1 2014-01-31 05:06:57 <nirom> Is there any system in development for P2P lending at the moment? Something like shared risk pool of bitcoin lending and borrowing?
2 2014-01-31 05:09:46 <Denim-jdev> Who would determine who is allowed to get a loan and who isn't...
3 2014-01-31 05:10:31 <swulf--> pool vote, I presume
4 2014-01-31 05:13:00 <Luke-Jr> why vote? just only loan from people who approve it
5 2014-01-31 05:14:02 <swulf--> The ripple style of lending, you mean?
6 2014-01-31 05:18:50 <nirom> I'm more looking for something decentralized
7 2014-01-31 05:19:55 <nirom> Denim: lending would be from someone to someone else direct, but the risk could be shared
8 2014-01-31 05:20:38 <nirom> This way, it could ease the requirements on the borrower, and the risk of the lender
9 2014-01-31 05:23:44 <Luke-Jr> nirom: that's .. um .. not sane?
10 2014-01-31 05:23:58 <Luke-Jr> "someone else" gets all the benefit of the loan, but no responsibility/risk?
11 2014-01-31 05:24:05 <Luke-Jr> s/no/only some of/
12 2014-01-31 05:24:25 <nirom> why not, if everyone cautioning the loan share a bit of the interest of the loan?
13 2014-01-31 05:24:57 <Luke-Jr> they should all agree to the loan too
14 2014-01-31 05:25:05 <Luke-Jr> in which case, it's from many someones direct
15 2014-01-31 05:25:05 <nirom> of course
16 2014-01-31 05:25:43 <Luke-Jr> disclaimer: I am not endorsing usury/interest.
17 2014-01-31 05:26:01 <nirom> let's say I wan to lend my bitcoin today, I have not choice but to get 100% exposure to risk
18 2014-01-31 05:26:37 <Denim-jdev> it'd be too easy to game the system
19 2014-01-31 05:26:37 <nirom> I'm thinking a bit like the insurance companies do with re-assurance
20 2014-01-31 05:27:01 <Denim-jdev> he rips you off
21 2014-01-31 05:27:01 <Denim-jdev> if you don't bear the risk, you loan to a friend
22 2014-01-31 05:27:01 <Denim-jdev> other people in the pool have to pay for it
23 2014-01-31 05:27:01 <nirom> how so?
24 2014-01-31 05:28:16 <nirom> then can someone begin to borrow without a complicated web of trust
25 2014-01-31 05:29:53 <nirom> what could be used to caution the borrowed amount?
26 2014-01-31 05:31:23 <nirom> today, I can lend my btc through a margin exchange, as long as I beleive the exchange operator won't rip me off. It is still centralized trust.
27 2014-01-31 05:31:45 <Luke-Jr> â¦
28 2014-01-31 05:32:10 <Luke-Jr> nirom: you cannot get more money out of something than you put into it
29 2014-01-31 05:32:28 <Luke-Jr> if Joe lends 5 BTC, and Fred runs off with it, Joe isn't getting it back
30 2014-01-31 05:32:32 <Luke-Jr> no matter how you do it
31 2014-01-31 05:32:35 <nirom> I know
32 2014-01-31 05:33:36 <nirom> this is risk management
33 2014-01-31 05:33:47 <nirom> there will always be bad dept
34 2014-01-31 05:34:30 <nirom> but it might be interesting to average the risk over many depts instead of a single one.
35 2014-01-31 05:37:14 <nirom> So have you heard of anything like that, existing project?
36 2014-01-31 05:39:04 <Denim-jdev> you'd need government regulation
37 2014-01-31 05:39:42 <nirom> I dought it would help....
38 2014-01-31 05:41:10 <ZjP> Unless you can impose a cost on those who run off with the principal (loss of non-trivially replacable reputation-- [credit score]) is sufficient
39 2014-01-31 05:41:18 <ZjP> Then you have nothing
40 2014-01-31 05:41:46 <nirom> but credit score doesn't cross countries
41 2014-01-31 05:41:58 <ZjP> You are confused. I am talking in general.
42 2014-01-31 05:43:21 <nirom> what about using existing coins as collateral?
43 2014-01-31 05:43:31 <nirom> would it make sense
44 2014-01-31 05:43:48 <nirom> Let's say I "freeze" 50 btc to borrow 100 btc
45 2014-01-31 05:43:56 <ZjP> Thank you for the free 50 btc.
46 2014-01-31 05:44:48 <nirom> well you wont be able to touch it until the network agrees to release it?
47 2014-01-31 05:45:11 <ZjP> I can touch the 100 BTC minus my 50 with I give up on since I'm not paying the 100 back equals a free 50 btc for me.
48 2014-01-31 05:45:18 <nirom> yep
49 2014-01-31 05:45:54 <nirom> then what can be used s collateral?
50 2014-01-31 05:46:25 <nirom> I've been thinking about this all week, it is making me crazy
51 2014-01-31 05:47:03 <harddata> does anyone in here fully understand conditional bitcoin transfers?
52 2014-01-31 05:47:29 <nirom> not yet
53 2014-01-31 05:47:37 <nirom> what are they
54 2014-01-31 05:47:55 <harddata> so my understanding is that you can put conditions on a bitcoin transfer
55 2014-01-31 05:48:01 <Denim-jdev> uses scripting to add extra functionality
56 2014-01-31 05:48:07 <harddata> yes
57 2014-01-31 05:48:21 <harddata> is it implemented at all currently?
