1 2012-10-03 01:13:20 <jgarzik> The potential for abuse of a GPUMAX-like scenario is clear
 2 2012-10-03 01:13:46 <jgarzik> But it is also reasonable for someone to want to pay extra, to get a non-standard transaction mined on mainnet
 3 2012-10-03 01:14:15 <jgarzik> Luke-Jr is the only outlet for that right now, and that is more altruism/activism than profit motive I think
 4 2012-10-03 01:14:28 <jgarzik> it is disappointing there is not more of a market
 5 2012-10-03 01:21:04 <gmaxwell> you don't need to gpumax to get that.
 6 2012-10-03 01:22:14 <gmaxwell> besides gpumax has been paying premiums of 20 to 50 BTC per block, other >100% ask no questions have paid premiums of 5-8 BTC/block.
 7 2012-10-03 01:22:43 <gmaxwell> pretty sure if you drop your nonstandard txn with a 5 BTC fee you'll find it getting included right quick.
 8 2012-10-03 01:23:34 <jgarzik> gmaxwell: who will take it... and how to find them?
 9 2012-10-03 01:23:47 <jgarzik> it certainly won't get relayed
10 2012-10-03 01:24:05 <gmaxwell> post it on the forum. :P
11 2012-10-03 01:24:10 <jgarzik> heh
12 2012-10-03 01:43:25 <jgarzik> hum
13 2012-10-03 01:43:35 <jgarzik> where to find big endian test machines in the year 2012?
14 2012-10-03 01:44:10 <jgarzik> SF compile farm is long dead, and qemu is lame
15 2012-10-03 01:44:18 <maaku> ps3?
16 2012-10-03 01:44:22 <maaku> ppc mac
17 2012-10-03 01:48:17 <jgarzik> 
ACTION goes to look in the garage for a possible boat anchor^H^H^HPowerPC G5

18 2012-10-03 02:00:55 <asa1024> I once owned a 150 pound shortwave radio transmitter with vacuum tubes
19 2012-10-03 02:01:02 <asa1024> now THAT was a nice boat anchor
20 2012-10-03 02:03:59 <asa1024> I'm surprised someone sold it to an 18 year old non ham operator without checking to make sure what I was going to do with it
21 2012-10-03 02:04:20 <asa1024> I wanted to do some pirate broadcasting but the more I read into it the more I thought best not to do it.
22 2012-10-03 03:14:07 <SomeoneWeird> how long are btc addresses?
23 2012-10-03 03:14:22 <SomeoneWeird> how long *can they be
24 2012-10-03 03:57:55 <weex> SomeoneWeird: haven't seen one over 33 chars but if we're talking database just make it a varchar(64)...visually give it 35-40 chars and you should be ok
25 2012-10-03 04:06:20 <jgarzik> https://en.bitcoin.it/wiki/Address
26 2012-10-03 04:06:27 <jgarzik> huh, I thought the max was 35, not 34
27 2012-10-03 04:18:39 <SomeoneWeird> yeah theres a few conflicting sources jgarzik
28 2012-10-03 05:43:06 <edcba> 2) A few suggestions of how to improve bitcoin:
29 2012-10-03 05:43:13 <edcba> is that a good idea ?
30 2012-10-03 05:45:02 <edcba> different hash function according to merkle hash
31 2012-10-03 05:46:51 <edcba> only problem is to know if that construct is really secure
32 2012-10-03 05:47:57 <edcba> anyway i shouldn't read suggestions from ppl suggesting also to introduce third parties in bitcoin
33 2012-10-03 06:01:28 <sipa> why do you want a function to generate hash functions?
34 2012-10-03 06:01:42 <sipa> i'm sure humans are better at it than random functions
35 2012-10-03 06:01:58 <edcba> to advantage cpu than gpu ?
36 2012-10-03 06:02:19 <sipa> you like botnets?
37 2012-10-03 06:02:25 <edcba> yes
38 2012-10-03 06:02:46 <edcba> you really think botnets don't have gamers in it ? :)
39 2012-10-03 06:03:02 <sipa> sure, they do
40 2012-10-03 06:03:16 <edcba> you do like government controlled currencies ?
41 2012-10-03 06:03:23 <sipa> no
42 2012-10-03 06:03:52 <edcba> see you don't like easy paralleliasble mining :)
43 2012-10-03 06:04:36 <edcba> but anyway i doubt doing some wacky hash generating hash fn is the way to do it
44 2012-10-03 06:04:50 <sipa> yes it's a separate issue
45 2012-10-03 06:05:21 <sipa> there certainly may have been other functions that could have been chosen for the proof-of-work, which may or may not have advantages over the current one
46 2012-10-03 06:05:43 <edcba> sha3 !
47 2012-10-03 06:05:53 <edcba> didn't really look at it yet
48 2012-10-03 06:06:12 <sipa> sha3 certainly doesn't have advantages over sha2 for us now
49 2012-10-03 06:06:38 <edcba> and it's faster in hardware
50 2012-10-03 06:07:03 <sipa> that said, changing that now requires a hard fork, and that will not happen unless a very serious attack against SHA256 is found (or there are reasons why such an attack would be viable in the future)
51 2012-10-03 06:07:32 <sipa> and having a function to generate functions (what does that even mean?) sounds completely wacko to me, sorry :)
52 2012-10-03 06:07:43 <kinlo> I wonder if a hard fork is manageable anyway
53 2012-10-03 06:08:17 <edcba> yes if everyone agrees...
54 2012-10-03 06:09:29 <sipa> we had a hell of a time even agreeing about something trivial like BIP16/BIP17
55 2012-10-03 10:05:04 <edcba> someone know how does timekoin work ?
56 2012-10-03 10:05:21 <edcba> it claims no highend computer to work ?
57 2012-10-03 10:07:41 <edcba> technical pdf has too much ".php" in it
58 2012-10-03 10:07:44 <edcba> must be shit
59 2012-10-03 12:38:22 <gavinandresen> sipa: what's the URL for your stats on client versions?
60 2012-10-03 12:40:49 <Eliel> sipa: function to generate functions ... sounds like perfectly normal coding style in Haskell :P