1 2011-03-21 00:02:13 <da2ce7> https://www.bitcoin.org/smf/index.php?topic=4721.msg68784#msg68784
  2 2011-03-21 00:02:27 <da2ce7> ;;bc,stats
  3 2011-03-21 00:02:29 <gribble> Current Blocks: 114327 | Current Difficulty: 76193.9710474 | Next Difficulty At Block: 114911 | Next Difficulty In: 584 blocks | Next Difficulty In About: 4 days, 17 hours, 13 minutes, and 52 seconds | Next Difficulty Estimate: 66491.75235706
  4 2011-03-21 00:03:14 <da2ce7> above ^ posted a bitcoin address refrenece specfication idea.
  5 2011-03-21 00:03:51 <joepie91> ... base 58?
  6 2011-03-21 00:04:06 <joepie91> 58 - 10 = 48
  7 2011-03-21 00:04:09 <joepie91> ?
  8 2011-03-21 00:04:12 <joepie91> am I missing something?
  9 2011-03-21 00:04:40 <joepie91> I assume you use digits, lowercase a-z, uppercase a-z
 10 2011-03-21 00:04:45 <joepie91> in which case it would be... base 62?
 11 2011-03-21 00:04:58 <joepie91> 10 + ( 26 * 2 )
 12 2011-03-21 00:05:02 <joepie91> or am I failing at math? :P
 13 2011-03-21 00:05:41 <da2ce7> base 58 is what the bitcoin addresses use.
 14 2011-03-21 00:05:42 <da2ce7> :P
 15 2011-03-21 00:05:51 <da2ce7> hmm I hope I got the maths right.
 16 2011-03-21 00:05:57 <da2ce7> I did it by hand.
 17 2011-03-21 00:06:29 <joepie91> then I wonder what characters are used for bitcoin addresses.. :/
 18 2011-03-21 00:06:30 <joepie91> or rather
 19 2011-03-21 00:06:34 <joepie91> what characters are not used
 20 2011-03-21 00:07:17 <da2ce7> https://github.com/bitcoin/bitcoin/blob/master/base58.h
 21 2011-03-21 00:07:24 <da2ce7> 0OIl
 22 2011-03-21 00:07:49 <joepie91> ahhh
 23 2011-03-21 00:07:55 <joepie91> ambiguity with 1 and 0?
 24 2011-03-21 00:07:56 <joepie91> :P
 25 2011-03-21 00:07:59 <da2ce7> yep
 26 2011-03-21 00:08:16 <joepie91> makes sense, lol
 27 2011-03-21 00:11:41 <joepie91> is anyone else annoyed by the fact that if you want to wget a tar.gz from sourceforge you need to go to the download page on sf.net, then have to copy the download link while in the meantime it starts downloading the archive to your OWN computer while you don't even need it there?
 28 2011-03-21 00:11:50 <joepie91> instead of them just providing a direct link to the archive...
 29 2011-03-21 00:15:43 <da2ce7> eeek, I have more photos than your mum on a good porn shoot gets to process :/
 30 2011-03-21 00:16:01 <da2ce7> 568 photos from one gig.
 31 2011-03-21 00:16:30 <joepie91> ./bitcoind: error while loading shared libraries: libgthread-2.0.so.0: cannot open shared object file: No such file or directory
 32 2011-03-21 00:16:31 <joepie91> lolwut
 33 2011-03-21 00:18:11 <joepie91> ah, was missing a lib
 34 2011-03-21 00:23:35 <genjix> i am annoyed by all the laggy ugly javascript on clunky websites trying to be flashy
 35 2011-03-21 00:23:45 <joepie91> heh
 36 2011-03-21 00:23:49 <genjix> just offer a link you pretentious bastards
 37 2011-03-21 00:24:00 <genjix> my browser knows what to do.
 38 2011-03-21 00:24:01 <joepie91> only use javascript when it improves the user experience :)
 39 2011-03-21 00:24:50 <joepie91> and only use flash when it's really really really REALLY necessary
 40 2011-03-21 00:28:26 <andrew12> ahoy
 41 2011-03-21 00:30:15 <joepie91> ohai
 42 2011-03-21 00:34:29 <Mango-chan> so
 43 2011-03-21 00:45:33 <gribble> Error: The "Later" plugin is loaded, but there is no command named "jgarzik" in it.  Try "list Later" to see the commands in the "Later" plugin.
 44 2011-03-21 00:45:33 <jgarzik> ;;later jgarzik test
 45 2011-03-21 00:45:47 <gribble> Error: The "Later" plugin is loaded, but there is no command named "send" in it.  Try "list Later" to see the commands in the "Later" plugin.
 46 2011-03-21 00:45:47 <jgarzik> ;;later send jgarzik test
 47 2011-03-21 00:46:00 <gribble> The operation succeeded.
 48 2011-03-21 00:46:00 <jgarzik> ;;later tell jgarzik test
 49 2011-03-21 00:46:30 <gribble> The operation succeeded.
 50 2011-03-21 00:46:30 <jgarzik> ;;later tell BlueMatt I use Fedora 14's outstanding mingw32 support to build cpuminer Windows packages entirely on Linux.
 51 2011-03-21 00:55:41 <joepie91> later tell is some sort of memoserv?
 52 2011-03-21 00:56:18 <Lachesis> i suppose
 53 2011-03-21 00:56:19 <luke-jr> yes
 54 2011-03-21 00:56:21 <Lachesis> it does what it sounds like
 55 2011-03-21 00:56:30 <joepie91> heh
 56 2011-03-21 00:56:51 <joepie91> okay, backtrace is sort of disappointing me
 57 2011-03-21 00:57:05 <joepie91> they are now re-releasing the same logs they already re-released to gawker
 58 2011-03-21 00:57:13 <joepie91> which were originally published by crowdleaks
 59 2011-03-21 00:57:27 <joepie91> in a pdf, with a watermark
 60 2011-03-21 00:57:39 <joepie91> ... how many times can you try to attentionwhore off one set of logs >.<
 61 2011-03-21 00:58:35 <theymos> What logs?
 62 2011-03-21 01:00:23 <joepie91> logs from a supposed "headquarters"
 63 2011-03-21 01:00:32 <jgarzik> offtopic logs, I'm guessing?  :)
 64 2011-03-21 01:00:35 <joepie91> related to anonymous
 65 2011-03-21 01:00:44 <joepie91> to the hbgary leak, more specifically
 66 2011-03-21 01:17:56 <joepie91> o/
 67 2011-03-21 01:17:58 <joepie91> dns werking
 68 2011-03-21 01:18:05 <joepie91> http://www.paybtc.com
 69 2011-03-21 01:21:18 <jgarzik> where's ssl? :)
 70 2011-03-21 01:25:25 <Blitzboom> joepie91: webwallet service?
 71 2011-03-21 01:25:36 <joepie91> nop
 72 2011-03-21 01:25:40 <joepie91> direct payment processing
 73 2011-03-21 01:25:50 <joepie91> you create a payment page by entering your actual bitcoin address and the amount
 74 2011-03-21 01:25:51 <genjix> joepie91: nice. so i dont need a bitcoind running
 75 2011-03-21 01:25:56 <joepie91> either through an API or through the site itself
 76 2011-03-21 01:26:00 <joepie91> you get a payment url
 77 2011-03-21 01:26:09 <genjix> joepie91: are you using php?
 78 2011-03-21 01:26:14 <noagendamarket> nice
 79 2011-03-21 01:26:24 <joepie91> you give the customer the url, he gets a gateway address... as soon as he sends the money, the gateway makes a callback or sends an email and IMMEDIATELY sends the payment to the actual address
 80 2011-03-21 01:26:26 <joepie91> genjix: yup
 81 2011-03-21 01:26:32 <joepie91> and ssl is not yet there, waiting for validation
 82 2011-03-21 01:26:39 <genjix> ok i need to give you some tips
 83 2011-03-21 01:26:58 <genjix> because you're going to have problems using bitcoin by default in php
 84 2011-03-21 01:27:31 <joepie91> oh?
 85 2011-03-21 01:27:32 <joepie91> howso?
 86 2011-03-21 01:27:37 <genjix> https://en.bitcoin.it/wiki/API_tutorial_%28JSON-RPC%29#Precision
 87 2011-03-21 01:27:43 <genjix> you need to read that
 88 2011-03-21 01:27:48 <luke-jr> joepie91: JSON-RPC is terribly broken, etc etc
 89 2011-03-21 01:27:55 <luke-jr> joepie91: but it's not THAT hard to workaround
 90 2011-03-21 01:28:07 <luke-jr> at least not this particular problem
 91 2011-03-21 01:28:12 <genjix> well if you're dealing with people's money then the php json encodes values as floats
 92 2011-03-21 01:28:40 <luke-jr> joepie91: genjix's "solution" is also broken though
 93 2011-03-21 01:28:42 <genjix> joepie91: also if you're dealing with money anyway, you should be using ints
 94 2011-03-21 01:28:54 <genjix> luke-jr: how is it broken. stop spreading false info.
 95 2011-03-21 01:29:16 <joepie91> whoa, wait
 96 2011-03-21 01:29:17 <luke-jr> oh, maybe not.
 97 2011-03-21 01:29:19 <joepie91> information overload
 98 2011-03-21 01:29:23 <joepie91> lol
 99 2011-03-21 01:29:24 <luke-jr> genjix: why is GMP needed?
100 2011-03-21 01:29:31 <joepie91> what exactly is broken again?
101 2011-03-21 01:29:36 <luke-jr> joepie91: JSON-RPC
102 2011-03-21 01:30:01 <joepie91> speaking of which
103 2011-03-21 01:30:02 <joepie91> http://jsonrpcphp.org/
104 2011-03-21 01:30:03 <joepie91> broken
105 2011-03-21 01:30:07 <genjix> luke-jr: to handle large ints. it does accurate division .etc
106 2011-03-21 01:30:11 <joepie91> the irony...
107 2011-03-21 01:30:27 <genjix> joepie91: 1) you should be handling the money as ints (use GMP)
108 2011-03-21 01:30:34 <joepie91> GMP?
109 2011-03-21 01:30:43 <theymos> I prefer bcmath.
