--- Log opened Tue Nov 26 00:00:08 2013
00:00 < phantomcircuit> hmm
00:00 < phantomcircuit> true
01:48 < Ryan52> cfields: Oh, okay. Sorry about that, I should have provided a status update, I totally fell asleep last night trying to rest my eyes before doing so... I'm on my way out to play cards now, but I'll comment on your commit (or alternate preferred means of providing those details?) tonight.
01:49 < Ryan52> cfields: And no problem regarding missing my pong, please let me know if the results are urgent, and I can try to put a rush on it.
02:28 < cfields> Ryan52: nah, nothing urgent. I just don't want it to go stale
02:29 < warren> cfields: has anyone tested your new gitian targets?  sorry I'm  too swamped personally now.
02:30 < cfields> warren: not yet. I'm trying to sucker Ryan52 into it :)
02:30 < warren> We'll toss some coins to Ryan52 for doing it and improving it if necessary.
02:31 < cfields> great
02:32 < cfields> Ryan52: i suppose you're not able to trigger the osx db corruption?
02:32 < warren> cfields: I'm pushing builds to all the people who complained to ask for testing of your patch
02:33 < cfields> ok
02:33 < warren> litecoin too
02:33 < cfields> warren: it's a stab in the dark, but there's some logic to it
02:56 < Ryan52> cfields: yeah, understandable, I'll give it a try once I figure out gitian. WRT mac osx, I wish, but my client has been happily downloading blocks for days (perhaps starting with a bootstrap would have somehow been more effective for testing?).
02:56 < Ryan52> s/mac osx/mac osx corruption/
02:56 < warren> Ryan52: bootstrap.dat will be no different in testing as corruption people experience seems to be mostly after full sync
02:57 < Ryan52> warren: yeah, that was my assumption, I wasn't sure if it was valid. thanks for confirmation.
02:58 < Ryan52> I thought maybe throwing stuff at it while it is busy downloading/validating/etc might make reproducing faster, but it was a long shot.
02:59 < warren> nah, reproducers have been as simple as "clean shutdown" and start it again
03:13 < warren> cfields: so folks want qt5 with autotools, but there's no qt5 in macports yet?
03:14 < Emcy> anyone think its worth getting a topic for bitcoin put on usenet
03:15 < warren> no
03:15 < Emcy> :(
03:15 < wumpus> warren: demand for qt5 doesn't come only from mac
03:17 < warren> wumpus: I know
03:17 < wumpus> the newer ubuntus also come with qt5 by default
03:17 < warren> and lack db48?
03:18 < wumpus> yes
03:18 < warren> is anyone going to get rid of bdb?
03:18 < wumpus> as it is now, it looks like we're getting rid of the wallet before getting rid of bdb
03:19 < Emcy> only the wallet uses bdb
03:19 < wumpus> (well, at least to make wallet optional, and use nowalletmode by default)
03:19 < wumpus> yes Emcy
03:19 < Emcy> its simple, we kill the wallet and we kill the bdb lol
03:21 < wumpus> it wouldn't solve the problem, of course, as people that want to use the wallet are still stuck with bdb, but the wallet component will always need it for backwards compatibility anyhow :/
03:22 < wumpus> we can't just say 'hey, you can't use your old wallet.dat's anymore!'
03:23 < warren> cfields: do still plan on redoing the win32 gitian deps?
03:23 < warren> cfields: tarball instead of zip, version upgrades, etc.
03:24 < Emcy> look sipa will get around to it one day
03:24 < Emcy> theres in only one of him
03:24 < wumpus> I'm not sure Emcy, I think he lost interest in the wallet part as well, he's focused on improving the block handling now
03:25 < Emcy> rightly so
03:25 < wumpus> which is also a much higher priority
03:25 < wumpus> there are many wallets, but there is only one full node
03:25 < Emcy> i wouldnt like to see the wallet part fall into such disrepair that someone jsut says fuck it and comments it all out one day
03:26 < Emcy> i suppose thats a danger if no one works on it for years
03:26 < wumpus> the nice way would be to seperate it out into a different part
03:26 < Emcy> i was surprised to see gavins old headers first branch on the github the other day too like 2 years old, so it could happen
03:27 < wumpus> we just don't have enough interested developers
03:27 < Emcy> wumpus i wouldnt like to see a functioning reference client be fractured into parts
03:28 < Emcy> im sure it makes sense to nerds but thats how you turn the project into a plaything for just the tech elite
03:28 < wumpus> *everyone* with C++ skills could say any day "hey, let's improve the wallet" and improve the code and submit a pull
03:28 < Emcy> well there are like 150 people with commits on the github but most of those are one or 2
03:29 < gmaxwell> seems to be a lot more interest in reimplementing the basic underlying stuff.
03:29 < wumpus> but as it looks now, on the long term we need to focus the bitcoin projects on its core responsibilities
03:29 < wumpus> which is the P2P and block chain handling
03:29 < Emcy> but yes im surprised there hasnt been more interest from all the rest of the talented people out there
03:29 < wumpus> gmaxwell: yes, many people get stuck in the 'let's reimplement this to learn' part
03:30 < Emcy> with bitcoin being like a new frontier of computer sceince or whatever......thought that would attract the brainboxes. Perhaps even academia for more than shitty papers here and there
03:31 < midnightmagic> Emcy: Nobody wants to be the one that broke bitcoin.
03:31 < wumpus> Emcy: and splitting up the project doesn't have to mean anything changes for end users, we couuld still package a full node with wallet if there is demand for that, it will just consist of multiple parts internally
03:34 < wumpus> Emcy: well there is lots of focus on bitcoin as a currency or speculation vehicle, but almost none on the open source project
03:34 < Emcy> right
03:35 < Emcy> goes along with my suspicians that all the price wanking is hurting bitcoin the project in subtle ways
03:35 < wumpus> we're extremely good at attracting traders and gamblers though, but expecting them to learn to code between their adrenaline binges would be expecting too much :-)
03:35 < midnightmagic> :)
03:36 < Emcy> but there are many parts of the sytem that can only be engineered on in situ
03:36 < Emcy> because no one knows how the fuck it will behave otherwise
03:38 < Emcy> i wonder if some team could get an EU grant to do bitcoin stuff
03:38 < Emcy> theres a team that gets EU money to code a functioning 100% decentralised torrent client i think
03:39 < Emcy> tribler? I think thier solutions was MOAR DHT though
03:39 < gmaxwell> maybe there should be a gambling interface that shims into GCC.  "Will this line of code compile? Bet now!"
03:40 < wumpus> hehe, and a code editor that combines statements using a slot machine
03:41 < gmaxwell> spliting out the wallet is important for a lot of reasons. It is somewhat crazy that our private key handling wallet process is exposed to the internet. With the wallet seperated we could do a lot better sandboxing of all the processes.
03:41 < warren> gmaxwell: that might be the cause of the macos x corruption
03:41 < Emcy> and threading of things that should be threaded
03:41 < gmaxwell> oh some sandboxing thing?
