1 2011-06-25 00:11:37 <gmaxwell> Anyone with a yubikey know how many challenge responses it can do per second? I'm thinking it might make a reasonable dirt cheap solution for website password validation that can't be stolen by a cracker.
  2 2011-06-25 00:12:27 <gmaxwell> E.g. you store salt,yubikey(iteratedexpensivehash(password+salt))
  3 2011-06-25 00:15:30 <tcatm> gmaxwell: it's limited to about 24 Hz by the usb keyboard interface
  4 2011-06-25 00:15:50 <tcatm> gmaxwell: no idea whether the crypto limit is lower or higher
  5 2011-06-25 00:17:55 <gmaxwell> hah alas. Even that would be a bit slow. Though perhaps challenge response mode doesn't have the keyboard limit?
  6 2011-06-25 00:18:35 <tcatm> could be, but I doubt the crypto IC is powerful enough to be much faster
  7 2011-06-25 00:21:23 <tcatm> gmaxwell: do you know whether it works with linux?
  8 2011-06-25 00:22:11 <gmaxwell> Yes, it does.
  9 2011-06-25 00:23:09 <gmaxwell> They have code here: https://github.com/Yubico/python-yubico
 10 2011-06-25 00:23:27 <tcatm> looks like the fastest API call is reading the serial number which takes 50ms
 11 2011-06-25 00:24:10 <tcatm> the paper says that can be done synchronously as it is fast. thus I'd guess the crypto calls are much slower
 12 2011-06-25 00:24:40 <gmaxwell> well in some modes you have to press the button, thus the need for async
 13 2011-06-25 00:24:42 <jrmithdobbs> gmaxwell: how does that work without having a preseed like rsa's stuff?
 14 2011-06-25 00:24:43 <jrmithdobbs> or does it
 15 2011-06-25 00:25:04 <gmaxwell> jrmithdobbs: IIRC you can load the hmac key into it, you just can't read it out again.
 16 2011-06-25 00:25:12 <jrmithdobbs> oh nice
 17 2011-06-25 00:25:21 <jrmithdobbs> why didn't someone do that years ago?
 18 2011-06-25 00:25:40 <jrmithdobbs> has anyone actually tried hard to get the hmac back off?
 19 2011-06-25 00:26:04 <tcatm> there is a 30-250
 20 2011-06-25 00:26:14 <tcatm> ms delay depending on the operation executed
 21 2011-06-25 00:26:39 <tcatm> page 7 of Yubikey Client COM API.doc
 22 2011-06-25 00:27:32 <gmaxwell> so you generate a hmac key offline, load it (and back it up so you don't lose all passwords if it fails), then plug the sucker into the website. no more password theft. But sadly 30-250ms is way too slow.
 23 2011-06-25 00:28:11 <gmaxwell> jrmithdobbs: who knows? obviously it should be hardened against that, but it's probably not impossible. People are using them for software licensing dongles though, so I imagine that its been attacked.
 24 2011-06-25 00:28:41 <gmaxwell> In the context of a webserver the ability to attack it would be pretty limited (no physical access), but it's too slow.
 25 2011-06-25 00:29:40 <jrmithdobbs> gmaxwell: think so? how much they cost?
 26 2011-06-25 00:29:57 <jrmithdobbs> $25 each?
 27 2011-06-25 00:30:04 <tcatm> how does the OTP (the 44 char keyboard input) work? I mean, how does it ensure the passphrase can be used only once?
 28 2011-06-25 00:30:09 <jrmithdobbs> where's the whitesheet i'm not seeing power reqs
 29 2011-06-25 00:30:11 <gmaxwell> haha ... I suppose you're right.
 30 2011-06-25 00:30:30 <jrmithdobbs> just by 16 of them
 31 2011-06-25 00:30:31 <gmaxwell> tcatm: it has an internal counter, so it only generates it once, the server side needs to remember the current counter position.
 32 2011-06-25 00:30:33 <jrmithdobbs> and a hub
 33 2011-06-25 00:30:48 <tcatm> gmaxwell: ah, okay
 34 2011-06-25 00:30:52 <jrmithdobbs> 50 for 1000 even!
 35 2011-06-25 00:30:59 <gmaxwell> jrmithdobbs: yea, would probably still be cheaper than the commercial security modules... kinda ghetto though!
 36 2011-06-25 00:31:06 <tcatm> now if I could use it for GPG... :)
 37 2011-06-25 00:31:34 <jrmithdobbs> gmaxwell: how are people actually using it in the wild though? there's no clock on it so you can't use it like a traditional token
 38 2011-06-25 00:31:38 <jrmithdobbs> unless i'm missing something
 39 2011-06-25 00:32:02 <jrmithdobbs> sure it generates a OTP but if it can't be predicted on the other side ...
 40 2011-06-25 00:32:26 <gmaxwell> The otp is keyed. The other side knows the key.
 41 2011-06-25 00:33:22 <gmaxwell> In OTP mode, it outputs counter,sha1(key+counter)  The server knows the key and the last counter used.
 42 2011-06-25 00:35:19 <jrmithdobbs> gmaxwell: so what happens if you push the button too many times and send it to the wrong place?
 43 2011-06-25 00:35:36 <jrmithdobbs> it just makes sure the counter is > last time?
 44 2011-06-25 00:36:18 <gmaxwell> I believe so. Yea, it's not as good as the rsa tokens. But it's also $25 with no server costs.
 45 2011-06-25 00:36:33 <gmaxwell> the challenge response mode prevents that kind of stupidity, but it needs client software.
 46 2011-06-25 00:36:48 <gmaxwell> I the normal one way OTP mode it just acts as a usb keyboard and works everywhere.
 47 2011-06-25 00:37:11 <BitcoinForNewegg> you can also generate a few passcodes
 48 2011-06-25 00:37:14 <BitcoinForNewegg> and keep them with u
 49 2011-06-25 00:37:30 <jrmithdobbs> that's not useful
 50 2011-06-25 00:39:08 <jrmithdobbs> be really cool if they'd implement one that appeared as a microsdcard
 51 2011-06-25 00:39:19 <jrmithdobbs> usable in phone
 52 2011-06-25 00:39:31 <dehuman> ?
 53 2011-06-25 00:39:38 <BitcoinForNewegg> ?
 54 2011-06-25 00:39:57 <BitcoinForNewegg> RSA secureid tokens can be software, or a flash drive sized thingy
 55 2011-06-25 00:40:07 <dehuman> microsd dont see how that would work
 56 2011-06-25 00:40:29 <jrmithdobbs> dehuman: i can't think of how it could either
 57 2011-06-25 00:40:30 <dehuman> usb hid class is kind of neat tho
 58 2011-06-25 00:40:50 <dehuman> but yah liek you said its not really feasible
 59 2011-06-25 00:40:51 <jrmithdobbs> dehuman: just saying, my phone is where i'd want things like that the most ;p
 60 2011-06-25 00:40:56 <dehuman> cause of the reason you just enumerated
 61 2011-06-25 00:41:02 <dehuman> needs to work across devices
 62 2011-06-25 00:41:31 <dehuman> is there a bluetooth hid class?
 63 2011-06-25 00:41:35 <dehuman> there is isnt there
 64 2011-06-25 00:41:50 <gmaxwell> yes, of course, people use keyboards with phones all the time.
 65 2011-06-25 00:42:04 <dehuman> a bluetooth + usb key fob would work for many modern smartphone and computer
 66 2011-06-25 00:42:06 <jrmithdobbs> i wouldn't say all the time ;p
 67 2011-06-25 00:42:07 <gmaxwell> the yubico forum has some post where people are talking about bluetooth adaptors.
 68 2011-06-25 00:42:20 <dehuman> er is
 69 2011-06-25 00:42:26 <dehuman> sorry dunno why i said isnt
 70 2011-06-25 00:42:56 <dehuman> i meant' duh of course y0, serial, audio, keyboard + mouse'
 71 2011-06-25 00:42:56 <jrmithdobbs> oh really, didn't know usb hid -> bt hid adapters existed
 72 2011-06-25 00:43:30 <dehuman> well you'd just have whatever uc thing with both usb hid class and bt transceiver
 73 2011-06-25 00:43:47 <dehuman> i dunno about the usb to bt hid adapters exist but that makes sense i suppose
 74 2011-06-25 00:43:53 <dehuman> plug your favorite keyboard into bluetooth
 75 2011-06-25 00:43:53 <jrmithdobbs> dehuman: no you need batteries in it then
 76 2011-06-25 00:44:11 <dehuman> well you'd have to have batteries for bluetooth
 77 2011-06-25 00:44:12 <gmaxwell> yubico actually has a more proper HSM that is actually ment for this ghetto thing I'm suggesting, but it's $500, which is enormously cheap compared to whatever else... but it's not so cheap that joe-random-bitcoin-service would actually deploy it.
 78 2011-06-25 00:44:12 <jrmithdobbs> makes it much larger if replacable and limited lifetime if not
 79 2011-06-25 00:44:14 <dehuman> just make ti a fob
 80 2011-06-25 00:44:17 <dehuman> and have it charge usb
 81 2011-06-25 00:44:27 <jrmithdobbs> which kills two of their major selling points
 82 2011-06-25 00:44:30 <dehuman> with a small li-ion or something
 83 2011-06-25 00:44:38 <BitcoinForNewegg> so mtgox shoulda reopened site by now?
 84 2011-06-25 00:44:48 <pedroleone> any cool stuff to do with gpu's besides mining and cracking passwords?
 85 2011-06-25 00:44:49 <dehuman> well thats just reality of a phone interface jrsmithdob
 86 2011-06-25 00:44:56 <gmaxwell> http://static.yubico.com/var/uploads/pdfs/YubiHSM%20Manual%202011-04-12.pdf the fine manual
 87 2011-06-25 00:45:06 <jrmithdobbs> dehuman: hence why i was saying microsd
 88 2011-06-25 00:45:11 <dehuman> can an android phone act as an usb host?