110 2011-03-21 01:30:45 <genjix> 2) you need to use my branch of bitcoin
111 2011-03-21 01:30:51 <genjix> theymos: not good.
112 2011-03-21 01:30:54 <luke-jr> genjix: PHP doesn't have 64-bit numbers?
113 2011-03-21 01:31:00 <theymos> genjix: Why not?
114 2011-03-21 01:31:22 <genjix> theymos: if you divide 10 btc / 3 then you will get 3.3333333 (truncated).
115 2011-03-21 01:31:31 <jgarzik> genjix: CBigNum does not work for you?
116 2011-03-21 01:31:41 <jgarzik> genjix: we don't need GMP
117 2011-03-21 01:31:45 <genjix> if using large ints then you'll get 3333333333 and the remainder which you can then handle.
118 2011-03-21 01:31:49 <joepie91> genjix: I suppose your branch does not provide binary downloads?
119 2011-03-21 01:31:52 <luke-jr> jgarzik: CBigNum in PHP?
120 2011-03-21 01:31:54 <genjix> you don't want to be leaking money.
121 2011-03-21 01:32:18 <Kiba> The bitcoinweekly.com is ready for operation
122 2011-03-21 01:32:38 <Kiba> :D
123 2011-03-21 01:32:42 <jgarzik> luke-jr: I guess I misunderstood.  I thought genjix was talking about the official bitcoin client.
124 2011-03-21 01:32:53 <genjix> ok
125 2011-03-21 01:32:58 <luke-jr> genjix: does int(value * 1e8) really not work in PHP?
126 2011-03-21 01:33:15 <jgarzik> python does bignums and web quite well :)
127 2011-03-21 01:33:33 <genjix> except when you do 10//3 :p
128 2011-03-21 01:33:49 <genjix> you should be handling the modulus too
129 2011-03-21 01:33:57 <joepie91> 1. genjix: do you provide binary downloads for your branch, and if not, how do I compile it on Debian?
130 2011-03-21 01:34:03 <jgarzik> python does mod
131 2011-03-21 01:34:09 <genjix> yep
132 2011-03-21 01:34:22 <genjix> joepie91: you compile it same way as normal bitcoin
133 2011-03-21 01:34:29 <genjix> make -f makefile.unix bitcoind
134 2011-03-21 01:34:39 <joepie91> errr, what packages are required to build it?
135 2011-03-21 01:34:44 <genjix> luke-jr: i don't think so
136 2011-03-21 01:34:45 <joepie91> never been able to found a list of packages for debian
137 2011-03-21 01:34:48 <joepie91> err
138 2011-03-21 01:34:49 <joepie91> find*
139 2011-03-21 01:34:52 <joepie91> (it's late, lol)
140 2011-03-21 01:35:01 <genjix> because you get values like 0.999999999999999
141 2011-03-21 01:35:02 <luke-jr> joepie91: building bitcoind is a pain. you shouldn't need any modifications
142 2011-03-21 01:35:07 <joepie91> ...
143 2011-03-21 01:35:14 <genjix> joepie91: i have instructions on the dev subforum
144 2011-03-21 01:35:14 <joepie91> now do I need genjix branch or not? >.<
145 2011-03-21 01:35:20 <luke-jr> joepie91: I highly doubt it.
146 2011-03-21 01:35:42 <genjix> joepie91: http://www.bitcoin.org/smf/index.php?topic=4572.0
147 2011-03-21 01:36:06 <joepie91> errr, ubuntu =/= debian
148 2011-03-21 01:36:11 <joepie91> half the packages does not even exist in debian
149 2011-03-21 01:36:21 <theymos> genjix: Wouldn't you truncate, anyway?
150 2011-03-21 01:36:23 <genjix> joepie91: yes you do. if you're handling other people's money then it's super bad if you're leaking money.
151 2011-03-21 01:36:28 <genjix> theymos: no.
152 2011-03-21 01:36:40 <genjix> i would send the money back to their account or do something
153 2011-03-21 01:36:56 <genjix> i seriously hope you aren't dropping money from people's accounts....
154 2011-03-21 01:37:44 <luke-jr> joepie91: just feed all monetary values you get from JSON-RPC through ^
155 2011-03-21 01:37:56 <luke-jr> and when you send JSON-RPC, you need to do the reverse: value / 1e8
156 2011-03-21 01:38:07 <genjix> that's an awful hack.
157 2011-03-21 01:38:12 <genjix> round(...)
158 2011-03-21 01:38:12 <joepie91> mh
159 2011-03-21 01:38:24 <joepie91> sounds like the hack I used to store ratings in my mysql database
160 2011-03-21 01:38:28 <luke-jr> genjix: it works, and better than your hack :P
161 2011-03-21 01:38:37 <joepie91> (for some quickly hacked together project)
162 2011-03-21 01:38:41 <genjix> how is it better thna my 'hack'
163 2011-03-21 01:38:44 <luke-jr> joepie91: also, keep in mind that real bitcoin units are 1/100000000 of a BTC
164 2011-03-21 01:38:50 <genjix> my version is 100% accurate
165 2011-03-21 01:38:51 <joepie91> luke-jr: aware
166 2011-03-21 01:39:00 <genjix> yours is "mostly" accurate.
167 2011-03-21 01:39:04 <ArtForz> ... so is his
168 2011-03-21 01:39:18 <andrew12> are any of you guys on a mac?
169 2011-03-21 01:39:22 <joepie91> tbh I do not see why * 1e8 would not be accurate
170 2011-03-21 01:39:26 <luke-jr> joepie91: so when you accept user input, or render it, you need to do some formatting
171 2011-03-21 01:39:33 <luke-jr> joepie91: you *don't* want to do /1e8 for formatting
172 2011-03-21 01:39:45 <joepie91> lol, howso
173 2011-03-21 01:39:53 <luke-jr> probably want to convert to string, insert a decimal point, and trim extra 0s
174 2011-03-21 01:40:06 <luke-jr> joepie91: 10000000 / 1e8 = 0.0999999999999
175 2011-03-21 01:40:18 <joepie91> ah.
176 2011-03-21 01:40:42 <ArtForz> luke-jr: now limit the print to 8 decimal places
177 2011-03-21 01:40:51 <luke-jr> it's because 0.1 cannot be converted to binary as-is
178 2011-03-21 01:41:10 <genjix> that's such a cavalier attitude
179 2011-03-21 01:41:23 <luke-jr> ArtForz: that might work, not sure. might be implementation-specific too
180 2011-03-21 01:41:33 <ArtForz> just dont do internal math using floats ffs
181 2011-03-21 01:41:49 <joepie91> k, it might be because it's late, but I'm completely lost here on what to do now, due to contradicting advice :P Now, what is the 1-2-3 list of things to do that will be most accurate?
182 2011-03-21 01:42:01 <genjix> so why does the json rpc return floats then and not strings?
183 2011-03-21 01:42:13 <genjix> why the resistance to returning strings (which works)
184 2011-03-21 01:42:14 <luke-jr> joepie91: whenever you speak JSON-RPC, use round(amount * 1e8), and amount / 1e8
185 2011-03-21 01:42:32 <genjix> otherwise you have to make all these workaround hacks for ruby/php/python
186 2011-03-21 01:42:39 <luke-jr> joepie91: whenever you speak to humans, convert to string and do some magic formatting tricks
187 2011-03-21 01:42:42 <genjix> joepie91: read that wiki link i posted.
188 2011-03-21 01:42:46 <ArtForz> actually for python JSON it's easy as shit
189 2011-03-21 01:42:51 <luke-jr> genjix: because that's just as broken as JSON-RPC is now
190 2011-03-21 01:42:54 <ArtForz> for python json-rpc ... not so much
191 2011-03-21 01:43:09 <genjix> ArtForz: https://en.bitcoin.it/wiki/API_tutorial_%28JSON-RPC%29#Python
192 2011-03-21 01:43:12 <devrandom> https://secure.wikimedia.org/wikipedia/en/wiki/Floating_point#Accuracy_problems
193 2011-03-21 01:43:15 <genjix> i write my own serviceproxy
194 2011-03-21 01:43:27 <genjix> which uses decimal.Decimal
195 2011-03-21 01:43:33 <ArtForz> yep
196 2011-03-21 01:43:50 <ArtForz> I did something a bit uglier
197 2011-03-21 01:43:55 <joepie91> if I'd use your branch, genjix, then what packages do I install? half the packages in your list does not exist in Debian
198 2011-03-21 01:44:07 <ArtForz> fake http on top of socket, using json
199 2011-03-21 01:44:11 <luke-jr> yawn
200 2011-03-21 01:44:23 <genjix> joepie91: erm well you need to get them working somehow to build bitcoin anyway
201 2011-03-21 01:44:24 <devrandom> floats don't even have 8 decimal digits
202 2011-03-21 01:44:34 <joepie91> genjix: what is not there is not there, lol
203 2011-03-21 01:44:45 <genjix> ArtForz: and also ruby has no way to encode json in anything except floats
204 2011-03-21 01:44:52 <ArtForz> eww
205 2011-03-21 01:44:54 <joepie91> debian does not have all the packages that ubuntu has
206 2011-03-21 01:44:59 <genjix> https://en.bitcoin.it/wiki/API_tutorial_%28JSON-RPC%29#Ruby
207 2011-03-21 01:45:09 <joepie91> know what, I'll just try the hack approach first and just run some tests
208 2011-03-21 01:45:14 <genjix> response: python/php/perl/ruby is broken. WONTFIX.
209 2011-03-21 01:45:15 <joepie91> and if that turns out not to work...
210 2011-03-21 01:45:17 <genjix> dumb.
211 2011-03-21 01:45:41 <genjix> joepie91: then i will not use your service for anything critical.
212 2011-03-21 01:45:47 <genjix> it will just be a toy...
213 2011-03-21 01:45:55 <ArtForz> genjix: like... mtgox?
214 2011-03-21 01:46:06 <joepie91> genjix.
