1 2012-04-21 00:35:24 <copumpkin> is it normal for the "splash screen" of bitcoin-qt to be sitting around behind the main window forever? (on mac os)
  2 2012-04-21 00:36:03 <luke-jr> copumpkin: I don't think so O.o
  3 2012-04-21 00:36:21 <copumpkin> http://snapplr.com/14pf
  4 2012-04-21 00:36:26 <copumpkin> I just noticed when moving my window out of the way
  5 2012-04-21 00:37:17 <Joric> i started to skip ';', damn python
  6 2012-04-21 00:38:18 <luke-jr> copumpkin: report a bug
  7 2012-04-21 00:40:51 <copumpkin> luke-jr: hmm, when I restarted the program it didn't stick around
  8 2012-04-21 00:41:10 <luke-jr> copumpkin: do you use virtual desktops? maybe related
  9 2012-04-21 00:41:46 <copumpkin> well, I have an extra desktop, but the bitcoin window never moved off the main one
 10 2012-04-21 00:42:05 <luke-jr> copumpkin: maybe you were on the other one while it was starting?
 11 2012-04-21 00:42:11 <copumpkin> could be, I guess
 12 2012-04-21 00:42:31 <copumpkin> I'll experiment with it a bit
 13 2012-04-21 00:43:27 <copumpkin> aha, yeah
 14 2012-04-21 00:43:35 <copumpkin> if I switch between desktops while it loads, it gets confused
 15 2012-04-21 00:44:04 <copumpkin> where do bitcoin-qt bug reports belong?
 16 2012-04-21 00:44:18 <luke-jr> copumpkin: Github has an issue tracker
 17 2012-04-21 00:44:55 <copumpkin> ah, thanks
 18 2012-04-21 00:54:27 <etotheipi_> How did Bitcoin decide that port 8333 and 18333 were good ports to use?  and how did they know they weren't going to be in use by other processes?
 19 2012-04-21 00:56:30 <k9quaint> many coin flips were involved in the selection
 20 2012-04-21 00:56:44 <luke-jr> lol\n3387630
 21 2012-04-21 00:57:16 <etotheipi_> luke-jr: I've been doing that
 22 2012-04-21 00:57:31 <etotheipi_> but I don't know if it's bad etiquette to just take one
 23 2012-04-21 00:57:37 <luke-jr> it is
 24 2012-04-21 00:57:40 <etotheipi_> or if there's some intelligent process to use
 25 2012-04-21 00:57:41 <luke-jr> you're supposed to ask the IANA for one
 26 2012-04-21 00:57:52 <luke-jr> but good luck with that
 27 2012-04-21 00:58:07 <etotheipi_> ahh... okay so I'll just do some coinflips then
 28 2012-04-21 00:59:34 <gmaxwell> nope.
 29 2012-04-21 01:00:10 <gmaxwell> http://www.wolframalpha.com/input/?i=primes+%3C+65536
 30 2012-04-21 01:00:49 <gmaxwell> 65521 seems not taken, enh?
 31 2012-04-21 01:01:08 <luke-jr> but that's in the whole "temporary port numbers for outbound connections" rnge
 32 2012-04-21 01:02:15 <gmaxwell> well the largest prime < 1024 is 1021. less than 32768 is 32749.
 33 2012-04-21 01:04:36 <etotheipi_> is there a maximum port number I should be using?
 34 2012-04-21 01:05:40 <SomeoneWeird> 65535
 35 2012-04-21 01:06:19 <etotheipi_> so I was using 63331 ... someone had issues with it in Windows
 36 2012-04-21 01:06:34 <etotheipi_> they reported that everything about 62200 fails
 37 2012-04-21 01:06:48 <etotheipi_> and 61199 and lower succeeds
 38 2012-04-21 01:07:00 <SomeoneWeird> well... that's windows
 39 2012-04-21 01:07:01 <SomeoneWeird> :P
 40 2012-04-21 01:07:03 <etotheipi_> s/about/above/g
 41 2012-04-21 01:07:33 <etotheipi_> p.s. port 822888 worked in linux
 42 2012-04-21 01:08:37 <gmaxwell> etotheipi_: I'm sure you got 822888%65536 or something like that.