58 2014-01-31 05:48:25 <Denim-jdev> one classic example is 2 out of 3 sig
59 2014-01-31 05:48:28 <Denim-jdev> yes
60 2014-01-31 05:48:33 <nirom> so you release the btc when a condition is met
61 2014-01-31 05:48:48 <harddata> so i understand how to do that with multisig
62 2014-01-31 05:49:02 <harddata> but if I wanted to use realitykeys
63 2014-01-31 05:49:07 <harddata> to validate a piece of information
64 2014-01-31 05:49:25 <harddata> is there a way I could make it self-authorizing (so I wouldn't need an arbiter?)
65 2014-01-31 05:50:06 <harddata> basically, I want to send a transaction where instead of having a 3rd signature, it only sends if bitcoin hits a price of 1200 USD/BTC
66 2014-01-31 05:50:31 <ZjP> nope.
67 2014-01-31 05:50:49 <ZjP> that would require someone to act as the bitcoin price oracle
68 2014-01-31 05:50:56 <Denim-jdev> problem is you have no way of inputting what bitcoin price is
69 2014-01-31 05:51:10 <harddata> https://www.realitykeys.com/
70 2014-01-31 05:51:13 <harddata> that's how
71 2014-01-31 05:51:24 <harddata> assuming they're reliable
72 2014-01-31 05:51:27 <harddata> and i figured it out
73 2014-01-31 05:51:31 <harddata> they publish the private key
74 2014-01-31 05:51:32 <ZjP> And that's your third party....
75 2014-01-31 05:51:35 <Denim-jdev> that's basically a 3rd signature...
76 2014-01-31 05:51:57 <harddata> so what other things can I do with scripting besides multisignature
77 2014-01-31 05:52:08 <nirom> good night all
78 2014-01-31 05:52:35 <Denim-jdev> I think two way transactions that only go through together
79 2014-01-31 06:08:04 <harddata> or transactions that lock
80 2014-01-31 06:15:49 <mappum> how long are typical reorgs? seems like there are a few orphan blocks, but do any ever get blocks after them?
81 2014-01-31 06:16:13 <mappum> not counting intentional fork attacks
82 2014-01-31 06:16:32 <mappum> *a few orphan blocks per day
83 2014-01-31 06:16:59 <jcorgan> https://blockchain.info/charts/n-orphaned-blocks
84 2014-01-31 06:17:24 <jcorgan> actually, i don't think that's what you were asking
85 2014-01-31 06:17:44 <mappum> i'm looking for depth of non-main forks
86 2014-01-31 06:17:51 <jcorgan> yeah
87 2014-01-31 06:41:12 <Diablo-D3> http://www.marscoin.org/
88 2014-01-31 06:41:15 <Diablo-D3> now thats interesting
89 2014-01-31 07:04:54 <Luke-Jr> Diablo-D3: the idea is, yeah, but the params suck
90 2014-01-31 07:06:14 <Diablo-D3> yeah
91 2014-01-31 07:06:16 <Diablo-D3> what I should do
92 2014-01-31 07:06:22 <Diablo-D3> is a pure science coin
93 2014-01-31 07:06:30 <Diablo-D3> that privately funds next gen science
94 2014-01-31 07:08:34 <Luke-Jr> Diablo-D3: yawn.
95 2014-01-31 07:08:41 <Luke-Jr> if people want to fund next gen science, they can just fund it.
96 2014-01-31 07:08:56 <Diablo-D3> fine, how about just taking over the global financial system?
97 2014-01-31 07:14:28 <ZjP> I think that's called bitcoin
98 2014-01-31 07:16:11 <harddata> what's the most innovative use of the blockchain that you've seen
99 2014-01-31 07:16:51 <Diablo-D3> ZjP: I dont think bitcoin can make it
100 2014-01-31 09:15:48 <epscy> re MarsCoin I don't get it
101 2014-01-31 09:19:01 <wumpus> just one of the many wacky altcoin ideas
102 2014-01-31 09:21:02 <epscy> it's the "move the coin to mars at some point in the future bit i don't get"
103 2014-01-31 09:21:05 <wumpus> it appears that the altcoins thrive on some viral message or meme, rather sensible implementation ideas, and there is a lot of competition
104 2014-01-31 09:21:19 <epscy> surely that would just lead to a fork
105 2014-01-31 09:21:28 <epscy> EarthMarsCoin and MarsMarsCoin
106 2014-01-31 09:21:41 <wumpus> humanity going to mars is a very popular meme
107 2014-01-31 09:21:45 <epscy> but you are right, probably best not to take it seriously
108 2014-01-31 09:23:31 <wumpus> technically it doesn't make sense to start a coin for mars before people are actually on mars and trading...
109 2014-01-31 09:24:12 <wumpus> they aren't even simulating the lightspeed delay involved
110 2014-01-31 09:26:58 <wumpus> so, another pointless rename instead of a useful experiment
111 2014-01-31 11:34:10 <stonecoldpat> anyone here from lancaster uk?
