1 2012-02-26 00:00:04 <davex__> oh i see.
  2 2012-02-26 00:00:15 <sipa> When encrypting the wallet, the unencrypted versions of the new-style keys (with compressed
  3 2012-02-26 00:00:18 <davex__> so, the unencrypted keys just weren't getting removed before.
  4 2012-02-26 00:00:25 <sipa> pubkey, introduced in 0.6.0) were not removed.
  5 2012-02-26 00:01:33 <sipa> How many \03key lines did you have?
  6 2012-02-26 00:01:53 <davex__> there are 3 of them
  7 2012-02-26 00:02:04 <davex__> lots of \04ckey's
  8 2012-02-26 00:02:21 <sipa> Ok, good.
  9 2012-02-26 00:02:42 <sipa> Remove the \03key lines, plus the ones that follow them.
 10 2012-02-26 00:02:50 <sipa> Do the same for \04pool lines.
 11 2012-02-26 00:03:09 <davex__> done
 12 2012-02-26 00:03:12 <davex__> oh
 13 2012-02-26 00:03:30 <davex__> and those are for...?
 14 2012-02-26 00:03:46 <sipa> (reason for this: the public keys for the encrypted compressed keys, are wrong; you don't want to use them, but they are only reserve keys now, so just throw them out so they won't be used)
 15 2012-02-26 00:04:14 <davex__> ok, so each  \04pool line and the line that follows it
 16 2012-02-26 00:04:15 <davex__> ?
 17 2012-02-26 00:04:26 <sipa> The proper way is looking which 3 \04ckey lines correspond to the removed \03key lines, and removing those.
 18 2012-02-26 00:04:41 <sipa> But that's a lot more work than just flushing the keypool (removing the pool lines)
 19 2012-02-26 00:04:47 <davex__> oh
 20 2012-02-26 00:04:47 <sipa> Yes.
 21 2012-02-26 00:05:15 <davex__> ok done
 22 2012-02-26 00:05:48 <sipa> Ok, now do: db4.8_load -f yourdumpfile wallet.dat.new
 23 2012-02-26 00:08:16 <davex__> there we go.  Thanks a lot.
 24 2012-02-26 01:11:55 <gribble> New news from bitcoinrss: sipa opened pull request 900 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/900>
 25 2012-02-26 01:27:40 <C0ldDecember2012> anyone having trouble synching with the network on 0.5.2 mac os x?
 26 2012-02-26 01:28:15 <gmaxwell> C0ldDecember2012: Can you be more specific? What behavior are you seeing?
 27 2012-02-26 01:29:28 <C0ldDecember2012> im opening bitcoin-qt that i downloaded from the downloads section... its just the .dmg... i have the blue bar at the bottom making very little progress with standard preferences... i left it for two hours and still wasn't close to syncing
 28 2012-02-26 01:29:36 <C0ldDecember2012> synching*
 29 2012-02-26 01:30:19 <nanotube> how many total blocks do you have so far, C0ldDecember2012
 30 2012-02-26 01:30:31 <nanotube> it does take a while for the initial blockchain download/check
 31 2012-02-26 01:30:45 <gmaxwell> Depending on your disk speed and other factors it can take a day to completely sync. Also, the indicator can be a bit misleading because it always starts from 0%.
 32 2012-02-26 01:30:47 <nanotube> there are currently ,,bc,blocks blocks, so you get to wait until you have them all.
 33 2012-02-26 01:30:48 <gribble> 168424
 34 2012-02-26 01:31:28 <luke-jr> C0ldDecember2012: initial as in, the first time you run it; it won't take that long every time
 35 2012-02-26 01:31:29 <gmaxwell> nanotube: I'm really not a fan of -qt's behavior of hiding those counts. Makes techsupport a lot harder.