 43 2012-04-21 01:09:01 <etotheipi_> ahhh
 44 2012-04-21 01:20:47 <luke-jr> etotheipi_: you want something between 1024 and 32768
 45 2012-04-21 01:21:01 <luke-jr> below 1024 is reserved for root
 46 2012-04-21 01:21:08 <luke-jr> above 32768 is reserved for outgoing connections
 47 2012-04-21 01:24:55 <etotheipi_> ahh...good to know!
 48 2012-04-21 02:22:28 <phantomcircuit> bleh
 49 2012-04-21 02:22:44 <phantomcircuit> there is some weird bug with downloading the blockchain were the client gets offsync from the other side
 50 2012-04-21 02:22:49 <phantomcircuit> and every block is an orhan
 51 2012-04-21 02:22:51 <phantomcircuit> orphan
 52 2012-04-21 02:23:19 <phantomcircuit> i think it happens when you're in the middle of an initial download and a new block is generated
 53 2012-04-21 02:23:33 <gmaxwell> phantomcircuit: its harmless at least.
 54 2012-04-21 02:23:47 <phantomcircuit> well yes and no
 55 2012-04-21 02:23:48 <gmaxwell> just leaves messy logs.
 56 2012-04-21 02:23:52 <phantomcircuit> nothing breaks but it's slooow
 57 2012-04-21 02:24:00 <phantomcircuit> i have no idea why
 58 2012-04-21 02:24:24 <gmaxwell> I didn't think it was actually slow. You sure that its not just the normal slowness of the late in chain syncing?
 59 2012-04-21 02:24:44 <phantomcircuit> pretty sure
 60 2012-04-21 02:24:52 <phantomcircuit> it's a lot faster now
 61 2012-04-21 02:25:00 <phantomcircuit> although that could just be luck of the draw faster peer
 62 2012-04-21 02:26:33 <phantomcircuit> ;;bc,blocks
 63 2012-04-21 02:26:44 <gribble> 176551
 64 2012-04-21 02:26:49 <phantomcircuit> huh
 65 2012-04-21 02:26:54 <phantomcircuit> im a thousand blocks behind
 66 2012-04-21 02:27:04 <phantomcircuit> but payments i made today are showing confirmed
 67 2012-04-21 02:27:18 <phantomcircuit> oh
 68 2012-04-21 02:27:22 <phantomcircuit> right nvm that makes sense
 69 2012-04-21 04:02:54 <Steve__> hi does anyone know an easy way to extract my public keys from my wallet
 70 2012-04-21 04:03:14 <Steve__> I don't want the hash, I'd like to import the full byte array into another program
 71 2012-04-21 04:08:42 <Joric> Steve, https://github.com/joric/pywallet replace public_key_to_bc_address(x) to x.encode('hex')
 72 2012-04-21 04:09:15 <Steve__> thanks joric, i'll check it out
 73 2012-04-21 05:15:40 <dwon> etotheipi_: From http://www.iana.org/assignments/service-names-port-numbers/service-names-port-numbers.xml:
 74 2012-04-21 05:15:58 <dwon> Port numbers are assigned in various ways, based on three ranges: System
 75 2012-04-21 05:16:00 <dwon> are assigned by IANA using the "Expert Review" process, as per
 76 2012-04-21 05:16:02 <dwon> [RFC6335].  Dynamic Ports are not assigned.
 77 2012-04-21 05:16:28 <dwon> The registration procedures for service names and port numbers are
 78 2012-04-21 05:16:29 <dwon> described in [RFC6335].