112 2014-01-31 11:41:29 <epscy> they all hang out in #bitcoin-uk-lancs
113 2014-01-31 11:46:01 <stonecoldpat> epscy: lol
114 2014-01-31 11:51:21 <t7> hehe
115 2014-01-31 13:21:24 <sneak> /a
116 2014-01-31 13:21:26 <sneak> hi
117 2014-01-31 13:21:27 <sneak> oops
118 2014-01-31 13:21:29 <sipa> /b
119 2014-01-31 13:21:39 <sneak> sipa: 'sup
120 2014-01-31 13:21:45 <sneak> did you miami?
121 2014-01-31 13:21:53 <sneak> ACTION had too much work to be able to go
122 2014-01-31 13:21:55 <sipa> no
123 2014-01-31 13:36:50 <jgarzik> mornin'
124 2014-01-31 13:36:59 <jgarzik> Miami appeared fun for others
125 2014-01-31 14:19:43 <gavinandresen> I just tagged and pushed v0.9.0rc1
126 2014-01-31 14:20:42 <gavinandresen> ⦠and am working through getting gitian-building up and running again in my VirutalBox environment
127 2014-01-31 14:21:14 <michagogo> cloud|gavinandresen: \o/
128 2014-01-31 14:21:35 <michagogo> cloud|I'll see if I can get it built and signed before I need to go
129 2014-01-31 14:37:24 <michagogo> cloud|Okay, looks like I don't have enough time :-/
130 2014-01-31 14:38:43 <michagogo> cloud|With the new pulling out linux dependencies thing, plus the fact that we're doing win32, I need to do all the gbuilds for all the inputs. I'll get it built and sigs pushed tomorrow night.
131 2014-01-31 14:45:42 <TD> hey there
132 2014-01-31 14:45:56 <TD> ACTION feels ....... unemployed
133 2014-01-31 14:53:11 <petertodd> TD: ha, you too?
134 2014-01-31 14:53:25 <TD> yup!
135 2014-01-31 14:53:28 <petertodd> gavinandresen: comments on dust-b-gone?
136 2014-01-31 14:53:29 <TD> 7.5 years at google, over today
137 2014-01-31 14:53:39 <petertodd> TD: heh, we're all quitting dayjobs...
138 2014-01-31 14:53:46 <TD> it's the new black :)
139 2014-01-31 14:54:09 <petertodd> TD: got a plan?
140 2014-01-31 14:54:46 <TD> i have. it's a cunning plan. it's also .... not going to be public for a little while yet. suffice it to say, it involves more of what i do now
141 2014-01-31 14:55:04 <petertodd> TD: ha, good luck
142 2014-01-31 14:55:17 <TD> as the HR lady escorted me out she asked what i'd be working on, and said "oh cool, i look forward to reading about what you get up to in a positive context"
143 2014-01-31 14:55:19 <TD> lol
144 2014-01-31 14:55:50 <petertodd> sounds like HR alright... being from a smaller company my one joked that she better not see me in the paper in handcuffs
145 2014-01-31 14:55:52 <gavinandresen> ⦠in a positive context....
146 2014-01-31 14:56:29 <gavinandresen> petertodd: I sent dust-b-gone some old Faucet wallet dust last week, haven't checked to see if it got swept
147 2014-01-31 14:56:44 <petertodd> gavinandresen: it just did
148 2014-01-31 14:57:00 <petertodd> gavinandresen:
149 2014-01-31 14:57:01 <petertodd> bd5c650f11b0c67365efc6dfb039fbf23f9187455d1e9283b6dab83ccc7cb719
150 2014-01-31 14:57:43 <gavinandresen> petertodd: great! the big issue with dust-b-gone or any other tool that requires direct access to your wallet is trust⦠I can read the dust-b-gone code and trust it, but most people can't
151 2014-01-31 14:58:07 <petertodd> gavinandresen: indeed, not an easy one to solve either
152 2014-01-31 14:58:15 <gavinandresen> (afk for a bit, moving kidsâ¦.)
153 2014-01-31 14:58:42 <petertodd> gavinandresen: fwiw I noticed btc guild mined a OP_RETURN dust-b-gone a few days back; must have upgraded
154 2014-01-31 14:59:01 <TD> petertodd: you're going to work on mastercoin?
155 2014-01-31 14:59:48 <petertodd> TD: yup, mostly but not exclusively them
156 2014-01-31 15:00:18 <TD> interesting. doing what? my impression was they hadn't really got much code beyond some scraps put together by bounty hunters
157 2014-01-31 15:00:19 <petertodd> TD: fortunately they seem to agree that we're better off growing the decentralized finance space in general rather than fighting over marketshare
158 2014-01-31 15:00:24 <TD> (if you don't mind me asking)
159 2014-01-31 15:01:01 <petertodd> TD: hence why my job title is chief scientist - as i put it I said Mastercoin was an idea sufficiently stupid as to be indistinguishable from fraud, and JR replied "great! wanna fix it?"
160 2014-01-31 15:01:08 <TD> haha
161 2014-01-31 15:01:15 <TD> well, good to know we're on the same page there then :)
162 2014-01-31 15:03:05 <petertodd> indeed - msc will likely either be based on ethereum, or implement something like it
163 2014-01-31 15:03:48 <petertodd> which is interesting because that can lead to a msc team working on UI's and other stuff to actually use the tech, and a ethereum(alike) team working on underlying implementation tech
164 2014-01-31 15:03:53 <TD> yeah so far ethereum is the only alt coin i've seen that has both an interestingly different design and actual code too. though i'm not really convinced about the design, still. but we'll see.
