-02 01:46:01 <MC1984> im surpised all the pools havent banned dice tx yet though
108 2013-02-02 01:46:04 <Luke-Jr> HM: by artificially inflating their transaction volume, they can probably be making investors thing they're a big hit
109 2013-02-02 01:46:24 <andytoshi> hey, anyone here in or around austin tx?
110 2013-02-02 01:46:27 <HM> investors are dumb then
111 2013-02-02 01:46:33 <HM> you could reimplement SD in a day
112 2013-02-02 01:46:36 <HM> maybe 2
113 2013-02-02 01:46:40 <Luke-Jr> HM: I could do it BETTER in a day
114 2013-02-02 01:46:48 <Luke-Jr> 
ACTION grumbles about stupid laws in the USA

115 2013-02-02 01:47:03 <HM> haha gambling sucks there
116 2013-02-02 01:47:07 <gmaxwell> Not just investors, but also other gamblers.. there is a lot of social proof to gambling. Gamblers are all looking for an 'edge' (and foolish enough to think one exists when one doesn't)... a lot of people using something is a sign there is an edge.
117 2013-02-02 01:47:08 <HM> move to antigua
118 2013-02-02 01:47:11 <HM> problem solved
119 2013-02-02 01:47:43 <gmaxwell> plus all the griping the technical people do about that site is free promotion.
120 2013-02-02 01:47:48 <Luke-Jr> I really hope someone implements a Bitcoin-friendly Dice soon
121 2013-02-02 01:47:54 <HM> gmaxwell: there *are* egdes in sports...particular if you're in to arbitrage
122 2013-02-02 01:47:55 <Luke-Jr> (somewhere it's legal to do so)
123 2013-02-02 01:48:03 <gmaxwell> There are plenty of transaction efficient gambling sites.
124 2013-02-02 01:48:08 <sipa> Luke-Jr: most likely that already exist(ed), and nobody every heard about it :)
125 2013-02-02 01:48:13 <gmaxwell> hm: sure, but not in that kind of thing.
126 2013-02-02 01:48:25 <Luke-Jr> gmaxwell: sure, I mean that have all the benfits of dice
127 2013-02-02 01:48:48 <HM> Dice....couldn't pick a more boring game
128 2013-02-02 01:48:52 <Luke-Jr> lol
129 2013-02-02 01:48:55 <HM> i mean at least emulate a lottery
130 2013-02-02 01:48:58 <gmaxwell> I mean, I like to point people to https://ragecoin.appspot.com/ as an example of very efficient gambling.
131 2013-02-02 01:49:07 <HM> have a fancy flash ball machine animation
132 2013-02-02 01:49:12 <gmaxwell> hm: thats kinda the point.. the addicts need nothing else..
133 2013-02-02 01:49:15 <HM> fancy music, girl in a red dress
134 2013-02-02 01:49:16 <HM> tada
135 2013-02-02 01:50:22 <gmaxwell> if you're the kind of person who will blow a bunch on repeated plays at 98% EV your only entertainment is the adreline rush and the dopamine spike... yet another 'drug' based service, I guess. :P
136 2013-02-02 01:50:23 <HM> gmaxwell: just hit the 500x there l
137 2013-02-02 01:50:55 <HM> haha
138 2013-02-02 01:51:56 <HM> i quite like poker
139 2013-02-02 01:52:25 <gmaxwell> Maybe I shold spend some times writing up why it's also irrational to play games of chance which have small >100% EVs... perhaps that will help more people understand that gambling is foolish.
140 2013-02-02 01:52:29 <HM> but games of pure chance, in the casino, eesh
141 2013-02-02 01:52:49 <HM> gmaxwell: the kelly criterion
142 2013-02-02 01:53:21 <gmaxwell> Right.
143 2013-02-02 01:53:21 <sipa> what worries is me, is that even people who perfectly understand the statistics behind it, seem to play it
144 2013-02-02 01:54:22 <MC1984> gmaxwell if you dont understand gambling
145 2013-02-02 01:54:38 <MC1984> look up people who shoot up krokodil in russia
146 2013-02-02 01:54:40 <gmaxwell> I am less concerned about that??? its a smaller set of people... and at least they know what they are doing. There absolutely are a large number of people who think they actually have an edge, and also many people who totally don't get that the house edge is _per game_ and it adds up.
147 2013-02-02 01:56:10 <HM> meh. i'd rather spend ??1/day on a lottery ticket than on a newspaper
148 2013-02-02 01:56:44 <HM> but  i'd prefer to buy neither ;)
149 2013-02-02 01:59:54 <Luke-Jr> does anyone understand what Diapolo is trying to tell me on the Proposals pull? can someone explain? >_<
150 2013-02-02 02:00:43 <MC1984> i bought a scratchcard once
151 2013-02-02 02:00:46 <MC1984> it cost ?2
152 2013-02-02 02:00:50 <MC1984> i won ?4
153 2013-02-02 02:00:55 <MC1984> i never bought another once again
154 2013-02-02 02:00:58 <HM> good story.
155 2013-02-02 02:00:59 <MC1984> thats how you do gambling
156 2013-02-02 02:01:16 <HM> Cripes does ragecoin actually use MD5 :|
157 2013-02-02 02:02:02 <gmaxwell> HM: harmless in its usage.. more amusing was that their nonces used to only be about 40 bits... so it would have been possible to make a table.