 89 2011-06-25 00:45:19 <jrmithdobbs> dehuman: even though i can't think of how it'd work since afaik sd doesn't pass any power
 90 2011-06-25 00:45:23 <dehuman> but thats mass storage client class ro whatever
 91 2011-06-25 00:45:44 <jrmithdobbs> can if running cyanogenmod on certain hardware
 92 2011-06-25 00:45:46 <dehuman> there is a controller that provides a specific client class for mass storage, additionally it would have to grok a fileystem since phone expects to see that
 93 2011-06-25 00:45:51 <jrmithdobbs> it doesn't implement a software usb host
 94 2011-06-25 00:46:19 <jrmithdobbs> dehuman: that's fine. vfat. one file. return value when file is read.
 95 2011-06-25 00:46:22 <jrmithdobbs> that's not the hard part
 96 2011-06-25 00:46:29 <dehuman> now maybe if you did some kludge where there was a clietn on the phone or pc that wrote 'files' to the device to communicate with it and it wrote back otp or wahtever
 97 2011-06-25 00:46:46 <dehuman> but you'd need some middleware
 98 2011-06-25 00:46:51 <jrmithdobbs> yes of course
 99 2011-06-25 00:46:56 <dehuman> client is ikcy bnut eh
100 2011-06-25 00:47:03 <jrmithdobbs> you'd need middleware but the interface would actually be simpler than usb hid
101 2011-06-25 00:47:06 <jrmithdobbs> if you could get power to it.
102 2011-06-25 00:47:23 <dehuman> oh yah its powered you can just get power off the usb rail no matter what
103 2011-06-25 00:47:33 <dehuman> not alot but you dont need a lot
104 2011-06-25 00:49:26 <dehuman> i like the usb hid, it types in yoru password for you, no client
105 2011-06-25 00:49:32 <dehuman> any hid would be the same
106 2011-06-25 00:49:53 <gmaxwell> dehuman: since it can't do challenge response that limits the security though.
107 2011-06-25 00:50:27 <gmaxwell> e.g. I can borrow your yubikey and spool out a bunch of one-time passwords.
108 2011-06-25 00:50:33 <dehuman> eh
109 2011-06-25 00:50:45 <dehuman> can phones do usb cdc ?
110 2011-06-25 00:50:50 <dehuman> usb serial?
111 2011-06-25 00:50:53 <gmaxwell> but it is nice that it can do that driverless, and with a driver it can do challenge response.
112 2011-06-25 00:50:56 <gmaxwell> No clue.
113 2011-06-25 00:51:01 <jrmithdobbs> dehuman: none that i know of
114 2011-06-25 00:51:06 <jrmithdobbs> well
115 2011-06-25 00:51:09 <dehuman> then you can do a client that could talk to it
116 2011-06-25 00:51:14 <jrmithdobbs> old se phones can do bt serial
117 2011-06-25 00:51:15 <dehuman> and there is bluetooth serial profile too
118 2011-06-25 00:51:27 <jrmithdobbs> but android/iphone have the shittiest bt stacks EVER
119 2011-06-25 00:51:33 <jrmithdobbs> and those are the two that matter
120 2011-06-25 00:51:40 <dehuman> well thats why bt is a ghetto whore in the first place
121 2011-06-25 00:51:47 <dehuman> bt stacks are all over the place
122 2011-06-25 00:51:59 <dehuman> when you buy a bluetooth radio thats what you are paying for if you pay retail
123 2011-06-25 00:52:07 <dehuman> you can get the radios for like $2
124 2011-06-25 00:52:32 <dehuman> but microsoft even provides a free serial profile iirc
125 2011-06-25 00:52:57 <jrmithdobbs> bluez implements one
126 2011-06-25 00:53:16 <jrmithdobbs> serial, hid, file transfer iirc
127 2011-06-25 00:54:10 <jrmithdobbs> gmaxwell: i dunno if you use password + otp to aquire say, a kerberos token, the challenge response part isn't as important
128 2011-06-25 00:54:16 <jrmithdobbs> kerberos solves so many of these damned problems
129 2011-06-25 00:54:20 <jrmithdobbs> why don't people use it :(
130 2011-06-25 00:55:46 <dehuman> i think android can do SPP with BlueToothSocket
131 2011-06-25 00:56:28 <jrmithdobbs> spp?
132 2011-06-25 00:57:08 <gmaxwell> jrmithdobbs: someone can still pull off a bunch of otp keys using their laptop then themselves a krb ticket later when they feel like it.
133 2011-06-25 00:57:11 <dehuman> bluetooth serial port profile
134 2011-06-25 00:57:18 <dehuman> iphone doesnt and wont
135 2011-06-25 00:57:35 <dehuman> but there are others like PAN or maybe AVRCP that might be able to be used for other than intended purposes
136 2011-06-25 00:58:37 <jrmithdobbs> gmaxwell: except a sanely setup krb server only allows 5-10 minutes before expiration and if the real user uses a otp before they can the iterator will have already advanced
137 2011-06-25 00:58:51 <jrmithdobbs> gmaxwell: it mitigates it greatly
138 2011-06-25 00:59:08 <gmaxwell> Okay, fair enough.
139 2011-06-25 00:59:22 <jrmithdobbs> <3 kerberos
140 2011-06-25 00:59:27 <gmaxwell> Krb also solves the "server has to know the secret, so you can't use it to authenticate to multiple things" issue.
141 2011-06-25 00:59:37 <jrmithdobbs> (says the one man on earth with a non-microsoft kerberos realm setp in his apartment)
142 2011-06-25 00:59:43 <gmaxwell> (since only the krb server knows)
143 2011-06-25 00:59:44 <jrmithdobbs> s/setp/setup/
144 2011-06-25 00:59:55 <jrmithdobbs> yup
145 2011-06-25 01:00:14 <gmaxwell> jrmithdobbs: I've run krb at home, but not for years now... two accounts plus ssh key auth sort of removed the need.
146 2011-06-25 01:00:59 <jrmithdobbs> gmaxwell: ;p
147 2011-06-25 01:01:43 <jrmithdobbs> gmaxwell: did you know most major browsers actually support kerberos auth now?
148 2011-06-25 01:01:59 <jrmithdobbs> safari / firefox / ie
149 2011-06-25 01:02:14 <jrmithdobbs> not such about chrome but probably not seeing as it's ssl implementation is halfassed
150 2011-06-25 01:02:18 <jrmithdobbs> s/such/sure/
151 2011-06-25 01:02:57 <jrmithdobbs> gmaxwell: and still. NOONE USES IT.
152 2011-06-25 01:03:00 <jrmithdobbs> it makes me nerd rage.
153 2011-06-25 01:03:01 <jrmithdobbs> :(
154 2011-06-25 01:03:58 <gmaxwell> what n!?!@?
155 2011-06-25 01:05:47 <jrmithdobbs> oh wow
156 2011-06-25 01:05:51 <jrmithdobbs> color me shocked
157 2011-06-25 01:05:52 <jrmithdobbs> http://code.google.com/p/chromium/issues/detail?id=19
158 2011-06-25 01:06:02 <jrmithdobbs> chrome does support it. at least on windows not sure about other platforms
159 2011-06-25 01:06:12 <jrmithdobbs> gmaxwell: yup! you can thank microsoft for that actually.
160 2011-06-25 01:06:29 <jrmithdobbs> gmaxwell: they dropped their retarded NTLM browser auth and use the AD kerberos token now
161 2011-06-25 01:06:59 <jrmithdobbs> gmaxwell: nobody outside of AD-centric intranets uses it though
162 2011-06-25 01:08:00 <jrmithdobbs> gmaxwell: combined with dns-based kerberos realm discovery and/or sane realm names or the realm+host data being provided by the users it's actually feasible to use it on the internet
163 2011-06-25 01:08:32 <jrmithdobbs> assuming your target audience can get a valid krb token from an internet accessible server, of course
164 2011-06-25 01:08:44 <jrmithdobbs> (so, .edu basically, at this point)
165 2011-06-25 01:09:32 <gmaxwell> it couldn't work any less well than openid
166 2011-06-25 01:09:59 <gmaxwell> (at least I've found openid to your own site is basically non-functional, just about everything only works right against the few major services they support)
167 2011-06-25 01:10:01 <jrmithdobbs> works much better
168 2011-06-25 01:10:09 <jrmithdobbs> i never "got" openid
169 2011-06-25 01:10:22 <jrmithdobbs> hey google: just throw up so goddamned kerberos servers
170 2011-06-25 01:10:54 <jrmithdobbs> gmaxwell: that's because it's not a well defined spec and no two providers implement it the same way
171 2011-06-25 01:11:01 <gmaxwell> Well, it attempts to solve the same need. Except its from the web world so its hacky shit.
172 2011-06-25 01:12:04 <jrmithdobbs> ya but it's crappy and not based on strong, well-proven, crypto if i remember how openid works correctly
173 2011-06-25 01:12:24 <Namegduf> Huh.
174 2011-06-25 01:12:28 <Namegduf> I've had the opposite experience.
175 2011-06-25 01:12:41 <jrmithdobbs> Namegduf: of what?
176 2011-06-25 01:12:46 <Namegduf> OpenID.
177 2011-06-25 01:12:50 <upb> AHAHAHHAHAHA
178 2011-06-25 01:12:52 <upb> We have found that balances on some older accounts look significantly incorrect when compared with the old database. At this time we do not know what caused the balances to be off
179 2011-06-25 01:13:03 <gmaxwell> Namegduf: have you actually run you own.
180 2011-06-25 01:13:03 <Namegduf> I've not used that much OpenID stuff, but it's all liked my phpMyOpenId installation fine.
