1 2011-07-26 00:00:10 <SerajewelKS> 10000000.000001 appears to be the smallest amount that the json-rpc interface can convey without corruption
2 2011-07-26 00:00:29 <SerajewelKS> (well, not smallest, but smallest portion with a 10000000 component)
3 2011-07-26 00:01:04 <jgarzik> SerajewelKS: well if I had my druthers I would just present the int64 value, with zero decimals
4 2011-07-26 00:01:16 <SerajewelKS> jgarzik: that's not entirely safe either
5 2011-07-26 00:01:58 <jgarzik> SerajewelKS: it's the natively stored value, so nothing is safer
6 2011-07-26 00:02:01 TheZimm has joined
7 2011-07-26 00:02:10 <SerajewelKS> jgarzik: but integers don't exist in json
8 2011-07-26 00:02:38 <Marf> json is crap?
9 2011-07-26 00:02:46 <jgarzik> SerajewelKS: decoder issue, that all existing implementations can handle
10 2011-07-26 00:03:03 <jgarzik> SerajewelKS: a smart decoder will give you an int64
11 2011-07-26 00:03:13 <Marf> json is no crap?
12 2011-07-26 00:03:16 <SerajewelKS> and a stupid one will give you a float (single)
13 2011-07-26 00:03:43 <SerajewelKS> if you get a string, you can at least convert that however you want instead of relying on the behavior of code that may not be under your control
14 2011-07-26 00:05:03 <SerajewelKS> ((float)1000000000000001d) --> 1E+15
15 2011-07-26 00:05:22 <jrmithdobbs> and spec says it should be treated as float
16 2011-07-26 00:05:29 <jrmithdobbs> so the "stupid" decoders are the ones that don't
17 2011-07-26 00:05:52 <SerajewelKS> well in some languages, "float" is double-precision
18 2011-07-26 00:05:56 <SerajewelKS> it's a bit ambiguous
19 2011-07-26 00:06:03 sacredchao has quit (Ping timeout: 250 seconds)
20 2011-07-26 00:06:24 <SerajewelKS> but the point is that a json "number" is open to some interpretation with respect to target type by the decoder, while a string is not
21 2011-07-26 00:06:56 <SerajewelKS> therefore, using strings to convey 64-bit integers would be smarter than using the json number type
22 2011-07-26 00:07:12 dvide has quit ()
23 2011-07-26 00:07:32 sacredchao has joined
24 2011-07-26 00:07:48 soap has quit (Ping timeout: 255 seconds)
25 2011-07-26 00:07:51 <jrmithdobbs> i agree but dead horse
26 2011-07-26 00:07:52 theorb has joined
27 2011-07-26 00:07:55 <SerajewelKS> regardless of the "smart" decoders that will heuristically determine the best numeric type to use, the dumb ones will not
28 2011-07-26 00:08:27 <jrmithdobbs> and the dumb ones are still adhering to the spec, so calling them "dumb" isn't really correct
29 2011-07-26 00:08:48 theorbtwo has quit (Ping timeout: 276 seconds)
30 2011-07-26 00:08:58 theorb is now known as theorbtwo
31 2011-07-26 00:09:15 <SerajewelKS> jgarzik: i would at least suggest the following compromise: (1) wherever a BTC amount is sent in the json data, send it as a string too: {'amount': 100000000, 'amount_str': '100000000'}. (2) wherever a number is accepted in input, also accept a string.
32 2011-07-26 00:09:42 <SerajewelKS> if your json implementation does tricksy stuff, you can use 'amount'. if it does spec stuff, you can use the string.
33 2011-07-26 00:10:17 <x6763> prof7bit: there's a few bits of information about the info in the wallet here: https://github.com/gavinandresen/bitcointools/blob/master/NOTES.txt
34 2011-07-26 00:10:58 <SerajewelKS> accepting strings is as easy as tweaking AmountFromValue to check for strings
35 2011-07-26 00:14:18 <SerajewelKS> jgarzik: am i making some sense here, or is the numeric/non-string satoshi representation basically set in stone at this point?
36 2011-07-26 00:14:53 <prof7bit> x6763: thanks. found it in the zip file
37 2011-07-26 00:16:01 <jgarzik> SerajewelKS: There is a large cost in breaking everybody's bitcoin website. Any change to the existing, working API must have some clear value.
38 2011-07-26 00:16:17 <SerajewelKS> jgarzik: the clear value is that the current output can be WRONG
39 2011-07-26 00:16:36 <SerajewelKS> when you're dealing with financial data, wrong values are a good reason to change anything
40 2011-07-26 00:16:54 <prof7bit> i still believe for such an important file as the wallet a simpler file format should have been choosen.
41 2011-07-26 00:18:04 <SerajewelKS> jgarzik: if the output was not wrong, i wouldn't be raising this issue. but i'm working on a wallet provider, and there is a demonstrable vulnerability related to rounding in bitcoind.
42 2011-07-26 00:19:38 <prof7bit> not this crazy bdb of which 5474 different versions exist and with an an api that is overly complicated
43 2011-07-26 00:20:21 nhodges has quit (Ping timeout: 250 seconds)
44 2011-07-26 00:20:48 thefinn93 has quit (Remote host closed the connection)
45 2011-07-26 00:21:08 <b4epoche_> and the entire thing should have been written in pascal?
46 2011-07-26 00:21:13 thefinn93 has joined
47 2011-07-26 00:21:17 <jgarzik> SerajewelKS: ALL CAPS does not balance the cost of breaking everybody's website with an API change. In practice, in the field, this issue is very small and not causing any major impact.
48 2011-07-26 00:21:17 <prof7bit> no.
49 2011-07-26 00:21:17 thefinn93 has quit (Read error: Connection reset by peer)
50 2011-07-26 00:21:43 <jgarzik> SerajewelKS: any such change has the potential to create far more problems
51 2011-07-26 00:21:46 <SerajewelKS> jgarzik: i am emphasizing the fact that bitcoind can lie about the value of transactions
52 2011-07-26 00:22:09 copumpkin has quit (Quit: Computer has gone to sleep.)
53 2011-07-26 00:22:21 <prof7bit> or actually, yes.
54 2011-07-26 00:22:34 <prof7bit> would have solved a lot of problems
55 2011-07-26 00:22:34 <jgarzik> SerajewelKS: your point has long been acknowledged
56 2011-07-26 00:22:41 <jgarzik> SerajewelKS: what is more relevant is cost of change
57 2011-07-26 00:23:16 <lfm> SerajewelKS: any ui can lie about what it is doing, that is the nature of trojans
58 2011-07-26 00:23:19 <SerajewelKS> jgarzik: what you are proposing (int64 representation of satoshis) is considerably costlier in terms of existing software maintenance than representing the fractional value as a string
59 2011-07-26 00:23:44 <jgarzik> SerajewelKS: "if I had my druthers" ... implying .. obviously I don't
60 2011-07-26 00:23:47 <SerajewelKS> lfm: if you don't understand what we are talking about then please don't attempt to "contribute"
61 2011-07-26 00:24:10 <jgarzik> SerajewelKS: please parse the entire statement, rather than cherry picking
62 2011-07-26 00:24:15 <lfm> I am not attempting to contribute, I am attempting to distract.
63 2011-07-26 00:24:19 <prof7bit> what is costly about an int64?
64 2011-07-26 00:24:33 <SerajewelKS> prof7bit: the cost would be converting existing software that isn't expecting integers
65 2011-07-26 00:24:34 <jgarzik> prof7bit: Wal-Mart prices are insane
66 2011-07-26 00:24:48 <jgarzik> prof7bit: you can get an int32 pretty cheap, though
67 2011-07-26 00:25:09 zapnap has quit (Ping timeout: 252 seconds)
68 2011-07-26 00:25:17 <SerajewelKS> jgarzik: i was under the impression that you did have your druthers :) anyway, do you think that adding a string representation alongside the number representation would be sufficiently unintrusive?
69 2011-07-26 00:25:28 <lfm> get a few billion of em for 450
70 2011-07-26 00:25:36 <lfm> $50
71 2011-07-26 00:26:16 <lfm> none of us here designed bitcoin.
72 2011-07-26 00:26:33 eian has quit (Quit: Leaving)
73 2011-07-26 00:26:35 <jgarzik> SerajewelKS: if (a big if) there is an API change in that area, it might be reasonable. I wouldn't object at any rate.
74 2011-07-26 00:26:39 <prof7bit> when i look into the protocol specification there it is stored as a 64 bit integer. so i would say this is the natural representation of bitcoins. they all come in int64 when they come into existence.
75 2011-07-26 00:26:46 <SerajewelKS> jgarzik: alright, i might code up a patch then
76 2011-07-26 00:27:03 <SerajewelKS> arguments with patches speak louder than arguments alone :)
77 2011-07-26 00:27:38 <lfm> accountants much prefer integers to floats
78 2011-07-26 00:27:53 <SerajewelKS> prof7bit: indeed. but the argument i'm making is that json doesn't define a 64-bit integer type, it defines a 32-bit floating point type, depending on how you read the spec.
79 2011-07-26 00:28:02 <jrmithdobbs> jgarzik: actually he proposed a solution that would be backwards compat, add a _string or similar version of every number output and input (could make it an option in the call)
80 2011-07-26 00:28:12 <SerajewelKS> jrmithdobbs: that's exactly what i'm doing
81 2011-07-26 00:28:24 <prof7bit> then encode it in base10
82 2011-07-26 00:28:31 <SerajewelKS> jrmithdobbs: except that for input, it would accept both numbers AND strings
83 2011-07-26 00:28:41 <lfm> 32 bit float is useless. maybe a 64 bit float would work but it would actually be a collosal pain
84 2011-07-26 00:28:49 <jrmithdobbs> SerajewelKS: right
85 2011-07-26 00:28:50 <prof7bit> base10 is even human readable.
86 2011-07-26 00:28:58 <SerajewelKS> lfm: correct, but again, json doesn't have integer types
87 2011-07-26 00:29:27 <forrestv> SerajewelKS, why not implement your idea? (alternate amount_str field) but except with int? eg. bitcoind produces both 'amount' and 'amount_int' and accepts either
88 2011-07-26 00:29:27 sacredchao has quit (Ping timeout: 250 seconds)
89 2011-07-26 00:29:32 <lfm> char string representations would work ok also but the pedants would complain even more about wasting bits.
90 2011-07-26 00:29:48 <SerajewelKS> forrestv: because they are not interchangeable
91 2011-07-26 00:30:00 <prof7bit> json is a text format. whats the problem with writing a number to a text file and parse it again later?
92 2011-07-26 00:30:08 <SerajewelKS> forrestv: there are two things wrong with that: 'amount_int' would still have to be represented as a string.
93 2011-07-26 00:30:20 <forrestv> SerajewelKS, accept them as different fields - a client would either provide amount or amount_int in a request
94 2011-07-26 00:30:24 <SerajewelKS> forrestv: because otherwise, a json parser is free to read it as a 32-bit float.
95 2011-07-26 00:30:34 <SerajewelKS> forrestv: arguments to json-rpc are positional, not named
96 2011-07-26 00:30:46 <SerajewelKS> ("ordinal" or whatever)
97 2011-07-26 00:30:49 <forrestv> ah, good point
98 2011-07-26 00:31:02 ujjain has joined
99 2011-07-26 00:31:45 <SerajewelKS> so if it accepted "satoshi" representation as string and "1 BTC" representation as float, then you would have to pay real careful attention to your variable types, because 10 and '10' would be different values (a difference in magnitude of 10^8)
100 2011-07-26 00:31:50 sacredchao has joined
101 2011-07-26 00:32:02 <SerajewelKS> this is my argument for sticking with 1-BTC units
102 2011-07-26 00:32:06 <lfm> SerajewelKS: converting char strings to integers works fine, why bother to change it. Just cuz a few of the json libs try to force some floats or doubles on you has really not much to do with the protocol.
103 2011-07-26 00:32:14 <SerajewelKS> satoshi would arguably be cleaner, but 1-BTC is less confusing
104 2011-07-26 00:32:17 <riush> make it a one-time option - compile time or bitcoin.conf
105 2011-07-26 00:32:29 <SerajewelKS> lfm: go read the json spec
106 2011-07-26 00:32:49 <prof7bit> then encode the integer as base64 or something. saves even some bandwith.
107 2011-07-26 00:33:14 <lfm> take em as char strings
108 2011-07-26 00:33:29 <SerajewelKS> prof7bit: part of the beauty of json is that it's human-readable, which is a great debugging aid
109 2011-07-26 00:33:50 <prof7bit> then use a base10 string
110 2011-07-26 00:33:52 <forrestv> SerajewelKS, what json parsing library are you using?
111 2011-07-26 00:33:56 <SerajewelKS> prof7bit: representing bitcoin amounts as 'xxxxxxxx.xxxxxxxx' in JSON solves all of the problems, without introducing other complexity
112 2011-07-26 00:33:59 viggi has quit (Read error: Connection reset by peer)
113 2011-07-26 00:34:10 <lfm> thats what it is really now, is base 10 char strings.
114 2011-07-26 00:34:14 <SerajewelKS> forrestv: my argument is not specific to any particular library
115 2011-07-26 00:35:17 <prof7bit> or better yet: file a bug against the json parser if it messes up the "big" integer
116 2011-07-26 00:35:25 <lfm> SerajewelKS: the point is really you can fix the problem with the json spec easier than you can change bitcoin these days.
117 2011-07-26 00:35:28 <prof7bit> thsi would solve the problem
118 2011-07-26 00:35:30 <SerajewelKS> prof7bit: i would, if it were a bug
119 2011-07-26 00:35:59 <lfm> fork json then
120 2011-07-26 00:36:21 <SerajewelKS> lfm: are you trolling or what?
121 2011-07-26 00:36:26 <lfm> yup
122 2011-07-26 00:36:28 <jrmithdobbs> he's halfway serious
123 2011-07-26 00:36:29 <jrmithdobbs> tho
124 2011-07-26 00:37:02 <forrestv> SerajewelKS, i don't see any requirement for numbers to be decoded into floats in the spec
125 2011-07-26 00:37:03 <jrmithdobbs> prof7bit: no the parser is operating to spec, as has been said 20 bajillion times this is a flaw in bitcoin's json-rpc
126 2011-07-26 00:37:13 <SerajewelKS> i'm not arguing that the json spec could be improved. i'm looking at this pragmatically: what's the simplest change that would correct the issues without breaking compatibility or readability?
127 2011-07-26 00:37:14 <lfm> just allow json to pass you numeric arg as a string
128 2011-07-26 00:37:27 <lfm> decimal encoded string
129 2011-07-26 00:37:45 nhodges has joined
130 2011-07-26 00:38:55 viggi has joined
131 2011-07-26 00:39:25 sacarlson has quit (Ping timeout: 250 seconds)
132 2011-07-26 00:39:55 <lfm> the protocol sends it as a string anyway across the ports
133 2011-07-26 00:39:59 <SerajewelKS> forrestv: hmm, i thought it was in there. let me look around.
134 2011-07-26 00:40:00 <prof7bit> the pragmatic programmer will encode the number as a string when there is no numeric data type that can hold these values.
135 2011-07-26 00:40:08 osmosis has joined
136 2011-07-26 00:40:19 eastender has joined
137 2011-07-26 00:41:37 <lfm> well int64 holds em
138 2011-07-26 00:41:40 TuxBlackEdo has quit (Ping timeout: 252 seconds)
139 2011-07-26 00:42:25 <lfm> and in fact doubles hold em too, just not exact and you have to be very carefull what you do with them
140 2011-07-26 00:43:58 <lfm> you only really need 51 bits
141 2011-07-26 00:44:48 Marf has quit (Ping timeout: 258 seconds)
142 2011-07-26 00:44:59 moa7 has joined
143 2011-07-26 00:50:21 <lfm> I suspect that is on reason Satoshi designed btc to never be more than 21 million with 8 digit fraction, so it WOULD fit in a double
144 2011-07-26 00:50:23 copumpkin has joined
145 2011-07-26 00:52:08 FractalUniverse has quit (Quit: time for ZZZZZzzzzzzzzzzzzzz)
146 2011-07-26 00:52:32 <SerajewelKS> if satoshi was designing BTC amounts to be used with floating-point types, he is an idiot
147 2011-07-26 00:52:44 <SerajewelKS> you don't use floating-point types for money. period.
148 2011-07-26 00:53:18 <SerajewelKS> or you shouldn't anyway. you might be able to get away with it, but it's still kind of a bad idea...
149 2011-07-26 00:53:26 <lfm> yes, that is a good rule. It is very hard to see when it might be possible to break it.
150 2011-07-26 00:53:28 <prof7bit> you can also round them back to integers
151 2011-07-26 00:53:34 karnac has quit (Quit: karnac)
152 2011-07-26 00:54:31 sacarlson has joined
153 2011-07-26 00:54:31 <SerajewelKS> jrmithdobbs: my string-output patch is coming along, the bulk of the work is done. handling input will be less fun.
154 2011-07-26 00:54:39 karnac has joined
155 2011-07-26 00:56:52 <SerajewelKS> lfm: interestingly, neither C# nor python seem to be able to encode 10000000.00000001 as a double
156 2011-07-26 00:57:30 <lfm> void putval unit64 val, char *buf, int buflen) {
157 2011-07-26 00:57:31 <lfm> snprintf(buf, buflen, "%ull.%08ull", val / 100000000, val % 100000000);
158 2011-07-26 00:57:31 <lfm> }
159 2011-07-26 00:57:52 <SerajewelKS> lfm: that's roughly what i've already done, only with C++ strings
160 2011-07-26 00:58:32 <lfm> missed a (
161 2011-07-26 00:58:43 <SerajewelKS> hmm, there are some cases where the alternate string representation won't work -- getreceivedbyaddress for example does not return a dictionary, so it will have to return either a number or a string
162 2011-07-26 01:00:08 <upb> if the devs are worried about breaking existing apps, isnt there a way to version the interface?
163 2011-07-26 01:00:26 <upb> maybe just do function2 etc if there isnt :)
164 2011-07-26 01:00:38 karnac has quit (Quit: karnac)
165 2011-07-26 01:01:19 Marf has joined
166 2011-07-26 01:01:24 karnac has joined
167 2011-07-26 01:01:46 <lfm> oh the format should be %llu.%08llu instead of what I wrote
168 2011-07-26 01:02:07 zeph97 has quit (Quit: ZNC - http://znc.sourceforge.net)
169 2011-07-26 01:02:11 <SerajewelKS> upb: that's one option, yes. unfortunately, json-rpc is rather limited so conveying interface version information might not be straightforward.
170 2011-07-26 01:03:40 agricocb has quit (Quit: Leaving.)
171 2011-07-26 01:03:43 <upb> oh
172 2011-07-26 01:03:49 <prof7bit> var
173 2011-07-26 01:03:49 <prof7bit> x : Extended;
174 2011-07-26 01:03:49 <prof7bit> begin
175 2011-07-26 01:03:49 <prof7bit> X := StrToFloat('10000000.00000001');
176 2011-07-26 01:03:49 <prof7bit> writeln(x);
177 2011-07-26 01:03:50 <prof7bit> end.
178 2011-07-26 01:03:52 <prof7bit> 1.0000000000000010E+0007
179 2011-07-26 01:03:54 <SerajewelKS> one option would be to unshift {'_btc_rpc_version':1} into the argument list, and inside bitcoind check if the first argument is an object with _version set; if so, shift the argument and set the version number, etc
180 2011-07-26 01:04:15 <SerajewelKS> check for _btc_rpc_version rather
181 2011-07-26 01:04:28 fakap has joined
182 2011-07-26 01:04:58 karnac has quit (Client Quit)
183 2011-07-26 01:05:06 <upb> nah that is a really bad way
184 2011-07-26 01:05:37 <SerajewelKS> upb: it's not pretty, no. it's about the only option without extending the json-rpc spec.
185 2011-07-26 01:05:42 <upb> it would probably break unserializing the arguments in static typed langs
186 2011-07-26 01:05:46 karnac has joined
187 2011-07-26 01:06:07 <SerajewelKS> upb: uhm, bitcoind rpc already accepts structures. look at sendmany.
188 2011-07-26 01:06:25 <upb> oh i havent actually looked at how the interface looks :)
189 2011-07-26 01:06:34 <SerajewelKS> upb: static typing is just as capable of complex object structures as dynamic typing
190 2011-07-26 01:06:34 <upb> so every function just takes a struct ?
191 2011-07-26 01:06:45 <upb> sure
192 2011-07-26 01:06:47 <SerajewelKS> upb: not all functions do. just sendmany.
193 2011-07-26 01:06:57 <lfm> but like you said getreceivedbyaddress isnt a struct
194 2011-07-26 01:06:58 <upb> but if the arguments are determined by method name
195 2011-07-26 01:07:00 <upb> it would break
196 2011-07-26 01:07:17 <SerajewelKS> upb: sendmany is ['sending_account', {'addr1': 1.00, 'addr2': 5.00}]
197 2011-07-26 01:08:09 <upb> so thats 2 arguments or 1 ?
198 2011-07-26 01:08:22 <SerajewelKS> upb: two, the first is a string and the second is a dictionary
199 2011-07-26 01:08:42 <SerajewelKS> oh and to be pedantic, the quotes should be " instead of '
200 2011-07-26 01:08:43 <upb> okay i wasnt sure how the wrapping is done in json-rpc
201 2011-07-26 01:08:55 <SerajewelKS> JSON can nest structures arbitrarily deep
202 2011-07-26 01:08:59 <upb> yes
203 2011-07-26 01:09:06 MrTiggr has joined
204 2011-07-26 01:09:33 <upb> now imagine that you have a json-rpc library that tries to unserialize the arguments to types based on the method name
205 2011-07-26 01:09:36 <upb> it would break
206 2011-07-26 01:09:37 <SerajewelKS> i really like the idea of versioning. i'm trying to figure out how it would be implemented in a clean way though.
207 2011-07-26 01:10:02 <SerajewelKS> upb: that depends on the library and how it treats unexpected dictionary members
208 2011-07-26 01:10:11 <upb> yep
209 2011-07-26 01:10:14 <SerajewelKS> upb: most libraries i have seen ignore keys they don't recognize
210 2011-07-26 01:10:36 <upb> oh okay
211 2011-07-26 01:10:45 <SerajewelKS> upb: but i really don't think that the bitcoin devs consider adding keys a problem. the "proxy" key was added to the getinfo response recently.
212 2011-07-26 01:12:12 <SerajewelKS> upb: you know, there's a really fucking simple versioning mechanism that i've completely ignored
213 2011-07-26 01:12:23 <upb> anyway when you stash the interface version into arguments youre forcing the user to do their own dispatching
214 2011-07-26 01:12:27 <SerajewelKS> upb: the URL. the existing API is "/". the second version could be "/2".
