1 2012-03-11 00:00:34 <tcatm> http://188.138.99.157/stuff/qtvert8.png
  2 2012-03-11 00:01:02 <sipa> ;;bc,blocks
  3 2012-03-11 00:01:03 <gribble> 170596
  4 2012-03-11 00:01:15 <mcorlett> sipa: I'm seeing a whole bunch of 11DbExceptions
  5 2012-03-11 00:01:54 <mcorlett> http://pastebin.com/dg1Gxsii
  6 2012-03-11 00:02:08 <sipa> mcorlett: ok, that's what i wanted to see
  7 2012-03-11 00:02:23 <sipa> mcorlett: can you shut down your node, and compile my minireorg branch?
  8 2012-03-11 00:02:37 <sipa> https://github.com/sipa/bitcoin/tree/minireorg
  9 2012-03-11 00:03:07 <mcorlett> Sure thing.
 10 2012-03-11 00:03:43 <sipa> and then run it in the same way (but keep a backup of blk0001.dat and blkindex.dat, if you didn't make those yet)
 11 2012-03-11 00:06:39 <mcorlett> sipa: How do I disable UPNP when compiling, again?
 12 2012-03-11 00:07:04 <sipa> make USE_UPNP=
 13 2012-03-11 00:07:23 <mcorlett> Ok, here we go again.
 14 2012-03-11 00:08:15 <mcorlett> I still have the blockchain that's stuck on my Windows box, is that good enough for a "backup"?
 15 2012-03-11 00:08:21 <sipa> sure
 16 2012-03-11 00:10:13 <mcorlett> Running your branch now.
 17 2012-03-11 00:10:21 <sipa> With -connect= ?
 18 2012-03-11 00:10:44 <mcorlett> Yeah, but I didn't kill the old process nicely it seems.
 19 2012-03-11 00:10:51 <mcorlett> Bitcoin: Unable to bind to port 8333 on this computer.
 20 2012-03-11 00:11:00 <sipa> ah, kill it
 21 2012-03-11 00:11:13 <Diablo-D3> KHEEL EET
 22 2012-03-11 00:12:21 <mcorlett> Trying again!
 23 2012-03-11 00:13:18 <mcorlett> Looks like we've got reorganization.
 24 2012-03-11 00:13:45 <mcorlett> Oh yes!
 25 2012-03-11 00:13:52 <mcorlett> It's happenin'.
 26 2012-03-11 00:14:28 <Samuel> tcatm: How is it coming along?
 27 2012-03-11 00:14:41 <tcatm> http://188.138.99.157/stuff/qtvert8.png :)
 28 2012-03-11 00:15:45 <Diablo-D3> looking kind of osx there
 29 2012-03-11 00:16:43 <Samuel> tcatm: Nice! I think their needs to just be a few more pixels of space between the icon and text. Can you make the corners of the app be rounded?
 30 2012-03-11 00:17:35 <tcatm> I can't add a space there easily. It would require writing some code and manual painting methods.
 31 2012-03-11 00:18:03 <Samuel> oh, ok
 32 2012-03-11 00:18:11 <tcatm> No idea whether I can make them round. My window manager couldn't handle that anyway.
 33 2012-03-11 00:18:16 <sipa> mcorlett: ah, good to hear!
 34 2012-03-11 00:19:25 <Samuel> brb
 35 2012-03-11 00:19:33 <sipa> mcorlett: thanks you!
 36 2012-03-11 00:19:38 <sipa> *thank
 37 2012-03-11 00:20:06 <mcorlett> sipa: No worries, glad I could help.
 38 2012-03-11 00:28:41 <JorgePasada> How big is the current blockchain size wise?
 39 2012-03-11 00:28:59 <Tuxavant> sipa thank you.. confirmations starting to show up. got 2 now
 40 2012-03-11 00:29:21 <Samuel> tcatm: http://cl.ly/1i3o2W393S472s391A02 Can you make these changes?
 41 2012-03-11 00:33:46 <tcatm> Umm, the underlined letters are there for a reason.
 42 2012-03-11 00:34:41 <Samuel> Just seems useless, OS X hasn't had those things in years. I guess its just an appearance  thing
 43 2012-03-11 00:35:46 <tcatm> It's highly useful if you control a computer mostly with a keyboard.
 44 2012-03-11 00:36:21 <Samuel> Well, ok
 45 2012-03-11 00:37:42 <Samuel> tcatm: I'll get more icons for the buttons (copy, add, etc.)
 46 2012-03-11 00:42:36 <tcatm> Where should we show the current balance? Maybe still in the sidebar?
 47 2012-03-11 00:43:14 <Samuel> Can to have it in the top like i have in my screenshot? http://cl.ly/3l3V321y3d0z3S2H2x1e
 48 2012-03-11 00:43:36 <Samuel> Maybe to the right side since there is stuff on the right
 49 2012-03-11 00:44:06 <Samuel> (Where bar with File, Settings, Help)
 50 2012-03-11 00:45:46 <Samuel> The text for the tab that is selected in the sidebar should be bold. Can you do that?
 51 2012-03-11 00:51:48 <tcatm> I can, but then the longest text will be cut-off on the right when it is selected.
 52 2012-03-11 00:54:31 <neofutur> JorgePasada: more than one GB
 53 2012-03-11 00:55:00 <neofutur> -rw-------  1 neofutur neofutur 1.1G 2012-03-10 12:05 blk0001.dat
 54 2012-03-11 00:55:18 <neofutur> reaching 1.5 GB
 55 2012-03-11 00:56:31 <neofutur> imho light clients need some kind of btrusted lockchain provider
 56 2012-03-11 00:59:38 <Samuel_> Hmm, I reloaded the page now it says my name is registered so im Samuel_
 57 2012-03-11 01:00:04 <sipa> your old self is still here
 58 2012-03-11 01:00:06 <neofutur> if you registered you need to identify to use the nickname
 59 2012-03-11 01:00:18 <sipa> the server didn't realize the old one is not connected anymore
 60 2012-03-11 01:00:21 <neofutur> /msg nickserv help ghost
 61 2012-03-11 01:00:37 <neofutur> /msg nickserv help identify
 62 2012-03-11 01:00:53 <neofutur> ah the ghost is gone
 63 2012-03-11 01:01:06 <Samuel> Haha!
 64 2012-03-11 01:01:12 <Samuel_> What?
 65 2012-03-11 01:01:42 <Samuel_> Can I change back to just Samuel?
 66 2012-03-11 01:01:48 <mcorlett> /nick Samuel
 67 2012-03-11 01:02:15 <mcorlett> It's registered by someone, and I take it it's not you.
 68 2012-03-11 01:02:26 <Samuel> its registered?
 69 2012-03-11 01:02:32 <mcorlett> That person can kick you off at will.
 70 2012-03-11 01:02:45 <Samuel> oh wait
 71 2012-03-11 01:02:51 <Samuel> now im back to just Samuel
 72 2012-03-11 01:02:57 <Samuel> weird...
 73 2012-03-11 01:03:35 <Samuel> SO i want to register this so no one else can have it. What do I type for that?
 74 2012-03-11 01:03:51 <mcorlett> You can't do that. As I said, someone already owns that nickname.
 75 2012-03-11 01:04:08 <mcorlett> Choose something more unique.
 76 2012-03-11 01:04:17 <Samuel> Oh, then how am I Samuel right now?
 77 2012-03-11 01:04:39 <mcorlett> The person isn't online right now, and hasn't enabled the option of forcing you to identify.
 78 2012-03-11 01:04:55 <Samuel> oh, I guess I just keep doing this
 79 2012-03-11 01:05:14 <mcorlett> (I have. If you switch to my nickname and fail to identify within 30 seconds, you're kicked off.)
 80 2012-03-11 01:05:51 <userjgg> samuel: are you thinking in more ui changes?
 81 2012-03-11 01:06:18 <Samuel> What exactly do you mean by that?
 82 2012-03-11 01:07:46 <userjgg> samuel: the changes for now is just icons and color?
 83 2012-03-11 01:07:55 <userjgg> right?
 84 2012-03-11 01:08:16 <userjgg> or  are you thinking in more things?
 85 2012-03-11 01:08:28 <Samuel> Yeah... I guess maybe tomorrow we will do more of the inside parts
 86 2012-03-11 01:09:23 <userjgg> samuel: is it possible remove the menus?
 87 2012-03-11 01:09:24 <Samuel> We need to figure out what should go on the overview tab besides your balance, any ideas?
 88 2012-03-11 01:09:26 <userjgg> file
 89 2012-03-11 01:09:30 <userjgg> help
 90 2012-03-11 01:09:32 <userjgg> ...
 91 2012-03-11 01:10:03 <userjgg> and add a config menu on sidebar
 92 2012-03-11 01:10:13 <Samuel> I would like that, but you'll have to ask tcatm (not sure if he's still here.)
 93 2012-03-11 01:10:29 <userjgg> ok
 94 2012-03-11 01:11:24 <userjgg> i think should exist a settings menu down address book
 95 2012-03-11 01:11:57 <luke-jr> is GUIUtil::setupAmountWidget used, ever? O.o
 96 2012-03-11 01:12:02 <Samuel> You mean below address book?
 97 2012-03-11 01:12:16 <userjgg> yes below
 98 2012-03-11 01:12:28 <userjgg> my english is bad
 99 2012-03-11 01:12:42 <Samuel> yeah we may do that if we get rid of the task bar
100 2012-03-11 01:13:24 <tcatm> I think the taskbar will stay for now.
101 2012-03-11 01:13:42 <tcatm> luke-jr: Doesn't look like it is used anywhere.
102 2012-03-11 01:13:50 <luke-jr> >_<
103 2012-03-11 01:14:11 <userjgg> i like the google chrome settings design page
104 2012-03-11 01:20:38 <tcatm> Balance top right does not work.
105 2012-03-11 01:21:06 <userjgg> samuel: do you already saw safebit google chrome addon
106 2012-03-11 01:21:55 <Samuel> userjgg: No I have not heard of a Chrome add on called safebit, ill take a look at it
107 2012-03-11 01:22:08 <userjgg> it's a bitcoin wallet app
108 2012-03-11 01:22:50 <Samuel> hmm interesting
109 2012-03-11 01:24:10 <Samuel> I would use it if had at least half decent design.
110 2012-03-11 01:24:28 <Samuel> I hate it, personally
111 2012-03-11 01:24:31 <userjgg> do you have screenshot
112 2012-03-11 01:24:42 <userjgg> i can't see
113 2012-03-11 01:24:50 <Samuel> of the add on?
114 2012-03-11 01:24:57 <userjgg> slow like cell phome hardware
115 2012-03-11 01:25:01 <userjgg> yes
116 2012-03-11 01:25:33 <userjgg> i can't browse chrome addons page
117 2012-03-11 01:25:37 <Samuel> https://lh3.googleusercontent.com/0xRAnnf3n9OV63JVMjAwga7awiw-m0SZ1Dcvfng6pfqA4-iDhYW7utTm87DR68T1pfWJXrrSKg=s640-h400-e365 there is their snap they used for the extention store
118 2012-03-11 01:26:09 <userjgg> will see
119 2012-03-11 01:26:32 <Samuel> ok, maybe the deisgn isn't that bad but the functionality isn't smooth, snappy, or intuitive
120 2012-03-11 01:27:04 <userjgg> i like clean design
121 2012-03-11 01:27:26 <userjgg> maybe i liked because it's clean
122 2012-03-11 01:27:38 <Samuel> I suppose its clean
123 2012-03-11 01:28:28 <Samuel> it barely functions, i'll stick with Flexcoin and the software
124 2012-03-11 01:29:17 <Samuel> Do you use any online wallets?
