1 2013-05-30 00:26:51 <graingert> gmaxwell: the first 50 bitcoins, those in the genesis block. They're unspendable right?
2 2013-05-30 00:27:02 <graingert> due to a bug?
3 2013-05-30 00:27:16 <gmaxwell> graingert: no idea if it was a bug.
4 2013-05-30 00:27:21 <gmaxwell> But yes, they're unspendable.
5 2013-05-30 00:27:25 <graingert> why's that?
6 2013-05-30 00:27:36 <graingert> out by one error?
7 2013-05-30 00:27:43 <graingert> absolute refusal to pre-mine
8 2013-05-30 00:27:44 <graingert> ?
9 2013-05-30 00:27:46 <gmaxwell> Because the code generated the genesis block directly and never inserted that coinbase transaction into the database.
10 2013-05-30 00:28:10 <graingert> lol?
11 2013-05-30 00:28:53 <gmaxwell> It may have been a mistake, but otoh, it seems a little unlikely that the code could have been tested at all without noticing that the coins couldn't be spent, unless the genesis block coins used a key that wasn't in any wallet used for testing.
12 2013-05-30 00:30:28 <warren> gmaxwell: the boost issue
13 2013-05-30 00:37:02 <warren> gmaxwell: kill -9 required to exit, need to -salvagewallet often
14 2013-05-30 01:04:51 <super3> was anyone at the bitpay bitcoin meetup today?
15 2013-05-30 01:05:07 <Luke-Jr> super3: off-topic
16 2013-05-30 01:06:00 <super3> yes. slightly...
17 2013-05-30 01:09:43 <super3> so hey, whats up?
18 2013-05-30 01:12:32 <super3> has anyone important taken a look at vhf's commit yet? https://github.com/bitcoin/bitcoin/pull/2679
19 2013-05-30 01:22:40 <sipa> super3: seems sane to me, but i let GUI stuff to people who work on that :)
20 2013-05-30 01:27:34 <super3> sipa, who would be the gui people?
21 2013-05-30 01:29:48 <sipa> wumpus, Diapolo
22 2013-05-30 01:37:11 <super3> sipa, also you are awesome
23 2013-05-30 01:41:26 <owowo> https://www.btproof.com/ <--IsDust() ;o)
24 2013-05-30 01:45:22 <gigavps> is there any reason to continue to include coinbaseaux->flags into a pools coinbase transactions?
25 2013-05-30 01:49:24 <sipa> owowo: bah, yet another one...
26 2013-05-30 01:50:12 <owowo> exactly, why cant ppl use NMC for thing like this...
27 2013-05-30 01:52:51 <sipa> or a dedicated side-chain, like chronobit does
28 2013-05-30 01:53:31 <sipa> which provides A) zero impact on UTXO set B) strictly bounded O(1)-per-block effect on blockchain size C) higher granularity
29 2013-05-30 03:00:12 <Micha> iPhone|Kinda confusion how github emails come from full names and not usernames, as is displayed on the site
30 2013-05-30 03:01:40 <Micha> iPhone|So instead of getting emails from Diapolo, laanwj, and sipa, they come from Philip, Wladimir, and Pieter.
31 2013-05-30 03:01:51 <sipa> Sorry.
32 2013-05-30 03:02:22 <Micha> iPhone|Lol, you don't have any reason to apologize
33 2013-05-30 03:02:50 <sipa> that's why i used a capital and correct punctuation
34 2013-05-30 03:03:14 <Micha> iPhone|Huh?
35 2013-05-30 03:03:26 <sipa> for some reason, when used unexpectedly, it seems to convey irony imho :)
36 2013-05-30 03:04:12 <Micha> iPhone|Interesting, I tend to just use it automatically in most cases.
37 2013-05-30 06:18:05 <Micha> iPhone|If anyone cares, that false positive for bitcoin accessing the dyndns ip checker should be fixed shortly. http://i.imgur.com/R3oijur.png
38 2013-05-30 07:36:02 <kinlo> is it normal for the keypool reserve to keep going up in debug.log?
39 2013-05-30 08:10:01 <BlueMatt> ACTION just noticed the hash160 hex of his donation address has dead in it
40 2013-05-30 08:10:10 <BlueMatt> too bad I missed the beef
41 2013-05-30 08:51:24 <X-Factor> damnit, I hate it when I get into a coding zone, and look up and realize I've been coding all night and is light out now
42 2013-05-30 08:54:21 <nsh> worse things happen at sea :)
43 2013-05-30 08:59:20 <X-Factor> yes I'm sure that is true, heh
44 2013-05-30 09:10:54 <warren> sipa: https://github.com/sipa/secp256k1/commit/561b0e1044efb71edf06f3d27e6a783203238cfa I'm curious, what is the purpose of this?
45 2013-05-30 09:16:15 <nsh> 152 200 } // FTFY
46 2013-05-30 09:29:42 <jgm> Y'know, if people are going to keep attempting to put junk in to the blockchain like btproof how about building something in to the protocol which allows it but accounts for it, for example by setting a value for it and ensuring that the coin comes back to the bitcoin foundation? You could also create a separate address space for it so that it wouldn't need to clog up UTXO but would still be visible in the blockchain
47 2013-05-30 09:34:19 <warren> sipa: has anyone tried secp256k1 on win32 yet?
48 2013-05-30 10:16:44 <wallet43> cant we pay mining pools to include some arbitrary data in the coinbase ?
49 2013-05-30 10:17:46 <nsh> you could also pay a boiler repairman to resuscitate your dead baby; that doesn't make it a good idea
50 2013-05-30 10:24:25 <kinlo> wallet43: what data would you want to include?
51 2013-05-30 10:24:37 <kinlo> wallet43: I'd recommend looking into namecoin to store data in the blockchain
52 2013-05-30 10:56:50 <wallet43> mostly porn but also some timestamps
53 2013-05-30 10:57:29 <wallet43> im thinking of services like https://www.btproof.com/??
54 2013-05-30 10:58:04 <wallet43> if they d pay some miners to include the hashes, there would less utxo
55 2013-05-30 11:02:07 <BlueMatt> like clockwork..."[TEST] This is a test alert..."
56 2013-05-30 11:02:38 <melvster> wallet43: arbitrary data?
57 2013-05-30 11:02:59 <MKCoin> These test broadcasts always get my attention immediately, so I guess that's a success
58 2013-05-30 11:03:35 <BlueMatt> last week's had negative priority so it barely showed up, but today it actually popped up on my desktop
59 2013-05-30 11:04:17 <eps> non tx related data in the blockchain is a bad idea IMHO
60 2013-05-30 11:04:27 <MKCoin> I think after the chain fork, everyone is just preparing for some sort of implosion, so it was a bit of a blessing in disguise by increasing alertness
61 2013-05-30 11:04:57 <wallet43> melvster: well only hashes
62 2013-05-30 11:05:21 <melvster> wallet43: you can already put hashes in the block chain
63 2013-05-30 11:05:42 <wallet43> by creating tx
64 2013-05-30 11:05:55 <melvster> the block chain *IS* a timestamp server
65 2013-05-30 11:06:13 <melvster> satoshi is very clear about that in his white paper
66 2013-05-30 11:06:56 <eps> i'm butthurt about 9GB
67 2013-05-30 11:07:41 <kjj> yes, for loose definitions of "time" and "stamp"
68 2013-05-30 11:07:43 <melvster> i think he may also have made a forum post about it
69 2013-05-30 11:08:48 <wumpus> test alert broadcast ftw
70 2013-05-30 11:08:50 <darsk1ez> and a looser definition of "server"
71 2013-05-30 11:08:51 <darsk1ez> :)
72 2013-05-30 11:09:03 <melvster> lol
73 2013-05-30 11:09:26 <kjj> it would be better to say that it is an "orderstamping" service
74 2013-05-30 11:09:46 <melvster> no it wouldnt
75 2013-05-30 11:09:51 <kjj> as in, bitcoin only has a very crude notion of what time things happen, but a very solid notion of what order they happen in
76 2013-05-30 11:09:54 <melvster> white paper section 3
77 2013-05-30 11:09:57 <melvster> 'The solution we propose begins with a timestamp server.'
78 2013-05-30 11:10:25 <melvster> in fact section 3 is called 'Timestamp server'
79 2013-05-30 11:10:30 <kjj> I've read the paper. I'm not arguing aginst it, I'm describing how the system works in reality
80 2013-05-30 11:12:46 <melvster> as you prefer, im content with satoshi's description ...