165 2014-01-31 15:04:39 <petertodd> vitalik and some of the other ethereum team live in the same city as me so...
166 2014-01-31 15:04:39 <sipa> i'm unsure about ethereum scalability and incentives, but indeed, it's one of the few that actually tries to experiment with more than a few constants or hash algorithms
167 2014-01-31 15:05:09 <TD> yes, that's very convenient
168 2014-01-31 15:05:11 <petertodd> sipa: yup, and I think the scalability can be fixed
169 2014-01-31 15:06:25 <petertodd> sipa: what I'm not so sure is that the finances can be fixed - obviously forking it is trivial and could wipe out early investors - true of msc too
170 2014-01-31 15:13:14 <TD> lol
171 2014-01-31 15:13:15 <TD> http://www.forbes.com/sites/andygreenberg/2014/01/31/silk-road-vendor-filing-claim-for-seized-bitcoins-argues-he-sold-only-legal-items/
172 2014-01-31 15:13:24 <TD> the lulz to be had from that will be epic to watch
173 2014-01-31 15:14:14 <helo> no doubt
174 2014-01-31 15:15:52 <petertodd> 6ca0e5c777b67927ef1ab958bcc2b80adcc14dbe6b3a77988c98366bda5391b0 <- yup, btcguild is definitely running 0.9
175 2014-01-31 15:19:32 <TD> cool
176 2014-01-31 15:21:14 <petertodd> good chance to try out a double-spend attack - ~35% odds per try of a success
177 2014-01-31 15:25:34 <TD> how so?
178 2014-01-31 15:26:06 <petertodd> TD: trivial to submit to http://eligius.st/~wizkid057/newstats/pushtxn.php, wait a few seconds to propagate, and then broadcast normally
179 2014-01-31 15:26:22 <TD> oh, you mean because it's still considered non-standard
180 2014-01-31 15:26:38 <petertodd> yup, and 0.9 density is way below the percolation theshold
181 2014-01-31 16:17:44 <jcorgan> URL for the proposed 0.90rc1 release notes? i can't find them in the repo
182 2014-01-31 16:19:14 <sipa> https://gist.github.com/laanwj/8693503
183 2014-01-31 16:19:31 <jcorgan> tks
184 2014-01-31 16:22:24 <jcorgan> the cli option changed from -tor to -onion, but the config file option stayed onlynet=Tor, correct?
185 2014-01-31 16:22:41 <jcorgan> it chokes on onlynet=onion
186 2014-01-31 16:31:45 <jcorgan> nm, it seems these are separate options
187 2014-01-31 16:34:15 <wumpus> yes, onlynet=tor (not Tor)
188 2014-01-31 16:53:32 <stonecoldpat> does anyone here talk to bitcoin using java? (JSON-RPC) and if so, what api do you recommend for doing it ?
189 2014-01-31 16:57:18 <kinlo> yes, self-written :)
190 2014-01-31 16:58:17 <kinlo> you can do that in like 10 lines of code so...
191 2014-01-31 17:11:36 <kuzetsa> stonecoldpat: json is a style of rest API
192 2014-01-31 17:11:46 <kuzetsa> you don't need java to use rest APIs (json or otherwise)
193 2014-01-31 17:12:41 <kuzetsa> http://en.wikipedia.org/wiki/Representational_state_transfer <-- rest APIs (multiple references in the wikipedia article to json, but it's not the only type of rest)
194 2014-01-31 17:13:17 <stonecoldpat> kinlo: having an attempt at it now
195 2014-01-31 17:13:31 <kuzetsa> for json itself, the "jansson" library (c library) works quite well --> http://www.digip.org/jansson/
196 2014-01-31 17:13:33 <stonecoldpat> kuzetsa: Yeah, i'm trying to use the rpc calls, to get some of the JSON data
197 2014-01-31 17:13:36 <kuzetsa> jansson saves you quite a lot of lines of code
198 2014-01-31 17:13:40 <stonecoldpat> ill have a look at that now
199 2014-01-31 17:14:02 <kuzetsa> ãµãrï½ï½ãï½ï½ï½
200 2014-01-31 17:14:07 <kuzetsa> ãï½ï½
201 2014-01-31 17:14:14 <stonecoldpat> i normally use PHP for JSON stuff, so a bit unfamiliar with doing it in java
202 2014-01-31 17:14:20 <kuzetsa> sorry, accidentally switched my keyboard input IME to japanese
203 2014-01-31 17:14:25 <stonecoldpat> haha
204 2014-01-31 17:14:36 <stonecoldpat> i had that installed when i played ffxi centuries ago
205 2014-01-31 17:15:12 <kuzetsa> stonecoldpat: you don't need java
206 2014-01-31 17:15:17 <kuzetsa> java has nothing to do with json
207 2014-01-31 17:15:32 <stonecoldpat> oh i know, but i want to do some analysis on the json data, so I thought it may be a bit faster than php
208 2014-01-31 17:15:48 <stonecoldpat> so make the rpc calls to get the blockchain data, then do some work on it
209 2014-01-31 17:15:58 <kuzetsa> any json parsing library will be fine
210 2014-01-31 17:16:58 <kuzetsa> I can't remember what I was trying to type but messed up by having the wrong language turned on :(
211 2014-01-31 17:21:08 <kinlo> stonecoldpat: I do use a json library
212 2014-01-31 17:21:47 <kinlo> you don't want to do the json decoding yourself, you cannot do json encoding in 10 lines, certainly if you want to be fully correct
213 2014-01-31 17:22:22 <stonecoldpat> yeah, I wouldn't be doing the decoding
214 2014-01-31 17:22:30 <stonecoldpat> I just mean, what way do people connect to the bitcoind server?