 36 2012-02-26 01:31:49 <C0ldDecember2012> ok it doesmt show blocks
 37 2012-02-26 01:31:52 <nanotube> if you put your mouse over it, it shows, gmaxwell
 38 2012-02-26 01:31:54 <nanotube> C0ldDecember2012: ^
 39 2012-02-26 01:32:03 <nanotube> gmaxwell: that said yea, it'd be better if they were shown by default
 40 2012-02-26 01:32:09 <gmaxwell> nanotube: I know, but ^ whats what I was refering to.
 41 2012-02-26 01:32:17 <gmaxwell> :)
 42 2012-02-26 01:32:27 <nanotube> ????
 43 2012-02-26 01:32:36 <gmaxwell> wumpus: are you actually opposed to showing those status things?
 44 2012-02-26 01:34:26 <C0ldDecember2012> ok let me try this way
 45 2012-02-26 01:34:45 <gmaxwell> C0ldDecember2012: please mouseover the progress indicator and tell us how many blocks you have.
 46 2012-02-26 01:35:22 <C0ldDecember2012> ok one min i quit out of it... n00b impatient mistake... you can tell im a mac person :P
 47 2012-02-26 01:35:40 <gmaxwell> hah
 48 2012-02-26 01:36:42 <C0ldDecember2012> it says 134186 blocks downloaded upon just opening it
 49 2012-02-26 01:36:56 <gmaxwell> okay. Is the number increasing?
 50 2012-02-26 01:37:16 <gmaxwell> (might take a few seconds to start increasing again since you just restarted it)
 51 2012-02-26 01:37:23 <C0ldDecember2012> yeah 134225
 52 2012-02-26 01:37:30 <C0ldDecember2012> so im actually almost there
 53 2012-02-26 01:37:56 <C0ldDecember2012> 165000~ blocks of transaction history
 54 2012-02-26 01:39:00 <gmaxwell> Well the early blocks go much faster there isn't much in them. It may still take you a couple hours to get fully synced up. After that you should resync quickly.
 55 2012-02-26 01:39:19 <C0ldDecember2012> thanks guys!
 56 2012-02-26 06:41:07 <drkurubit> hows everyone doing
 57 2012-02-26 07:38:17 <wumpus> gmaxwell: yes, I don't want those big numbers in the main interface, having them in a popup or special window is ok
 58 2012-02-26 07:39:48 <gmaxwell> wumpus: maybe some about dialog that gives some troubleshooting info all in one place?
 59 2012-02-26 07:39:59 <wumpus> the idea is that most people, most of the time are not doing tech support
 60 2012-02-26 07:40:11 <wumpus> yes that's great
 61 2012-02-26 07:40:44 <wumpus> I also want to add a rpc console
 62 2012-02-26 07:41:19 <gmaxwell> the hover thing is really hard to communicate, even when there isn't a language barrier (but when there is we might as well give up. We had to ask someone over a dozen times the other day). Perhaps the dialog will be easier to get people to try.
 63 2012-02-26 07:41:30 <gmaxwell> wumpus: along with RPC console: log viewer.
 64 2012-02-26 07:41:41 <wumpus> yes
 65 2012-02-26 07:42:54 <wumpus> aside from in the menu, it should probably open when you click on one of the status icons in the status bar
 66 2012-02-26 07:44:06 <gmaxwell> Good call.
 67 2012-02-26 07:45:42 <wumpus> let's try to get those in for 0.7.0, I don't want anything to hold up 0.6.0 any further :)
 68 2012-02-26 07:48:53 <gmaxwell> Yea, I wasn't thinking of anything for 0.6  though if I get too many more users that take a dozen requests to get their block count, I'm going to start stabbing them.