81 2013-05-30 11:13:05 <melvster> but yes
82 2013-05-30 11:13:06 <kjj> the difference isn't huge
83 2013-05-30 11:13:16 <melvster> the motivation is to get things in the right order
84 2013-05-30 11:13:19 <melvster> to prevent double spend
85 2013-05-30 11:14:00 <melvster> 'In
86 2013-05-30 11:14:00 <melvster> this paper, we propose a solution to the double-spending problem using a peer-to-peer distributed
87 2013-05-30 11:14:00 <melvster> timestamp server to generate computational proof of the chronological order of transactions.'
88 2013-05-30 11:19:17 <kjj> all I'm saying is that the "timestamp server" in the whitepaper is not like any other "timestamp server" that people are familiar with. for one thing, it is not terribly concerned with time.
89 2013-05-30 11:20:16 <melvster> 'I like Hal Finney's idea for user-friendly timestamping. Convert the hash of a file to a bitcoin address and send 0.01 to it' -- Satoshi
90 2013-05-30 11:20:41 <melvster> https://bitcointalk.org/index.php?topic=2162.msg28533#msg28533
91 2013-05-30 11:21:19 <melvster> cc wallet43 ^^
92 2013-05-30 11:21:52 <kjj> yes, I've actually implemented a timestamping system in exactly that way.
93 2013-05-30 11:22:14 <melvster> kjj: did you make the bitproof site?
94 2013-05-30 11:22:33 <kjj> no, I never released mine
95 2013-05-30 11:22:38 <melvster> ah ok
96 2013-05-30 11:22:49 <melvster> well you could argue the genesis block is a timestamp hash
97 2013-05-30 11:23:21 <kjj> yes it is, but, again, only in a very crude sense
98 2013-05-30 11:23:35 <melvster> i guess the fear is that it's yet another reason to attack bitcoin if you are relying on it to save timestamps... or maybe it's another reason to protect bitcoin!
99 2013-05-30 11:23:56 <kjj> we can be reasonably sure that bitcoin was released at some time after that headline was published. but that's all we can really say with certainty
100 2013-05-30 11:24:04 <melvster> yes
101 2013-05-30 11:24:14 <melvster> and i think it has a huge proof of work there too
102 2013-05-30 11:24:32 <kjj> heh. not huge at all by modern standards
103 2013-05-30 11:24:48 <melvster> true
104 2013-05-30 11:25:47 <kjj> there has been a lot of speculation around that block, and the early blocks in general. I followed it, but not carefully enough to come up with a strong opinion on the matter
105 2013-05-30 11:25:49 <melvster> I dont mean to be contrary, but i know that satoshi worked on that white paper for at least 10 years, so I really see it as a kind of personal bitcoin bible :)
106 2013-05-30 11:26:29 <kjj> hmm. most documents that take 10 years to write are far longer.
107 2013-05-30 11:26:44 <melvster> i suspect he cut some things out :)
108 2013-05-30 11:27:48 <eps> you know he worked on it for 10 years?
109 2013-05-30 11:29:05 <melvster> i dont know for sure, but it's something i believe from talking to people that i trust
110 2013-05-30 11:29:37 <kjj> plenty of people have been interested in electronic cash for 20+ years. I could see him working on the idea for 10 years. but I'd be amazed if the paper took more than a week or two to write (once the idea was fully formed)
111 2013-05-30 11:30:21 <eps> the paper to me seems like something that was written after having written some code
112 2013-05-30 11:30:36 <kjj> but again, minor disagreement. it comes down to how you account for the time. if you want to include idea-work in paper-time, I can't really argue against it other than to say that I disagree
113 2013-05-30 11:30:45 <eps> trying to get everything straight in his head, before continuing an early version
114 2013-05-30 11:30:58 <melvster> yes we know he wrote about 18 months of code before releasing the paper, but he was active on the crypto group since at least 1998
115 2013-05-30 11:31:36 <eps> how do you know that?
116 2013-05-30 11:31:53 <melvster> as i said, from talking to people i trust
117 2013-05-30 11:31:58 <eps> uh ok
118 2013-05-30 11:32:41 <eps> my feeling has always been that he was a relatively young guy, who found the idea of crypto currencies appealing
119 2013-05-30 11:32:48 <eps> of course i can't prove any of that
120 2013-05-30 11:32:53 <melvster> agree
121 2013-05-30 11:32:58 <melvster> well
122 2013-05-30 11:33:05 <melvster> depends what you mean by young
123 2013-05-30 11:33:22 <Goonie_> Luke-Jr: Do you have a chart for the 0.8.2 client propagation? I'd like to know how many percent of clients already enforce the new fee rules.
124 2013-05-30 11:33:48 <eps> like early 20s
125 2013-05-30 11:34:01 <melvster> dan kaminsky speculated that satoshi was 3-8 quants out of london or new york at the conference
126 2013-05-30 11:34:19 <eps> i still think he is a single guy
127 2013-05-30 11:34:30 <melvster> eps: yes me too
128 2013-05-30 11:34:43 <melvster> and quants aren't as smart as satoshi
129 2013-05-30 11:34:58 <nsh> the internet is always telling me that single cryptographic geniuses in my area are looking to meet
130 2013-05-30 11:35:28 <eps> bitcoin is impressive but it isn't perfect, it's a very left field idea as well, it fits the MO of a lone polymath
131 2013-05-30 11:35:31 <melvster> nsh: nice ... is that looking to meet without revealing their identity ? :P
132 2013-05-30 11:35:53 <kjj> Dan likes to talk out of his ass. it gained him a bunch of fame during a formative period in his career, and now he doesn't dare stop
133 2013-05-30 11:35:57 <eps> yeah looking not to meet would be more realistic
134 2013-05-30 11:35:58 <nsh> they seem to use a revolving stock image system, so it's a safe assumption
135 2013-05-30 11:36:17 <melvster> yes I only know of one quant that could have pulled off something like bitcoin and that's Peter Kliedman, but I doubt it's hin
136 2013-05-30 11:36:19 <melvster> him
137 2013-05-30 11:37:28 <melvster> kjj: lol ... yeah he rubbished bitcoin when he first saw it, now he's a fanboy ... go figure ... he also said proof of work function would not last the end of year with 100% Probability lol ... but he's entertaining, that's for sure!
138 2013-05-30 11:40:36 <melvster> the 'future of bitcoin' panel was interesting
139 2013-05-30 11:40:52 <nsh> synopsis?
140 2013-05-30 11:41:27 <melvster> lots covered https://www.youtube.com/watch?v=_qdr_Z3hrqQ
141 2013-05-30 11:41:40 <melvster> fellow traveler is awesome
142 2013-05-30 11:42:42 <melvster> he's another one that's been working in this space 10 years
143 2013-05-30 11:42:47 <melvster> +
144 2013-05-30 11:46:00 <nsh> oh, hmm
145 2013-05-30 11:46:42 <kjj> wow. amazon really fucked up their site
146 2013-05-30 12:02:04 <binaryFate> hi there
147 2013-05-30 12:04:20 <altgribble> Goonie_ was last seen in #bitcoin-dev 1 week, 2 days, 14 hours, 23 minutes, and 22 seconds ago: <Goonie_> what's the plan for 0.8.2? will there be an additional rc2?
148 2013-05-30 12:04:20 <BlueMatt> ;;seen Goonie_
149 2013-05-30 12:04:30 <BlueMatt> :(
150 2013-05-30 12:08:15 <Goonie_> BlueMatt: I'm here
151 2013-05-30 12:09:52 <michagogo> Hmm, someone should fix the topic
152 2013-05-30 12:10:22 <BlueMatt> oh, hey
153 2013-05-30 12:11:21 <kjj> nice
154 2013-05-30 12:11:27 <Goonie_> bluematt: missing " | The cake is a lie" (-:
155 2013-05-30 12:11:40 <BlueMatt> heh
156 2013-05-30 12:11:53 <BlueMatt> Goonie_: pm?
157 2013-05-30 12:12:03 <Goonie_> sure, can you login jabber?
158 2013-05-30 12:12:31 <BlueMatt> hmm...I think i have a jabber account on my vps, but I cant get to it from here...
159 2013-05-30 12:13:22 <nsh> you can create a jabber account with near zero friction using pidgin(+otr)
160 2013-05-30 12:13:36 <nsh> (it's just a checkbox to create account when entering credentials)
161 2013-05-30 12:14:37 <jouke_> Hmmm, it isn't possible to get the raw transaction of a spent wallet transaction? (I don't keep the full index on that node)
162 2013-05-30 12:16:13 <helo> i thought sent and received transactions were stored in wallet.dat
163 2013-05-30 12:19:44 <jouke_> helo: I thought so too, but raw transaction doesn't return the information.