215 2011-07-26 01:12:45 <SerajewelKS> that's so damn clean i need to go change my underwear
216 2011-07-26 01:12:58 <upb> yes it is, i've done that with soap
217 2011-07-26 01:13:08 <SerajewelKS> alrightly, well let me just hack that up
218 2011-07-26 01:13:09 <upb> so you really have a separate service for each version
219 2011-07-26 01:13:21 <lfm> underwear and soap, ya should work fine.
220 2011-07-26 01:13:47 SecretSJ has quit (Quit: If you think nobody cares, try missing a few payments)
221 2011-07-26 01:15:03 <upb> ha ha ha
222 2011-07-26 01:15:05 <upb> not :P
223 2011-07-26 01:15:14 eian has joined
224 2011-07-26 01:16:56 soap has joined
225 2011-07-26 01:17:15 <SerajewelKS> uh oh, we highlighted soap
226 2011-07-26 01:19:20 <SerajewelKS> oh good lord, bitcoind uses the same HTTP response parsing code for both client and server mode. bad bitcoind.
227 2011-07-26 01:20:00 Zagitta has quit (Ping timeout: 260 seconds)
228 2011-07-26 01:20:42 rynx has quit (Ping timeout: 255 seconds)
229 2011-07-26 01:21:03 Marf has quit (Quit: Nettalk6 - www.ntalk.de)
230 2011-07-26 01:23:46 <eian> Are there any other C++ implementations of bitcoind?
231 2011-07-26 01:24:00 <eian> (That work?)
232 2011-07-26 01:24:23 <BlueMatt> not yet
233 2011-07-26 01:26:47 <eian> Crypto++ has its own Integer class - I was wondering if Satoshi's CBigNum was somehow optimized?
234 2011-07-26 01:27:43 <BlueMatt> if you are asking about optimization, that is so far down the list its just not even worth thinking about
235 2011-07-26 01:27:48 <eian> haha
236 2011-07-26 01:28:01 <eian> Is there a reason he rolled his own BigInt class?
237 2011-07-26 01:28:09 <eian> I was wondering about the design decision
238 2011-07-26 01:28:25 * BlueMatt wouldnt know, Im assuming there was something he needed
239 2011-07-26 01:28:35 <upb> anything possible in bitcoin is home rolled :)
240 2011-07-26 01:28:55 <eian> upb: this makes me sad
241 2011-07-26 01:29:27 <eian> upb: reading some of the code ...it makes me want to club a baby seal
242 2011-07-26 01:29:30 <upb> the serialization is actually pretty nice
243 2011-07-26 01:29:36 <upb> but i wonder why he did that
244 2011-07-26 01:31:08 <BlueMatt> quite a bit of the code is actually pretty nice, its just not modularized well
245 2011-07-26 01:31:32 osmosis has quit (Ping timeout: 252 seconds)
246 2011-07-26 01:31:36 sacredchao has quit (Quit: KVIrc KVIrc Equilibrium 4.1.1, revision: 5829, sources date: 20110403, built on: 2011-05-07 15:04:44 UTC http://www.kvirc.net/)
247 2011-07-26 01:31:49 TheZimm has quit (Quit: Computer has gone to sleep.)
248 2011-07-26 01:32:43 <eian> I've never really done multiplatform work (outside of Java). Maybe that's why I'm having trouble reading this stuff
249 2011-07-26 01:32:56 <eian> So many macros :(
250 2011-07-26 01:33:22 <soap> why "uh oh" over highlighting me?
251 2011-07-26 01:33:40 sacredchao has joined
252 2011-07-26 01:33:50 <SerajewelKS> soap: we brought up SOAP and about a minute later you joined :)
253 2011-07-26 01:34:09 <eian> This is me nitpicking, but I feel this actually makes readability more difficult(from util.h): #define loop for (;;)
254 2011-07-26 01:34:23 osmosis has joined
255 2011-07-26 01:34:30 <eian> why not just use while(true) or something
256 2011-07-26 01:34:50 rynx has joined
257 2011-07-26 01:34:52 <BlueMatt> thats one of the very smallest code reability things right now
258 2011-07-26 01:34:53 <eian> or even for(;;) is easier to comprehend
259 2011-07-26 01:35:02 <eian> yeah, I know. I'm bitching
260 2011-07-26 01:35:06 <eian> haha sorry
261 2011-07-26 01:35:12 <soap> ahh, yes, this machine has had 99+% uptime the last two years, but it's suffering something.
262 2011-07-26 01:38:19 <SerajewelKS> jrmithdobbs: i thing i just implemented versioning. *crosses fingers*
263 2011-07-26 01:38:25 <SerajewelKS> think*
264 2011-07-26 01:40:23 Akiron has joined
265 2011-07-26 01:41:57 <BlueMatt> next version of rpc should be...not rpc
266 2011-07-26 01:42:01 <jrmithdobbs> BlueMatt: do you not get that guy on lists' comments about dnssec?
267 2011-07-26 01:42:09 <SerajewelKS> BlueMatt: hah, no kidding
268 2011-07-26 01:42:31 <BlueMatt> jrmithdobbs: what did he say about dnssec?
269 2011-07-26 01:43:04 <jrmithdobbs> BlueMatt: because if a zone is signed and you have a validating resolver under your control you can be 100% sure that there is no MITM
270 2011-07-26 01:43:05 <BlueMatt> jrmithdobbs: he said nothing about dnssec
271 2011-07-26 01:43:16 <BlueMatt> jrmithdobbs: who has a resolver under their control?
272 2011-07-26 01:43:20 <jrmithdobbs> me
273 2011-07-26 01:43:22 <BlueMatt> maybe 1%
274 2011-07-26 01:43:29 <BlueMatt> no... .01%
275 2011-07-26 01:43:44 <BlueMatt> so...no one, aka its secure for such a tiny minority it doesnt matter
276 2011-07-26 01:43:45 <jrmithdobbs> still your comments show a clear lack of understanding of dnssec
277 2011-07-26 01:43:47 b4epoche has quit (Read error: Operation timed out)
278 2011-07-26 01:43:47 b4epoche_ is now known as b4epoche
279 2011-07-26 01:43:57 b4epoche_ has joined
280 2011-07-26 01:44:13 <BlueMatt> tell me how a mitmer at starbucks could *not* steal all your coins in this situation?
281 2011-07-26 01:44:22 <BlueMatt> I understand dnssec just fine
282 2011-07-26 01:44:22 <jrmithdobbs> because the bitcoin code itself (or any client making use of that feature) could implement the dnssec stuff to validate itself
283 2011-07-26 01:44:28 <BlueMatt> bah
284 2011-07-26 01:44:33 <SerajewelKS> upb: i have basic versioning in place :)
285 2011-07-26 01:44:35 <BlueMatt> wtf would we do that?
286 2011-07-26 01:44:35 <jrmithdobbs> trivially
287 2011-07-26 01:44:41 <jrmithdobbs> to prevent mitm
288 2011-07-26 01:44:41 karnac has quit (Read error: Connection reset by peer)
289 2011-07-26 01:44:45 <jrmithdobbs> obviously
290 2011-07-26 01:44:45 wirehead has joined
291 2011-07-26 01:44:48 <BlueMatt> that is just pointlessly complicated
292 2011-07-26 01:44:57 <jrmithdobbs> if bitcoin ever ends up using dns for anything
293 2011-07-26 01:44:58 <BlueMatt> better solution: dont use dns here
294 2011-07-26 01:45:00 karnac has joined
295 2011-07-26 01:45:14 <jrmithdobbs> uris still use dns
296 2011-07-26 01:45:19 <jrmithdobbs> so it's no different
297 2011-07-26 01:45:22 <upb> SerajewelKS: nice, do you have it on github ?:P
298 2011-07-26 01:45:28 <BlueMatt> plus dnssec fails open...
299 2011-07-26 01:45:32 <SerajewelKS> upb: no, i'm still polishing it
300 2011-07-26 01:45:35 <upb> ah
301 2011-07-26 01:45:36 <jrmithdobbs> fails what?
302 2011-07-26 01:45:46 <BlueMatt> so if you did do it client side, you would have to wait a couple more years before it would be secure
303 2011-07-26 01:45:56 <BlueMatt> if a zone is not dnssec signed, it still passes validation
304 2011-07-26 01:45:59 <BlueMatt> aka fails open
305 2011-07-26 01:46:25 <jrmithdobbs> that's not a "failure" that's by design
306 2011-07-26 01:46:33 <BlueMatt> you could set your implementation to fail close, but even some root servers dont have it yet...
307 2011-07-26 01:46:34 <jrmithdobbs> it's up to zone maintainers whether they wish to sign?
308 2011-07-26 01:46:36 E-sense has quit (Ping timeout: 250 seconds)
309 2011-07-26 01:47:08 <BlueMatt> yea, thats fine, except that you still have to wait for the root servers
310 2011-07-26 01:47:19 <BlueMatt> if it fails open, and root servers dont all have it, then you are fucked
311 2011-07-26 01:47:23 <jrmithdobbs> no you don't
312 2011-07-26 01:47:36 <jrmithdobbs> dlv
313 2011-07-26 01:47:36 <BlueMatt> yes you do
314 2011-07-26 01:47:40 <lfm> jrmithdobbs: but do impostors need to sogn then, just send you an unsigned result
315 2011-07-26 01:47:40 <jrmithdobbs> no. you don't.
316 2011-07-26 01:48:10 <SerajewelKS> upb: if you access the service as /bitcoin.vN then it uses version N
317 2011-07-26 01:48:21 <upb> :)
318 2011-07-26 01:48:26 afed is now known as gasjews
319 2011-07-26 01:48:42 <BlueMatt> if you dont have a root server without dnssec, then a mitm attacker can simply intercept everything and not sign anything and poof
320 2011-07-26 01:48:44 <jrmithdobbs> lfm: if validation fails a well behaving client treats sig check failures as NXDOMAIN
321 2011-07-26 01:49:01 <BlueMatt> if you have known dnssec keys for root servers, then you can verify everything if you force dnssec
322 2011-07-26 01:49:16 <jrmithdobbs> BlueMatt: except that the isc dlv registry lives on a tld that is signed
323 2011-07-26 01:49:40 <BlueMatt> except that no one uses DLV
324 2011-07-26 01:49:50 <jrmithdobbs> um, yes, they do
325 2011-07-26 01:50:02 <jrmithdobbs> dlv has been in use for years long before any tlds were signed
326 2011-07-26 01:50:11 <jrmithdobbs> the isc dlv specifically.
327 2011-07-26 01:50:17 <BlueMatt> its used in theory, not on a broad scale in practice
328 2011-07-26 01:50:30 <jrmithdobbs> it's used by anyone implementing dnssec
329 2011-07-26 01:50:47 <BlueMatt> which is...almost no one
330 2011-07-26 01:51:16 <BlueMatt> in any case, this is just way overcomplicated, this means having an entire dns resolver built into bitcoin
331 2011-07-26 01:51:21 <SerajewelKS> upb: http://pastie.org/2271588
332 2011-07-26 01:51:22 <BlueMatt> in other words...wtf?
333 2011-07-26 01:51:33 <jrmithdobbs> no actually it doesn't
334 2011-07-26 01:51:38 <SerajewelKS> tada, api versioning
335 2011-07-26 01:52:22 bitcoinbulletin has quit (Remote host closed the connection)
336 2011-07-26 01:52:30 <jrmithdobbs> BlueMatt: there are libs that provide the functionality
337 2011-07-26 01:52:33 <upb> wow, the json-rpc client proxy is named Server?
338 2011-07-26 01:52:41 <upb> or the method returning one
339 2011-07-26 01:52:41 <BlueMatt> jrmithdobbs: again, how is that not overengineering
340 2011-07-26 01:52:48 <BlueMatt> jrmithdobbs: yes, there are libs, but there are better solutions
341 2011-07-26 01:53:13 <BlueMatt> just because you can make something work, does not at all mean you should
342 2011-07-26 01:53:32 <jrmithdobbs> BlueMatt: i don't agree with that guy's use of TXT records but auto-discovered local-net SRV records would actually be quite handy.
343 2011-07-26 01:53:48 <BlueMatt> ...
344 2011-07-26 01:53:59 <jrmithdobbs> it would mean i could get rid of my damned bitcoin client wrapper scripts
345 2011-07-26 01:54:02 <jrmithdobbs> for instance
346 2011-07-26 01:54:03 <BlueMatt> again, there are much, much, much better solutions for mapping name->address
347 2011-07-26 01:54:12 <jrmithdobbs> like?
348 2011-07-26 01:54:29 bitcoinbulletin has joined
349 2011-07-26 01:54:31 <BlueMatt> like an https server serving addresses
350 2011-07-26 01:54:33 <jrmithdobbs> there's better solutions than dns for name->address? seriously?
351 2011-07-26 01:54:38 <jrmithdobbs> that still uses dns?
352 2011-07-26 01:54:41 <jrmithdobbs> wtf are you talking about
353 2011-07-26 01:54:42 Diablo-D3 has joined
354 2011-07-26 01:54:49 <BlueMatt> you cant spoof an ssl cert
355 2011-07-26 01:54:59 <jrmithdobbs> so? you can't spoof srv records on my network
356 2011-07-26 01:55:13 rynx has quit (Quit: Leaving)
357 2011-07-26 01:55:13 <BlueMatt> you can give different addresses per request
358 2011-07-26 01:55:28 <jrmithdobbs> you can with dns as well?
359 2011-07-26 01:55:36 <BlueMatt> not really
360 2011-07-26 01:55:39 <jrmithdobbs> yes really
361 2011-07-26 01:55:52 <BlueMatt> you could set ttl 0 but you dont know who is asking, etc
362 2011-07-26 01:56:00 <BlueMatt> you know the name of one of their resolvers
363 2011-07-26 01:56:31 eoss has quit (Remote host closed the connection)
364 2011-07-26 01:56:33 <BlueMatt> I just have yet to see an advantage to dns over https here
365 2011-07-26 01:56:50 <jrmithdobbs> no need for 3rd party trust for the cert
366 2011-07-26 01:56:51 <BlueMatt> no one has given me one and people have been talking about dns-bitcoin for a while now
367 2011-07-26 01:57:06 <upb> but why go with https when you could easily use ftps here ?
368 2011-07-26 01:57:13 <BlueMatt> if ssl certs are broke, there are much, much worse stuff going on here
369 2011-07-26 01:57:14 <upb> or gopher over tls
370 2011-07-26 01:57:28 denisx has quit (Remote host closed the connection)
371 2011-07-26 01:57:38 <jrmithdobbs> BlueMatt: if you want to actually create trust with certs you need a ca
372 2011-07-26 01:57:44 <jrmithdobbs> or you need to use a 3rd party ca
373 2011-07-26 01:57:46 denisx has joined
374 2011-07-26 01:57:47 <BlueMatt> yes, Im aware of that
375 2011-07-26 01:58:02 <BlueMatt> but there is an existing infrastructure for that
376 2011-07-26 01:58:08 <yebyen> has anyone tried zookeeper?
377 2011-07-26 01:58:09 <jrmithdobbs> running your own ca is a pain in the ass and requires more work per-client than dnssec
378 2011-07-26 01:58:10 <yebyen> for anything
379 2011-07-26 01:58:18 <BlueMatt> wtf would you run your own ca here?
380 2011-07-26 01:58:33 <jrmithdobbs> because i'm talking about local network?
381 2011-07-26 01:58:46 <jrmithdobbs> why the fuck would you pay $150+ for a cert from a ca for local usage?
382 2011-07-26 01:58:57 <BlueMatt> if you are doing this entirely local, you have to run your own dnssec root trust too
383 2011-07-26 01:59:13 <jrmithdobbs> no, since i have trusted resolvers locally ...
384 2011-07-26 01:59:15 <BlueMatt> and if you are doing this locally, just write the damn address down
385 2011-07-26 01:59:35 <wasabi1> Running a CA is easy. For me.
386 2011-07-26 01:59:45 <BlueMatt> then you still have to buy a domain and sign all that crap too
387 2011-07-26 01:59:52 E-sense has joined
388 2011-07-26 01:59:53 <BlueMatt> which really is no different from getting a ca signed cert
389 2011-07-26 01:59:54 <wasabi1> Tangent tangent tangent
390 2011-07-26 02:00:00 wolfspraul has joined
391 2011-07-26 02:00:03 <BlueMatt> also, there is that ssl cert that is free
392 2011-07-26 02:00:07 <jrmithdobbs> trusting the anchor on . is fine since if the anchor on . gets fucked and org. gets fucked and someone compromises my authoratative servers, AND someone compromises my resolvers ... well
393 2011-07-26 02:00:15 <jrmithdobbs> i've got MUCH bigger things to worry about and have pissed off the wrong people
394 2011-07-26 02:00:24 <BlueMatt> same with ssl
395 2011-07-26 02:00:50 <upb> wasabi1: you run a commercial ca ?:P
396 2011-07-26 02:01:54 denisx_ has joined
397 2011-07-26 02:02:03 <jrmithdobbs> BlueMatt: tbqh i think you just like to argue about things you don't understand and reinvent the wheel instead of reusing existing infrastructure that works, but w/e
398 2011-07-26 02:02:13 <BlueMatt> lol, ok then
399 2011-07-26 02:02:13 <jrmithdobbs> BlueMatt: next you're going to tell me that using krb for auth is pointless
400 2011-07-26 02:02:28 <jrmithdobbs> since using dns for name resolution is pointless
401 2011-07-26 02:02:45 <BlueMatt> no, Im trying to argue one solution is better than another because one offers features the other doesnt and has no significant draw backs
402 2011-07-26 02:02:47 <wasabi1> upb, yup.
403 2011-07-26 02:02:56 <upb> cool
404 2011-07-26 02:03:14 <wasabi1> Specific problem case. Just thought I'd brag on a tangent about how my life is so awesome.
405 2011-07-26 02:03:28 <jrmithdobbs> BlueMatt: running a name resolution server over https with custom code vs just adding an srv record is simpler? how exactly?
406 2011-07-26 02:03:47 <BlueMatt> https with custom code? no
407 2011-07-26 02:03:53 <BlueMatt> you can just put up a txt file if you want
408 2011-07-26 02:03:59 <upb> which vendors hsms are you using if you can say
409 2011-07-26 02:04:03 <jrmithdobbs> but then it can't change dynamically as you demanded
410 2011-07-26 02:04:04 <BlueMatt> but those who want *can* do more
411 2011-07-26 02:04:12 <BlueMatt> vs dns were you cant really
412 2011-07-26 02:04:20 <jrmithdobbs> whereas it can with dns it can well enough with no extra work
413 2011-07-26 02:04:23 <BlueMatt> or...it doesnt work as well for certain cases
414 2011-07-26 02:04:34 <wasabi1> Are ya'll talking about seeding of IPs for bc? I assume you are.
415 2011-07-26 02:04:43 <BlueMatt> no name->address mapping
416 2011-07-26 02:04:48 <jrmithdobbs> so you throw out 80% of the solution because it doesn't fit a .1% use case?
417 2011-07-26 02:04:50 <wasabi1> In general?
418 2011-07-26 02:04:56 <jrmithdobbs> wasabi1: yes fucking lol
419 2011-07-26 02:05:23 <BlueMatt> no, Im saying throw out one solution in favor of another that has an advantage for probably 25% of people
420 2011-07-26 02:05:32 <wasabi1> Use an existing P2P DNS system. Booyah!
421 2011-07-26 02:05:37 <wasabi1> Probably not.
422 2011-07-26 02:05:38 <BlueMatt> remember the bitcoin community is full of people who think they are being anonymous and want to be
423 2011-07-26 02:06:15 denisx has quit (Ping timeout: 276 seconds)
424 2011-07-26 02:06:16 denisx_ is now known as denisx
425 2011-07-26 02:06:56 * wasabi1 spent a week playing around with PNRP awhile ago.
426 2011-07-26 02:07:06 <SerajewelKS> upb / jgarzik: http://pastie.org/2271650
427 2011-07-26 02:07:17 <SerajewelKS> this is THE* solution
428 2011-07-26 02:07:19 <SerajewelKS> * my
429 2011-07-26 02:09:31 <upb> cool :D
430 2011-07-26 02:10:18 <wasabi1> SerajewelKS: You should make my miner Mono compatible.
431 2011-07-26 02:10:28 <wasabi1> Heh.
432 2011-07-26 02:10:30 <SerajewelKS> wasabi1: it's not? O_o
433 2011-07-26 02:10:34 <SerajewelKS> tsk
434 2011-07-26 02:10:34 <wasabi1> Got me. Haven't tried.
435 2011-07-26 02:10:47 <wasabi1> I'm pretttttty sure the mixed mode assembly ain't gonna works.
436 2011-07-26 02:10:57 <wasabi1> I did an SSE version.
437 2011-07-26 02:11:06 <SerajewelKS> it won't work
438 2011-07-26 02:11:07 <upb> damn youre a mono dev?
439 2011-07-26 02:11:16 <upb> thought i'd seen your nick somewhere
440 2011-07-26 02:11:32 <wasabi1> Yeah I know. I was being facetious.
441 2011-07-26 02:11:37 <SerajewelKS> upb: me? i'm not an official developer OF mono. i've worked with it a lot and contributed patches.
442 2011-07-26 02:11:42 <wasabi1> I have Cloo working though.
443 2011-07-26 02:11:47 <upb> ic:)
444 2011-07-26 02:11:51 <wasabi1> That should probably work. I need to test it.
445 2011-07-26 02:11:54 <wasabi1> I'm using a LOT of 4.0 stuff.
446 2011-07-26 02:12:20 <wasabi1> Actually... am I? Maybe I toned that down a bit in the last rearch.
447 2011-07-26 02:12:25 <wasabi1> It's been two weeks since I looked at it.
448 2011-07-26 02:12:37 <upb> SerajewelKS: any chance you worked on the tls or soap implementations there ?
449 2011-07-26 02:12:48 <upb> or the cryptography part
450 2011-07-26 02:12:58 <SerajewelKS> upb: no, i don't know enough about ssl/crypto type stuff and soap makes me vomit
451 2011-07-26 02:13:07 <jrmithdobbs> as it should
452 2011-07-26 02:13:09 <wasabi1> YOu're not supposed to swallow it.
453 2011-07-26 02:13:18 <SerajewelKS> wasabi1: that's what he said?
454 2011-07-26 02:13:24 <wasabi1> Why would he say that?
455 2011-07-26 02:13:31 <jrmithdobbs> lol
456 2011-07-26 02:13:39 <wasabi1> He's an idiot.