215 2011-03-21 01:46:16 <joepie91> at this moment I have no clue how to compile bitcoind on debian
216 2011-03-21 01:46:24 <joepie91> and apparently noone has sufficient instructions
217 2011-03-21 01:46:37 <joepie91> I can play around with my debian box for days on end to figure out how to get stuff working
218 2011-03-21 01:46:43 <joepie91> or I can first try the shorter approach
219 2011-03-21 01:46:46 <genjix> mtgox uses floats? i cannot see their source so i dont know.
220 2011-03-21 01:46:47 <genjix> anyway mtgox is not a bastion of good coding
221 2011-03-21 01:46:50 <joepie91> run sufficient tests to verify it works
222 2011-03-21 01:46:58 <genjix> note the brute forcing of people's passwords and putting password strings in GET requests
223 2011-03-21 01:47:02 <joepie91> and _if_ it turns out it doesn't work
224 2011-03-21 01:47:03 <joepie91> ->
225 2011-03-21 01:47:05 <ArtForz> genjix: well, unless they just randomly round wrong, they use floats.
226 2011-03-21 01:47:06 <joepie91> next step
227 2011-03-21 01:47:10 <luke-jr> joepie91: they're both hacks
228 2011-03-21 01:47:16 <luke-jr> joepie91: genjix's hack is uglier :P
229 2011-03-21 01:47:18 <genjix> ArtForz: yeah that's pretty shitty.
230 2011-03-21 01:47:37 <joepie91> I have no idea how genjix' hack works so whether it's ugly or not is not something I know, nor something I really care about a lot
231 2011-03-21 01:47:47 <joepie91> what interests me is reliability
232 2011-03-21 01:47:48 <genjix> if i'm handling people's money then i would *never* use floats.
233 2011-03-21 01:47:57 <joepie91> if the simple hack works 100% accurate, there is no problem
234 2011-03-21 01:48:00 <joepie91> if it doesn't, there is a problem
235 2011-03-21 01:48:02 <luke-jr> it does.
236 2011-03-21 01:48:11 <joepie91> tests will tell.
237 2011-03-21 01:48:13 <luke-jr> the hard part is really displaying it for humans
238 2011-03-21 01:48:28 <devrandom> serious financial systems never use floats
239 2011-03-21 01:48:50 <luke-jr> anyhow, I've been working on a real solution
240 2011-03-21 01:49:00 <luke-jr> it takes time, and the simple hack for JSON-RPC is good enough for now
241 2011-03-21 01:49:09 <theymos> I can't think of any common cases where you would divide money in a situation where truncating would not be OK.
242 2011-03-21 01:49:46 <devrandom> bitcoin requires 50+ bits to represent correctly
243 2011-03-21 01:49:51 <devrandom> I'm not sure that even fits a double
244 2011-03-21 01:49:53 <luke-jr> you mean < 52 :P
245 2011-03-21 01:50:08 <luke-jr> devrandom: it works fine in a 64-bit double, if everything is an integer
246 2011-03-21 01:50:18 <luke-jr> err, double is by definition 64-bit, so yes
247 2011-03-21 01:50:31 <devrandom> you foget that the exponent takes up bits
248 2011-03-21 01:50:32 <luke-jr> devrandom: the problem is that values like 0.1 *cannot* be repesented exactly
249 2011-03-21 01:50:35 <ArtForz> double has 53 bits of mantissa
250 2011-03-21 01:50:39 <luke-jr> devrandom: no, I didn't forget
251 2011-03-21 01:50:56 <luke-jr> devrandom: the other problem is that BTC is strictly HUMAN-SIDE, and should not be used for internals
252 2011-03-21 01:51:03 <ArtForz> biggest possible amount fits in 51 bits
253 2011-03-21 01:51:10 <devrandom> ok
254 2011-03-21 01:51:19 <joepie91> simple question: if I limit the amount of decimals for a transaction to 2 (so that 0.01 is the smallest possible value), does the problem exist?
255 2011-03-21 01:51:26 <luke-jr> joepie91: yes
256 2011-03-21 01:51:31 <luke-jr> joepie91: also, that would be mean
257 2011-03-21 01:51:43 <luke-jr> joepie91: such a limitation makes TBC compatibility basically impossible
258 2011-03-21 01:51:49 <devrandom> you shouldn't do that, because it will stop working once deflation requires smaller fractions
259 2011-03-21 01:51:54 <joepie91> aware
260 2011-03-21 01:52:02 <joepie91> I wouldn't ever implement it as a permanent solution, lol
261 2011-03-21 01:52:19 <luke-jr> joepie91: what are you writign anyhow?
262 2011-03-21 01:52:28 <joepie91> payment gateway
263 2011-03-21 01:52:39 <joepie91> not an e-wallet thing or anything
264 2011-03-21 01:52:53 <joepie91> basically just a "middle man" gateway that flips the switch when a payment has been sent
265 2011-03-21 01:52:57 <joepie91> so that you can automate for example webshops
266 2011-03-21 01:53:01 <joepie91> without needing to run bitcoind
267 2011-03-21 01:53:02 <luke-jr> ah
268 2011-03-21 01:53:14 <luke-jr> how do you do that without e-wallet?
269 2011-03-21 01:53:14 <TheKid> so a webshop would integrate your software
270 2011-03-21 01:53:16 <joepie91> it immediately forwards the payment, and no account is required to use the system
271 2011-03-21 01:53:18 <joepie91> yup
272 2011-03-21 01:53:22 <luke-jr> i c
273 2011-03-21 01:53:36 <TheKid> and forwards the payment to wherever, while also sending a message to the site that payment was received
274 2011-03-21 01:53:38 <TheKid> right?>
275 2011-03-21 01:53:42 <joepie91> you will be able to create payment requests through both an API and a form on the site
276 2011-03-21 01:53:43 <joepie91> yup
277 2011-03-21 01:53:51 <joepie91> or send an email with a predefined text
278 2011-03-21 01:53:57 <joepie91> based on what the payment "creator" sets
279 2011-03-21 01:54:10 <luke-jr> joepie91: will you support QR-Codes?
280 2011-03-21 01:54:26 <joepie91> it will use a double callback system, where the target of the callback will have to make a request back to the gateway to confirm the callback was legit
281 2011-03-21 01:54:36 <joepie91> luke-jr: in what way? I haven't seen anything in regards to QR-codes and btc yet
282 2011-03-21 01:54:40 <joepie91> :P*
283 2011-03-21 01:54:41 <luke-jr> eg, a QR-Code embeds http://yourdomain/path/to/php, which generates a new address and redirects to bitcoin:address?amount=&
284 2011-03-21 01:54:56 <joepie91> wait, I don't follow
285 2011-03-21 01:55:10 <joepie91> rephrase? :P
286 2011-03-21 01:55:17 <luke-jr> joepie91: QR-Codes embed URIs
287 2011-03-21 01:55:32 <luke-jr> so your merchant would have a QR-Code that points to a script on your site
288 2011-03-21 01:55:37 <joepie91> ah
289 2011-03-21 01:55:45 <luke-jr> your site generates a payment address, and uses a standard HTTP Redirect to a bitcoin URI
290 2011-03-21 01:55:56 <joepie91> well, the API will just return a page URL
291 2011-03-21 01:56:00 <joepie91> which is essentially the payment page
292 2011-03-21 01:56:00 <luke-jr> customer's phone pops up a transaction confirmation
293 2011-03-21 01:56:03 <luke-jr> they click OK
294 2011-03-21 01:56:06 <joepie91> I guess I could make it return a qr as well
295 2011-03-21 01:56:10 <joepie91> orw ell
296 2011-03-21 01:56:13 <joepie91> link to a qr
297 2011-03-21 01:56:34 <joepie91> hmm... for that to work
298 2011-03-21 01:56:35 <luke-jr> joepie91: well, this way, the customer only needs a bitcoin-compatible phone
299 2011-03-21 01:56:38 <luke-jr> not a full browser
300 2011-03-21 01:56:42 <joepie91> would the QR target url need to send a 302 redirect?
301 2011-03-21 01:56:56 <joepie91> to the bitcoin uri, I mean
302 2011-03-21 01:57:01 <noagendamarket> this is what paypal does
303 2011-03-21 01:57:02 <devrandom> ArtForz - looks like BTC are represented as binary coded decimal in the protocol?
304 2011-03-21 01:57:04 <noagendamarket> btw
305 2011-03-21 01:57:21 <luke-jr> joepie91: right
306 2011-03-21 01:57:22 <noagendamarket> you just set a price and copy the html
307 2011-03-21 01:57:29 <ArtForz> ?
308 2011-03-21 01:57:32 <joepie91> lol, paypal
309 2011-03-21 01:57:38 <luke-jr> devrandom: that is self-contradictign :P
310 2011-03-21 01:57:40 <joepie91> they have some retarded system with POST requests etc
311 2011-03-21 01:57:44 <luke-jr> devrandom: in the p2p protocol, it's an int64
312 2011-03-21 01:57:47 <noagendamarket> ya
313 2011-03-21 01:57:57 <joepie91> PayPal is absolutely horrible to work with
314 2011-03-21 01:57:58 <joepie91> srsly
315 2011-03-21 01:58:13 <noagendamarket> I like th fact you can just copy and paste html and it adds a button :)-
316 2011-03-21 01:58:16 <luke-jr> joepie91: this *should* use a POST request&
317 2011-03-21 01:58:19 <devrandom> 80 FA E9 C7 00 00 00 00                           - 33.54 BTC (3354000000)
318 2011-03-21 01:58:23 <joepie91> no, I mean for the button
319 2011-03-21 01:58:24 <devrandom> https://en.bitcoin.it/wiki/Protocol_specification
320 2011-03-21 01:58:30 <luke-jr> joepie91: it should.
