--- Log opened Mon Sep 09 00:00:20 2013
00:01 < gmaxwell> HM: depends on what you mean by SPV.. also you.
00:01 < gmaxwell> keep in mind that I have an easy out here: I can just forget about bitcoin.
00:01 < gmaxwell> (which I will likely do at some point, in fact— if I am to make a prediction)
00:02 < HM> Bummer
00:02 < gmaxwell> HM: if you (being the generalized representative of all man kind) are unwilling to take any cost at all to increase the collective security, then I don't think an improvement is possible.
00:03 < gmaxwell> If you're willing to take some small cost, which happens to currently be less than running a full node, then perhaps there are some things that can be done... but it's not clear to me that anyone will do them: too easy to just walk away from bitcoin.
00:03 < gmaxwell> Worse: if someone were to do such a thing they'd be personally better off (failing the CI) to go do it in an altcoin where they could go own a bunch of it upfront.
00:04 < HM> That's the thing though. Individual disregard for security effects the network in Bitcoin.
00:04 < gmaxwell> (plus have a much easier time doing it: to improve anything in bitcoin involves convincing a lot of people: some who are actually opposed to decenteralization, many who are just clueless, etc.. vs an altcoin you can just put it in.. Fiat Lux.)
00:05 < sipa> ok, let's start egocoin
00:05 < HM> Really you should be objecting to me using a thin client because it puts *your* security at risk
00:05 < HM> in theory
00:05 < sipa> depends what you're comparing it to
00:06 < gmaxwell> it's complicated, if your alternative is no bitcoin at all I'd rather you use the spv client.
00:06 < sipa> you running a thin client vs you not using bitcoin at all isn't a decrease in security
00:06 < sipa> !hi5
00:06 < HM> lol
00:07 < gmaxwell> Thats why I gave those CI passing examples above "I'll run a SPV wallet only, if and only if I honestly don't have the resources to run a full node" and "I'll run both a SPV wallet (on my phone) and a regular one elsewhere"
00:08 < gmaxwell> (the latter means that full nodes need to be cheap enough to run for many people, but at least thats a pure technical challenge)
00:09 < HM> the problem is, more and more people will likely be introduced to Bitcoin through thin clients or hosted wallets
00:10 < HM> you're then trying to argue for additional work that sees no personal or immediate benefit
00:11 < HM> that's a tougher position to start in than having people run full nodes and then saying "this is why you can have nice things"
00:12 < gmaxwell> maybe, we ultimately don't need _everyone_ to run one. But the problem is that what you're describing to me is basically where ~no one runs one.  Who would have more incentive to than you esp in a world where most people already weren't?
00:12 < gmaxwell> And I dunno about trying to argue: I'm just stating what I think is the logical conclusion. The things I am observing are telling me that bitcoin is doomed.
00:13 < HM> Some people can walk away from Bitcoin, some can't or won't for other reasons.
00:13 < HM> those that can't will run full nodes if it came to a crunch
00:13 < HM> those that can, know this
00:13 < gmaxwell> sure, and something called "bitcoin" might exist forever, but it wouldn't be the thing that I would call bitcoin.
00:13 < gmaxwell> This isn't clear to me.
00:13 < amiller> gmaxwell, thanks for listening to my idea and describing it as hiding the income, i hadn't thought of it that way but that totally helps
00:13 < gmaxwell> "if it came to a crunch" is too late.
00:14 < amiller> i'm going to use it immediately and remember to give you credit :o
00:14 < gmaxwell> amiller: thanks! I hope you're able to come up with something interesting!
00:14 < amiller> mooncoin
00:14  * amiller drifts off into space
00:14 < HM> fyi, i don't think bitcoin is doomed
00:15 < amiller> it's hard not to worry that the whole internet is doomed these days :/
00:15 < gmaxwell> HM: You could say that the users of the USD would reject inflation. Except that they don't. Incrementally everyone happily agrees a little inflation is A OK and good for expident interests.