03:42 < warren> gmaxwell: no, "Will this line of code compile? Bet now!"
03:42 < wumpus> or with speculation 'this line of code is now worth 3 mBTC, invest in it to make it worth more!'
03:42 < gmaxwell> ah!
03:42 < gmaxwell> lol
03:42 < Emcy> as long as you can still compile a binary with all the parts forming a functioning client, it would be ok
03:42 < Emcy> as long as it doesnt end up like PGP
03:42 < gmaxwell> and when someone shows up and wants to rewrite some big chunk of the wallet it would be much easier to accept ... or even offer in parallel.
03:42 < gmaxwell> right now its a pain to run more than one or two bitcoin daemons on a host.. but it would be nice to be able to try out a couple different wallet types.
03:43 < wumpus> right, that's the advantage of modularization
03:43 < gmaxwell> "oh, you want the ultra fancy mode with coin control? no problem run wallet-advanced-qt "
03:43 < wumpus> keep the consensus part in a locked-down repository, and allow freeer experimentation with the rest
03:45 < wumpus> and 'berkelydb wallet' and 'fancy new append-only wallet' could exist in parallel for a while
03:45 < gmaxwell> http://www.jocm.us/uploadfile/2013/1125/20131125103803901.pdf < 20 bits per second across 20 meters, using high pitched (e.g. near inaudiable) sound from a laptop.  So much for your airgapped wallet. :P
03:45 < wumpus> (I forgot "deterministic" somewhere in there)
03:48 < warren> I'm pushing Bitcoin and Litecoin mac builds to the people who complained about corruption.
03:48 < Emcy> noticed the trend against modularity or even customisation in end user software though?
03:48 < warren> we'll see ...
03:48 < wumpus> An Acoustical Multi-hop Keylogger
03:49 < wumpus> warren: hopefully
03:50 < Emcy> gmaxwell what was the mic?
03:50 < warren> gmaxwell: white noise jamming
03:50 < wumpus> wouldn't white noise be easy to filter out?
03:50 < gmaxwell> warren: with a sutiable design thats actually very hard.
03:50 < gmaxwell> Emcy: it's between laptops, they built a mesh network
03:51 < Emcy> thats frankly cool and how could we use this power for good instead of evil
03:51 < wumpus> I guess it could also work with webcams and screen patterns, at least if they're in each others line of sight
03:52 < gmaxwell> warren: it's possible to have modulation schemes that are very difficult to jam without knowing the right cryptographic keys. ... and jamming those is basically only effective via overpowering.
03:52 < warren> sounds about as monetizable as the patented qrcodes that contain audio recordings.  The MBA program tried to make us do a consulting project on saving that stupid company.  I dropped the class because I didn't appreciate the professor using his own crappy portfolio companies.
03:52 < Emcy> i could hear 21khz last time i checked with a signal generator though so im good :)
03:53 < Emcy> wumpus there was a casio watch in the 90s that used flashing bars on your monitor for data downlaoding
03:53 < gmaxwell> Emcy: yes I can too, _however_ your and my sensitivity to that frequency is very weak even though we can hear it.
03:54 < Emcy> and gmaxwell is right about jamming, im reminded of how GPS gets a signal through from orbit with some sort of notch frequency thing
03:54 < gmaxwell> Emcy: which means the computer can be moderately loud but still seem quiet to us.
03:54 < gmaxwell> Emcy: orbit is easy! it's stright up1
03:54 < gmaxwell> and gps birds have a fairly low orbits. :P
03:55 < Emcy> gmaxwell the sound coming from old tube displays frequently caused me physical pain
03:55 < Emcy> i lucked out being born in time for the age of flatscreens
03:56 < gmaxwell> Emcy: there is some crazy person a few blocks from where I live that has some kind of pest repeller or something that produces a ~20KHz tone. I plug my ears when I walk by.. it's super loud.
03:57 < warren> does it work?
03:57 < gmaxwell> it repels me, does that count?
03:58 < warren> that could mean it is successful, it just doesn't discriminate
03:59 < wumpus> so it keeps humans away as well, terrific
04:00 < Emcy> some shops here started putting those things above the door specifically to keep chav kids from hanging around the front drinking
04:01 < Emcy> obviously completely discriminatory and not just against chavs
04:01 < gmaxwell> non chavs can afford earplugs?
04:02 < wumpus> their reasoning is probably that the younger kids hearing is still more sensitive to those frequencies
04:02 < gmaxwell> it's generally true, though also women.
04:02 < gmaxwell> which might not be their goal
04:02 < Emcy> lol im not putting in earplugs just to go into your shitty overpriced spar
04:03 < Emcy> the ironinc thing is everything in these shops is overpriced except the cheap gutrot cider
04:03 < wumpus> here they use classical music for that, in some places
04:03 < Emcy> lol really
04:04 < gmaxwell> "Well, we still have bums at the door; but at least they're classy bums now"
04:05 < wumpus> hah
04:09 < warren> http://download1.rpmfusion.org/~warren/bitcoin-0.8.5-OMG4/macosx/
04:09 < warren> http://download1.rpmfusion.org/~warren/litecoin-0.8.5.2-rc6/macosx/
04:11 < Emcy> omg?
04:12 < warren> Emcy: OMG!
04:13 < warren> grr
04:13 < warren> I built the wrong branch
04:17 < deantrade> gmaxwell: thanks.  I'm guessing I won't bother you as much posting in here, but I'll still be able to toss my ideas out on more experimental things
04:20 < deantrade> Did you see my comment on trashing DiskBlockIndex.pnext and storing height->mainChainBlockHash key/value table instead?
04:20 < Emcy> just remember the first rule of wizards
04:20 < Emcy> you dont talk about wizards
04:23 < deantrade> I thought it was something more along the lines of "People believe invalid things for all sorts of motivations, such as fear, instead of believing by evidence/reason."
04:24 < Emcy> wheres that from
04:24 < deantrade> Sword of Truth series
04:50 < deantrade> On UTXO aging:  I was thinking of having a system like this: The money supply inflates at some constant % increase each year.  Say something like 1% per year.  Then in 70 years (when the money supply doubles) you do a reverse stock split, and anything that gets rounded to zero coin value is thrown out.
04:52 < deantrade> 2% per year: 35 years to double money supply.  3% per year: 23.5 years.  4% per year: 18 years.
04:54 < gmaxwell> "inflates" so what winner are you going to transfer the results of everyone in the economies labor to?
04:55 < gmaxwell> ... because if you simply scale the value of all existing coins, none will ever fall under your thresholding.
04:56 < deantrade> Lets say you had 1 satoshi.  If there was a reverse stock split where 2 satoshis were then stored as 1 satoshi, then 1/2 = 0.
04:57 < deantrade> Winner of the inflation would be the block solvers (via proof of work).