181 2011-06-25 01:13:09 <jrmithdobbs> gmaxwell: rofl http://code.google.com/p/chromium/issues/detail?id=19
182 2011-06-25 01:13:09 <Namegduf> *ID
183 2011-06-25 01:13:11 <Namegduf> Yes.
184 2011-06-25 01:13:12 <jrmithdobbs> er
185 2011-06-25 01:13:15 <jrmithdobbs> https://support.mtgox.com/entries/20224998-huge-bitcoin-sell-off-due-to-a-compromised-account-rollback
186 2011-06-25 01:13:21 <gmaxwell> Namegduf: yea, it doesn't actually work with lots of sites.
187 2011-06-25 01:13:35 <jrmithdobbs> also google randomly changes your uid
188 2011-06-25 01:13:53 <jrmithdobbs> so actually working properly with google (arguably the most important/common provider) is a nightmare
189 2011-06-25 01:14:01 <jrmithdobbs> you can randomly lose access to your account and shit
190 2011-06-25 01:14:18 <jrmithdobbs> (from the other side of the conversation)
191 2011-06-25 01:14:27 <jrmithdobbs> upb: i knew that 24 hours ago
192 2011-06-25 01:14:29 <Namegduf> Ick.
193 2011-06-25 01:14:42 <upb> jrmithdobbs: how ?
194 2011-06-25 01:14:51 <Namegduf> gmaxwell: I think I've only tried Simple Machine Forums and britcoin, so you're probably right
195 2011-06-25 01:14:59 <jrmithdobbs> upb: he said it in #mtgox
196 2011-06-25 01:15:02 <upb> oh
197 2011-06-25 01:15:09 <gmaxwell> Namegduf: or it behaves weird, for example on google gerrit you can login via it but your access is readonly if you're coming from phpmyopenid.
198 2011-06-25 01:15:16 <jrmithdobbs> upb: THERE IS NO SQLI
199 2011-06-25 01:15:22 <jrmithdobbs> upb: PAY NO ATTENTION TO THE MAN BEHIND THE CURTAIN
200 2011-06-25 01:15:28 <Namegduf> gmaxwell: Weird.
201 2011-06-25 01:16:19 <jrmithdobbs> upb: http://pastebin.com/HGssM2L7
202 2011-06-25 01:16:39 <gmaxwell> Namegduf: there are also sites that simply block anything that isn't one of the major sites: too many sibyls.
203 2011-06-25 01:17:01 <jrmithdobbs> ya, openid is a failure
204 2011-06-25 01:17:30 <Namegduf> That sounds like a failure, yeah.
205 2011-06-25 01:18:00 <gmaxwell> "usingly negative by a lot" uh oh.
206 2011-06-25 01:18:52 <gmaxwell> wtf?
207 2011-06-25 01:27:47 <PatrikR> openid is good in theory, but causes too many problems in practice...
208 2011-06-25 01:28:28 <jrmithdobbs> no actually it's not even good in theory
209 2011-06-25 01:28:34 <jrmithdobbs> it seriously solves the same problem kerberos does.
210 2011-06-25 01:28:45 <jrmithdobbs> kerberos has been around 20 years, is well vetted, and is backed by strong crypto
211 2011-06-25 01:40:12 <sacarlson> why would I get the error error: {"code":-1,"message":"value is type str, expected real"}  when I pass the value in params[1] with 1.23 with this function int64 nAmount = AmountFromValue(params[1]); ?
212 2011-06-25 01:40:43 <gmaxwell> pretty halarious, the nutbag "omg we gotta fork bitcoin" guy showed up in #bitcoin-mining as "kevinJ" and claimed to be not the same original nutbag  "I talked to him and he was right"; yea too bad for him his hostmask was the same. hehe
213 2011-06-25 01:42:08 <Graet> lol
214 2011-06-25 01:44:23 <grug> lol
215 2011-06-25 01:44:38 <grug> there are too many retards t hat claim that bitcoin needs forking
216 2011-06-25 01:45:24 <zamgo> so I forked bitcointools, to text out exporting wallet to SQL     https://github.com/zamgo/bitcointools
217 2011-06-25 01:45:27 <zamgo> it kinda works
218 2011-06-25 01:45:50 <zamgo> exporting blockchain to SQL is next... but that's non-trivial
219 2011-06-25 01:46:29 <gmaxwell> grug: he's probably quiting and reoining trying to change hist hostmask now. :)
220 2011-06-25 01:48:37 <lfm> zamgo why? block chain is already in a database
221 2011-06-25 01:48:48 <zamgo> why ask why?
222 2011-06-25 01:49:00 <zamgo> it's it a very flat database
223 2011-06-25 01:49:06 <lfm> zamgo you like wasting your time?
224 2011-06-25 01:49:16 <zamgo> wow
225 2011-06-25 01:49:23 <zamgo> welcome to ignore lfm
226 2011-06-25 01:49:51 <lfm> ok, he fails to deny it
227 2011-06-25 01:50:02 <jrmithdobbs> lfm: blockchain is not in a database
228 2011-06-25 01:50:11 <jrmithdobbs> lfm: an index of the blockchain is
229 2011-06-25 01:50:11 <lfm> bdb
230 2011-06-25 01:50:14 <jrmithdobbs> nope
231 2011-06-25 01:50:23 <jrmithdobbs> blk0001.dat is not bdb
232 2011-06-25 01:50:39 <jrmithdobbs> look at it, i didn't realise this until the other day either
233 2011-06-25 01:50:53 <lfm> blk0001 with blkindex
234 2011-06-25 01:56:44 <assassindrake> what happens if you send coins to a non-existant address?
235 2011-06-25 01:56:57 <jrmithdobbs> all addresses exist so long as the checksum verifies
236 2011-06-25 01:57:27 <ByteCoin> the coins wait until someone generates a key pair for that address
237 2011-06-25 01:57:29 <jrmithdobbs> and the magic number is valid
238 2011-06-25 01:57:37 <lfm> assassindrake: if you can figure out how to make up a non-existant address then the coins are basiclly lost
239 2011-06-25 01:58:05 <assassindrake> so mistyping an address cant happen?
240 2011-06-25 01:58:11 <jrmithdobbs> not really
241 2011-06-25 01:58:19 <ByteCoin> can happen - see my post on the forum
242 2011-06-25 01:58:22 <ByteCoin> not likely though
243 2011-06-25 01:58:25 <jrmithdobbs> it's theorhetically possible
244 2011-06-25 01:58:29 <jrmithdobbs> but ya not likely
245 2011-06-25 01:58:38 <assassindrake> ok
246 2011-06-25 01:58:49 <lfm> like 1111111111111111111114oLvT2 is a valid address. random addresses are probably invalid, they have a checksum part inside
247 2011-06-25 01:58:55 <assassindrake> my next question was would it show up in the block but i guess you answered that
248 2011-06-25 01:59:13 <doublec> it's easy to send namecoins to a bitcoin address and viceversa
249 2011-06-25 01:59:17 <doublec> quite a few people have done it
250 2011-06-25 01:59:22 <ByteCoin> 1ByteCoinAddressesMatch1kpCxNXmHKW
251 2011-06-25 01:59:24 <ByteCoin> 1ByteCoinAddressesMatch1kpCWNXmHKW
252 2011-06-25 01:59:29 <ByteCoin> both valid addresses
253 2011-06-25 02:00:05 <egecko> but the namecoins wouldnt be recognized in the official chain right?
254 2011-06-25 02:00:43 <zamgo> and easy to recover namecoins sent to bitcoin address....
255 2011-06-25 02:00:46 <zamgo> and vice versa
256 2011-06-25 02:00:48 <lfm> all those funny looking addresses above are in the block chain
257 2011-06-25 02:01:09 <doublec> egecko: right
258 2011-06-25 02:01:32 <lfm> but you wont "recover" coins sent to them
259 2011-06-25 02:01:33 <doublec> zamgo: yep, I've done it for a few users who've sent funds to the wrong type of address in the exchange
260 2011-06-25 02:02:07 <zamgo> yes, got some code already that can spit out namecoin/bitcoin/testnet addresses for a keypair
261 2011-06-25 02:02:45 <lfm> oh ok but you cant recover coins sent to 1111111111111111111114oLvT2 cuz there never was a private key for it
262 2011-06-25 02:03:23 <ByteCoin> there never was a public key for it either for that matter
263 2011-06-25 02:03:30 <lfm> ya true
264 2011-06-25 02:03:56 <ByteCoin> lfm: How come I don't see you post on the forum?
265 2011-06-25 02:04:13 <cuddlefish> Hey guys, I have a strange proposal.
266 2011-06-25 02:04:21 <lfm> I have a little bit but I dont follow it much
267 2011-06-25 02:04:22 <cuddlefish> why do we need public-key crypto
268 2011-06-25 02:04:28 <ByteCoin> ok
269 2011-06-25 02:05:04 <lfm> cuddlefish: you mean in general or you mean for bitcoin?
270 2011-06-25 02:05:12 <cuddlefish> lfm: for Bitcoin
271 2011-06-25 02:05:43 <ByteCoin> lfm: Gavin's looking for ways to stop people losing their bitcoins if their computers get hacked
272 2011-06-25 02:06:02 <ByteCoin> He's willing to change the client and introduce new transaction types etc..
273 2011-06-25 02:06:10 <lfm> cuddlefish: hmm not sure how to explain, it is basiclly the only way we know to prove ownership of some BTC
274 2011-06-25 02:06:25 <cuddlefish> lfm: Well, it's the only way you know :P
275 2011-06-25 02:07:21 <ByteCoin> cuddlefish: Briefly, you need to be able to sign something without the info in your signature enabling someone else to get your key
276 2011-06-25 02:07:34 <lfm> cuddlefish: it also stops other people claiming ownership even after they're spent.
277 2011-06-25 02:07:40 <cuddlefish> ByteCoin: I know. i'm just writing this out
278 2011-06-25 02:08:08 <ByteCoin> cuddlefish: You did ask why PK crypto.. that's the reason.