125 2012-03-11 01:29:23 <tcatm> http://188.138.99.157/stuff/qtvert9.png
126 2012-03-11 01:29:27 <userjgg> samuel: look the amount field of safebit
127 2012-03-11 01:29:54 <userjgg> the us dollar
128 2012-03-11 01:30:12 <Samuel> its a little C symbol
129 2012-03-11 01:30:35 <userjgg> i think the switch between btc, mbtc .. should be like that
130 2012-03-11 01:31:37 <userjgg> i already used the online wallet of blockchain.info
131 2012-03-11 01:31:42 <Samuel> tcatm: nice! can the amount of bitcoins be left-aligned like the text above it?
132 2012-03-11 01:32:10 <sipa> Interesting, a designer and a developer collaborating via a public IRC channel.
133 2012-03-11 01:32:25 <Samuel> Hell ya!
134 2012-03-11 01:32:34 <tcatm> http://188.138.99.157/stuff/qtvert9l.png
135 2012-03-11 01:33:31 <Samuel> The Views text doesn't really fit for a cattegory name... maybe "My Account"
136 2012-03-11 01:33:54 <Samuel> The red outline is just temporary right?
137 2012-03-11 01:34:12 <tcatm> The red outline is my window manager marking the active window.
138 2012-03-11 01:34:22 <Samuel> ok
139 2012-03-11 01:36:13 <Samuel> Can the box be white so it sticks out on the gray background? Also maybe more "roundness" to the corners, try 5px
140 2012-03-11 01:36:33 <tcatm> Which box?
141 2012-03-11 01:36:45 <Samuel> The "Wallet" box
142 2012-03-11 01:37:20 <Samuel> and the Recent transactions can be in a box too, but theirs nothing their so yeah...
143 2012-03-11 01:37:31 <tcatm> Sure. I haven't really touched that part yet.
144 2012-03-11 01:37:38 <Samuel> Ok
145 2012-03-11 01:38:09 <Samuel> Im not liking the header text, bold helvetica maybe?
146 2012-03-11 01:38:11 <tcatm> I'd like to move them around a little. Maybe switch from vertical split to horiz. split?
147 2012-03-11 01:38:45 <Samuel> hmmm
148 2012-03-11 01:39:07 <Samuel> I like the current vertical split
149 2012-03-11 01:39:52 <tcatm> Let's first decide on whether the balance in the sidebar should be left or right aligned.
150 2012-03-11 01:40:39 <Samuel> left!
151 2012-03-11 01:40:58 <Samuel> liek how it is in the most recent snap
152 2012-03-11 01:41:01 <Samuel> *like
153 2012-03-11 01:41:43 <tcatm> I prefer right. Hrm.
154 2012-03-11 01:42:17 <tcatm> Reason: When right aligned digits won't move around when balance changes
155 2012-03-11 01:42:50 <Samuel> Ahh, I see
156 2012-03-11 01:42:56 <Samuel> Hmm
157 2012-03-11 01:42:56 <tcatm> So when downloading new blocks with increasing balance it won't jiggle but steadily increase.
158 2012-03-11 01:43:23 <Samuel> Thats a great point, I would have never thought of that
159 2012-03-11 01:43:28 <sipa> tcatm: i think the (number of) digits after the decimal is more likely to change
160 2012-03-11 01:43:32 <Samuel> Ok, go with right
161 2012-03-11 01:43:40 <sipa> So I say left. Nah.
162 2012-03-11 01:43:45 <userjgg> can the borders be roundness? (all borders)
163 2012-03-11 01:44:07 <tcatm> That's another thing to consider... I'd prefer fixed number of decimals (depending on BTC, mBTC, ???BTC setting)
164 2012-03-11 01:44:28 <sipa> In accounting, you typically want exact numbers.
165 2012-03-11 01:44:45 <Samuel> well if it fixed then their wont be giggling right?
166 2012-03-11 01:45:23 <tcatm> Exact numbers have trailing zeros.
167 2012-03-11 01:45:43 <sipa> Good point. And that's ugly.
168 2012-03-11 01:46:08 <Samuel> Hmm
169 2012-03-11 01:46:20 <tcatm> Hrm. Depends. Hopefully, we'll use mBTC be default within a few years so it will look a lot less ugly.
170 2012-03-11 01:46:53 <JFK911> more like MBTC after the crash
171 2012-03-11 01:47:06 <Samuel> Im not familiar with mBTC (bitcoin newbie here) what is it?
172 2012-03-11 01:47:23 <JFK911> SI unit prefix applied to the BTC unit.
173 2012-03-11 01:47:28 <tcatm> What about some "pseudo intelligent" trailing digits algorithm? If there are < 3 trailing digits, show do %.2f, else %.8f?
174 2012-03-11 01:47:45 <sipa> Maybe you can have more "units", like "BTC (8 decimals)", "BTC (4 decimals)", "BTC (2 decimals)", "mBTC (5 decimals)", "mBTC (2 decimals)", "uBTC (2 decimals)".
175 2012-03-11 01:47:58 <sipa> With BTC (2 decimals) the default.
176 2012-03-11 01:49:03 <tcatm> That's confusing! I had some coins on an exchange that rounded the amount so I couldn't withdraw the full amount even though it would have let me enter it.
177 2012-03-11 01:49:15 <sipa> Good point.
178 2012-03-11 01:52:24 <tcatm> Tricky... I wish we had switched to mBTC last year.
179 2012-03-11 01:53:56 <userjgg> samuel: i think i prefer the overview page background in white color
180 2012-03-11 01:54:00 <sipa> Samuel: mBTC = millibitcoin
181 2012-03-11 01:54:12 <userjgg> just my opinion
182 2012-03-11 01:54:42 <userjgg> i don't like only box in white
183 2012-03-11 01:55:17 <tcatm> I found an old wallet. Now we have better numbers in the screenshots :)
184 2012-03-11 01:56:06 <Samuel> userjgg: I like the gray because I would ideally like it to be like this: http://cl.ly/2c061n3n0N1g3x2Q1w3z
185 2012-03-11 01:56:18 <Samuel> so the white box stands out
186 2012-03-11 01:56:48 <Samuel> That looks much nicer than just a plain list of numbers
187 2012-03-11 01:57:49 <userjgg> it's good too, but i prefer all white. but i'm ok with it
188 2012-03-11 01:58:10 <tcatm> sipa: Would it be possible to have a user defined string (walletlabel) encoded in wallet.dat?
189 2012-03-11 01:58:31 <sipa> tcatm: for addresses?
190 2012-03-11 01:59:01 <userjgg> samuel: the orange icon is great
191 2012-03-11 01:59:15 <Samuel> yeah, you suggested that, right?
192 2012-03-11 01:59:19 <tcatm> No, just for the wallet. I.e. the text "MY WALLET" in http://188.138.99.157/stuff/qtvert9.png could be changed by the user and stored in the file so when one loads a different wallet.dat he knows which wallet is opened.
193 2012-03-11 01:59:30 <userjgg> samuel: yes
194 2012-03-11 02:00:12 <sipa> tcatm: sure, that's easy
195 2012-03-11 02:00:24 <sipa> tcatm: but adding support for more wallets is harder :)
196 2012-03-11 02:00:31 <Samuel> tcatm: I sent more icons to you, not sure if the X once fits ni with the rest though
197 2012-03-11 02:00:52 <tcatm> True, but at least users who have lots of wallets and copy them around to activate them won't be that confused anymore.
198 2012-03-11 02:01:08 <tcatm> Samuel: Got them. Thanks!
199 2012-03-11 02:01:33 <Samuel> I'm going to make adjustments to the X so it matches the rest
200 2012-03-11 02:11:25 <luke-jr> Samuel: why C when everyone's using B?? ?
201 2012-03-11 02:11:47 <tcatm> We'll stick to BTC.
202 2012-03-11 02:12:26 <luke-jr> speaking of which, I just finished TBC support for Bitcoin-Qt
203 2012-03-11 02:12:46 <sipa> *crickets*
204 2012-03-11 02:13:47 <luke-jr> under 500 LOC :P
205 2012-03-11 02:13:56 <luke-jr> 7 files changed, 247 insertions(+), 25 deletions(-)
206 2012-03-11 02:14:12 <tcatm> Does it simplify Balance formatting?
207 2012-03-11 02:14:31 <luke-jr> tcatm: actually, I was about to mention that
208 2012-03-11 02:14:54 <tcatm> Cool. Where's the code? Maybe we can throw away the TBC stuff and use it :)
209 2012-03-11 02:14:55 <luke-jr> it does simplify amount entry at least
210 2012-03-11 02:15:01 <luke-jr> tcatm: no, that would be evil
211 2012-03-11 02:15:25 <luke-jr> zero reason to omit TBC
212 2012-03-11 02:15:33 <[Tycho]> Where is the "TBC stuff" ?
213 2012-03-11 02:15:35 <luke-jr> it'd just be more mindless bigotry
214 2012-03-11 02:16:00 <luke-jr> https://github.com/luke-jr/bitcoin/commit/69850957c238b2954c2dd0c941ef111334cf54e1
215 2012-03-11 02:18:19 <[Tycho]> Do anyone remembers the bug when old client accepted duplicate inputs in TX
216 2012-03-11 02:18:20 <[Tycho]> ?
217 2012-03-11 02:19:36 <sipa> [Tycho]: yes
218 2012-03-11 02:20:03 <[Tycho]> Where exactly is the code that doesn't allows accepting such TXes to memorypool ?
219 2012-03-11 02:21:28 <sipa> [Tycho]: this is the commit that introduced that check: https://github.com/bitcoin/bitcoin/commit/33208fb5575d76a19163e830617eaaf32dbacda8
220 2012-03-11 02:22:38 <[Tycho]> Thanks.
221 2012-03-11 02:22:46 <tcatm> Can someone send a few coins to mfpMcLEExnHFeP12CsY34MbrVR1G8GAxE7 ?
222 2012-03-11 02:23:42 <sipa> tcatm: how many?
223 2012-03-11 02:24:12 <Samuel> I can send ya 0.03, lol
224 2012-03-11 02:24:29 <sipa> I have 4700 tnBTC
225 2012-03-11 02:24:51 <Samuel> tnBTC? Now whats that?
226 2012-03-11 02:24:59 <sipa> testnet bitcoin
227 2012-03-11 02:24:59 <tcatm> Doesn't matter how much. I just want to see whether the Qt signal to update the QLabel works :)
228 2012-03-11 02:25:24 <luke-jr> tcatm: sent you 1 GTBC
229 2012-03-11 02:25:27 <tcatm> Thanks!
230 2012-03-11 02:25:36 <sipa> tcatm: sent you some
231 2012-03-11 02:25:53 <luke-jr> hmm
232 2012-03-11 02:26:06 <luke-jr> I presume it's intentional that the "Send" confirmation prompt always uses BTC, regardless of selected unit?
233 2012-03-11 02:26:13 <tcatm> Oh, it looks like currentBalance will only update with confirmations.
234 2012-03-11 02:26:16 <Samuel> Ok, so how do you use these "fake" bitcoins?
235 2012-03-11 02:26:27 <sipa> Samuel: run the client with option -testnet
236 2012-03-11 02:26:42 <sipa> it's a completely parallel world, only the coins are worthless
237 2012-03-11 02:27:01 <luke-jr> & and difficulty has weird rules
238 2012-03-11 02:27:05 <Samuel> Ah, I see. great way to explain it
239 2012-03-11 02:27:40 <tcatm> What's the blockrate on testnet?
240 2012-03-11 02:28:29 <tcatm> Yay. it works! :)
241 2012-03-11 02:28:37 <sipa> tcatm: undefined.