457 2011-07-26 02:13:40 <SerajewelKS> dunno, it just sounds like something he might say
458 2011-07-26 02:14:14 <upb> ok
459 2011-07-26 02:14:20 <wasabi1> I've been lacking on the Linux desktop department for a bit. Doing way too much real work in VS... so haven't tested it.
460 2011-07-26 02:14:24 <SerajewelKS> upb: now i get to parse nnnnnnnn.nnnnnnnn as int64. woot.
461 2011-07-26 02:14:24 <wasabi1> I used Diablo's kernel.
462 2011-07-26 02:14:30 <wasabi1> And it runs. ANd produces valid hashes.
463 2011-07-26 02:14:46 <SerajewelKS> wasabi1: virtualbox.org. go get it.
464 2011-07-26 02:14:52 <wasabi1> True, true.
465 2011-07-26 02:14:53 [7] has quit (Disconnected by services)
466 2011-07-26 02:15:04 TheSeven has joined
467 2011-07-26 02:16:06 <SerajewelKS> upb: i have to be really careful how i parse this :S
468 2011-07-26 02:17:44 BlueMatt has quit (Quit: Ex-Chat)
469 2011-07-26 02:17:59 denisx has quit (Quit: denisx)
470 2011-07-26 02:19:41 pierce1 is now known as pierce
471 2011-07-26 02:20:53 nefario has joined
472 2011-07-26 02:24:09 AStove has quit (Ping timeout: 255 seconds)
473 2011-07-26 02:29:27 thefinn93 has joined
474 2011-07-26 02:44:52 osmosis has quit (Quit: Leaving)
475 2011-07-26 02:45:52 agricocb has joined
476 2011-07-26 02:52:42 <SerajewelKS> upb: i win
477 2011-07-26 02:53:33 <SerajewelKS> upb: http://pastie.org/2271842
478 2011-07-26 02:54:33 <eian> whatever that is, it doesn't look good
479 2011-07-26 02:54:41 <eian> you broke something :(
480 2011-07-26 02:54:51 <SerajewelKS> what did i break?
481 2011-07-26 02:55:03 <jrmithdobbs> that looks awesome
482 2011-07-26 02:55:18 <eian> sorry, I was trying to make a funny
483 2011-07-26 02:55:22 <upb> :D
484 2011-07-26 02:55:22 <SerajewelKS> oh :)
485 2011-07-26 02:55:23 <eian> I'm tired
486 2011-07-26 02:55:28 <upb> good night/morning/midday
487 2011-07-26 02:55:29 <jrmithdobbs> SerajewelKS: so / will always use v0?
488 2011-07-26 02:55:34 <SerajewelKS> jrmithdobbs: yes
489 2011-07-26 02:55:41 <SerajewelKS> jrmithdobbs: as the patch author, i am declaring v0 and v1 equal
490 2011-07-26 02:56:54 <jrmithdobbs> fair
491 2011-07-26 03:01:25 Katapult_ has joined
492 2011-07-26 03:03:56 Katapult has quit (Ping timeout: 252 seconds)
493 2011-07-26 03:04:46 Nicksasa has quit (Read error: Operation timed out)
494 2011-07-26 03:04:49 tcoppi has quit (Read error: Operation timed out)
495 2011-07-26 03:05:26 tcoppi has joined
496 2011-07-26 03:05:42 tcoppi has quit (Remote host closed the connection)
497 2011-07-26 03:06:44 Katapult_ is now known as Katapult
498 2011-07-26 03:07:01 <SerajewelKS> i'm staging a branch against 0.3.24
499 2011-07-26 03:07:17 Nicksasa has joined
500 2011-07-26 03:09:21 thefinn93 has quit (Quit: BAI)
501 2011-07-26 03:09:34 <jgarzik> SerajewelKS: putting a new API at a new URL seems a fair and standard way to do things
502 2011-07-26 03:10:25 tcoppi has joined
503 2011-07-26 03:12:21 Tril has quit (Ping timeout: 252 seconds)
504 2011-07-26 03:13:45 Clipse has quit (Read error: Connection reset by peer)
505 2011-07-26 03:17:11 senseles has quit (Read error: Connection reset by peer)
506 2011-07-26 03:17:22 senseles has joined
507 2011-07-26 03:17:46 zapnap has joined
508 2011-07-26 03:17:52 copumpkin is now known as draino
509 2011-07-26 03:18:22 draino is now known as copumpkin
510 2011-07-26 03:19:45 Guest39969 has joined
511 2011-07-26 03:20:28 TheZimm has joined
512 2011-07-26 03:20:37 Akiron has quit (Disconnected by services)
513 2011-07-26 03:22:12 jimon has quit (Ping timeout: 255 seconds)
514 2011-07-26 03:22:38 mrb_ has quit (Ping timeout: 252 seconds)
515 2011-07-26 03:22:44 mrb_ has joined
516 2011-07-26 03:23:18 <eian> Anyone here know how to exponentiate using crypto++? (I can obviously multiple in a loop, but that defeats the purpose). I'm using CryptoPP::Integer
517 2011-07-26 03:23:27 <eian> multiply*
518 2011-07-26 03:27:31 <SerajewelKS> upb: https://github.com/bitcoin/bitcoin/pull/431
519 2011-07-26 03:27:49 <SerajewelKS> hopefully this pull request will receive some attention, unlike my listsinceblock pull request, which has sat there bit-rotting
520 2011-07-26 03:29:39 <luke-jr> SerajewelKS: that one *should* bitrot
521 2011-07-26 03:29:53 <SerajewelKS> mmhmm
522 2011-07-26 03:30:20 <luke-jr> SerajewelKS: it looks like a rewrite of my RPCv1 branch, but without fixing the actual problem
523 2011-07-26 03:30:49 <SerajewelKS> the problem being that it doesn't use TBC?
524 2011-07-26 03:31:05 <luke-jr> SerajewelKS: the problem being that it is trying to use human units for a low-level API
525 2011-07-26 03:31:27 <SerajewelKS> JSON is low-level?
526 2011-07-26 03:31:29 <luke-jr> yes
527 2011-07-26 03:31:33 <SerajewelKS> go away, troll
528 2011-07-26 03:31:40 <eian> haha
529 2011-07-26 03:31:43 <luke-jr> also, your claimed Python issues don't exist
530 2011-07-26 03:31:44 <lfm> luke-jr you mean it should just use satoshis?
531 2011-07-26 03:31:50 <luke-jr> lfm: yes
532 2011-07-26 03:32:03 eoss has joined
533 2011-07-26 03:32:27 <luke-jr> lfm: like everything else ;)
534 2011-07-26 03:33:04 <lfm> well almost everything
535 2011-07-26 03:33:20 <luke-jr> at least everything in bitcoind
536 2011-07-26 03:34:24 <lfm> luke-jr ya , i guess it is the old debate, is the json a purely computer interface or is it a human interface too.
537 2011-07-26 03:34:44 <luke-jr> pretty sure that was never argued: it's obviously a computer interface
538 2011-07-26 03:35:02 <luke-jr> the only argument was that fixing it didn't justify breaking compatibility, nor adding a versioning layer
539 2011-07-26 03:35:17 <lfm> the raw json out put is exposed to human eyes tho thru the bitcoind command line commands
540 2011-07-26 03:35:38 <luke-jr> sure, but that's a test/debug interface, not meant for actual use
541 2011-07-26 03:36:07 <SerajewelKS> it's meant for use by CLI people
542 2011-07-26 03:36:19 <luke-jr> SerajewelKS: no it isn't.
543 2011-07-26 03:36:28 <forrestv> SerajewelKS, you can tell the python json parser to convert json numbers to Decimals ...
544 2011-07-26 03:36:31 <lfm> a human test debug interface, and it is prefectly feasible for humans to use it for bitcoin operations, I know I do, but maybe I am too weird to be considered human
545 2011-07-26 03:36:43 <SerajewelKS> forrestv: using which mechanism?
546 2011-07-26 03:36:47 <luke-jr> forrestv: that wouldn't work like people assume though
547 2011-07-26 03:36:50 <forrestv> SerajewelKS, one of the examples on http://docs.python.org/library/json.html
548 2011-07-26 03:36:57 <luke-jr> SerajewelKS: it's default for bitcoinrpc module
549 2011-07-26 03:36:58 <lfm> the bitcoind comand line comands
550 2011-07-26 03:37:14 <SerajewelKS> forrestv: neat
551 2011-07-26 03:37:22 <luke-jr> lfm: I do too, but someone using programic interfaces should know how to read them too
552 2011-07-26 03:37:31 <forrestv> luke-jr, what do you mean?
553 2011-07-26 03:37:35 <luke-jr> forrestv: even if you parse it as Decimal, you still need to round it
554 2011-07-26 03:37:51 <luke-jr> forrestv: bitcoind can send 0.03999999999999999 when it means 0.04
555 2011-07-26 03:38:09 <luke-jr> at least in theory
556 2011-07-26 03:38:21 <lfm> luke-jr well I spoze it could use satoshis but it would be kinda a pain
557 2011-07-26 03:38:36 <luke-jr> lfm: not at all. and the code's already been done for months
558 2011-07-26 03:39:08 <luke-jr> or you mean a pain to read?
559 2011-07-26 03:39:16 <lfm> without breaking aps that use the api?
560 2011-07-26 03:39:35 <luke-jr> lfm: my RPCv1 branch used a version for it
561 2011-07-26 03:39:55 <lfm> ya a pain to read. I spoze bitcoind could add some extra ui stuff to display amounts more reasonably anyway.
562 2011-07-26 03:39:56 <luke-jr> so ver 0 (default) was the old way, and ver 1 was satoshis
563 2011-07-26 03:40:14 <luke-jr> bitcoin-cli should display it nicer anyway :p
564 2011-07-26 03:40:24 <lfm> I think 0 and 1 are the same. 2 is satoshis
565 2011-07-26 03:40:45 <luke-jr> lfm: not in my original code, at least
566 2011-07-26 03:40:57 <SerajewelKS> lfm: if you're referring to my patch, 0/1 are the same and 2 uses string representation (but still uses a decimal point)
567 2011-07-26 03:40:57 <SerajewelKS> changing it to satoshis would be pretty easy
568 2011-07-26 03:41:00 <lfm> oh what SerajewelKS did
569 2011-07-26 03:41:04 <luke-jr> Spesmilo still has support for RPCv1, though I don't even bother anymore
570 2011-07-26 03:41:26 <luke-jr> SerajewelKS: there's no reason/use for the stringification as satoshis ;)
571 2011-07-26 03:42:06 <luke-jr> unless you're aiming to workaround the bugs in jansson, which is IMO a terrible reason
572 2011-07-26 03:42:55 <SerajewelKS> luke-jr: except that 1000000000000001 can't be represented in 32-bit floating-point, which is a perfectly legal json implementation. the problem is that json numbers don't convey any sort of precision or bit-length information, which is a serious deficiency for financial information.
573 2011-07-26 03:43:11 <lfm> as a string, just take out the . to get satoshis
574 2011-07-26 03:43:23 <luke-jr> SerajewelKS: it's reasonable to assume JSON Numbers must be able to represent ECMAScript Numbers
575 2011-07-26 03:43:26 <SerajewelKS> the json spec is intentionally vague about it, and that's no good for monetary units
576 2011-07-26 03:43:33 <luke-jr> lfm: doesn't work like that
577 2011-07-26 03:43:41 <SerajewelKS> luke-jr: reasonable != specified behavior
578 2011-07-26 03:44:00 <SerajewelKS> (specifically, the spec specifies no behavior at all)
579 2011-07-26 03:45:08 TheZimm has quit (Quit: Computer has gone to sleep.)
580 2011-07-26 03:45:24 <SerajewelKS> when dealing with vague number types, using strings is preferred to "some kind of number thingy"
581 2011-07-26 03:46:28 <luke-jr> better to just drop JSON altogether than worry about that IMO
582 2011-07-26 03:46:45 <luke-jr> or define the requirement as "JSON with at least 51 bits of number resolution"
583 2011-07-26 03:46:53 <luke-jr> (which is every implementation except jansson afaik)
584 2011-07-26 03:47:02 <luke-jr> actually, if you throw a "." on the end, jansson included
585 2011-07-26 03:47:15 <SerajewelKS> perhaps
586 2011-07-26 03:48:28 <luke-jr> IF strings are to be used, however, it is IMO logical to say "anything other than digits should be ignored"
587 2011-07-26 03:48:42 <luke-jr> then you can send a . for BTC, and even commas for thousand separators
588 2011-07-26 03:48:46 EPiSKiNG- has quit (Ping timeout: 255 seconds)
589 2011-07-26 03:49:04 <luke-jr> while also guaranteeing full satoshi resolution to parsers that just ignore thos
590 2011-07-26 03:49:17 <luke-jr> (ie, it'd have to be 0-padded on the right)
591 2011-07-26 03:49:50 <SerajewelKS> well my implementation always uses eight digits right of the decimal point, so for output you can just strip the .
592 2011-07-26 03:50:01 <SerajewelKS> input does consider the . though, so it would need to be tweaked
593 2011-07-26 03:50:29 <SerajewelKS> there could always be a suffix for satoshis; then you have satoshis with full compat with older clients
594 2011-07-26 03:50:51 <luke-jr> stringifying is not compatible no matter what you do
595 2011-07-26 03:51:11 <SerajewelKS> hence the versioning
596 2011-07-26 03:51:54 <luke-jr> RPCv1 had versioning and satoshis ☺
597 2011-07-26 03:53:13 <luke-jr> if you wanted to go crazy, you could do /?amount=strBTC /?amount=str /?amount= (with the default being /?amount=BTC) ;)
598 2011-07-26 03:53:32 <SerajewelKS> luke-jr: that idea did cross my mind :P
599 2011-07-26 03:54:14 <luke-jr> all of this seems to suggest one thing though: binary protocols are simplest :p
600 2011-07-26 03:54:58 <lfm> iiieee binary endianess protocols?
601 2011-07-26 03:55:03 <SerajewelKS> JSON is plain simple if you represent everything as strings, bools, null, or objects
602 2011-07-26 03:55:33 <SerajewelKS> the number type is useless since the spec leaves the implementing type open-ended
603 2011-07-26 03:55:49 <lfm> ok for 0 to 64000
604 2011-07-26 03:55:58 <jgarzik> heh
605 2011-07-26 03:56:11 <lfm> or 0 to 32000 I mean
606 2011-07-26 03:56:16 <SerajewelKS> 32767
607 2011-07-26 03:56:35 <lfm> you shouldnt assume it is binary based
608 2011-07-26 03:56:49 <luke-jr> lfm: there is a standard network endian, kthx
609 2011-07-26 03:56:52 <lfm> it might be running on a IBM 1401!
610 2011-07-26 03:56:54 <SerajewelKS> if it's not binary based then the 32000 division is meaningless
611 2011-07-26 03:57:10 <SerajewelKS> luke-jr: which is conveniently the opposite of most computers that exist today :(
612 2011-07-26 03:57:22 <lfm> ya, so?
613 2011-07-26 03:57:23 <luke-jr> SerajewelKS: I agree, quite convenient.
614 2011-07-26 03:57:31 <luke-jr> SerajewelKS: it forces people to write sane code. in theory.
615 2011-07-26 03:58:00 <JFK911> most computers? isnt x86 the only one?
616 2011-07-26 03:58:06 <luke-jr> JFK911: fail
617 2011-07-26 03:58:10 <lfm> or the MIX 1009 in its decimal version(s)
618 2011-07-26 03:58:19 <SerajewelKS> luke-jr: i think he's derping on purpose
619 2011-07-26 03:58:28 <luke-jr> maybe
620 2011-07-26 03:58:29 <lfm> derping?
621 2011-07-26 03:58:35 <luke-jr> but it's only fun if someone responds
622 2011-07-26 03:58:38 <luke-jr> so I went ahead
623 2011-07-26 03:58:56 <JFK911> iirc intel's the only one pigheaded enough
624 2011-07-26 03:59:01 <JFK911> it serves them right to have to adapt for the network
625 2011-07-26 03:59:10 <jrmithdobbs> lfm: original M
626 2011-07-26 03:59:12 <jrmithdobbs> ierr
627 2011-07-26 03:59:13 <JFK911> since they made pci difficult for everyone else with their endianness
628 2011-07-26 03:59:23 <JFK911> they had to come out with dual endian stuff
629 2011-07-26 03:59:25 <lfm> JFK911: the IBM 360/370 etc are little endian also
630 2011-07-26 03:59:27 <jrmithdobbs> lfm: original MIX for life. MMIX is a bastardization! ;p
631 2011-07-26 03:59:41 <JFK911> really? they had a funky word size iirc
632 2011-07-26 03:59:44 <jrmithdobbs> tapes and punch cards are good enough for everything!
633 2011-07-26 03:59:52 <lfm> jfk 32 bits
634 2011-07-26 04:00:08 <luke-jr> jgarzik: so still on the same decision (not worth changing JSON-RPC)?
635 2011-07-26 04:00:53 <lfm> JFK911: the "funkyest" part if the IBM 360 line was it generally did not use ASCII, it used ibm ebcidic
636 2011-07-26 04:00:57 <JFK911> right
637 2011-07-26 04:01:08 <JFK911> and wont ebcdic characters pack unevenly into that wordsize?
638 2011-07-26 04:01:14 <JFK911> arent they 6 bits or something?
639 2011-07-26 04:01:15 <jrmithdobbs> lfm: do not speak those foul words
640 2011-07-26 04:01:24 <luke-jr> IBM 5100 is fun
641 2011-07-26 04:01:32 <luke-jr> it ships with a supercomputer* emulator
642 2011-07-26 04:01:54 <jgarzik> luke-jr: there is no 'decision'. in my opinion, not speaking for anyone else, SerajewelKS's changing the URL to /v2 or whatever seems like a reasonable API switch... but that still has the "supporting multiple APIs" downside, and who knows if the community will like it
643 2011-07-26 04:01:56 <lfm> naw ebcidc was 8 bit. hollerith was 12 bit tho and would pack oddly if you were using them as raw data.
644 2011-07-26 04:01:59 <jgarzik> or use it, more importantly
645 2011-07-26 04:02:11 <jgarzik> if everybody stays on APIv0, it's just dead weight
646 2011-07-26 04:02:16 <luke-jr> jgarzik: so why didn't we do it when I implemented it months ago?
647 2011-07-26 04:02:47 <jgarzik> luke-jr: same reasons just listed. Maybe I missed it, but I did not see any community buy-in.
648 2011-07-26 04:02:50 <luke-jr> http://en.wikipedia.org/wiki/IBM_5100#Emulator_in_microcode
649 2011-07-26 04:02:57 eoss has quit (Remote host closed the connection)
650 2011-07-26 04:03:23 <JFK911> luke-jr: oh i didnt know that thing was a hypervisor
651 2011-07-26 04:03:35 <JFK911> they were really early with this vision
652 2011-07-26 04:03:43 <lfm> JFK911: there was quite a few systems that liked to use 6 bit char sets tho to save memory. the 360 wasnt one tho
653 2011-07-26 04:03:45 <JFK911> too bad it took the world 30+ years to catch up
654 2011-07-26 04:03:56 <luke-jr> JFK911: there's a lot of stuff like that ;P
655 2011-07-26 04:03:58 <JFK911> lfm: right, even early telecom circuits used five bit code
656 2011-07-26 04:04:07 <luke-jr> also: LOL @ See also: John Titor
657 2011-07-26 04:04:13 <luke-jr> jgarzik: does Linux support IBM 5100 ? :p
658 2011-07-26 04:04:26 <JFK911> luke-jr: doing it again: compare 3270 forms terminal with a web browser and contrast vs. character terminal
659 2011-07-26 04:04:29 <lfm> JFK911: ya the baudau code
660 2011-07-26 04:05:20 <JFK911> did telex tape ribbons store raw baudots?
661 2011-07-26 04:05:31 <JFK911> i remember they had a static row of dots down one column
662 2011-07-26 04:05:50 <lfm> JFK911: telex made both baudau and ascii terminals
663 2011-07-26 04:05:52 <JFK911> but last time i saw one, the coolest thing about the machine was the bucket of chads
664 2011-07-26 04:06:08 <lfm> baudot, right
665 2011-07-26 04:06:47 <lfm> JFK911: yup both paper tapes and punch cards had that "feature" you were always perpared for weddings
666 2011-07-26 04:08:10 <lfm> JFK911: ya the "parity" bit was often just hard coded to a one
667 2011-07-26 04:09:02 <lfm> that would have been on the 8 bit ascii versions, the 5 level tapes for baudot needed all 5 columns
668 2011-07-26 04:09:52 <lfm> the ascii teletypes often did not support lower case chars, theyed just print upper case for both
669 2011-07-26 04:12:13 thefinn93 has joined
670 2011-07-26 04:12:16 <JFK911> pretty easy to throw away the bit for lower case
671 2011-07-26 04:12:55 thefinn93 has quit (Client Quit)
672 2011-07-26 04:13:10 <lfm> there was card punches that only had upper case too iirc
673 2011-07-26 04:13:21 thefinn93 has joined
674 2011-07-26 04:18:29 zapnap has quit (Remote host closed the connection)
675 2011-07-26 04:19:24 sgornick has quit (Read error: Connection reset by peer)
676 2011-07-26 04:21:54 sgornick has joined
677 2011-07-26 04:37:03 eastender has quit (Quit: Nettalk6 - www.ntalk.de)
678 2011-07-26 04:40:17 aviadbd has joined
679 2011-07-26 04:40:34 <aviadbd> resending a question i asked at #bitcoin:
680 2011-07-26 04:40:42 <aviadbd> when i start RPCMiner, I get the following message:
681 2011-07-26 04:40:49 <aviadbd> Target = 00000000ffffffffffffffffffffffffffffffffffffffffffffffffffffff
682 2011-07-26 04:40:54 <aviadbd> now, that in itself isnt odd,
683 2011-07-26 04:41:00 <aviadbd> but the target here has only 62 characters, which makes it 31 bytes. shouldn't it be 32?
684 2011-07-26 04:41:21 TuxBlackEdo has joined
685 2011-07-26 04:47:01 <lfm> aviadbd: yes it should be 32 bytes
686 2011-07-26 04:48:10 zleeb has quit (Quit: Leave me alone. kthx)
687 2011-07-26 04:48:23 RBecker has quit (Quit: You care. You're there for me. You love me so much, and I never want to let it go. You are the one truly amazing person. MDR 3/6/11 <3)
688 2011-07-26 04:48:56 <aviadbd> lfm: so where's the missing byte?
689 2011-07-26 04:49:08 wood has joined
690 2011-07-26 04:49:26 <lfm> youd have to look at your rpcminer to figure it out I guess
691 2011-07-26 04:49:34 eian has quit (Quit: Leaving)
692 2011-07-26 04:49:36 <aviadbd> the code you mean?
