[12:02:02] --- LOGGING STARTED
[12:48:57] --- alfredh has joined
[12:55:19] --- Willi has joined
[13:02:37] --- touch has joined
[13:03:14] --- cgn has joined
[13:03:55] --- cullenfluffyjennings@gmail.com has joined
[13:04:15] <cullenfluffyjennings@gmail.com> Doing Document status update
[13:04:33] <cullenfluffyjennings@gmail.com> We have an mileston to adopt an application behavior document
[13:04:33] --- dyork has joined
[13:04:49] <cullenfluffyjennings@gmail.com> Should we consdier adopting draft-ford-behave-app
[13:18:28] <cullenfluffyjennings@gmail.com> We are on JDR's draft
[13:18:28] --- narten has joined
[13:18:44] <cullenfluffyjennings@gmail.com> large discution around how large of protection a STUN fingerprint
[13:19:16] <cullenfluffyjennings@gmail.com> question is how big ?
[13:19:52] --- loughney has joined
[13:20:13] <cullenfluffyjennings@gmail.com> TCP has rate 2^32 after other checks
[13:20:46] <cullenfluffyjennings@gmail.com> One note is SHA1 is slow, likely need another hash fucntion
[13:24:17] <cullenfluffyjennings@gmail.com> EKR explained a way to pick the max size of the hash -
[13:24:43] <cullenfluffyjennings@gmail.com> Also sugested use something that was not exactly CRC to reduce odds of collision
[13:25:14] --- davidbryan@gmail.com has joined
[13:25:37] --- sarolaht has joined
[13:30:17] --- ldondeti has joined
[13:33:13] <ldondeti> On Cullen's note about SHA-1 being slow, I just looked this up and on a safenet chip SHA-1 is at 1580 Mbps and MD-5 at 1970 Mbps; surely SHA-1 is slower, but the difference in performance is not really that significant
[13:34:21] <ldondeti> And SHA-256 is faster than SHA-1, rated at 1970 Mbps, same as MD5
[13:35:11] <cullenfluffyjennings@gmail.com> Yah but how fast is CRC32
[13:35:43] <cullenfluffyjennings@gmail.com> I got a small FPGA that can likely to it at 40Gbps
[13:35:54] <ldondeti> :)
[13:36:00] <ldondeti> sure can't beat CRC
[13:36:36] <cullenfluffyjennings@gmail.com> Well I got a parity check that is even faster than that :-)
[13:36:42] <ldondeti> didn't have the full context, but we can't compare a cryptographic hash function and an error code function, can we?
[13:36:56] <ldondeti> right, different classes of functions though
[13:37:08] <cullenfluffyjennings@gmail.com> Yah - the context was we don't need a crypo hash
[13:37:39] <ldondeti> oh, ok
[13:38:15] <ldondeti> then there is no need for even considering SHA-1 or anything like that unless there are CPU cycles available for free ... cool
[13:38:17] <ldondeti> thanks
[13:43:23] <cullenfluffyjennings@gmail.com> REVIEWes - Bruce Lowekamp
[13:43:38] <cullenfluffyjennings@gmail.com> Philip ,
[13:45:09] <cullenfluffyjennings@gmail.com> Sorry - just for the jabber notes later - who agreed to review the STUN bis document?
[14:04:03] --- bhoeneis has joined
[14:07:45] --- jft has joined
[14:09:49] --- alfredh has left
[14:22:33] --- jft has left
[14:23:03] <touch> is the TCP document up yet?
[14:24:03] <cgn> No, still turn
[14:24:05] --- davidbryan@gmail.com has left: Logged out
[14:25:14] <touch> I'm in BTNS, I'd like to pop over when TCP comes up - please post (thanks, in advance)
[14:25:43] <cgn> Ok, I will send a note.
[14:39:20] <cgn> draft-ietf-behave-tcp-01 is coming up NOW!
[14:39:27] <touch> thanks - coming now.
[14:39:37] --- touch has left
[14:41:37] --- touch has joined
[15:02:27] --- cgn has left: Logged out
[15:02:41] --- ldondeti has left: Disconnected.
[15:02:48] --- touch has left
[15:06:24] --- loughney has left
[15:08:36] --- sarolaht has left
[15:22:46] --- dyork has left
[15:29:54] --- Willi has left
[15:34:15] --- cullenfluffyjennings@gmail.com has left
[15:44:37] --- dumdidum has joined
[15:52:49] --- bhoeneis has left
[16:00:14] --- narten has left
[16:15:18] --- dumdidum has left: Replaced by new connection