279 2011-06-25 02:08:33 <lfm> cuddlefish: you think you know another way?
280 2011-06-25 02:08:54 <cuddlefish> yep
281 2011-06-25 02:09:01 <cuddlefish> Fixing a few flaws though
282 2011-06-25 02:09:03 <lfm> good luck with that
283 2011-06-25 02:09:11 <ByteCoin> can be done I think without PK crypto but keys are huge I think...
284 2011-06-25 02:10:23 <lfm> might be some way with diffie-helman key exchange or something but it wou;ld be way more clubersome I think
285 2011-06-25 02:10:39 <lfm> clumbersome
286 2011-06-25 02:10:48 <ByteCoin> dh still public key
287 2011-06-25 02:10:57 <lfm> well not really
288 2011-06-25 02:11:03 <ByteCoin> yes really...
289 2011-06-25 02:11:19 <lfm> dh doesnt do signature afaik
290 2011-06-25 02:11:51 <sacarlson> I'm getting this error error: {"code":-1,"message":"value is type str, expected real"}  with 3 different version I have compiled for bitcoind settxfee .001 does the present build of bitcoind display this?  maybe I have a lib problem?
291 2011-06-25 02:11:53 <ByteCoin> Not directly..
292 2011-06-25 02:12:09 <ByteCoin> But even DSA can be traced back to DH
293 2011-06-25 02:13:10 <lfm> sacarlson: ya so you're passing a string instead of a real. so?
294 2011-06-25 02:13:11 <ByteCoin> Ok I grant that when most people say DH they think of key agreement which has no signature scheme
295 2011-06-25 02:13:35 <sacarlson> lfm: so how do I pass a real on the command line?
296 2011-06-25 02:13:41 <ByteCoin> brb
297 2011-06-25 02:14:17 <lfm> sacarlson: what command line?
298 2011-06-25 02:14:42 <sacarlson> lfm: bitcoind settxfee .001 ;  as I type on my command line
299 2011-06-25 02:14:49 <lfm> try 0.001
300 2011-06-25 02:15:06 <sacarlson> lfm: ok cool I'll try that thanks
301 2011-06-25 02:15:37 <lfm> some packages say numbers need to start with digits
302 2011-06-25 02:15:45 <sacarlson> lfm: nope same error bitcoind settxfee 0.001
303 2011-06-25 02:17:10 <lfm> sacarlson: works for me. I dont know what you are doing different
304 2011-06-25 02:17:39 <lfm> "version" : 32300,
305 2011-06-25 02:17:40 <sacarlson> lfm: must be the version I'm running or my libs I guess
306 2011-06-25 02:18:41 <sacarlson> lfm: "version" : 32300,  but the last I tried was with the mod for escrow so I'll try check out a different one
307 2011-06-25 02:23:06 <lfm> it seems to me it is actually a bug to demand it as a real snce it theoreticlly could get inaccurate, not that you are likely to want to set the fee to 20 million btc where youd see the loss of accuracy
308 2011-06-25 02:25:16 <gmaxwell> lfm: not just that, it encourages stupid behavior on the part of users. E.g. they'll end up using single precision float
309 2011-06-25 02:26:39 <gmaxwell> and then when people get righteous, 'how dare you use float for money you idiot!' they'll feel misguided by the api.
310 2011-06-25 02:27:20 <sacarlson> lfm: well I'm happy with a work around if you have one
311 2011-06-25 02:27:49 <lfm> maybe the fact sacarlson is getting the error in a test version means someone is part way thru fixing it
312 2011-06-25 02:28:09 <gmaxwell> then they'll switch to 32 bit ints and die in an overflow. :)
313 2011-06-25 02:28:15 <lfm> sacarlson: go back to the "production" verson of 0.3.23
314 2011-06-25 02:28:18 <sacarlson> does it work in version 0c97aa9e141f5c23dc9f80e5fb4c8297d8a6b4dd published may 17th?
315 2011-06-25 02:28:50 <lfm> whats that version? a md5sum or ??
316 2011-06-25 02:29:05 <lfm> I get 4d79c45086d17b38015c1babce1f5018  bitcoin-0.3.23-linux.tar.gz
317 2011-06-25 02:29:12 <lfm> for my tar.gz file
318 2011-06-25 02:29:26 <jgarzik> the sums I posted are SHA1
319 2011-06-25 02:29:48 <sacarlson> lfm: on my gitg it says it's the sha version
320 2011-06-25 02:30:25 <lfm> k I get 5ca82c5b694fc432b4b342dde5cd1b145e265f13  bitcoin-0.3.23-linux.tar.gz
321 2011-06-25 02:30:31 <lfm> for sha1sum
322 2011-06-25 02:32:45 <jgarzik> lfm: yep that matches sums posted in http://forum.bitcoin.org/index.php?topic=16553.msg215481#msg215481
323 2011-06-25 02:35:02 <sacarlson> lfm: yes I'm using https://github.com/bitcoin/bitcoin.git
324 2011-06-25 02:36:16 <sacarlson> oh IC your giving me the sha1 of the tar file no I was looking for the github checkout point
325 2011-06-25 02:36:52 <lfm> sacarlson: k, see if you can try the tar file version
326 2011-06-25 03:55:53 <sacarlson> I found what I was doing wrong in my error: {"code":-1,"message":"value is type str, expected real"} ,  I used an older version of bitcoind to do the communication on rpc,  I never realized that could change anything
327 2011-06-25 04:18:34 <CIA-103> DiabloMiner: Patrick McFarland master * r354b532 / (2 files in 2 dirs):
328 2011-06-25 04:25:37 <gribble> 133176
329 2011-06-25 04:25:37 <phantomcircuit> ;;bc,blocks
330 2011-06-25 04:28:32 <jrmithdobbs> was just about to do that lol
331 2011-06-25 04:28:44 <jrmithdobbs> status: blocks still too damned big
332 2011-06-25 04:28:46 <jrmithdobbs> ugh
333 2011-06-25 04:37:02 <egecko> blocks prolly arent gonna be getting smaller ;)
334 2011-06-25 06:05:19 <egecko> aww bummer man, peter falk died
335 2011-06-25 06:11:51 <[Tycho]> Wow, first evil mining botnets :)
336 2011-06-25 06:15:18 <jgarzik> [Tycho]: ?
337 2011-06-25 06:16:13 <[Tycho]> Hello, jgarzik.
338 2011-06-25 06:16:24 <[Tycho]> What's your question ?
339 2011-06-25 06:16:42 <jgarzik> <[Tycho]> Wow, first evil mining botnets :)   <<--  to what do you refer?
340 2011-06-25 06:18:30 <[Tycho]> To a some evil botnets that I detected this morning.
341 2011-06-25 06:18:45 <[Tycho]> One is pretty impressive, peaking at 35 GH/s on CPUs.
342 2011-06-25 06:25:17 <[Tycho]> Are you scared ?
343 2011-06-25 07:09:44 <prof7bit> how do you detect botnets?
344 2011-06-25 07:11:40 <prof7bit> or do you own them? ;-)
345 2011-06-25 07:12:00 <doublec> sudden appearance of a ton of cpu's mining at the same time?
346 2011-06-25 07:12:00 <prof7bit> and detect that they actually work?
347 2011-06-25 07:19:35 <[Tycho]> It's somewhat obvious when I see 5000 PCs on one worker account.
348 2011-06-25 07:19:54 <[Tycho]> Well, it was 5000 five hours ago, now more.
349 2011-06-25 07:20:32 <[Tycho]> Also, one of my users from russian forum catched a sample.
350 2011-06-25 07:24:28 <sipa> nothing compared to the MM :)
351 2011-06-25 07:25:22 <[Tycho]> I think it's more than 10000 now.
352 2011-06-25 07:31:58 <vegard> that's really interesting.
353 2011-06-25 07:36:34 <[Tycho]> Also, his trojan is not really secure. He already wrote a message to me telling that someone changed his account password :)
354 2011-06-25 07:40:03 <[Tycho]> Who is the owner of bitcoincharts ?
355 2011-06-25 08:02:26 <prof7bit> maybe after 25 years successfully suppressing the evil part in me it is now time to rethink this and maybe join the dark side...
356 2011-06-25 08:05:11 <wump> you're giving up your shining white armor?
357 2011-06-25 08:05:57 <diki> I thought that even when the diff was 7-8 million, blocks would still be found at the same place, however after the new diff, the blocks with <1mill shares have decreased...
358 2011-06-25 08:06:13 <sipa> wump == wumpus?
359 2011-06-25 08:06:21 <diki> blocks would still be found at the same place<-becasue of how pools work
360 2011-06-25 08:06:25 <wump> yes
361 2011-06-25 08:06:28 <diki> s/place/pace
362 2011-06-25 08:06:43 <wump> hey my name changed
363 2011-06-25 08:07:00 <sipa> what is the nInitialBlockThreshold = 10000 in your patch?
364 2011-06-25 08:07:29 <wump> it is an arbitrary magic difference value, how recent a block needs to be to be called 'initial'
365 2011-06-25 08:08:04 <sipa> hmmm, ok
366 2011-06-25 08:08:05 <wump> without such a thing, if the approximation would be close to the real thing, the client would claim to be in initial download mode forever
367 2011-06-25 08:08:22 <sipa> the question is: if you've been offline for a day
368 2011-06-25 08:08:46 <sipa> shouldn't the client switch to "initial download"-like mode to protect you from doing transactions with outdated data either?
369 2011-06-25 08:08:49 <wump> yes it'd go into initial download mode again to sync up :) in my case it'd mean it shows a progress bar
370 2011-06-25 08:08:50 <wump> yes
371 2011-06-25 08:09:05 <wump> so 10000 is prob to much
372 2011-06-25 08:09:19 <sipa> well, nInitialBlockTreshold should maybe be more like 144 (=1 day) ?