215 2014-01-31 17:22:38 <stonecoldpat> like, an example of a post request to get the mining info
216 2014-01-31 17:22:43 <stonecoldpat> and to pass in parameters
217 2014-01-31 17:23:01 <stonecoldpat> https://en.bitcoin.it/wiki/API_reference_(JSON-RPC) for java just gives an example of how to use the authenticator - but nothing else
218 2014-01-31 17:24:14 <CheckDavid> Hi
219 2014-01-31 17:24:18 <CheckDavid> Hi, I am looking to hire a web developer to create something similar to localbitcoins.com for peer 2 peer e-currency exchange, with escrow, without a local focus.
220 2014-01-31 17:25:53 <TD> stonecoldpat: you can also parse the block chain directly using bitcoinj
221 2014-01-31 17:26:03 <TD> stonecoldpat: depends what kind of analysis you want to do
222 2014-01-31 17:26:26 <stonecoldpat> TD: oh really? i was reading it and from what I gathered it didnt depend on the full blockchain
223 2014-01-31 17:26:55 <TD> it doesn't *require* the full block chain, but it can still download and process it.
224 2014-01-31 17:27:10 <TD> in git master there's also (since today) a postgres backend that indexes the UTXO set
225 2014-01-31 17:27:15 <TD> so you can query address balances
226 2014-01-31 17:28:27 <stonecoldpat> oh thats cool, for Bitcoinj, you still run bitcoind in the background?
227 2014-01-31 17:28:37 <TD> no
228 2014-01-31 17:28:53 <TD> it's independent of bitcoind (well, it will go find one from the p2p network to download the block chain from of course)
229 2014-01-31 17:29:03 <TD> so if you have a local synced bitcoind it's faster to download it from that local node
230 2014-01-31 17:29:54 <stonecoldpat> oh thats awesome, i'm having a look at the github now for it, will download it and give it a spin
231 2014-01-31 17:30:10 <TD> please read the docs on the website for more info
232 2014-01-31 17:30:59 <stonecoldpat> will do, its already warned me that it does not qualify me to write production applications ;)
233 2014-01-31 17:31:22 <TD> after the getting started tutorial, nope, hehe
234 2014-01-31 17:31:27 <TD> though it doesn't sound like you want to throw other peoples money around
235 2014-01-31 17:31:39 <TD> there's an IRC channel for it too, #bitcoinj
236 2014-01-31 17:31:53 <stonecoldpat> no no just want to learn how to manipulate the blockchain data and do stuff with it
237 2014-01-31 17:32:27 <TD> sure
238 2014-01-31 17:32:40 <Goonie> gavinandresen: can you have a look at the log of your payment request generator?
239 2014-01-31 17:34:33 <gavinandresen> Goonie: I could⦠why?
240 2014-01-31 17:35:04 <Goonie> gavin: I sent in a paymentmessage some minutes ago. txid is ad024a4ffde70a06dd4110b5c964b230292849aab51c3c861bcba93d8bda0d8c
241 2014-01-31 17:35:17 <Goonie> I did not receive an ack. Maybe you can see why?
242 2014-01-31 17:37:21 <gavinandresen> Goonie: could be a problem on the server end, I see: PHP Warning: MemcachePool::get(): Invalid key in /home/gavin/public_html/payACK.php on line 22
243 2014-01-31 17:37:49 <Goonie> Do you think you can fix it?
244 2014-01-31 17:38:18 <gavinandresen> Goonie: yes, should be able to fix it in the next few hours
245 2014-01-31 17:38:35 <Goonie> gavinandresen: nice, thanks
246 2014-01-31 17:38:52 <TD> Goonie: have you received an ACK before?
247 2014-01-31 17:38:59 <TD> i mean, i'm wondering if this is a bug in our code rather than gavin's
248 2014-01-31 17:39:08 <TD> gavinandresen: are you back in the USA now?
249 2014-01-31 17:39:21 <Goonie> TD: I just implemented submitting a payment message via HTTP(s)
250 2014-01-31 17:39:22 <gavinandresen> TD: yes, back home
251 2014-01-31 17:39:36 <Goonie> TD: So no, I did not receive an ack via http(s) before
252 2014-01-31 17:39:41 <Goonie> TD: but via BT it worked
253 2014-01-31 17:39:46 <Goonie> I mean *works*
254 2014-01-31 17:39:58 <TD> that wasn't to gavin's server. gavinandresen, are you using merchant_data as the memcache key?
255 2014-01-31 17:40:27 <gavinandresen> TD: I don't remember. All the server code is on github.....