 69 2012-02-26 10:26:29 <gribble> New news from bitcoinrss: mad opened issue 901 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/issues/901>
 70 2012-02-26 14:11:35 <kish> make -f makefile.unix
 71 2012-02-26 14:11:36 <kish> g++ -c -pthread -Wextra -Wno-sign-compare -Wno-char-subscripts -Wno-invalid-offsetof -Wformat-security -g -DNOPCH -I/home/ly/bitcoin2/bitcoin/src -DUSE_UPNP=0 -DUSE_SSL -fno-stack-protector -fstack-protector-all -Wstack-protector -Wl,-z,relro -Wl,-z,now -D_FORTIFY_SOURCE=2 -O2 -MMD -o obj/net.o net.cpp
 72 2012-02-26 14:11:41 <kish> net.cpp:18:32: fatal error: miniupnpc/miniwget.h: No such file or directory
 73 2012-02-26 14:11:43 <kish> compilation terminated.
 74 2012-02-26 14:11:46 <kish> make: *** [obj/net.o] Error 1
 75 2012-02-26 14:11:49 <kish> why the fuck do i get that
 76 2012-02-26 14:12:07 <sipa> Because you don't have libminiupnpc installed.
 77 2012-02-26 14:12:29 <sipa> You can compile using USE_UPNP= to not use it.
 78 2012-02-26 14:13:02 <kish> seems to work
 79 2012-02-26 14:13:12 <kish> thanks :)
 80 2012-02-26 14:13:20 <sipa> See doc/build-unix.txt
 81 2012-02-26 14:13:28 <kish> i skimmed it
 82 2012-02-26 14:42:46 <etotheipi_> sipa, how is proof-of-ownership implemented in 0.6?  I want to create a similar challenge-response interface for Armory, but I'm not sure how exactly I want to implement it
 83 2012-02-26 14:43:51 <etotheipi_> I feel like, if I just use a random 16 bytes, you open up the slim possibility that the fake-owner will change the challenge to another for which he has seen the signature and the challenger won't notice the difference
 84 2012-02-26 14:44:43 <sipa> Proof-of-ownership?
 85 2012-02-26 14:44:48 <sipa> You mean message signing?
 86 2012-02-26 14:45:04 <etotheipi_> I thought 0.6 was implementing a method for proving that you own a particular address
 87 2012-02-26 14:45:21 <sipa> Yes, you can sign a message using a key associated with a certain address.
 88 2012-02-26 14:45:27 <etotheipi_> so I'm envisioning a manual challenge-response dialog
 89 2012-02-26 14:45:43 <sipa> But if you want to use it for proof-of-ownership, you need more, like guaranteeing freshness.
 90 2012-02-26 14:46:15 <etotheipi_> okay, so 0.6 only provides signing and verification of arbitrary key data?
 91 2012-02-26 14:46:20 <sipa> Yes.
 92 2012-02-26 14:46:32 <sipa> (0.5 too, by the way, just not in the GUI)
 93 2012-02-26 14:46:56 <etotheipi_> so I'm trying to go one step further and provide a canned challenge response... which would require the challenger to provide a unique string to sign
 94 2012-02-26 14:47:15 <sipa> For example, you'll typically want to add the date to the message.
 95 2012-02-26 14:47:53 <etotheipi_> "Make someone prove they own an address:  give them this" : "----Begin-Challenge---- Addr:  1Ghke34ds... Challenge: Feb 28, 12:30am ----End-Challenge----"
 96 2012-02-26 14:48:46 <sipa> It'd be nice if you could implement the same low-level signing function, but it may be a bit challenging, as it uses key recovery.
 97 2012-02-26 14:48:50 <etotheipi_> maybe that's all I need, is just to plug in a high-res date
 98 2012-02-26 14:49:35 <etotheipi_> sipa:  http://dl.dropbox.com/u/1139081/dlgBtcCalc_reduced.png
 99 2012-02-26 14:50:12 <etotheipi_> I just use Bitcoin-like signing:  double-sha256 for the hash before ECDSA signature
100 2012-02-26 14:50:41 <etotheipi_> so advanced users who know what they're doing can use this interface for challenge-response, I want to provide a more-direct form of it
101 2012-02-26 14:51:02 <etotheipi_> and, if the challenge- and response-packets can be made to be interoperable, that would be great
102 2012-02-26 14:52:10 <sipa> etotheipi_: just looked it up
103 2012-02-26 14:52:28 <sipa> etotheipi_: bitcoin's signmessage function uses a 65-byte signature, encoded as 120 hex characters
104 2012-02-26 14:52:46 <etotheipi_> 130 hex chars?