164 2013-05-30 12:20:26 <helo> so you can see it in listtransactions, but getrawtransaction doesn't return it?
165 2013-05-30 12:21:10 <helo> jouke_: odd... same behavior here
166 2013-05-30 12:21:27 <helo> i feel like i've been able to getrawtransaction for my wallet transactions in the past
167 2013-05-30 12:22:05 <jouke_> +1
168 2013-05-30 12:26:56 <kjj> if you want to retain that behavior in 0.8+, you need to enable full indexing
169 2013-05-30 12:29:51 <helo> as long as it was intended :)
170 2013-05-30 12:30:25 <helo> so the wallet doesn't actually contain the full transaction payload?
171 2013-05-30 12:31:02 <helo> surely with the growth of my testnet wallet, it must
172 2013-05-30 12:31:21 <helo> it's nearly as big as MAX_BLOCKSIZE
173 2013-05-30 12:32:27 <jouke_> helo: your wallet is almost 1mb?
174 2013-05-30 12:33:54 <helo> yes
175 2013-05-30 12:34:28 <damo22> i have identified an issue that could be detrimental to bitcoin, its just a problem with the client software, nothing major just its compilation
176 2013-05-30 12:34:41 <helo> damo22: details?
177 2013-05-30 12:34:45 <kjj> what's the issue?
178 2013-05-30 12:35:02 <damo22> openssl is not entirely free software
179 2013-05-30 12:35:07 <damo22> with ECC
180 2013-05-30 12:35:25 <kaniini> uhh, what?
181 2013-05-30 12:35:37 <kaniini> oh. fsf/member
182 2013-05-30 12:35:47 <kaniini> okay, that explains the lack of sanity.
183 2013-05-30 12:35:53 <damo22> :)
184 2013-05-30 12:36:12 <kaniini> i think openssl has greater issues than software freedom
185 2013-05-30 12:36:16 <kaniini> have you looked at the codebase?
186 2013-05-30 12:36:22 <kaniini> it is seriously awful
187 2013-05-30 12:36:45 <damo22> i havent read it just been trying to compile the client
188 2013-05-30 12:36:53 <damo22> using totally free software
189 2013-05-30 12:37:08 <kaniini> damo22: i do not believe the ECC code to be any less free than any other component of OpenSSL.
190 2013-05-30 12:38:33 <kaniini> specifically, all the ECC code has this statement:
191 2013-05-30 12:38:35 <kaniini> 61 * Portions of the attached software ("Contribution") are developed by
192 2013-05-30 12:38:37 <kaniini> 62 * SUN MICROSYSTEMS, INC., and are contributed to the OpenSSL project.
193 2013-05-30 12:38:51 <kaniini> which places it all, in it's entirety, under the same license as OpenSSL itself.
194 2013-05-30 12:39:14 <damo22> ahh ok
195 2013-05-30 12:39:24 <kaniini> i.e. BSD license with the 2 additional advertising clauses
196 2013-05-30 12:39:46 <kaniini> some 'free' distributions disable it because they are afraid it is patented
197 2013-05-30 12:39:54 <damo22> maybe the version of openssl you have uses the sun code and people compiling bitcoin should use this too
198 2013-05-30 12:40:19 <kaniini> damo22: the version of openssl i have is just mainline as distributed by them
199 2013-05-30 12:40:26 <kaniini> damo22: what distribution is this
200 2013-05-30 12:40:51 <damo22> uh, its not really a distro so
201 2013-05-30 12:41:01 <kaniini> damo22: where did your openssl source come from
202 2013-05-30 12:41:17 <kaniini> and... further...
203 2013-05-30 12:41:21 <kaniini> openssl is not free software
204 2013-05-30 12:41:25 <damo22> i better check
205 2013-05-30 12:41:26 <kaniini> it has advertising clauses
206 2013-05-30 12:42:08 <kaniini> (though, i do not understand why bitcoin uses openssl at all. surely crypto++ would be a better fit?)
207 2013-05-30 12:42:20 <damo22> i just dont want anyone getting in trouble over some stupid patent
208 2013-05-30 12:44:12 <kaniini> ahh, so the issue is not a matter of software freedom at all, and instead, bullshit
209 2013-05-30 12:44:14 <kaniini> gotcha.
210 2013-05-30 12:44:31 <kaniini> damo22: well, i have some bad news for you!
211 2013-05-30 12:44:45 <kaniini> damo22: bitcoin extensively uses elliptic curve cryptography
212 2013-05-30 12:44:52 <damo22> in my opinion, i think something as public and important as a p2p distributed cryptocurrency client should not use any non-free software
213 2013-05-30 12:44:58 <kaniini> well...
214 2013-05-30 12:45:02 <kaniini> it doesn't!
215 2013-05-30 12:45:16 <kaniini> i mean
216 2013-05-30 12:45:18 <kaniini> it uses openssl
217 2013-05-30 12:45:23 <kaniini> but that's politically non-free
218 2013-05-30 12:45:37 <damo22> yeah
219 2013-05-30 12:45:40 <kaniini> either way...
220 2013-05-30 12:45:47 <kaniini> replacing openssl does not solve your problem
221 2013-05-30 12:45:49 <kaniini> as, basically
222 2013-05-30 12:45:54 <kaniini> the entirety of bitcoin crypto
223 2013-05-30 12:45:56 <kaniini> is ECC
224 2013-05-30 12:46:06 <kaniini> so, if you do not allow ECC cryptography
225 2013-05-30 12:46:10 <kaniini> you can't use bitcoin
226 2013-05-30 12:46:12 <damo22> i realise that,
227 2013-05-30 12:46:18 <kaniini> then... what's the bug?
228 2013-05-30 12:46:34 <damo22> which is why i raised the concern
229 2013-05-30 12:46:46 <kjj> kaniini, you ever think about condensing your sentences into single lines? :)
230 2013-05-30 12:48:11 <kaniini> damo22: i think that the FSF is making itself obsolete through fighting things this way instead of creating real solutions
231 2013-05-30 12:48:38 <Vinnie_win> OpenSSL sucks a big fat cock
232 2013-05-30 12:48:41 <kjj> that's the typical view of people that don't care about software freedom
233 2013-05-30 12:48:51 <Vinnie_win> Fuck OpenSSL and fuck anyone who programs in C
234 2013-05-30 12:49:08 <helo> Vinnie_win: i don't want to :(
235 2013-05-30 12:49:15 <damo22> well if the bitcoin client could not be compiled without using a totally closed piece of code, bitcoin could be destroyed by anyone with enough power to buy that code
236 2013-05-30 12:49:16 <Vinnie_win> It wasn't an imperative foolio
237 2013-05-30 12:49:27 <kaniini> damo22: uhh
238 2013-05-30 12:49:30 <kaniini> damo22: but, it's not
239 2013-05-30 12:49:38 <kjj> FSF takes a hard stance on a single issue, and they stick by it, come hell or high water. that's laudable, in my opinion.
240 2013-05-30 12:49:40 <kaniini> damo22: the ECC code in OpenSSL is fully open-source
241 2013-05-30 12:49:42 <Vinnie_win> Who will help me implement SSL in C++ using cryptopp ?
242 2013-05-30 12:49:45 <damo22> i know its not, which is a great thing
243 2013-05-30 12:49:53 <damo22> i commend you guys on your work
244 2013-05-30 12:50:14 <kaniini> i do not understand your complaint
245 2013-05-30 12:50:20 <kaniini> again -- the ECC code in OpenSSL is fully open-source
246 2013-05-30 12:50:40 <kjj> open source != free software
247 2013-05-30 12:50:45 <Vinnie_win> You mean to say that ECC is provided under a "permissive license" (i.e. MIT, ISC)
248 2013-05-30 12:50:57 <Vinnie_win> http://en.wikipedia.org/wiki/Permissive_free_software_licence
249 2013-05-30 12:51:07 <Vinnie_win> Versus GPL which is not a permissive license (but is open source / free)
250 2013-05-30 12:51:41 <kaniini> Vinnie_win: i can read, modify, and redistribute the source, it is therefore 'open source'. i did not say it was 'free software'.
251 2013-05-30 12:52:07 <kaniini> Vinnie_win: really, i've been doing this a very long time. i know what i'm doing.