04:58 < gmaxwell> deantrade: right, so depending on how much coin was lost you might be transferring some huge percentage of the economy to miners— who sets that and keeps it sane?. this doesn't sound like a grand plan.
04:59 < deantrade> No, when you reverse stock split nothing goes to the miners
04:59 < jtimon> that's escheatment we have thought about it for freicoin (which has demurrage instead of inflation)
05:00 < jtimon> at some point some outputs will approach zero value
05:00 < jtimon> should we still allow to spend them?
05:01 < jtimon> the anti-dust stuff limits that a little bit but you still can't take them out of the utxo set
05:01 < jtimon> with custom assets the problem gets worse I guess
05:02 < go1111111> gmaxwell: the miners would bid up the mining difficulty until they were barely making a profit. a 2% per year inflation forever (once the existing rate would otherwise fall below 2%) would just enhance the strength of the network, assuming it didn't cause people to be less enthusiastic about the currency
05:02 < deantrade> I don't think you guys are getting what I'm trying to say
05:02 < jtimon> I don't know, the incentives are complex, but didn't users already paid the transaction fees?
05:03 < jtimon> deantrade I really think I do, just replacing inflation with demurrage
05:04 < jtimon> in freicoin, for example, there will always be 5% (aprox) demurrage going to miners
05:04 < jtimon> in this case you don't need to "double the minimum expressable quantity"
05:04 < jtimon> because some outputs nominal values will go under 1 satoshi through demurrage
05:06 < gmaxwell> go1111111: yea, or cause starvation because 40% of the econonmy is going to build spheres capture all the sun's energy to power miners.
05:06 < deantrade> Yes, it is effectively demurrage.  I guess the problem with this system is that its not really a great thing that the miners get paid via inflation rather than via tx fees.
05:08 < jtimon> gmaxwell maybe 5% is "too much security", we've discussed that a lot on the freicoin forums
05:08 < jtimon> that ultimately depends on monetary velocity
05:09 < gmaxwell> jtimon: it depends also on prior lost coins and a bunch of other factors.
05:09 < jtimon> with V=10, 5% of the nominal supply is less than 49% with V=1
05:09 < jtimon> with demurrage lost coins are eventually recycled (even without escheatment)
05:10 < jtimon> in any case, many freicoiners believe that we need another p2p distribution mechanism besides mining
05:10 < deantrade> I guess maybe the best solution would be that unspent transactions would just be thrown out after X number of years.  And then coins would just need to be made more divisible if that became a problem.
05:11 < deantrade> So when you throw out coins, it effectively reduces the money supply, causing people to realize they now own a larger fraction of the total money supply.
05:12 < jtimon> I've heard that proposal before "escheatment if funds aren't moved in X blocks"
05:13 < jtimon> that would solve the "utxo size problem" too
05:14 < jtimon> but I'm not sure something like that is even necessary
05:14 < deantrade> But its not that the coins go to someone in particular who gets richer.  It just reduces the money supply, making everyone realize they are now effectively some fraction "richer".
05:15 < jtimon> as explained by go1111111 nobody gets necessarily richer
05:15 < jtimon> if you give coins to miners you subsidize security, that's all
05:16 < jtimon> in prefect competition profits still tend to zero
05:16 < jtimon> but your proposal is polemic
05:17 < gmaxwell> jtimon: well not quite because there is no perfect competition, so everyone with friction along that path are taking their tax.
05:17 < deantrade> Well, if the coins weren't spent for 100 years, then the market probably already adjusted to the lower effective money supply, then like if the original miners who forgot/lost their private keys all get their coins thrown out, people will then know for sure the money supply actually is smaller.
05:18 < jtimon> think about paper wallets, physical representations of bitcoin...
05:18 < jtimon> gmaxwell there is perfect competition in theory
05:19 < jtimon> and bitcoin's "demand for security" is extremely elastic
05:20 < gmaxwell> security is basically a perfect lemon market. You only need any at all except in hindsight.
05:20 < deantrade> jtimon: on that note, I was thinking that eventually people will make altcoins with all sorts of different fixed inflation rates (fixed per ledger), and then let the market decide which inflation rate they want to use.
05:22 < deantrade> I wish bitcoin didn't have such drastic changes in block reward...  50 25 12.5...  its a big deal when transaction fees are significantly less than inflation block reward
05:23 < deantrade> I mean to say, it shoulda been made more continual, no?
05:23 < gmaxwell> seemed to work out okay in practice.
05:24 < gmaxwell> piecewise constant has certian planning and accounting advantages.
05:24 < deantrade> In practice it didn't really matter too much to the miners.  But when the next transition hits, miners will have tro do lots of planning yea on what kind of hardware they want to buy and run.
05:24 < jtimon> in freicoin it decreases linearly
05:25 < jtimon> gmaxwell some have said that the first reward halving caused the following "bubble"
05:25 < deantrade> In just one block the reward for mining is going to half when it had been the same for 4 years, that is going to have a big effect on network hash rate when it happens
05:25 < jtimon> deantrade not necessarily, it can also affect prices, or both or a combination
05:26 < gmaxwell> jtimon: the following bubble was pretty long after (three months?)
05:26 < gmaxwell> jtimon: if so, uh. well I am not complaining.
05:26 < deantrade> jtimon: No, I don't think so.  Bitcoin is valuable because it is better than other currencies/money/banking systems.
05:26 < deantrade> Maybe bitcoin's halving just brought lots of media attention and more confidence to the system because it was maturing.
05:27 < jtimon> I think it was Impaler who speculated that that was the time it took for the markets to "feel the lack of new bitcoins coming"
05:27 < jtimon> according to him, miners speculated as much as they could but then they had to sell some part to pay the bills
05:28 < jtimon> I think liear would have been better but I don't think it is a big deal really
05:29 < deantrade> Linear?  I'm not sure what you mean.  Do you mean a more continual reward reduction rather than one step every 4 years?
05:29 < gmaxwell> jtimon: I'm skeptical, market volume was a pretty big multiple of the newly mined coins by then (oddly it seems lower now) but I guess its unknowable.
05:30 < jtimon> deantrade yes, in frc it is reduced every block until it is not reduced anymore
05:30 < gmaxwell> the biggest argument against the half operation that I have is that it creates a pretty big incentive to orphan the last block!
05:31 < gmaxwell> but arguably a continuious formula makes for a much smaller incentive to do that constantly instead of only a couple times in the system's life.
05:31 < jtimon> gmaxwell yeah I don't know, Impaler or galambo (I think was Impaler) made some numbers I think, but I agree is probably unknowable
05:32 < jtimon> never thought about it that way
05:32 < gmaxwell> jtimon: it's also hard to sort out because we actually changed who was mining at that time.
05:33 < gmaxwell> When the 50->25 change happened I was watching eagerly to see if we'd get stuck warring for the last 50 btc block. :P
05:33 < jtimon> I have no idea, but it was an interesting hypothesis
05:33 < gmaxwell> certantly we had miners which were large enough to where doing so would have been rational.