00:16 < gmaxwell> HM: I don't know that its doomed, but patterns are suggestive to me that its initial argument for existance is not likely to be upheld. I'd probably just give it 50/50. I would give it less, but I've seen some remarkable arguments from people I normally wouldn't have expected to "get it" that show they really do get the motivations for such a system... and I get a bit of hope from that.
00:17 < HM> I don't think people worry about central banks
00:17 < HM> they see them as benign
00:17 < HM> it's high street and investment banks that people think of as the devil
00:17 < HM> so just use cash, right?
00:18 < gmaxwell> HM: if you've not read… http://p2pfoundation.ning.com/forum/topics/bitcoin-open-source
00:19 < gmaxwell> But I absolutely do think that a lot of people worry about central banks, for a great many reasons... (some more dumb than others— e.g. inflation seems to be the most worried about thing, but its probably one of the less harmful things. They engage in incredible acts of economic distortion on a worldwide scale.)
00:21 < gmaxwell> We've invented what amounts to national-scale indentured servitude substantially via the excessive economic power of central banks, which props up worldwild wealth inequality. (And even if I were some kind of inhumane pure libertarian I would sill observe that wealth distributions which are too unequal are not just socially unfair, they're objectively inefficient and lower mankinds longterm odds of survival.)
00:24 < HM> I've not really made up my mind on all that stuff
00:26 < gmaxwell> well the proof that too much inequality is inefficient is pretty simple, though it doesn't tell you exactly where too much is:  at some point the have-nots will find it in their best interest to take by force, and so increasing amounts of resources have to be diverted to preventing that or defending from it. In the end, cooperation has certian fundimental efficiencies. :)
00:33 < HM> competition and war is also a fairly good equaliser
00:34 < HM> if you're in competition or in a battle, the stable state is parity
00:35 < HM> if you suck, you'll die. if you're too good, they die. either way the competition ends. if it is to continue then parity has to be achieved.
00:36 < HM> it's not clear to me how cooperation is naturally efficient
00:36 < HM> you can pool resources but you can do that through trade
00:38 < Luke-Jr> cooperation/dependency is efficient because you have each person specialise in their one task
00:39 < Luke-Jr> (on the other hand, it has scaling issues at a point)
00:43 < HM> I guess the bottom line is, i'm not yet convinced that monetising debt and loose cannon central banks are really a new kind of threat to society
00:44 < HM> forcing the money supply to grow means you're forcing people to fill the world with goods and services or face desperate inflation and unemployment and mockery from other nations or whatever
00:45 < HM> but everyones doing it, so it's a bit more like an arms race than anything else
00:45 < amiller> the whole concept of money is a wrong design, it only works if it's "universal," by definition, and it's unrealistic to make it universal
00:46 < HM> i can't agree with that
00:47 < amiller> if economists realized how difficult even 'barter' is to enforce without police, the whole mythic story of money would be a lot different
00:48 < HM> if every nation had their own sovereign bitcoin
00:49 < HM> nobody would be able to set an exchange
00:50 < amiller> that's invalid, there's plenty of exchange between bitcoin and the other goofball knockoffs
01:00 < HM> I'm not so sure it'd work. I think a bit of currency manipulation, while negative overall, can help to smooth off the sharp edges when it comes to trade imbalances
01:07 < HM> i guess what you mean by universal is the €
01:07 < HM> not a universal bitcoin that can't be floated.
02:02 < gmaxwell> Someone should create a P2SH address that serves as a bounty for a SHA-256 collision. The script would be something like "OP_2DUP OP_NUMNOTEQUAL OP_VERIFY OP_SHA256 OP_SWAP OP_SHA256 OP_EQUALVERIFY"
02:05 < sipa> ha
02:06 < gmaxwell> We can do this for SHA1 too. :P which is actually more exciting, since that one is more likely to happen.
02:06 < sipa> is there a sha1 opcode?
02:06 < gmaxwell> yup.
02:06 < sipa> orly?
02:07 < gmaxwell> OP_RIPEMD160 / OP_SHA1 / OP_SHA256 / OP_HASH160 / OP_HASH256   (the last is sha256^2 of course)
02:07 < gmaxwell> I think the idea was to use them for binding external systems.