252 2013-05-30 12:52:30 <kaniini> kjj: that does not matter. bitcoin is also not 'free software'
253 2013-05-30 12:52:38 <porquilho> ;;ticker
254 2013-05-30 12:52:40 <altgribble> BTCUSD ticker | Best bid: 131.49983, Best ask: 131.50000, Bid-ask spread: 0.00017, Last trade: 131.50000, 24 hour volume: 21335.73758034, 24 hour low: 128.01000, 24 hour high: 132.71988, 24 hour vwap: 130.87539
255 2013-05-30 12:52:55 <Vinnie_win> "open source" includes a lot of different licenses
256 2013-05-30 12:53:05 <Vinnie_win> GPL is open source as is MIT but only one of them is permissive
257 2013-05-30 12:53:41 <kaniini> Vinnie_win: for the purposes of conversation though, damo22 is trying to claim the ECC code is not legally kosher
258 2013-05-30 12:53:55 <kaniini> Vinnie_win: it is, however, legally kosher. that is what i was trying to convey.
259 2013-05-30 12:54:11 <kaniini> damo22: i have now checked openssl 0.9.8 as well
260 2013-05-30 12:54:13 <Vinnie_win> kaniini: I see
261 2013-05-30 12:54:21 <helo> anyone care to think about (and share) the implications of the work function being a composition of various hash functions, determined in each block by some subset of data from previous blocks (eg hash-of-last-2016-nonces)?
262 2013-05-30 12:54:27 <kaniini> damo22: all ECC code is under the openssl license itself.
263 2013-05-30 12:54:35 <damo22> i am a fsf associate member, nothing special, neither am i a lawyer, but i want to help make sure bitcoin devs dont get in trouble since there is a lot of money in bitcoin
264 2013-05-30 12:55:19 <kaniini> damo22: frankly, i would be more concerned about financial patent trolls than anything else with bitcoin
265 2013-05-30 12:55:29 <helo> is it feasible that even in such a convoluted system, eventually someone could make an asic powerful enough and overpower everyone else?
266 2013-05-30 12:55:36 <kaniini> helo: Yes
267 2013-05-30 12:55:49 <kaniini> helo: but, it is not likely.
268 2013-05-30 12:56:03 <kaniini> helo: as there are other stakeholders working to do the same
269 2013-05-30 12:56:14 <kaniini> helo: so, any breakthrough like that would be temporary at best
270 2013-05-30 12:56:14 <kjj> helo: strange hashing systems make difficulty unpredictable
271 2013-05-30 12:56:40 <jgarzik> damo22, the patent issue is an old and known issue
272 2013-05-30 12:56:51 <jgarzik> Fedora distro elides ECDSA from their openssl pkg
273 2013-05-30 12:57:29 <damo22> jgarzik: thanks, i read your post on one of the old fedora bug things
274 2013-05-30 12:57:58 <damo22> or was it on a bitcoin forum
275 2013-05-30 12:59:17 <jgarzik> damo22, Most likely, the ECDSA curve used by bitcoin is not under the scary patent zone in EC (but of course Disclaimer: This Is Not An Official Determination By Lawyers)
276 2013-05-30 12:59:41 <BlueMatt> who actually owns those patents?
277 2013-05-30 12:59:54 <kaniini> jgarzik: indeed, secp256k1 curve is public domain
278 2013-05-30 13:00:02 <kjj> mostly Certicom and the NSA
279 2013-05-30 13:00:07 <kjj> http://en.wikipedia.org/wiki/ECC_patents
280 2013-05-30 13:00:12 <jgarzik> what kjj said ;p
281 2013-05-30 13:00:16 <kaniini> NSA patents are public domain
282 2013-05-30 13:00:26 <kaniini> they have to be under US law
283 2013-05-30 13:00:31 <damo22> i find it outrageous that a company can own a "curve"
284 2013-05-30 13:01:23 <kjj> I'm not sure that such a patent has ever been tested
285 2013-05-30 13:01:24 <kaniini> i agree
286 2013-05-30 13:01:52 <kaniini> either way secp256k1 is public domain
287 2013-05-30 13:01:57 <BlueMatt> hmm, RIM owns them
288 2013-05-30 13:02:06 <kaniini> (i used it in my ECDSA-CHALLENGE SASL mechanism for this reason)
289 2013-05-30 13:02:26 <BlueMatt> RIM bought Certicom + multiple others who owned stuff
290 2013-05-30 13:02:29 <BlueMatt> quite a few others
291 2013-05-30 13:02:46 <BlueMatt> here's hoping someone "good" buys up RIM's patent portfolio
292 2013-05-30 13:04:02 <damo22> why cant we just use hyperbolic curves instead
293 2013-05-30 13:04:07 <damo22> or
294 2013-05-30 13:04:10 <kaniini> well
295 2013-05-30 13:04:12 <kaniini> because
296 2013-05-30 13:04:18 <kaniini> bitcoin already exists
297 2013-05-30 13:04:33 <kaniini> and you know, changing it, kinda would require starting over from scratch
298 2013-05-30 13:04:44 <kaniini> maybe you should start a GNU Money
299 2013-05-30 13:04:56 <damo22> :P
300 2013-05-30 13:05:00 <kaniini> which uses hyperbolic curves instead
301 2013-05-30 13:05:06 <kaniini> then stallman can start his own silk-road
302 2013-05-30 13:05:24 <wumpus> there is gnu cash :p
303 2013-05-30 13:05:36 <kaniini> "oh, we'll sell -- support -- for those drugs!"
304 2013-05-30 13:06:04 <damo22> thats funny
305 2013-05-30 13:06:34 <damo22> thanks for having me here
306 2013-05-30 13:06:46 <kaniini> actually, i don't think bitcoin is compatible with stallman's view of humanity
307 2013-05-30 13:07:00 <kjj> he seems to think it is
308 2013-05-30 13:07:05 <damo22> kaniini: how so?
309 2013-05-30 13:07:07 <kaniini> well, he would like to get rid of money in general
310 2013-05-30 13:07:33 <kaniini> perhaps he sees it as well, "okay, it's money, but it takes control away from the bastards"
311 2013-05-30 13:07:59 <damo22> i would think he likes the ability to trade things
312 2013-05-30 13:08:22 <damo22> so he can buy GNU compatible hardware
313 2013-05-30 13:08:41 <kaniini> damo22: he would prefer a world where things were freely exchanged without the necessity of trade
314 2013-05-30 13:08:50 <kaniini> or, at least, he did last decade.
315 2013-05-30 13:09:20 <kaniini> but like many things, i am sure his views evolve too
316 2013-05-30 13:09:41 <kaniini> and, i have not really kept up with them since giving up on FSF/GNU when they effectively killed DotGNU :P
317 2013-05-30 13:10:06 <jgarzik> anyway
318 2013-05-30 13:10:11 <damo22> goodbye
319 2013-05-30 13:10:12 <nsh> +1
320 2013-05-30 13:10:16 <jgarzik> since we're merging stuff
321 2013-05-30 13:10:20 <jgarzik> let's get back on topic
322 2013-05-30 13:14:13 <binaryFate> Which channel should I go for technical questions?
323 2013-05-30 13:14:35 <kjj> possibly this one. what's the question?
324 2013-05-30 13:15:15 <binaryFate> I'm experimenting with addresses, trying to figure out entirely how they are created. I cannot reproduce the example on that page: https://en.bitcoin.it/wiki/Technical_background_of_version_1_Bitcoin_addresses
325 2013-05-30 13:15:41 <binaryFate> step 2 fails for me. When I SHA-256 the public key, I have something different.
326 2013-05-30 13:16:01 <kjj> make sure you are working with binary, and not hex. also, pay attention to which end is which
327 2013-05-30 13:17:38 <binaryFate> Both input and output are hex, no?
328 2013-05-30 13:18:02 <kjj> you may want to see them in hex to make your life easier, but the operations all have to happen on the binary version
329 2013-05-30 13:18:53 <kjj> as in, you don't hash the ASCII string "045086...", you hash the 520-bit integer that the string represents
330 2013-05-30 13:19:29 <binaryFate> Ok I see... Any idea which module could help me to do it python? I use python-crypto, but don't see anything for binary hashing handling
331 2013-05-30 13:20:00 <kjj> I don't know python, but the transformation is trivial to write.
332 2013-05-30 13:20:59 <binaryFate> Yes, but my problem is not converting but that I don't know how to store such large numbers
333 2013-05-30 13:22:11 <binaryFate> I'll figure out, thanks for the anwers :)
334 2013-05-30 13:22:50 <kjj> according to wikipedia, "the built-in int (3.x) / long (2.x) integer type is of arbitrary precision"
335 2013-05-30 13:25:55 <binaryFate> afaiu, it only means it will use the longest possible type available on the machine, 64bytes on a 64bytes machine for instance. But I might be wrong
336 2013-05-30 13:27:24 <jgarzik> ok, merged a bunch
337 2013-05-30 13:27:28 <jgarzik> test test test bitcoin.git HEAD
338 2013-05-30 13:27:34 <lupine> ...