256 2014-01-31 17:40:32 <Goonie> Ah indeed I do not copy merchant_data already
257 2014-01-31 17:40:44 <TD> Goonie: you shouldn't have to. the PaymentSession code is meant to do that
258 2014-01-31 17:40:50 <gavinandresen> https://github.com/gavinandresen/paymentrequest/tree/master/php
259 2014-01-31 17:41:20 <gavinandresen> ⦠err, I mean https://github.com/gavinandresen/paymentrequest/tree/master/php/demo_website
260 2014-01-31 17:41:27 <TD> thanks
261 2014-01-31 17:41:35 <Goonie> TD: I'm testing with the app, not wallettool
262 2014-01-31 17:42:13 <TD> yeah so the memcache key is the merchant_data
263 2014-01-31 17:42:21 <TD> Goonie: doesn't matter. your code is using PaymentSession, right?
264 2014-01-31 17:42:37 <Goonie> TD: no it isn't
265 2014-01-31 17:42:40 <gavinandresen> would be good if it told you if it can't find the key⦠"patches welcome"
266 2014-01-31 17:43:02 <Goonie> TD: just for pki validation so far
267 2014-01-31 17:43:20 <TD> well, that's the bug then. not any issue with gavin's code. you need to be using PaymentSession.
268 2014-01-31 17:43:44 <TD> that's what it's there for, i mean :) you can override the sendPayment(URI, Protos.Payment) method to customize the HTTP behaviour
269 2014-01-31 17:43:58 <TD> like to replace it with your own HTTP wrapper code
270 2014-01-31 17:50:20 <Goonie> gavinandresen: I send to https://bitcoincore.org/~gavin/payACK.php , I get a 404: Not found
271 2014-01-31 17:50:30 <TD> Goonie: does the overriding of the method give you what you need?
272 2014-01-31 17:50:47 <TD> Goonie: that error also means the merchant_data resulted in no hit in the memcache
273 2014-01-31 17:52:15 <TD> Goonie: don't try and reimplement this yourself, you need to be using PaymentSession otherwise it's not going to work right. so let me know what's needed so you can use it. i know about the HTTP thing, subclassing it for your needs should work fine. anything else?
274 2014-01-31 17:52:36 <Goonie> TD: I'm afraid PaymentSession is far from what I need
275 2014-01-31 17:52:57 <TD> in what ways? the only one i remember is changing how http requests are made
276 2014-01-31 17:53:04 <Goonie> TD: Remember everything is already implemented because I implemented it for the BT protocol.
277 2014-01-31 17:53:25 <Goonie> TD: Now I'm just configuring the parts in a different way
278 2014-01-31 17:53:49 <TD> you should have been able to subclass PaymentSession to do BT as well. it sounds like your reimplementation is incomplete ...
279 2014-01-31 17:54:26 <Goonie> well it is obviously
280 2014-01-31 17:55:46 <TD> :(
281 2014-01-31 17:56:17 <TD> i think if you override protected ListenableFuture<Ack> sendPayment(final URL url, final Protos.Payment payment) then you should be able to use PaymentSession for any kind of transport that supports two way communication: http, bluetooth, nfc, etc
282 2014-01-31 17:56:50 <Goonie> Futures in general don't work for Android
283 2014-01-31 17:57:17 <TD> i am skeptical of this claim :)
284 2014-01-31 17:57:25 <Goonie> Because if your activity goes away there is no way to re-create the same future
285 2014-01-31 18:04:07 <TD> Goonie: you realise you can just stuff something into a global variable, right?
286 2014-01-31 18:04:25 <TD> Goonie: in your activity, have a static global that holds the future for the currently pending payment request submission
287 2014-01-31 18:04:37 <TD> if the activity is reconstructed, it just reads the future out of the global and continues
288 2014-01-31 18:04:52 <Goonie> Hacky hacky...
289 2014-01-31 18:05:13 <Goonie> Did you know that Android can have multiple instances of the same activity?
290 2014-01-31 18:06:39 <spokesz> I'm looking for an exchange api with similar functionality as mtgox's v1 trades api w/ 'since' parameter
291 2014-01-31 18:07:59 <TD> it's not hacky
292 2014-01-31 18:08:03 <TD> it's how things are meant to work
293 2014-01-31 18:08:11 <TD> your app is only designed to be submitting a single payment at once
294 2014-01-31 18:08:29 <TD> yes, it can create multiple instances of the same object when rotating and things. but only one is meant to be active on screen
295 2014-01-31 18:09:44 <TD> Goonie: you have a singleton application instance, right? it's not any different to that
296 2014-01-31 18:09:45 <pierreatwork> Regarding the payment protocol, did anyone have any input on the proposed extension of BIP-0070 for recurring payments?
297 2014-01-31 18:10:18 <TD> pierreatwork: sure, people replied on the thread
298 2014-01-31 18:11:23 <pierreatwork> TD: right, but not on Stéphane's follow-up email
299 2014-01-31 18:11:58 <pierreatwork> We were eager to start hacking on it but wanted to make sure we're headed in a sane direction
300 2014-01-31 18:13:51 <TD> ok, i just replied
301 2014-01-31 18:14:09 <Goonie> TD: I think we had that discussion already.