105 2012-02-26 14:52:51 <sipa> Yes.
106 2012-02-26 14:53:01 <sipa> the first byte is 27 + a 3-bit flag.
107 2012-02-26 14:53:20 <sipa> Then follows r is 32 bytes (big endian), then s in 32 bytes (big endian).
108 2012-02-26 14:55:37 <sipa> The 1-bit tells whether the signer's pubkey is odd or even, the 2-bit is used to distinguish the two possible keys in case they are two possible recoveries, and 4-bit says whether the pubkey is compressed or not.
109 2012-02-26 14:57:33 <sipa> See section 4.1.6 in the SEC specification for more information :)
110 2012-02-26 14:58:09 <etotheipi_> gah... I don't have compressed public keys yet
111 2012-02-26 14:58:24 <sipa> Doesn't matter, in that case bit 4 is always 0 :)
112 2012-02-26 14:59:32 <etotheipi_> So I was thinking of something like this:  http://pastebin.com/1u5KxvDR
113 2012-02-26 14:59:44 <etotheipi_> that gives a straightforward ASCII way to send these through email
114 2012-02-26 15:00:34 <sipa> The reason for the flag byte is that it allows key recovery, so you don't need to include the public key in the message.
115 2012-02-26 15:00:37 <sipa> *response
116 2012-02-26 15:00:54 <sipa> (nor the address)
117 2012-02-26 15:02:22 <etotheipi_> oooh, right
118 2012-02-26 15:02:29 <etotheipi_> to recover the public key from the signature
119 2012-02-26 15:02:42 <sipa> Yes, that's key recovery.
120 2012-02-26 15:02:53 <sipa> https://github.com/bitcoin/bitcoin/blob/master/src/key.cpp#L45
121 2012-02-26 15:03:07 <etotheipi_> why did we switch to compressed public keys when we could've just used 1 byte + key recovery?
122 2012-02-26 15:03:30 <sipa> It has nothing to do with compressed public keys.
123 2012-02-26 15:03:36 <sipa> Key recovery is always possible.
124 2012-02-26 15:03:49 <sipa> Oh, I misread your remark.
125 2012-02-26 15:03:54 <etotheipi_> I know... so why bother even including a publick ey in the tx at all?
126 2012-02-26 15:03:56 <sipa> The reason is backward compatibility.
127 2012-02-26 15:04:22 <etotheipi_> oh... I forget that there are 3 years of previous clients to worry about
128 2012-02-26 15:04:24 <etotheipi_> :)
129 2012-02-26 15:05:13 <sipa> http://people.xiph.org/~greg/addr.compare.html
130 2012-02-26 15:06:50 <etotheipi_> what is the cost of key recovery compared to, say, verification?
131 2012-02-26 15:07:15 <etotheipi_> oh, I guess I could count the EC multiplies in the code you referenced :)
132 2012-02-26 15:07:22 <sipa> 5% slower
133 2012-02-26 15:07:53 <sipa> Anyway, got to go.
134 2012-02-26 15:07:55 <etotheipi_> so key recovery approximately doubles verification time?
135 2012-02-26 15:08:03 <etotheipi_> okay, thanks sipa
136 2012-02-26 15:08:04 <sipa> No, it increases it with 5%
137 2012-02-26 15:08:16 <etotheipi_> so it basically costs nothing to include key recovery?
138 2012-02-26 15:08:25 <sipa> You don't need to verify afterwards anymore; just compare the recovered pubkey to the claimed address.