339 2013-05-30 13:29:06 <kjj> binaryFate: grab a copy of pywallet and see how other people are doing it. apparently no python users are around here to give you the correct answer
340 2013-05-30 13:31:08 <jgarzik> well, https://github.com/jgarzik/python-bitcoinlib/
341 2013-05-30 13:31:10 <jgarzik> https://github.com/jgarzik/pynode/
342 2013-05-30 13:33:10 <sipa> warren: the tweaking is the low-level operation needed for type-2 deterministic wallet schemes (includjng bip32)
343 2013-05-30 13:34:51 <sipa> Vinnie_win: fwiw, i'm working on a librarg thag efficiently implements secp256k1 math from scratch, so you don't need openssl (the origin reason was better performance though)
344 2013-05-30 13:35:22 <kaniini> sipa: that is interesting
345 2013-05-30 13:35:30 <kaniini> sipa: can you do it in straight C?
346 2013-05-30 13:35:55 <kaniini> if we can make it mimic the openssl API, it could be provided as a dropin replacement for openssls which do not have ECDSA support
347 2013-05-30 13:36:10 <Diablo-D3> you mean like osx?
348 2013-05-30 13:36:14 <Diablo-D3> fucking apple >_>
349 2013-05-30 13:36:52 <sipa> kaniini: it doesn't mimick the openssl api (which is way way overkill for something that only needs one curve), and it is straight C
350 2013-05-30 13:37:28 <kaniini> sipa: it still interests me, as ecdsa-nist256p-challenge uses secp256k1 obviously
351 2013-05-30 13:37:30 <kaniini> :)
352 2013-05-30 13:37:51 <Diablo-D3> sipa, kaniini
353 2013-05-30 13:38:21 <Diablo-D3> whats the hardest message signing curve out there
354 2013-05-30 13:38:31 <sipa> kaniini: no it doesn't
355 2013-05-30 13:38:43 <sipa> nist p256 is secp256R1
356 2013-05-30 13:38:52 <sipa> https://github.com/sipa/secp256k1/blob/master/include/secp256k1.h
357 2013-05-30 13:39:16 <sipa> Diablo-D3: there is a standard 521-bit curve (yes, not 512)
358 2013-05-30 13:39:28 <Diablo-D3> yeah Im aware some curves have weird bits
359 2013-05-30 13:40:52 <kaniini> oh, right
360 2013-05-30 13:40:56 <Diablo-D3> also, someones asking me this in another channel, is there something on the wiki that explains how coins are actually transmitted to other people?
361 2013-05-30 13:41:08 <Diablo-D3> ACTION doesnt feel like explaining it to them
362 2013-05-30 13:41:17 <kaniini> sipa: still your code can be used to implement most of secp256r1 :)
363 2013-05-30 13:41:30 <sipa> kaniini: no
364 2013-05-30 13:41:45 <sipa> it's very specifically optimized for k1
365 2013-05-30 13:42:13 <sipa> well you could adapt it of course, but it would not be trivial
366 2013-05-30 13:42:59 <nsh> jgarzik, what are you current intentions for pynode?
367 2013-05-30 13:45:45 <jgarzik> nsh, full or SPV node
368 2013-05-30 13:46:00 <nsh> excellent
369 2013-05-30 13:46:03 <jgarzik> nsh, pynode is ideally a tool that is adaptable to whatever your monitoring etc. needs are
370 2013-05-30 13:46:15 <nsh> ACTION nods
371 2013-05-30 13:46:24 <jgarzik> nsh, Part of the goal is to grow python-bitcoinlib to be "anything bitcoin related, in python"
372 2013-05-30 13:46:36 <nsh> i think i'll have a look through the code and see if it's accessible enough for me to attempt contributions
373 2013-05-30 13:46:59 <jgarzik> e.g. ideally I would like pywallet to use python-bitcoinlib
374 2013-05-30 13:47:14 <nsh> this would probably be an easier way of getting to grips with node implementation than studying C++ code until my brain implodes
375 2013-05-30 13:47:18 <nsh> it makes sense
376 2013-05-30 13:47:48 <nsh> this was the general sense i got from the future of bitcoin talk
377 2013-05-30 13:48:16 <nsh> -- importance of standardizing library code for more secure and facilitated application development
378 2013-05-30 13:48:31 <roconnor> sipa: I think I'm strongly leaning towards botnets are better than specialized hardware for POW. :)
379 2013-05-30 13:49:07 <sipa> roconnor: unsure
380 2013-05-30 13:49:23 <roconnor> sipa: botnets can be fixed; but a hardware mining oligary cannot.
381 2013-05-30 13:50:15 <sipa> the question is probably ultimately which has the largeat scaling advantage
382 2013-05-30 13:50:24 <Diablo-D3> yes which means
383 2013-05-30 13:50:31 <sipa> and that's likely an oligarch
384 2013-05-30 13:50:32 <Diablo-D3> I need to write DiabloCoin :3
385 2013-05-30 13:50:34 <roconnor> sipa: really? You think scaling is the issue?
386 2013-05-30 13:50:50 <sipa> s/scaling/network effect/
387 2013-05-30 13:51:13 <roconnor> sipa: I don't quite see what you mean by scaling in this regard.
388 2013-05-30 13:51:20 <eps> Diablo-D3: let me in on the pre-mine
389 2013-05-30 13:51:45 <sipa> i mean: is there a more-than-linear reward effect for the operator, in function of hashpower
390 2013-05-30 13:52:07 <rdponticelli> You'll allways have some form of inequality
391 2013-05-30 13:52:25 <rdponticelli> I guess diversity is good
392 2013-05-30 13:52:39 <sipa> i don't care about inequality
393 2013-05-30 13:52:44 <rdponticelli> At least an heterogeneous ecosystem will balance itself
394 2013-05-30 13:53:16 <rdponticelli> sipa: You don't?
395 2013-05-30 13:53:19 <roconnor> inequality isn't my concern. Collusion is my (minor) concern.
396 2013-05-30 13:53:36 <rdponticelli> Huge inequality is centeralization
397 2013-05-30 13:54:06 <roconnor> sipa: what is the more than linear reward with hardware mining?
398 2013-05-30 13:54:24 <sipa> the fact that hardware is close to eachother
399 2013-05-30 13:54:30 <MC1984_> kaminsky seemed pretty sure on himself that sha256 pow is not long for this world
400 2013-05-30 13:54:32 <sipa> so you can synchronize work
401 2013-05-30 13:54:43 <sipa> and not compete with yourself
402 2013-05-30 13:54:45 <MC1984_> actually hes a bit scary when talking
403 2013-05-30 13:55:04 <sipa> MC1984_: no offence, but i thought he was a nutcase
404 2013-05-30 13:55:09 <roconnor> MC1984_: kaminsky's comment is what got me thinking about POW again; although I found his prediction laughable.
405 2013-05-30 13:55:23 <MC1984_> yeah he was......animated
406 2013-05-30 13:55:37 <MC1984_> hes not a dumbass though
407 2013-05-30 13:55:58 <sipa> he is not
408 2013-05-30 13:56:09 <roconnor> I'm not entirely sure what qualified him to be up on this particular panel though.
409 2013-05-30 13:56:15 <nsh> the main thing i took away was that a diverse collection of different hashtypes akin to a "basket of currencies" may be a very good thing
410 2013-05-30 13:56:28 <MC1984_> cos he initially talked so much shit about bitcoin a few years ago?
411 2013-05-30 13:56:34 <nsh> in the future, not necessarily within bitcoin itself, but as part of the larger digital commodity economy
412 2013-05-30 13:56:37 <MC1984_> i think he cares about it much more than he wants to let on
413 2013-05-30 13:56:41 <sipa> but i think he sometimes needs to make outrageous claims
414 2013-05-30 13:56:45 <sipa> it's his job
415 2013-05-30 13:56:49 <sipa> or his character
416 2013-05-30 13:56:54 <bitanarchy> anyone familiar with hardware wallet companies with shares?
417 2013-05-30 13:57:05 <phantomcircuit> sipa, both
418 2013-05-30 13:57:18 <phantomcircuit> also i laughed when he said he couldn't hack it
419 2013-05-30 13:57:21 <roconnor> I've been trying to learn to ignore "expert" predictions.