373 2011-06-25 08:09:38 <sipa> and we shouldn't keep calling it "initial" download either, probably
374 2011-06-25 08:09:46 <wump> that'd be better -- but I just erred on the safe site
375 2011-06-25 08:09:58 <sipa> sure, i was just curious for the reason behind the constant
376 2011-06-25 08:10:30 <wump> indeed, it would only be initial to the client startup.. not initial to the persistent instance
377 2011-06-25 08:10:38 <sipa> maybe we can change it do "Synchronizing with block chain..." or something
378 2011-06-25 08:10:50 <wump> good idea
379 2011-06-25 08:10:56 <sipa> or even "Synchronizing with network..."
380 2011-06-25 08:11:03 <wump> even better
381 2011-06-25 08:11:06 <sipa> i really like the idea of a progress bar
382 2011-06-25 08:11:26 <wump> yep it takes the guessing out of it
383 2011-06-25 08:11:33 <wump> and makes it more like bittorrent :-)
384 2011-06-25 08:11:46 <sipa> indeed
385 2011-06-25 08:12:00 <prof7bit> "We have found that balances on some older accounts look significantly incorrect when compared with the old database. At this time we do not know [...]"  <-- LOL
386 2011-06-25 08:12:27 <sipa> i wonder if "really old" means: from before mt acquired mtgox
387 2011-06-25 08:13:28 <prof7bit> "No money is lost"
388 2011-06-25 08:15:43 <wump> yes he worded that really badly
389 2011-06-25 08:16:00 <wump> I hope he restores my btc balance back to that time though ;)
390 2011-06-25 08:19:30 <sipa> you can already check your balance on the site
391 2011-06-25 08:19:35 <sipa> mine is definitely correct
392 2011-06-25 08:21:48 <sipa> ;;later tell jgarzik how did you come up with the mixed sha512+sha256 key derivation function used in scratchcards? is it published/standardized anywher?
393 2011-06-25 08:21:49 <gribble> The operation succeeded.
394 2011-06-25 08:26:44 <wump> my balance is correct too
395 2011-06-25 08:39:23 <sacarlson> does this guy groffer ever chat here?  the guy who created the escrow commit? https://github.com/groffer/bitcoin/commit/83707c8dd4573bb958f9e504fb6263c8fa1ef942  if so do we know what he uses as a nick here?
396 2011-06-25 08:39:38 <gribble> Error: "lastseen" is not a valid command.
397 2011-06-25 08:39:38 <sacarlson> ;;lastseen groffer
398 2011-06-25 08:44:37 <picci> sacarlson: ;;seen is the command
399 2011-06-25 08:44:49 <gribble> groffer was last seen in #bitcoin-dev 1 week, 0 days, 8 hours, 45 minutes, and 55 seconds ago: <groffer> if you have time to review my code, I'd be happy for any comments
400 2011-06-25 08:44:49 <sacarlson> ;;seen groffer
401 2011-06-25 08:44:59 <sacarlson> picci: thanks
402 2011-06-25 08:59:14 <yorick> what is the amount of listtransactions calls I can make every second?
403 2011-06-25 09:02:40 <yorick> also, does "listreceivedbyaddress" only list things once there are confirmations?
404 2011-06-25 09:08:00 <yorick> what is the difference between listreceivedbyaddress and listtransactions?
405 2011-06-25 09:11:26 <Happy0> listtransactions includes send transactions, amirite, yorick?
406 2011-06-25 09:11:36 <yorick> hrm, that might be true
407 2011-06-25 09:11:42 <yorick> but then the name is confusing
408 2011-06-25 09:31:29 <carli2> hi, how can i transfer a private key - public key pair from one computer to an other?
409 2011-06-25 09:33:13 <sipa> for now, the only way is moving wallet files (and there are always risks in doing so)
410 2011-06-25 09:33:36 <sipa> i have a patch that can export a private key using RPC
411 2011-06-25 09:33:41 <sipa> and import it again elsewhere
412 2011-06-25 09:36:15 <carli2> how hard is it to crack a private key? do you think it will be crackable in 2033?
413 2011-06-25 09:37:06 <vegard> no
414 2011-06-25 09:40:41 <erus`> yes it will
415 2011-06-25 09:41:12 <erus`> its crackable now
416 2011-06-25 09:41:18 <erus`> just not cost effective
417 2011-06-25 09:48:47 <Sebastan> in chatlog I have read that finding a valid block will result in 50btc always. Is that correct? If so how does this match the 10 minute timespan? Are the 10 mins only a value that is targetted so that the difficulty is changed to reach 10 mins?
418 2011-06-25 09:49:58 <jtaylor> yes a block is 50btc for the solver and yes difficulty changes so one gets created every 10 min on average
419 2011-06-25 09:51:16 <Sebastan> and who is calculating the difficulty? I mean if its not zentralized in some way everyone would have different values isnt it?
420 2011-06-25 09:51:34 <sipa> Sebastan: it's a simple formula that's directly calculated from the block chain
421 2011-06-25 09:51:45 <sipa> since everyone has the same block chain, everyone finds the same number
422 2011-06-25 09:52:35 <sipa> erus`: it'd require at least 2^128 attempts
423 2011-06-25 09:53:53 <Sebastan> ok... so if someone has the last found block (found itself or got it from the network) he has the exact difficulty too...
424 2011-06-25 09:54:36 <sipa> the difficulty of each block is strictly determined by the previous blocks in the same chain, so there is no problem
425 2011-06-25 09:54:55 <sipa> erus`: that would take 10^18 years with the current bitcoin mining hardware
426 2011-06-25 09:55:19 <erus`> what about every computer in the world
427 2011-06-25 09:55:20 <sipa> if one attempt took as long as a bitcoin mining double-sha256 (it's a lot more, i believe), that is
428 2011-06-25 09:55:35 <sipa> maybe 10^14 years?
429 2011-06-25 09:56:16 <erus`> what about spending 5 years creating so much special hardware that it only takes 10 years
430 2011-06-25 09:56:24 <erus`> checkmate.
431 2011-06-25 09:56:59 <sipa> try as you may
432 2011-06-25 09:57:28 <Sebastan> and even though its said its unlikely that more than one people work on the same branch what happens when 2 miners find a matching hash nearly at the same time and give themselfes the 50btc then probably the one wins that propagates faster? having a earlier timestamp included wouldnt matter...
433 2011-06-25 09:57:54 <jtaylor> one of the solutions will be invalid
434 2011-06-25 09:58:49 <Sebastan> invalid? cant there be more than one solution that is below difficulty? or do you mean one is taken as invalid because the first propagated his result faster?
435 2011-06-25 09:59:03 <sipa> Sebastan: the block chain is actually a block tree
436 2011-06-25 09:59:22 <sipa> if two people find the next block simulateneously, you get a split in the chain
437 2011-06-25 09:59:36 <sipa> and part of the network will continue working on one side, and another part on the other
438 2011-06-25 09:59:40 <sipa> that is no problem
439 2011-06-25 09:59:57 <sipa> as soon as one of those finds a new block, that will becomes the new best, and win everywhere
440 2011-06-25 10:04:32 <carli2> does the network stay splittet then?
441 2011-06-25 10:04:54 <minus> *split
442 2011-06-25 10:05:05 <minus> (yes, i love correcting people)
443 2011-06-25 10:07:02 <sipa> 13:59:57 < sipa> as soon as one of those finds a new block, that will becomes the new best, and win everywhere
444 2011-06-25 10:07:13 <sipa> -s
445 2011-06-25 10:07:31 <sipa> the rule is: accept the first best block you see as master
446 2011-06-25 10:08:07 <sipa> so if two nodes find a same successor block at the same time, they will both be equally good, and nodes will assume the one they saw first is the best
447 2011-06-25 10:08:42 <sipa> but as soon as one of the chains gets extended, it is strickly better than the other chain, and a "reorganization" happens in the other chain
448 2011-06-25 10:09:15 <Sebastan> sipa: so only one client will keep his 50 btc? or will both get them?
449 2011-06-25 10:09:32 <sipa> only the one whose chain gets extended gets the reward
450 2011-06-25 10:09:49 <sipa> and the unlucky one's block doesn't exist anymore in the chain that eventually wins
451 2011-06-25 10:09:58 <sipa> so there is nothing left to give him a reward
452 2011-06-25 10:10:25 <Sebastan> i thought it would become a new branch.
453 2011-06-25 10:11:53 <Sebastan> but then again when 2 clients find a block at the same time the one will win that first gets the network to think he was first...
454 2011-06-25 10:12:26 <sipa> well both get the reward in the chain they themselves consider best, obviously
455 2011-06-25 10:12:40 <sipa> only after a while the fork is resolved, and one of them wins, overwriting the other
456 2011-06-25 10:14:17 <carli2> so can i branch from an earlier block and find enough blocks to become better than the current tip?
457 2011-06-25 10:17:19 <sipa> yes, if you have more than 50% of the network computing power
458 2011-06-25 10:18:57 <Sebastan> sipa: so when i would work on my own branch i could raise the chance to keep my money... so a poolowner could let his miners work on his branch to ensure that his branch will win and he keeps the money while the other one would lost his btcs.
459 2011-06-25 10:19:30 <sipa> he can't keep that up unless he has close to 50% of the network power
460 2011-06-25 10:19:59 <sipa> but it's a known and possible attack, yes
461 2011-06-25 10:20:23 <Sebastan> why is that? i mean only one of these 2 branches can survive. and one survive when a new block is found in it isnt it?
462 2011-06-25 10:20:44 <sipa> the longest one survives
463 2011-06-25 10:20:55 <Sebastan> ok...