139 2012-02-26 15:08:31 <etotheipi_> ooh
140 2012-02-26 15:08:33 <etotheipi_> cool
141 2012-02-26 15:10:30 <etotheipi_> for later:  I still wouldn't mind making a challenge-response packet like this... regardless of whether it includes compressed/uncompressed/no public keys
142 2012-02-26 15:10:55 <etotheipi_> Maybe we can standardize on it, somehow...
143 2012-02-26 15:11:48 <kish> where is satoshi
144 2012-02-26 15:17:18 <gribble> New news from bitcoinrss: laanwj opened pull request 902 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/902>
145 2012-02-26 15:28:43 <badluck> satoshi is in a safe house in belize
146 2012-02-26 16:23:39 <C0ldDecember2012> hey guys... is there a special stop command for diablominer on mac osx?
147 2012-02-26 16:23:52 <Diablo-D3> no, there is a generic command
148 2012-02-26 16:23:55 <Diablo-D3> press control-c
149 2012-02-26 16:24:17 <C0ldDecember2012> mhash: 1.7/1.6 | accept: 1 | reject: 0 | hw error: 0
150 2012-02-26 16:24:18 <C0ldDecember2012> i just get this over and over again
151 2012-02-26 16:24:41 <Diablo-D3> wow, thats pretty slow dude
152 2012-02-26 16:25:01 <Diablo-D3> what gpu?
153 2012-02-26 16:25:39 <C0ldDecember2012> haha yeah im just trying to see if mining is actually an option NVidea 9400m which i know is bad... but im my thought is i can run it before going to bed and at least do something with my comp while im not using it
154 2012-02-26 16:26:13 <C0ldDecember2012> so do i just wait till its done hashing?
155 2012-02-26 16:26:39 <Diablo-D3> [12:23:54] <Diablo-D3> press control-c
156 2012-02-26 16:27:26 <C0ldDecember2012> gotcha
157 2012-02-26 16:28:09 <C0ldDecember2012> so is that just an error telling me its really slow?
158 2012-02-26 16:28:16 <Diablo-D3> no
159 2012-02-26 16:28:20 <Diablo-D3> thats a persistent status line
160 2012-02-26 16:28:30 <Diablo-D3> you've done something wrong if its scrolling
161 2012-02-26 16:29:28 <C0ldDecember2012> ok on startup do i just leave it?
162 2012-02-26 16:29:37 <Diablo-D3> yeah.
163 2012-02-26 16:29:38 <Diablo-D3> also
164 2012-02-26 16:29:41 <Diablo-D3> do me a favor
165 2012-02-26 16:29:45 <C0ldDecember2012> ok
166 2012-02-26 16:29:46 <Diablo-D3> try -na to see if its faster
167 2012-02-26 16:30:44 <C0ldDecember2012> whats that? control -na? or do i type it on startup?
168 2012-02-26 16:31:03 <Diablo-D3> add that to the arguments
169 2012-02-26 16:32:57 <C0ldDecember2012> how do i add arguments (lost)
170 2012-02-26 16:33:52 <C0ldDecember2012> ok so all ive managed to do is my initial configuration with the host my machine and my miner
171 2012-02-26 16:35:44 <Diablo-D3> you type them after ./DiabloMiner-OSX.sh
172 2012-02-26 16:36:35 <C0ldDecember2012> ok
173 2012-02-26 16:37:17 <C0ldDecember2012> i started the app in finder so that is probably why i don't understand configuration in command lines... let me try to open in terminal and go from there
174 2012-02-26 16:37:47 <Diablo-D3> DiabloMiner isnt meant to be ran the way you did it.
175 2012-02-26 16:38:50 <C0ldDecember2012> yeah its one of those (theres your problem) moments for me
176 2012-02-26 16:39:15 <C0ldDecember2012> im reading in threads about the proper way to do it for mac os x... and i thought i kinda was tech savy lol
177 2012-02-26 16:48:57 <C0ldDecember2012> thanks for the help guys im going to learn more and im sure ill be back asking the pros for more help.... honestly this was more help than any threads i found
178 2012-02-26 19:15:39 <luke-jr> w00t, I confused github! https://github.com/luke-jr/eloipool/network
179 2012-02-26 19:16:33 <RedEmerald> whats wrong with it?