05:33 < deantrade> I was just looking at the FAQ on freicoin.  I disagree with a lot of what the author has to say, his philosophy.  It flies in the face of Austrian Economics.
05:33 < jtimon> yeah, we have to rewrite those faqs to somthing more neutral
05:34 < jtimon> r000n wrote those faqs
05:34 < deantrade> For example: "But money is created by the government, isn't it?"  You say the government doesn't make the money, but that's not quite right.
05:34 < jtimon> I wrote ones before but then they were assimilated into the about page...
05:34 < deantrade> The Federal Government's Military and Citizen Police enforce the US's monopoly on money in the US and in international trade
05:34 < jtimon> it's not very well expressed
05:35 < jtimon> but comercial banks create most of the money, even if the state enforced that privilege
05:35 < deantrade> In exchange, the Federal Reserve prints them lots of money for thier protection racket.
05:35 < deantrade> Yea, I agree, the commercial banks also with their FDIC default protection get to print lots of money for themselves too
05:36 < jtimon> the treasury could print the money directly without needing to "exchange" anything with the fed
05:36 < deantrade> Yea but that would be less confusing, and they like to keep the sheeple confused
05:36 < jtimon> that's what "greenbackers", positive money and other monetary reformist propose
05:37 < jtimon> what backs paper money is the state and its promise to tax you on that currency
05:37 < deantrade> Anyways, yea the government is the enforcer of the monopoly money, the gov steals from gold backed private banks (NORFED/egold/1933)
05:37 < jtimon> not anything in the feds balance sheet
05:38 < deantrade> No, what backs paper money is that using paper money and banks increases our productivity via productivity gains in specialization and trade
05:38 < jtimon> that's what back all money, but yes, true
05:38 < deantrade> Its just that there is a monopoly enforcement on USD, so we have to use USD to get those productivity gains
05:39 < jtimon> what i mean is that state money (like any other money) doesn't need any backing
05:40 < jtimon> and the goverment could take all the seignoriage for itself instead of giving it to the banking cartel
05:40 < deantrade> I agree, only for money to have reliable limited supply and for it to be easily/most efficent in trading is what makes money valuable as money
05:40 < jtimon> it doesn't even need to impose a monopoly
05:41 < deantrade> Hm, but the banking cartel is kind of like the smart people, and the government is just pandering politicans who do what the cartel wants.
05:42 < jtimon> yeah, the politicians don't rule
05:43 < deantrade> Freicoin says that the underlying cause of the boom/bust cycle is the entrenchment of the financial elite...  so it then concludes that for people to be able to own durable valubable things for a long time is bad.
05:43 < deantrade> That is invalid.
05:43 < deantrade> The boom/bust cycle is caused by monopoly money enforcement + money supply manipulation.
05:45 < jtimon> no, what causes monetary cycles is nominally everlasting money's incapability of producing zero interest rates when real capital yields naturally drop that low
05:45 < jtimon> keynes didn't solved the problem, but the problem is older than him
05:45 < jtimon> there was monetary cycles with gold
05:46 < jtimon> we really need to correct the fact, thank you for pointing that out
05:47 < deantrade> "There was monetary cycles with gold"-> not so much when there were private banks, there were local and chain defaults, and booms from bankers increasing their reserve ratios... but nothing like what the Federal Reserve can do.
05:47 < jtimon> probably you learn more about free-money by reading directly from Gesell
05:48 < jtimon> well, I'm not historian
05:49 < jtimon> but when do you say monetary cicles started?
05:50 < jtimon> Gesell, predicted hyperiflation as the unoavoidable end of keynes-like schemes, yet was strongly against gold and blamed it for cycles
05:50 < deantrade> "money's incapability of producing zero interest rates when real capital yields drop that low"-> uh... in the free market... every durable good has an interest rate that directly corresponds to how much value over time it brings to the market owners as demand and people's strength of desire to own something now rather than later.
05:51 < deantrade> Monetary cycles start when banks loan out at higher rates then they can afford to stay in business without defaulting.
05:51 < deantrade> When banks loan more out (higher reserve ratios) (lower interest rates)
05:52 < jtimon> deantrade so called "time preference" theory of interest is based on the fallacy that everybody prefers things in the present over things in the future
05:52 < jtimon> just because everybody prefers dollars and gold in the present than in the future
05:53 < deantrade> If people don't care when they have something then interest rates go lower.  That doesn't make it invalid/fallacy, you are just confirming what I am saying.
05:53 < jtimon> interest rates, like any other price, depends on supply and demand
05:54 < deantrade> But if people want things more right away then interest rates go up.
05:54 < jtimon> capital yields are profits, and depend on competition, not in the intrinsic properties of the real capital
05:54 < jtimon> the more factories there are, the less each one of them yields
05:54 < deantrade> Agreed on last 2 statements.
05:55 < jtimon> and if people prefer things in the future they go negative? that can't happen with gold, usd or btc
05:55 < jtimon> money DOES HAVE and effect on people's time preference, more than the other way around
05:55 < deantrade> It can happen.  You are assuming that the market purchasing power of one thing verses another stays constent
05:56 < jtimon> no, I'm not assuming that
05:56 < jtimon> let me rephrase
05:56 < jtimon> the more factories that produce a given good, the less each one of them yields
05:57 < jtimon> sorry, I'll come back in a minute
05:58 < deantrade> The more of something you have, the less market value each additional thing will have.
06:01 < deantrade> Q: Why does Freicoin use demurrage? I'm worried that my coins will just fade away.
06:01 < deantrade> A: Worried? It's a good thing. If the amount of money  was stable, people would prefer saving it as opposed to spending it.  This would decrease the quantity of money circulated, which in turn  would act against the main purpose of money - a medium of exchange. With  demurrage in place, you should think about money as it's meant to be,  not as a storage medium of wealth.
06:03 < jtimon> not a verygood answer, I agree
06:03 < deantrade> Invalid...  even in bitcoin, even if the value goes up over time, if a person has lots of it, he may still want to trade some of them in exchange for other things he wants more.  Just because money is becoming more valuable over time, this has no effect on whether it is suitable for trade.  Suitability for day to day trading is more just about security/transaction fee/speed of transaction.
06:03 < jtimon> but let me explain you Gesell's theory on interest
06:04 < jtimon> you agree that capital accumulation leads to lower real capital yields, no?
06:04 < jtimon> that in turn leads to lower prices of consuming goods
06:04 < deantrade> Only if the capital is just a copy of the previous capital.  But if the capital is an improvement over existing, then no.
06:05 < jtimon> the owners of the "old" capital have to compete with the owners of the "improved capital"
06:05 < deantrade> But I would agree that goods can generally be made cheeper if more capital/durable goods exist and are suitable for the particular goods to be produced efficiently.