 79 2012-04-21 05:59:38 <da2ce7> Updated Builds:
 80 2012-04-21 05:59:39 <da2ce7> https://bitcointalk.org/index.php?topic=77301
 81 2012-04-21 06:32:52 <gribble> New news from bitcoinrss: laanwj opened pull request 1129 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/1129>
 82 2012-04-21 06:34:19 <gmaxwell> If anyone cares: I have updated fedora 16 rpms for openssl with the recent openssl cve fix in them (plus enabling ECDSA, of course): http://people.xiph.org/~greg/openssl/fedora16/
 83 2012-04-21 06:35:03 <gmaxwell> (fedora hasn't actually released the fix yet it's still in their updates testing pipeline, my rpm is just theirs with ECDSA reenabled and -DPURIFY )
 84 2012-04-21 06:35:24 <gmaxwell> (hmph, https link is better since people won't check signatures: https://people.xiph.org/~greg/openssl/fedora16/ )
 85 2012-04-21 06:55:24 <seco> <luke-jr> so N-of-M not-necessarily-developers, paid a reasonable fee, with a week/month long warning to all clients with the key being recovered :P
 86 2012-04-21 06:55:52 <seco> i thought about that idea of wallet.dat recovery again, and find any kind of centralisation very critical
 87 2012-04-21 06:58:21 <seco> but sending an alert to the sending addresses of all coins received on the wallet (e.g. by storing last m used public addresses from where coins came from) with some kind of OTP, it should be enough to say: get N<M of those guys cou received coins together who give you the OTP, and you can recover your lost wallet.dat by -recover OTP1 OTP2 OTP3 ...  : wallet-owner and senders can only come together successfully if the wallet-owner is still the same, is
 88 2012-04-21 06:58:37 <seco> (OTP= OneTimePassword)
 89 2012-04-21 06:59:51 <seco> would need some kind of BIP, with that "user initited alert of wallet-recovery with generated bunch of OTPs", but however it is not bound to a central point
 90 2012-04-21 07:02:45 <seco> like only by getting 30% of the last Bitcoinusers from which you got your coins last, giving you theit wallet.dat-OTP for you, you could recover your lost wallet.dat
 91 2012-04-21 07:08:29 <seco> some kind of bitcoinaddress-to-bitcoinaddress communication; Bitcoin chat lol
 92 2012-04-21 09:14:38 <gribble> New news from bitcoinrss: laanwj opened issue 1130 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/issues/1130>
 93 2012-04-21 13:09:10 <gribble> 176613
 94 2012-04-21 13:09:10 <sipa> ;;bc,blocks
 95 2012-04-21 14:56:22 <Diablo-D3> everyone just sign this damned thing so we can fuck the banks https://wwws.whitehouse.gov/petitions/#!/petition/push-fannie-mae-and-freddie-mac-issue-principal-reductions-underwater-homeowners/qtS3crg7
 96 2012-04-21 15:38:28 <t7> Diablo-D3: so they can post another stupid responce
 97 2012-04-21 15:39:06 <k9quaint> Diablo-D3: sorry, I bought mortgage backed securities after they crashed
 98 2012-04-21 15:39:13 <k9quaint> I don't want principle reductions
 99 2012-04-21 15:39:38 <k9quaint> let them foreclose instead, screws the banks and freddie, bondholders get paid ;)
100 2012-04-21 15:40:27 <Diablo-D3> hehhhhhh
101 2012-04-21 15:43:54 <t7> entangled particles are the most stupid thing ever
102 2012-04-21 15:44:06 <t7> once you know the state of 1 know know the state of both
103 2012-04-21 15:44:10 <t7> but you cant set it
104 2012-04-21 15:44:22 <t7> good for random numbers i guess
105 2012-04-21 16:09:23 <_W_> t7, not really - there's no need to entangle particles for randomness. If there's something unknown about a single particle, that's just as good as something unknown about two entangled particles
106 2012-04-21 16:09:46 <t7> distributed randomness ?