158 2013-02-02 02:02:50 <HM> it doesn't lookt hat long now
159 2013-02-02 02:03:06 <HM> md5(x,y,x ...are those the reel results?
160 2013-02-02 02:03:09 <gmaxwell> He fixed it after I convinced him it was insecure... but it actually took some work to convince him.
161 2013-02-02 02:03:31 <gmaxwell> By the end I was a bit irritated with myself that I didn't just exploit it. :P
162 2013-02-02 02:04:34 <HM> looks to be about 70 bits now
163 2013-02-02 02:04:53 <HM> plus whatever the reals are
164 2013-02-02 02:04:56 <HM> reels*
165 2013-02-02 02:05:49 <gmaxwell> yea, you only need to precompute for a small number of reels. E.g. every one of the highest return options.. then spin until the next will be one of those and play that round.
166 2013-02-02 02:05:58 <gmaxwell> since it costs nothing to spin without playing.
167 2013-02-02 02:06:55 <HM> it's not provably fair is it?
168 2013-02-02 02:10:11 <HM> he should put a timestamp in the session alongside a mac to ensure nobody has a feasible time in which to brute it
169 2013-02-02 02:13:54 <muhoo> wow, whatever changes you guys made in the latest HEAD of bitcoinj, block chain download speed increased by an insane amount, like 100x
170 2013-02-02 02:14:18 <sipa> muhoo: BIP37
171 2013-02-02 02:14:26 <sipa> i suppose
172 2013-02-02 02:15:19 <muhoo> i also note it is periodically attempting to redownload the chain, on a timer, looks like, even if it hasn't changed. haven't seen where in the code that could be (wag would be PeerGroup.java)
173 2013-02-02 02:17:23 <ralphtheninja> hey, I'm interested in helping out with developing bitcoind/bitcoin-qt, where do I start? :)
174 2013-02-02 02:17:34 <Luke-Jr> ralphtheninja: testing pullreqs
175 2013-02-02 02:17:46 <muhoo> yep, looks like the bloom filter code is in here now. fantastic!
176 2013-02-02 02:18:43 <ralphtheninja> Luke-Jr: check, can you point me to a FAQ or similar?
177 2013-02-02 02:19:02 <Luke-Jr> ralphtheninja: you know how to use git?
178 2013-02-02 02:19:14 <ralphtheninja> Luke-Jr: yeah :)
179 2013-02-02 02:19:40 <Luke-Jr> ralphtheninja: https://github.com/bitcoin/bitcoin/pulls
180 2013-02-02 02:19:47 <ralphtheninja> Luke-Jr: obviously there is a shitload of stuff that I don't understand, but there must be small stuff that I can start with
181 2013-02-02 02:19:58 <Luke-Jr> ralphtheninja: testing is the big bottleneck
182 2013-02-02 02:20:15 <Luke-Jr> ralphtheninja: if you really want to start writing code right away, unit test cases
183 2013-02-02 02:20:22 <ralphtheninja> Luke-Jr: gotcha, I'll poke around a bit and see if I can find something
184 2013-02-02 02:20:57 <HM> what's the bloom filter for?
185 2013-02-02 02:21:06 <HM> i noticed that in the codebase
186 2013-02-02 02:21:18 <sipa> HM: filtering transactions in blocks
187 2013-02-02 02:21:23 <sipa> HM: read BIP 37
188 2013-02-02 02:21:29 <muhoo> for spv
189 2013-02-02 02:21:48 <HM> k
190 2013-02-02 02:22:11 <sipa> it means SPV clients don't need to download all transactions
191 2013-02-02 02:25:06 <HM> hmm indeed
192 2013-02-02 02:25:34 <HM> what's the scenario where a peer doesn't feel like doing this work?
193 2013-02-02 02:25:38 <HM> just disconnect?
194 2013-02-02 02:25:51 <muhoo> light clients on slow networks
195 2013-02-02 02:25:55 <muhoo> i.e. android phone
196 2013-02-02 02:26:10 <muhoo> or a server running on a cheap-ass 1BTC/quarter VPS :-)
197 2013-02-02 02:26:21 <muhoo> with 500MB RAM
198 2013-02-02 02:26:59 <HM> no sorry, i meant what happens if someone requests a peer to filter and the peer says "forget it"
199 2013-02-02 02:27:16 <muhoo> i don't think the peer is filtering, the client is.
200 2013-02-02 02:27:18 <sipa> HM: if they don't implement it, they will ignore it
201 2013-02-02 02:27:24 <gmaxwell> 19:07 < HM> it's not provably fair is it?
202 2013-02-02 02:27:25 <sipa> muhoo: peers are peers :)
203 2013-02-02 02:27:28 <HM> right. so basically it's advisary
204 2013-02-02 02:27:49 <sipa> HM: a getdata for inv type 3 will be considered ignored
205 2013-02-02 02:27:55 <HM> the SPV client will have to detect peers not being polite and disconnect them
206 2013-02-02 02:27:57 <sipa> eh, non-existing
207 2013-02-02 02:27:59 <gmaxwell> No, I told him how to make it provably fair and he worked on it some and I think it was just more work than he wanted to do considering the traffic it gets (not much)
208 2013-02-02 02:28:26 <HM> gmaxwell: lol ok.
209 2013-02-02 02:28:45 <sipa> 
ACTION goes in stand-by mode