693 2011-07-26 04:49:41 <lfm> ya
694 2011-07-26 04:49:44 <aviadbd> hmph
695 2011-07-26 04:49:47 <aviadbd> i was hoping to avoid that
696 2011-07-26 04:49:58 <aviadbd> for now i just assume there are prefixed zeros to the string that i don't see
697 2011-07-26 04:50:07 <aviadbd> (or rather, that rpcminer adds them automatically)
698 2011-07-26 04:50:15 <aviadbd> that's a reasonable assumption I think
699 2011-07-26 04:50:19 <aviadbd> even though it makes the target harder
700 2011-07-26 04:50:26 <aviadbd> so it might be that there are trailing FF's
701 2011-07-26 04:50:28 <aviadbd> ..
702 2011-07-26 04:50:32 RBecker has joined
703 2011-07-26 04:50:37 <aviadbd> damn. i have to go to the code. :/
704 2011-07-26 04:50:50 Joric has joined
705 2011-07-26 04:50:55 <lfm> ya, most pools use target with just 8 zeros
706 2011-07-26 04:52:07 <aviadbd> so its reasonable to assume trailing FF's?
707 2011-07-26 04:52:21 <lfm> which pool is it?
708 2011-07-26 04:52:23 wirehead has quit (Ping timeout: 246 seconds)
709 2011-07-26 04:53:12 RenaKunisaki has quit (Ping timeout: 252 seconds)
710 2011-07-26 04:53:47 <aviadbd> eligius
711 2011-07-26 04:54:04 RenaKunisaki has joined
712 2011-07-26 04:55:04 <aviadbd> sigh - gotta go. thanks !
713 2011-07-26 04:55:43 aviadbd has quit (Quit: thanks)
714 2011-07-26 05:07:27 blueadept has quit (Quit: Leaving)
715 2011-07-26 05:11:24 rethaw has joined
716 2011-07-26 05:13:12 Beccara has quit (Read error: Connection reset by peer)
717 2011-07-26 05:17:11 TheZimm has joined
718 2011-07-26 05:22:11 Beccara has joined
719 2011-07-26 05:25:11 TheZimm has quit (Quit: Computer has gone to sleep.)
720 2011-07-26 05:26:53 josephcp has joined
721 2011-07-26 05:30:17 Beccara has quit (Ping timeout: 240 seconds)
722 2011-07-26 05:33:41 sanchaz has quit (Ping timeout: 250 seconds)
723 2011-07-26 05:36:59 Zarutian has quit (Quit: Zarutian)
724 2011-07-26 05:38:51 pimpinganiteasy has quit (Ping timeout: 240 seconds)
725 2011-07-26 05:40:31 sanchaz has joined
726 2011-07-26 05:40:43 <upb> http://anonymity-in-bitcoin.blogspot.com/2011/07/bitcoin-is-not-anonymous.html
727 2011-07-26 05:40:48 <upb> ^- really interesting stuff
728 2011-07-26 05:41:46 <josephcp> Is there an easy function/patch/rcp-call which can give the bitcoin balance in unspent txout ids (instead of by account or address)?
729 2011-07-26 05:41:47 RobinPKR has quit (Quit: RobinPKR)
730 2011-07-26 05:42:00 <josephcp> rpc
731 2011-07-26 05:42:26 Beccara has joined
732 2011-07-26 05:43:30 RobinPKR has joined
733 2011-07-26 05:44:56 <josephcp> just wondering if there's an existing patch before i start writing my own rpc call....
734 2011-07-26 05:47:14 kreal- has quit ()
735 2011-07-26 05:56:43 kreal- has joined
736 2011-07-26 05:58:36 kreal- has left ()
737 2011-07-26 05:59:02 kreal- has joined
738 2011-07-26 05:59:17 josephcp has quit (Changing host)
739 2011-07-26 05:59:17 josephcp has joined
740 2011-07-26 05:59:54 <yebyen> if i have a testnet sending a million bitcoins, the transaction sits as unconfirmed for a while...
741 2011-07-26 05:59:59 <yebyen> is there any way to push it through?
742 2011-07-26 06:00:22 <yebyen> something easy, like send it to all of the nodes, would be nice...
743 2011-07-26 06:00:53 <yebyen> i'm assuming it's just too big, and none of the nodes can solve all of the blocks quickly
744 2011-07-26 06:01:03 <yebyen> and the answer is no
745 2011-07-26 06:01:18 RazielZ has joined
746 2011-07-26 06:02:12 <josephcp> bitcoins don't work that way, it's as easy to send a million bitcoin as it is sending 0.00000001 bitcoin in a block, i think you're confusing a block and a bitcoin?
747 2011-07-26 06:02:32 <yebyen> the bitcoins are received, they just never are confirmed
748 2011-07-26 06:02:33 <josephcp> or do you mean a million transactions?
749 2011-07-26 06:02:51 <yebyen> i mean a million bitcoins, most of the bitcoins on the network really
750 2011-07-26 06:02:57 <yebyen> i think i lost about 500k
751 2011-07-26 06:03:05 <yebyen> misplaced
752 2011-07-26 06:03:16 <yebyen> but i still have 1mil
753 2011-07-26 06:03:16 <josephcp> are they unconfirmed?
754 2011-07-26 06:03:25 <yebyen> 1mil unconfirmed
755 2011-07-26 06:03:30 <yebyen> i keep sending them from node to node
756 2011-07-26 06:04:04 <yebyen> they were mature blocks, i'm guessing nothing has been able to collect the tx fees and that's why it's unconfirmed
757 2011-07-26 06:04:18 <josephcp> what was your tx fee
758 2011-07-26 06:04:37 <yebyen> i'm not sure...
759 2011-07-26 06:04:59 <yebyen> let me load up the node that sent them two sends ago
760 2011-07-26 06:05:55 <yebyen> 0.48
761 2011-07-26 06:06:05 <yebyen> assuming the next one sent with the same tx fee
762 2011-07-26 06:06:21 <yebyen> both transactions are unconfirmed
763 2011-07-26 06:06:24 RobinPKR has quit (Quit: RobinPKR)
764 2011-07-26 06:06:27 <josephcp> dumb question: is your client connected to other nodes?
765 2011-07-26 06:06:34 <yebyen> there are 4 nodes
766 2011-07-26 06:06:45 <yebyen> each under 1MHash/sec
767 2011-07-26 06:06:52 <josephcp> and how long have you waited?
768 2011-07-26 06:06:57 <yebyen> except for the big one, my desktop does 2.8MHash/sec
769 2011-07-26 06:06:57 <josephcp> testnet is solving blocks really slow
770 2011-07-26 06:07:07 <yebyen> this is a private testnet
771 2011-07-26 06:07:09 <josephcp> OH
772 2011-07-26 06:07:17 <josephcp> OHOHOH
773 2011-07-26 06:07:38 RobinPKR has joined
774 2011-07-26 06:07:38 <josephcp> let me re-read everything
775 2011-07-26 06:07:40 <yebyen> around version 0.3.11.0 beta
776 2011-07-26 06:08:05 <yebyen> i made the binaries incompatible with standard coins following a tutorial that gavin released before that time
777 2011-07-26 06:08:13 <yebyen> and lost the incompatible changes, now i just have the binaries
778 2011-07-26 06:09:48 <yebyen> so, here i see 07/07/2011 unconfirmed 989700 credit
779 2011-07-26 06:09:58 <imsaguy> no confirmations?
780 2011-07-26 06:10:13 <yebyen> zero
781 2011-07-26 06:10:55 <yebyen> followed by 07/25/2011 sending 1000000.48 (0.48 tx fee)
782 2011-07-26 06:11:26 <yebyen> and now i just have a million testnet coins floating around unconfirmed wherever they stop
783 2011-07-26 06:11:50 zamgo_ has joined
784 2011-07-26 06:11:56 <imsaguy> well
785 2011-07-26 06:12:04 <imsaguy> you could reset the blockchain
786 2011-07-26 06:12:19 <imsaguy> if no nodes are registering the transaction
787 2011-07-26 06:12:24 <imsaguy> then the original wallet still owns
788 2011-07-26 06:13:49 <josephcp> well i think the easy way to go about it if all the nodes are rejecting the transaction would be to delete everything except for the wallet.dat on the computer that sent it no?
789 2011-07-26 06:14:01 <yebyen> they've been generating together
790 2011-07-26 06:14:18 <yebyen> i'm not sure any node is completely responsible for a million bitcoins
791 2011-07-26 06:14:28 <josephcp> but you sent it from one computer right?
792 2011-07-26 06:14:29 <imsaguy> lol
793 2011-07-26 06:14:47 <yebyen> somewhere i will find the first send that was unconfirmed, i guess
794 2011-07-26 06:14:47 <imsaguy> if you sent a million bitcoins, one wallet had to be the container
795 2011-07-26 06:15:18 <imsaguy> if no wallet had a millioin coins, then thats why the transaction never confirmed.. you can't send coins you don't have
796 2011-07-26 06:15:21 Jefff2 has quit (Ping timeout: 240 seconds)
797 2011-07-26 06:15:23 <imsaguy> million*
798 2011-07-26 06:15:45 <josephcp> clear out everything except for one computer (minus wallet.dat), resend the blockchain to the other computers, delete everything (except for wallet.dat) in the first computer
799 2011-07-26 06:16:05 citiz3n has quit (Ping timeout: 246 seconds)
800 2011-07-26 06:16:07 <imsaguy> exactly
801 2011-07-26 06:16:07 <josephcp> i mean it's a hacky way to go about it, but you can probably do that in 10 minutes :-P
802 2011-07-26 06:16:16 <imsaguy> I had to do it in RL before
803 2011-07-26 06:16:19 <yebyen> ok, i'm willing to try it
804 2011-07-26 06:17:38 <josephcp> it might not work if you discovered a weird bug and the transactions are accepted but not processed for some reason, but if you're doing straight sends i doubt that would crop up
805 2011-07-26 06:17:44 <yebyen> addr.dat, blk0001.dat, blkindex.dat, db.log, database, debug.log...
806 2011-07-26 06:17:53 <yebyen> all getting deleted on all but one computer
807 2011-07-26 06:18:08 <imsaguy> you don't have to all computers
808 2011-07-26 06:18:13 <imsaguy> just the one that sent the coins originally
809 2011-07-26 06:19:14 <yebyen> if i'm not sure which one that is...
810 2011-07-26 06:19:26 <yebyen> (one machine, probably the original, won't launch X clients)
811 2011-07-26 06:19:38 <yebyen> does it really matter which one?
812 2011-07-26 06:20:05 <yebyen> i would rather delete the block chains on the computers that don't have a million anymore
813 2011-07-26 06:20:22 <josephcp> no it doesn't matter, just make sure you don't delete the wallet on any of the computers
814 2011-07-26 06:20:32 <josephcp> and one computer has the blockchain at all times
815 2011-07-26 06:20:34 <josephcp> at least
816 2011-07-26 06:20:36 <imsaguy> gotta have a blockchain somehwere
817 2011-07-26 06:20:40 <imsaguy> otherwise, the wallet's won't matter
818 2011-07-26 06:20:46 <josephcp> imsaguy: hehehehe
819 2011-07-26 06:20:50 <yebyen> that makes sense
820 2011-07-26 06:21:20 <imsaguy> whichever blockchain is left, will by default be the hardest
821 2011-07-26 06:21:29 <imsaguy> and the million coins will be somewhere
822 2011-07-26 06:21:34 <imsaguy> either sent/confirmed or not sent
823 2011-07-26 06:21:37 <imsaguy> but not 'lost'
824 2011-07-26 06:21:54 <yebyen> there was a wallet that was lost, but that was long before the 1mil was generated
825 2011-07-26 06:21:59 <yebyen> i might have had 200k or 500k in it
826 2011-07-26 06:22:09 <imsaguy> without the wallet, they are lost for good
827 2011-07-26 06:22:11 <imsaguy> no private keys
828 2011-07-26 06:22:39 Folklore has quit (Ping timeout: 250 seconds)
829 2011-07-26 06:23:19 Folklore has joined
830 2011-07-26 06:24:40 <yebyen> the three with no block chains are downloading it now
831 2011-07-26 06:24:41 <yebyen> at least one is
832 2011-07-26 06:24:51 <imsaguy> it all depends on how they peered
833 2011-07-26 06:25:00 <imsaguy> give it time
834 2011-07-26 06:25:29 <yebyen> of course i left the block chain on the computer with the least bandwidth upstream...
835 2011-07-26 06:25:36 <yebyen> that I'm sitting in front of
836 2011-07-26 06:25:42 <josephcp> if it still doesn't work, try changing your MIN_RELAY_TX_FEE and MIN_TX_FEE to 0 in main.h
837 2011-07-26 06:25:50 <imsaguy> can't
838 2011-07-26 06:25:53 <imsaguy> only has binaries left
839 2011-07-26 06:25:54 <yebyen> lost the source
840 2011-07-26 06:25:55 <yebyen> yeah
841 2011-07-26 06:25:59 <josephcp> oh i see
842 2011-07-26 06:26:01 <imsaguy> yeah
843 2011-07-26 06:26:10 <imsaguy> doh
844 2011-07-26 06:26:15 <yebyen> if i can't confirm this million, i might as well start with the testnet in a box from scratch
845 2011-07-26 06:26:20 <imsaguy> yeah
846 2011-07-26 06:26:23 <imsaguy> reset the difficult
847 2011-07-26 06:26:29 <imsaguy> revert to the current testnet
848 2011-07-26 06:26:47 <yebyen> i'm not interested in connecting to arbitrary testnet nodes, :)
849 2011-07-26 06:26:51 <imsaguy> no
850 2011-07-26 06:27:00 <yebyen> or downloading their blockchain
851 2011-07-26 06:27:01 <imsaguy> I meant grab a copy of the current testnet source
852 2011-07-26 06:27:04 <yebyen> right
853 2011-07-26 06:27:06 <imsaguy> and start fresh
854 2011-07-26 06:27:08 Jefff has joined
855 2011-07-26 06:27:18 <yebyen> i tried it on my netbook, it's got some new features
856 2011-07-26 06:27:24 <yebyen> accounts
857 2011-07-26 06:28:32 <imsaguy> so why all the time and effort on testnet when you could be making real coins?
858 2011-07-26 06:28:33 <josephcp> hrm i've always wondered, what happens when there's two transactions which spend the same coins on the network? do existing nodes relay the second transaction?
859 2011-07-26 06:28:45 <imsaguy> it'd be a race condition
860 2011-07-26 06:28:48 <yebyen> with 4MHash/sec?
861 2011-07-26 06:28:52 <imsaguy> eventually, one would be accepted, the other not
862 2011-07-26 06:29:04 <imsaguy> lol yebyen, anything's better than nothing?
863 2011-07-26 06:29:08 <josephcp> if the first one is on a node and is sent the second one does it relay the transaction to other nodes or does it drop it?
864 2011-07-26 06:29:24 <imsaguy> probably drops it
865 2011-07-26 06:29:27 <imsaguy> but I dunno
866 2011-07-26 06:29:32 <josephcp> yeah i've always wondered that
867 2011-07-26 06:29:43 <imsaguy> ask yebyen to try it ;)
868 2011-07-26 06:29:45 <yebyen> imsaguy: i think i'm more likely to make a json-rpc application that earns money somehow...
869 2011-07-26 06:29:50 <imsaguy> lol
870 2011-07-26 06:29:52 <josephcp> hehe exactly why i wondered :-P
871 2011-07-26 06:30:00 <imsaguy> just start some website
872 2011-07-26 06:30:07 <imsaguy> everyone else is doing it
873 2011-07-26 06:30:25 <yebyen> i have cornered the market on bitcoin craps
874 2011-07-26 06:30:42 <yebyen> someone approached me about it and he's actually built a craps game, but not hoked it up to bitcoin yet
875 2011-07-26 06:30:57 <yebyen> just wants me to write a pretty UI
876 2011-07-26 06:31:00 <yebyen> lol
877 2011-07-26 06:31:24 <imsaguy> lol
878 2011-07-26 06:31:27 <yebyen> i have the google results
879 2011-07-26 06:31:30 <imsaguy> of course
880 2011-07-26 06:31:39 <imsaguy> google can be bought ;)
881 2011-07-26 06:31:40 <yebyen> "so we will be able to buy flying pigs with our quest gold"
882 2011-07-26 06:31:41 <SuprTiggr> http://www.youtube.com/watch?v=sP5BNT5Fpvg
883 2011-07-26 06:31:50 DD- has joined
884 2011-07-26 06:32:43 <yebyen> i've moved on to experimenting with zookeeper
885 2011-07-26 06:32:58 <yebyen> since it does things that memcached shouldn't
886 2011-07-26 06:33:11 <imsaguy> zoocoins?
887 2011-07-26 06:33:18 <imsaguy> oops, wrong channel
888 2011-07-26 06:35:14 <yebyen> halfway there...
889 2011-07-26 06:35:28 <yebyen> should be able to tell you all if it worked before I go to bed
890 2011-07-26 06:36:26 <imsaguy> awesome
891 2011-07-26 06:37:38 <jgarzik> SerajewelKS: atoi()? really? in remote, attacker-facing code?
892 2011-07-26 06:39:47 <SerajewelKS> jgarzik: there are no buffer overflow vulnerabilities in atoi to my knowledge. what vulnerabilities are you aware of?
893 2011-07-26 06:40:06 <rethaw> ya whats wrong with atoi?
894 2011-07-26 06:42:46 <doublec> atoi has no way of reporting an error
895 2011-07-26 06:42:53 <SerajewelKS> the only issues with atoi that i'm aware of are "it doesn't report errors" and "it might overflow," neither of which i give a shit about
896 2011-07-26 06:42:57 <doublec> so long string numbers can overlfow, etc
897 2011-07-26 06:43:30 <SerajewelKS> if someone uses /bitcoin.v234zomg i don't care. they'll get version 234. what's wrong with that?
898 2011-07-26 06:43:54 <jgarzik> SerajewelKS: you can pass negative numbers to it, and you don't check for that
899 2011-07-26 06:43:57 <jgarzik> sigh
900 2011-07-26 06:43:58 <jgarzik> C 101
901 2011-07-26 06:44:10 <SerajewelKS> because i don't CARE about negative numbers. API version -1 is equivalent to 0.
902 2011-07-26 06:44:22 <SerajewelKS> show me how this is a problem
903 2011-07-26 06:44:26 <jgarzik> not true throughout the code
904 2011-07-26 06:44:37 <SerajewelKS> throughout my code it is
905 2011-07-26 06:45:17 <SerajewelKS> jgarzik: considering that bitcoind considers "FUCKOFF bazinga NOODLE" a valid HTTP request, you are really worried about this?
906 2011-07-26 06:45:25 copumpkin has quit (Ping timeout: 252 seconds)
907 2011-07-26 06:45:43 copumpkin has joined
908 2011-07-26 06:45:49 <yebyen> haha bazinga...
909 2011-07-26 06:45:53 <zamgo_> heh
910 2011-07-26 06:46:06 <SerajewelKS> i check the version in exactly one place: if (api_version < 2)
911 2011-07-26 06:46:11 <zamgo_> 408 Noodle Not Found
912 2011-07-26 06:47:34 <SerajewelKS> jgarzik: if you'd feel more comfortable if i checked for negative numbers and conversion errors then ok, but can we also check the HTTP method and version then?
913 2011-07-26 06:47:48 <SerajewelKS> even though none of it /really/ matters?
914 2011-07-26 06:48:42 QueryTom3000 has joined
915 2011-07-26 06:54:10 <yebyen> re-downloading the block chain to the computer with a million bitcoins...
916 2011-07-26 06:54:34 <SerajewelKS> well, i need to go to bed. if the only complaint with my patch is that they can enter negative version numbers, then i guess i did good, since negative version numbers will behave like a zero. :)
917 2011-07-26 06:56:58 <RenaKunisaki> is it normal for bitcoind to use 256MB of RAM?
918 2011-07-26 06:57:11 <doublec> no, it's usually higher
919 2011-07-26 06:57:13 <SerajewelKS> RenaKunisaki: it's normal for it to use way more than that, in my experience
920 2011-07-26 06:57:31 * SerajewelKS high-fives doublec
921 2011-07-26 06:57:47 <doublec> hehe
922 2011-07-26 07:08:33 Joric has quit ()
923 2011-07-26 07:09:08 wirehead has joined
924 2011-07-26 07:09:08 Shuro__ is now known as Shuro
925 2011-07-26 07:09:38 Shuro is now known as Guest89930
926 2011-07-26 07:10:23 Guest89930 is now known as Shuro_
927 2011-07-26 07:11:32 danbri_ is now known as danbri
928 2011-07-26 07:13:45 dr_win has quit (Remote host closed the connection)
929 2011-07-26 07:14:31 dr_win has joined
930 2011-07-26 07:14:58 BTCTrader_ is now known as BTCTrader
931 2011-07-26 07:18:04 nefario has left ()
932 2011-07-26 07:18:23 <yebyen> well, one block has been generated since all of the nodes re-downloaded the blockchain, and my 1mil is still 0/unconfirmed
933 2011-07-26 07:19:28 <yebyen> but it's chugging pretty hard, the khash/s is half of what it usually looks like
934 2011-07-26 07:19:56 <da2ce7> EXCEPTION: 22DbRunRecoveryException
935 2011-07-26 07:20:06 <da2ce7> clean install of bitcoind
936 2011-07-26 07:20:17 <da2ce7> what(): DbEnv::open: DB_RUNRECOVERY: Fatal error, run database recovery
937 2011-07-26 07:28:46 thefinn93 has quit (Quit: BAI)
938 2011-07-26 07:30:03 Akinava is now known as away!~lis@babylon.saf-14.ru|Akinava
939 2011-07-26 07:30:07 <random_cat> i hate that error;
940 2011-07-26 07:30:34 <da2ce7> i'm gonna upload the entire .bitcoin folder...
941 2011-07-26 07:30:40 <da2ce7> don't have anything private in it.
942 2011-07-26 07:34:12 Joric has joined
943 2011-07-26 07:34:53 MrTiggr has quit (Ping timeout: 252 seconds)
944 2011-07-26 07:40:49 <Joric> how do you think what are all those 18k transactions
945 2011-07-26 07:40:50 <Joric> http://www.bitcoinmonitor.com
946 2011-07-26 07:41:12 <Joric> scroll back a little
947 2011-07-26 07:42:14 zamgo_ has left ()
948 2011-07-26 07:43:36 <Joric> it started from 18579 BTC and ended at 17577, 28 transactions overall, each next smaller than the previous one
949 2011-07-26 07:47:33 mrb_ has quit (Ping timeout: 264 seconds)
950 2011-07-26 07:47:41 <da2ce7> http://69.64.53.25/bitcoin_userdata.tar.bz
951 2011-07-26 07:47:46 mrb_ has joined
952 2011-07-26 07:47:48 <da2ce7> has all the log and everything
953 2011-07-26 07:48:46 <Joric> da2ce7, what is this?