464 2011-06-25 10:21:04 <sipa> if the rest of the world works on the longest chain, and you work on your own, you'll never keep up with the rest
465 2011-06-25 10:21:10 <sipa> unless you're close to 50%
466 2011-06-25 10:23:10 <Sebastan> sipa: I thought i have read in http://code.google.com/p/bitcoinj/source/browse/trunk/src/com/google/bitcoin/core/Block.java#210 that you can try to find a block for every part of the tree. But what you say sounds like you only have a chance to earn money with mining when you work on the longest chain...
467 2011-06-25 10:23:36 <Sebastan> line-number doesnt match...
468 2011-06-25 10:23:49 <sipa> of course you can try everywhere
469 2011-06-25 10:24:10 <sipa> and if you consider your own block the best tip, then in your own view of the world, you'll be rich
470 2011-06-25 10:24:23 <sipa> but you need to get the rest of the world to agree with you if you want to be able to spend it
471 2011-06-25 10:24:24 <Sebastan> *lol*
472 2011-06-25 10:29:09 <Sebastan> sipa:what you say sounds to me like when the speed with that you interact with the network is more important than I thought... because your best chance to find a valid block would be to get the possibly last valid block in the chain fast as hell to work on it and to spread it into the net as fast as you can again. if so it sounds to me like there are possibilities to raise your chances...
473 2011-06-25 10:31:43 <sipa> Sebastan: you're right
474 2011-06-25 10:32:23 <Sebastan> sipa: finally it looks like I understand the thing a bit... ;)
475 2011-06-25 10:36:28 <kika_> where is magicaltux?
476 2011-06-25 10:36:45 <MagicalTux> here
477 2011-06-25 10:37:17 <kika_> MagicalTux: can you restore my mtgox account? because it sent me a link telling me that the proof i provided is not enough, so i just submitted more proof
478 2011-06-25 10:37:29 <kika_> MagicalTux: i hope my account be restored soon :)
479 2011-06-25 10:37:59 <MagicalTux> kika_: there's 200ish accounts pending verification, unfortunately I'm kind of busy right now
480 2011-06-25 10:38:28 <kika_> MagicalTux: so only 200 accounts pending vertification? or 2000 or 20000 ?
481 2011-06-25 10:38:38 <kika_> i dont know what the ish means sorry hehe
482 2011-06-25 10:38:41 <MagicalTux> 200
483 2011-06-25 10:38:48 <MagicalTux> "ish" means "about"
484 2011-06-25 10:38:57 <BitcoinForNewegg> my balance is below 0
485 2011-06-25 10:38:59 <BitcoinForNewegg> :(
486 2011-06-25 10:39:28 <sipa> MagicalTux: opening is still planned for GMT 15:00 ?
487 2011-06-25 10:40:16 <kika_> MagicalTux: i see so my account is on the list of those 200 accounts, i hope it be restored soon, otherwise ill let you know
488 2011-06-25 10:40:43 <BitcoinForNewegg> wow u are doing them by hand? good luck
489 2011-06-25 10:40:53 <kika_> MagicalTux: i provieded valid proof so i dont know really why you guys are not re-opening it
490 2011-06-25 10:41:27 <BitcoinForNewegg> what if someone else provided valid proof
491 2011-06-25 10:41:29 <kika_> MagicalTux: i think a link sent to my email to allow me change my mtgox password would be eh enough proof that im the owner of the email address and mtgox account
492 2011-06-25 10:41:38 <BitcoinForNewegg> woudl you rather he just randomly pick one to give ur account too?
493 2011-06-25 10:41:57 <dinox> hmm, how do you see diff between pushes at github?
494 2011-06-25 10:42:01 <kika_> BitcoinForNewegg: no, because im the only owner of my email address i only have access there
495 2011-06-25 10:42:16 <BitcoinForNewegg> what if u used same password for email as for mtgox, or a simple one
496 2011-06-25 10:42:18 <kika_> BitcoinForNewegg: if mtgox sends an email to my email address that that proves that its me
497 2011-06-25 10:42:23 <BitcoinForNewegg> how woudl he know ur email wasent hacked
498 2011-06-25 10:42:41 <kika_> BitcoinForNewegg: i dont use the same password for different accounts
499 2011-06-25 10:42:46 <BitcoinForNewegg> just saying, some delays might be well needed
500 2011-06-25 10:42:52 <BitcoinForNewegg> but he doesnt know that
501 2011-06-25 10:42:53 <dinox> like, if I want to see diff c8bbe75...bc0cc85 in sipa's pushes
502 2011-06-25 10:43:42 <sipa> dinox: "git diff c8bbe75...bc0cc85"
503 2011-06-25 10:47:28 <dinox> thanks sipa, that implies being in showwallet branch?
504 2011-06-25 10:47:56 <sipa> no
505 2011-06-25 10:50:25 <dinox> sipa: ah, you need whole id: git diff c774b1697690c3f94ccd...bc0cc8581e67d32f2d3f
506 2011-06-25 10:50:33 <dinox> great work btw
507 2011-06-25 10:50:47 <sipa> works fine with just the 6 first hex characters here
508 2011-06-25 10:50:53 <vegard> try .. instead of ... ?
509 2011-06-25 10:56:00 <dinox> hmm, yeah worked with first 6 hex chars now
510 2011-06-25 10:56:51 <sipa> dinox: you still have the problem with the incorrect balances with the current showwallet branch?
511 2011-06-25 11:00:07 <dinox> 1 min, I need to test your current update
512 2011-06-25 11:04:14 <ericmock> vragnaroda: you around?
513 2011-06-25 11:04:31 <vragnaroda> yes
514 2011-06-25 11:05:41 <ericmock> hey, I learned something that might affect you....  I was using db5.1 which is not backwards compatible with db4.8 which the other mac clients use
515 2011-06-25 11:06:43 <ericmock> were you able to go back to wxbitcoin?
516 2011-06-25 11:07:20 <vragnaroda> ericmock: not at first; i copied the wallet and got rid of the other stuff
517 2011-06-25 11:07:38 <vragnaroda> then it re-dl'd the blockchain and everything worked
518 2011-06-25 11:07:47 <ericmock> sorry about that.
519 2011-06-25 11:08:05 <vragnaroda> ericmock: that's ok; that's what happens with alpha software
520 2011-06-25 11:08:10 <ericmock> I didn't realize that 5.1 what clandestinely upgrade the db
521 2011-06-25 11:09:10 <ericmock> so, wallet.dat was not affected?  or you have used a copy of it?
522 2011-06-25 11:09:19 <vragnaroda> one of the many reasons i think there should be a separate bitcoinlib (preferably in C or something)
523 2011-06-25 11:09:39 <ericmock> yea.
524 2011-06-25 11:09:40 <vragnaroda> ericmock: i had another copy of the wallet
525 2011-06-25 11:09:57 <ericmock> okay, good...  that's what I was mostly worried about
526 2011-06-25 11:11:07 <vragnaroda> (an ada implementation of bitcoin would be interesting)
527 2011-06-25 11:12:34 <dinox> sipa: Can you give me a privkey and corresponding address to import? I can't get the merge working with my removeprivkey function
528 2011-06-25 11:13:06 <dinox> and without that I cant generate an address without including it in wallet
529 2011-06-25 11:14:04 <dinox> *testnet-keys* of course
530 2011-06-25 11:15:45 <sipa> dinox: you run a second instance :)
531 2011-06-25 11:15:48 <sipa> *can
532 2011-06-25 11:16:14 <sipa> and i don't have any testnet funds here
533 2011-06-25 11:17:06 <dinox> hmm, I'll try that
534 2011-06-25 11:17:43 <sipa> also, is testnet usable to test with? (not sure about how often blocks are found there)
535 2011-06-25 11:17:57 <dinox> btw, you have added the pwalletMain stuff, I'm a java dev and not c++ dev, can you explain what that means?
536 2011-06-25 11:18:11 <dinox> sipa: I think it works, it did yesterday
537 2011-06-25 11:19:16 <sipa> well, i added a CWallet class to bundle all wallet operations and data structures (instead of a ton of globals)
538 2011-06-25 11:19:30 <sipa> now rpc and gui still need a 'the wallet' to interact with
539 2011-06-25 11:19:41 <sipa> pwalletMain is a pointer to 'the wallet' :)
540 2011-06-25 11:19:51 <dinox> ah
541 2011-06-25 11:20:07 <sipa> eventually, when gui and rpc becomes nice classes themselves, they'll carry their own pwallet in a field, instead of this global
542 2011-06-25 11:21:58 <sipa> dinox: how do you mean you think it works?
543 2011-06-25 11:22:38 <sipa> yesterday the same version was online as a few weeks ago, so if that caused incorrect balances for you, it should still have carried the bug yesterday
544 2011-06-25 11:22:51 <dubbz82> so
545 2011-06-25 11:23:08 <dubbz82> i quick hacked up a c# program to pull data from the tradehill api
546 2011-06-25 11:23:09 <dinox> It's the stuff with pointers I'm not completely confident with... I guess I get used to it
547 2011-06-25 11:23:12 <dubbz82> i just gotta parse it up
548 2011-06-25 11:23:19 <dubbz82> and make it into a useful program
549 2011-06-25 11:23:20 <dubbz82> :P
550 2011-06-25 11:23:41 <dinox> sipa: Yeah, the update was quite big
551 2011-06-25 11:24:00 <dubbz82> downside to that, is i gotta get going to work...so i won't be able to hack that out later
552 2011-06-25 11:26:21 <ericmock> I'm trying to figure out more about this db5.1 vs. db4.8 issue...
553 2011-06-25 11:26:47 <ericmock> if I do a db_dump on database files created with either version it shows VERSION=3
554 2011-06-25 11:27:39 <ericmock> I'm assume '3' is the database format, and would have expected different versions
555 2011-06-25 11:28:49 <vegard> is it possible for the same bitcoin address to have two (or more) base58 representations by adding/removing leading (trailing) zeroes?