210 2013-02-02 02:29:48 <HM> sipa: you get to choose your own hash tweak?
211 2013-02-02 02:30:20 <HM> i guess that's fine in a bloom filter.
212 2013-02-02 02:34:26 <HM> 
ACTION rolls over and goes to sleep

213 2013-02-02 02:37:53 <bsdunx> Question is the block transaction history count in the UI actually an approximation of blocks remaining?
214 2013-02-02 02:41:14 <gmaxwell> bsdunx: if you're referring to the block count in the progress indicator, it's an approximation. Yes.
215 2013-02-02 02:42:40 <bsdunx> Okay, I'll put the tilde back in, working on fixing the pluralization for the en translation.
216 2013-02-02 04:14:25 <jgarzik> cool
217 2013-02-02 04:14:34 <jgarzik> 2 out of 3 exmulti machines are on http://blockchain.info/hub-nodes
218 2013-02-02 04:14:42 <jgarzik> I need to finish brd ASAP, and boost my stats
219 2013-02-02 04:14:52 <Diablo-D3> p2pool needs more blocks
220 2013-02-02 04:15:04 <MC1984> whats exmulti
221 2013-02-02 04:15:52 <jgarzik> MC1984: my one-man corp
222 2013-02-02 04:16:42 <MC1984> eh
223 2013-02-02 04:20:28 <MC1984> should p2pool be getting more blocks now because of jeffs thing
224 2013-02-02 04:23:14 <mappum> he's not mining it there
225 2013-02-02 04:24:49 <Luke-Jr> I love how PayPal tells you how to bypass the password prompt after you get it wrong twice.. -.-
226 2013-02-02 04:26:11 <mappum> oh god, and they hold all my money
227 2013-02-02 04:26:17 <mappum> that isnt btc
228 2013-02-02 04:28:41 <Luke-Jr> 
ACTION wonders why he has 22 compilers installed <.<