954 2011-07-26 07:48:56 <da2ce7> EXCEPTION: 22DbRunRecoveryException
955 2011-07-26 07:49:09 <da2ce7> all the userdata from the system
956 2011-07-26 07:49:18 <da2ce7> it was a fresh install so no private data
957 2011-07-26 07:49:41 <Joric> how did it happen?
958 2011-07-26 07:49:52 <da2ce7> just ran bitcoind
959 2011-07-26 07:49:55 <da2ce7> on clean install
960 2011-07-26 07:50:18 <da2ce7> then it complained about not having the bitcoin.conf... so I made it... then on the next run it was failing on a db error
961 2011-07-26 07:51:00 <da2ce7> bitcoin-0.3.24
962 2011-07-26 07:51:04 karnac has quit (Quit: karnac)
963 2011-07-26 07:51:09 <da2ce7> using 64bit
964 2011-07-26 07:51:24 <Joric> wallet is fine
965 2011-07-26 07:51:30 <Joric> just delete other files
966 2011-07-26 07:52:02 <da2ce7> I know... just wondered if anyone wanted to investigate the bug that made this error
967 2011-07-26 07:55:58 EPiSKiNG- has joined
968 2011-07-26 08:02:35 Cryo has quit (Quit: Terminated with extreme prejudice - dircproxy 1.2.0)
969 2011-07-26 08:03:41 Shuro_ is now known as Shuro
970 2011-07-26 08:04:21 Shuro is now known as Guest64976
971 2011-07-26 08:04:33 Guest64976 is now known as Shuro_
972 2011-07-26 08:04:59 josephcp has quit (Quit: leaving)
973 2011-07-26 08:07:12 d1g1t4l has joined
974 2011-07-26 08:07:12 Beccara has quit (Ping timeout: 260 seconds)
975 2011-07-26 08:07:15 <da2ce7> is there a command to check bitcoind's verson?
976 2011-07-26 08:08:53 <da2ce7> ah don't worry... found it.
977 2011-07-26 08:10:31 _Silverpike_ has quit (Ping timeout: 260 seconds)
978 2011-07-26 08:14:27 sacredchao has quit (Ping timeout: 250 seconds)
979 2011-07-26 08:28:27 molecular has joined
980 2011-07-26 08:30:41 kish` has joined
981 2011-07-26 08:31:26 dvide has joined
982 2011-07-26 08:33:00 Folklore has quit ()
983 2011-07-26 08:33:31 kish`_ has quit (Ping timeout: 250 seconds)
984 2011-07-26 08:35:10 hugolp has joined
985 2011-07-26 08:37:42 OneTimePad has quit (Max SendQ exceeded)
986 2011-07-26 08:43:54 AndyBr has quit ()
987 2011-07-26 08:46:26 Silverpike has joined
988 2011-07-26 08:55:11 Fireball has joined
989 2011-07-26 08:55:17 larsivi has quit (Ping timeout: 240 seconds)
990 2011-07-26 08:56:16 redhatzero has joined
991 2011-07-26 08:56:55 fnord0 has quit (Ping timeout: 250 seconds)
992 2011-07-26 08:57:34 Marf has joined
993 2011-07-26 08:58:41 larsivi has joined
994 2011-07-26 09:05:51 dr_win has quit (Remote host closed the connection)
995 2011-07-26 09:09:58 fnord0 has joined
996 2011-07-26 09:10:05 Cherothald has quit (Ping timeout: 276 seconds)
997 2011-07-26 09:14:45 erus` has joined
998 2011-07-26 09:15:24 josephcp has joined
999 2011-07-26 09:21:29 dr_win has joined
1000 2011-07-26 09:23:56 Silverpike has quit (Ping timeout: 276 seconds)
1001 2011-07-26 09:33:06 <da2ce7> wow... bitcoin taks ages to compile.
1002 2011-07-26 09:34:48 asuk has joined
1003 2011-07-26 09:36:17 TD has joined
1004 2011-07-26 09:39:45 bitcoinbulletin has quit (Ping timeout: 264 seconds)
1005 2011-07-26 09:39:45 ahihi2 has quit (Ping timeout: 264 seconds)
1006 2011-07-26 09:40:39 Clipse has joined
1007 2011-07-26 09:44:37 bitcoinbulletin has joined
1008 2011-07-26 09:45:00 altamic has joined
1009 2011-07-26 09:45:00 altamic has quit (Client Quit)
1010 2011-07-26 09:54:18 rethaw has quit (Quit: rethaw)
1011 2011-07-26 09:58:21 datagutt has joined
1012 2011-07-26 09:59:30 Clipse has quit (Read error: Connection reset by peer)
1013 2011-07-26 09:59:48 enquirer has joined
1014 2011-07-26 10:00:17 shLONG has joined
1015 2011-07-26 10:01:35 TheAncientGoat has joined
1016 2011-07-26 10:05:22 larsivi has quit (Ping timeout: 240 seconds)
1017 2011-07-26 10:06:30 m00p has joined
1018 2011-07-26 10:13:23 senseles has quit (Read error: Connection reset by peer)
1019 2011-07-26 10:13:32 senseles has joined
1020 2011-07-26 10:13:32 Clipse has joined
1021 2011-07-26 10:14:21 TD has quit (Quit: TD)
1022 2011-07-26 10:16:24 <jtaylor> welcome to the wonderful world of c++ ^^
1023 2011-07-26 10:20:25 shLONG has quit ()
1024 2011-07-26 10:20:41 shLONG has joined
1025 2011-07-26 10:21:07 <josephcp> you could try adding the compilation flag -j 4 if you have multiple cpu cores, i think it compiles with only one thread?
1026 2011-07-26 10:21:39 <josephcp> or -pipe
1027 2011-07-26 10:22:03 ahbritto_ has quit (Ping timeout: 250 seconds)
1028 2011-07-26 10:24:01 <Joric> dive into python!
1029 2011-07-26 10:25:07 TiggrBot has quit (Ping timeout: 250 seconds)
1030 2011-07-26 10:25:20 suriv has quit (Ping timeout: 250 seconds)
1031 2011-07-26 10:25:46 zeropointo has quit (Ping timeout: 250 seconds)
1032 2011-07-26 10:25:55 ahbritto_ has joined
1033 2011-07-26 10:28:14 zeropointo has joined
1034 2011-07-26 10:29:00 Marf has quit (Read error: Connection reset by peer)
1035 2011-07-26 10:29:11 <Joric> http://dump.bitcheese.net/files/amefezo/dive-into-python.png
1036 2011-07-26 10:30:54 QueryTom3000 has left ()
1037 2011-07-26 10:32:08 suriv has joined
1038 2011-07-26 10:34:53 dbitcoin has joined
1039 2011-07-26 10:39:13 x6763 has quit (Remote host closed the connection)
1040 2011-07-26 10:39:31 Tril has joined
1041 2011-07-26 10:47:04 x6763 has joined
1042 2011-07-26 10:47:10 <erus`> learn you a haskell!
1043 2011-07-26 10:47:47 <erus`> funny pic :P
1044 2011-07-26 10:48:54 AStove has joined
1045 2011-07-26 10:54:01 Tril has quit (Ping timeout: 252 seconds)
1046 2011-07-26 10:54:15 Clipse has quit (Read error: Connection reset by peer)
1047 2011-07-26 10:56:41 asuk has quit (Ping timeout: 276 seconds)
1048 2011-07-26 10:57:17 moa7 has quit (Quit: Page closed)
1049 2011-07-26 10:57:46 BlueMatt has joined
1050 2011-07-26 10:59:36 dbitcoin has quit (Quit: Ex-Chat)
1051 2011-07-26 11:00:54 <prof7bit> help me port it to pascal. then it will compile in 1.5 seconds.
1052 2011-07-26 11:03:25 <erus`> we are porting to vb5 with a access backend
1053 2011-07-26 11:03:34 <erus`> webscale!
1054 2011-07-26 11:03:36 <Joric> nasty!
1055 2011-07-26 11:05:21 Shuro_ is now known as Shuro
1056 2011-07-26 11:05:37 <prof7bit> erus`: vb5 is 1997, are you trying to be funny?
1057 2011-07-26 11:06:23 pimpinganiteasy has joined
1058 2011-07-26 11:06:32 <erus`> trying
1059 2011-07-26 11:07:46 <Joric> you started first
1060 2011-07-26 11:10:19 <prof7bit> no
1061 2011-07-26 11:10:41 pimpinganiteasy has quit (Ping timeout: 255 seconds)
1062 2011-07-26 11:11:06 Tril has joined
1063 2011-07-26 11:13:56 Clipse has joined
1064 2011-07-26 11:15:53 nus has quit (Ping timeout: 240 seconds)
1065 2011-07-26 11:20:46 nefario has joined
1066 2011-07-26 11:29:09 denisx has joined
1067 2011-07-26 11:33:42 iddo has quit (Ping timeout: 260 seconds)
1068 2011-07-26 11:35:12 iddo has joined
1069 2011-07-26 11:38:03 nus has joined
1070 2011-07-26 11:39:56 molecular has quit (Ping timeout: 255 seconds)
1071 2011-07-26 12:00:32 Rabbit67890 has joined
1072 2011-07-26 12:04:32 BlueMatt has quit (Quit: Ex-Chat)
1073 2011-07-26 12:08:32 asuk has joined
1074 2011-07-26 12:08:59 BlueMatt has joined
1075 2011-07-26 12:17:13 agricocb has quit (Quit: Leaving.)
1076 2011-07-26 12:17:13 uber` has quit (Read error: Connection reset by peer)
1077 2011-07-26 12:19:48 Stryker has quit (Ping timeout: 260 seconds)
1078 2011-07-26 12:25:55 bobd0bb has joined
1079 2011-07-26 12:33:14 <prof7bit> regarding "confirmations": I see two numbers being mentioned more often than others, the one is 6 blocks and the other one is 120 blocks. Are these just arbitrary recommendations (based on guesses) or is there actually something hardcoded in the protocol that would make it impossible after 120 blocks to reorganize the chain and if it happened the world would end or something like that?
1080 2011-07-26 12:33:50 <mtrlt> no, it's hardcoded in the client
1081 2011-07-26 12:34:24 <jtaylor> no its arbitrary
1082 2011-07-26 12:34:25 <mtrlt> and if a chain reorg happened after 120, people would just lose bitcoins :p
1083 2011-07-26 12:34:48 <jtaylor> 6 blocks is just a reasonable guess based on random walk
1084 2011-07-26 12:34:55 <mtrlt> yep
1085 2011-07-26 12:35:19 <jtaylor> but the value is hardcoded, but there is nothing preventing a block with more confirmations to be invalidated
1086 2011-07-26 12:35:56 <jtaylor> e.g. by someone having more than 50% of the hashing power
1087 2011-07-26 12:36:12 <jtaylor> its very unlikely for someone with less to manage that
1088 2011-07-26 12:38:04 agricocb has joined
1089 2011-07-26 12:38:38 <prof7bit> has anybody done the math already how "often" the 6 and the 120 blocks event would be expected to happen?
1090 2011-07-26 12:38:53 <jtaylor> there is a calculation in the satoshi paper
1091 2011-07-26 12:39:48 <edcba> 6 ?
1092 2011-07-26 12:41:01 Stellar has joined
1093 2011-07-26 12:41:13 <prof7bit> edcba: i don't know, i ust hear the number 6 mentioned very often and I have not yet looked into the code to determine what *exactly* it allows/disallows/assumes after 6 and after 120
1094 2011-07-26 12:41:58 <jtaylor> nothing disallows it, just the probablilty that it happens by chance is very low with > 6 confirmations
1095 2011-07-26 12:42:08 <jtaylor> one confirmation is also pretty safe, but I don't know the number
1096 2011-07-26 12:43:18 <prof7bit> i used the words "allow/disallow" because the word "hardcoded" has been mentioned, so I assume that "someting" is done with that number
1097 2011-07-26 12:43:29 nefario has left ()
1098 2011-07-26 12:44:26 <jtaylor> no client just considers 6 confirmations as confirmed, and that number is could be hardcoded, but the value after which you consider your transaction "confirmed enough" is up to you
1099 2011-07-26 12:45:01 <jtaylor> e.g. if you sell something for 1 dollar you are probably happy with one confirmation, if you transfer a million you might want to wait for 100
1100 2011-07-26 12:45:18 <Rabbit67890> Or 259
1101 2011-07-26 12:45:23 <prof7bit> so there is actually no additional number 120 hardcoded or used for anything in the code
1102 2011-07-26 12:49:23 <jtaylor> I'm not really familiar with why a block requires 120 confirmations, but there is probably some reason
1103 2011-07-26 12:49:25 <CIA-103> libbitcoin: genjix * r49a6870bdbc9 / (6 files in 3 dirs): Use better looking overloaded name string_repr for printing methods.
1104 2011-07-26 12:49:27 <hugolp> jtaylor: is there a limit in the protocol on how many confirmations I need before I can resend a bitcoin?
1105 2011-07-26 12:49:33 minimoose has joined
1106 2011-07-26 12:50:06 <prof7bit> i'm going to physically delete old spent transaction outputs (and entire transactions) from disk once they have been spent in an x blocks old block. In case of a reorg I would have to redownload not only the other fork but also all old blocks that are referenced in the now invalidated branch because some deleted tx might be still unspent in this new parallel reality. So I need to find some reasonable number x to make this not
1107 2011-07-26 12:50:06 <prof7bit> happen every other week.
1108 2011-07-26 12:52:02 <prof7bit> and the continued mentioning of the 120 number made me curious about whether something special might happen at that number.
1109 2011-07-26 12:52:04 <edcba> prof7bit: there is a 120 and 6 numbers hardcoded
1110 2011-07-26 12:52:23 <edcba> but actually they derive from the arbitrary chosen 10 minutes and 2 weeks from satoshi
1111 2011-07-26 12:52:27 <jtaylor> hugolp: not sure, but probably not
1112 2011-07-26 12:52:55 <forrestv> 20 hours and 1 hour?
1113 2011-07-26 12:53:38 Eremes has joined
1114 2011-07-26 12:53:40 theorbtwo is now known as si_washer
1115 2011-07-26 12:53:42 <jtaylor> but why does a block need 120 and not 6?
1116 2011-07-26 12:53:56 si_washer is now known as theorbtwo
1117 2011-07-26 12:54:33 <prof7bit> and more important what exactly is done with the hardcoded 120 number, is it actually used for anything in the protocol?
1118 2011-07-26 12:55:17 brooss has joined
1119 2011-07-26 12:55:21 <prof7bit> (i could grep through the code but I am lazy and hope someone knows it off the top of his head)
1120 2011-07-26 12:56:12 <epscy> i doubt it, as someone said earlier i think they are just arbitary numbers chosen by the client to measure the "confirmedness" of the transaction
1121 2011-07-26 12:56:21 <jtaylor> I'm no dev, but the numbers 6 and 120 should ahve nothing to do with the protocol
1122 2011-07-26 12:56:28 <forrestv> prof7bit, generation transactions can't be spent within 100 blocks
1123 2011-07-26 12:56:58 <forrestv> 120 is just the 'soft' limit for spending generation transactions - clients don't allow you to spend them before that
1124 2011-07-26 12:57:54 <prof7bit> ok, this explains why I have never actually seen my client doing anything special after reaching the 120, i never generated a block myself so far
1125 2011-07-26 12:58:32 <epscy> it wouldn't do anything special, apart from mark the transaction as confirmed
1126 2011-07-26 12:59:33 hugolp has quit (Quit: KVIrc 4.1.1 Equilibrium http://www.kvirc.net/)
1127 2011-07-26 12:59:50 <prof7bit> epscy: and since i don't have generation transactions all my coins didn't appear to look any different before and after 120
1128 2011-07-26 13:00:36 <prof7bit> "all my coins" == zero-point-zero-something
1129 2011-07-26 13:02:22 asuk has quit (Ping timeout: 252 seconds)
1130 2011-07-26 13:02:36 Clipse has quit (Read error: Connection reset by peer)
1131 2011-07-26 13:03:09 <prof7bit> are there any records / mentions from observed real world chain forks of notable size in larger parts of the network?
1132 2011-07-26 13:04:09 somuchwin2 has joined
1133 2011-07-26 13:05:07 somuchwin has quit (Ping timeout: 260 seconds)
1134 2011-07-26 13:05:33 Joric has quit (Ping timeout: 246 seconds)
1135 2011-07-26 13:05:48 <prof7bit> any such event that really was noticed by more clients or had some impact or was otherwise worth remembering?
1136 2011-07-26 13:06:43 ProgramMax has left ()
1137 2011-07-26 13:07:41 <edcba> prof7bit: a 'fork' happened when bitcoin had some overflow vuln
1138 2011-07-26 13:08:15 asuk has joined
1139 2011-07-26 13:08:16 <edcba> we had to patch the vuln and start a new chain not including an exploited transaction
1140 2011-07-26 13:10:30 asuk_ has joined
1141 2011-07-26 13:10:46 Joric has joined
1142 2011-07-26 13:10:46 Joric has quit (Changing host)
1143 2011-07-26 13:10:46 Joric has joined
1144 2011-07-26 13:11:08 MrTiggr has joined
1145 2011-07-26 13:11:45 <prof7bit> does the client put out any special log messages in case of a larger chain reorg? maybe it would be interesting to record this separately and somehow try to collect this data to view the network "from above" to observe how they propagate and how they resolve and how big they become
1146 2011-07-26 13:13:45 <sacarlson> I seem to have merge mining working on my weeds and mergemineTEST test chains useing multicoin-exp branch
1147 2011-07-26 13:13:54 gp5st has joined
1148 2011-07-26 13:14:09 ahihi2 has joined
1149 2011-07-26 13:17:25 MrTiggr has quit (Ping timeout: 252 seconds)
1150 2011-07-26 13:17:34 ewal-otg has joined
1151 2011-07-26 13:17:50 MrTiggr has joined
1152 2011-07-26 13:18:18 anarchyx has quit (Ping timeout: 240 seconds)
1153 2011-07-26 13:21:42 <prof7bit> also there is one thing i wonder about how it impacts the network as a whole: only newly started clients tend to have open connection slots available, after a while they are full. If I use the DNS seed only I am connected to 10 clients in less than a second but if I draw random addresses from my db then it takes *very* long
1154 2011-07-26 13:22:25 <prof7bit> even after only a few minute its already notably slow.
1155 2011-07-26 13:22:29 <cjdelisle> That's because those addresses in your db are really stale, people use dynamic ip or whatever.
1156 2011-07-26 13:22:33 copumpkin has quit (Ping timeout: 252 seconds)
1157 2011-07-26 13:22:55 ewal-w has quit ()
1158 2011-07-26 13:22:58 copumpkin has joined
1159 2011-07-26 13:23:05 <cjdelisle> All of these problems were already scientifically solved by the kadimilla people :|
1160 2011-07-26 13:23:10 <prof7bit> i wonder how this statistically affects the topology of the network
1161 2011-07-26 13:24:04 <prof7bit> and how it weakens it (also regarding the likelihood of net splits)
1162 2011-07-26 13:24:22 MrTiggr has quit (Ping timeout: 260 seconds)
1163 2011-07-26 13:24:26 <BlueMatt> the dnsseeds try their best to find every node which is up and pass that out
1164 2011-07-26 13:24:38 <BlueMatt> and rotates through its list of nodes
1165 2011-07-26 13:25:58 m00p has quit (Ping timeout: 264 seconds)
1166 2011-07-26 13:26:14 <BlueMatt> though current dnsseeds only return addresses of nodes that are running 0.3.24+, at this point if you are not running 0.3.24+, you are being a dick, hurting the network, and generally causing problems
1167 2011-07-26 13:26:16 Stellar has quit (Ping timeout: 246 seconds)
1168 2011-07-26 13:27:00 MrTiggr has joined
1169 2011-07-26 13:27:31 <prof7bit> i'm running my own client (and you could call it a "parasitic" services=0 client, similar to bitcoinj)
1170 2011-07-26 13:27:31 <cjdelisle> What's wrong with old clients?
1171 2011-07-26 13:28:26 <cjdelisle> meaning what do they break...
1172 2011-07-26 13:28:30 <BlueMatt> cjdelisle: network problems
1173 2011-07-26 13:28:36 molecular has joined
1174 2011-07-26 13:30:07 <prof7bit> but I think the network should be able to handle this and *not* become problems. And I don't want to be called a dick because I have a leeching only client. The network *must* be robust enough to stand a reasonable number of such nodes.
1175 2011-07-26 13:30:49 <BlueMatt> leeching client isnt the problem if bitcoin can handle it (ie if you have fClient set), running an old version is like running a leeching client without fClient
1176 2011-07-26 13:30:54 <prof7bit> after all it is outlined in the satoshi paper that such clients may exist
1177 2011-07-26 13:31:05 <BlueMatt> which could cause problems
1178 2011-07-26 13:31:10 TiggrBot has joined
1179 2011-07-26 13:31:10 <BlueMatt> s/could/will/
1180 2011-07-26 13:31:12 <prof7bit> what is fclient?
1181 2011-07-26 13:31:34 <BlueMatt> the flag you are supposed to pass if you are running a client instead of a full node
1182 2011-07-26 13:31:58 mu-b has quit (Ping timeout: 264 seconds)
1183 2011-07-26 13:31:59 <prof7bit> is it documented? the last time i looked there was only NODE_NETWORK and no other flag
1184 2011-07-26 13:32:07 <prof7bit> so i set all flags to 0
1185 2011-07-26 13:32:13 <BlueMatt> not really, read the source
1186 2011-07-26 13:32:30 <BlueMatt> fClient is calculated as NODE_NETWORK
1187 2011-07-26 13:32:38 <BlueMatt> so if you dont set that, you are setting fClient
1188 2011-07-26 13:32:42 <BlueMatt> two words for the same thing
1189 2011-07-26 13:33:39 <BlueMatt> anyway, Im off...if you have NODE_NETWORK set, dont use anything older than 0.3.24
1190 2011-07-26 13:33:44 <prof7bit> <BlueMatt> fClient is calculated as NODE_NETWORK <-- you mean *not* NODE_NETWORK? Or did I understand you wrong?