06:06 < jtimon> exactly, capital yields are profits, and as such they should drop with competition
06:06 < jtimon> innovation can drive profits higher, but only temporarely
06:07 < jtimon> unless of course the state limits capital production somehow, creating rents
06:07 < jtimon> rents are profits that are somehow protected from competition
06:08 < jtimon> when capital yields are low
06:08 < deantrade> I wouldn't use that as the definition of "rent" normally, but for this conversation I will follow with your definition
06:08 < jtimon> ok
06:09 < deantrade> Err I would rather say "creating higher rent than a free market would have"
06:09 < jtimon>  to be more specific
06:09 < jtimon> http://en.wikipedia.org/wiki/Economic_rent
06:10 < jtimon> so what causes monetary cycles according to Gesell
06:10 < jtimon> ?
06:10 < jtimon> 1) competition drives real capital yields low
06:11 < jtimon> 2) savers start preferring to just hoard their money instead of lending at low rates
06:12 < jtimon> 3) that reduces monetary velocity, which causes price deflation, which further encourages hoarding vs lending (positive feedback loop, notsustainable for long)
06:12 < jtimon> 4) after enough capital destruction (maybe just by lack of mainteniance)
06:13 < jtimon> yields go up again and investment resumes, stopping capital and employment destruction
06:13 < jtimon> how this problem has tried to be solved?
06:14 < deantrade> I'm not sure where the problem is.  You say there is "capital and emplyment destruction"?  How?
06:14 < deantrade> How is someone hoarding gold/money bad for anyone?
06:15 < jtimon> keynesians have confused the real problem (basic interest/liquidity premium) with the symptom (deflation)
06:15 < jtimon> lack of investment causes capital destruction and therefore employment destruction
06:15 < deantrade> If Bill Gates made a billion dollars by creating everyone computers and trading for dollars, and then burried/permanently destroyed the dollars, then everyone else but Bill Gates would have made out well.
06:16 < jtimon> when you close a factory it will start to deteriorate
06:16 < jtimon> price delfation discourages investment (assuming monetary monopoly)
06:18 < deantrade> "Price deflation discourages investment" -> only when investments are not worthwhile despite the increasing purchase power of the entity with the savings.
06:18 < jtimon> no, always
06:19 < jtimon> think of any investment example
06:19 < jtimon> let's say a factory that will yield 5% of its nominal value
06:20 < jtimon> will you invest in that factory?
06:20 < jtimon> a) with 0% price inflation
06:20 < jtimon> b with 10% price deflation
06:21 < jtimon> with 10% price deflation you're better of keeping the "abstract wealth" money represents rather than investing in real wealth
06:21 < jtimon> keynesians have tried to "solve deflation" by replacing real savers with newly created money
06:21 < deantrade> Ok, agreed, but I'm still not seeing what the problem is.
06:22 < jtimon> the problem is capital yields should naturally tend to zero, but that's impossible with nominally everlasting money
06:22 < deantrade> If there are people who have needs, and they want to work for their goal attainments, then they will do that.  Whats to stop them?
06:23 < deantrade> "he problem is capital yields should naturally tend to zero" Should?  Do they?  And if they do, then how is this a problem?
06:23 < jtimon> the worker needs capital just like capital needs workers to operate it
06:23 < deantrade> Worker works, makes wage, then has capital.
06:23 < jtimon> money is not real capital, money is just a symbol of value
06:24 < deantrade> Money is real capital.
06:24 < deantrade> Money has market value.  Money is durable.
06:24 < jtimon> what can you build with money (without exchanging it for another thing)?
06:24 < jtimon> some monies are durable
06:25 < jtimon> beef has market value too, but it's not capital
06:25 < deantrade> I can build a pile of coins to swim in like scrooge mcduck
06:25 < jtimon> it's a consuming good
06:25 < jtimon> money is not a cosnuming good neither
06:25 < jtimon> you can build a swiming pool with some monies
06:26 < jtimon> you can't swim on bitcoin can you?
06:26 < deantrade> ok, then I don't really care what you use as a definition of "Capital"...  and your statment "money is not real capital, money is just a symbol of value" doesn't really mean anything to me.
06:26 < warren> this is getting absurd
06:26 < warren> NMC is at $5
06:26 < deantrade> I can put bitcoin on flash drives, and swim in flash drives.
06:27 < jtimon> that's what makes money capital, really?
06:27 < jtimon> by real capital I just mean producing goods
06:27 < jtimon> money is not a producing good not a consuming good
06:27 < jtimon> nor
06:28 < jtimon> it's just an implicit agreement between their users that facilitates trade
06:28 < deantrade> Here's what I'd use instead of your word "capital": resource.  Resource for attaining goals of individual humans.  Each human has a set of resources which are under his control.  He is continually "using" them as they are under his control.
06:28 < jtimon> but my concept is far more specific
06:28 < jtimon> consuming goods and raw materials are resources too
06:29 < deantrade> By him "using" them in whatever way he desires, it helps him attain his goals.
06:29 < jtimon> work is another resource
06:29 < deantrade> Work is not a resource.
06:29 < jtimon> how not?
06:29 < deantrade> Having an employee who can do work is a potential resource.  But the work is not a resource, only the product of the work is a resource.
06:30 < jtimon> ok, then your employees are resources, whatever
06:30 < deantrade> Owning a house for example, living in the house, continually brings market value to the owner.
06:30 < jtimon> my point is that "producing goods" are a very concrete type of resource
06:31 < jtimon> a house is real capital
06:31 < jtimon> you can rent it'
06:31 < deantrade> Uh, you can rent any object really.
06:32 < jtimon> well, living in a house is a consuming good, produced by the house, the producing good
06:33 < jtimon> and houses have a capital yield that comes from that
06:33 < jtimon> if interest rates are at 5%
06:34 < deantrade> And the money in your pocket is "producing" one value by giving you confidence/ability that you will be able to use it to trade at a future time for something else.
06:36 < deantrade> But lets say you can't think of anything you want in the future.  You just have needs now to fulfill.  Then you'd trade the money now, no matter the interest rate.
06:38 < deantrade> A well I donno why I was arguing with him.
06:39 < jtimon> sorry, deantrade, electricity went down
06:40 < deantrade> Noproblemo
06:41 < jtimon> what's the last thing you received?
06:42 < deantrade> If savers hoard, then the market value of the money goes up.  If they stop increasing their reserves, then the market value of money stabalizes.
06:43 < deantrade> I guess I just don't see what the problem is with savers saving.
06:43 < jtimon> the problem is not with savers saving but with savers hoarding
06:44 < deantrade> k, "hoarding".  What is wrong with "hoarding"?
06:44 < jtimon> if they lend or reinvest their savings in something real like they do when interests are high there's no problem
06:45 < jtimon> you're influencing the price of money, and you're getting a free insurance against uncertainty
06:45 < jtimon> that's a positive externality inherent in "durable money" and it is paid for by others
06:46 < jtimon> what neo-keynesians do is replace savers with newly printed money
06:46 < jtimon> manipulating interest rates
06:46 < deantrade> If you have money as savings, then that means you already did work to create something real in order to attain the money.  Then if you go to spend the money, you don't create, you only exchange.