229 2013-02-02 04:48:11 <bsdunx> Diablo-D3: curious, are you a BH member?
230 2013-02-02 04:48:24 <Diablo-D3> a what?
231 2013-02-02 04:48:52 <bsdunx> guess not, =p .. blizzhackers aka edgeofnowhere
232 2013-02-02 04:49:34 <gmaxwell> bsdunx: now you get ranted at that his nick predates blizzard. in 3. 2. 1. ...
233 2013-02-02 04:50:04 <bsdunx> gotcha
234 2013-02-02 04:50:41 <bsdunx> I only asked since another member is around.
235 2013-02-02 05:05:04 <jgarzik> 
ACTION wants to switch mining targets, but is doing a how-long-until-restart test... which it is surviving

236 2013-02-02 05:05:19 <jgarzik> and configuration changes require restart, due to apparent software bug
237 2013-02-02 05:05:27 <SomeoneWeird> heh
238 2013-02-02 05:05:59 <jgarzik> 
ACTION again pondered solo mining

239 2013-02-02 05:06:23 <Diablo-D3> bsdunx: my nick predates blizzard entertainment, co.
240 2013-02-02 05:06:23 <jgarzik> but ignoring variance, there is also the risk that this chip was never tested at a super high difficulty
241 2013-02-02 05:06:27 <Diablo-D3> gmaxwell: damnit
242 2013-02-02 05:06:38 <jgarzik> it does survive vardiff, with current diff around 32
243 2013-02-02 05:06:55 <gmaxwell> jgarzik: so solo mine with lower difficulty results handed back and ignored.
244 2013-02-02 05:07:11 <jgarzik> I suppose I could hack a proxy to do that
245 2013-02-02 05:07:17 <gmaxwell> any poolserver software will do that.
246 2013-02-02 05:07:34 <jgarzik> pool server would give me LP, too
247 2013-02-02 05:07:38 <jgarzik> or the equivalent thereof
248 2013-02-02 05:08:33 <jgarzik> 
ACTION wonders if he could do his own p2pool... i.e. use p2pool software, but within my own private network-of-1