242 2012-03-11 02:29:45 <tcatm> http://188.138.99.157/stuff/qtvert10.png Balance now updates :)
243 2012-03-11 02:30:05 <luke-jr> sipa: https://github.com/bitcoin/bitcoin/pull/929
244 2012-03-11 02:30:37 <Samuel> I'll get you some other icons to use other than those, they're kinda ugly
245 2012-03-11 02:30:41 <sipa> luke-jr: find me a significant number of users that care about it.
246 2012-03-11 02:31:04 <luke-jr> sipa: the point is to grow adoption of Bitcoin by providing another reason for people to use it
247 2012-03-11 02:31:22 <luke-jr> sipa: also, zero effect on people who don't use it
248 2012-03-11 02:31:32 <[Tycho]> luke-jr: but bitcoin is already adopted by you.
249 2012-03-11 02:31:35 <sipa> agree, but non-zero effect on the code size
250 2012-03-11 02:32:00 <sipa> and no i do not believe bitcoin's adoption will be increased by hordes of tonal fans suddenly because of this
251 2012-03-11 02:32:28 <tcatm> Anyone have some time to compile https://github.com/tcatm/bitcoin/tree/new-gui and see whether the Sidebar breaks with different styles or even on different platforms like OSX and windows?
252 2012-03-11 02:33:20 <gribble> New news from bitcoinrss: luke-jr opened pull request 929 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/929>
253 2012-03-11 02:37:38 <luke-jr> sipa: would tonal+dozenal be enough people?
254 2012-03-11 02:37:49 <tcatm> No :)
255 2012-03-11 02:39:23 <neofutur> somments and pull requests welcome on https://github.com/neofutur/bitcoin_simple_php_tools
256 2012-03-11 02:39:27 <neofutur> http://bitcoin.gw.gd/bitcoin_simple_php_tools/imageprice/pricetobtc.png?usdprice=100
257 2012-03-11 02:39:34 <XMPPwocky> nothing says secure like PHP
258 2012-03-11 02:39:47 <neofutur> http://bitcoin.gw.gd/bitcoin_simple_php_tools/imageprice/pricetobtc.txt.php?eurprice=100
259 2012-03-11 02:40:47 <XMPPwocky> bitcoin.gw.gd/bitcoin_simple_php_tools/textprice/pricetobtc.txt.php?usdprice=<script>alert("hi");</script> that was easy.
260 2012-03-11 02:40:48 <neofutur> ( still have to work the look of the image :P but it works )
261 2012-03-11 02:41:17 <neofutur> XMPPwocky: this is not working
262 2012-03-11 02:41:24 <neofutur> just saying "bad price"
263 2012-03-11 02:41:52 <neofutur> or I missed something ?
264 2012-03-11 02:42:05 <neofutur> anyone see a js alert ?
265 2012-03-11 02:42:22 <XMPPwocky> woops
266 2012-03-11 02:42:45 <neofutur> but feel free to report any real security problems on https://github.com/neofutur/bitcoin_simple_php_tools
267 2012-03-11 02:42:53 <tcatm> Samuel: Where do you think should we move the statusbar icons?
268 2012-03-11 02:43:37 <Samuel> tcatm: What icons are you referring to?
269 2012-03-11 02:44:30 <tcatm> http://188.138.99.157/stuff/qtvert10.png bottom right
270 2012-03-11 02:44:57 <Samuel> oh those.
271 2012-03-11 02:45:02 <Samuel> brb
272 2012-03-11 02:47:32 <Samuel> THe only place I can think of is the sidebar, but they might look good staying at the bottom if they were apart of a bar at the bottom instead of just being there. I'll mock it up, one sec
273 2012-03-11 02:53:46 <Samuel> tcatm: http://cl.ly/2J2n2F3i2R0V2y2j190u Thats where they should go
274 2012-03-11 02:54:28 <Samuel> tcatm: Should I get some icons for the signal bars and checkmark to match the rest?
275 2012-03-11 02:55:08 <Samuel> brb
276 2012-03-11 02:55:13 <tcatm> Hmm, there a lot of icons to replace.
277 2012-03-11 02:57:24 <Samuel> Like an icon for one bar and another for two bars so on and so forth?
278 2012-03-11 02:57:43 <tcatm> Yep. And a spinning double arrow when syncing the chain.
279 2012-03-11 02:58:14 <tcatm> https://github.com/tcatm/bitcoin/tree/new-gui/src/qt/res/icons These are all icons used.
280 2012-03-11 02:58:15 <Samuel> Are those just the default that Qt has?
281 2012-03-11 02:58:51 <tcatm> No, those are the ones Bitcoin uses.
282 2012-03-11 02:59:13 <Samuel> yeah, thats a lot to replace... darn
283 2012-03-11 03:16:16 <etotheipi_> question about Satoshi networking...
284 2012-03-11 03:16:36 <etotheipi_> I used to send the satoshi client a tx, then send it a getdata request for the txhash to see if it was accepted
285 2012-03-11 03:17:29 <etotheipi_> it's my way of checking whether the Satoshi client accepted, and thus forwarded, my transaction from Armory
286 2012-03-11 03:18:12 <etotheipi_> recently, it stopped doing it: it does accept and forward the transaction, but does not reply to the getdata request
287 2012-03-11 03:18:49 <etotheipi_> in debug.log, I see "AcceptToMemoryPool(): accepted 6c48189779", then "received getdata for: tx 6c481897797a73f6ac26" 6 times, then "socket closed"
288 2012-03-11 03:19:18 <etotheipi_> "disconnecting node 127.0.0.1:51666"
289 2012-03-11 03:19:50 <etotheipi_> from Armory, I see that the getdata message was sent, but I never received a reply from the Satoshi client
290 2012-03-11 03:20:28 <etotheipi_> why would it not reply?  and why would it disconnect localhost (I thought it wasn't supposed to do that)
291 2012-03-11 03:20:45 <etotheipi_> sipa, gmaxwell, luke-jr... I'm baffled by this, maybe one of you can help
292 2012-03-11 03:20:58 <sipa> The disconnecting is strange, it could be anti-DDoS, but it should report that in debug.log
293 2012-03-11 03:21:21 <etotheipi_> it appears that other users are reporting the same thing
294 2012-03-11 03:21:28 <sipa> Maybe it doesn't reply to getdata because it thinks/knows you already know about that transaction.
295 2012-03-11 03:22:10 <etotheipi_> is there a higher-debugging I can enable and try again?
296 2012-03-11 03:22:46 <sipa> no
297 2012-03-11 03:23:03 <tcatm> Samuel: Icons moved. Doesn't look that good, though. http://188.138.99.157/stuff/qtvert11.png
298 2012-03-11 03:23:33 <[Tycho]> "Warning: This is a warning" :)
299 2012-03-11 03:23:44 <etotheipi_> okay, I'll see if I can revert back to an older version that I know worked... and see if the debug log looks any different
300 2012-03-11 03:23:51 <userjgg> samuel: don't you think the numbers on recent transactions should be smaller in size than total on the wallet?
301 2012-03-11 03:25:28 <Samuel> userjgg: yeah, I like how I have it from my snap: http://cl.ly/2J2n2F3i2R0V2y2j190u
302 2012-03-11 03:25:29 <sipa> etotheipi_: i know of no changes to that mechanism
303 2012-03-11 03:25:59 <Samuel> tcatm: Yeah, not sure about it.
304 2012-03-11 03:26:32 <Samuel> I guess we keep it on the bottom right
305 2012-03-11 03:27:16 <userjgg> samuel: i think is not good be in the same size
306 2012-03-11 03:27:19 <tcatm> I'd like to remove the statusbar..
307 2012-03-11 03:27:30 <neofutur> dynamic btc prices for forum posts :
308 2012-03-11 03:27:31 <userjgg> should be smaller
309 2012-03-11 03:27:32 <neofutur> [img]http://bitcoin.gw.gd/bitcoin_simple_php_tools/imageprice/pricetobtc.png?usdprice=350[/img]
310 2012-03-11 03:27:46 <neofutur> example on https://bitcointalk.org/index.php?topic=1687.msg26007#msg26007
311 2012-03-11 03:28:08 <neofutur> your prices will change following the bitcoin price
312 2012-03-11 03:28:28 <Samuel> userjgg: So you think they should be the same size?
313 2012-03-11 03:28:32 <neofutur> comments and pull requests welcome on https://github.com/neofutur/bitcoin_simple_php_tools
314 2012-03-11 03:29:54 <etotheipi_> sipa, okay it looks like the old version still works, so I'll look closer at what changed (though I didn't think I changed any of that code...
315 2012-03-11 03:29:54 <userjgg> samuel: i think the numbers on recent transaction box should be smaller that the wallet box
316 2012-03-11 03:30:05 <userjgg> than
317 2012-03-11 03:30:19 <Samuel> userjgg: oh I see what you mean now
318 2012-03-11 03:30:30 <userjgg> the 100.000...
319 2012-03-11 03:30:42 <sipa> etotheipi_: which is the old version?
320 2012-03-11 03:31:20 <etotheipi_> sorry, I meant old version of armory
321 2012-03-11 03:31:38 <userjgg> samuel:the total of bitcoins on the wallet should be in evidence.
322 2012-03-11 03:31:44 <sipa> etotheipi_: ah, ok
323 2012-03-11 03:32:31 <etotheipi_> sipa, I actually see the exact same thing in the debug log:  "AcceptToMemoryPool(): accepted 826978bbb9" followed by 7 instances of "received getdata for: tx 826978bbb9923b1aaee2', then "socket closed" and "disconnecting"
324 2012-03-11 03:32:45 <etotheipi_> but somehow Armory got the tx packed this time
325 2012-03-11 03:33:17 <sipa> sure it is not armory that is closing the socket?
326 2012-03-11 03:33:37 <etotheipi_> sipa, I suppose it's possible, though I never intended to
327 2012-03-11 03:34:01 <etotheipi_> but what's weird is that I added debug output to the beginng of ever "dataReceived" call and "sendMessage" call
328 2012-03-11 03:34:04 <userjgg> samuel: on recent transaction box should have the icons of < and >
329 2012-03-11 03:34:21 <etotheipi_> sipa, I definitely only sent the getdata request once, and did receive the tx packet reply
330 2012-03-11 03:34:27 <etotheipi_> but the debug log doesn't indicate it
331 2012-03-11 03:34:31 <etotheipi_> (on the satoshi client)
332 2012-03-11 03:34:33 <userjgg> samuel: btc comming and going
333 2012-03-11 03:35:09 <etotheipi_> the satoshi client reports 7 getdata requests, then a disconnect, but never records having sent a tx packet
334 2012-03-11 03:35:19 <userjgg> samuel: the icon of comming in orange color
335 2012-03-11 03:35:49 <userjgg> samuel: and the going icon gray
336 2012-03-11 03:35:56 <Samuel> hmm, that might just be confusing to users
337 2012-03-11 03:35:58 <sipa> etotheipi_: it doesn't report that, it seems
338 2012-03-11 03:36:18 <userjgg> samuel: hum
339 2012-03-11 03:36:46 <Samuel> I was thinking "You sent ------" "You received ------"
340 2012-03-11 03:36:48 <sipa> etotheipi_: and it always should, if that transaction is in its relay memory
341 2012-03-11 03:37:16 <etotheipi_> sipa, ugh... well at least I have code that works, so I can compare directly what I changed... perhaps I'll find a subtle, eduacational networking nuance
342 2012-03-11 03:37:28 <userjgg> samuel: represented by name or icon?