06:46 < jtimon> but that's not sustainable
06:47 < jtimon> exactly, you deserve to receive something in exchange for whatever you previously provided to society
06:47 < deantrade> But for as long as you just hold the money, its like you just did all of that work in exchange for nothing, so the rest of society benefited at your expense
06:47 < jtimon> but why society must allow you to think what you want in exchange for as long as you want with no cost?
06:48 < deantrade> There is no gauruntee the money will have the same market purchasing power in the future.
06:48 < deantrade> There is no one forcing anyone to accept some amount of money for anything, its free trade
06:49 < jtimon> yeah, if many savers hoard, it will have an even greater market value in real terms
06:49 < jtimon> I'm assuming monetary monopoly all along
06:49 < jtimon> for example, a gold standard
06:49 < deantrade> Unless a new form of money is created that has better features, then the old money becomes worthless
06:49 < jtimon> there's some force here
06:50 < deantrade> Monetary monopoly: money monopolies do not last either.  They have lasted long time durations, but not forever.
06:50 < jtimon> with a free monetary market edflation is not that harmful because trade and investment can just occur in other currencies
06:51 < jtimon> with a free monetary market, let's say real capital yields drop to 1%
06:51 < deantrade> I agree, people can just chose to invest in whatever they want.  It would just be fraudulent to create a currency where you say it will have one inflation plan, and then later to do some different plan.
06:51 < jtimon> savers don't lend or invest bitcoins anymore
06:52 < jtimon> it doesn't matter, other savers will be happy to lend their frc at 0%interest
06:52 < deantrade> How was the gold standard forced?  Or do you mean in our current situation there is force?
06:52 < jtimon> in our current situation there is force, yes
06:53 < jtimon> and in the gold standard was the same monpoly
06:53 < jtimon> the legal tender was 1 gold mark or whatever
06:53 < deantrade> Savers only options right now is [US Tresuries, Stocks, or Land], gold, bitcoins, what else?  (In brackets = in a bubble)
06:53 < jtimon> dependin on the country
06:54 < jtimon> real capital
06:54 < deantrade> What did that mean though "legal tender"?  At one time it just meant "You can only call it a dollar if it is this many ounces of gold".
06:54 < jtimon> stocks could be counted as real capital, but I agree they're probably still in bubble prices
06:55 < jtimon> the problem is when you can only trade using thalers, whatever the quantity of silver that defines them
06:55 < deantrade> In a free market where banks/money was not a monopoly, "banks" would not be protected from default (their owners would be held liable to pay up), and banks would offer higher interest rates to money market accounts
06:56 < jtimon> interest rates would not be manipulated
06:56 < deantrade> But in the world as it is now, banks just print money and lend out at way lower interest rates than savers would be willing to accept.
06:56 < deantrade> And then banks offer pretty much 0% interest rate to savers.
06:57 < deantrade> So savers are stuck having to invest in US Treasuries, stocks, land (and gold/bitcoins for the smart ones)
06:57 < jtimon> but I think that with enough mutual credit currencies (usually 0% interest) and demurrage currencies like freicoin interests would tend to zero in a free market
06:58 < jtimon> I agree the current situation sucks
06:58 < jtimon> I believe it will end up just as Gesell predicted: hyperinflation
06:58 < deantrade> This is also Austrian Economist's prediction.
06:58 < jtimon> https://www.community-exchange.org/docs/Gesell/en/neo/part3/13.htm
06:59 < jtimon> Gesell, studied bohem-bawerk, he has more to do with Menger than with Keynes
07:00 < jtimon> in fact, he's closer to Menger than Mises in certain senses, like rejecting the notion of so called "intrinsic value"
07:00 < deantrade> Factories and farms etc... they don't just exist and produce the same amount of products at the same efficiency no matter the owner.
07:00 < jtimon> a dogma very often widespread among "austrians"
07:00 < deantrade> I reject "intrinsic value".
07:01 < deantrade> Value is only relative to one who acts to attain goals.
07:01 < jtimon> but the markets forces the operators of the "unefficient capitals" to change hands
07:02 < jtimon> that's good, unfortunately many goldbugs (and even bitcoiners) don't think like you
07:02 < deantrade> Right... and poor people who prove to be capable of operating them, but don't have the capital to buy them at the moment will look for a loan.
07:02 < jtimon> yes
07:03 < deantrade> And there are many rich people who die, and their children blow the money on drugs etc.
07:03 < jtimon> yes
07:04 < jtimon> there's no need to redistribute wealth from rich to poor, but it's completely necessary to stop redistributing wealth from the poor to the rich
07:04 < deantrade> So productive people live and die.  And when a poor productive person sees that they could vastly improve their life by just loaining some amount of money at some interest rate, then they will take the offer.
07:04 < jtimon> the problem is that some monetary systems impede that interest rate to be zero
07:05 < jtimon> which would represent optimal prosperity: maximum capital accumulation for society
07:06 < jtimon> would be the best position possible for workers (comparatively with capital)
07:07 < jtimon> well, negative itnerest rates would be "unfair for capital" but they're not natural even with demurrage
07:08 < deantrade> Interest rates should simply be chosen by the market.  Interest rates are chosen by two people who come together with differing resources and contracts to deliver at a later time, and fully mutually voluntary acceptence of the contract.
07:08 < jtimon> interest rates are voluntary and determined by the market with freicoin too
07:09 < jtimon> nobody forces you to dodge the demurrage fee by lending or investing
07:09 < deantrade> I'm not disagreeing with that.  I'm disagreeing with the idea that somehow having a money supply that is decreasing is necessarily bad, particularly when that money is just one competing currency when there are many others to chose from.
07:10 < jtimon> I think it's bad only if it's the only money
07:11 < jtimon> I don't think bitcoin will hurt society with its deflation because it will never be monopoly money
07:11 < jtimon> it just won't be as useful to society as it could be if it had demurrage
07:12 < deantrade> Useful to attain what?
07:12 < jtimon> economic development and prosperity
07:12 < deantrade> Economic development and prosperity of which group of people?
07:14 < jtimon> for everyone that produces and consumes
07:14 < jtimon> the higher the itnerest rates, the more everyone pays for what he consumes
07:15 < deantrade> Not necessarily.
07:15 < jtimon> the higher the interest rates the lower the precentage of good prices come from worker wages
07:15 < jtimon> name a single consuming good that doesn't include interest in its final selling price
07:17 < jtimon> for some that % is as high as 50%
07:17 < deantrade> When you pay over time with high interest rate, if the money supply is increasing more rapidly than the interest rate, then later when you pay the interest you potentially have to exchange a lower market value than what the money was worth when you agreed to the deal.