1191 2011-07-26 13:33:50 <BlueMatt> yea, sorry not
1192 2011-07-26 13:33:54 <prof7bit> ok
1193 2011-07-26 13:33:56 mu-b has joined
1194 2011-07-26 13:34:00 <BlueMatt> pfrom->fClient = !(pfrom->nServices & NODE_NETWORK);
1195 2011-07-26 13:34:05 <prof7bit> then I'm doing it correct
1196 2011-07-26 13:34:10 <BlueMatt> yes
1197 2011-07-26 13:34:17 <BlueMatt> sorry for the confusion
1198 2011-07-26 13:34:18 <BlueMatt> my bad
1199 2011-07-26 13:34:40 <prof7bit> bitcoinj also set this flag to 0
1200 2011-07-26 13:34:43 x6763 has quit (Ping timeout: 250 seconds)
1201 2011-07-26 13:34:45 WakiMiko has quit (Ping timeout: 250 seconds)
1202 2011-07-26 13:34:48 <BlueMatt> as it should
1203 2011-07-26 13:35:06 * BlueMatt -> gone
1204 2011-07-26 13:35:39 WakiMiko has joined
1205 2011-07-26 13:35:45 x6763 has joined
1206 2011-07-26 13:37:20 <prof7bit> are there notable differences in behavior already if it detects an fClient node? I'm not familiar with the codebase of the client. I am programming with the wiki documentation (and occasional look into bitcoinj sources ) only.
1207 2011-07-26 13:39:43 <prof7bit> s/the client/the reference client
1208 2011-07-26 13:39:46 Mr_Tiggr has joined
1209 2011-07-26 13:41:16 kluge has quit (Quit: ....)
1210 2011-07-26 13:41:52 MrTiggr has quit (Ping timeout: 260 seconds)
1211 2011-07-26 13:43:48 Mr_Tiggr is now known as MrTiggr
1212 2011-07-26 13:44:43 Cherothald has joined
1213 2011-07-26 13:45:42 <prof7bit> thats why I'm asking stupid questions here all the time because i need to know what it is supposed to do and not what it actually does.
1214 2011-07-26 13:45:46 larsivi has joined
1215 2011-07-26 13:46:55 kluge has joined
1216 2011-07-26 13:47:25 <upb> hmm, btw how does having multiple urls in bitcoind interfere with 'port forwardings' ?
1217 2011-07-26 13:48:18 b4epoche has quit (Quit: Computer has gone to sleep.)
1218 2011-07-26 13:48:19 b4epoche_ is now known as b4epoche
1219 2011-07-26 13:49:28 <erus`> does anyone elses github newsfeed just look like "bitcoin bitcoin bitcoin bitcoin bitcoin bitcoin bitcoin bitcoin bitcoin bitcoin bitcoin bitcoin bitcoin bitcoin bitcoin bitcoin bitcoin bitcoin "?
1220 2011-07-26 13:49:43 ewal-otg has quit (Quit: ewal-otg)
1221 2011-07-26 13:50:05 <Joric> oh good idea i should stop following
1222 2011-07-26 13:50:07 <prof7bit> bitcoin is the only project still using github?
1223 2011-07-26 13:50:35 <erus`> everyone loves github
1224 2011-07-26 13:50:46 <Joric> yes, bitcoin is the only project on github
1225 2011-07-26 13:50:56 <sacarlson> new updated merge mining results and documented method released http://forum.bitcoin.org/index.php?topic=24209.msg394289#msg394289 using MultiCoin-exp branch
1226 2011-07-26 13:50:56 <prof7bit> for suitable definitions of "everyone"
1227 2011-07-26 13:52:00 sacredchao has joined
1228 2011-07-26 13:52:03 <prof7bit> bitcoin and a zillion bitcoin forks with 2 lines difference
1229 2011-07-26 13:52:25 <da2ce7> ok... I'm tring to make bitcoin on my fedora box...
1230 2011-07-26 13:52:35 <da2ce7> however i'm getting /usr/bin/ld: cannot find -lboost_system
1231 2011-07-26 13:52:50 <da2ce7> and other ld errors...
1232 2011-07-26 13:52:56 <Joric> i'm like a week on github, has been forked twice already :)
1233 2011-07-26 13:53:00 <Joric> damn forkhub
1234 2011-07-26 13:53:11 <da2ce7> however I have boost-devel installed
1235 2011-07-26 13:54:25 <prof7bit> on ubuntu at least boost is split into a lot of packages
1236 2011-07-26 13:54:29 <doublec> sacarlson: good job!
1237 2011-07-26 13:54:56 <senseles> just download my centos 5.6 x64 binary
1238 2011-07-26 13:54:59 <senseles> it'll probably work for you
1239 2011-07-26 13:55:00 <prof7bit> maybe on fedora to
1240 2011-07-26 13:55:02 <senseles> search for it on the forums
1241 2011-07-26 13:55:04 <prof7bit> too
1242 2011-07-26 13:55:16 <sacarlson> doublec: thanks but it's all thanks to Vincent Durham who just released the needed fix
1243 2011-07-26 13:57:03 <upb> da2ce7: libboost-system1.42-dev: /usr/lib/libboost_system.a
1244 2011-07-26 13:57:03 <upb> libboost-system1.42-dev: /usr/lib/libboost_system.so
1245 2011-07-26 13:57:41 <upb> so install libboost-system-dev
1246 2011-07-26 14:00:35 <upb> hah, sipa has retracted his 'port forwarding' claim
1247 2011-07-26 14:03:09 Zagitta has joined
1248 2011-07-26 14:03:31 Stellar has joined
1249 2011-07-26 14:08:20 b4epoche_ has joined
1250 2011-07-26 14:08:30 marvinM has joined
1251 2011-07-26 14:08:36 zeropointo has quit (Quit: leaving)
1252 2011-07-26 14:08:44 <marvinM> ;;bc,stats
1253 2011-07-26 14:08:47 <gribble> Current Blocks: 138138 | Current Difficulty: 1690906.2047244 | Next Difficulty At Block: 139103 | Next Difficulty In: 965 blocks | Next Difficulty In About: 6 days, 4 hours, 30 minutes, and 10 seconds | Next Difficulty Estimate: 1822343.99876686
1254 2011-07-26 14:11:25 marvinM has quit (Client Quit)
1255 2011-07-26 14:11:58 koleg has joined
1256 2011-07-26 14:12:13 koleg has joined
1257 2011-07-26 14:15:27 kartmetal has quit (Ping timeout: 250 seconds)
1258 2011-07-26 14:17:05 erus`_ has joined
1259 2011-07-26 14:18:48 erus` has quit (Ping timeout: 260 seconds)
1260 2011-07-26 14:18:58 erus`_ is now known as erus`
1261 2011-07-26 14:22:07 johnlockwood_ has joined
1262 2011-07-26 14:22:56 johnlockwood_ has left ()
1263 2011-07-26 14:23:56 sacredchao has quit (Quit: KVIrc KVIrc Equilibrium 4.1.1, revision: 5829, sources date: 20110403, built on: 2011-05-07 15:04:44 UTC http://www.kvirc.net/)
1264 2011-07-26 14:25:28 <edcba> ;;bc,mtgox
1265 2011-07-26 14:25:29 <gribble> {"ticker":{"high":14.4304,"low":13.768,"avg":14.056022718,"vwap":14.039630236,"vol":21214,"last":13.9699,"buy":13.95103,"sell":13.9699}}
1266 2011-07-26 14:28:52 kartmetal has joined
1267 2011-07-26 14:29:36 copumpkin has quit (Quit: Computer has gone to sleep.)
1268 2011-07-26 14:30:31 Clipse has joined
1269 2011-07-26 14:30:58 p0s has joined
1270 2011-07-26 14:33:30 glassresistor has joined
1271 2011-07-26 14:33:30 glassresistor has quit (Changing host)
1272 2011-07-26 14:33:30 glassresistor has joined
1273 2011-07-26 14:35:05 Akinava is now known as Akinava|away
1274 2011-07-26 14:36:42 cjdelisle has quit (Ping timeout: 260 seconds)
1275 2011-07-26 14:36:43 <BlueMatt> prof7bit: yes, the current bitcoin client does react to the NODE_NETWORK flag
1276 2011-07-26 14:37:11 Joric has quit ()
1277 2011-07-26 14:37:55 <erus`> prof7bit is a non stop rubbish troll
1278 2011-07-26 14:37:58 vigilyn has quit (Quit: Leaving)
1279 2011-07-26 14:38:06 <erus`> its just boring
1280 2011-07-26 14:38:14 cjdelisle has joined
1281 2011-07-26 14:38:31 vigilyn has joined
1282 2011-07-26 14:41:02 Rabbit67890 has quit (Quit: Colloquy for iPad - http://colloquy.mobi)
1283 2011-07-26 14:43:28 <prof7bit> erus`: you are borig too
1284 2011-07-26 14:43:56 <prof7bit> nobody asked you about your opinion
1285 2011-07-26 14:45:09 asuk has quit (Quit: leaving)
1286 2011-07-26 14:45:09 asuk_ has quit (Quit: leaving)
1287 2011-07-26 14:45:37 asuk has joined
1288 2011-07-26 14:45:53 cjdelisle has quit (Ping timeout: 258 seconds)
1289 2011-07-26 14:45:58 <prof7bit> you just pop up here from time to time and insult other people. i haven't seen you contributing anything yet,
1290 2011-07-26 14:47:15 cjdelisle has joined
1291 2011-07-26 14:50:29 Clipse-b has joined
1292 2011-07-26 14:51:39 Clipse has quit (Ping timeout: 255 seconds)
1293 2011-07-26 14:53:33 minimoose has quit (Remote host closed the connection)
1294 2011-07-26 14:54:11 minimoose has joined
1295 2011-07-26 14:54:38 <b4epoche> erus`: leave him alone, he's 1 bit shy of a full byte
1296 2011-07-26 14:56:48 <BlueMatt> ;;seen gavinandresen
1297 2011-07-26 14:56:48 <gribble> gavinandresen was last seen in #bitcoin-dev 1 week, 1 day, 9 hours, 22 minutes, and 31 seconds ago: <gavinandresen> gmaxwell: very nice!
1298 2011-07-26 14:56:52 <BlueMatt> damn
1299 2011-07-26 14:57:32 <TuxBlackEdo> ;;seen gmaxwell
1300 2011-07-26 14:57:33 <gribble> gmaxwell was last seen in #bitcoin-dev 20 hours, 5 minutes, and 19 seconds ago: <gmaxwell> it might be an issue on some kind of supersnazzy asic miner where saving some communications cost for H=0 is worth it... but you'd want to do constant operations there of course.
1301 2011-07-26 14:58:17 DD- has quit ()
1302 2011-07-26 14:59:03 copumpkin has joined
1303 2011-07-26 14:59:09 Clipse has joined
1304 2011-07-26 14:59:23 cjdelisle has quit (Ping timeout: 240 seconds)
1305 2011-07-26 14:59:33 <epscy> ;;bc,stats
1306 2011-07-26 14:59:36 <gribble> Current Blocks: 138144 | Current Difficulty: 1690906.2047244 | Next Difficulty At Block: 139103 | Next Difficulty In: 959 blocks | Next Difficulty In About: 6 days, 3 hours, 2 minutes, and 48 seconds | Next Difficulty Estimate: 1822531.28413221
1307 2011-07-26 14:59:39 pierce has quit (Ping timeout: 250 seconds)
1308 2011-07-26 14:59:49 <epscy> did the difficulty drop last time?
1309 2011-07-26 15:00:02 <mtrlt> nope
1310 2011-07-26 15:00:42 <epscy> interesting
1311 2011-07-26 15:02:40 <upb> http://siyobik.info/main/pastebin/view/0/48
1312 2011-07-26 15:02:54 Clipse-b has quit (Ping timeout: 255 seconds)
1313 2011-07-26 15:02:58 <upb> ^- what do you think, could tuxes 'own dao' look anything like this crap?
1314 2011-07-26 15:06:23 WakiMiko_ has joined
1315 2011-07-26 15:07:02 larsivi has quit (Ping timeout: 252 seconds)
1316 2011-07-26 15:07:26 ewal-otg has joined
1317 2011-07-26 15:07:43 <Optimo> there's a bitcoin trader app on cydia ;p
1318 2011-07-26 15:08:28 <Optimo> it's for using your tradehill account
1319 2011-07-26 15:09:19 WakiMiko has quit (Ping timeout: 240 seconds)
1320 2011-07-26 15:09:58 cjdelisle has joined
1321 2011-07-26 15:09:59 <b4epoche> by whom?
1322 2011-07-26 15:10:38 <Optimo> Tyler (?)
1323 2011-07-26 15:10:44 <Optimo> I posted it
1324 2011-07-26 15:11:02 <Optimo> teeman seems to be his handle
1325 2011-07-26 15:11:04 <b4epoche> yea, figured. Just curious who wrote it, and way TH?
1326 2011-07-26 15:11:08 <Optimo> it's probably mentioned on teh forums
1327 2011-07-26 15:11:31 <Optimo> hopefully this will break the ice ;)
1328 2011-07-26 15:12:04 <b4epoche> do you know ifdid he try appstore?
1329 2011-07-26 15:12:20 <Optimo> doesn't seem like it
1330 2011-07-26 15:12:21 <vegard> nice, I know that guy :D
1331 2011-07-26 15:12:58 <vegard> he's tmandry here on freenode
1332 2011-07-26 15:13:07 <Optimo> vegard, as I'ev mentioned before I maintain thebigboss repo - came here orginally to recruit apps that apple rejects
1333 2011-07-26 15:13:10 <Optimo> ah cool
1334 2011-07-26 15:13:44 <Optimo> open sauced https://github.com/teeman/BitcoinTrader
1335 2011-07-26 15:14:29 <b4epoche> damn auto-correct
1336 2011-07-26 15:15:32 <b4epoche> Optimo: well got web sockets working in BitcoinTrader… it's coming along nicely
1337 2011-07-26 15:15:59 minimoose has quit (Ping timeout: 240 seconds)
1338 2011-07-26 15:16:02 <b4epoche> now I gotta figure out what to do with all the data
1339 2011-07-26 15:16:47 <Optimo> I gota update switchermod today
1340 2011-07-26 15:17:02 <Optimo> apaprently it breaks multitouch swipe gestures on ipad
1341 2011-07-26 15:20:32 <Optimo> lulz GeckoOS
1342 2011-07-26 15:22:04 SISUbtcX has joined
1343 2011-07-26 15:36:31 pierce has joined
1344 2011-07-26 15:37:15 ar4s has joined
1345 2011-07-26 15:39:38 koleg has quit (2!kvirc@79.133.132.158|Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/)
1346 2011-07-26 15:41:07 minimoose has joined
1347 2011-07-26 15:42:43 gasjews has quit (K-Lined)
1348 2011-07-26 15:47:19 nhodges has quit (Ping timeout: 250 seconds)
1349 2011-07-26 15:49:26 karnac has joined
1350 2011-07-26 15:51:05 <CIA-103> bitcoin: Pieter Wuille master * ra0eb1d1 / src/base58.h :
1351 2011-07-26 15:51:05 <CIA-103> bitcoin: Merge pull request #434 from laanwj/addkey_retval
1352 2011-07-26 15:51:05 <CIA-103> bitcoin: make SetHash160 return a value (as specified in the function signature) - https://github.com/bitcoin/bitcoin/commit/a0eb1d151f60ac49ce30042c7ae4039cf538992d
1353 2011-07-26 15:55:22 Prof_BiG_BanG has quit (Ping timeout: 240 seconds)
1354 2011-07-26 15:55:30 Guest39969 is now known as jimon
1355 2011-07-26 16:00:05 devon_hillard has joined
1356 2011-07-26 16:00:12 erus` has quit (Remote host closed the connection)
1357 2011-07-26 16:02:46 Titeuf_87 has joined
1358 2011-07-26 16:04:57 ewal-otg has quit (Quit: ewal-otg)
1359 2011-07-26 16:04:58 pierce has quit (Disconnected by services)
1360 2011-07-26 16:06:11 pierce1 has joined
1361 2011-07-26 16:10:04 ar4s has quit (Quit: ar4s)
1362 2011-07-26 16:11:51 altamic has joined
1363 2011-07-26 16:12:06 FractalUniverse has joined
1364 2011-07-26 16:14:56 pierce has joined
1365 2011-07-26 16:15:51 blueadept has joined
1366 2011-07-26 16:17:21 Sickness\ is now known as Rebecca_Black
1367 2011-07-26 16:17:36 Rebecca_Black is now known as RebeccaBlack
1368 2011-07-26 16:18:39 RebeccaBlack is now known as Acceber
1369 2011-07-26 16:19:10 nhodges has joined
1370 2011-07-26 16:19:57 altamic has quit (Ping timeout: 276 seconds)
1371 2011-07-26 16:21:08 cjdelisle has quit (Ping timeout: 260 seconds)
1372 2011-07-26 16:21:17 altamic has joined
1373 2011-07-26 16:23:01 cjdelisle has joined
1374 2011-07-26 16:23:44 erus` has joined
1375 2011-07-26 16:24:22 gp5st has quit (Ping timeout: 255 seconds)
1376 2011-07-26 16:28:09 larsivi has joined
1377 2011-07-26 16:28:59 jrmithdobbs has quit (Read error: Operation timed out)
1378 2011-07-26 16:30:41 jrmithdobbs has joined
1379 2011-07-26 16:32:03 thefinn93 has joined
1380 2011-07-26 16:35:09 QueryTom3000 has joined
1381 2011-07-26 16:36:51 estornudo has joined
1382 2011-07-26 16:38:28 <CIA-103> libbitcoin: genjix * r6b0d9f589428 / (2 files in 2 dirs): when organising the blockchain, match up outputs with inputs.
1383 2011-07-26 16:39:07 Clipse has quit (Read error: Connection reset by peer)
1384 2011-07-26 16:39:33 gp5st1 has joined
1385 2011-07-26 16:39:40 Incitatus has quit (Ping timeout: 255 seconds)
1386 2011-07-26 16:39:44 wardearia has quit (Ping timeout: 258 seconds)
1387 2011-07-26 16:48:12 d1g1t4l has quit (Remote host closed the connection)
1388 2011-07-26 16:49:51 larsivi has quit (Ping timeout: 276 seconds)
1389 2011-07-26 16:51:19 gp5st1 has quit (Quit: Leaving.)
1390 2011-07-26 16:51:58 gp5st1 has joined
1391 2011-07-26 16:52:38 Incitatus has joined
1392 2011-07-26 16:55:50 x6763 has quit (Remote host closed the connection)
1393 2011-07-26 16:56:13 zyb has quit (Quit: leaving)
1394 2011-07-26 16:56:30 x6763 has joined
1395 2011-07-26 16:56:30 wardearia has joined
1396 2011-07-26 16:59:01 ujjain has quit (Ping timeout: 255 seconds)
1397 2011-07-26 17:02:10 ujjain has joined
1398 2011-07-26 17:02:27 shLONG has quit ()
1399 2011-07-26 17:05:24 altamic has quit (Ping timeout: 240 seconds)
1400 2011-07-26 17:08:50 Clipse has joined
1401 2011-07-26 17:11:40 zyb has joined
1402 2011-07-26 17:12:02 <CIA-103> bitcoin: Wladimir J. van der Laan master * r690c723 / src/base58.h : make SetHash160 return a value (as specified in the function signature) - https://github.com/bitcoin/bitcoin/commit/690c723e5fef4de77a2f1cbc51c9b3166c509520
1403 2011-07-26 17:13:41 rethaw has joined
1404 2011-07-26 17:16:26 altamic has joined
1405 2011-07-26 17:16:26 altamic has quit (Changing host)
1406 2011-07-26 17:16:26 altamic has joined
1407 2011-07-26 17:18:36 tomat has quit (Read error: Connection reset by peer)
1408 2011-07-26 17:20:06 sytse has quit (Read error: Connection reset by peer)
1409 2011-07-26 17:20:10 sytse has joined
1410 2011-07-26 17:21:43 larsig_ has joined
1411 2011-07-26 17:21:58 <larsig_> o.0
1412 2011-07-26 17:25:27 ujjain has quit (Quit: /dev/null)
1413 2011-07-26 17:26:51 SecretSJ has joined
1414 2011-07-26 17:26:56 redhatzero has quit (Quit: Leaving.)
1415 2011-07-26 17:28:48 MrTiggr has quit (Ping timeout: 260 seconds)
1416 2011-07-26 17:29:01 MrTiggr has joined
1417 2011-07-26 17:30:58 NickelBot has quit (Ping timeout: 255 seconds)
1418 2011-07-26 17:31:29 knotwork has quit (Ping timeout: 258 seconds)
1419 2011-07-26 17:32:45 zamgo has joined
1420 2011-07-26 17:32:56 SISUbtcX has quit (Quit: SISUbtcX)
1421 2011-07-26 17:35:29 tomat has joined
1422 2011-07-26 17:37:46 rethaw has quit (Ping timeout: 240 seconds)
1423 2011-07-26 17:38:21 slothbear has joined
1424 2011-07-26 17:38:37 slothbear has left ()
1425 2011-07-26 17:38:54 knotwork has joined
1426 2011-07-26 17:39:39 thefinn93 has quit (Quit: BAI)
1427 2011-07-26 17:40:01 thefinn93 has joined
1428 2011-07-26 17:42:11 mosimo has joined
1429 2011-07-26 17:46:50 NickelBot has joined
1430 2011-07-26 17:47:16 <CIA-103> bitcoin: Pieter Wuille master * rb63241d / (src/script.cpp src/ui.cpp src/wallet.cpp):
1431 2011-07-26 17:47:16 <CIA-103> bitcoin: Bugfix: don't overuse limited ExtractAddress
1432 2011-07-26 17:47:16 <CIA-103> bitcoin: ExtractAddress was called with the keystore as argument in RPC and
1433 2011-07-26 17:47:16 <CIA-103> bitcoin: UI, limiting results to own keys. This caused empty "address"
1434 2011-07-26 17:47:16 <CIA-103> bitcoin: fields. - https://github.com/bitcoin/bitcoin/commit/b63241d4511896fcd65996ac7d9a5cb935118ca3
1435 2011-07-26 17:47:17 <CIA-103> bitcoin: Pieter Wuille master * raa7c5e7 / (src/script.cpp src/ui.cpp src/wallet.cpp):
1436 2011-07-26 17:47:17 <CIA-103> bitcoin: Merge pull request #435 from sipa/fixextract
1437 2011-07-26 17:47:18 <CIA-103> bitcoin: Bugfix: don't overuse limited ExtractAddress - https://github.com/bitcoin/bitcoin/commit/aa7c5e7ceb87d0309349b89c1ad5e388cf6828a4
1438 2011-07-26 17:53:58 Acceber is now known as Sickness\
1439 2011-07-26 17:54:09 BTCTrader has quit (Quit: BTCTrader)
1440 2011-07-26 17:55:15 sacredchao has joined
1441 2011-07-26 17:57:39 * jgarzik performs his standard morning exercises: stretch, walk dogs, email corrections to reporters who wrote incorrect bitcoin stories
1442 2011-07-26 18:00:13 <imsaguy2> lol
1443 2011-07-26 18:00:36 <imsaguy2> no S^3?