321 2011-03-21 01:58:31 <joepie91> a "paypal button" is actually an encrypted string in a hidden input
322 2011-03-21 01:58:35 <joepie91> with a paypal image as a "submit button"
323 2011-03-21 01:58:38 <devrandom> oh, never mind, read that incorrectly
324 2011-03-21 01:58:41 <luke-jr> joepie91: since you need to generate an address, it should be POST
325 2011-03-21 01:58:48 <joepie91> no, you don't
326 2011-03-21 01:58:54 <luke-jr> um, yes you do
327 2011-03-21 01:58:55 <joepie91> not for the actual button
328 2011-03-21 01:58:58 <joepie91> sigh
329 2011-03-21 01:59:00 <luke-jr> the target of it
330 2011-03-21 01:59:04 <joepie91> I am talking about the button you place on your _site_
331 2011-03-21 01:59:07 <joepie91> not on the paypal site
332 2011-03-21 01:59:18 <joepie91> if you create a donation button, that button is actually a submit button image
333 2011-03-21 01:59:18 <luke-jr> when I click the button, your page needs to generate an address for me to send to
334 2011-03-21 01:59:20 <joepie91> with a hidden input
335 2011-03-21 01:59:28 <joepie91> <joepie91>I am talking about the button you place on your _site_
336 2011-03-21 01:59:38 <luke-jr> joepie91: me too
337 2011-03-21 01:59:44 <joepie91> that does not in any way require a post request.
338 2011-03-21 01:59:50 <luke-jr> it does
339 2011-03-21 02:00:01 <joepie91> http://www.paybtc.com/pay/hfg8wsm
340 2011-03-21 02:00:04 <luke-jr> because the *target* needs to generate an address
341 2011-03-21 02:00:07 <joepie91> done
342 2011-03-21 02:00:09 <joepie91> no post needed.
343 2011-03-21 02:00:10 <luke-jr> joepie91: yes, you CAN do that, but it's WRONG
344 2011-03-21 02:00:15 <joepie91> how would that be wrong?
345 2011-03-21 02:00:32 <luke-jr> anything like that is supposed to be a POST
346 2011-03-21 02:00:37 <joepie91> *sigh*
347 2011-03-21 02:00:38 <joepie91> WHY
348 2011-03-21 02:00:42 <joepie91> would that need to be a post
349 2011-03-21 02:00:53 <luke-jr> because you're doing something when the request comes in
350 2011-03-21 02:00:58 <joepie91> gahhhhhhhhhhh
351 2011-03-21 02:00:58 <luke-jr> GET is strictly for read-only
352 2011-03-21 02:01:04 <joepie91> you don't get it
353 2011-03-21 02:01:09 <joepie91> you create a payment page.
354 2011-03-21 02:01:12 <luke-jr> -.-
355 2011-03-21 02:01:15 <joepie91> that generates a payment page url
356 2011-03-21 02:01:18 <joepie91> you get that url back
357 2011-03-21 02:01:18 <luke-jr> GET is like 'const' on a C++ method
358 2011-03-21 02:01:21 <joepie91> you link to that url
359 2011-03-21 02:01:24 <joepie91> customer goes there
360 2011-03-21 02:01:29 <joepie91> gets a "paybtc" page
361 2011-03-21 02:01:31 <joepie91> with a bitcoin address
362 2011-03-21 02:01:34 <joepie91> that is handled by paypal
363 2011-03-21 02:01:37 <joepie91> or a bitcoin uri
364 2011-03-21 02:01:39 <luke-jr> that's even worse
365 2011-03-21 02:01:45 <luke-jr> and yes, requires a POST there
366 2011-03-21 02:01:59 <joepie91> you have still not clearly explained why that would require a POST.
367 2011-03-21 02:02:13 <luke-jr> joepie91: because if the request is GET, it is invalid to act on it
368 2011-03-21 02:02:13 <theymos> It doesn't sound like there's any action being performed, so a POST isn't necessary.
369 2011-03-21 02:02:22 <joepie91> luke-jr
370 2011-03-21 02:02:22 <luke-jr> theymos: generate new address is an action
371 2011-03-21 02:02:24 <joepie91> the payment page
372 2011-03-21 02:02:25 <joepie91> is created
373 2011-03-21 02:02:29 <joepie91> by the server itself
374 2011-03-21 02:02:31 <luke-jr> theymos: adding a db entry is an action
375 2011-03-21 02:02:32 <joepie91> not by submitting a form
376 2011-03-21 02:02:37 <joepie91> it is done when the API is called
377 2011-03-21 02:02:44 <luke-jr> joepie91: but it's invalid for the server to create anything based on a GET
378 2011-03-21 02:02:53 <joepie91> good god
379 2011-03-21 02:02:59 <luke-jr> GET is only for retrieval of existing data
380 2011-03-21 02:03:08 <joepie91> you have absolutely no clue how http works, lol
381 2011-03-21 02:03:14 <[Tycho]> Do not create anything with GET :)
382 2011-03-21 02:03:16 <luke-jr> joepie91: no u
383 2011-03-21 02:03:19 <joepie91> the webshop software _itself_ makes an API call and gets back a payment url.
384 2011-03-21 02:03:28 <luke-jr> more importantly, what you just described requires the merchant to have scripts. totally unnecessary.
385 2011-03-21 02:03:30 <joepie91> when the API call is made
386 2011-03-21 02:03:38 <joepie91> the payment is added into the database
387 2011-03-21 02:03:46 <luke-jr> joepie91: that means the webshop itself needs to be a POST
388 2011-03-21 02:03:49 <joepie91> the webshop redirects the user to the payment url that was returned
389 2011-03-21 02:03:55 <joepie91> the user pays there
390 2011-03-21 02:03:56 <joepie91> et voila
391 2011-03-21 02:04:02 <luke-jr> joepie91: that's way harder than PayPal
392 2011-03-21 02:04:09 <luke-jr> and requires more expensive servers
393 2011-03-21 02:04:10 <joepie91> this is about as simple as it gets.
394 2011-03-21 02:04:13 <luke-jr> no
395 2011-03-21 02:04:20 <luke-jr> PayPal gives you a fixed piece of static code
396 2011-03-21 02:04:21 <devrandom> it's really about GET being idempotent... two GETs should not reach a different state than one
397 2011-03-21 02:04:39 <joepie91> luke-jr: so does this system.
398 2011-03-21 02:04:41 <devrandom> I think it's okay to generate an address, if you don't generate it again on a second GET
399 2011-03-21 02:04:44 <joepie91> just this piece of static code is not an entire form
400 2011-03-21 02:04:48 <joepie91> but just a url
401 2011-03-21 02:04:51 <[Tycho]> And it would be easier without API call from merchant.
402 2011-03-21 02:04:59 <joepie91> all address generation is handled by paybtc itself
403 2011-03-21 02:05:17 <joepie91> in the point from webshop page -> paybtc page, there is no requirement for a POST whatsoever
404 2011-03-21 02:05:17 <luke-jr> joepie91: it is illegal to generate an address in response to a GET request.
405 2011-03-21 02:05:37 <joepie91> sigh
406 2011-03-21 02:05:40 <[Tycho]> I agree. Stop abusing HTTP !
407 2011-03-21 02:05:42 <joepie91> I have explained it several times now
408 2011-03-21 02:05:46 <joepie91> the _API_
409 2011-03-21 02:05:49 <joepie91> makes a POST request
410 2011-03-21 02:05:53 <joepie91> the BROWSER
411 2011-03-21 02:05:55 <joepie91> makes a GET request
412 2011-03-21 02:05:58 <joepie91> it is _not_ the same request
413 2011-03-21 02:06:09 <joepie91> the API is the one that actually creates the payment page
414 2011-03-21 02:06:11 <luke-jr> joepie91: that means the server needs to make a request
415 2011-03-21 02:06:13 <joepie91> and gets back an url
416 2011-03-21 02:06:16 <[Tycho]> But something is created after browser request ?
417 2011-03-21 02:06:19 <joepie91> then the browser just has to go there
418 2011-03-21 02:06:21 <joepie91> no
419 2011-03-21 02:06:25 <joepie91> the browser does not create anything
420 2011-03-21 02:06:27 <luke-jr> joepie91: that means the server needs to make a request
421 2011-03-21 02:06:31 <joepie91> yes, it does
422 2011-03-21 02:06:36 <luke-jr> that's ridiculous
423 2011-03-21 02:06:42 <joepie91> as pretty much any webshop does.
424 2011-03-21 02:06:45 <joepie91> can do*
425 2011-03-21 02:06:47 <luke-jr> that's far more complex than PayPal and more expensive
426 2011-03-21 02:06:57 <joepie91> for gods sake, how would that be more expensive?
427 2011-03-21 02:06:58 <luke-jr> and SLOWER
428 2011-03-21 02:07:06 <luke-jr> joepie91: it requires a scriptable server
429 2011-03-21 02:07:09 <luke-jr> instead of static pages
430 2011-03-21 02:07:10 <[Tycho]> Why webshop should make any requests ?
431 2011-03-21 02:07:12 <joepie91> luke-jr
432 2011-03-21 02:07:19 <joepie91> sigh
433 2011-03-21 02:07:20 <joepie91> I give up
434 2011-03-21 02:07:21 <joepie91> srsly
435 2011-03-21 02:07:25 <luke-jr> static pages are far cheaper to host
436 2011-03-21 02:07:41 <luke-jr> and doable by ordinary people
437 2011-03-21 02:07:48 <joepie91> the address is not generated at the moment you send the api call
438 2011-03-21 02:07:53 <joepie91> just the database entry is made at that point
439 2011-03-21 02:07:57 <joepie91> when you actually visit the payment page
440 2011-03-21 02:07:57 <luke-jr> &&.
441 2011-03-21 02:08:01 <joepie91> the address is generated
442 2011-03-21 02:08:03 <joepie91> the btc address
443 2011-03-21 02:08:09 <luke-jr> THEN THE PAYMENT PAGE MUST BE A POST
444 2011-03-21 02:08:10 <joepie91> by the payment page itself
445 2011-03-21 02:08:13 <joepie91> no, it does not
446 2011-03-21 02:08:16 <luke-jr> yes it does
447 2011-03-21 02:08:18 <joepie91> it gets its data from the database
448 2011-03-21 02:08:29 <joepie91> then tell me, what data has to be POSTed?
449 2011-03-21 02:08:41 <luke-jr> it can be a POST without any data
450 2011-03-21 02:08:46 <luke-jr> but it still needs to be a POST
451 2011-03-21 02:08:49 <joepie91> then why the fuck would you do a post?