180 2012-02-26 19:18:16 <luke-jr> http://luke.dashjr.org/tmp/screenshots/snapshot75.png
181 2012-02-26 19:18:26 <luke-jr> check out the green branch
182 2012-02-26 19:35:52 <graingert> luke-jr: it's 404
183 2012-02-26 19:36:02 <luke-jr> graingert: see 2nd link
184 2012-02-26 19:36:09 <graingert> that seems a little odd
185 2012-02-26 19:37:08 <graingert> luke-jr: what does git-cola make of it?
186 2012-02-26 19:37:17 <luke-jr> git-cola?
187 2012-02-26 19:37:23 <graingert> a git gui
188 2012-02-26 19:37:51 <graingert> it has a branch visualizer too
189 2012-02-26 19:41:10 <luke-jr> not without gtk support in git
190 2012-02-26 19:41:32 <luke-jr> or tk
191 2012-02-26 19:52:49 <gribble> New news from bitcoinrss: dooglus opened issue 903 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/issues/903>
192 2012-02-26 19:53:29 <superjames> needa tad of help with json api
193 2012-02-26 19:54:16 <superjames> {"method":"listtransactions","params":[],"id":92141}
194 2012-02-26 19:54:19 <superjames> ^ works
195 2012-02-26 19:54:28 <superjames> {"method":"listtransactions","params":{"count":10,"from":0},"id":905137}
196 2012-02-26 19:54:30 <superjames> ^ doesnt
197 2012-02-26 20:02:49 <forrestv> superjames, arguments aren't named. do params=[10, 0]
198 2012-02-26 20:07:23 <superjames> hmm that does not work
199 2012-02-26 20:07:25 <superjames> the wiki says that [account] (as the first argument) is not mandatory
200 2012-02-26 20:07:42 <superjames> maybe it needs to be used when you want to use the count, from keys
201 2012-02-26 20:10:21 <forrestv> superjames, then supply '*' for account
202 2012-02-26 20:10:25 <forrestv> you can't skip arguments
203 2012-02-26 20:10:44 <superjames> ahh wildcards
204 2012-02-26 20:11:27 <superjames> works, thanks mate
205 2012-02-26 21:08:41 <copumpkin> are there any clients that support multisends from a csv spec?
206 2012-02-26 21:08:59 <copumpkin> so you can just have a big csv of what addresses you're sending what amounts to
207 2012-02-26 21:14:22 <Karmaon> copumpkin: just parse the csv line by line and pass the dictionary to bitcoind
208 2012-02-26 21:16:40 <copumpkin> Karmaon: sure, but I was just wondering if anything supported it directly
209 2012-02-26 21:46:18 <nanotube> copumpkin: directly, no. only the parser you'll make yourself. :)
210 2012-02-26 21:46:44 <copumpkin> hah, was just trying to help pirate out :)
211 2012-02-26 21:50:22 <gribble> New news from bitcoinrss: sipa opened pull request 904 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/904>
212 2012-02-26 21:54:23 <etotheipi_> sipa, gmaxwell, any recommendation for challenge-response implementation for "prove-you-own-this-address"?  I'm sitting here trying to decide what string to use to be signed by the responder
213 2012-02-26 21:54:39 <luke-jr> etotheipi_: why are they proving it?