1444 2011-07-26 18:00:44 erus` has quit (Read error: Connection reset by peer)
1445 2011-07-26 18:01:00 molecular has quit (Ping timeout: 240 seconds)
1446 2011-07-26 18:01:18 erus` has joined
1447 2011-07-26 18:01:42 molecular has joined
1448 2011-07-26 18:04:22 bitcoinbulletin has quit (Remote host closed the connection)
1449 2011-07-26 18:07:17 MrTiggr has quit (Read error: Connection reset by peer)
1450 2011-07-26 18:07:40 MrTiggr has joined
1451 2011-07-26 18:14:24 sipa has joined
1452 2011-07-26 18:14:34 karnac has quit (Quit: karnac)
1453 2011-07-26 18:14:38 bitcoinbulletin has joined
1454 2011-07-26 18:16:41 <sipa> gmaxwell: having addresses in your wallet without their private key is very useful, but it requires some other changes before it's useful (wallet disk format, signing with non-available private key causes export/dump to sign offline, ...)
1455 2011-07-26 18:17:06 <sipa> wumpus: seems i noticed the "To:" issue at the same time you did :)
1456 2011-07-26 18:17:46 Joric has joined
1457 2011-07-26 18:18:11 <wumpus> sipa: I wonder what the rationale was to make it this way
1458 2011-07-26 18:18:23 phuzion has quit (Ping timeout: 260 seconds)
1459 2011-07-26 18:18:25 <gmaxwell> sipa: what inspired that was someone drafting out a spec api for the required export / offline sign.
1460 2011-07-26 18:19:37 <sipa> wumpus: originally there were two function, one for extracting pubkeys, the other for extracting addresses - i merged them into a single function (when getting rid of mapPubKeys) that either took a keystore or not, and was a bit careless when invoking that function
1461 2011-07-26 18:20:09 <sipa> strange that i didn't notice, i did transactions when testing the patch
1462 2011-07-26 18:20:11 <wumpus> sipa: ah, right, I thought it had something to do with a new kind of transaction that required a different kind of key extraction
1463 2011-07-26 18:20:15 thefinn93 has quit (Quit: BAI)
1464 2011-07-26 18:21:20 <wumpus> but indeed I noticed all the key and address functions changed around
1465 2011-07-26 18:22:05 <sipa> wumpus: the reason for the keystore being passed to extractadress is for the future possibility where extractaddress returns multiple possible addresses for a single output
1466 2011-07-26 18:22:23 <sipa> wumpus: in that case, you sometimes want to be sure it's one of your own keys that is returned
1467 2011-07-26 18:22:26 <wumpus> sipa: if you don't pass the keystore in it segfaults :-)
1468 2011-07-26 18:22:35 <sipa> not anymore
1469 2011-07-26 18:22:39 TD has joined
1470 2011-07-26 18:22:56 <wumpus> ok, so for txouts we should not pass a keystore anymore?
1471 2011-07-26 18:23:55 <sipa> if you want to test whether a txout is to you, you should
1472 2011-07-26 18:24:09 <sipa> if you just want to request to whom the payment is, no
1473 2011-07-26 18:24:25 <sipa> and the first case is covered by IsMine() and co
1474 2011-07-26 18:24:33 <wumpus> I mean if you just want to see who the transaction goes to
1475 2011-07-26 18:24:36 <wumpus> ok thanks
1476 2011-07-26 18:24:59 <wumpus> this is pretty confusing, a transaction can have multiple outs and one txout could point to multiple addresses :-)
1477 2011-07-26 18:25:34 <sipa> maybe i should do a commit with a few lines of documentation for each function i've written or modified the past few months...
1478 2011-07-26 18:26:09 <wumpus> hehe, you could probably copy/paste a lot of documentation from irc logs :)
1479 2011-07-26 18:26:11 <jgarzik> sipa: does https://github.com/bitcoin/bitcoin/pull/220 remain fresh?
1480 2011-07-26 18:26:43 <sipa> jgarzik: it should still work
1481 2011-07-26 18:26:58 <sipa> though key removal needs some extra work still
1482 2011-07-26 18:27:29 TD_ has joined
1483 2011-07-26 18:29:01 TD has quit (Ping timeout: 255 seconds)
1484 2011-07-26 18:29:01 TD_ is now known as TD
1485 2011-07-26 18:30:12 phuzion has joined
1486 2011-07-26 18:36:57 minimoose has quit (Quit: minimoose)
1487 2011-07-26 18:38:02 BTCTrader has joined
1488 2011-07-26 18:38:27 BTCTrader is now known as Guest45980
1489 2011-07-26 18:39:59 Sedra has quit (Ping timeout: 258 seconds)
1490 2011-07-26 18:40:06 Marf has joined
1491 2011-07-26 18:42:17 Sedra has joined
1492 2011-07-26 18:44:00 Joric has quit ()
1493 2011-07-26 18:50:19 phuzion has quit (Ping timeout: 246 seconds)
1494 2011-07-26 18:51:59 jbalint has joined
1495 2011-07-26 18:52:46 klikklak has quit (Ping timeout: 258 seconds)
1496 2011-07-26 18:53:58 sacredchao has left ("No matter how dark the night, somehow the Sun rises once again")
1497 2011-07-26 18:54:07 phuzion has joined
1498 2011-07-26 18:54:21 TD has quit (Quit: TD)
1499 2011-07-26 18:54:25 superman2011 has quit ()
1500 2011-07-26 18:54:38 superman2016 has quit (Read error: Connection reset by peer)
1501 2011-07-26 18:56:17 osmosis has joined
1502 2011-07-26 18:56:18 superman2016 has joined
1503 2011-07-26 19:02:07 RazielZ has quit (Quit: Leaving)
1504 2011-07-26 19:05:38 devrandom has quit (Remote host closed the connection)
1505 2011-07-26 19:07:28 TheAncientGoat has quit (Ping timeout: 276 seconds)
1506 2011-07-26 19:12:07 <diki> when searching for transactions we hash the merkle root, which are basically transactions, right?
1507 2011-07-26 19:12:14 <diki> s/transactions 1/block
1508 2011-07-26 19:12:26 nhodges has quit (Remote host closed the connection)
1509 2011-07-26 19:13:28 Sedra has quit (Quit: ( IRC :: Quit ))
1510 2011-07-26 19:14:03 <diki> Also, merkle sounds stupid. Merklee sounds better
1511 2011-07-26 19:16:24 nhodges has joined
1512 2011-07-26 19:16:41 enquirer has quit (Quit: back soon)
1513 2011-07-26 19:17:14 <prof7bit> the merkle tree is used to prove that a transaction is part of this block. it consists of hashes of transactions
1514 2011-07-26 19:17:35 Marf has quit (Read error: Connection reset by peer)
1515 2011-07-26 19:17:40 cjdelisle has quit (Ping timeout: 240 seconds)
1516 2011-07-26 19:17:45 <jbalint> so what is a merkle branch?
1517 2011-07-26 19:17:54 <prof7bit> a part of the tree
1518 2011-07-26 19:18:20 dburns has joined
1519 2011-07-26 19:18:21 <dburns> anyone interested in 15 btc at 12ppusd each
1520 2011-07-26 19:19:24 <Namegduf> dburns: I'll give you 10 PPUSD each.
1521 2011-07-26 19:19:44 Sedra has joined
1522 2011-07-26 19:19:53 <prof7bit> you can leave away parts of the tree while still being able to verify the hashes from one particular leaf all the way down to the root.
1523 2011-07-26 19:19:56 <Namegduf> Changed my mind. 5.
1524 2011-07-26 19:20:26 <jbalint> but the complete tree refers to tx/blocks that are not part of the current block, right?
1525 2011-07-26 19:20:39 <CIA-103> bitcoin: Jeff Garzik master * ra139ed7 / src/db.cpp : CAddrDB::LoadAddresses: properly initialize CAddress ... https://github.com/bitcoin/bitcoin/commit/a139ed74f2a5764afbd678c0e4eb6143455a3a49
1526 2011-07-26 19:21:01 <prof7bit> no, each block has its tree.
1527 2011-07-26 19:21:20 <jbalint> ok... im not clear on this then. imma keep readin the code
1528 2011-07-26 19:22:13 <prof7bit> and all tx that are in this block have their place in this tree
1529 2011-07-26 19:22:26 <prof7bit> their hash
1530 2011-07-26 19:22:39 JRWR has joined
1531 2011-07-26 19:22:39 JRWR has quit (Changing host)
1532 2011-07-26 19:22:39 JRWR has joined
1533 2011-07-26 19:23:15 TD has joined
1534 2011-07-26 19:24:23 <jbalint> so to verify a tx, you have to calculate the whole tree for a block? otherwise, how do you know the root hash is correct?
1535 2011-07-26 19:25:01 <sipa> you need the path of hashes from that tx to the root
1536 2011-07-26 19:25:08 <prof7bit> you need only calculate the hashes that lead from this particular place down to the root
1537 2011-07-26 19:25:13 altamic has quit (Ping timeout: 276 seconds)
1538 2011-07-26 19:25:24 <diki> isnt it time for a forum upgrade?
1539 2011-07-26 19:25:30 <diki> i dont like the forum's pager function
1540 2011-07-26 19:25:33 <diki> and smf sucks in general
1541 2011-07-26 19:25:34 rethaw has joined
1542 2011-07-26 19:26:09 <TD> AFAIK the forum is basically running itself. the only people with admin access don't want to change anything about it
1543 2011-07-26 19:26:27 <jbalint> but at the root level, multiple branches are included in the hash, and to have the hash of those branches, you have to hash all the way up from the leaves of the tree
1544 2011-07-26 19:26:30 <diki> the only people with admin access don't want to change anything about it<-but i do
1545 2011-07-26 19:26:55 <diki> Now, so if there are no transactions...
1546 2011-07-26 19:27:01 <diki> how does one calculate a block?
1547 2011-07-26 19:27:14 <sipa> jbalint: in each node of the tree, there are two children - on of these children you know, the other must be given
1548 2011-07-26 19:27:29 <sipa> diki: there is always at least a generation tx
1549 2011-07-26 19:27:47 GMP has quit (Remote host closed the connection)
1550 2011-07-26 19:28:34 <diki> but....genesis block
1551 2011-07-26 19:28:37 <diki> no tx before it
1552 2011-07-26 19:28:52 <sipa> ?
1553 2011-07-26 19:28:53 <diki> so what were they hashing if there were no transactions?
1554 2011-07-26 19:29:05 <prof7bit> if you cut away an entire branch from the tree you need to leave the root of that branch intact. then you can still calculate all hashes from the remaining partt of the tree,
1555 2011-07-26 19:29:06 <sipa> the genesis block has a generation tx like every other block
1556 2011-07-26 19:29:23 <diki> but in order to find it you need to have a merkle root, no?
1557 2011-07-26 19:29:29 <diki> which are basically the transactions?
1558 2011-07-26 19:29:39 <jbalint> prof7bit: but when you get a block, you dont have anything except the root hash, right?
1559 2011-07-26 19:29:39 <sipa> if there is only one transaction, the hash of that transaction is the root
1560 2011-07-26 19:29:53 <diki> and if there is NO transaction?
1561 2011-07-26 19:30:01 <prof7bit> when you dwnload a block you get the tree and all tx
1562 2011-07-26 19:30:02 <diki> before the genesis block, no one had any bitcoins
1563 2011-07-26 19:30:05 <sipa> diki: as i told you, each block has a generation tx
1564 2011-07-26 19:30:06 <diki> thus no transactions
1565 2011-07-26 19:30:14 <sipa> otherwise it is not valid
1566 2011-07-26 19:30:15 <jbalint> prof7bit: thanks.
1567 2011-07-26 19:30:23 <diki> sipa, but before generating the genesis block there were no transactions, right?
1568 2011-07-26 19:30:34 <sipa> is that relevant?
1569 2011-07-26 19:30:41 <sipa> the genesis block has a generation tx
1570 2011-07-26 19:30:50 <sipa> so it has 1 tx
1571 2011-07-26 19:30:55 d1g1t4l has joined
1572 2011-07-26 19:30:59 <sipa> which is enough
1573 2011-07-26 19:31:05 <diki> so you hash a tx which doesnt even exist yet?
1574 2011-07-26 19:31:09 <prof7bit> you can prune the block by carefully pruning the tree while removng tx from it ans still be able to follow the hashes of the remainig tx down to the root
1575 2011-07-26 19:31:18 <sipa> diki: the genesis block defines that tx
1576 2011-07-26 19:32:18 <jbalint> prof7bit: when is this done? after the money in that tx is spent?
1577 2011-07-26 19:32:31 <prof7bit> the official client does not do this
1578 2011-07-26 19:33:10 <diki> is it possible to change the client to accept transactions which have a fee and ignore the fee-less ones?
1579 2011-07-26 19:33:15 <prof7bit> but the simplified verification method needs a simple way to prove that one particular tx belongs to a certain block
1580 2011-07-26 19:33:23 <diki> for if you find a block i mean
1581 2011-07-26 19:34:16 <jbalint> in this page https://en.bitcoin.it/wiki/Protocol_specification#block , it says only the merkle root hash is sent with the block. is the whole tree included?
1582 2011-07-26 19:34:25 <prof7bit> so for this you need the block header with merkle root and only the branch leading up to this one transaction, you dont need all other tx from this block and also not all other branches of the tree
1583 2011-07-26 19:35:08 <diki> if there are a lot of transactions in the network waiting to be included in a block(and i mean a lot) does that affect in any way block generation time?
1584 2011-07-26 19:35:11 <prof7bit> with the block header
1585 2011-07-26 19:37:07 sabalaba has joined
1586 2011-07-26 19:40:45 <prof7bit> i can't find it now too while quickly scanning through this wiki page it must be somewhere there, maybe analyze the code.
1587 2011-07-26 19:41:06 <jbalint> yes, thats what i will do
1588 2011-07-26 19:41:12 karnac has joined
1589 2011-07-26 19:41:21 macdada has joined
1590 2011-07-26 19:41:25 SISUbtc has joined
1591 2011-07-26 19:41:40 <prof7bit> block message is the next thing I implement, so I will find it soon.
1592 2011-07-26 19:42:06 thefinn93 has joined
1593 2011-07-26 19:42:25 Guest45980 is now known as BTCTrader
1594 2011-07-26 19:42:28 <prof7bit> bitcoinj sources might also help as a separate implementation written in ava
1595 2011-07-26 19:42:32 <prof7bit> +j
1596 2011-07-26 19:43:26 <jbalint> is this in development?
1597 2011-07-26 19:44:37 <prof7bit> its a separate project, you find it on googlecode. I haven't tried it yet but looked into the sources a few times already to see how they implemented some things
1598 2011-07-26 19:44:57 <sipa> diki: the number of transactions in the memory pool does not influence the speed of block mining
1599 2011-07-26 19:45:12 <sipa> as it is just one precalculated number that needs to be known, the merkle root
1600 2011-07-26 19:45:28 irssi has joined
1601 2011-07-26 19:46:45 irssi is now known as cjdelisle
1602 2011-07-26 19:46:49 koleg has joined
1603 2011-07-26 19:49:27 koleg has quit (Read error: Connection reset by peer)
1604 2011-07-26 19:49:28 koleg has joined
1605 2011-07-26 19:49:46 koleg has quit (2!~koleg@89.151.191.93|Read error: Connection reset by peer)
1606 2011-07-26 19:50:24 zeropointo has joined
1607 2011-07-26 19:50:34 joepie91 has joined
1608 2011-07-26 19:52:28 thefinn93 has quit (Quit: BAI)
1609 2011-07-26 19:54:58 osmosis has quit (Ping timeout: 240 seconds)
1610 2011-07-26 19:55:25 lyspooner has joined
1611 2011-07-26 19:58:04 thefinn93 has joined
1612 2011-07-26 19:58:55 <prof7bit> jbalint: http://code.google.com/p/bitcoinj/source/browse/trunk/src/com/google/bitcoin/core/Block.java#283
1613 2011-07-26 19:58:55 dburns has quit (Ping timeout: 252 seconds)
1614 2011-07-26 19:59:12 <jbalint> thanks prof7bit . looking
1615 2011-07-26 19:59:47 TheZimm has joined
1616 2011-07-26 20:00:43 <prof7bit> you can calculate it on the fly if you have all blocks. Then you need to remember only the root (and the rule in which exact order to sort them into the tree). only when you start removing tx from the block you need to remember enough of the hashes to still be able to calculate the remaining half of the tree.
1617 2011-07-26 20:00:55 osmosis has joined
1618 2011-07-26 20:00:56 <prof7bit> s/all blocks/all tx
1619 2011-07-26 20:01:17 nameless has joined
1620 2011-07-26 20:01:27 SISUbtc has quit (Quit: Colloquy for iPad - http://colloquy.mobi)
1621 2011-07-26 20:01:43 <sipa> BlueMatt: locking problem
1622 2011-07-26 20:02:00 <sipa> BlueMatt: there is no sane way of only locking cs_vMasterKey without cs_KeyStore
1623 2011-07-26 20:02:08 <jbalint> ok, but in the running program, lets say you get a new tx, you have to look up the block this is tx from. are you expected to have calculated (or have been given) the tree? (or enough of it?)
1624 2011-07-26 20:02:35 <TD> in the current protocol transactions do not come with branches
1625 2011-07-26 20:03:11 <sipa> jbalint: the program just keeps a database of all transactions and the blocks they are in
1626 2011-07-26 20:03:13 zamgo has left ()
1627 2011-07-26 20:03:21 <prof7bit> the lookup of the tx in the local storage is something different, there you would index the DB by the hashes or the inputs or outputs
1628 2011-07-26 20:03:29 larsivi has joined
1629 2011-07-26 20:03:30 <jbalint> ok, this is what i'm getting at. to verify that the tx was included in a previous block, you have to calculate the tree for that block
1630 2011-07-26 20:04:01 <sipa> no, just check your database of transactions
1631 2011-07-26 20:04:25 <sipa> well, an alternative would indeed be to know its merkle path, and look op the merkle root in your block database
1632 2011-07-26 20:04:44 <sipa> BlueMatt: as most places where cs_vMasterKey is locked, cannot be prevented from doing calls that require cs_KeyStore afterwards
1633 2011-07-26 20:04:44 MrTiggr has quit (Ping timeout: 260 seconds)
1634 2011-07-26 20:05:27 <sipa> BlueMatt: so we need to order those ("never lock cs_vMasterKey without having locked cs_KeyStore")
1635 2011-07-26 20:05:27 MrTiggr has joined
1636 2011-07-26 20:05:32 Happy0 has quit ()
1637 2011-07-26 20:05:51 <sipa> BlueMatt: but that would mean that change passphrase dialogs in the GUI block the entire keystore
1638 2011-07-26 20:05:54 imsaguy2 has quit (Ping timeout: 260 seconds)
1639 2011-07-26 20:08:03 <sipa> BlueMatt: so i think it's preferable to not do locking from the GUI for those cases, and live with the terrible race condition that one types the correct old passphrase, it's changed in between (through rpc), and that soon afterwards the changing of the passphrase fails
1640 2011-07-26 20:08:20 <prof7bit> to verify that a tx was included in a certain block (and if you don't have a full copy of that block) you need at least the block header with the merkle root and enough hashes to build the merkle branch leading from this root up to this tx. if thsi verifies than the tx does indeed belong to that block.
1641 2011-07-26 20:08:39 thefinn93 has quit (Remote host closed the connection)
1642 2011-07-26 20:08:54 <jbalint> where are the 'enough hashes' coming from?
1643 2011-07-26 20:09:05 <sipa> currently, nowhere
1644 2011-07-26 20:09:12 thefinn93 has joined
1645 2011-07-26 20:09:17 <jbalint> so you have to calc the whole tree for that block
1646 2011-07-26 20:09:25 <prof7bit> and if you can prove that it was part of a block then you have proven that the miners have found it worthy to be included in a block.
1647 2011-07-26 20:09:29 <sipa> jbalint: what is your use case?
1648 2011-07-26 20:10:06 <jbalint> uhm
1649 2011-07-26 20:10:11 <jrmithdobbs> s/miners/validators/
1650 2011-07-26 20:10:22 <jbalint> i guess i'm trying to see how a tx is validated before being included in a new block
1651 2011-07-26 20:10:35 <sipa> jbalint: in the current full client?
1652 2011-07-26 20:10:39 theorbtwo has quit (Ping timeout: 252 seconds)
1653 2011-07-26 20:10:42 <prof7bit> if you plan to store the entire block and never delete it you calculate it once to make sure the block is complete and then store it and from then on simply trust your local database.
1654 2011-07-26 20:10:43 <jbalint> sipa: sure
1655 2011-07-26 20:10:53 <sipa> jbalint: just check whether none if its inputs are spent
1656 2011-07-26 20:10:56 <jbalint> prof7bit: ok, that makes sense
1657 2011-07-26 20:11:03 <jbalint> sipa: thats too high level
1658 2011-07-26 20:11:08 <jbalint> of course thats the idea
1659 2011-07-26 20:11:11 <sipa> how so?
1660 2011-07-26 20:11:21 <sipa> you need a map of all unspent outputs
1661 2011-07-26 20:11:30 pierce has quit (Remote host closed the connection)
1662 2011-07-26 20:11:35 <sipa> and you verify whether all the inputs in a tx are there
1663 2011-07-26 20:11:46 <jbalint> because to do *that*, you need to verify first that it's in the block thats referenced as the previous output
1664 2011-07-26 20:11:47 <sipa> there is no need to look up a block or whatever for that
1665 2011-07-26 20:12:02 <sipa> ?
1666 2011-07-26 20:12:05 <prof7bit> the complicated thing with connecting inputs and outputs and verifying that it is ok is to do the OP_CHECKSIG dance
1667 2011-07-26 20:12:34 <jbalint> yeah, but to build a map of all unspect outputs, you need to iterate the WHOLE ENTIRE block chain, and calc the merkel trees (to verify tx's are in the block)
1668 2011-07-26 20:12:35 <prof7bit> and this is IMHO the most complex part of the entire protocol
1669 2011-07-26 20:12:51 <jbalint> yes, but that's more of a localized concern
1670 2011-07-26 20:12:55 <sipa> jbalint: no, just keep a database with all unspent outputs
1671 2011-07-26 20:13:02 <jbalint> where does this db come from?