107 2012-04-21 16:09:58 <_W_> entanglement is just a "hey, look, quantum mechanics is counterintuitive", it can't actually be /used/ for anything
108 2012-04-21 16:10:19 <_W_> t7, pseudorandom does the job
109 2012-04-21 16:10:25 <_W_> just start out with the same seed
110 2012-04-21 16:10:45 <gmaxwell> _W_: well, quantum key distribution is at least kinda nifyt on theoretical grounds, even if it does totally fail on pratical engineering grounds.
111 2012-04-21 16:10:48 <_W_> also doesn't involve any physical processes, you can transmit the seed as plain information
112 2012-04-21 16:11:45 <_W_> gmaxwell, I don't see how. The security has to be the same as for any physical key, and you can just then actually /have/ a physical key and the entanglement is superfluous
113 2012-04-21 16:11:50 <t7> I did some crypto stuff for playing poker without a dealer
114 2012-04-21 16:12:15 <t7> using commutative crypto, it was really cool 'mental poker'
115 2012-04-21 16:12:42 <kakobrekla> anyone seen this before?
116 2012-04-21 16:12:43 <kakobrekla> http://shrani.si/f/1b/qa/SWkTNoX/untitled.png
117 2012-04-21 16:12:48 <_W_> t7, yep, very interesting that you can in fact play fair poker without trust
118 2012-04-21 16:12:57 <gmaxwell> _W_: because no small seeded crypto system is provably secure, they all depend on the apparent hardness of varrious cipher structures.. and they do get broken from time to time.
119 2012-04-21 16:13:26 <gmaxwell> (one time pad is but then you have a distribution problem with the pad)
120 2012-04-21 16:13:41 <_W_> gmaxwell, I mean, you can use whatever quantum data you need *at the origin* (the place where you would create the entanglement in the first place) and then the entanglement is superfluous, as you can just securely encode it
121 2012-04-21 16:14:07 <_W_> this is independent, with no relevance, to my comment to t7 about pseudorandomness
122 2012-04-21 16:14:47 <gmaxwell> _W_: oh no, thats not so. The QKD channel is interception proof. (assuming the system doesn't have implementation bugs)  Effectively QKD gives you a DH like primitive which depends on the security of shoddy engineering vs shoddy math. ;)
123 2012-04-21 16:14:56 <_W_> if you  have some device containing entangled particles that can be read, said device must be protected just as good as, say, magnetic media that has the key encoded
124 2012-04-21 16:15:34 <_W_> gmaxwell, there is no channel. Information wise it is equivalent to two pieces of paper with the same writing on them separated by spacetime
125 2012-04-21 16:16:23 <gmaxwell> _W_: indeed, but the paper can keep growing forever, without risk of your CPRNG being compromised.
126 2012-04-21 16:16:31 <gmaxwell> (instead your QKD hardware gets compromised)
127 2012-04-21 16:16:36 <graingert> _W_: no it only needs to be protected while in use
128 2012-04-21 16:16:52 <graingert> _W_: OTP needs to be protected during delivery
129 2012-04-21 16:16:58 <_W_> erm, you're saying some entanglement with non-finite information content? I am not aware of anything such being backed by current theory
130 2012-04-21 16:17:00 <gmaxwell> Vs a one time pad where some annoying courier has to keep carrying more paper.
131 2012-04-21 16:17:09 <gmaxwell> _W_: Here, read: http://en.wikipedia.org/wiki/Quantum_key_distribution#BB84_protocol:_Charles_H._Bennett_and_Gilles_Brassard_.281984.29
132 2012-04-21 16:17:17 <graingert> _W_: I don't think QKD uses entanglement
133 2012-04-21 16:17:20 <graingert> just polarization
134 2012-04-21 16:17:42 <gmaxwell> graingert: ... yes it does. hush. :)
135 2012-04-21 16:17:44 <_W_> it starts out badly, talking of using entanglement to "transmit the information"
136 2012-04-21 16:18:02 <_W_> graingert, polarization is one of the properties that can be entangled
137 2012-04-21 16:18:49 <gmaxwell> _W_: yea yea, ignore the crappy writing, it's wikipedia. I probably should have just linked you to a paper on the protocol, but I figured everyone _except_ you would get techbabbled at anything proper.