07:18 < jtimon> that's why the inflation premium is a compenent of interest
07:19 < jtimon> sadly inflation indexes are usually manipulated nowdays
07:19 < deantrade> 2% CPI haha
07:20 < deantrade> Fred Monetary Base has been increasing at >30% per year for 5 years now (since 2008 housing financial crisis)
07:20 < jtimon> what I mean is that most people pay far more interests than they receive, even when they haven't borrowed any money
07:20 < wumpus> no matter the economic arguments for it, no one would have bought into bitcoin if it had demurrage; many people were already not taking it seriously for being "virtual", let alone if your holdings magically evaporate over time
07:21 < jtimon> wumpus yes, probably something like bitcoin was destined to be the first crypto
07:21 < wumpus> a future cryptocurrency could do it differently, but bitcoin had to be like this to work
07:21 < jtimon> there was a time when people believed that money couldn't be made of paper, now some people doubt that it can be made of bits
07:21 < wumpus> for example freicoin, had it not included the strange centralized contribution for every mined block
07:22 < jtimon> probably the first p2p currency had to have fully p2p distribution too, no matter how wasteful that is
07:23 < deantrade> Wasteful?
07:23 < jtimon> in terms of real resources, yes
07:23 < jtimon> it's subsidizing security
07:23 < deantrade> How is giving out practically worthless bitcoins (initially worthless) wasteful?
07:24 < wumpus> decentralized systems are by definition less efficient than centralized systems, but compensate for this with added robustness
07:24 < jtimon> no, mining like we're doing now is wasteful
07:24 < jtimon> wumpus, but when the 21 M are issued, fees should provide enough security
07:25 < deantrade> Mining is essential.  Prove of work that you earned the money.  You'd rather the bitcoins were handed out willy nilly like helocopter Ben?
07:25 < wumpus> I have decided for myself that I like the robustness more than the efficiency, but your opinion may vary
07:25 < jtimon> I prefer that they're are given to noprofits you freely decide to donate to like in freicoin, obviously
07:26 < jtimon> http://foundation.freicoin.org/
07:26 < jtimon> by the way, crypto-currencies related projects can be listed too even if they're not legally non-profits
07:27 < deantrade> You say "robustness", as if that doesn't also make it efficient.  Bitcoin is an extremely efficient value storage and value ownership transfer system.
07:27 < jtimon> bitcoin solves seignoriage by trying to destroy that value
07:28 < deantrade> For example, the Fed not being able to steal my bitcoins unlike they'd be able to steal my gold or my gold in a bank is really really awesome.
07:28 < jtimon> that's true for freicoin too
07:29 < jtimon> there's no need to give 100% of the initial supply to miners to have that
07:29 < jtimon> not to long ago gmaxwell was saying that 5% of the total supply anually for miners is wasteful
07:30 < jtimon> then the initial subsidies have to be much more wasteful
07:30 < jtimon> we need proof of work for security, totally agreed
07:30 < deantrade> "waste"?  wasting what?
07:31 < jtimon> but I don't think we need ti for issuance too
07:31 < deantrade> How else do issuance?  Have the group of developers that made it each get some number of coins?
07:31 < jtimon> electricity, conductors to build more asics than we need instead of general purpose computers
07:32 < jtimon> deantrade, it's not an easy issue
07:32 < jtimon> we've looked for ways in the freicoin forum and many have appeared
07:32 < jtimon> we hope to find a purely p2p one
07:33 < jtimon> for example, subsidizing scientific computations demonstrable with spark/scip
07:34 < deantrade> "electricity, conductors to build more asics than we need instead of general purpose computers" -> just like all of the effort used to mine and refine and make jewlrey out of gold.
07:34 < jtimon> isntead of getting money for random hashes, get it for submitting folding@home work units or something else
07:35 < deantrade> It takes effort to make bitcoins, they don't just come into existence willy nilly.
07:35 < jtimon> that's purely a design decision as demonstraded by freicoin
07:36 < jtimon> only 20% of the initial freicoins come to existence through mining
07:36 < jtimon> yet the freicoin network is secure and their miners get profits
07:37 < jtimon> miners don't create the value, users and merchants do
07:37 < deantrade> folding@home = centralized acceptor no?  I agree it would be cool if the work could be towards something like organic chemistry simulation, but I'm not really an expert in that field in order to really know if there is anything practical that we could do that would be hard to find a solution for but easy to verify work on.
07:39 < deantrade> Miners do create some value: we use the ledger made by the person who proves to do the most work, rather than just going with ledgers that are made by valueless lazy losers who just would want to double spend or spam
07:40 < jtimon> hard to find a solution for but easy to verify work -> that's what spark/scip is about
07:40 < jtimon> yes, miners provide security
07:41 < jtimon> I agree it's a complex problem
07:41 < jtimon> for now mining is the only 100% p2p distribution mechanism I know
07:42 < jtimon> all I'm saying is that if we had another one, issuance could be completely decoupled from mining
07:43 < jtimon> miners will live only on fees, and if that's not enough security then the system is not sustainable in the long run and the current subsidies are blinding us
07:44 < deantrade> Yea its a system that hasn't been tested yet.
07:45 < deantrade> Thinking about it while sleep deprived now...
07:51 < jtimon> think of ripple's xrp for example
07:51 < jtimon> their security mechanism is not pow
07:51 < jtimon> they could have distributed it through proof of work
07:52 < jtimon> but that wouldn't haven't made any economic sense
07:52 < jtimon> that doesn't mean the distribution mecanism they've chosen is cool
07:52 < deantrade> ripple is crap, not pow, its mob vote rule
07:53 < deantrade> good night!
07:53 < jtimon> Ryan Fugger's Ripple is a great concept
07:53 < jtimon> ripplelab's implementation is not good enough, I agree
07:53 < jtimon> ok, good night, I have to code
16:09 < warren> jgarzik: you had people at the office who could reproduce the mac corruption?
16:10 < warren> Please ask them to test the new builds?
16:10 < Ryan52> cfields: submitted my verification of sources in deterministic dmg builds, as a comment to your commit.
16:10 < cfields> Ryan52: great, thanks
16:11 < warren> Ryan52: URL?
16:11 < warren> cfields: zero reports of testing your memory barrier patch so far
16:11 < warren> from people who were able to make it fail
16:11 < cfields> warren: the more i read, the less convinced i am that it will do anything significant
16:11 < warren> cfields: oh?
16:11 < Ryan52> warren: https://github.com/theuni/bitcoin/commit/8a64fb98370ccc299d73111bbf97cdde23f681b1#commitcomment-4708754
16:12 < cfields> asm dumps seem to show that the asm memory hack is enough to get the compiler to avoid reordering
16:12 < warren> Ryan52: btw, for gitian you need to have an established GPG key and identify
16:12 < warren> identity*
16:14 < Ryan52> warren: Sadly, my 1024 bit key is not well connected and my 4096 key is not really connected at all. :(
16:14  * Ryan52 needs to get out more, or something.
16:15 < warren> Ryan52: I started a 8192 bit key only a few months ago
16:15 < warren> Ryan52: what state are you in now?
