/srv/irclogs.ubuntu.com/2013/10/16/#ubuntu-us-oh.txt

gilbertpaultag: that's awesome :) how many uninteresting keys did you generate before landing on that one?02:26
paultaggilbert: none, actually ;) (I had to implement huge chunks of OpenPGP though)02:28
paultagcurrently rewriting in C02:28
gilbertyou're writing your own pgp implementation?02:30
paultagNot for other people to use, I'm trying to learn more about OpenPGP / GPG in general02:31
paultagI do that by implementing RFCs02:31
paultagcalming, really02:31
gilbertpaultag: ur always picking the craziest projects ;)02:31
paultaggilbert: ♥    :)02:32
paultagYeah, totally02:32
paultaggilbert: with so many DDs changing their keys for some reason, I figure I should learn about the protocol, how some of this stuff works02:32
gilbertyou should name it pretty good paultag02:32
paultagperhaps PGP smartcards02:32
paultagthose are neat.02:32
paultagbwaahahhahahahahahha02:32
paultagthat's brilliant gilbert02:32
gilberthehe :)02:33
gilbertbtw, i can't stand reading RFCs.  you're a better man than i...02:33
paultagYeah, lota braindamage sometimes02:33
gilbertmuch less implementing them...02:34
paultagbut you do find fun corners of the spec that are prone to … abuse02:34
paultagsuch as that hash ;)02:34
paultagI'm good at homework :)02:34
gilberti see, so that was intentional...02:35
paultagyessir02:35
gilberturgh02:35
paultagperhaps my next one will be 0x0000004202:35
gilbertthat's ripe for abuse02:35
paultagYeah, asheesh abused this before me02:35
paultaghe made a new 4096 key with the same short ID as his old key02:35
paultageons ago02:35
paultagturns out we did it the same way02:36
gilbertoh, i had assumed he did that by generating a bunch of random keys until he got the last 8 to match...02:37
paultagso under the hood it is, but there are some … tricks … to speed it up02:37
paultagwithout actually gen-key'ing02:37
gilberti see02:37
paultagI fear abuse02:38
paultagI think everyone does02:38
paultagsince you can imagine someone doing this with every key in the keyservers02:38
paultagand even signing them in proper order too02:38
gilbertpaultag: u could be a hero by fixing the spec :)02:39
gilbert(if its even possible)02:39
paultagsadly not02:39
paultagthe real fix is to always use the full fingerprint02:39
paultagat least 16 bits of it02:40
gilbertyou mean 16 bytes, right?02:40
paultagerm02:40
paultagyes please02:40
paultag0xFF ftw02:40
paultagerm02:41
paultag0xFFFF02:41
paultagfuck I need sleep02:41
jrgiffordso... we have a twitter now.02:41
jrgiffordalso, paultag - nice key ID. :)02:41
paultagthanks jrgifford :)02:41
paultagI wish I did 1BADD00D02:41
paultagin retrospect02:41
paultagperhaps I'll revoke it for a 0x0000004202:41
gilbert1337D00D would also be nice :)02:42
paultagOh! another good idea02:43
paultag0xBAD1D3A502:43
gilbertyes02:43
paultagAlright, back to the spec02:44
paultagI'm in the middle of breaking GPG02:44
gilbertbad paultag02:44
paultaggpg: mpi larger than indicated length (0 bytes)02:44
paultaggpg: read_block: read error: invalid packet02:44
gilbertlater02:44
paultagwelcome to my hell02:44
gilberti assume the same issue affects subkeys?02:45
paultagyeah02:45
paultagI could issue subkeys with cooked IDs too I think02:45
paultagI'd need to adjust a few things, I think02:46
paultagI'm still getting through the RFC, so I've not even touched subkeys yet02:46
paultagok, for real, outie 500002:47
paultagmuch love02:47
* skellat is still up, watching Tekzilla04:01

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!