452 2011-03-21 02:08:50 <devrandom> luke-jr - read http://www.cs.tut.fi/~jkorpela/forms/methods.html
453 2011-03-21 02:08:50 <luke-jr> because you're doing something
454 2011-03-21 02:08:53 <joepie91> goddamnit
455 2011-03-21 02:08:54 <joepie91> seriously
456 2011-03-21 02:08:56 <joepie91> lrn2http
457 2011-03-21 02:08:56 <theymos> That seems reasonable. Just making one database entry isn't a problem. It's the repeatability of GET requests that causes problems.
458 2011-03-21 02:09:00 <[Tycho]> joepie91, just one question - will something bad happen if some script will request GET 100000 times in a row ?
459 2011-03-21 02:09:01 <luke-jr> joepie91: U
460 2011-03-21 02:09:20 <joepie91> [Tycho] yes, the IP will temporarily get blocked
461 2011-03-21 02:09:27 <joepie91> the request IP that is
462 2011-03-21 02:09:32 <luke-jr> devrandom: you're linking the wrong person
463 2011-03-21 02:09:49 <devrandom> luke-jr it *is* for you
464 2011-03-21 02:10:06 <luke-jr> joepie91: if your page does any action/change, it must use POST period
465 2011-03-21 02:10:13 <joepie91> it must not.
466 2011-03-21 02:10:13 <luke-jr> devrandom: no, because I agree with the page
467 2011-03-21 02:10:20 <joepie91> there is no technical requirement for that whatsoever
468 2011-03-21 02:10:21 <[Tycho]> joepie91, why don't you think that it will be easier without additional requests ?
469 2011-03-21 02:10:26 <devrandom> GET vs POST is about idempotent actions, not about whether there are changes made
470 2011-03-21 02:10:29 <luke-jr> joepie91: there is a STANDARDS requirement
471 2011-03-21 02:10:32 <joepie91> lol
472 2011-03-21 02:10:38 <luke-jr> devrandom: read the definition of idempotent
473 2011-03-21 02:10:41 <joepie91> let's bitch about semantics and make some horribly bloated errorprone code
474 2011-03-21 02:10:43 <joepie91> instead of an url
475 2011-03-21 02:10:53 <joepie91> have you ever ordered through for example dynadot?
476 2011-03-21 02:10:53 <luke-jr> joepie91: standards exist for a reason
477 2011-03-21 02:11:00 <devrandom> idempotent means that two actions are equivalent to one
478 2011-03-21 02:11:12 <luke-jr> devrandom: no
479 2011-03-21 02:11:14 <joepie91> if you ever ordered from dynadot and paid through paypal, you know exactly what this POST bullshit leads to
480 2011-03-21 02:11:20 <luke-jr> joepie91: POST works fine
481 2011-03-21 02:11:23 <devrandom> if he is generating just one address no matter how many hits, then GET is fine
482 2011-03-21 02:11:26 <joepie91> they want to redirect you automatically to the payment page on paypal
483 2011-03-21 02:11:27 <joepie91> BUT
484 2011-03-21 02:11:28 <joepie91> to do so
485 2011-03-21 02:11:31 <joepie91> they must submit a post request
486 2011-03-21 02:11:38 <joepie91> therefore they make a blank page with an invisible form
487 2011-03-21 02:11:43 <joepie91> and a javascript that automatically submits the form
488 2011-03-21 02:11:49 <joepie91> because there is no other fucking way
489 2011-03-21 02:11:55 <joepie91> to make someone pay through paypal
490 2011-03-21 02:12:07 <[Tycho]> You can just create a button for POST request.
491 2011-03-21 02:12:15 <joepie91> yes, and that is exactly the issue
492 2011-03-21 02:12:27 <joepie91> k, fine, write me off as an idiot
493 2011-03-21 02:12:53 <[Tycho]> I missed the start of this discussion, but looks wrong anyway.
494 2011-03-21 02:13:15 <luke-jr> joepie91: it is perfectly legal for a browser to hit your URI 1000 times if it is GET
495 2011-03-21 02:13:28 <joepie91> exact same for post.
496 2011-03-21 02:13:30 <luke-jr> joepie91: not to mention Google doing it for indexing
497 2011-03-21 02:13:32 <luke-jr> joepie91: no
498 2011-03-21 02:13:34 <[Tycho]> It will be rather web-spider :)
499 2011-03-21 02:13:38 <joepie91> lmfao
500 2011-03-21 02:13:43 <luke-jr> joepie91: POST can only be performed once
501 2011-03-21 02:13:47 <joepie91> lolno
502 2011-03-21 02:13:52 <[Tycho]> lolyes
503 2011-03-21 02:13:53 <theymos> GETs can be prefectched, which can often cause problems.
504 2011-03-21 02:13:57 <devrandom> joepie91 - he is right on that one
505 2011-03-21 02:14:01 <luke-jr> that's why when you go to reload a POSTed page, it warns you it will resubmit
506 2011-03-21 02:14:10 <devrandom> POST must only be redone with user approval
507 2011-03-21 02:14:18 <joepie91> luke-jr: the point is that it doesn't matter if the user goes to the same page twice
508 2011-03-21 02:14:23 <joepie91> no new database entries are made
509 2011-03-21 02:14:49 <luke-jr> http://www.htmlhelp.com/faq/cgifaq.2.html#9
510 2011-03-21 02:15:00 <joepie91> 1. webshop makes request and creates payment request / vendor uses form and creates payment request (for static sites)
511 2011-03-21 02:15:01 <joepie91> 4. customer goes to url, makes payment
512 2011-03-21 02:15:07 <[Tycho]> luke-jr, you shouldn't worry. Merchants will choose right service anyway :)
513 2011-03-21 02:15:13 <luke-jr> [Tycho]: good point
514 2011-03-21 02:15:33 <joepie91> luke-jr: where did you get the idea that the get request makes a database entry?
515 2011-03-21 02:15:50 <devrandom> joepie91 - if the browser prefetches that url, will anything unexpected happen from the point of view of the user?
516 2011-03-21 02:15:52 <luke-jr> joepie91: you said it generates an address.
517 2011-03-21 02:15:59 <joepie91> devrandom: nope
518 2011-03-21 02:16:02 <joepie91> luke-jr: yup.
519 2011-03-21 02:16:06 <joepie91> so?
520 2011-03-21 02:16:09 <devrandom> then GET is fine...
521 2011-03-21 02:16:09 <luke-jr> joepie91: that's the same thing
522 2011-03-21 02:16:09 <[Tycho]> joepie91, "vendor uses form and creates payment request" - ??ach time when something is going to be sold or once per item type ?
523 2011-03-21 02:16:50 <joepie91> [Tycho]: that depends. if the vendor wants to make callbacks (which indicates he already would have a scripting language available to him) then he should make a new payment request for every order
524 2011-03-21 02:17:05 <joepie91> if he just wants to send out an email with for example a download link, he can use the same payment page for every customer
525 2011-03-21 02:17:12 <luke-jr> joepie91: it doesn't indicate that.
526 2011-03-21 02:17:19 <joepie91> it doesn't indicate what?
527 2011-03-21 02:17:32 <luke-jr> his callbacks might be on another more expensive server than his storefront
528 2011-03-21 02:17:53 <joepie91> I have yet to see the first webshop that uses a non-scripting frontend and a scripting backend
529 2011-03-21 02:18:17 <joepie91> if a store has enough money to set up two servers in the first place
530 2011-03-21 02:18:25 <joepie91> he won't be having static hosting without any scripting
531 2011-03-21 02:18:27 <luke-jr> shared hosting.
532 2011-03-21 02:18:32 <joepie91> yes, what's with it?
533 2011-03-21 02:18:42 <joepie91> I can make an API call from my shared host just fine.
534 2011-03-21 02:18:53 <luke-jr> and even if he has 2 servers, enough traffic could possibly require static pages everywhere possible
535 2011-03-21 02:19:08 <joepie91> link me to one webshop who does that.
536 2011-03-21 02:19:25 <luke-jr> they wouldn't be open source most likely
537 2011-03-21 02:19:30 <joepie91> I didn't say that.
538 2011-03-21 02:19:48 <luke-jr> it's impossible to tell without source
539 2011-03-21 02:19:51 <joepie91> (and seriously, there are little cases where there would be SO much traffic and SO little budget that such an odd and impractical setup would be warranted)
540 2011-03-21 02:19:52 <joepie91> lolno
541 2011-03-21 02:19:56 <joepie91> that's bullshit.
542 2011-03-21 02:20:20 <joepie91> from extension + page behaviour (for example, are there dynamic elements) you can perfectly fine derive if it has a static frontend
543 2011-03-21 02:20:42 <joepie91> you are thinking from some completely impractical server setup that I have _never_ seen being used
544 2011-03-21 02:21:05 <joepie91> well, maybe there will be two stores on this globe who can't use the mentioned system, as opposed to tens of thousands who can (if not more)
545 2011-03-21 02:21:43 <[Tycho]> I don't think that it's really needed to support static-only sites, but still don't see why should merchant make API requests.
546 2011-03-21 02:22:04 <[Tycho]> Everything can be done stateless.
547 2011-03-21 02:22:15 <joepie91> because that would be the only way to create new payment requests without any risk of tampering with it.
548 2011-03-21 02:22:33 <joepie91> you can't make the pay button tell the gateway what the finally-to-be-sent email is
549 2011-03-21 02:22:33 <[Tycho]> What tampering do you mean ?
550 2011-03-21 02:22:48 <joepie91> assume I'd use the POST button to tell the gateway how the request should be made
551 2011-03-21 02:22:56 <joepie91> it would have to include the recipient address
552 2011-03-21 02:22:57 <joepie91> the amount
553 2011-03-21 02:23:03 <joepie91> and the callback or email that has to be sent
554 2011-03-21 02:23:23 <joepie91> would be quite easy to figure out what the final email to be sent is then?
555 2011-03-21 02:23:27 <[Tycho]> It should include only one magic number. Or merchant code + amount.
556 2011-03-21 02:23:28 <joepie91> without actually paying
557 2011-03-21 02:23:38 <joepie91> or as is seen with the old paypal trick
558 2011-03-21 02:23:39 <joepie91> changing to 1 cent
559 2011-03-21 02:23:54 <joepie91> yes, and that would require either the merchant submitting a form
560 2011-03-21 02:23:55 <joepie91> or an API call
561 2011-03-21 02:23:56 <[Tycho]> Merchant will check how much money was paid anyway.