302 2014-01-31 18:15:08 <pierreatwork> TD: Thanks! FYI, we'll like do BitcoinJ first
303 2014-01-31 18:15:10 <TD> probably :)
304 2014-01-31 18:15:35 <TD> pierreatwork: great! feel free to extend the wallet protobuf format to record subscription info
305 2014-01-31 18:15:51 <TD> Goonie: there we go, another reason to use PaymentSession .... people are going to extend it with new features.
306 2014-01-31 18:16:11 <pierreatwork> :-)
307 2014-01-31 18:16:42 <Goonie> TD: I will send patches as well
308 2014-01-31 18:16:55 <TD> Goonie: w.r.t the Future, if the user task switches away and the Android app is shut down whilst the HTTP request to do the payment submission is outstanding, I think it's OK to leave that as an edge case. most apps won't resume HTTP requests that were midflight if they are killed and restarted, no need for your app to be different in this regard
309 2014-01-31 18:17:20 <TD> so keeping the PaymentSession in a global variable is fine
310 2014-01-31 18:17:45 <TD> then you can have the AndroidHttpPaymentSession or BluetoothPaymentSession subclass that manages submitting the Payment and receiving the PaymentAck
311 2014-01-31 18:26:07 <Goonie> gavinandresen: I verified that I set the correct merchant data. Still I get the 404 not found. Your merchant data for my last request was [111, 47, -80, 1, -53, -3, 39, 84, -84, 94, -30, -54, 46, -2, 96, -62, 53, 50, 101, 97, 51, 99, 49, 102, 99, 53, 50, 56, 49]
312 2014-01-31 18:26:33 <TD> Goonie: can i see the code?
313 2014-01-31 18:26:42 <gfawkes_> *sigh* i fucking hate software development.. i thought it would be a great career, and for a brief time it has been
314 2014-01-31 18:27:06 <Goonie> TD: Yes, give me a minute
315 2014-01-31 18:28:09 <Diapolo> I finally managed it to re-setup Pidgin to use freenodes hidden service, took me that long I can only give a short "Hi there!" before I need to leave -_-.
316 2014-01-31 18:32:39 <Goonie> TD: I pushed it to my payment protocol branch. I don't know how much you already reviewed. The BIP70 work starts at commit 4962b8a7e175912aad88055d9cd55d664f8f794a
317 2014-01-31 18:33:37 <Goonie> Sending payment messages to HTTP is commit 32a8173f43c7f1e498515c74043c7c8af6135e49 and then c66765bda639fed64614da2828630b04452f249b
318 2014-01-31 18:34:43 <TD> for the intent, why don't you just serialize the entire PaymentRequest protobuf?
319 2014-01-31 18:34:50 <TD> seems like you're duplicating its structure
320 2014-01-31 18:35:42 <TD> Goonie: btw there isn't really any logging in gavin's code. wallet-tool works, so there's clearly something going wrong in your reimplementation
321 2014-01-31 18:36:45 <Goonie> did you verify wallet-tool just now?
322 2014-01-31 18:38:21 <Goonie> The PaymentIntent is an app-internal structure. I stores things that a PaymentRequest doesn't have
323 2014-01-31 18:39:01 <TD> yeah, sure, but you could also store the entire serialized protobuf. otherwise as new fields get added you'll have to keep adding code to parcel them up
324 2014-01-31 18:39:56 <Goonie> Well the other I'd need to round-trip via protobuf
325 2014-01-31 18:40:09 <Goonie> Parceling is Android native, so that's the better format to use
326 2014-01-31 18:40:56 <TD> why? it's almost exactly the same, but different and creates more work
327 2014-01-31 18:41:04 <Goonie> And I'd still needs parcels for the fields that are not in the protobuf, so its also more complicated
328 2014-01-31 18:42:06 <Goonie> I currently don't even have any protobuf generation in my project
329 2014-01-31 18:42:41 <TD> you don't need to generate anything .... you have read the data from somewhere, right? just keep it as a byte array and then turn it into data structures after the intent is loaded
330 2014-01-31 18:43:53 <Goonie> well I can change the way how it parcels at any time
331 2014-01-31 18:44:39 <Goonie> So far, it was just a matter of 2 minutes to create the parceling stuff
332 2014-01-31 18:45:55 <TD> yes i just tried wallet-tool and it worked
333 2014-01-31 18:46:41 <TD> hmm well more than two mins as it's not quite working yet :) i guess, could you print out the protobuf you receive from the server and the protobuf you upload? in string form, as that will print it out in a useful way
334 2014-01-31 18:46:52 <TD> that way we can check that merchant_data is really being duplicated exactly
335 2014-01-31 18:48:21 <Goonie> Hmmm, payment messages are not that complicated. Does Gavins generator require a memo from my side?
336 2014-01-31 18:48:27 <TD> no
337 2014-01-31 18:48:39 <TD> the code is here
338 2014-01-31 18:48:40 <TD> https://github.com/gavinandresen/paymentrequest/blob/master/php/demo_website/payACK.php
339 2014-01-31 18:48:55 <TD> as you can see the only path that creates 404 is if the memcache lookup for the merchant_data doesn't work
340 2014-01-31 18:49:01 <TD> implying it's not being sent correctly
341 2014-01-31 18:51:28 <Goonie> looks like merchantData is the only thing its using from the payment message
342 2014-01-31 18:52:55 <TD> yes
343 2014-01-31 19:00:23 <Goonie> gavinandresen, TD: I know what the issue probably is... I used a payment request from several days ago -- did you empty your database since then?