138 2012-04-21 16:19:16 <graingert> I was talking about BB84 protocol
139 2012-04-21 16:19:31 <graingert> which just uses non-orthogonal states
140 2012-04-21 16:20:41 <graingert> According to quantum mechanics (particularly quantum indeterminacy), no possible measurement distinguishes between the 4 different polarization states, as they are not all orthogonal. The only possible measurement is between any two orthogonal states (an orthonormal basis). So, for example, measuring in the rectilinear basis gives a result of horizontal or vertical. If the photon was created as horizontal or vertical (as a rectilinear eig
141 2012-04-21 16:20:43 <graingert> enstate) then this measures the correct state, but if it was created as 45??? or 135??? (diagonal eigenstates) then the rectilinear measurement instead returns either horizontal or vertical at random. Furthermore, after this measurement the photon is polarized in the state it was measured in (horizontal or vertical), with all information about its initial polarization lost.
142 2012-04-21 16:23:43 <graingert> although it's just a theory
143 2012-04-21 16:24:07 <_W_> gmaxwell, that seems to me to be independent of entanglement
144 2012-04-21 16:24:13 <graingert> yup
145 2012-04-21 16:27:48 <_W_> the no-cloning theorem only forbids the cloning of an unknown state. You can always just /read/ the state, /then/ clone it
146 2012-04-21 16:28:45 <_W_> (yes, this prevents you from giving the device on to its intended recipient - but you can just pretend to be that recipient now, so why would you bother)
147 2012-04-21 16:40:47 <graingert> _W_: the public channel has to be absolutely public
148 2012-04-21 16:40:55 <graingert> _W_: so recording it in the blockchain would do it
149 2012-04-21 16:42:56 <gmaxwell> _W_: yes, bb84 is but E91 is not.
150 2012-04-21 16:44:10 <gmaxwell> _W_: as I said, it's effectively DH  with the same kind of properties except it depends on the engineering of your hardware rather than the believed hardness of some mathmatical puzzle.
151 2012-04-21 16:51:21 <gribble> New news from bitcoinrss: laanwj opened pull request 1131 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/1131>
152 2012-04-21 17:15:03 <gmaxwell> :(  0.4% of binary downloaders download the signature files on sourceforge.
153 2012-04-21 17:17:50 <lianj> ^^
154 2012-04-21 17:29:09 <t7> i build from source and inspect every line
155 2012-04-21 17:29:48 <t7> if they can change your sourceforge files, cant they change the sigs?
156 2012-04-21 17:33:26 <gmaxwell> t7: ...
157 2012-04-21 17:33:36 <gmaxwell> Thats why they're signatures no, they can't change them.
158 2012-04-21 17:34:13 <t7> dont i need your public key?
159 2012-04-21 17:34:38 <gmaxwell> t7: Yes, but you can verify public keys out of band and hold them over time.
160 2012-04-21 17:37:20 <gmaxwell> e.g. if you've validated almost any public key you can find a chain to give you high confidence of mine:  e.g. here is the path between me and the key used to sign OpenSSL: http://webware.lysator.liu.se/jc/wotsap/wots/latest/paths/0xB0413BFA-0xF295C759.png?size=2000x700
161 2012-04-21 17:38:22 <gmaxwell> It actually should be shorter, because I know Bradley Kuhn (he's in the third level up from the bottom) and I are cross signed, but for some reason this tool doesn't know that.
162 2012-04-21 18:25:32 <Diablo-D3> https://bitcointalk.org/index.php?topic=77469.msg862654#msg862654
163 2012-04-21 19:18:49 <graingert> gmaxwell: what's the key for the arrows?