420 2013-05-30 13:57:25 <phantomcircuit> cause at the time i was aware of at least 1 serious issue
421 2013-05-30 13:57:31 <phantomcircuit> which has since been fixed of course
422 2013-05-30 13:58:43 <roconnor> I seem to recall Bruce Schiender predicted that there would be a collision found in sha-1 within the year ... a few years ago.
423 2013-05-30 13:59:35 <nsh> the future is notoriously one of hardest segments of time to make accurate predictions about
424 2013-05-30 13:59:51 <roconnor> sipa: so by closeness you mean avoiding work on a block level that has already been completed but you haven't heard of yet?
425 2013-05-30 13:59:53 <sipa> lol
426 2013-05-30 13:59:54 <jgarzik> all my predictions come true
427 2013-05-30 14:00:03 <jgarzik> of course, I only make predictions on obvious, sure things ;p
428 2013-05-30 14:00:03 <sipa> roconnor: yes
429 2013-05-30 14:00:27 <jgarzik> Dan Kam is backing away from my 10.0 BTC bet, that PoW algo will remain
430 2013-05-30 14:00:32 <roconnor> sipa: do you really think that delay is signficant?
431 2013-05-30 14:00:42 <roconnor> jgarzik: *l* good bet.
432 2013-05-30 14:00:48 <sipa> roconnor: not with current block sizes
433 2013-05-30 14:01:02 <helo> jgarzik: :D
434 2013-05-30 14:01:07 <nsh> jgarzik, why is OP_2ROT (in particular) in the python-bitcoinlib TODO? are the other stack opcodes implemented?
435 2013-05-30 14:01:16 <bitanarchy> what kind of device does bitcoin trezor use?
436 2013-05-30 14:01:21 <roconnor> sipa: as in larger block sizes will need more validation effort?
437 2013-05-30 14:01:49 <sipa> roconnor: yes, transfer/validation
438 2013-05-30 14:02:05 <roconnor> sipa: as in one day the amount of time to transfer a block will be dwarfed by the time needed to validate the block?
439 2013-05-30 14:02:52 <jgarzik> nsh, just last on the list, and nobody seems to use it, so no failed tests ;p
440 2013-05-30 14:02:58 <roconnor> I now understand your idea about POW demonstrating validation speed.
441 2013-05-30 14:02:59 <nsh> fair enough
442 2013-05-30 14:03:08 <jgarzik> nsh, all other working opcodes are implemented -- pynode can already validate mainnet chain, and testnet3 chain
443 2013-05-30 14:03:08 <nsh> seems a bit abstruse
444 2013-05-30 14:03:09 <helo> isn't the average amount of time needed to validate a block relatively consistent?
445 2013-05-30 14:03:18 <jgarzik> nsh, pynode does full verification already
446 2013-05-30 14:03:18 <nsh> jgarzik, great
447 2013-05-30 14:03:28 <nsh> ACTION nods
448 2013-05-30 14:03:43 <roconnor> helo: it should be proportional to the number of coins spent I would think.
449 2013-05-30 14:03:53 <roconnor> more technically the number of signatures that need to be checked.
450 2013-05-30 14:04:47 <nsh> (where time is measured in CPU cycles?)
451 2013-05-30 14:05:42 <roconnor> nsh: well I'm thinking of validating on a fixed computer, so CPU cycles should be proportional to wall-clock time.
452 2013-05-30 14:06:19 <nsh> roconnor, there may be other bottlenecks depending on implementation
453 2013-05-30 14:06:42 <nsh> disk I/O, particularly
454 2013-05-30 14:07:48 <bitanarchy> Is anyone from the trezor team here?
455 2013-05-30 14:08:45 <MC1984_> NIN reference?
456 2013-05-30 14:08:45 <MC1984_> the hell does trezor mean any way
457 2013-05-30 14:08:54 <bitanarchy> its a hardware wallet
458 2013-05-30 14:08:54 <nsh> MC1984_, http://bitcointrezor.com/
459 2013-05-30 14:08:57 <nsh> (hardware wallet)
460 2013-05-30 14:09:04 <MC1984_> i know what it is
461 2013-05-30 14:09:11 <MC1984_> it has a strange name
462 2013-05-30 14:09:13 <bitanarchy> it should be open source software...
463 2013-05-30 14:09:22 <nsh> bitanarchy, i don't recall seeing either of their handles (stick/slush) on irc, but they may use different nicks
464 2013-05-30 14:10:56 <bitanarchy> i am wondering of these hardware wallet devices are already on the market or custom made
465 2013-05-30 14:11:36 <jgarzik> nsh, slush visits #bitcoin-dev
466 2013-05-30 14:11:46 <nsh> ah, my bad
467 2013-05-30 14:11:49 <jgarzik> he runs mining.bitcoin.cz
468 2013-05-30 14:12:15 <nsh> right
469 2013-05-30 14:13:37 <tumak> MC1984_: http://www.youtube.com/watch?v=c-IxXblfXHs ... this is what trezor means :)
470 2013-05-30 14:15:30 <TheLordOfTime> bitcoin-qt 0.8.2..
471 2013-05-30 14:15:31 <TheLordOfTime> stable?
472 2013-05-30 14:15:40 <TheLordOfTime> like, i won't lose the blockchain or my wallet if i upgrade?
473 2013-05-30 14:15:48 <TheLordOfTime> BlueMatt: ^ also make a saucy version for the PPA!
474 2013-05-30 14:17:44 <MC1984_> why would you lose anything
475 2013-05-30 14:18:00 <MC1984_> ofc backup that wallet any way
476 2013-05-30 14:20:40 <TheLordOfTime> MC1984_: because linux has a habit for breaking things :P
477 2013-05-30 14:20:47 <TheLordOfTime> MC1984_: at least, sometimes.
478 2013-05-30 14:20:53 <TheLordOfTime> MC1984_: fortunately the upgrade went smoothly.
479 2013-05-30 14:21:02 <TheLordOfTime> ... now if only bluematt would make a build in the PPA for saucy.
480 2013-05-30 14:21:12 <TheLordOfTime> (yes i'm persesverating, but i know three bitcoiners who use saucy :/)
481 2013-05-30 14:21:15 <MC1984_> whats saucy
482 2013-05-30 14:21:21 <TheLordOfTime> ubuntu dev build
483 2013-05-30 14:21:29 <TheLordOfTime> i'm basically at this point just harassing BlueMatt to do things :P
484 2013-05-30 14:21:38 <MC1984_> they called it saucy?
485 2013-05-30 14:21:40 <TheLordOfTime> since his email got lost in the ether so i can't email him like he wants.
486 2013-05-30 14:21:45 <TheLordOfTime> saucy salamander i think
487 2013-05-30 14:21:49 <helo> with ubuntu (particularly newest ubuntu) if you're going from self-compiled to a release version, your wallet may be incompatible
488 2013-05-30 14:21:52 <TheLordOfTime> TBH i prefer numeric designations
489 2013-05-30 14:22:12 <TheLordOfTime> helo: which is EXACTLY why BlueMatt needs to launch a 13.10 build onto the PPAs
490 2013-05-30 14:22:17 <TheLordOfTime> so that upgrades work right :P
491 2013-05-30 14:22:18 <MC1984_> christ ubuntu
492 2013-05-30 14:22:18 <MC1984_> how long will it take them to get to "masturbating marsupial"
493 2013-05-30 14:22:44 <TheLordOfTime> saucy salamander is a better name than "speedy scootaloo" which was also suggested >.>
494 2013-05-30 14:22:52 <BlueMatt> TheLordOfTime: yes, yes...
495 2013-05-30 14:22:56 <TheLordOfTime> tbh i want to find the person that recommended that there name so I can beat them with a stick
496 2013-05-30 14:23:06 <TheLordOfTime> but that's just because biased against ponies in my ubuntu.
497 2013-05-30 14:23:15 <BlueMatt> I also broke the 0.8.2 uploads for other versions, so only like half got pushed...
498 2013-05-30 14:23:31 <TheLordOfTime> BlueMatt: the version for precise works afaict :)
499 2013-05-30 14:23:40 <nsh> jgarzik, in coredefs.py, what are the magic literals within NETWORKS = { .... } ? (for mainnet/testnet3) doesn't seem to be used
500 2013-05-30 14:23:45 <TheLordOfTime> ... oh that reminds me, where can i file a feature request for bitcoin-qt?
501 2013-05-30 14:24:01 <TheLordOfTime> a shouldn't-be-that-hard-to-implement feature request.