249 2013-02-02 05:08:42 <gmaxwell> Right. Some people solo using p2pool in an isolated mode... but since you were having issues with 10 second shares.
250 2013-02-02 05:08:47 <gmaxwell> Yep you certantly can.
251 2013-02-02 05:11:07 <jgarzik> at the next config change, I definitely want to go back to p2pool.  it seemed like "btcaddr+10" username plus forrestv's p2pool patch did the trick for the most part.  I think the new physical configuration/lab placement may fix some p2pool issues.
252 2013-02-02 05:12:29 <gmaxwell> Bummer you didn't get your first block on p2pool, would have been neat to have a bunch of people paid directly in the first asic mined block. :)
253 2013-02-02 05:13:36 <bsdunx> Would refactoring stuff out of main.cpp be welcomed?
254 2013-02-02 05:14:24 <andytoshi> i'd support it, fwiw
255 2013-02-02 05:14:27 <andytoshi> (exactly nothing)
256 2013-02-02 05:14:31 <jgarzik> bsdunx: refactoring is done slowly on an as-needed basis.  there is general agreement that refactoring is needed.
257 2013-02-02 05:14:50 <jgarzik> however, refactoring really should not be a first-project
258 2013-02-02 05:15:02 <jgarzik> but something tackled after messing around with the codebase for a while
259 2013-02-02 05:15:18 <jgarzik> too often it's a "refactor just so I can read the damn thing" effort ;p
260 2013-02-02 05:15:43 <bsdunx> getting rid of the globals?
261 2013-02-02 05:16:21 <jgarzik> bsdunx: sure, but again, slowly (or at least in many commits, not One Big Commit)
262 2013-02-02 05:16:37 <jgarzik> and it has to make sense... sometimes a global is simply easy and efficient
263 2013-02-02 05:16:49 <jgarzik> and the "clean" alternative winds up more complex.
264 2013-02-02 05:17:03 <jgarzik> as with anything, life is a zen balance of many factors, costs, benefits :)
265 2013-02-02 05:17:18 <andytoshi> support i submit a patch with translates all the code to lisp...
266 2013-02-02 05:17:21 <jgarzik> newbies should start with one global
267 2013-02-02 05:17:26 <jgarzik> and work their way up :)
268 2013-02-02 05:17:27 <Luke-Jr> f = lambda b = None, a = 0: a if b is None else lambda c = None: f(c, a + b)
269 2013-02-02 05:17:32 <Luke-Jr> ^ and people say Perl is hard to read
270 2013-02-02 05:17:49 <jgarzik> tiny cleanups are great for newbies (new to the project, I mean)
271 2013-02-02 05:17:56 <jgarzik> big sweeping changes, not so much, even if cosmetic
272 2013-02-02 05:18:29 <gmaxwell> Refactors are hard to audit. If we accepted them easily from new contributors they'd be the best way to slip in a security vulnerablity.
273 2013-02-02 05:18:31 <andytoshi> Luke-Jr: fibonacci?
274 2013-02-02 05:18:53 <Luke-Jr> andytoshi: f(1)(2)(3)() = 1 + 2 + 3
275 2013-02-02 05:19:05 <jgarzik> 
ACTION ponders crowdfunding

276 2013-02-02 05:19:23 <Luke-Jr> 
ACTION ponders why someone would use p2pool for soloing when Eloipool works :P

277 2013-02-02 05:19:30 <jgarzik> i.e. exMULTI wants X BTC to run M public nodes "for the public good"
278 2013-02-02 05:19:37 <andytoshi> 
ACTION keeps trying to trace that haskell

279 2013-02-02 05:19:40 <jgarzik> anybody can add BTC to the given transaction
280 2013-02-02 05:19:47 <gmaxwell> Luke-Jr: as mentioned: people can't figure how how to run it. :P
281 2013-02-02 05:19:49 <jgarzik> 
ACTION wonders how easy that is, in Bitcoin-Qt UI