164 2012-04-21 19:19:01 <graingert> key as in color->semantic
165 2012-04-21 19:19:11 <graingert> not pub lickey
166 2012-04-21 19:19:55 <graingert> http://webware.lysator.liu.se/jc/wotsap/wots/latest/paths/0xF295C759-0xB0413BFA.png?size=2000x700
167 2012-04-21 19:20:35 <gmaxwell> ah, it's higlighting the strongest path and the strongest path (no hops in common) which is completely orthorgonal with the first path.
168 2012-04-21 19:20:47 <MC1984> gmaxwell how do i verify your builds against the signature?
169 2012-04-21 19:20:54 <graingert> gmaxwell: an arrow from a-> b means a signs b
170 2012-04-21 19:21:32 <gmaxwell> graingert: yes.  Ah, actually on the graph you linked to there is only one path (from him to me)
171 2012-04-21 19:22:02 <graingert> I can spot 4 paths
172 2012-04-21 19:22:03 <gmaxwell> oh. hm. no there are two in that direction, I don't know why it didn't make both green.
173 2012-04-21 19:22:12 <graingert> green is the trust level
174 2012-04-21 19:22:33 <gmaxwell> graingert: I'm counting totally orthorgonal paths (no two people in common)
175 2012-04-21 19:22:40 <graingert> I see
176 2012-04-21 19:23:10 <gmaxwell> (otherwise when paths have someone in common that someone could be faking it and you couldn't tell)
177 2012-04-21 19:23:10 <graingert> any faqs on how to get into the WOT
178 2012-04-21 19:23:42 <gmaxwell> graingert: get signed by someone else in it.. There are keysigning events at most free software conferences, linux users groups, network operators events (like nanog).
179 2012-04-21 19:23:54 <graingert> ie are there any useful keysigining parties
180 2012-04-21 19:24:09 <gmaxwell> You create a key with your real identity on it, and you show up with slips of paper with the fingerprint on it.. show people your ID, and they later sign your key and you do the same in reverse.
181 2012-04-21 19:24:24 <graingert> I have been to a mini one in my uni
182 2012-04-21 19:24:30 <graingert> (very mini it was two people)
183 2012-04-21 19:24:39 <gmaxwell> there are many about 60% of people at keysignings flake out and don't bother finishing it... but they work all the same.
184 2012-04-21 19:25:23 <gmaxwell> graingert: is there a linux users group near you? you can nag them to hold a keysigning and if they haven't done one in a long time they probably will.
185 2012-04-21 19:25:41 <graingert> the slips of paper thing is hilarious, both because people do it and that it is actually necessary
186 2012-04-21 19:26:10 <gmaxwell> It works pretty well and it's a lot more secure than trying to sit there and sign things on your laptop.
187 2012-04-21 19:26:34 <gmaxwell> I have my fingerprint on my business cards, which makes things easier.
188 2012-04-21 19:27:27 <graingert> obviously. Moving ones key around is a bit silly
189 2012-04-21 19:27:46 <gmaxwell> (e.g. I don't have to plan to go to a keysigning.. I can just show up and I always have my fingerprint on paper)
190 2012-04-21 19:28:11 <graingert> yes I could do with business cards
191 2012-04-21 19:28:12 <gmaxwell> Some keysigning parties don't use the paper they have people stand up and read off their fingerprint.. but thats way less efficient and I don't think its that common anymore.
192 2012-04-21 19:28:13 <graingert> I'm just too lazy
193 2012-04-21 19:28:28 <graingert> I hope they've stopped using keyID
194 2012-04-21 19:28:55 <gmaxwell> oh you mean using the 32bit IDs? I've never seen people only validate those at keysignings.
195 2012-04-21 19:29:27 <graingert> there was a nice paper on spoofing those
196 2012-04-21 19:29:44 <gmaxwell> yes, by the person on the hop nearest to me on the rightmost path in that graph
197 2012-04-21 19:30:03 <gmaxwell> (Asheesh Laroia)
198 2012-04-21 19:30:03 <graingert> asheesh?