16:16 < TD> cfields: hmm i thought you said you checked that and saw reorderings across the barrier
16:16 < Ryan52> Gah, guess I didn't even get my 4096 key usable before it was obsolete, that is sad. I actually do have one more signature sitting in my INBOX from recently, though, when I get the chance to pull out the disk with the private key storage from my safe.
16:16 < cfields> TD: i do with some compilers, with some flags
16:16 < Ryan52> warren: Oregon
16:17  * Ryan52 travels to Washington frequently, too
16:17 < cfields> TD: i think the change is necessary, but i'm not sure that it will affect the way we currently build
16:17 < TD> ok
16:17 < TD> hmm too bad. i thought that might have been it, for a moment
16:18 < cfields> the asm diffs are pretty hard to read since it changes a bunch of stuff around. it's still possible they're necessary, but i'm not as confident as i was yesterday
16:44 < maaku> Ryan52: 4096 bits is obsolete?
16:45 < cfields> warren: have you heard of the corruption happening to anyone running a 64bit osx binary?
16:46 < sipa> yes, gavin is running a 64-bit build and has seen corruption
16:46 < maaku> cfields: yes, it's happened to me
16:46 < cfields> the current corruption? not the ones that have been fixed already?
16:46 < sipa> i have no idea what has been fixed
16:47 < sipa> i don't have any iThings
16:47 < maaku> it was either v0.8.3 or v0.8.5, but not trunk
16:47 < cfields> the fdatasync commits a while ago, i meant
16:47 < maaku> i don't know about gavin
16:47 < cfields> ok
16:47 < sipa> gavin has seen corruption on master a month ago or so
16:47 < maaku> gavinandresen: ^^
16:48 < phantomcircuit> Ryan52, 4096 bit keys are probably safe for a decade assuming no major quantum computing improvements
17:20 < amiller> hey does anyone here understand iddo's protocol er adam back's version of it
17:21 < amiller> adam3us, does this post work and not rely on currently disabled opcodes? https://bitcointalk.org/index.php?topic=277048.msg3210328#msg3210328
17:22 < amiller> i don't see how you can guarantee that you can add a + b, does addition in the script just do overflow with mod?
17:27 < amiller> so each number must be 4 bytes or else it fails
17:27 < amiller> and it's signed
17:28 < amiller> what you'd want to do is typecheck the b preimage of h(b)
17:28 < amiller> in other words you don'tw ant to allow  b to be a 5 byte number, becuase then b could claim his side but a wouldn't be able to take it anyway
17:28 < amiller> so it would time out
17:28 < gavinandresen> hmm?  I've had three instances of corruption over the last six months, all running at-the-time-git-HEAD, compiled 64-bit with clang3.3 (my development environment builds)
17:29 < amiller> but you can typecheck b by making b have to do an OP_ADD 0 to it, which guarantees that if B can spend it, A can spend it
17:33 < warren> cfields: the distributed bitcoin binaries are built with xcode 3.2.x on MacOS 10.6.x
17:34 < warren> cfields: gcc-4.2 based
17:34 < cfields> gavinandresen: ok, thanks
17:43 < warren> cfields: so would test results of my builds with your patch built by gcc-4.2 32bit be useful?
17:44 < cfields> warren: erm, what else would they be built with?
17:44 < warren> cfields: you're asking about 32bit vs 64bit and different compilers
17:45 < warren> cfields: these are built in the "standard" release way
17:45 < warren> gavinandresen: quick question
17:45 < cfields> warren: i saw an oddity in leveldb that would only be a problem for 32bit builds. If the error manifests in 64bit as well, there's no need to investigate it
17:47 < warren> gavinandresen: https://github.com/bitcoin/bitcoin/blob/master/doc/release-process.md   my Bitcoin builds using this documented process ending with "contrib/macdeploy/macdeployqtplus" works in creating .dmg's, but their plist is missing menu options.  Are you following exactly this documented process?
17:48 < gavinandresen> warren: yes, I copy and paste from that document when I do the builds
17:49 < gavinandresen> warren: … and that's really not a bitcoin-wizards type of question.
18:11 < adam3us> amiller: iddo wrote it up later in the same bct thread as https://bitcointalk.org/index.php?topic=277048.msg3220019#msg3220019 and there he used (A xor B) mod 2 == 1 as the test.  however xor is disabled.   similarly add is enabled but aborts if the input is > 4 bytes.  i guess thats a problem too because a hash of a 32 bit number is trivial to brute force.  cat is also disabled, and hash takes one input only
18:13 < amiller> adam3us, so what about what i suggested
18:13 < amiller> you can hash a 4 byte number
18:13 < amiller> ah but that's not very high entropy.
18:13 < adam3us> amiller: yes but then its brute forceable and cat is disabled etc yep
18:13 < amiller> you can basically enumerate the hashes of the 4 bytes
18:13 < amiller> so without cat you're screwd
18:13 < amiller> okay right.
18:21 < adam3us> amiller: maybe you could do something with p2sh - if that gives you a way to hash a random value and a 0 or 1 bool
18:28 < warren> cfields: perhaps its time to submit the osx cross gitian as a PR?  Mark it "DO NOT COMMIT" at first.  More visibility for review?
18:29 < warren> cfields: although please add the equivalent of https://github.com/bitcoin/bitcoin/pull/3191
18:30 < gmaxwell> Note that gavin has had corruption on a newer toolchain, but very rarely. So perhaps _yet another bug_
18:31 < warren> gmaxwell: with the memory barrier patch?
18:31 < gmaxwell> warren: I am relatively confident that the issue in question doesn't exist in a sufficiently new toolchain.
18:32 < cfields> warren: i got the mac icon fixed...
18:32 < cfields> is it possible that's a regression since .8 branch?
18:32 < warren> cfields: possible, my Bitcoin 0.8 branch has stuff from master
18:32 < Ryan52> maaku, phantomcircuit: no, I just didn't realize moving to 8192 bits was a thing so soon.
18:33 < Ryan52> perhaps obselete for new keys? or not even?
18:33 < warren> Ryan52: you need to edit the gnupg source and rebuild to be able to generate 8192 bit keys
18:34 < cfields> warren: http://pastebin.com/raw.php?i=DdNtY5ia
18:34 < sipa> what is moving to 8192 bits?
18:34 < Ryan52> warren: oh, wow, so you are just super future-proofed.
18:34 < Ryan52> sipa: gpg keys
18:34 < cfields> yea, that's a regression from the qt5 commit
18:34 < sipa> why not to ECC? :(
18:35 < warren> Ryan52: not really.  xkcd 538 is much easier.
18:35 < sipa> but with 8192 bit RSA, even xkcd 538 will be obsolete!
18:36 < warren> The cost of xkcd 538 cracking is constant at any bit length.
18:36 < warren> O(1)
18:36 < Ryan52> heh
--- Log closed Wed Nov 27 00:00:07 2013