282 2013-02-02 05:19:58 <Luke-Jr> gmaxwell: well, I did document the less obvious bits
283 2013-02-02 05:20:00 <gmaxwell> p2pool is easy, has reasonable instructions, very easy to mergemine with..
284 2013-02-02 05:20:08 <jgarzik> i.e. add coins to existing ANYONECANPAY transaction, from the UI
285 2013-02-02 05:20:09 <gmaxwell> Luke-Jr: oh, good.
286 2013-02-02 05:20:57 <gmaxwell> jgarzik: it's not hard to do them from the rawtxn api, but ANYONECANPAY is kinda kludgy since you can't add change.
287 2013-02-02 05:21:55 <andytoshi> guys, is there a good way to determine what the input to a txout should be?
288 2013-02-02 05:22:00 <andytoshi> it seems like bitcoind just has some templates
289 2013-02-02 05:22:19 <andytoshi> rather, what the input should be shaped like
290 2013-02-02 05:22:25 <jgarzik> hmmm
291 2013-02-02 05:22:33 <jgarzik> users could send X BTC to themselves
292 2013-02-02 05:22:41 <gmaxwell> jgarzik: beter perhaps to have a multipayer txn (like my 'I taint rich' thread) and a simple cgi that lets people join into one.
293 2013-02-02 05:22:48 <jgarzik> then add precisely X BTC to the ANYONECANPAY TX
294 2013-02-02 05:23:05 <andytoshi> i'd like to some sorta explain program for raw transactions
295 2013-02-02 05:23:05 <jgarzik> hmmm
296 2013-02-02 05:23:13 <gmaxwell> jgarzik: yep, thats how you have to use ANYONECANPAY.
297 2013-02-02 05:23:24 <gmaxwell> https://bitcointalk.org/index.php?topic=139581.0 ('I taint rich' thread)
298 2013-02-02 05:25:37 <gmaxwell> a simple cgi that let people provide a txid/vout they want to put up (and the amount if its not fixed).... and an email address to nag when its time to sign might work. at least if the number of participants is fairly modest (too many and you run into issues with people vanishing)
299 2013-02-02 05:26:29 <jgarzik> makes me want to write a simple bot, instead of a cgi
300 2013-02-02 05:26:51 <gmaxwell> annoyingly raw txn run into IRC message length limits.
301 2013-02-02 05:27:11 <jgarzik> SD-like
302 2013-02-02 05:27:21 <jgarzik> just get bitcoins over the blockchain.  return them, if level not reached.
303 2013-02-02 05:27:34 <jgarzik> (i.e. contract not funded)
304 2013-02-02 05:28:21 <gmaxwell> Also see: http://www.logarithmic.net/pfh/rspp  if you're considering crowdfunding type things.
305 2013-02-02 05:36:50 <jgarzik> 
ACTION wishes he could come up with some fantastic incentive that spreads bitcoin relay nodes throughout the world ;p

306 2013-02-02 05:37:22 <gmaxwell> enable pay-to-ip again and then send funds to randomly selected healthy nodes. :P
307 2013-02-02 05:38:28 <petertodd> include replacable unsigned-for outputs with sighash single and tell nodes to redirect the funds to themselves and get them to miners ASAP
308 2013-02-02 05:38:35 <petertodd> (nodes == miners in this case...)
309 2013-02-02 05:39:09 <petertodd> and then make these special tx's be dependent on tonnes of other ones confirming
310 2013-02-02 05:39:22 <jgarzik> if I had the cash, I'd rent servers from small shops scattered widely, just to run full nodes and nothing else
311 2013-02-02 05:39:29 <gmaxwell> petertodd: dunno why miners wouldn't replace those outputs.
312 2013-02-02 05:39:48 <petertodd> gmaxwell: that's why I said nodes == miners...
313 2013-02-02 05:40:03 <gmaxwell> ah, well you don't have to pay miners. The network pays miners.
314 2013-02-02 05:40:13 <gmaxwell> And you could just create high fee txn to do that.
315 2013-02-02 05:40:17 <petertodd> point is to incentivising getting the dependent tx's mined, so the special ones can be
316 2013-02-02 05:40:52 <petertodd> and incentivising in such a way that *non-supportng* miners have a reason to make this happen
317 2013-02-02 05:41:15 <petertodd> basically it's a bridge to better tx fee handling and tit-for-tat relay protocols
318 2013-02-02 05:43:36 <two_dollar_btc> btc going back to $2
319 2013-02-02 05:44:03 <gmaxwell> two_dollar_btc: you've found the wrong channel.
320 2013-02-02 05:44:29 <two_dollar_btc> sorry brothers.... sell off now.
321 2013-02-02 05:45:29 <gmaxwell> two_dollar_btc: troll elsewhere, no one here cares.
322 2013-02-02 05:46:05 <jgarzik> <Eleuthria> So the upgraded bitcoind should reduce stales a bit.  I could also do the other trick that a lot of other pools do...just send a coinbase-only block at LP so I don't have to wait for bitcoind.
323 2013-02-02 05:46:11 <jgarzik> 
ACTION never knew that trick

324 2013-02-02 05:46:52 <petertodd> gmaxwell: Did you see my message responding to your fidelity bank token idea?