199 2012-04-21 19:30:23 <graingert> http://www.asheesh.org/note/debian/short-key-ids-are-bad-news.html
200 2012-04-21 19:30:37 <graingert> although everyone trusts debian on their security
201 2012-04-21 19:31:01 <graingert> TL;DR: This now gives two results: gpg --recv-key 70096AD1
202 2012-04-21 19:31:12 <graingert> this should really be a concept in papers
203 2012-04-21 19:31:18 <graingert> abstracts are way too long these days
204 2012-04-21 19:32:43 <gmaxwell> well, the refresh is actually a vulnerabilitiy.
205 2012-04-21 19:38:33 <graingert> http://www.l3s.de/~olmedilla/events/MTW06_papers/presentation04/presentation04.html
206 2012-04-21 19:40:10 <gmaxwell> meh. if that interests you join #bitcoin-wot  ... but I'm mostly not enthused by wot systems. They have some usefulness but it's limited.
207 2012-04-21 19:42:13 <graingert> meh bitcoin wot
208 2012-04-21 19:42:22 <graingert> pgp is already big enough
209 2012-04-21 19:42:41 <graingert> it might be handy to support pgp->bitcoin and bitcoin->pgp WOT links
210 2012-04-21 19:42:52 <gmaxwell> the pgp wot is mostly useful to the extent that it's useful because the people using it follow some norms about what keys they'll sign.
211 2012-04-21 19:42:57 <graingert> and the best way to do that is probably through RDF and some sort of reasoner
212 2012-04-21 19:43:16 <graingert> yes the values are in the trust levels
213 2012-04-21 19:43:21 <graingert> and the semantics of those values
214 2012-04-21 19:44:00 <gmaxwell> So for the most part I know that pgp signatures mean these people me in person and showed state issued id of some kind.  When people don't do this, I see it happen (when they sign my key without checking) and I knot not to trust paths via them.
215 2012-04-21 19:44:25 <gmaxwell> But all it really does is confirmes identities.. if you try to use it for things beyond that meh.
216 2012-04-21 19:44:51 <sipa> gmaxwell: actually, there a pgp path between us?
217 2012-04-21 19:47:11 <gmaxwell> I don't think so /me looks
218 2012-04-21 19:48:25 <gmaxwell> oh, there is!
219 2012-04-21 19:48:32 <gmaxwell> http://webware.lysator.liu.se/jc/wotsap/wots/latest/paths/0xB0413BFA-0x1DAAC974.png
220 2012-04-21 19:49:01 <gmaxwell> And two the other way: http://webware.lysator.liu.se/jc/wotsap/wots/latest/paths/0x1DAAC974-0xB0413BFA.png
221 2012-04-21 19:49:21 <Diablo-D3> what, are we playing six degrees of bacon?
222 2012-04-21 19:49:48 <gmaxwell> sipa: seems that Petr Machata either hasn't signed your key, or if he has this server doesn't know about the signature.
223 2012-04-21 19:49:56 <sipa> Diablo-D3: six degrees of zimmerman
224 2012-04-21 19:50:13 <gmaxwell> Best pgp practice is that when you sign someone's key you don't post it yourself: instead you email it to them and let them post it.
225 2012-04-21 19:50:20 <sipa> gmaxwell: i participated in a keysigning at fosdem two years ago
226 2012-04-21 19:50:27 <sipa> yes, i know
227 2012-04-21 19:50:35 <gmaxwell> This proves that they controlled the relevant email address at that point in time.. but a lot of people don't bother pushing out the updated keys.
228 2012-04-21 19:50:37 <Diablo-D3> the BEST pgp practice is to not fucking sign keys
229 2012-04-21 19:50:47 <sipa> there's a script for that even
230 2012-04-21 19:51:06 <sipa> gmaxwell: email it encrypted to their key, even
231 2012-04-21 19:51:17 <gmaxwell> Diablo-D3: Are you feeling left out? want me to come up to maine and sign your key?