502 2013-05-30 14:24:12 <kjj> there are no feature requests. just write the code and submit a pull request
503 2013-05-30 14:24:15 <BlueMatt> Ill fix them later today, and do saucy when....I dunno, sometime
504 2013-05-30 14:24:15 <chmod755> TheLordOfTime, i guess you could post it on the forums
505 2013-05-30 14:24:33 <BlueMatt> maybe Ill be inspired, but Im moving this weekend, so...may not get around to it
506 2013-05-30 14:24:39 <TheLordOfTime> chmod755: forums... forums... link?
507 2013-05-30 14:24:40 <TheLordOfTime> :P
508 2013-05-30 14:24:53 <jgarzik> nsh, certainly they are used. Look at the definition of NetMagic. That corresponds to msg_start. grep for 'msg_start'
509 2013-05-30 14:25:05 <chmod755> there should be a place where people can offer btc to get a feature implemented or a bug fixed
510 2013-05-30 14:25:36 <nsh> jgarzik, ah ok, ty
511 2013-05-30 14:25:49 <kjj> you can do that on the forums (bitcointalk.org) or the dev mailing list. bounties don't work very well, but you can sure try
512 2013-05-30 14:26:41 <jgarzik> indeed
513 2013-05-30 14:27:18 <jgarzik> Just about the only bounties that work are stupidly large bounties for quick work
514 2013-05-30 14:28:10 <kjj> also, unless you write software yourself, you should never attempt to estimate the effort required for a feature.
515 2013-05-30 14:28:14 <jgarzik> Problems common to bounties: unclear specifications, how to resolve multiple independent competitors, ...
516 2013-05-30 14:30:00 <kjj> now I'm curious though. what feature do you want?
517 2013-05-30 14:30:41 <nsh> ACTION read yesterday some of the bitcointalk "give us better forum software" bounty thread (no particular progress after one year and (today's) 5 million USD offering
518 2013-05-30 14:31:10 <nsh> proof of coding work (in terms of timestamped coffee stains) should be marked to market
519 2013-05-30 14:31:27 <chmod755> timestamped coffee stains???
520 2013-05-30 14:31:43 <jgarzik> hah
521 2013-05-30 14:31:47 <helo> if escrow was used for that bounty, surely there would be progress at this point
522 2013-05-30 14:32:09 <jgarzik> see above. even with funds in hand, bounties can be problematic.
523 2013-05-30 14:32:13 <kjj> theymos IS escrow
524 2013-05-30 14:33:00 <jgarzik> How to resolve multiple people who want to working on feature X? (obviously there are solutions, but people may not be able to agree amicably) For a non-programmer, it is difficult to even write a specification that CLEARLY defines the conditions to claim the bounty.
525 2013-05-30 14:33:19 <chmod755> helo, .... and maybe when someone agrees to work on it, it should be locked. and there has to be some progress otherwise it should be open for anyone again....
526 2013-05-30 14:34:10 <helo> that sounds like a decent idea
527 2013-05-30 14:34:56 <kjj> this topic was discussed extensively at the conference
528 2013-05-30 14:35:43 <kjj> at one point, Mike, Yifu and a couple of others were discussing it at length and were unable to come up with a workable bounty system
529 2013-05-30 14:36:17 <nsh> you could probably overlay a bounty system onto a set of unit tests using as a customization to a versioning system. but it would require significant up-front work
530 2013-05-30 14:36:23 <kjj> If I recall, they decided to find competent programmers and offer them contract work
531 2013-05-30 14:38:15 <kjj> it isn't that bounties can't work, or haven't worked occasionally in the past. the problem is that they are just not a good model for doing anything serious
532 2013-05-30 14:39:04 <jgarzik> Bounties are the lazy person's way to do contracting
533 2013-05-30 14:39:21 <jgarzik> People have money, but don't want to spend the time putting together a formal spec, finding people to do the work, etc.
534 2013-05-30 14:40:04 <tumak> when you have to think things like formal spec its a lost cause already
535 2013-05-30 14:40:09 <kjj> Jerry Pournelle is a big advocate of bounties for government work, but those in general have less of a specification problem than software
536 2013-05-30 14:40:19 <tumak> i can imagine bounties work only among more or less peers
537 2013-05-30 14:40:26 <tumak> where one outsources to another because of lack of time
538 2013-05-30 14:40:36 <jgarzik> well there are varying definitions of "formal" ;-)
539 2013-05-30 14:40:57 <jgarzik> A big problem with bounties in the past is an unclear spec. Person does some work, wants the money, and then the bounty offer-er says the job is not done
540 2013-05-30 14:41:05 <tumak> kjj: tendering process is a bit different beast though
541 2013-05-30 14:41:18 <tumak> jgarzik: aka "client from hell syndrome"
542 2013-05-30 14:41:25 <tumak> changing their mind continously etc :)
543 2013-05-30 14:41:27 <eps> jgarzik: that is the big problem with all software development
544 2013-05-30 14:41:34 <eps> jgarzik: heard of the software crisis?
545 2013-05-30 14:41:50 <jgarzik> eps: Yes. It's called the Satoshi codebase
546 2013-05-30 14:42:28 <eps> jgarzik: 70% of all large IT projects either aren't delivered, deliver partially or go over budget
547 2013-05-30 14:42:43 <tumak> ACTION calls bullshit
548 2013-05-30 14:43:02 <tumak> eps: s/IT projects/government IT projects/
549 2013-05-30 14:43:43 <kjj> Meh. that sounds about right to me, even without the government qualifier
550 2013-05-30 14:43:48 <eps> yeah
551 2013-05-30 14:43:55 <eps> good software stands out
552 2013-05-30 14:44:04 <eps> like iOS and android
553 2013-05-30 14:44:04 <yubrew> if you could create tests against the bounty, that could help
554 2013-05-30 14:44:17 <yubrew> unit, functional, integration
555 2013-05-30 14:44:21 <eps> everyone is like "oh wow, software doesn't have to be crappy?"
556 2013-05-30 14:45:04 <eps> i am not a blanket fan of agile, but the one thing i like about it is rapid prototyping and regularly shipping code
557 2013-05-30 14:45:15 <tumak> yubrew: if the bounty issuer is actually capable of unit tests, there are rarely problems with such arrangements
558 2013-05-30 14:45:20 <eps> it's the only way i can see that keeps on top of technical debt
559 2013-05-30 14:45:27 <tumak> shit hits the fan when coder meets clueless client from hell
560 2013-05-30 14:45:46 <yubrew> tumak fair enough haha
561 2013-05-30 14:45:49 <kjj> on the other hand, writing software to satisfy tests is rarely more difficult than just writing the tests in the first place
562 2013-05-30 14:46:36 <eps> kjj: yeah often when creating an API, writing the tests takes longer because it forces you to think about how it will be used
563 2013-05-30 14:47:07 <tumak> well, unit tests check only small components of the project
564 2013-05-30 14:47:15 <tumak> to "test" it as large, you actually use regression tests
565 2013-05-30 14:47:27 <tumak> ie collected cases from past which segfaulted your app
566 2013-05-30 14:47:53 <yubrew> don't really know testing well
567 2013-05-30 14:48:14 <yubrew> only familiar with web app testing (e.g. unit, functional, integration, behavioral)
568 2013-05-30 14:49:35 <tumak> hm, webdesign ut?
569 2013-05-30 14:49:39 <tumak> you mean like, uh, acid test? :)
570 2013-05-30 14:50:02 <yubrew> don't know what acid is, but sounds like a hell of a drug :)
571 2013-05-30 14:50:18 <tumak> http://acid3.acidtests.org/
572 2013-05-30 14:50:22 <tumak> it shows how high you are :)
573 2013-05-30 14:51:28 <jgarzik> web app testing? Is this #php-user-dev? :)
574 2013-05-30 14:51:45 <tumak> jgarzik: of course
575 2013-05-30 14:51:57 <tumak> you test if your web design is pixel-exact across browsers!
576 2013-05-30 14:52:21 <yubrew> looking for something small to do on bitcoin
577 2013-05-30 14:52:40 <yubrew> catch is, i know languages that are irrelevant to bitcoin atm haha
578 2013-05-30 15:03:54 <Luke-Jr> Goonie_: not specifically for 0.8.2, no
579 2013-05-30 15:06:56 <owowo> was gribble hacked?
580 2013-05-30 15:09:04 <nsh> owowo, what makes you ask?
581 2013-05-30 15:09:36 <owowo> there is a altgribble in #bitcoin-analysis
582 2013-05-30 15:09:49 <owowo> <altgribble> forgot: lolol r00t wuz hurr
583 2013-05-30 15:09:49 <owowo> <forgot> ;;bc,stats
584 2013-05-30 15:10:03 <owowo> doing things like this
585 2013-05-30 15:10:29 <runeks_> Is there a way to limit the memory usage of bitcoind?