343 2012-03-11 03:37:44 <etotheipi_> sipa, though the getdata packets look identical, besides the checksum and the txhash
344 2012-03-11 03:37:52 <Samuel> the text would say, not sure if I'd like an icon
345 2012-03-11 03:38:11 <userjgg> samuel: understand
346 2012-03-11 03:38:16 <sipa> etotheipi_: hmm
347 2012-03-11 03:38:26 <sipa> etotheipi_: by the way, i got the HMAC-SHA512 key derivation implemented and unit tested
348 2012-03-11 03:38:31 <etotheipi_> SOB
349 2012-03-11 03:38:40 <etotheipi_> it looks like endianness got switched somewhere
350 2012-03-11 03:38:47 <sipa> in your code?
351 2012-03-11 03:38:55 <etotheipi_> yes
352 2012-03-11 03:38:56 <userjgg> samuel: it's good
353 2012-03-11 03:38:59 <sipa> bummer :)
354 2012-03-11 03:39:18 <Samuel> userjgg: What's good?
355 2012-03-11 03:39:44 <userjgg> samuel: only names like you sent, you received
356 2012-03-11 03:39:57 <etotheipi_> wtf...
357 2012-03-11 03:40:13 <etotheipi_> or, at least it printed to console differently...
358 2012-03-11 03:40:36 <Samuel> userjgg: Ahh yes,  I'm changing some things. I'll send you the new snap when its done.
359 2012-03-11 03:40:47 <userjgg> ok
360 2012-03-11 03:41:11 <etotheipi_> I HATE ENDIANNESS IN BITCOIN
361 2012-03-11 03:45:16 <sipa> there, there..
362 2012-03-11 03:47:11 <etotheipi_> sipa, it looks to me like the hash works in the getdata packet when in little-endian
363 2012-03-11 03:47:19 <etotheipi_> but that doesn't sound right... I might've just mixed things up
364 2012-03-11 03:50:46 <etotheipi_> even more mysteriously, I see not a single line of code changed between the working and non-working versions that has to do with endianness...
365 2012-03-11 03:50:54 <etotheipi_> okay, my problem now, thanks for listening :)
366 2012-03-11 03:54:29 <Joric> etotheipi_!
367 2012-03-11 03:55:29 <Joric> etotheipi_, are you going to rewrite blockchain loader without loading it completely into ram
368 2012-03-11 03:56:10 <Samuel> userjgg: http://cl.ly/0s06081E342q2N0C460p I was thinking green or red bars under the transactions for receiving and sending
369 2012-03-11 03:56:26 <Samuel> Got to go. See ya!
370 2012-03-11 03:56:39 <userjgg> samuel: i'll tell this now
371 2012-03-11 03:57:17 <Joric> i was trying simple blockchain scanner https://github.com/joric/pyblockchain it takes ~10 minutes to parse while loading takes only 20 seconds :( i guess all because of disk IO
372 2012-03-11 03:58:15 <Joric> i mean i may load those 1.2 gb into ram in 20 seconds, but parsing it from disk takes ages
373 2012-03-11 03:59:15 <Joric> maybe it would be faster to load/parse by 100-250 mb chunks or something
374 2012-03-11 03:59:30 <etotheipi_> Joric...I don't follow entirely
375 2012-03-11 03:59:42 <etotheipi_> Armory currently loads the blockchain into RAM in one giant chunk
376 2012-03-11 03:59:57 <etotheipi_> then it can scan for new address balances in less than 1 s
377 2012-03-11 04:00:07 <Joric> etotheipi_, it doesn't work on 2gb machines already
378 2012-03-11 04:00:20 <etotheipi_> if it's reading from disk, it takes 20-60 sec depending on your disk IO
379 2012-03-11 04:00:46 <etotheipi_> I will be rewriting it to mmap the file... which one machines with enough RAM, will take the same (nearly instantaneous, after loading)
380 2012-03-11 04:00:46 <phantomcircuit> armory is quickly going to stop working
381 2012-03-11 04:00:47 <phantomcircuit> as in
382 2012-03-11 04:00:51 <phantomcircuit> like next month
383 2012-03-11 04:00:54 <etotheipi_> phantomcircuit, I agree
384 2012-03-11 04:01:12 <phantomcircuit> etotheipi_, what is armory written in?
385 2012-03-11 04:01:23 <etotheipi_> phantomcircuit, Armory's blockchain scanning was originally written as a data structures optimization exercise for fun
386 2012-03-11 04:01:24 <sipa> etotheipi_: what if the client crashes or get killed during a write?
387 2012-03-11 04:01:44 <etotheipi_> sipa, right now Armory doesn't write anything to blk0001.dat
388 2012-03-11 04:01:55 <sipa> ah, of course
389 2012-03-11 04:02:17 <etotheipi_> I will be re-writing the blockchain scanning to not load everythign into RAM
390 2012-03-11 04:02:32 <etotheipi_> and keep some extra data around for wallets so it can find data quickly
391 2012-03-11 04:02:43 <tcatm> What do you think about moving the "downloading blocks" progressbar to the sidebar?
392 2012-03-11 04:03:16 <phantomcircuit> etotheipi_, a hint, in general there is little reason to worry about the integrity of the block chain on the local node so long as you can identify corruption
393 2012-03-11 04:03:20 <phantomcircuit> (which is easy)
394 2012-03-11 04:03:54 <etotheipi_> phantomcircuit, I do have a blockchain integrity checker
395 2012-03-11 04:04:11 <phantomcircuit> i think you misunderstood me
396 2012-03-11 04:04:20 <etotheipi_> I haven't figured out exactly how I'm going to do all this yet... but at the moment it will be a full rescan on every load (as now), but only mmap'ing it
397 2012-03-11 04:04:25 <Joric> i'm thinking of writting a cached file class to load/parse using 500mb portions
398 2012-03-11 04:04:50 <phantomcircuit> Joric, dont it'll be slower than just relying on the OS's page cache
399 2012-03-11 04:05:00 <etotheipi_> in fact, my long-term plan may be to completely re-do it
400 2012-03-11 04:05:01 <phantomcircuit> since you'll end up with duplicate caching
401 2012-03-11 04:05:48 <phantomcircuit> etotheipi_, can you reliably verify transactions
402 2012-03-11 04:06:29 <Joric> phantomcircuit, dunno why but i may load those 1.2 gb in ram in 20 seconds but when i'm parsing sequentally it takes 10 minutes
403 2012-03-11 04:06:53 <etotheipi_> phantomcircuit, I don't know what you mean by "verify"...  I don't do full verification, I let the Satoshi client do that for me
404 2012-03-11 04:07:11 <etotheipi_> but I do check leading zeros on headers and make sure all hashes match the way I expect
405 2012-03-11 04:07:13 <jrmithdobbs> Joric: because it's just an mmap it's not actually getting loaded
406 2012-03-11 04:07:14 <phantomcircuit> Joric, when you say load into ram, do you mean simply reading the file sequentially into ram using a small number of read() calls?
407 2012-03-11 04:07:26 <Joric> jrmithdobbs, oh
408 2012-03-11 04:07:27 <jrmithdobbs> or that
409 2012-03-11 04:07:32 <jrmithdobbs> one of the two is the problem ;p
410 2012-03-11 04:07:33 <phantomcircuit> etotheipi_, oh so you dont do the hard part...
411 2012-03-11 04:07:36 <phantomcircuit> yeah
412 2012-03-11 04:08:02 <etotheipi_> phantomcircuit, not yet
413 2012-03-11 04:08:09 <phantomcircuit> Joric, load the block chain into memory
414 2012-03-11 04:08:15 <phantomcircuit> then test parsing time
415 2012-03-11 04:08:25 <phantomcircuit> you want to force the file into the page cache
416 2012-03-11 04:08:34 <etotheipi_> but i have everythign in place, I *could* do it, but that could be a HUGE investment of time, that's not necessary until I cut the umbilical cord to the Satoshi client
417 2012-03-11 04:08:37 <phantomcircuit> just do something like dd if=blk0001.dat of=/dev/null
418 2012-03-11 04:09:12 <etotheipi_> Joric, the way Armory does it, it literally does one giant block copy into RAM
419 2012-03-11 04:09:15 <phantomcircuit> etotheipi_, i suspect you are vastly under estimating the complexity of doing full transaction verification as well as the enormous performance problem that hitting disk is
420 2012-03-11 04:09:19 <phantomcircuit> but whatever carry on
421 2012-03-11 04:09:33 <etotheipi_> phantomcircuit, that's exactly why I haven't done it yet
422 2012-03-11 04:09:48 <phantomcircuit> to be completely honest
423 2012-03-11 04:10:06 <phantomcircuit> armory seems like it's a waste of time as there is a clear and obvious expiration date
424 2012-03-11 04:10:18 <etotheipi_> phantomcircuit, that's a pretty narrow view
425 2012-03-11 04:10:27 <phantomcircuit> you're basically racing against average device memory space
426 2012-03-11 04:10:30 <etotheipi_> phantomcircuit, if I died today, you would be right
427 2012-03-11 04:10:56 <etotheipi_> but I'm not sitting idly by... I'm aware that I'm running out of time to upgrade Armory
428 2012-03-11 04:11:09 <phantomcircuit> etotheipi_, just add an rpc call to bitcoin to get the balance at an address where bitcoind has just the address
429 2012-03-11 04:11:14 <etotheipi_> it's been in my plan.. and I'd hardly consider the app to be a waste of time
430 2012-03-11 04:11:15 <phantomcircuit> it would be order of magnitude easier
431 2012-03-11 04:11:47 <jrmithdobbs> what are we talking about exactly
432 2012-03-11 04:11:49 <Joric> etotheipi_, you don't have and additional hash map for tx hashes, do you?
433 2012-03-11 04:11:56 <etotheipi_> phantomcircuit, I'd have to be retarded not to recognize what you're talking about
434 2012-03-11 04:12:08 <etotheipi_> and I'd have to be retarded to believe that somehow it's sustainable
435 2012-03-11 04:12:19 <etotheipi_> I never said this was the end-all
436 2012-03-11 04:12:35 <phantomcircuit> jrmithdobbs, armory is a pseudo client, basically he loads the entire block chain into ram and doesn't verify it, then uses the information to produce transactions
437 2012-03-11 04:13:44 <jrmithdobbs> phantomcircuit: best (worst) of both worlds, split it all out to individual files in a huge hardlink tree and mmap is necessary O(n) lookups and the OS does all your caching for you. (remember to adjust ulimit -n) ;p
438 2012-03-11 04:13:57 <jrmithdobbs> s/is necc/as/
439 2012-03-11 04:14:00 <Joric> i'm going to write my own stream handler and finetune the ram/disk part of it )
440 2012-03-11 04:14:06 <phantomcircuit> jrmithdobbs, lolllllll
441 2012-03-11 04:14:16 <jrmithdobbs> phantomcircuit: it'd actually work pretty well depending on how you chunked it
442 2012-03-11 04:14:26 <jrmithdobbs> so long as you managed open fds sanely
443 2012-03-11 04:14:32 <Joric> and i'm not going to keep blockchain in ram, i'm going to use hash maps for addresses / transactions
444 2012-03-11 04:14:38 <jrmithdobbs> way over complex tho
445 2012-03-11 04:15:58 <phantomcircuit> jrmithdobbs, actually
446 2012-03-11 04:16:10 <phantomcircuit> sqlite with fsync completely off doesn't perform poorly
447 2012-03-11 04:16:48 <phantomcircuit> the performance of bitcoind can be massively improved if you realize that you dont really need to guarantee the integrity of your own blockchain
448 2012-03-11 04:16:54 <phantomcircuit> as you can always check it against the networks
449 2012-03-11 04:17:02 <Joric> tried sqlite, found out you have to call commit() as seldom as possible )
450 2012-03-11 04:17:14 <phantomcircuit> Joric, yeah commit causes an fsync
451 2012-03-11 04:17:20 <phantomcircuit> a modern hdd will give you
452 2012-03-11 04:17:26 <phantomcircuit> MAYBE 20 fsync calls/second
453 2012-03-11 04:18:02 <phantomcircuit> so take everything your running combine that with fsync's from page cache being flushed and well you have maybe like 7 of them for sqlite
454 2012-03-11 04:18:18 <Joric> not sure sqlite is perfectly good for blockchain bsd db seems fine
455 2012-03-11 04:19:12 <Joric> there's no indexing by address though
456 2012-03-11 04:19:50 <phantomcircuit> sqlite is perfectly capable if you know the gotchas in bitcoin
457 2012-03-11 04:20:22 <jrmithdobbs> doesn't genjix have it working?