562 2011-03-21 02:24:02 <joepie91> not if it's automated.
563 2011-03-21 02:24:10 <joepie91> that is the entire point of this gateway
564 2011-03-21 02:24:12 <joepie91> to automate it
565 2011-03-21 02:24:13 <[Tycho]> Submitting a form is fine.
566 2011-03-21 02:24:16 <joepie91> well
567 2011-03-21 02:24:18 <joepie91> then what is the issue?
568 2011-03-21 02:24:35 <joepie91> that is exactly the system I proposed, only is ALL data stored on the server, not just the amount and merchant code
569 2011-03-21 02:24:37 <[Tycho]> User should submit a form, not the site.
570 2011-03-21 02:24:41 <joepie91> it would require the same degree of interaction
571 2011-03-21 02:24:45 <joepie91> ..
572 2011-03-21 02:25:00 <joepie91> the merchant still has to create the payment button, right?
573 2011-03-21 02:25:06 <[Tycho]> Yes.
574 2011-03-21 02:25:08 <joepie91> exactly.
575 2011-03-21 02:25:23 <joepie91> and that is _exactly_ what the API / form (depending on what kind of site) does in my idea
576 2011-03-21 02:25:36 <joepie91> either he uses the API if he has scripting
577 2011-03-21 02:25:43 <joepie91> or he uses the form on paybtc.com itself to create a payment request
578 2011-03-21 02:25:59 <[Tycho]> No, you said that merchant's SERVER should make a request to your API
579 2011-03-21 02:26:06 <LobsterMan> ;;bc,stats
580 2011-03-21 02:26:12 <joepie91> <joepie91>1. webshop makes request and creates payment request / vendor uses form and creates payment request (for static sites)
581 2011-03-21 02:26:15 <joepie91> ^
582 2011-03-21 02:26:16 <gribble> Current Blocks: 114335 | Current Difficulty: 76193.9710474 | Next Difficulty At Block: 114911 | Next Difficulty In: 576 blocks | Next Difficulty In About: 4 days, 16 hours, 9 minutes, and 36 seconds | Next Difficulty Estimate: 66275.25708289
583 2011-03-21 02:26:33 <joepie91> as I have said a few times
584 2011-03-21 02:26:35 <joepie91> either the API
585 2011-03-21 02:26:37 <joepie91> or the on-site form
586 2011-03-21 02:28:34 <joepie91> so again, I fail to see where a POST request would come into the picture anywhere here, except for the (optional) API call
587 2011-03-21 02:28:49 <joepie91> (or the form that the merchant fills in)
588 2011-03-21 02:29:37 <[Tycho]> You create a database entry on GET. That's not good.
589 2011-03-21 02:29:44 <joepie91> no, I do not
590 2011-03-21 02:29:46 <[Tycho]> If it's not the counter :)
591 2011-03-21 02:29:53 <joepie91> for the umpteenth time
592 2011-03-21 02:29:58 <joepie91> the database request is created
593 2011-03-21 02:30:00 <joepie91> err
594 2011-03-21 02:30:01 <joepie91> entry*
595 2011-03-21 02:30:05 <joepie91> when the API call is made
596 2011-03-21 02:30:08 <joepie91> or when the merchant fills in the form
597 2011-03-21 02:30:17 <joepie91> which are both POST requests
598 2011-03-21 02:30:21 <joepie91> when the BUYER follows the payment link
599 2011-03-21 02:30:27 <joepie91> he does a GET request
600 2011-03-21 02:30:30 <[Tycho]> Ok.
601 2011-03-21 02:30:36 <joepie91> and the resulting page will provide him with a bitcoin address that is generated at that time
602 2011-03-21 02:30:47 <joepie91> getting the payment data from the db
603 2011-03-21 02:31:09 <[Tycho]> The bitcoin addres will be the same if he refreshes that page ?
604 2011-03-21 02:31:24 <joepie91> no, that would make payments for static sites impossible
605 2011-03-21 02:31:52 <joepie91> who cannot generate a new payment request every time a customer wants to buy
606 2011-03-21 02:31:52 <[Tycho]> So if he clicks on "Refresh", he will got another bitcoin address ?
607 2011-03-21 02:31:54 <joepie91> yes
608 2011-03-21 02:32:04 <joepie91> the point is that that does not matter
609 2011-03-21 02:32:05 <[Tycho]> With GET ? That's so wrong.
610 2011-03-21 02:32:10 <joepie91> why would taht be wrong?
611 2011-03-21 02:32:15 <joepie91> and not just "because standards"
612 2011-03-21 02:32:23 <[Tycho]> You should not generate anything on GET.
613 2011-03-21 02:32:24 <joepie91> because that is a bullshit reason if standards impair usability
614 2011-03-21 02:32:36 <joepie91> it's not even standards
615 2011-03-21 02:32:41 <luke-jr> joepie91: never heard of PGP?
616 2011-03-21 02:32:41 <[Tycho]> Standarts are here to IMPROVE usability.
617 2011-03-21 02:32:42 <joepie91> it's a recommendation, it's semantics
618 2011-03-21 02:32:57 <joepie91> exactly, they are here to improve them
619 2011-03-21 02:33:07 <joepie91> so if they don't, it's dumb to go nitpicking over the type of request
620 2011-03-21 02:33:12 <joepie91> if it's clear one works better than the other
621 2011-03-21 02:33:35 <[Tycho]> joepie91, how many times user is allowed to load that page ?
622 2011-03-21 02:33:38 <luke-jr> joepie91: there is NOTHING wrong with POST
623 2011-03-21 02:33:39 <joepie91> in fact, whereas a simple get request will work everywhere, a post button will give issues with various mobile browsers\n347457
624 2011-03-21 02:33:52 <joepie91> [Tycho]: I'd say 3 times for the same payment url, after that he gets a temp IP block
625 2011-03-21 02:34:13 <luke-jr> joepie91: pretty sure my browser will send stuff 3x in some cases
626 2011-03-21 02:34:19 <luke-jr> GETs that is
627 2011-03-21 02:34:25 <joepie91> luke-jr: there is nothing wrong with POST itself, there is something wrong with stuffing POST buttons and invisible forms on pages
628 2011-03-21 02:34:28 <joepie91> to go to another page
629 2011-03-21 02:34:33 <luke-jr> no, there isn't.
630 2011-03-21 02:34:37 <joepie91> there is.
631 2011-03-21 02:34:38 <luke-jr> that's perfectly valid
632 2011-03-21 02:34:39 <[Tycho]> Pretty sure someone will refresh it, may be because of internet lag or something else.
633 2011-03-21 02:34:50 <joepie91> [Tycho]: they will, of course
634 2011-03-21 02:34:53 <joepie91> which is why it's on 3
635 2011-03-21 02:34:55 <joepie91> or maybe even more
636 2011-03-21 02:34:57 <luke-jr> idiot
637 2011-03-21 02:35:33 <joepie91> know what?
638 2011-03-21 02:35:35 <joepie91> we'll see if it works.
639 2011-03-21 02:37:08 <joepie91> when it;s running, send some 0.01btc payments to yourself
640 2011-03-21 02:37:10 <joepie91> so you see it works
641 2011-03-21 02:37:14 <joepie91> there are no transaction costs, anyway
642 2011-03-21 02:38:43 <genjix> im really happy since i've been running lots of Q&A for the last few days on my exchange trying to break it
643 2011-03-21 02:39:24 <genjix> and the only problems are either not my fault (google/yahoo) or due to user error (me entering incorrect SQL commands to simulate a deposit)
644 2011-03-21 02:39:35 <genjix> bodes well.
645 2011-03-21 02:43:05 <joepie91> night all
646 2011-03-21 02:56:38 <genjix> Blitzboom: i got the fundz :)
647 2011-03-21 02:57:11 <Blitzboom> good to hear!
648 2011-03-21 03:02:46 <naskashy> but the GUI don't opened on fluxbox
649 2011-03-21 03:03:33 <naskashy> just under gnome
650 2011-03-21 03:07:30 <Aciid> genjix: http://i.imgur.com/ZdkAD.png
651 2011-03-21 03:07:37 <Aciid> it's shit like this with the current ISP
652 2011-03-21 03:08:13 <Aciid> in the middle of a 10GB transfer it may come up with those shits every 6hours or so, for no apparent reason
653 2011-03-21 03:09:09 <Aciid> *Server ISP
654 2011-03-21 03:09:55 <Aciid> ridicilous
655 2011-03-21 03:12:15 <Aciid> ah got the the userdir slashed out now
656 2011-03-21 03:12:32 <Aciid> I didn't have the php5.conf loaded which told about the userdir php_engine
657 2011-03-21 03:17:09 <Aciid> hmm I wonder is it possible to sudo only some commands
658 2011-03-21 03:17:40 <Aciid> excuse my paranoia
659 2011-03-21 03:18:51 <genjix> Aciid: i dont get it
660 2011-03-21 03:19:20 <genjix> also: not linux :p
661 2011-03-21 03:19:47 <Aciid> genjix: pondering about the possibility to add a user which could sudo. without giving me a headache
662 2011-03-21 03:19:57 <Aciid> I think I got it
663 2011-03-21 03:20:04 <Aciid> I must try
664 2011-03-21 03:23:34 <Aciid> genjix: my god it works
665 2011-03-21 03:24:48 <Aciid> if I propose you to try to live with me , should you accept this mere apology. for all the shenningans I've caused?
666 2011-03-21 03:24:53 <Aciid> genjix:
667 2011-03-21 03:25:17 <Kiba> finally
668 2011-03-21 03:25:21 <Kiba> I can rest now
669 2011-03-21 03:25:24 <luke-jr> genjix: I think I finally fixed all the bugs with confirmation counting XD
670 2011-03-21 03:25:56 <luke-jr> testing tongiht before I commit
671 2011-03-21 03:27:35 <genjix> luke-jr: cool
672 2011-03-21 03:27:43 <genjix> been working on other stuff for time being :p
673 2011-03-21 03:30:34 <Kiba> http://bitcoinweekly.com finally have content!