1672 2011-07-26 20:13:03 <prof7bit> the merkle you do onyl when you receive a block
1673 2011-07-26 20:13:07 <sipa> jbalint: you keep it
1674 2011-07-26 20:13:27 <jbalint> sipa: you build it from scratch, which requires verification of the complete past transaction history?
1675 2011-07-26 20:13:29 <sipa> sorry, a database with all unspent outputs from transactions in the current best chain
1676 2011-07-26 20:13:33 <sipa> yes
1677 2011-07-26 20:13:39 <jbalint> thats what i just said
1678 2011-07-26 20:13:40 <sipa> you do that while downloading the block chain
1679 2011-07-26 20:13:47 datagutt has quit (Quit: Computer has gone to sleep.)
1680 2011-07-26 20:13:51 <jbalint> so we're in agreement
1681 2011-07-26 20:13:54 <sipa> no
1682 2011-07-26 20:14:05 <jbalint> what is inconsistent?
1683 2011-07-26 20:15:22 dr_win has quit (Remote host closed the connection)
1684 2011-07-26 20:15:30 <jbalint> sipa: what are we not agreeing about?
1685 2011-07-26 20:15:34 <sipa> well i don't see where you need a merkle path or tree to verify a tx, or create the database
1686 2011-07-26 20:16:03 SISUbtc has joined
1687 2011-07-26 20:16:04 <sipa> of course, you need to do the merkle tree hashing once when receiving the block, to check that its merkle root matches with the txs in it
1688 2011-07-26 20:16:15 <jbalint> ok, so you get a block somewhere in the middle of the chain, to verify that the given tx's are in that block (of which you initially only have the merkle root hash), you have to calc the whole tree
1689 2011-07-26 20:16:29 <jbalint> right
1690 2011-07-26 20:16:31 <prof7bit> the merkle tree should be checked directly after receiving it to make sure the block is complete
1691 2011-07-26 20:16:37 <jbalint> so to create the databas,e you need to verify the block's txs
1692 2011-07-26 20:16:45 <sipa> but once that is done, and other consistency checks succeed, you know the block is valid
1693 2011-07-26 20:16:46 <prof7bit> then you can throw it away
1694 2011-07-26 20:16:53 <sipa> and you can mark all txouts of that block as available
1695 2011-07-26 20:16:54 <jbalint> ok...
1696 2011-07-26 20:17:26 <sipa> and for further verification of blocks and txs, you need to check that all inputs of all the txs in it are available
1697 2011-07-26 20:17:39 <sipa> that implies that the tx you received is not yet in the block chain, by the way
1698 2011-07-26 20:17:42 <prof7bit> and then you have some clever indexing scheme in your database to quickly find the outputs that belong to the inputs and find your unspent outputs
1699 2011-07-26 20:18:15 <sipa> BlueMatt: hmm, idea: *first* lock cs_vMasterKey, and then cs_KeyStore
1700 2011-07-26 20:18:27 <jbalint> prof7bit: ;)
1701 2011-07-26 20:18:34 Daviey has quit (Excess Flood)
1702 2011-07-26 20:18:49 altamic has joined
1703 2011-07-26 20:18:52 <jbalint> sipa: ok. thanks
1704 2011-07-26 20:18:56 altamic has quit (Changing host)
1705 2011-07-26 20:18:56 altamic has joined
1706 2011-07-26 20:19:32 <prof7bit> and after all outputs of a tx have been spent after a while (few hundred blocks) you can even throw the entire tx away to make room on the hdd
1707 2011-07-26 20:19:37 imsaguy2 has joined
1708 2011-07-26 20:19:57 JRWR has quit (Read error: Connection reset by peer)
1709 2011-07-26 20:20:56 <jbalint> but if that block is requested by another participant, you have to supply it, right?
1710 2011-07-26 20:21:21 <prof7bit> only if you run a full node (NODE_NETWORK)
1711 2011-07-26 20:21:58 <prof7bit> but then you shoudl do all verification so that you don't just relay things from hearsay without checking it
1712 2011-07-26 20:22:04 zeropointo has quit (Quit: leaving)
1713 2011-07-26 20:22:46 malnilion has quit (Ping timeout: 252 seconds)
1714 2011-07-26 20:22:59 <jbalint> ok
1715 2011-07-26 20:23:04 macdada has left ()
1716 2011-07-26 20:23:18 <prof7bit> and in this case you *cannot* deloete anything on your disk. all data must be there to serve old blocks to others.
1717 2011-07-26 20:24:04 <jbalint> alright
1718 2011-07-26 20:25:20 zeropointo has joined
1719 2011-07-26 20:25:45 riush has quit (Ping timeout: 260 seconds)
1720 2011-07-26 20:26:05 SISUbtc has quit (Quit: Colloquy for iPad - http://colloquy.mobi)
1721 2011-07-26 20:26:13 Daviey has joined
1722 2011-07-26 20:26:38 Daviey has quit (Excess Flood)
1723 2011-07-26 20:26:42 <CIA-103> bitcoinj: hearn@google.com * r164 /trunk/src/com/google/bitcoin/core/Block.java: Fix a bogus comment in checkMerkleBranch(). http://bitcoinj.googlecode.com/svn-history/r164/
1724 2011-07-26 20:26:46 altamic has quit (Ping timeout: 276 seconds)
1725 2011-07-26 20:28:45 kmio has joined
1726 2011-07-26 20:29:55 MrTiggr has quit (Read error: Connection reset by peer)
1727 2011-07-26 20:31:04 MrTiggr has joined
1728 2011-07-26 20:31:33 koleg has quit (3!kvirc@79.133.132.158|Read error: Connection reset by peer)
1729 2011-07-26 20:35:16 ersi has quit (Changing host)
1730 2011-07-26 20:35:17 ersi has joined
1731 2011-07-26 20:35:32 JRWR has joined
1732 2011-07-26 20:35:32 JRWR has quit (Changing host)
1733 2011-07-26 20:35:32 JRWR has joined
1734 2011-07-26 20:36:49 thefinn93 has quit (Ping timeout: 255 seconds)
1735 2011-07-26 20:37:13 Daviey has joined
1736 2011-07-26 20:39:50 JRWR has quit (Ping timeout: 250 seconds)
1737 2011-07-26 20:39:54 dburns has joined
1738 2011-07-26 20:39:59 rethaw has quit (Quit: rethaw)
1739 2011-07-26 20:40:14 TheZimm has quit (Quit: Textual IRC Client: http://www.textualapp.com/)
1740 2011-07-26 20:40:25 MrTiggr has quit (Ping timeout: 240 seconds)
1741 2011-07-26 20:41:12 dburns has quit (Client Quit)
1742 2011-07-26 20:41:14 JRWR has joined
1743 2011-07-26 20:41:14 JRWR has quit (Changing host)
1744 2011-07-26 20:41:14 JRWR has joined
1745 2011-07-26 20:42:09 MrTiggr has joined
1746 2011-07-26 20:42:48 TheZimm has joined
1747 2011-07-26 20:47:20 osmosis has quit (Ping timeout: 240 seconds)
1748 2011-07-26 20:47:36 TheZimm has quit (Client Quit)
1749 2011-07-26 20:51:45 ByronJohnson has quit (Ping timeout: 240 seconds)
1750 2011-07-26 20:53:02 hippich has quit (Ping timeout: 255 seconds)
1751 2011-07-26 20:53:21 TD has quit (Ping timeout: 240 seconds)
1752 2011-07-26 20:54:23 <BlueMatt> sipa: no, you should be able to change the password without locking KeyStore
1753 2011-07-26 20:54:34 <BlueMatt> sipa: not 100% sure I coded it that way, but you /should/ be able to...
1754 2011-07-26 20:54:40 <BlueMatt> wait...no it wasnt coded that way
1755 2011-07-26 20:54:47 <BlueMatt> hmmm....no Im wrong
1756 2011-07-26 20:55:26 <BlueMatt> sipa: or...maybe just remove cs_MasterKey again
1757 2011-07-26 20:55:36 <BlueMatt> sipa: as you suggested originally, and just use cs_keyStore
1758 2011-07-26 20:55:40 theorbtwo has joined
1759 2011-07-26 20:56:25 TD has joined
1760 2011-07-26 21:02:23 JRWR has quit (Read error: Connection reset by peer)
1761 2011-07-26 21:04:02 glassresistor has quit (Remote host closed the connection)
1762 2011-07-26 21:04:51 <sipa> BlueMatt: a cs on masterkey is necessary, but maybe not public
1763 2011-07-26 21:06:30 <sipa> maybe a separate cs for "prevent locking"
1764 2011-07-26 21:06:38 <sipa> that is public
1765 2011-07-26 21:06:47 JRWR has joined
1766 2011-07-26 21:06:47 JRWR has quit (Changing host)
1767 2011-07-26 21:06:47 JRWR has joined
1768 2011-07-26 21:07:20 <BlueMatt> sipa: why not just lock the keystore and say instead of locking master key, just lock key store and use it for both
1769 2011-07-26 21:07:56 lyspooner has quit (Read error: Connection reset by peer)
1770 2011-07-26 21:10:12 JRWR has quit (Read error: Connection reset by peer)
1771 2011-07-26 21:10:23 osmosis has joined
1772 2011-07-26 21:10:59 nus has quit (Read error: Connection reset by peer)
1773 2011-07-26 21:11:19 osmosis has quit (Read error: Connection reset by peer)
1774 2011-07-26 21:12:01 nus has joined
1775 2011-07-26 21:12:51 <sipa> BlueMatt: because locking that way is quite drastic
1776 2011-07-26 21:13:25 <sipa> hey, that's it: a referenxe counter
1777 2011-07-26 21:13:27 <BlueMatt> well are you not going to have to lock both in all situations?
1778 2011-07-26 21:13:28 b4epoche_ has quit (Quit: Computer has gone to sleep.)
1779 2011-07-26 21:13:55 <sipa> as long as some operation uses the store, don't lock
1780 2011-07-26 21:14:07 ByronJohnson has joined
1781 2011-07-26 21:14:28 <BlueMatt> mmm, that works
1782 2011-07-26 21:20:19 JRWR has joined
1783 2011-07-26 21:20:19 JRWR has quit (Changing host)
1784 2011-07-26 21:20:19 JRWR has joined
1785 2011-07-26 21:22:39 MrTiggr has quit (Read error: Connection reset by peer)
1786 2011-07-26 21:23:04 MrTiggr has joined
1787 2011-07-26 21:24:52 JRWR has left ()
1788 2011-07-26 21:25:08 <sipa> i'llexcept that requires a condition variable...
1789 2011-07-26 21:27:07 <jgarzik> BlueMatt: has IO- launched their mining pool, and posted additional trust info?
1790 2011-07-26 21:27:51 <sipa> jgarzik: ?
1791 2011-07-26 21:28:24 BlueMattBot has quit (Ping timeout: 264 seconds)
1792 2011-07-26 21:29:08 <jgarzik> sipa: BlueMatt submitted a pull req for IO-'s dns seed. IMO, to be added as a DNS seed, there needs to be a -lot- of trust involved. At least to me, IO- is an unknown quantity. No idea how likely, or unlikely, he is to inject evil node addresses.
1793 2011-07-26 21:29:18 <jgarzik> sipa: This question is very relevant now that DNS seeding is default
1794 2011-07-26 21:30:30 BlueMatt has quit (Ping timeout: 260 seconds)
1795 2011-07-26 21:37:05 estornudo has quit (Ping timeout: 250 seconds)
1796 2011-07-26 21:37:24 BlueMatt has joined
1797 2011-07-26 21:39:31 dr_win has joined
1798 2011-07-26 21:41:55 jgarzik has quit (Quit: upgrades)
1799 2011-07-26 21:45:20 thefinn93 has joined
1800 2011-07-26 21:46:21 BlueMattBot has joined
1801 2011-07-26 21:48:49 Raccoon has quit ()
1802 2011-07-26 21:53:29 Mr_Tiggr has joined
1803 2011-07-26 21:55:23 sanity has joined
1804 2011-07-26 21:55:43 agricocb has quit (Quit: Leaving.)
1805 2011-07-26 21:57:09 MrTiggr has quit (Ping timeout: 260 seconds)
1806 2011-07-26 21:57:50 asuk has quit (Ping timeout: 255 seconds)
1807 2011-07-26 21:58:16 vigilyn has quit (Quit: Leaving)
1808 2011-07-26 21:58:56 Raccoon has joined
1809 2011-07-26 21:59:37 gp5st1 has quit (Quit: Leaving.)
1810 2011-07-26 22:03:14 eoss has joined
1811 2011-07-26 22:09:00 moa7 has joined
1812 2011-07-26 22:09:49 TuxBlackEdo has quit (Ping timeout: 258 seconds)
1813 2011-07-26 22:09:59 vigilyn has joined
1814 2011-07-26 22:12:34 imsaguy has quit (Read error: Connection reset by peer)
1815 2011-07-26 22:15:39 agricocb has joined
1816 2011-07-26 22:16:20 imsaguy has joined
1817 2011-07-26 22:21:02 imsaguy has quit (Ping timeout: 246 seconds)
1818 2011-07-26 22:21:37 imsaguy has joined
1819 2011-07-26 22:25:05 Mr_Tiggr has quit (Ping timeout: 276 seconds)
1820 2011-07-26 22:25:23 Mr_Tiggr has joined
1821 2011-07-26 22:25:55 Titeuf_87 has quit (Quit: Ex-Chat)
1822 2011-07-26 22:27:05 larsivi has quit (Ping timeout: 255 seconds)
1823 2011-07-26 22:27:45 skeledrew1 has joined
1824 2011-07-26 22:30:14 skeledrew has quit (Ping timeout: 255 seconds)
1825 2011-07-26 22:31:01 vigilyn has quit (Quit: Leaving)
1826 2011-07-26 22:31:50 jackjack___ has joined
1827 2011-07-26 22:35:50 xelister has joined
1828 2011-07-26 22:38:00 p0s has quit (Remote host closed the connection)
1829 2011-07-26 22:38:51 wirehead has quit (Read error: Operation timed out)
1830 2011-07-26 22:40:23 bitcoinbulletin has quit (Remote host closed the connection)
1831 2011-07-26 22:40:28 jackjack___ has quit (Quit: Page closed)
1832 2011-07-26 22:44:32 bitcoinbulletin has joined
1833 2011-07-26 22:44:41 sanity has quit (Ping timeout: 252 seconds)
1834 2011-07-26 22:47:26 TD has quit (Quit: TD)
1835 2011-07-26 22:51:30 vigilyn has joined
1836 2011-07-26 22:51:56 Mr_Tiggr has quit (Ping timeout: 276 seconds)
1837 2011-07-26 22:52:28 Mr_Tiggr has joined
1838 2011-07-26 22:54:47 b4epoche_ has joined
1839 2011-07-26 22:56:17 GMP has joined
1840 2011-07-26 22:57:56 MrTiggr has joined
1841 2011-07-26 22:58:16 xelister has quit (Read error: Operation timed out)
1842 2011-07-26 22:58:58 b4epoche_ has quit (Read error: Connection reset by peer)
1843 2011-07-26 23:00:04 b4epoche_ has joined
1844 2011-07-26 23:00:20 Mr_Tiggr has quit (Ping timeout: 260 seconds)
1845 2011-07-26 23:02:38 AStove has quit (Ping timeout: 255 seconds)
1846 2011-07-26 23:03:01 sabalaba has quit (Ping timeout: 240 seconds)
1847 2011-07-26 23:03:03 erus` has quit (Quit: ChatZilla 0.9.87 [Firefox 5.0/20110615151330])
1848 2011-07-26 23:06:30 datagutt has joined
1849 2011-07-26 23:09:33 zeropointo has quit (Quit: leaving)
1850 2011-07-26 23:09:38 Mr_Tiggr has joined
1851 2011-07-26 23:09:38 d1g1t4l has quit (Remote host closed the connection)
1852 2011-07-26 23:12:32 MrTiggr has quit (Ping timeout: 255 seconds)
1853 2011-07-26 23:14:42 mosimo has quit (Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com ))
1854 2011-07-26 23:19:19 dvide has quit ()
1855 2011-07-26 23:21:17 sabalaba has joined
1856 2011-07-26 23:21:36 Superbest has joined
1857 2011-07-26 23:31:28 gjs278 has quit (Ping timeout: 258 seconds)
1858 2011-07-26 23:31:33 FractalUniverse has quit ()
1859 2011-07-26 23:32:11 gjs278 has joined
1860 2011-07-26 23:32:47 copumpkin has quit (Quit: Computer has gone to sleep.)
1861 2011-07-26 23:33:11 nhodges has quit (Ping timeout: 250 seconds)
1862 2011-07-26 23:36:08 molecular has quit (Ping timeout: 276 seconds)
1863 2011-07-26 23:37:11 mtrlt_ has joined
1864 2011-07-26 23:38:34 gjs278 has quit (Ping timeout: 252 seconds)
1865 2011-07-26 23:40:29 mtrlt has quit (Ping timeout: 276 seconds)
1866 2011-07-26 23:42:14 Mr_Tiggr has quit (Ping timeout: 255 seconds)
1867 2011-07-26 23:46:14 <IO-> BlueMatt ping
1868 2011-07-26 23:46:28 <BlueMatt> IO-: da pong
1869 2011-07-26 23:46:48 * BlueMatt is distracted...but here
1870 2011-07-26 23:47:33 <IO-> www.unitedminers.com , "open beta"
1871 2011-07-26 23:47:55 <IO-> its up and live and production and I have a few randoms running
1872 2011-07-26 23:48:13 <IO-> i'm moving my miners over to it and double checking the managed rig instructions with a couple beta testers
1873 2011-07-26 23:48:14 <BlueMatt> nice
1874 2011-07-26 23:48:17 <IO-> thats still heavy beta
1875 2011-07-26 23:48:27 <IO-> but regular old worker L&P is all stable
1876 2011-07-26 23:48:31 <IO-> thats nothing special
1877 2011-07-26 23:49:00 Superbest has quit (Quit: It's time to man up and end this)
1878 2011-07-26 23:49:15 <BlueMatt> "We utilize Private Cloud technology with high availability and fault tolerant hosts." <-- anyone who says we use the cloud just tends to sound stupid...say you run your own servers, it always sounds better
1879 2011-07-26 23:49:18 <BlueMatt> dedicated servers
1880 2011-07-26 23:49:27 <upb> Private Cloud ?:P
1881 2011-07-26 23:49:36 <imsaguy> buzz words
1882 2011-07-26 23:49:37 <imsaguy> thats all
1883 2011-07-26 23:49:46 <IO-> yup, 100% buzz
1884 2011-07-26 23:49:51 <IO-> it's an ESX cluster
1885 2011-07-26 23:50:05 <IO-> so VM's float around between nodes as need be
1886 2011-07-26 23:50:17 <BlueMatt> still, better to say, hosted on multiple dedicated servers
1887 2011-07-26 23:50:23 <upb> imo too
1888 2011-07-26 23:50:26 <BlueMatt> doesnt mean the bitcoin stuff is on a dedicated server
1889 2011-07-26 23:50:29 <upb> ahh youre that guy
1890 2011-07-26 23:50:31 <BlueMatt> esx is though
1891 2011-07-26 23:51:05 <BlueMatt> still...looks awsome
1892 2011-07-26 23:51:18 * BlueMatt wishes he had the gpus to get anything done on a pool anymore
1893 2011-07-26 23:51:38 <Eliel> oh, managed mining too. I haven't seen that before :)
1894 2011-07-26 23:52:04 <Eliel> I won't be using it but I can see how other might find it tempting.
1895 2011-07-26 23:52:07 <IO-> its just a dedicated linux install that runs on a USB/SATA/SCSI boot drive and connects to me
1896 2011-07-26 23:52:15 <IO-> iphone/android web page made for fingers
1897 2011-07-26 23:52:26 <BlueMatt> still that looks really awsome
1898 2011-07-26 23:52:36 <BlueMatt> if I could spell awesome that is
1899 2011-07-26 23:52:36 <IO-> you can restarted it, change OC levels, change critical temp watchdog level, things like that
1900 2011-07-26 23:52:45 <IO-> all from your phone
1901 2011-07-26 23:52:54 <BlueMatt> thats really cool
1902 2011-07-26 23:52:56 hahuang65 has joined
1903 2011-07-26 23:53:06 <IO-> i've been using it on all my rigs for over 2 months now so it's been tested in the real world
1904 2011-07-26 23:53:14 <BlueMatt> have you added it to the list on the wiki and posted on the forum, or are you still waiting a bit for that?
1905 2011-07-26 23:53:15 <IO-> certainly not bug free
1906 2011-07-26 23:53:20 <IO-> wiki yes
1907 2011-07-26 23:53:31 <IO-> my forum post is made but not posted
1908 2011-07-26 23:53:39 <IO-> i want word back from a beta tester of mine on the managed rig
1909 2011-07-26 23:53:53 <IO-> i had a SQL problem he found and I fixed it but i want a confirmation
1910 2011-07-26 23:53:58 <IO-> i want to be stable out the door
1911 2011-07-26 23:54:01 <BlueMatt> ah, ok
1912 2011-07-26 23:54:10 <BlueMatt> damn this really looks awesome
1913 2011-07-26 23:54:22 <IO-> i just got a txt my rig is down
1914 2011-07-26 23:54:28 <IO-> over heated a bit ago
1915 2011-07-26 23:54:36 <IO-> it'll start up again automatically in about 10min
1916 2011-07-26 23:54:43 <BlueMatt> nice
1917 2011-07-26 23:56:29 <Eliel> I'm not using any especially reliable pools but there's automatic failover so I don't need to worry about pools being up. My miners are quite stable.
1918 2011-07-26 23:56:42 <IO-> one of my todo lists is to add some kinda click through disclaimer about the managed rig and OC'ing
1919 2011-07-26 23:56:46 <IO-> i guess i need it, i dont know
1920 2011-07-26 23:56:49 <Eliel> I haven't needed to touch them in ... over a week
1921 2011-07-26 23:57:03 a______ has joined
1922 2011-07-26 23:57:13 a______ has quit (Client Quit)
1923 2011-07-26 23:57:22 <IO-> i haven't touched my rigs in over 1.5 months, they self heal and self regulate
1924 2011-07-26 23:57:45 <Eliel> I haven't had these in this exact configuration for much longer than a week so :P
1925 2011-07-26 23:57:57 <IO-> when i was in active development mode i was rebooting 10 times a day :)
1926 2011-07-26 23:58:50 <IO-> someones already playing around
1927 2011-07-26 23:59:05 <Eliel> my GPUs are overclocked but in a conservative way. I prefer the systems stable than getting every last drop of power out.
1928 2011-07-26 23:59:29 <IO-> I started off too much and had to scale back