458 2012-03-11 04:20:39 <jrmithdobbs> with libbitcoin
459 2012-03-11 04:21:05 <phantomcircuit> he's got a postgresql interface
460 2012-03-11 04:21:25 <phantomcircuit> he isn't batching enough so it isn't that fast
461 2012-03-11 04:21:29 <phantomcircuit> but it could be with some effort
462 2012-03-11 04:21:49 <phantomcircuit> when there is a block reorg that results in a TON of updates
463 2012-03-11 04:21:59 <phantomcircuit> and iirc he was calling commit after each one
464 2012-03-11 04:22:55 <jrmithdobbs> ya, large reorgs need to drop the locking until complete and then verify
465 2012-03-11 04:23:20 <sipa> jrmithdobbs: ?
466 2012-03-11 04:23:39 <phantomcircuit> well im going to go back to my HFT engine
467 2012-03-11 04:23:40 <phantomcircuit> ;)
468 2012-03-11 04:23:50 <jrmithdobbs> sipa: if stored as sql
469 2012-03-11 04:23:57 <sipa> ah
470 2012-03-11 04:24:21 <jrmithdobbs> but the rdbms does the "verify" step anyways
471 2012-03-11 04:24:22 <Joric> for now i'm just looking for the fastest way to reindex blockchain
472 2012-03-11 04:24:54 <sipa> Joric: in the satoshi client, or what?
473 2012-03-11 04:25:24 <jrmithdobbs> sipa: err, that was supposed to be 'locking model', eg, a reorg would be one big commit in sql
474 2012-03-11 04:25:28 <Joric> as in binary concatenation of blocks
475 2012-03-11 04:25:39 <Joric> not really just want to gather some stats
476 2012-03-11 04:26:22 <sipa> jrmithdobbs: it currently is in the satoshi client too, and it shouldn't be (always)
477 2012-03-11 04:26:25 <Joric> build a net of addresses, etc
478 2012-03-11 04:26:26 <phantomcircuit> jrmithdobbs, yeah that's what i was telling him
479 2012-03-11 04:26:56 <phantomcircuit> the python client i was working on did that properly
480 2012-03-11 04:27:10 <phantomcircuit> but there is some bug in the reorg that i cant be asked to track down
481 2012-03-11 04:27:11 <phantomcircuit> so
482 2012-03-11 04:27:13 <phantomcircuit> whatever
483 2012-03-11 04:28:45 <Joric> maybe i just need to build addrindex.dat along with blkindex.dat )
484 2012-03-11 04:29:02 <jrmithdobbs> Joric: that's in the wallet?
485 2012-03-11 04:29:16 <Joric> the global one
486 2012-03-11 04:29:27 <jrmithdobbs> what
487 2012-03-11 04:29:40 <jrmithdobbs> oh, nm
488 2012-03-11 04:30:11 <tcatm> Does anybody know why all this connect() lines where there (effectively showing the main window when switching tabs)? https://github.com/tcatm/bitcoin/commit/92b217d0dabf38bab2469635ba938bc7c49b616e
489 2012-03-11 04:39:01 <Samuel> I'm back
490 2012-03-11 04:42:06 <tcatm> Great :)
491 2012-03-11 04:42:24 <tcatm> We still need to figure out where to put those icons :)
492 2012-03-11 04:43:01 <Samuel> Yes
493 2012-03-11 04:43:57 <forsetifox> Get rid of them and have a thingy that slides down and displays useful numbers underneath the buttons on the left.
494 2012-03-11 04:44:10 <tcatm> "slides down"?
495 2012-03-11 04:44:21 <forsetifox> Click button. Numbers appear!
496 2012-03-11 04:44:31 <tcatm> Good idea.
497 2012-03-11 04:45:00 <tcatm> It should still show small icons, though.
498 2012-03-11 04:49:50 <tcatm> Is it possible to unlock the wallet using the GUI?
499 2012-03-11 04:52:36 <Samuel> Maybe this looks better: http://cl.ly/1j2u1p0x1h3h2M0C3k26
500 2012-03-11 04:52:42 <sipa> tcatm: no
501 2012-03-11 04:52:53 <sipa> only as-needed
502 2012-03-11 04:53:52 <tcatm> I see. If I unlock using RPC will it be unlocked for the GUI, too?
503 2012-03-11 04:53:59 <sipa> yes
504 2012-03-11 04:54:21 <sipa> (though maybe the GUI is not smart enough, and still asks for a password, not sure)
505 2012-03-11 04:56:35 <tcatm> Looks like it is smart enough.
506 2012-03-11 04:57:09 <tcatm> Samuel: We need a good pair of lock icons. One indicating locked state and one for unlocked wallet.
507 2012-03-11 04:58:12 <Joric> Samuel, 12k bitcoins look great, who's wallet it is?
508 2012-03-11 04:58:47 <Samuel> Its fake lol (123456789)
509 2012-03-11 04:58:59 <tcatm> Joric: It's basically QLabe("123,123.123456789 BTC"); :D
510 2012-03-11 04:59:46 <Samuel> tcatm: So the locked or unlock icon would go next to the checkmark and signal icons?
511 2012-03-11 05:00:36 <tcatm> Samuel: Maybe. What do you think about putting them right to MY WALLET?
512 2012-03-11 05:01:07 <Samuel> Let me understand what these icons mean first
513 2012-03-11 05:01:32 <tcatm> Locked = Wallet encrypted and passphrase not entered
514 2012-03-11 05:02:04 <tcatm> Unlocked = Wallet encrypted but passphrase entered so coins can be spent without entering the passphrase
515 2012-03-11 05:02:14 <Samuel> I see thanks
516 2012-03-11 05:02:41 <Samuel> I think a lock icon should go to the right of any of the tabs it applys to
517 2012-03-11 05:03:09 <Samuel> Since it affects the send tab it would be grayed out with the lock icon
518 2012-03-11 05:03:32 <Samuel> And same with other tabs
519 2012-03-11 05:03:36 <tcatm> It also applies to the RPC server.
520 2012-03-11 05:04:27 <Samuel> I don't know what that is... but I think you get my point
521 2012-03-11 05:04:29 <tcatm> The current workflow is: Click 'send coins', enter address and amount, click send, enter passphrase, coins get sent.
522 2012-03-11 05:05:48 <Samuel> ok
523 2012-03-11 05:08:23 <Samuel> gosh Photoshop keeps quitting on me. I'm going to restart my computer, brb
524 2012-03-11 05:14:06 <Samuel> I'm back again
525 2012-03-11 05:14:13 <tcatm> sipa: Is it possible to test warnings on testnet?
526 2012-03-11 05:14:38 <tcatm> http://188.138.99.157/stuff/qtvert12.png no more statusbar!
527 2012-03-11 05:16:10 <Samuel> The loading bar is cramped aganst the edges of the window, give it some padding on all sides
528 2012-03-11 05:16:49 <tcatm> That's just temporary because it had to go somewhere. :)
529 2012-03-11 05:19:31 <Samuel> oh ok
530 2012-03-11 05:20:02 <tcatm> Maybe we could combine it with the spinning arrows icon in the sidebar.
531 2012-03-11 05:20:35 <Samuel> I have a new green check icon, I guess I'll keep making these until all of them match our new design. I'll send it to you
532 2012-03-11 05:22:51 <Joric> how about calculating progressbar from what has been loaded / left to load not as now
533 2012-03-11 05:23:21 <Samuel> brb
534 2012-03-11 05:24:06 <tcatm> Joric: Isn't that how it works now?
535 2012-03-11 05:24:32 <Joric> i guess not it starts from zero every launch
536 2012-03-11 05:25:09 <tcatm> Nope. Recent versions start from 99% IIRC.
537 2012-03-11 05:29:41 <Samuel> back
538 2012-03-11 05:36:19 <Samuel> tcatm: Did you get my green check replacement? (It's to replace this one https://github.com/tcatm/bitcoin/blob/new-gui/src/qt/res/icons/synced.png)
539 2012-03-11 05:45:49 <wumpus> sipa: the gui will notice when the wallet is unlocked through RPC, and update the lock icon
540 2012-03-11 05:48:42 <Samuel> wumpus: What about just not displaying the "unlocked" icon when its unlocked? The user would know its locked if is.
541 2012-03-11 05:49:23 <Samuel> *would know its UNlocked if it is
542 2012-03-11 05:49:31 <wumpus> there are three states: unencrypted (no icon), encrypted and locked, and encrypted and unlocked
543 2012-03-11 05:50:00 <Samuel> Ahh, ok. That makes sense
544 2012-03-11 05:50:36 <wumpus> encrypted and locked, and encrypted and unlocked have an icon... so you can still see the wallet is encrypted even though it is temporarily unlocked
545 2012-03-11 06:04:26 <wumpus> tcatm: looks nice
546 2012-03-11 06:14:28 <Samuel> Still here tcatm?
547 2012-03-11 06:14:39 <Samuel> I have lock and unlock icons
548 2012-03-11 09:03:14 <UukGoblin> I'm trying to remind myself how merged mining worked...
549 2012-03-11 09:03:21 <UukGoblin> does namecoin have to look at bitcoin blocks at all?
550 2012-03-11 09:04:09 <UukGoblin> if it's what I think it is, i.e. you try to find a target bitcoin block with namecoin data in the coinbase... what happens when there's more than 1 aux chain? How does namecoin know what data is hashed in coinbase?
551 2012-03-11 09:05:35 <UukGoblin> oh, and what's going on with BIP16/17 in the end?
552 2012-03-11 09:05:42 <UukGoblin> (as a separate question)
553 2012-03-11 09:15:22 <neofutur> (12:01) < UukGoblin> does namecoin have to look at bitcoin blocks at all?
554 2012-03-11 09:15:33 <neofutur> yes you need a bitcoind and a namecoind
555 2012-03-11 09:15:45 <neofutur> you ll have more answers on #namecoin
556 2012-03-11 09:16:00 <neofutur> https://github.com/vinced/namecoin/blob/mergedmine/contrib/merged-mine-proxy
557 2012-03-11 09:17:23 <UukGoblin> neofutur, I know what I need, I asked how does namecoin know what other altchains were mined
558 2012-03-11 09:17:51 <UukGoblin> and does it actually have to look at bitcoin blocks - and if so, how does it connect to bitcoin?
559 2012-03-11 09:18:31 <UukGoblin> I'm trying to understand how exactly does merged mining work, not just for namecoin but other chains too
560 2012-03-11 09:18:39 <UukGoblin> oh - and is there a list of chain IDs somewhere?
561 2012-03-11 09:41:47 <denisx> since the deployment of the BIP30 patch I had 2 blocks not accepted by bitcoind, is that possible?
562 2012-03-11 09:42:29 <Eliel> UukGoblin: As I understood it, the namecoin blocks are not necessarily valid bitcoin blocks since the difficulties differ.
563 2012-03-11 09:43:21 <Eliel> UukGoblin: so, namecoin needs to understand bitcoin's blockformat enough to verify that the bitcoin block's hash that contains the namecoin block hash matches the difficulty.