556 2011-06-25 11:28:57 <sipa> no
557 2011-06-25 11:29:15 <sipa> that would change the number of bytes its decoded form has
558 2011-06-25 11:29:33 <vegard> but the decoded form is just a number, isn't it?
559 2011-06-25 11:29:43 <sipa> no, it's a byte vector
560 2011-06-25 11:29:50 <sipa> with a known length
561 2011-06-25 11:29:59 <sipa> and this length must be 25 bytes for an address
562 2011-06-25 11:30:48 <vegard> in the bitcoin code it seems to be a CBigNum which is converted to/from a byte array
563 2011-06-25 11:31:15 <sipa> yes
564 2011-06-25 11:31:31 <sipa> but adding a '1' in front, will add 0-byte to the decoded byte array
565 2011-06-25 11:53:20 <aristidesfl> ericmock: its a trojar
566 2011-06-25 11:53:22 <aristidesfl> trojan
567 2011-06-25 11:53:53 <ericmock> eh? 5.1...  those Berkeley bastards...  hmm, but I'm one of them ;-)
568 2011-06-25 11:55:00 <aristidesfl> ericmock: is it opensource?
569 2011-06-25 11:57:58 <chmod755> ;;seen gavinandresen
570 2011-06-25 11:57:58 <gribble> gavinandresen was last seen in #bitcoin-dev 19 hours, 16 minutes, and 16 seconds ago: <gavinandresen> x6763: http://forum.bitcoin.org/index.php?topic=9495.0
571 2011-06-25 12:01:20 <dinox> sipa: Ok got my function to work now but getbalance still showing incorrect amount
572 2011-06-25 12:01:41 <dinox> restart bitcoind and getbalance is correct
573 2011-06-25 12:03:02 <dinox> It's like this: getbalance=0->importprivkey ...->getbalance=0
574 2011-06-25 12:03:18 <dinox> ->restart bitcoind->getbalance=10
575 2011-06-25 12:03:59 <sipa> dinox: and getbalance "*" ?
576 2011-06-25 12:04:25 <dinox> that shows correct amount entire time
577 2011-06-25 12:04:53 <dinox> but since sendtoaddress uses GetBalance(), I cant send funds
578 2011-06-25 12:05:59 <CIA-103> DiabloMiner: Patrick McFarland master * r8aba968 / (2 files in 2 dirs): Further fixes - http://bit.ly/iVTQXn
579 2011-06-25 12:09:21 <dinox> sipa: Here is the removeprivkey commit: https://github.com/dinox/bitcoin/commit/dcf6f4066584e81e041d98ff89397d87153a3046
580 2011-06-25 12:09:38 <dinox> which has the same bug as your importprivkey
581 2011-06-25 12:10:35 <sipa> before the import, is there *any* tx in the wallet?
582 2011-06-25 12:11:49 <dinox> sipa: no
583 2011-06-25 12:12:10 <dinox> listtransactions returns empty array
584 2011-06-25 12:13:03 <sipa> after import, how many confirmations to your tx?
585 2011-06-25 12:13:11 <dinox> 344
586 2011-06-25 12:14:12 <sipa> does using importwallet have the same problem?
587 2011-06-25 12:14:35 <dinox> sipa: I haven't tried that
588 2011-06-25 12:15:35 <sipa> so, create a file with as contents: '{"keys":[{"sec":"5..."}]}', and pass that filename to ./bitcoin importwallet <filename>
589 2011-06-25 12:19:06 <dinox> sipa: http://paste.pocoo.org/show/419137/
590 2011-06-25 12:19:52 <sipa> so, same problem?
591 2011-06-25 12:19:58 <dinox> yep
592 2011-06-25 12:20:46 <dinox> restart fixes it here too
593 2011-06-25 12:21:15 <kika_> does anyone know how i can mark bad sectors on a hard disk ? i want to basically mark those bad sectors as wrong so i can continue using the hard disk
594 2011-06-25 12:21:55 <andyfletcher> if you are getting bad sectors the disk is on the way out. Bad sector counts will rapidly increase
595 2011-06-25 12:22:02 <sipa> kika_: if your hard drive is less than 10 years old,  it will do that automatically, and start using spare sectors
596 2011-06-25 12:22:17 <sipa> if you still see them, it has run out of spare sectors and you should throw the drive away
597 2011-06-25 12:22:20 <kika_> sipe: yes it uses SMART
598 2011-06-25 12:22:34 <kika_> sipa: it uses SMART right?
599 2011-06-25 12:22:45 <sipa> smart is a way to query that information yes
600 2011-06-25 12:22:47 <kika_> sipe: why throw the drive away?
601 2011-06-25 12:23:15 <sipa> because it will soon become worse
602 2011-06-25 12:23:30 <kika_> sipa: seems like it ran out of spare sectors i think because linux is throwing me the SMART bad sectors warning, it says the hd has many bad sectors, backup data and replace disk
603 2011-06-25 12:23:39 <sipa> kika_: then do so
604 2011-06-25 12:23:42 <kika_> sipa: isnt anything i can do to continue using the disk ?
605 2011-06-25 12:23:49 <sipa> you can, but it's very risky
606 2011-06-25 12:24:01 <kika_> sipa: what risk? i dont mind to format
607 2011-06-25 12:24:10 <sipa> no, risk losing your data
608 2011-06-25 12:24:17 <ericmock> I swapped libdb5.1 for libdb4.8 and now I get Bitcoin: Error loading addr.dat; Error loading blkindex.dat; Error loading wallet.dat
609 2011-06-25 12:24:19 <kika_> sipa: i dont have any valuable data
610 2011-06-25 12:24:20 <sipa> some filesystems have a way for marking sectors as bad
611 2011-06-25 12:24:29 <ericmock> even though those are the files freshly created
612 2011-06-25 12:24:30 <kika_> sipa: on linux you know how i can do that?
613 2011-06-25 12:24:44 <edcba> not having valuable data ???
614 2011-06-25 12:24:52 <kika_> sipa: i dont mind if i loose data, i dont have any data on it, i can just format it
615 2011-06-25 12:25:09 <kika_> sipa: i just have a fresh fedora install on it
616 2011-06-25 12:25:10 <edcba> the problem is that it may corrupt os silently
617 2011-06-25 12:25:28 <kika_> i had to reinstall fedora because i wasnt able to boot it anymore
618 2011-06-25 12:25:35 <sipa> dinox: can you modify CWallet::GetBalance() is wallet.cpp to print out pcoin->IsFinal,->IsConfirmed and ->GetAvailableCredit ?
619 2011-06-25 12:25:39 <sipa> *in
620 2011-06-25 12:25:41 <kika_> probably because the os got corrupt silently
621 2011-06-25 12:25:42 <ericmock> I think this might be why I moved to libdb5.1 in the first place.  Anyone have any idea what the problem might be?
622 2011-06-25 12:26:14 <sipa> dinox: i'd like to know why those tx's aren't counted while they are obviously in the wallet
623 2011-06-25 12:27:14 <edcba> not booting anymore is not what i called 'silent' :)
624 2011-06-25 12:27:30 <ericmock> this db crap is pissing me off...  I'm moving back to 5.1
625 2011-06-25 12:28:19 <dinox> sipa: sure
626 2011-06-25 12:38:46 <sipa> dinox: any result?
627 2011-06-25 12:40:38 <CIA-103> DiabloMiner: Patrick McFarland master * rc4ed37f / src/main/resources/DiabloMiner.cl : Final fix, frakenkernel isa go - http://bit.ly/lneY9A
628 2011-06-25 12:40:50 <vegard> so it seems to me that if you pushed the signature using OP_PUSHDATA, then bitcoin would not be able to verify it correctly
629 2011-06-25 12:41:08 <sipa> why not?
630 2011-06-25 12:41:26 <vegard> because it does: scriptCode.FindAndDelete(CScript(vchSig));
631 2011-06-25 12:41:45 <vegard> let me refine my statement. if you push a signature of
632 2011-06-25 12:42:00 <sipa> ah right
633 2011-06-25 12:42:05 <vegard> ~75 bytes using OP_PUSHDATA*, then CScript(vchSig) would not use the OP_PUSHDATA opcode
634 2011-06-25 12:42:10 <sipa> indeed
635 2011-06-25 12:42:32 <sipa> that findanddelete hack is really horrible
636 2011-06-25 12:42:34 <dinox> sipa: no, I cant get it to write out anything
637 2011-06-25 12:42:52 <sipa> dinox: also nothing in debug.log ?
638 2011-06-25 12:42:54 <dinox> i run it with -debug flag and tail -f debug.log
639 2011-06-25 12:42:56 <dinox> no
640 2011-06-25 12:43:41 <sipa> could you paste the code you added?
641 2011-06-25 12:43:44 <vegard> I suppose there should be some sort of test suite for this kind of thing. not so much to test bitcoin, but to test other implementations
642 2011-06-25 12:44:16 <sipa> dinox: sorry i'm not helping to do the test myself, but i'm kinda doing other things as well now
643 2011-06-25 12:45:05 <dinox> sipa: np, I want to learn some on the way too
644 2011-06-25 12:46:11 <sipa> anyway, it's hard to believe getbalance isn't called or doesn't perform the loop over wallet transactions, and listtransactions shows they are there
645 2011-06-25 12:46:30 <sipa> so i'd like to see how you tried to output those pcoin properties
646 2011-06-25 12:50:24 <dinox> sipa: printf("GetBalance() %s %s %s",pcoin->IsFinal(),pcoin->IsConfirmed(),pcoin->GetAvailableCredit());
647 2011-06-25 12:50:51 <dinox> at line 560 in wallet.cpp
648 2011-06-25 12:51:26 <sipa> line numbers are hard to match up
649 2011-06-25 12:51:53 <dinox> sipa: http://paste.pocoo.org/show/419148/
650 2011-06-25 12:52:28 <sipa> try turning that into %i %i %llu
651 2011-06-25 12:52:37 <sipa> the returned things aren't strings
652 2011-06-25 12:53:18 <dinox> hmm, it seems to work when I'm not using testnet
653 2011-06-25 12:53:45 <sipa> define "work"?