586 2013-05-30 15:10:36 <nsh> runeks_, rlimit
587 2013-05-30 15:11:05 <sipa> lol
588 2013-05-30 15:11:24 <sipa> "is there a way to secure my computer from hackers?" - "unplug ethernet cable"
589 2013-05-30 15:11:33 <owowo> true
590 2013-05-30 15:11:36 <sipa> runeks_: not in any way you'd like
591 2013-05-30 15:11:41 <runeks_> :(
592 2013-05-30 15:11:50 <sipa> runeks_: which version are you running?
593 2013-05-30 15:11:54 <runeks_> It gets killed consuming 572740kB of memory.
594 2013-05-30 15:12:04 <nsh> that's like half a facebook tab
595 2013-05-30 15:12:15 <runeks_> sipa: 0.8.1
596 2013-05-30 15:12:19 <sipa> try 0.8.2
597 2013-05-30 15:12:26 <sipa> memory usage is reduced significantly
598 2013-05-30 15:12:32 <sipa> especially with many network connections
599 2013-05-30 15:12:47 <sipa> though i can't promise it will stay under 500 MB
600 2013-05-30 15:13:13 <runeks_> sipa: Cool. I'll try that. I don't think it has any network connections at that point though. It should be busy building blockchain/UTXO index.
601 2013-05-30 15:17:02 <yubrew> runeks_ i found increasing the swap also worked
602 2013-05-30 15:17:17 <runeks_> yubrew: Is it fast enough then?
603 2013-05-30 15:18:21 <yubrew> fast in what way?
604 2013-05-30 15:18:47 <runeks_> yubrew: Are you running bitcoind on the Raspberry Pi?
605 2013-05-30 15:18:57 <yubrew> no i am not haha
606 2013-05-30 15:19:03 <runeks_> Ok.
607 2013-05-30 15:19:08 <yubrew> i'm running on a micro instance of ec2
608 2013-05-30 15:19:13 <runeks_> That's my plan.
609 2013-05-30 15:19:16 <runeks_> yubrew: How much RAM?
610 2013-05-30 15:19:23 <yubrew> 500mb
611 2013-05-30 15:19:31 <yubrew> try upgrading
612 2013-05-30 15:19:33 <yubrew> first
613 2013-05-30 15:19:41 <runeks_> Upgrading what?
614 2013-05-30 15:19:46 <runeks_> Oh bitcoind?
615 2013-05-30 15:19:53 <yubrew> 0.8.2
616 2013-05-30 15:19:57 <yubrew> but i think increasing swap is a more permanent solution
617 2013-05-30 15:20:22 <yubrew> do you have a compiler error?
618 2013-05-30 15:20:35 <nsh> (starting your conference talk by reading a massive wall of text in a dull monotone: not commonly advised :/)
619 2013-05-30 15:20:42 <runeks_> Yeah I'm building 0.8.2 now.
620 2013-05-30 15:20:42 <yubrew> something like this: 'internal compiler error: Killed (program cc1plus)'
621 2013-05-30 15:20:47 <nsh> (re: https://www.youtube.com/watch?v=8Od4tV88oR8&list=PLUOP0P68GJ3BGjfqoLLnzAefk3ZzXQtJ7 )
622 2013-05-30 15:20:59 <runeks_> yubrew: I'm cross compiling it on my x86 system. So it's fine.
623 2013-05-30 15:21:04 <yubrew> ah ok
624 2013-05-30 15:21:11 <kjj> generally, compiling takes more memory than running
625 2013-05-30 15:21:17 <runeks> But I've seen that error before when building on the Pi.
626 2013-05-30 15:21:27 <TheLordOfTime> yubrew: how much data usage does bitcoind have on that ec2 micro instance?
627 2013-05-30 15:21:33 <runeks> kjj: Especially when you have a single file with 12MB C++ code like Armory does :)
628 2013-05-30 15:21:52 <TheLordOfTime> and CPU usage, too... :P
629 2013-05-30 15:22:03 <yubrew> haha, yes i compiled the damn thing on a micro.. not the best of ideas
630 2013-05-30 15:22:09 <yubrew> https://gist.github.com/yubrew/5572930
631 2013-05-30 15:22:20 <kjj> I keep an ancient box running, and somewhere around 0.6 or 0.7 I had to upgrade it from 512MB to 1GB otherwise the compiles would take days to finish
632 2013-05-30 15:22:20 <yubrew> this is what i did to increase swap, and it worked fine afterwards
633 2013-05-30 15:23:33 <runeks> yubrew: Yeah I had to do that too when building Armory. But I'd really prefer to not have to use a swap file. The Pi is slow enough as it is :\\
634 2013-05-30 15:23:47 <runeks> Got 0.8.2 built. Gonna try it out now.
635 2013-05-30 15:24:00 <yubrew> let us know how it goes
636 2013-05-30 15:24:53 <kjj> well, I'm out for a week again. feels like I just got back to work after the bitcoin conference, and now I'm off to a work conference
637 2013-05-30 15:25:55 <runeks> yubrew: Will do.
638 2013-05-30 15:45:01 <dansmith_btc> Hi, I'm generating pub/priv key using https://github.com/weex/addrgen It gives me a priv key like L4C1FUabYPBRJ8pBzzqtxuhGPT89adyA9mZqAVuJ6n1rkYB562b9 . I need to convert it to WIF, so I need to know what is the name of such a priv key format?
639 2013-05-30 15:45:19 <sipa> dansmith_btc: it _is_ WIF
640 2013-05-30 15:45:34 <sipa> if your client rejects it, it means they don't support compressed pubkeys
641 2013-05-30 15:45:41 <dansmith_btc> sipa, why doesnt it start with "5"?
642 2013-05-30 15:45:57 <sipa> because only WIF privkeys for uncompressed pubkeys start with 5
643 2013-05-30 15:46:19 <sipa> and converting 9though trivial) it doesn't make sense, as that would change the associated address
644 2013-05-30 15:47:36 <dansmith_btc> blockchain.info requires uncompressed apparently Method: importprivkey
645 2013-05-30 15:47:36 <dansmith_btc> Description: Import a private key into your bitcoin wallet. Private key must be in wallet import format (Sipa) beginning with a '5'.
646 2013-05-30 15:47:36 <dansmith_btc> Parameters: (String privateKey)
647 2013-05-30 15:48:11 <gmaxwell> dansmith_btc: welp, sounds like blockchain.info is really limited, you should nag your vendor to fix their support
648 2013-05-30 15:48:28 <gmaxwell> dansmith_btc: it's not possible to meaninfully 'convert' that.
649 2013-05-30 15:48:47 <gmaxwell> It's alread in WIF and the site you are using apparently doesn't convert the type of key it is.
650 2013-05-30 15:48:57 <sipa> s/convert/support/
651 2013-05-30 15:49:10 <gmaxwell> right.
652 2013-05-30 15:49:31 <sipa> wow, i didn't know b.i didn't even support compressed pubkeys
653 2013-05-30 15:49:49 <dansmith_btc> is there a python function which generates pub/uncompressed priv key pair?
654 2013-05-30 15:50:26 <sipa> i'm sure there are many, but you really shouldn't be using uncompressed keys anymore :(
655 2013-05-30 15:50:51 <sipa> (i know, compatibility with certain sites is a good reason, but please get them to support compressed ones instead...)
656 2013-05-30 15:51:16 <dansmith_btc> sipa, if b.i uses bitcoind which accepts WIFs by defult, why would they change that to uncompressed?
657 2013-05-30 15:51:35 <sipa> they're not using bitcoind
658 2013-05-30 15:51:51 <dansmith_btc> ok
659 2013-05-30 15:51:58 <gmaxwell> It's really troubling that popular services aren't keeping up with bitcoin ecosystem changes from, what, 2 years ago?
660 2013-05-30 15:51:59 <sipa> at least not for their wallet implementation
661 2013-05-30 15:52:06 <tumak> some lovely tracebacks now and then indicate something java based
662 2013-05-30 15:52:36 <jgarzik> "If It Aint Broke Dont Fix It" status quo momentum is powerful
663 2013-05-30 15:52:49 <jgarzik> plus I imagine blockchain.info guy is busy making double-spend gadgets
664 2013-05-30 15:53:11 <sipa> Compressed pubkeys
665 2013-05-30 15:53:11 <sipa> Date: Mon Nov 21 02:46:28 2011 +0100