564 2012-03-11 09:55:25 <UukGoblin> Eliel, exactly, it needs to know that the namecoin block hash is included in the low-diff bitcoin block, right? But how does it know that, when the namecoin data is hashed with other chains' data in the coinbase?
565 2012-03-11 09:57:30 <Graet> the winning share is submitted to both bitcoind and namecoind - afaik they dont need to talk to each other
566 2012-03-11 09:58:00 <Graet> any other alt chain would have its shares submitted to its *coind
567 2012-03-11 09:58:40 <UukGoblin> hrm yeah, but how can the *coind see that its data was in fact in the bitcoin low-diff block?
568 2012-03-11 09:59:10 <UukGoblin> cause when you merkle-hash all the chains together, you get a unreadable hash in coinbase
569 2012-03-11 09:59:26 <UukGoblin> what tells all the *coinds where their hash is in the auxchain merkle tree?
570 2012-03-11 09:59:56 <UukGoblin> sorry, I probably should've asked that in the beginning
571 2012-03-11 10:00:10 <UukGoblin> (but I wasn't sure how to formulate the question)
572 2012-03-11 10:15:09 <Eliel> UukGoblin: merged mined namecoin blocks obviously need to include all the data needed to verify them.
573 2012-03-11 10:15:32 <Eliel> including merkle branches
574 2012-03-11 10:17:09 <UukGoblin> right. So where could I find specs of this data, and how it's communicated? :-)
575 2012-03-11 10:20:13 <Eliel> other than namecoin source code, I have no pointers to give for those questions :)
576 2012-03-11 10:21:36 <UukGoblin> yeah, I was afraid it'd come to that ;-) ok, thanks anyway :-)
577 2012-03-11 10:22:10 <UukGoblin> anyway it looks like the merged-mining-proxy needs some way of communicating data from other altchains back to namecoind
578 2012-03-11 10:23:32 <Eliel> luke-jr should be pretty informed about it though. He might have a pointer or two to give. I think he said he coded merged mining support from the scratch for Eligius because he didn't like the existing code.
579 2012-03-11 10:23:55 <UukGoblin> cool thanks :-)
580 2012-03-11 10:28:07 <Eliel> UukGoblin: have you seen this yet? https://en.bitcoin.it/wiki/Merged_mining_specification
581 2012-03-11 10:32:10 <UukGoblin> Eliel, yeah
582 2012-03-11 10:32:37 <neofutur> its really undomented. really : https://github.com/vinced/namecoin/blob/mergedmine/contrib/merged-mine-proxy
583 2012-03-11 10:33:30 <UukGoblin> I'll have a look, thanks all :-)
584 2012-03-11 10:35:15 <Stary2001> Does the bitcoin Qt client do verification of blocks in another thread?
585 2012-03-11 10:35:38 <Stary2001> because my blockchain was downloading slowly and I had to up priority to fix it]
586 2012-03-11 10:36:08 <denisx> the slowest thing is the disk IO
587 2012-03-11 10:36:45 <Stary2001> hm.. yeah i guess
588 2012-03-11 12:30:40 <Joric> i'm trying to reindex blockchain want to write my own hash map that would be faster than in STL (for finding transactions/addresses) what hash should i use? rehashing hash256^2 seems shady, should i use self-balancing trees or something?
589 2012-03-11 12:36:42 <Zarutian> Joric: why not a trie?
590 2012-03-11 12:39:20 <Zarutian> Joric: with adaptable length of the prefixes at each node.
591 2012-03-11 12:40:17 <Joric> well, i'll try
592 2012-03-11 12:40:45 <Joric> will check it against rb-tree in std::map
593 2012-03-11 12:55:55 <Joric> linus delivers https://lkml.org/lkml/2012/3/8/495
594 2012-03-11 13:07:24 <sipa> tcatm: afaik gavin can create testnet alerts, yes
595 2012-03-11 13:39:44 <SpeedBus> heyy
596 2012-03-11 13:54:08 <gribble> New news from bitcoinrss: sipa opened pull request 930 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/930>
597 2012-03-11 14:29:47 <TD> are there any forum moderators around?
598 2012-03-11 14:34:55 <archo47> uhm no
599 2012-03-11 14:51:10 <sipa> Eliel: I know little about Qt, so I can't review the code, but it sounds reasonable to merge for 0.6
600 2012-03-11 14:52:15 <Eliel> I know about as little about Qt as you do I think :)
601 2012-03-11 14:52:33 <Eliel> thankfully, that patch didn't need anything Qt specific.
602 2012-03-11 14:54:29 <denisx> why is the wallet.dat flushed with every new block?
603 2012-03-11 14:57:14 <TD> the wallet contains the current chain head
604 2012-03-11 14:57:42 <denisx> ah, ok
605 2012-03-11 14:58:13 <denisx> Flushed wallet.dat 284ms
606 2012-03-11 14:58:26 <denisx> but that looks too long for just saving it
607 2012-03-11 14:59:07 <Eliel> who knows what the bdb is doing.
608 2012-03-11 15:02:35 <lodewijkadlp> I'm going to be selling bitcoins in a user-friendly manner. Can someone advise upon wire-transfers and chargebacks within the EU and from elsewhere towards the EU? I might be willing to accept chargebacks if there's reasonable legal framework to handle them.
609 2012-03-11 15:08:17 <etotheipi_> sipa, I got distracted yesterday and slid past your HMAC deterministic wallet comment
610 2012-03-11 15:08:34 <etotheipi_> can you send me that unit test?   I do plan to implement it, or something close
611 2012-03-11 15:09:23 <TD> lodewijkadlp: you will be running an exchange, in other words?
612 2012-03-11 15:09:34 <sipa> etotheipi_: it's implemented here: https://github.com/sipa/bitcoin/blob/detwallet/src/test/detwallet_tests.cpp
613 2012-03-11 15:09:48 <TD> lodewijkadlp: SEPA transfers can be reversed, as far as I know, eg if the source bank account was compromised.
614 2012-03-11 15:09:57 <TD> lodewijkadlp: so it's dangerous and you will need good risk analysis and verification procedures
615 2012-03-11 15:10:27 <lodewijkadlp> TD in a way. It will have more significant fee's and will not keep credit. Enter bitcoin address, enter money. Or vise versa. Done.
616 2012-03-11 15:10:54 <TD> yeah, so you have to be careful because that will make you an attractive target for cashing out hacked bank accounts
617 2012-03-11 15:11:37 <TD> it's bogus and wrong but banks expect _you_ to block such breaches and give the money back, even though you can't make the source banks more secure. if you receive fraudulently obtained funds too often you will be cut off/shut down
618 2012-03-11 15:11:55 <lodewijkadlp> are there numbers available on the probability? What kind of authentication should I do?
619 2012-03-11 15:11:56 <TD> what is your real goal? There may be better ways to achieve it. Like, with better p2p coin exchanger systems
620 2012-03-11 15:12:20 <TD> well you'd have to do all that research yourself. your bank could advise (they had better know what you are doing if you get any serious volume and banks are a PITA to work with)
621 2012-03-11 15:12:46 <etotheipi_> thanks, sipa
622 2012-03-11 15:13:15 <sipa> etotheipi_: it's possible that i do some changes to it still (bytecode had some comments about it)
623 2012-03-11 15:13:29 <TD> personally i'd rather see somebody set up a really good person-to-person trading system. just having a decent map-based website where individuals can register would be good. several were created but none achieved critical mass
624 2012-03-11 15:13:36 <lodewijkadlp> I've created a user-friendly e-wallet (bitvau.lt, testing fase right now. Please break it and mail admin@bitvau.lt how you did it) (not even (d)dos proof, don't try that please). And people need to get bitcoins with at least the same ease as the wallet works.
625 2012-03-11 15:14:11 <etotheipi_> sipa, no worries... just let me know when you do
626 2012-03-11 15:15:01 <etotheipi_> sipa, I just decided that if I'm going to upgrade the wallet, I'm going to do do a LOT of upgrades...
627 2012-03-11 15:15:08 <etotheipi_> so I'll need to revamp my unit tests
628 2012-03-11 15:15:13 <TD> lodewijkadlp: no offense but if you want people to store money with you, you need to get native english speakers to proof-read your material. eg check the spelling of "encryption"
629 2012-03-11 15:15:31 <TD> services like yours have a chequered past with the bitcoin community
630 2012-03-11 15:15:37 <TD> so you're going to be swimming upstream with it
631 2012-03-11 15:15:46 <etotheipi_> sipa, I'll try to support multiple chains, but I'm finding it complicated:  not so much in code, but in terms of how to present it
632 2012-03-11 15:16:16 <lodewijkadlp> I'm going to be working today on accepting iDeal, a Dutch-only chargeback-free for e-shops payment system. There seems to be an equivalent for some other EU countries. I'll just continue research.
633 2012-03-11 15:16:33 <sipa> etotheipi_: if your wallet concept corresponds to a single chain, no problem; it would already be nice if they are compatible at that level
634 2012-03-11 15:16:53 <TD> lodewijkadlp: be skeptical of anything that claims to be "chargeback free"
635 2012-03-11 15:16:55 <TD> see: dwolla
636 2012-03-11 15:17:19 <sipa> etotheipi_: i believe the use cases for more chains are important, so i'm thinking about how to implement those in the satoshi client, but i suppose that is up to the creator of the application
637 2012-03-11 15:17:51 <etotheipi_> sipa, I'll be implementing multiple chains, I just don't know how I'm going to enable them, if at all
638 2012-03-11 15:18:01 <lodewijkadlp> I believe I caught that in my development already. I'll restart the server soon. I make no claims to accurate spelling though.
639 2012-03-11 15:18:11 <etotheipi_> if I enable them, it may just be as if they are multiple wallets
640 2012-03-11 15:18:38 <etotheipi_> because I'm finding the concept of "subwallets" to be difficult to present in a non-confusing way
641 2012-03-11 15:18:54 <sipa> i mean: it would be nice if you use the same derivation system; so you can import my [m=...]/n/0 node, and have a wallet that corresponds to the the external chain of my n'th account
642 2012-03-11 15:19:00 <TD> lodewijkadlp: make sure you have an absolutely watertight contract with iDeal if you use them, because otherwise you could lose a lot of money. they have to provide iron-clad guarantees of being chargeback free written into your contract at minimum
643 2012-03-11 15:19:08 <etotheipi_> sipa, I plan on doing that :)
644 2012-03-11 15:19:19 <lodewijkadlp> I've read about the Dwolla things but iDeal is a between-banks system. I'm quite confident that it's actually chargeback free. Another advantage is that people that can use it are likely Dutch, which I can sue.
645 2012-03-11 15:20:19 <sipa> etotheipi_: i'm thinking about standardizing a serialization for such tree nodes as well
646 2012-03-11 15:20:31 <sipa> "extended public key" and "extended private key" or so
647 2012-03-11 15:20:46 <lodewijkadlp> Thanks for the heads up though. If there's any more tips regarding accepting payment you can always contact me at admin or lodewijkadlp at bitvau.lt. I'll lurk for a while still.