232 2012-04-21 19:51:24 <graingert> that's handy
233 2012-04-21 19:51:29 <graingert> it verifies their email also
234 2012-04-21 19:51:40 <Diablo-D3> gmaxwell: heh, no
235 2012-04-21 19:51:48 <Diablo-D3> I used to have a gpg key
236 2012-04-21 19:51:48 <neofutur> gmaxwell: http://webware.lysator.liu.se/jc/wotsap/wots/latest/paths/0xB0413BFA-0x690B4E07.png
237 2012-04-21 19:51:51 <Diablo-D3> Im not sure where it went
238 2012-04-21 19:51:53 <neofutur> ;)
239 2012-04-21 19:53:01 <sipa> gmaxwell: if you ever meet gavin or jeff in real life, get them to sign keys :)
240 2012-04-21 19:53:03 <graingert> I can be sure to trust you not to muchk up then
241 2012-04-21 19:53:24 <graingert> gmaxwell: did you not go to the world bitcoin conference by onlyonetv?
242 2012-04-21 19:53:26 <gmaxwell> sipa: yea, it's on my todo.
243 2012-04-21 19:54:16 <gmaxwell> Hm. someone should make a tool that analizes the WOT and bugs you about connections you would be expected to have but don't.
244 2012-04-21 19:54:35 <Diablo-D3> ha ha you said anal
245 2012-04-21 19:54:52 <graingert> gmaxwell: well if that pgp-wot -> RDF was here it would totally work
246 2012-04-21 19:54:59 <graingert> as long as you remember to FOAF everyone
247 2012-04-21 19:55:57 <gmaxwell> graingert: well, for example, lots of regulars here are WOT disconnected midnightmagic, luke, gavin, for example.
248 2012-04-21 19:56:04 <graingert> yes
249 2012-04-21 19:56:11 <graingert> but how would you know who should be linked?
250 2012-04-21 19:56:42 <gmaxwell> graingert: it wouldn't work for people who are completely disconnected. .. but e.g. most of the people I've signed in the strong set are bidi signed with some people I'm missing.
251 2012-04-21 19:57:59 <gmaxwell> Bradley Kuhn for example, who I know I exchanged with but it must have fallen through at some point.
252 2012-04-21 19:58:37 <gmaxwell> (maybe the picture of me in my key blew up his autoemailing script. :) )
253 2012-04-21 19:59:07 <graingert> SELECT DISTINCT ?name ?key
254 2012-04-21 19:59:09 <graingert> WHERE {
255 2012-04-21 19:59:11 <graingert> ?x rdf:type foaf:Person.
256 2012-04-21 19:59:13 <graingert> ?x wot:hasKey ?key
257 2012-04-21 19:59:15 <graingert> ?x foaf:knows {{your uri}}
258 2012-04-21 19:59:17 <graingert> }
259 2012-04-21 20:00:05 <sipa> gmaxwell: i have a picture in my key as well, and i did get over a hundred signatures via e-mail after fosdem two years ago
260 2012-04-21 20:00:24 <gmaxwell> graingert: ah, I see is the idea that the pgp WOT forms a backbone with these less trusted FOAF links at the edges?
261 2012-04-21 20:01:16 <gmaxwell> sipa: yep. You have a photo.
262 2012-04-21 20:02:04 <sipa> bah, i should change it :)
263 2012-04-21 20:04:07 <gmaxwell> mine is the same picture I have on journal articles, so I guess thats good.
264 2012-04-21 20:04:22 <gmaxwell> though it's old and I've unfortunately gained weight since then.
265 2012-04-21 20:06:31 <sipa> created: 1999-11-27
266 2012-04-21 21:13:21 <graingert> gmaxwell: yes
267 2012-04-21 21:13:55 <graingert> gmaxwell: although all it is is a serialization of PGP data to rdf
268 2012-04-21 21:13:57 <graingert> and it should all work