02:11 < gmaxwell> I'm a little surprised that no such bounties already exist.
02:11 < gmaxwell> I looked through every singual unusual script in the utxo set and there is basically nothing interesting there.
02:12 < gmaxwell> (well, there were 5 anyone-can-takes which I took, and a single puzzle one for 0.09 btc which I solved. And I will not be surprised if petertodd made that one)
02:13 < gmaxwell> there may be other puzzles but they're ones that involve just guessing the keys.
10:37 < Luke-Jr> psst, come play https://github.com/chronokings/chronokings on testnet with me :P
10:43 < michagogo> Luke-Jr: What is it?
10:43 < Luke-Jr> michagogo: blockchain-based game
10:44 < michagogo> What do I need to get it working?
11:21 < Luke-Jr> michagogo: qmake && make
11:23 < michagogo> Luke-Jr: On Windows?
11:45 < Luke-Jr> michagogo: good luck :P
11:45 < michagogo> Luke-Jr: lol
11:45 < michagogo> What are the dependancies?
11:45 < michagogo> I guess I could boot up my vm
11:45 < Luke-Jr> same as Namecoin-Qt I guess
11:46 < michagogo> Namecoin-Qt?
11:50 < michagogo> Luke-Jr: Hmm, just noticed https://github.com/chronokings/chronokings/blob/master/contrib/easywinbuilder/README.md
11:50 < michagogo> I'll give it a try
14:12 < michagogo> Luke-Jr: Meh, can't get it to build
14:13 < michagogo> Not even in an Ubuntu vm
22:29 < gmaxwell> sipa: in your libsecp256k1 can't you eliminate/highly abbreviate the secp256k1_ge_is_valid when using a compressed key?  I think for our curve the definition of being a valid point is not being the point at infinity, and the same y criteria as public key recovery.
22:30 < gmaxwell> sipa: if so, that should make your code a bit faster for compressed keys. (actually, I think it would make compressed keys almost the same speed as non-compressed ones)
22:32 < gmaxwell> (I was looking to see if various things checked that the public point was valid— seeing if I can break anything with the fact that the twist of secp256k1 can easily have the DLP solved for it)
23:32 < gmaxwell> (I was impressed the the 'factor' command could handle 1286578769603068245382716924545379906921918859152521322839515520912848165551 )
23:32 < gmaxwell> (a 256 bit number)
23:39 < gmaxwell> Luke-Jr: did you get that game working? I think I just got it compiled.
23:40 < Luke-Jr> gmaxwell: yeah, played for a bit and got bored :P
23:40 < Luke-Jr> guess I can hop back on
23:40 < gmaxwell> do I just run the resulting executable does it need a node running?
23:40 < gmaxwell> is it namecoin based?
23:41 < Luke-Jr> it seems to just work
23:41 < Luke-Jr> yes
23:41 < Luke-Jr> bah, someone killed me again I think
23:41 < Luke-Jr> gmaxwell: note all activity is testnet
23:41  * Luke-Jr ponders if Eligius should have rate-limited it
23:41 < Luke-Jr> (mining, I mean)
23:42 < gmaxwell> oh, its namecoin testnet?
23:42 < Luke-Jr> I hope not the real namecoin one!
23:42 < Luke-Jr> I would feel bad for polluting it
23:42 < gmaxwell> I don't appear to be connected.
23:43 < Luke-Jr> hrm
23:43 < gmaxwell> got a node I can -connect?
23:44 < Luke-Jr> wait, UPnP is broken
23:44 < Luke-Jr> so that won't work :/
23:44 < Luke-Jr> try 192.241.222.65
23:45 < Luke-Jr> hrm
23:45 < Luke-Jr> my client keeps crashing due to memory allocation failures :/
23:46 < gmaxwell> hacking directly on the client is a dumb way to implement something like this. :P
23:47 < Luke-Jr> :p
23:48 < gmaxwell> yea, can't reach that host.
--- Log closed Tue Sep 10 00:00:23 2013