648 2012-03-11 15:20:52 <TD> like i said:  EU banks have been observed to reverse wire transfers. it doesn't matter if iDeal is between banks. you cannot sue anyone for this because the most common failure mode is that a bank account is hacked/phished by somebody unknown who then tries to cash out via bitcoin. the owner of the bank account in question is innocent
649 2012-03-11 15:21:12 <TD> i'd recommend talking to jered kenna or mark karpeles about their experiences before even thinking about this
650 2012-03-11 15:21:29 <TD> interfacing bitcoin with the banking system is VERY hard, risky, and some people have lost a lot of money trying
651 2012-03-11 15:21:42 <TD> that's why i'd suggest focusing on non-banking ways to get bitcoins to people. like good p2p exchange support
652 2012-03-11 15:21:47 <etotheipi_> sipa, well I'm not doing it just yet:  in fact I'll wait for the Satoshi client to do it, because I have enough other priorities ATM
653 2012-03-11 15:22:42 <sipa> etotheipi_: sure
654 2012-03-11 15:22:53 <etotheipi_> but I'll be happy to discuss it and battle the concepts... I want them to be interoperable
655 2012-03-11 15:23:17 <sipa> etotheipi_: it won't be released before 0.7.0, and 0.6.0 isn't even there :)
656 2012-03-11 15:23:44 <etotheipi_> sipa, great :)  I need to get the RAM reduction efforts moving forward (and perhaps a longer-term solution, as well)
657 2012-03-11 15:24:04 <etotheipi_> so I'm glad I have time before I have to think about this :)
658 2012-03-11 15:31:13 <etotheipi_> sipa, how do you feel about libcoin?  as in... what is it's future in the Bitcoin world?
659 2012-03-11 15:31:41 <etotheipi_> I'm assuming it is *not* being considered for any official place Satoshi client...?
660 2012-03-11 15:32:41 <etotheipi_> I guess, I'm looking at long term blockchain mgmt solutions for Armory, and libcoin seems like a good place to leverage existing work...
661 2012-03-11 15:35:36 <sipa> the libcoin author has done a really nice job with refactoring the satoshi code, imho
662 2012-03-11 15:36:21 <sipa> he did a few things i'm wary about (he wasn't aware reading from an STL data structure required a lock, if it can be simultaneously written to, e.g.)
663 2012-03-11 15:36:46 <sipa> but afaik it is GPL, so we can't just merge it back
664 2012-03-11 15:38:05 <etotheipi_> sipa, I guess that's a big question on my mind:  I would love to merge/leverage code *especially* if it will be maintained and up-to-date
665 2012-03-11 15:41:23 <jrmithdobbs> sipa: ya, i  haven't even looked at his code because saw gpl3
666 2012-03-11 15:41:27 <etotheipi_> although, in the really long term, I would love to accommodate blockchain compression/pruning -- I assume there's no plans for that -- have any of the existing ideas been discussed as a serious "Bitcoin upgrade"?  (such as including unspent-output-tree hashes in the coinbases)
667 2012-03-11 15:43:41 <Joric> added cpp/sqlite parsers https://github.com/joric/pyblockchain
668 2012-03-11 15:44:05 <Joric> etotheipi_, i'm trying to implement memory cached file )
669 2012-03-11 15:44:47 <etotheipi_> Joric, and I'm working on reading the encrypted wallet.dat for migration
670 2012-03-11 15:45:41 <Joric> well there's aes using 3 different libs http://pastebin.com/XWibUePh
671 2012-03-11 15:46:04 <etotheipi_> Joric, I'm just going to leverage the AES I have in my code
672 2012-03-11 15:46:26 <etotheipi_> as I looked through pywallet, I think I figured out how it's done well enough just to integrate it into armory
673 2012-03-11 15:46:38 <etotheipi_> when I'm done, I'll toss you a few BTC :)
674 2012-03-11 15:46:38 <Joric> python version is quite slow i was mostly using openssl dynamic lib
675 2012-03-11 15:47:53 <jrmithdobbs> Joric: you just need to upgrade your system to something that gives openssl hardware accel ;p
676 2012-03-11 15:48:13 <jrmithdobbs> (*BSD or the /dev/crypto patches for linux)
677 2012-03-11 15:48:57 <sipa> etotheipi_: it seems my hmac-sha512 implementation is fully compliant (it passes the rfc test cases)
678 2012-03-11 15:49:27 <sipa> i couldn't find a library that provided it
679 2012-03-11 15:50:14 <etotheipi_> sipa, compliant is good :)  I haven't tried implementing it, but I did mean to ask if you were using the "standard" HMAC algorithm
680 2012-03-11 15:50:45 <sipa> originally I wasn't, but it was hardly any work to make it compliant, so why not
681 2012-03-11 15:53:54 <sipa> etotheipi_: btw, if we do a "serious upgrade" (read: blockchain forking, full network upgrade required), why bother putting it in the coinbase, instead of just an extra block header field?
682 2012-03-11 15:54:18 <etotheipi_> sipa, because I didn't think anyone was considering that kind of upgrade
683 2012-03-11 15:54:54 <etotheipi_> plus, you don't need more than a few of the miners to do it... you don't need every coinbase to hold it
684 2012-03-11 15:55:07 <etotheipi_> in fact, you probably only need 1 in ever 2016 blocks to have it
685 2012-03-11 15:55:18 <tcatm> Block header changes will require rewriting many mining algorithms as most of them assume the nonce to be at a specified position.
686 2012-03-11 15:56:09 <sipa> I doubt that's the problem.
687 2012-03-11 15:57:29 <Joric> added example output to chart.py - http://goo.gl/A53P0
688 2012-03-11 15:58:09 <etotheipi_> I really like the idea of letting the blockchain pruning happen "gracefully" -- i.e. people can hold onto it if they want to, or can start pruning and verify against hundreds of other nodes
689 2012-03-11 15:58:13 <TD> i don't see why block chain pruning needs changes to the protocol
690 2012-03-11 15:58:31 <sipa> TD: it doesn't
691 2012-03-11 15:58:54 <etotheipi_> but I think it's going to require *someone* to start putting data in coinbases
692 2012-03-11 15:59:22 <TD> why? it's a purely local optimization. so we have some archived snapshots of the unpruned, full block chain
693 2012-03-11 15:59:26 <etotheipi_> or maybe just an extra network message for exchanging this info
694 2012-03-11 15:59:28 <TD> you can download them to bootstrap a new node
695 2012-03-11 16:00:03 <TD> the only protocol change is at the network level. nodes need to be able to state they can't serve the chain. we already have it for client mode anyway
696 2012-03-11 16:00:12 <TD> can't serve the full chain, sorry
697 2012-03-11 16:00:24 <TD> you can obviously keep the last few thousand blocks around unpruned or whatever so nodes that leave temporarily can come back
698 2012-03-11 16:00:38 <etotheipi_> TD, I'm thinking about the far future, where it may not be feasible except for the full, super-nodes to hold the whole blockchain
699 2012-03-11 16:01:22 <TD> end users will be on lightweight clients. miners, merchants, bitbanks, whatever, full nodes with pruning to keep the disk space requirements in check. assuming anyone cares about the disk space requirements by that point
700 2012-03-11 16:01:33 <TD> it doesn't seem like a high priority for a long time, at least.
701 2012-03-11 16:03:04 <etotheipi_> it may not be, but I believe it's a critical success factor for Bitcoin in the long-term, which means I wouldn't mind figuring it out, and even trying to implement a solution before it becomes critical
702 2012-03-11 16:03:28 <TD> sure
703 2012-03-11 16:04:14 <etotheipi_> but arguably, I *do* have other priorities right now... I'm just thinking that if I re-implement my blockchain code, it would be cool to reimplement it with pruning
704 2012-03-11 16:04:28 <etotheipi_> but not until there is a *solid* agreeable solution to the problem
705 2012-03-11 16:05:10 <etotheipi_> but I might be living in utopia to believe that will happen
706 2012-03-11 16:05:28 <sipa> well, you can implement pruning; i do not think it's that hard, but you won't be able to answer to getdata/block messages
707 2012-03-11 16:05:36 <sipa> i.e., not function as a full node anymore
708 2012-03-11 16:06:40 <etotheipi_> sipa, understood
709 2012-03-11 16:07:00 <etotheipi_> my concern is more to do with the security of pruning:  I don't want to even tackle it until I know that there are ways to verify against other nodes
710 2012-03-11 16:07:24 <etotheipi_> I feel like there's too many subtle things that can go-wrong/be-attacked if you just prune by yourself
711 2012-03-11 16:08:07 <luke-jr> tcatm: blockchain-fork upgrade will probably be merged-mined-only anyway, so block header won't be directly hashes
712 2012-03-11 16:08:09 <luke-jr> hashed*
713 2012-03-11 16:36:32 <neofutur> searching partners for a a public btcprice server http://p.b.gw.gd/ip/p.php?usdprice=100
714 2012-03-11 16:58:10 <Raccoon> what would be the bit length of the number 10000000000000000000000000
715 2012-03-11 16:58:56 <Raccoon> or 10^25
716 2012-03-11 16:59:10 <Raccoon> looking for 2^X
717 2012-03-11 16:59:43 <etotheipi_> Raccoon, it's approximately a factor of 3 difference:  10^25 is *approx* 2^75
718 2012-03-11 17:00:14 <Raccoon> hmm, wouldn't that be 9^25 then?
719 2012-03-11 17:02:10 <etotheipi_> it's actually a little bit more... factor is 3.3
720 2012-03-11 17:02:19 <etotheipi_> 10^25 is about 2^83
721 2012-03-11 17:03:31 <Raccoon> thanks.
722 2012-03-11 17:03:37 <etotheipi_> Raccoon, you can use any calculator to do the computation for any base
723 2012-03-11 17:03:49 <etotheipi_> you can compute it exactly by doing   log(10^25)/log(2)
724 2012-03-11 17:04:18 <etotheipi_> log(X)/log(Y) = Z -->   Y^Z=X
725 2012-03-11 17:04:53 <Raccoon> thanks :)
726 2012-03-11 17:05:10 <etotheipi_> np
727 2012-03-11 17:06:12 <Raccoon> would you agree to my statements at http://bitcoin-kamikaze.com/?userID=53130&secretKey=tk7P2OTsXUenScsp
728 2012-03-11 17:06:32 <Raccoon> taking http://i51.tinypic.com/9ifpma.png  as the discussed system
729 2012-03-11 17:06:35 <Joric> does not compute
730 2012-03-11 17:06:54 <Joric> funny i have no access to that page
731 2012-03-11 17:08:03 <Raccoon> Try http://bitcoin-kamikaze.com/ then click on MiniChat
732 2012-03-11 17:08:36 <Joric> looks like probably ip filtering i may access it via anonymizer
733 2012-03-11 17:09:58 <Raccoon> I suggest that using a hash key in contrast to the one in that image
734 2012-03-11 17:10:00 <Raccoon> T(2)wo fIv(5)e Th(3)ReE (4)FouR ON(1)E Si(6)x TWo(2) tHrE(3)E gIRaFfE
735 2012-03-11 17:10:28 <Raccoon> offering roughly 2^83 bits of entropy while making it improbable to MD5 collide
736 2012-03-11 17:10:46 <Joric> Raccoon, check out 'honesty proof' on http://goo.gl/2q5My )
737 2012-03-11 17:11:09 <Raccoon> "Entropy in the above example is roughly 27^8 (lotto picks with shuffled number positioning) * 2^35 (mixed case) * 1000 (dictionary common nouns). That comes to 10^25"
738 2012-03-11 17:12:32 <Raccoon> i'm not a fan of that one either.  though its shorter length makes it harder to exploit MD5
739 2012-03-11 17:25:48 <tcatm> http://188.138.99.157/stuff/qtvert13.png Lock icon moved to "MY WALLET" near balance.
740 2012-03-11 17:26:55 <Joric> what happened with window header and window frame?
741 2012-03-11 17:27:39 <tcatm> That's the window managers job.
742 2012-03-11 17:28:38 <[Tycho]> Connectivity meter is way too close the edge
743 2012-03-11 17:28:49 <[Tycho]> *to
744 2012-03-11 17:36:32 <TD> sipa: what do you think about making bitcoin announce the contents of its memory pool with an inv at startup
745 2012-03-11 17:37:29 <sipa> hmm, i don't see any reason not to
746 2012-03-11 17:38:47 <sipa> by "at startup" you mean "when you receive a connection" ?