214 2012-02-26 21:54:40 <etotheipi_> it needs to be fresh... so I'm thinking hi-res date would be good
215 2012-02-26 21:55:00 <etotheipi_> but I"m not sure if that's enough
216 2012-02-26 21:55:01 <luke-jr> people shouldn't sign a date by itself
217 2012-02-26 21:55:30 <etotheipi_> luke-jr, one example is people are donating to me, expecting to get rewards, but everyone can see all the txs to the donation address
218 2012-02-26 21:55:36 <etotheipi_> there's nothing stopping them from claiming it's their own
219 2012-02-26 21:56:41 <etotheipi_> i.e. I could trivially setup a script to send out an email "hey, I just donated" as soon as they see a transaction hit the donation address
220 2012-02-26 21:56:41 <luke-jr> etotheipi_: "<date>: I sent transaction <txid>, and wish to request favour for <?>."
221 2012-02-26 21:56:58 <sipa> if you only sign the date, the value of the response is equal to "the owner of this address recently used the signature function"
222 2012-02-26 21:57:24 <sipa> sorry, slightly more
223 2012-02-26 21:57:35 <etotheipi_> sipa, right... that's why I want to include something else, say... personal or unique... especially something the challenger could remember
224 2012-02-26 21:57:46 <luke-jr> etotheipi_: "<date>: I sent transaction <txid>, and wish to request favour for <?>."
225 2012-02-26 21:57:53 <etotheipi_> while users can enter whatever they want... I should have a default
226 2012-02-26 21:58:04 <sipa> etotheipi_: what about a nonce signed by yourself?
227 2012-02-26 21:58:15 <sipa> etotheipi_: when that comes back, you're sure that the challenge came from you
228 2012-02-26 21:58:32 <etotheipi_> sipa, that's the kind of idea I was looking for
229 2012-02-26 21:58:46 <etotheipi_> luke-jr, your idea is good, except that it's a lot of work to collect that info
230 2012-02-26 21:59:06 <etotheipi_> I mean, it's not hard, but it's unnecessary to go find the tx ID and write a personal message...
231 2012-02-26 21:59:44 <sipa> etotheipi_: I'm still not sure what you'd actually need such a proof for, though.
232 2012-02-26 22:00:36 <etotheipi_> sipa, I am not convinced this is going to be an epic feature... but I have seen cases where it would be useful
233 2012-02-26 22:03:26 <etotheipi_> plus, I thought the Satoshi client was implementing something like this (in a GUI)
234 2012-02-26 22:05:01 <etotheipi_> https://en.bitcoin.it/wiki/Contracts#Example_2:_Escrow_and_dispute_mediation :  "The merchant challenges the mediator with a random nonce. The mediator signs the nonce with the private form of K2, thus proving it really belongs to merchant."
235 2012-02-26 22:06:08 <etotheipi_> so that's actually a very useful situation:  someone is adding a third party to your contract, you need to verify that the third-party key belongs to a known, impartial party
236 2012-02-26 22:09:18 <etotheipi_> sipa, I guess it doesn't even have to be a signed msg... it could just be an address in your wallet
237 2012-02-26 22:10:09 <etotheipi_> and you would reject/warn if the signature is signing something that isn't in your wallet
238 2012-02-26 23:50:47 <luke-jr> etotheipi_: &
239 2012-02-26 23:51:21 <etotheipi_> luke-jr, what's up?
240 2012-02-26 23:54:19 <luke-jr> [17:58:46] <etotheipi_> luke-jr, your idea is good, except that it's a lot of work to collect that info
241 2012-02-26 23:54:21 <luke-jr> not really
242 2012-02-26 23:55:01 <etotheipi_> luke-jr, I am having second thoughts about how to implement this... I didn't feel it was necessary to collect TxIDs, you just needed anything unique
243 2012-02-26 23:55:27 <etotheipi_> but I'm starting to think about holes in this whole idea, so I need to think things through a bit more
244 2012-02-26 23:55:48 <sipa> luke-jr, etotheipi_: i think you both want signmessage for different purposes
245 2012-02-26 23:56:06 <luke-jr> sipa: I agree that signmessage as-is is not suitable for transaction-specific stuff
246 2012-02-26 23:56:32 <luke-jr> sipa: really, a way is needed to sign from a specific txn, and verify against that txn