674 2011-03-21 03:39:25 <naskashy> Nice article !
675 2011-03-21 03:39:28 <naskashy> ;)
676 2011-03-21 03:46:52 <[Tycho]> ;;bc,gen 290000
677 2011-03-21 03:47:17 <[Tycho]> Kiba, so where is the content ?
678 2011-03-21 03:47:22 <gribble> The expected generation output, at 290000 Khps, given current difficulty of 76193.9710474 , is 3.82825867141 BTC per day and 0.159510777975 BTC per hour.
679 2011-03-21 03:48:56 <Kiba> [Tycho]: it's on http://bitcoinweekly.com
680 2011-03-21 03:49:02 <[Tycho]> Oh, i'm supposed to click on the title..
681 2011-03-21 04:03:13 <Kiba> somebody is polluting witcoin with sexting stuff
682 2011-03-21 04:03:22 <Kiba> but I am gald that they have to pay
683 2011-03-21 04:03:25 <Kiba> for their pollution
684 2011-03-21 04:04:25 <devrandom> ArtForz - do you know why stuff like OP_XOR was disabled?  I can't think how that can be used to cause a crash
685 2011-03-21 04:04:47 <devrandom> OP_MUL and OP_CAT are pretty obvious
686 2011-03-21 04:05:16 <theymos> Satoshi probably became paranoid.
687 2011-03-21 04:05:40 <[Tycho]> Kiba, it's xellister
688 2011-03-21 04:06:14 <Kiba> what make you think that?
689 2011-03-21 04:06:45 <devrandom> theymos - I see
690 2011-03-21 04:23:57 <theymos> Gavin did yet another intervew: http://tinyurl.com/482qpgr
691 2011-03-21 04:26:51 <Kiba> go Gavin!
692 2011-03-21 04:29:11 <naskashy> Lately, what is the total number of blocks ?
693 2011-03-21 04:29:25 <theymos> ;;bc,stats
694 2011-03-21 04:29:41 <gribble> Current Blocks: 114350 | Current Difficulty: 76193.9710474 | Next Difficulty At Block: 114911 | Next Difficulty In: 561 blocks | Next Difficulty In About: 4 days, 12 hours, 18 minutes, and 15 seconds | Next Difficulty Estimate: 66456.77131848
695 2011-03-21 04:30:44 <naskashy> thanks !! i already got 114223
696 2011-03-21 04:30:52 <naskashy> :)
697 2011-03-21 04:31:43 <theymos> Did you solve that display issue you were having?
698 2011-03-21 04:32:39 <Kiba> hmm
699 2011-03-21 04:32:41 <Kiba> witcoin is slow loading
700 2011-03-21 04:33:06 <naskashy> nope, i'm on gnome
701 2011-03-21 04:33:17 <Kiba> had anybody been able to load witcoin?
702 2011-03-21 04:33:28 <theymos> naskashy: You should post about it on the forum.
703 2011-03-21 04:33:32 <ArtForz> devrandom: nope
704 2011-03-21 04:34:30 <Kiba> well, it was down
705 2011-03-21 04:35:04 <noagendamarket> maybe it was your connection :)
706 2011-03-21 04:35:38 <Kiba> wee!
707 2011-03-21 04:35:41 <Kiba> I made money!
708 2011-03-21 04:35:43 <theymos> It was also down for me. It looked like someone was changing something, since I got a blank page.
709 2011-03-21 04:36:16 <noagendamarket> you dos'd it lol
710 2011-03-21 04:36:25 <Kiba> it seem that all my posts made me money now :D
711 2011-03-21 04:36:39 <Kiba> witcoin posting is a profitable enterprise for kiba
712 2011-03-21 04:36:45 <Kiba> virtually all his posts made him money :D
713 2011-03-21 04:36:49 <theymos> Witcoin should hide NSFW from the default homepage.
714 2011-03-21 04:37:17 <noagendamarket> yeah we are tyalking about that
715 2011-03-21 04:37:21 <Kiba> I don't like the sexting content and wonders who would vote for them
716 2011-03-21 04:37:30 <Kiba> poor quality
717 2011-03-21 04:38:31 <noagendamarket> I dont see anyone voting for them
718 2011-03-21 04:39:04 <theymos> Some of them have 3 votes.
719 2011-03-21 04:39:46 <theymos> He's probably using alt accounts. The quality is low.
720 2011-03-21 04:39:59 <Kiba> not sexy at all :(
721 2011-03-21 04:40:07 <Kiba> blah, let him spend his money
722 2011-03-21 04:40:43 <noagendamarket> I think someone had a domain that received lots of text messages like that by mistake
723 2011-03-21 04:41:07 <Kiba> well, spammers don't get a free lunch right?
724 2011-03-21 04:41:26 <noagendamarket> no they dont
725 2011-03-21 04:41:31 <noagendamarket> its posted on a category that costs .01
726 2011-03-21 04:41:43 <noagendamarket> so must have money to burn imo
727 2011-03-21 04:43:28 <Kiba> noagendamarket: hopefully nobody upvote his dreadful content
728 2011-03-21 04:44:28 <Kiba> night
729 2011-03-21 05:35:15 <euclid> url for witcoin?
730 2011-03-21 05:36:27 <Spenvo> ? www.witcoin.com
731 2011-03-21 05:37:55 <euclid> sorry I'm new
732 2011-03-21 05:38:45 <Spenvo> it's cool!  bitcoin.witcoin.com is a forum for bitcoin topics :)
733 2011-03-21 05:39:44 <euclid> do you mind if I ask some questions?
734 2011-03-21 05:39:52 <Kiba> it's crazy people can post thousand of topics about bitcoin!
735 2011-03-21 05:40:11 <Kiba> it's just a digital currency!
736 2011-03-21 05:40:26 <Kiba> euclid: ask away
737 2011-03-21 05:41:14 <euclid> Is the 5770 the only way to go?
738 2011-03-21 05:41:21 <Kiba> don't know anything about mining
739 2011-03-21 05:41:52 <euclid> how to you post 'thoughts' like that
740 2011-03-21 05:42:25 <euclid> with a * i mean
741 2011-03-21 05:42:36 <euclid> *thinking
742 2011-03-21 05:42:58 <Kiba> /me whateveryouwant
743 2011-03-21 05:44:07 <euclid> Is mtgox the only major exchange?
744 2011-03-21 05:44:40 <Spenvo> it's the biggest, but coinpal is well trusted
745 2011-03-21 05:44:44 <theymos> It's the most major, but there are a few others, such as Bitcoin Central and Bitcoin Market.
746 2011-03-21 05:45:23 <genjix> and soon to be britcoin :)
747 2011-03-21 05:45:49 <genjix> euclid: /me is thinking
748 2011-03-21 05:45:52 <euclid> Is mining profitable using 5770's? if so how many?
749 2011-03-21 05:46:19 <euclid> *dollar signs in his eyes
750 2011-03-21 05:46:22 <euclid> fuck
751 2011-03-21 05:46:32 <ArtForz> 5970 is where it's at
752 2011-03-21 05:47:12 <ArtForz> or undervolted 5870
753 2011-03-21 05:47:19 <genjix> ;;bc,seen MagicalTux
754 2011-03-21 05:47:19 <gribble> Error: "bc,seen" is not a valid command.
755 2011-03-21 05:47:21 <ArtForz> 5770 power efficiency is ~30% worse
756 2011-03-21 05:47:23 <ZenMondo> I think it reaches a point that mining a coin will be more expensive than what you can spend it on.
757 2011-03-21 05:47:24 <genjix> ;;seen MagicalTux
758 2011-03-21 05:47:24 <gribble> MagicalTux was last seen in #bitcoin-dev 1 day, 3 hours, 49 minutes, and 23 seconds ago: <MagicalTux> (anyway cya)
759 2011-03-21 05:47:38 <ArtForz> ZenMondo: that would be... weird
760 2011-03-21 05:47:58 <MagicalTux> genjix: I'm here
761 2011-03-21 05:47:59 <MagicalTux> xD
762 2011-03-21 05:48:23 <genjix> oh hey :p any chance to get setup on the vps yet?
763 2011-03-21 05:48:35 <genjix> dissipate is poking me on PM
764 2011-03-21 05:49:10 <euclid> I did some math when the price was around 0.90 and it would take 4 5770's and about 3 months to make back your investment, if the value goes up past a dollar all coins mined now will be more valuable than coins generated in the future, correct? Or is my logic fanciful
765 2011-03-21 05:49:38 <genjix> your logic is fanciful but im not sure if it's wrong.
766 2011-03-21 05:49:44 <ArtForz> one coin is worth as much as any other coin
767 2011-03-21 05:49:58 <ArtForz> doesnt matter if it was mined today or 2 years ago
768 2011-03-21 05:50:10 <euclid> in terms of electricity and hardware I think not...
769 2011-03-21 05:50:26 <MagicalTux> er ?
770 2011-03-21 05:50:33 <MagicalTux> genjix: you need anything else?
771 2011-03-21 05:50:36 <ArtForz> huh?
772 2011-03-21 05:50:43 <genjix> MagicalTux: i never got the login details or anything.
773 2011-03-21 05:51:01 <MagicalTux> genjix: I told you, reinstall your vps and the root password will be the password displayed on the interface for vnc
774 2011-03-21 05:51:02 <euclid> If difficulty increases coins become harder to come by... or am I missing something, I am new
775 2011-03-21 05:51:09 <ArtForz> well, that doesnt change the value of the coin, just the cost to generate it
776 2011-03-21 05:51:14 <ArtForz> sure
777 2011-03-21 05:51:29 <ZenMondo> Anyone here besides [Noodles] on Second Life?
778 2011-03-21 05:51:49 <euclid> But thats my point, gold found on the ground is more valuable than gold 1000ft below the ground
779 2011-03-21 05:51:58 <ArtForz> no, it's not
780 2011-03-21 05:51:59 <euclid> even though gold = gold
781 2011-03-21 05:52:15 <genjix> MagicalTux: there's nothing under hosting...