747 2012-03-11 17:51:07 <[Tycho]> Hello, luke-jr
748 2012-03-11 17:52:46 <TD> sipa: yes, sorry
749 2012-03-11 17:53:04 <TD> sipa: it means if somebody sends you money, then you start the client, you will see it appear immediately
750 2012-03-11 17:53:11 <TD> currently you have to wait for a block if you miss the broadcast
751 2012-03-11 17:54:16 <sipa> I'm wondering if there could be any downside or other reason why Satoshi chose not to do that.
752 2012-03-11 17:55:12 <luke-jr> hi [Tycho]
753 2012-03-11 18:01:06 <TD> sipa: i can't think of any
754 2012-03-11 18:01:11 <sipa> gmaxwell: do you think we should put specific version numbers in rc's?
755 2012-03-11 18:01:13 <TD> sipa: he probably assumed you'd run the software all the time
756 2012-03-11 18:01:27 <sipa> TD: haha, yes, and be mining all the time as well.
757 2012-03-11 18:01:45 <TD> the whole address thing came quite late in the design, i think
758 2012-03-11 18:02:46 <sipa> I think so too; the name "scriptSig" and "scriptPubKey" seem to give a hint in that direction
759 2012-03-11 18:02:53 <gmaxwell> TD: so if the new peer is behind on the chain, the memory pool txn will end up as orphans on the peer.
760 2012-03-11 18:03:10 <TD> yes
761 2012-03-11 18:03:19 <TD> it's more useful for lightweight clients
762 2012-03-11 18:03:27 <gmaxwell> They'll also nail their peers with copies of the ones they take.
763 2012-03-11 18:03:45 <gmaxwell> so it would increase inv traffic somewhat, though I don't know if that will matter much.
764 2012-03-11 18:04:16 <sipa> If you're worried about it, you could limit that rule to cases where the incoming connection advertizes with less or equal number of blocks as you have.
765 2012-03-11 18:04:56 <TD> it'd mean one extra inv per connection
766 2012-03-11 18:05:01 <TD> not a big deal, really
767 2012-03-11 18:06:07 <gmaxwell> hm. Will it tend to make transactions that don't confirm immortal?
768 2012-03-11 18:06:23 <sipa> Probably.
769 2012-03-11 18:06:27 <gmaxwell> e.g. because nodes always coming online and offline will reinv them and introduce them to their peers who've forgotten them.
770 2012-03-11 18:06:40 <sipa> But they live +- forever in miner's nodes as well now.
771 2012-03-11 18:07:11 <sipa> I think unconfirming transactions is a separate problem anyway.
772 2012-03-11 18:07:41 <gmaxwell> I thought they aged out (though, er. I don't remember actually seeing the code for that I do know that you can replace a transaction which doesn't confirm if you stop announcing it and wait a while)
773 2012-03-11 18:08:19 <sipa> In any case, only you and the receiver ever rebroadcast them.
774 2012-03-11 18:08:56 <sipa> So if neither of you does, it will die out in the network communication. But I suppose it may live for a long time in some nodes' memory pools still.
775 2012-03-11 18:08:59 <gmaxwell> sipa: yes, but the reinving would have that effect.
776 2012-03-11 18:09:06 <sipa> Yes.
777 2012-03-11 18:09:19 <sipa> Anyway, got to go.
778 2012-03-11 18:24:04 <luke-jr> sipa: but&
779 2012-03-11 18:42:18 <TD> can anyone see an issue with adding a "pong" that is generated in response to a ping?
780 2012-03-11 18:42:42 <t7> are there other near complete implementations of bitcoin?
781 2012-03-11 18:42:43 <andytoshi> let's people know you're alive, and maybe your IP?
782 2012-03-11 18:43:01 <andytoshi> why do routers ignore ICMP pings these days?
783 2012-03-11 18:43:24 <andytoshi> t7: depends what you mean by "near"
784 2012-03-11 18:43:36 <andytoshi> i think armory is fairly complete
785 2012-03-11 18:44:56 <TD> t7: of a full node? no
786 2012-03-11 18:45:04 <TD> t7: of a lightweight node? bitcoinj isn't bad
787 2012-03-11 18:45:52 <andytoshi> i think mtgox and some other folks have their own private nodes
788 2012-03-11 18:46:05 <andytoshi> for what that's worth
789 2012-03-11 18:46:28 <t7> its just that the bitcoin source is a bit huge and not very modular
790 2012-03-11 18:46:35 <andytoshi> i concur
791 2012-03-11 18:46:55 <andytoshi> i'm working on a cleaner version written in D, but it's nowhere near complete
792 2012-03-11 18:47:05 <andytoshi> and it's a very low-priority project compared to schoolwork and research
793 2012-03-11 18:48:11 <andytoshi> having said that, the satoshi client isn't terrible - it's a million times better than it was a year ago
794 2012-03-11 18:48:18 <andytoshi> the devs have done a lot of cleanup
795 2012-03-11 18:50:58 <t7> i think there are too many unsafe functions in the code (memcpy etc)
796 2012-03-11 18:51:38 <andytoshi> well, 'unsafe' just means you have to be careful
797 2012-03-11 18:51:54 <t7> everyone makes mistakes
798 2012-03-11 18:52:02 <andytoshi> true, but bitcoin is very heavily tested
799 2012-03-11 18:52:57 <t7> i wonder if anyone has tried implementing with coq or something
800 2012-03-11 18:53:24 <andytoshi> i doubt it.. you could ;)
801 2012-03-11 18:53:54 <t7> i remember roconner was working on a haskell client ages ago
802 2012-03-11 18:55:31 <Eliel> I think he progressed pretty far with it. However, he hasn't been here for 3 weeks now.
803 2012-03-11 18:55:32 <t7> i think c# or java or something a little 'safer' would be a step in the right direction
804 2012-03-11 18:56:18 <andytoshi> java's memory model is a nightmare
805 2012-03-11 18:56:30 <andytoshi> i wouldn't want to manage a 1-2Gb blockchain with it
806 2012-03-11 18:56:43 <andytoshi> you might be able to get some traction starting a c# client
807 2012-03-11 19:03:13 <luke-jr> did roconnor ever publish his work?
808 2012-03-11 19:03:38 <denisx> when I send {"method":"getwork","params":[],"id":7548705} to bitcoind the id should be part of the answer, right?
809 2012-03-11 19:04:58 <Eliel> luke-jr: darcs repo is here: http://r6.ca/Purecoin
810 2012-03-11 19:07:30 <luke-jr> e darcs deps & :/
811 2012-03-11 19:08:21 <Eliel> I suspect you got yourself a nice big compilation job if you're using gentoo :)
812 2012-03-11 19:08:30 <Eliel> especially if you haven't used haskell before
813 2012-03-11 19:11:56 <andytoshi> i think most people have used haskell these days.. functional programming is coming back into vogue
814 2012-03-11 19:12:06 <andytoshi> as a complexity managment strategy
815 2012-03-11 19:12:06 <t7> does anyone know a binary file pastebin?
816 2012-03-11 19:12:18 <andytoshi> just base64 it :P
817 2012-03-11 19:12:26 <midnightmagic> uh.
818 2012-03-11 19:12:32 <midnightmagic> how do I get past 170052?
819 2012-03-11 19:12:37 <andytoshi> the short answer is, the FBI keeps shutting them down
820 2012-03-11 19:12:42 <midnightmagic> or 170059 or whatever.
821 2012-03-11 19:12:48 <Eliel> t7: yeah, megaupload was like that.
822 2012-03-11 19:12:51 <midnightmagic> upgrading to 0.6.0rc2 doesn't correct it.
823 2012-03-11 19:12:59 <andytoshi> midnightmagic: sipa has a patch that might help
824 2012-03-11 19:13:15 <midnightmagic> andytoshi: can I go backwards to an earlier version?
825 2012-03-11 19:13:19 <Eliel> t7: well, dropbox works for that though.
826 2012-03-11 19:13:30 <andytoshi> that is, a branch on which he dropped the complexity of a blockchain reorg, to avoid database limits
827 2012-03-11 19:13:46 <midnightmagic> gah brutal.
828 2012-03-11 19:13:49 <andytoshi> midnightmagic: no, the chain is goofed, i believe
829 2012-03-11 19:14:02 <midnightmagic> well i guess i shut down my miners for the time being then. :-( oh well.
830 2012-03-11 19:14:12 <midnightmagic> thanks andytoshi. guess the electricity bill could use a break.
831 2012-03-11 19:14:36 <andytoshi> if you want, you can compile sipa's minireorg branch and use that
832 2012-03-11 19:14:42 <andytoshi> which might get you past 170059
833 2012-03-11 19:14:50 <andytoshi> it worked for me, though i've been informed it's not a sure thing
834 2012-03-11 19:14:54 <luke-jr> Eliel: I *used* to have darcs installed. Not sure when I lost it :p
835 2012-03-11 19:14:58 <midnightmagic> okay, thanks, i'll do that later.
836 2012-03-11 19:16:51 <Eliel> how do you add labels to issues on github? I'd like to add bug and gui labels into my pull request https://github.com/bitcoin/bitcoin/pull/928
837 2012-03-11 19:23:59 <yellowhat> has anyone info on the status of the testnet? is 48968 the current height?
838 2012-03-11 19:36:51 <tribbler> c
839 2012-03-11 19:45:23 <denisx> anyone here using BIP30 already?
840 2012-03-11 19:45:43 <denisx> [Tycho]: how is your BIP30 test going?
841 2012-03-11 19:45:52 <[Tycho]> Hello.
842 2012-03-11 19:45:59 <[Tycho]> I think it's fine.
843 2012-03-11 19:46:22 <denisx> I have now 3 blocks which got not accepted by bitcoind
844 2012-03-11 19:46:33 <denisx> telling me the nBits are not correct
845 2012-03-11 19:46:41 <[Tycho]> It's not 15.03 yet :
846 2012-03-11 19:46:54 <denisx> there is some small part not depending on that
847 2012-03-11 19:46:57 <gmaxwell> denisx: are you talking about testnet?
848 2012-03-11 19:47:00 <denisx> no
849 2012-03-11 19:48:35 <gmaxwell> denisx: the bip30 patches don't touch anything related to nbits checking IIRC.
850 2012-03-11 19:49:14 <denisx> ok
851 2012-03-11 19:49:45 <gmaxwell> Some poolserver software (E.g. pushpool, IIRC) sends up solutions which are not actually blocks.
852 2012-03-11 19:50:06 <gmaxwell> Could it be thats what you're seeing and you're only noticing it now because you're actually paying attention?
853 2012-03-11 19:50:16 <denisx> yeah, push pool does that too, it is called better hash
854 2012-03-11 19:50:24 <denisx> has the first five bytes zero
855 2012-03-11 19:50:32 <denisx> and these are logged in my database
856 2012-03-11 19:50:58 <denisx> and I check if they fit the real target
857 2012-03-11 19:51:40 <denisx> gmaxwell: it started with my BIP30 deployment
858 2012-03-11 19:51:50 <denisx> but it can also be coincidence
859 2012-03-11 19:51:55 <denisx> Iam just checking
860 2012-03-11 19:53:42 <luke-jr> no problems here
861 2012-03-11 19:53:52 <gmaxwell> Ah. Well the bip30 patch itsef doesn't appear to carry any danger of rejections due to nbits. If you've made other upgrades at the same time, then perhaps those but I can't think of anything other than the testnet specific rule changes.
862 2012-03-11 19:54:56 <denisx> can someone take a look? http://pastie.org/3573194
863 2012-03-11 19:55:09 <denisx> thats the debug.output from bitcoind
864 2012-03-11 19:58:20 <denisx> in the debug output I can see that the hash is smaller than the target and it also lists the correct nBits