654 2011-06-25 12:54:13 <dinox> debug output*
655 2011-06-25 12:54:42 <sipa> eh?
656 2011-06-25 12:55:52 <dinox> ie, bitcoind actually writes anything to debug.log when running ./bitcoind -debug -daemon but not when running ./bitcoind -debug -daemon -testnet
657 2011-06-25 12:57:18 <sipa> are you looking at the right debug.log ?
658 2011-06-25 12:58:19 <dinox> damn, testnet has separate dir, right?
659 2011-06-25 12:58:44 <sipa> yes
660 2011-06-25 12:59:37 <sipa> subdir 'testnet' of the real directory
661 2011-06-25 13:02:08 <dinox> okay, I get "GetBalance() 1 1 0"
662 2011-06-25 13:02:38 <dinox> on each of the 2 txes
663 2011-06-25 13:06:47 <sipa> dinox: in CWallet::AddToWallet() in wallet.cpp, above '// Write to disk', add the line: "if (fInsertedNew) wtx.MarkDirty();"
664 2011-06-25 13:08:15 <dinox> sipa: okay, compiling...
665 2011-06-25 13:11:39 <dinox> sipa: Same result, getbalance=0 and "GetBalance() 1 1 0"
666 2011-06-25 13:12:10 <sipa> ok, can you also output pcoin->GetCredit() ?
667 2011-06-25 13:12:29 <dinox> sure
668 2011-06-25 13:12:43 <dinox> btw, why doesn't it rescan from genesisblock?
669 2011-06-25 13:13:21 <sipa> heh?
670 2011-06-25 13:14:10 <sipa> oh, that's at startup
671 2011-06-25 13:14:26 <badmanu> shouldnt mt.gox be online for 14 minutes now? did i miss something?
672 2011-06-25 13:14:45 <sipa> dinox: the rescan done by importprivkey isn't reported
673 2011-06-25 13:15:02 <dinox> sipa: But I'm using importwallet now
674 2011-06-25 13:15:46 <sipa> oh, right
675 2011-06-25 13:16:19 <sipa> if the dumpfile doesn't contain information about block heights, it assumes there are no former transactions, and it won't rescan
676 2011-06-25 13:16:26 <sipa> can you test with importprivkey too?
677 2011-06-25 13:17:34 <dinox> yeah, http://paste.pocoo.org/show/419168/
678 2011-06-25 13:20:10 <JRWR> it sure is taking some time for a encrypted wallet.dat feature to come into main branch
679 2011-06-25 13:21:00 <sipa> JRWR: we're working on it
680 2011-06-25 13:21:10 <sipa> dinox: just to be sure, pcoin->GetAvailableCredit(false) and pcoin->GetCredit(false) ?
681 2011-06-25 13:21:56 <JRWR> sipa: if you need a win32 beta tester, I have a strange way of always finding little bugs in the system :)
682 2011-06-25 13:21:58 <erus`> JRWR: how is the wallet encrypted? does the user have to type a password when they start bitcoin?
683 2011-06-25 13:22:23 <sipa> erus`: yes
684 2011-06-25 13:22:24 <BlueMatt> JRWR: well, Im rewriting it, and since I just got back from vacation, it means its gonna take a bit
685 2011-06-25 13:22:32 <BlueMatt> (well, Im just starting rewrite today)
686 2011-06-25 13:22:38 <BlueMatt> well, rewrite #2
687 2011-06-25 13:22:40 <erus`> sipa:  how long would that take to crack?
688 2011-06-25 13:22:53 <JRWR> erus`: thats up to the devs, but i could see it being a timeout to reencrypt
689 2011-06-25 13:23:00 <BlueMatt> erus`: the idea if you have to enter password at send-time
690 2011-06-25 13:23:12 <sipa> erus`: if your password is secure, cracking will be very hard
691 2011-06-25 13:23:14 <BlueMatt> or you can enter password once on rpc, and tell it to store it for x seconds
692 2011-06-25 13:23:33 <erus`> that sounds like a good feature
693 2011-06-25 13:23:43 <erus`> back to my telnet bitcoin bank :)
694 2011-06-25 13:23:55 <JRWR> lol
695 2011-06-25 13:24:09 <JRWR> erus`: anon SSH? or pure telnet?
696 2011-06-25 13:24:35 <dinox> sipa: wtf... it seems working now
697 2011-06-25 13:24:36 <erus`> pure telnet. resistant to css/sql injection and more
698 2011-06-25 13:24:47 <JRWR> what about sniffing?
699 2011-06-25 13:25:04 <erus`> dont let people sniff your connection
700 2011-06-25 13:25:05 <JRWR> telnet is send in the clear, no one will like that
701 2011-06-25 13:25:06 <sipa> dinox: that's possible, that false means it should disregard the cached value, and recompute it
702 2011-06-25 13:25:26 <dinox> ah, I think we have it there then
703 2011-06-25 13:25:52 <dinox> both import and remove seems to work fine now
704 2011-06-25 13:25:56 <JRWR> erus`: you should be able to make a anon ssh session,
705 2011-06-25 13:26:15 <JRWR> erus`: its why SSL exists :)
706 2011-06-25 13:26:35 <erus`> jrwr im gonna make a website in the end but for now im just setting up a telnet thing to make sure it work
707 2011-06-25 13:26:39 <erus`> ssh is a good idea
708 2011-06-25 13:27:26 <JRWR> the website is where you can set password or give pubkey
709 2011-06-25 13:27:36 <JRWR> I would op for default to pubkey
710 2011-06-25 13:30:52 <Kiba`> hey guys
711 2011-06-25 13:30:56 <Kiba`> got a funny comic for ya
712 2011-06-25 13:31:05 <Kiba`> http://www.bitcoinweekly.com/comics/bitsquirrel
713 2011-06-25 13:31:09 <Kiba`> Bitsquirrel!
714 2011-06-25 13:31:27 <wump> lol Kiba`
715 2011-06-25 13:32:46 <erus`> lol
716 2011-06-25 13:33:34 <kika_> does anyone know how i can shutdown x on fedora 14 ?
717 2011-06-25 13:37:56 <kiba> wump, erus`: maybe you guys should donate some bitcents http://forum.bitcoin.org/index.php?topic=22023.0
718 2011-06-25 13:38:09 <wump> yep i will
719 2011-06-25 13:38:18 <erus`> i have 0.011 bitcoins
720 2011-06-25 13:38:33 <erus`> and that took me days, so i cant donate anything yet
721 2011-06-25 13:40:19 <JFK911> MagicalTux: nice ticker.php; firefox wants to 'save as'
722 2011-06-25 13:40:40 <JRWR> save it
723 2011-06-25 13:40:44 <wump> kiba: done
724 2011-06-25 13:40:46 <JRWR> and steal db password
725 2011-06-25 13:40:59 <erus`> hurrr
726 2011-06-25 13:58:56 <upb> sup
727 2011-06-25 14:00:45 <kiba> upb: you see my bitcoin comic?
728 2011-06-25 14:01:08 <ericmock> libdb sux
729 2011-06-25 14:01:23 <upb> looking :)
730 2011-06-25 14:03:01 <upb> haha kiba
731 2011-06-25 14:03:38 <Happy0> kiba: link? :P
732 2011-06-25 14:03:45 <jgarzik> sipa: I just came up with an algorithm that Hal and other people on the thread thought was OK
733 2011-06-25 14:03:46 <Happy0> oh wait
734 2011-06-25 14:03:50 <Happy0> just saw it, sorry
735 2011-06-25 14:03:56 <Happy0> the link that is
736 2011-06-25 14:04:09 <Happy0> hahaha
737 2011-06-25 14:04:10 <jgarzik> sipa: the goal was something computationally expensive to brute force
738 2011-06-25 14:04:10 <luke-jr> so is there actually code for deterministic wallets? or just a theory?
739 2011-06-25 14:04:11 <Happy0> very nice
740 2011-06-25 14:04:17 <kiba> Happy0: http://bitcoinweekly.com/comics/bitsquirrel
741 2011-06-25 14:04:28 <Happy0> yeah kiba, like i said, i scrolled up and saw it... good comic =p
742 2011-06-25 14:04:38 <sipa> jgarzik: and why not just iterated sha512 ?
743 2011-06-25 14:04:50 <kiba> and don't forget to donate a few bitcents if you like it: http://forum.bitcoin.org/index.php?topic=22023.0
744 2011-06-25 14:05:29 <Happy0> kiba: haha, i wish i had any to donate =p
745 2011-06-25 14:05:51 <Happy0> alas i have nothing
746 2011-06-25 14:05:51 <jgarzik> sipa: it's nice to use multiple algorithms, in case one has problems
747 2011-06-25 14:07:33 <jgarzik> sipa: remember that scratch card may have as little as 64 bits of password.  with that few bits, I wanted something difficult to brute force
748 2011-06-25 14:08:22 <sipa> sure, but that's a useful thing to have in any case
749 2011-06-25 14:17:54 <ericmock> jgarzik: what's your take on the db5.x vs. db4.x issue.  I can't seem to get anything to work using 4.x
750 2011-06-25 14:19:06 <ericmock> whenever I link to 4.8 and have bitcoin generate new databases, it generates them but then quits complaining Bitcoin: Error loading addr.dat, etc.
751 2011-06-25 14:20:31 <ericmock> seems like it can't load the database it just generated.  Is there something else that needs swapped besides libdb_cxx-5.1.dylib with libdb_cxx-4.8.dylib?
752 2011-06-25 14:25:46 <BTCOE__> Does anyone know anything about KYC requirements for Bitcoin related projects that also handles USD?