782 2011-03-21 05:52:22 <MagicalTux> genjix: login to autovps.net with your kalyhost login then
783 2011-03-21 05:52:25 <ArtForz> it's still the same value
784 2011-03-21 05:52:27 <MagicalTux> (will be easier)
785 2011-03-21 05:52:31 <genjix> ok thanks
786 2011-03-21 05:52:33 <ArtForz> you keep mixing up cost and value
787 2011-03-21 05:52:33 <ZenMondo> its not that its worth more its that it cost less to obtain.
788 2011-03-21 05:52:55 <euclid> I though they were directly related
789 2011-03-21 05:53:01 <ArtForz> errr... no
790 2011-03-21 05:53:22 <genjix> MagicalTux: how can I get Ubuntu btw ? :p
791 2011-03-21 05:53:33 <genjix> debian is ancient.
792 2011-03-21 05:53:43 <MagicalTux> genjix: use virtual cd loader and install via vnc
793 2011-03-21 05:53:48 <genjix> ok
794 2011-03-21 05:54:06 <MagicalTux> note that once you change the cd you may have to switch off then start again the vps
795 2011-03-21 05:54:08 <euclid> Rather a mining operation is more profitable when the gold is easier to obtain
796 2011-03-21 05:54:11 <genjix> oh sweet
797 2011-03-21 05:54:16 <ArtForz> yep
798 2011-03-21 05:54:20 <MagicalTux> (use the power menu for that)
799 2011-03-21 05:54:28 <genjix> ok autovps is really nice
800 2011-03-21 05:54:36 <genjix> how comes you don't have this by default?
801 2011-03-21 05:55:52 <euclid> ArtForz, you mean (hardware + electricity) /Mhash
802 2011-03-21 05:55:59 <ArtForz> no
803 2011-03-21 05:56:07 <euclid> ArtForz, o
804 2011-03-21 05:56:15 <ArtForz> generated btc * market price / mhash
805 2011-03-21 05:57:02 <ArtForz> imo mining is still way to profitable to level off soon
806 2011-03-21 05:57:09 <euclid> ArtForz, I don't understand bitcoin economics well enough...
807 2011-03-21 05:57:22 <ArtForz> ;;bc,gen 650000
808 2011-03-21 05:57:23 <gribble> The expected generation output, at 650000 Khps, given current difficulty of 76193.9710474 , is 8.58057978074 BTC per day and 0.357524157531 BTC per hour.
809 2011-03-21 05:57:46 <euclid> haha thats so cool
810 2011-03-21 05:57:48 <ArtForz> thats one 5970, power cost... about $0.80/day at $0.1/kWh
811 2011-03-21 05:58:46 <euclid> hmm
812 2011-03-21 05:58:51 <ArtForz> so power cost is... about $0.093/btc
813 2011-03-21 05:59:01 <ArtForz> ;;bc,mtgox
814 2011-03-21 05:59:03 <gribble> {"ticker":{"high":0.79,"low":0.7405,"vol":4467,"buy":0.7407,"sell":0.7675,"last":0.7724}}
815 2011-03-21 05:59:13 <euclid> that makes sense
816 2011-03-21 05:59:20 <ArtForz> while market price is north of $0.74/btc
817 2011-03-21 06:00:17 <euclid> What do you think the price will be in 4 months say? 1 year ? 4 years?
818 2011-03-21 06:00:30 <euclid> upwards of a dollar?
819 2011-03-21 06:01:02 <euclid> I mean I know this is speculation but for shits and giggles
820 2011-03-21 06:01:06 <ArtForz> no clue
821 2011-03-21 06:01:49 <ArtForz> as long as a decent majority of miners uses GPUs and pays for their power, I'm not concerned
822 2011-03-21 06:01:57 <euclid> fair enough
823 2011-03-21 06:02:13 <theymos> That's pretty profitable. I'm surprised difficulty is going down.
824 2011-03-21 06:02:22 <ArtForz> theymos: yeah, kinda surprising
825 2011-03-21 06:02:47 <Blitzboom> psychology
826 2011-03-21 06:02:53 <ArtForz> ofc I'm using ASICs that use < 10% of the power of GPUs for the same hashrate
827 2011-03-21 06:03:21 <euclid> Where can you get ASICs for bitcoin?
828 2011-03-21 06:03:31 <ArtForz> I had em manufactured
829 2011-03-21 06:03:37 <euclid> lol your the man!
830 2011-03-21 06:04:06 <euclid> Would you happen to be selling said money machine?
831 2011-03-21 06:04:09 <ArtForz> ~19.2Ghps online, another 19.2Ghps arriving in ~7wks
832 2011-03-21 06:05:10 <theymos> Are you going to buy more of the same type after you get that?
833 2011-03-21 06:05:24 <euclid> are you an EE?
834 2011-03-21 06:05:27 <ArtForz> yep
835 2011-03-21 06:05:37 <ArtForz> probably not, I dont like the cooling issues
836 2011-03-21 06:05:51 <euclid> ArtForz, my hat is off to you
837 2011-03-21 06:06:19 <Blitzboom> EE=?
838 2011-03-21 06:06:32 <euclid> Electrical Engineer
839 2011-03-21 06:06:34 <ArtForz> I'm hoping xilinx gets their ass off the ground, artix7 FPGAs look *very* interesting
840 2011-03-21 06:07:59 <ArtForz> the 165 should be big enough to fit a complete pipelined bitcoinhash engine
841 2011-03-21 06:08:36 <ArtForz> err, the 175
842 2011-03-21 06:08:56 <ArtForz> if they're at least as fast as spartan6s, thats ~150Mhps
843 2011-03-21 06:09:12 <euclid> So GPU mining will soon be a thing of the past?
844 2011-03-21 06:09:45 <ArtForz> biggest spartan6 = lx150 is just a bit too small to fit a complete bitcoinhash engine
845 2011-03-21 06:09:55 <ArtForz> and virtexes are WAY too expensive
846 2011-03-21 06:11:19 <ArtForz> if XC7A175T costs less than $200, it should be a pretty good deal
847 2011-03-21 06:11:19 <euclid> Who makes the bitcoinhash engine?
848 2011-03-21 06:11:40 <ArtForz> erm, I did
849 2011-03-21 06:11:58 <euclid> I presume that its not opensource
850 2011-03-21 06:12:04 <ArtForz> nope
851 2011-03-21 06:12:14 <euclid> was it tough to make>
852 2011-03-21 06:12:21 <ArtForz> not really
853 2011-03-21 06:12:52 <ArtForz> more like tedious
854 2011-03-21 06:12:56 <euclid> It just computes the hash functions?
855 2011-03-21 06:12:59 <ArtForz> yup
856 2011-03-21 06:13:23 <ArtForz> well, it also does the nonce incrementing and H==0 comparing
857 2011-03-21 06:14:01 <euclid> That would be the proof I'm assuming?
858 2011-03-21 06:14:18 <ArtForz> well, it's difficulty 1 proof of work aka a pool share
859 2011-03-21 06:14:39 <ArtForz> host then checks against real target
860 2011-03-21 06:14:58 <[Tycho]> How do you interface your chips to host?
861 2011-03-21 06:15:03 <euclid> and this goes on until it hits the current difficulty?
862 2011-03-21 06:15:15 <ArtForz> spartan6 w/ full speed USB
863 2011-03-21 06:15:41 <ArtForz> 8 ASICs are connected to a spartan6 lx16
864 2011-03-21 06:16:20 <ArtForz> which just acts as a glorified usb<->custom serial protocol translator
865 2011-03-21 06:16:42 <Diablo-D3> heh
866 2011-03-21 06:16:56 <ArtForz> then 4 of those are connected to a 4-port usb hub
867 2011-03-21 06:19:15 <ArtForz> so each module has 8 ASICs, 4 modules in a 2U, connected to controller PC via single high-speed USB link
868 2011-03-21 06:19:38 <euclid> whats lx16?
869 2011-03-21 06:19:45 <[Tycho]> Will be there a photo available ?
870 2011-03-21 06:19:48 <ArtForz> erm, the model
871 2011-03-21 06:19:53 <ArtForz> nope
872 2011-03-21 06:19:55 <euclid> gotcha
873 2011-03-21 06:20:26 <ArtForz> so each 2U = 32 ASICs @ 200Mhz = 6.4Ghps, consuming ~300W
874 2011-03-21 06:20:34 <ArtForz> well, average
875 2011-03-21 06:20:44 <ArtForz> when ambient temp rises it's closer to 350W
876 2011-03-21 06:20:56 <euclid> whats the price tag on a rig like that?
877 2011-03-21 06:21:17 <ArtForz> compared to GPUs, way too much
878 2011-03-21 06:21:21 <[Tycho]> Why no photos ? :)
879 2011-03-21 06:23:43 <euclid> Are you the only one with this approach ?
880 2011-03-21 06:23:50 <euclid> currently
881 2011-03-21 06:23:51 <ArtForz> dunno, so far looks like it
882 2011-03-21 06:24:24 <lfm> Art is the only one talking about it, if there are more
883 2011-03-21 06:24:50 <euclid> How many people at any given time would you say are mining?
884 2011-03-21 06:24:58 <genjix> over 9000
885 2011-03-21 06:25:08 <ArtForz> 42
886 2011-03-21 06:25:12 <ArtForz> iirc I was also the first guy with a GPU farm
887 2011-03-21 06:25:21 <euclid> lol
888 2011-03-21 06:25:45 <euclid> the answer to life and how many penises i have
889 2011-03-21 06:26:46 <euclid> you guys ever hear of reprap?
890 2011-03-21 06:26:52 <ArtForz> yeah
891 2011-03-21 06:26:59 <euclid> into it?
892 2011-03-21 06:27:02 <ArtForz> nah
893 2011-03-21 06:27:35 <genjix> is that good?
894 2011-03-21 06:27:53 <genjix> seems to me like a big gluegun
895 2011-03-21 06:27:54 <euclid> I want to try building one, think people would dig a store that sold parts that accepted bitcoin
896 2011-03-21 06:28:03 <ArtForz> imo plastic extrusion = toy
897 2011-03-21 06:28:06 <genjix> how do they protect against deformation?
898 2011-03-21 06:28:12 <euclid> heated bed
899 2011-03-21 06:28:50 <genjix> ?
900 2011-03-21 06:29:05 <genjix> so what happens if i try to manufacture princess leia?