1 2014-07-18 00:06:23 <maaku> blubberbop: BlueMatt is at 30,000 feet
 2 2014-07-18 00:08:57 <sipa> maaku: i'm impressed by your altitudinal accuracy
 3 2014-07-18 00:09:30 <gwillen> heh, I'm over new jersey, I wonder if he's nearby
 4 2014-07-18 00:09:43 <gmaxwell> laser tag
 5 2014-07-18 04:00:14 <phantomcircuit> <gmaxwell> [00:09:43] laser tag
 6 2014-07-18 04:00:16 <phantomcircuit> zero context amusement
 7 2014-07-18 07:27:33 <wumpus> hehe
 8 2014-07-18 08:45:06 <michagogo> 
ACTION pokes his head in

 9 2014-07-18 08:45:15 <michagogo> Greetings from Slovenia!
10 2014-07-18 08:46:52 <wumpus> greetings back from the Netherlands!
11 2014-07-18 08:50:53 <damethos> wumpus, give us some weed
12 2014-07-18 08:50:57 <damethos> :p
13 2014-07-18 09:00:16 <michagogo> wumpus: yeah, but you live there :P
14 2014-07-18 09:16:43 <wumpus> damethos: hah it's all weed, windmills and wooden shoes here right?
15 2014-07-18 09:17:42 <damethos> hehe exactly
16 2014-07-18 09:17:52 <damethos> and toulips
17 2014-07-18 12:59:00 <jokoon> hello
18 2014-07-18 12:59:29 <jokoon> is it possible to use the existing bitcoins nodes to develop another protocol on top of it ?
19 2014-07-18 12:59:39 <jokoon> like for example host other types of data ?
20 2014-07-18 13:06:31 <jgarzik> jokoon, no
21 2014-07-18 13:06:50 <jokoon> any other protocols that do that ?
22 2014-07-18 13:06:52 <jokoon> like dogecoin ?
23 2014-07-18 16:15:28 <jcorgan> at the time AddBlockToIndex is called in AcceptBlockHeader, are all the block validation tests complete?
24 2014-07-18 16:15:44 <jcorgan> i know this is pre-reorg logic
25 2014-07-18 16:17:52 <sipa> no
26 2014-07-18 16:18:06 <sipa> not the tests that require the utxo set
27 2014-07-18 16:18:34 <sipa> but header connectivity, context-independent transaction tests and proof of work have
28 2014-07-18 16:20:12 <jgarzik> 
ACTION still wants to find a data[base|structure] that permits efficient viewing of the UTXO set, as it existed when $hash was the tip of the best chain.

29 2014-07-18 16:20:18 <jgarzik> for my own edification if nothing else.
30 2014-07-18 16:20:35 <jcorgan> thanks.  i'm trying to locate the point of "greatest trust"; that is, where is that an incoming block has had every test applied and has passed
31 2014-07-18 16:20:36 <jgarzik> ie. given block $hash, return utxo set
32 2014-07-18 16:22:57 <sipa> jgarzik: not hard... use a functional treemap (never overwriting entries, just building new versions that reuse nodes from the previous version if unchanged)
33 2014-07-18 16:23:16 <sipa> jgarzik: but it'll grow faster in memory usage than the blockchain itself
34 2014-07-18 16:23:19 <jgarzik> sipa, or more coarsely, a snapshotting filesystem
35 2014-07-18 16:23:35 <sipa> that's pretty much how snapshotting filesystems work, yes
36 2014-07-18 16:23:42 <sipa> at least zfs does
37 2014-07-18 16:31:35 <jcorgan> orphan blocks get put into a map for later processing, and don't go through all the validation in AcceptBlock until they have a predecessor, correct?
38 2014-07-18 16:32:55 <sipa> until they have _all_ predecessors, yes
39 2014-07-18 16:33:07 <jcorgan> right
40 2014-07-18 16:33:07 <sipa> also, headersfirst removes the concept of orphan blocks :)
41 2014-07-18 16:34:44 <jcorgan> heh, yeah