1 2010-12-09 00:01:25 <jgarzik> Right now the current chain has a monopoly on the entire mindshare of timestamped proof-of-work (distributed notary) service.  There are many reasons why we should /not/ assume bitcoin is a one-size-fits-all solution.  Yes mainline bitcoin has a strong block chain, but there are plenty of parties who would likely want to invest in a solid DNS chain.
 2 2010-12-09 00:01:26 <jgarzik> The existence of one chain does not imply there should never be another.
 3 2010-12-09 00:01:28 <tcatm> Can we remove OP_DROP from script?
 4 2010-12-09 00:02:16 <ArtForz> we should be able to
 5 2010-12-09 00:02:41 <tcatm> Or does satoshi have any ideas of where it might be useful for transactions?
 6 2010-12-09 00:03:07 <ArtForz> we used to have other useful opcodes, too
 7 2010-12-09 00:03:26 <ArtForz> notice how much of the math opcodes are disabled
 8 2010-12-09 00:03:40 <jgarzik> script was a neat idea, but it's basically too easy to craft algorithmic attacks with arbitrary bytecode
 9 2010-12-09 00:04:06 <MT`AwAy> jgarzik: well, you can't do any loop, it limits the risks
10 2010-12-09 00:04:10 <ArtForz> okay, whats the possible attack scenario of OP_AND ?
11 2010-12-09 00:04:24 <kiba> what a scripting system?
12 2010-12-09 00:04:45 <jgarzik> ArtForz: just wasting a bit of cpu
13 2010-12-09 00:04:53 <jgarzik> not huge without loops
14 2010-12-09 00:05:17 <ArtForz> I can understand disabling string ops, shifts, multiply, ...
15 2010-12-09 00:05:42 <ArtForz> because those make it pretty easy to create large values with a few opcodes
16 2010-12-09 00:05:45 <tcatm> Where in the source can I find isStandard() (does github have a search function?)?
17 2010-12-09 00:05:53 <ArtForz> but basic boolean ops?
18 2010-12-09 00:06:25 <MT`AwAy> I don't think anything can use more cpu than checksig
19 2010-12-09 00:06:29 <gavinandresen> tcatm: one sec, I need to update my git for latest svn...
20 2010-12-09 00:06:36 <jgarzik> tcatm: main.h
21 2010-12-09 00:06:41 <kiba> I wonder
22 2010-12-09 00:06:45 <jgarzik> tcatm: latest SVN
23 2010-12-09 00:06:47 <kiba> should bitcoin have unit tests?
24 2010-12-09 00:06:59 <ArtForz> yeah, but for example op_lshift allowed creating BIG bigints