344 2014-01-31 19:01:06 <TD> ah. it's a memcache
345 2014-01-31 19:01:13 <TD> try with a fresh request
346 2014-01-31 19:01:40 <Goonie> so how long does it hold the "session"?
347 2014-01-31 19:02:25 <gavinandresen> $memcache->set($id, $params['ACK_message'], 0, 60*60*24);
348 2014-01-31 19:02:32 <gavinandresen> ⦠so 24 hours
349 2014-01-31 19:02:42 <Goonie> thanks, that explains
350 2014-01-31 19:03:44 <TD> ah yes. i figured you'd be testing by scanning the qrcode and getting a fresh request
351 2014-01-31 19:08:25 <Goonie> Nice, with a fresh request it works.
352 2014-01-31 19:08:34 <TD> great
353 2014-01-31 19:09:26 <Goonie> TD, gavinandresen: we should really add some numeric value to the ack message... currently I expect the string "ack" for a positive result, and "nack" for a negative.
354 2014-01-31 19:10:34 <TD> the string where?
355 2014-01-31 19:10:38 <TD> you shouldn't try to parse the memo field
356 2014-01-31 19:10:45 <TD> a PaymentACK means the merchant accepted it. there is no NACK
357 2014-01-31 19:11:28 <Goonie> TD: You suggested the numeric value yourself
358 2014-01-31 19:12:03 <TD> ... for a future version. don't try and require some magic strings in the memo fields, that won't work
359 2014-01-31 19:12:34 <Goonie> TD: Yeah I'll probably change that
360 2014-01-31 19:13:16 <TD> yeah don't go beyond the spec. maybe in future there will be some explicit way to communicate back errors but currently there is none
361 2014-01-31 19:14:55 <Goonie> Hu? Sure I will go beyond spec, and I already to that. Its a field for experimentation.
362 2014-01-31 19:17:43 <TD> yeah ok, sure, i meant don't go beyond it in ways that would make things incompatible with other implementations
363 2014-01-31 19:18:01 <gavinandresen> aclocal-1.11: couldn't open directory `m4': No such file or directory
364 2014-01-31 19:18:01 <gavinandresen> Anybody know why I'm getting an error gitian building linux-deps.yml ? Error is: CDPATH="${ZSH_VERSION+.}:" && cd . && /bin/bash /home/ubuntu/build/qrencode-3.4.3/use/missing --run aclocal-1.11 -I m4
365 2014-01-31 19:19:16 <gavinandresen> I'm sure I'm missing a dependency in the VM, but I can't figure out what it is (I've got make / autoconf / m4....)
366 2014-01-31 19:30:29 <TD> ACTION hates sonos
367 2014-01-31 22:53:20 <BlueMatt> how do tor peers seed currently?
368 2014-01-31 22:53:29 <BlueMatt> isnt there a static set of onion addresses
369 2014-01-31 22:53:34 <BlueMatt> or do they still use dnsseeds?
370 2014-01-31 22:53:50 <sipa> they still use dnsseeds, but by connecting to them rather than resolving them
371 2014-01-31 22:54:07 <sipa> and send a getaddr, wait for a response, and disconnect
372 2014-01-31 22:54:33 <kinlo> does bitcoin already support ipv6 dnsseeds?
373 2014-01-31 22:54:42 <sipa> yes
374 2014-01-31 22:55:02 <sipa> but it can only learn about onion peers through addr messafes
375 2014-01-31 22:55:06 <sipa> messages
376 2014-01-31 22:55:45 <gmaxwell> we really ought to add some onion-seeds. E.g. like the seeds lists but with onion hosts.
377 2014-01-31 22:56:10 <kinlo> what about adding txt-record based dns seeds? that would allow portnumbers too
378 2014-01-31 22:56:23 <BlueMatt> sipa: ahh...ewwww
379 2014-01-31 22:56:25 <gmaxwell> otherwise bootstraping with onlynet=tor is basically non-functional.
380 2014-01-31 22:56:38 <BlueMatt> sipa: be better off connecting to a static onion node and then sending addr and disconnecting
381 2014-01-31 22:56:47 <sipa> why?
382 2014-01-31 22:56:50 <BlueMatt> gmaxwell: yes, that should be pretty trivial to do
383 2014-01-31 22:56:57 <BlueMatt> sipa: not mitm-able by your exit node
384 2014-01-31 22:58:14 <gmaxwell> static onion is superior because it is a single point of trust instead of something where any of N parties could cheat.
385 2014-01-31 22:58:33 <sipa> fair enough
386 2014-01-31 22:59:12 <gmaxwell> In particular, tor reuses exit circuits, so its likely that if you make N connections out all at once they'll end up going through the same exit.
387 2014-01-31 22:59:18 <kinlo> that requires a trusted onion server, but that is probably more secure then the current dnsseed
388 2014-01-31 22:59:34 <gmaxwell> well, N trusted ones, no need for there to only be one.
389 2014-01-31 22:59:53 <gmaxwell> and then they become 'only moderately trusted'.
390 2014-01-31 23:06:46 <jcorgan> +1 for onion seeds; i've never been able to bring up an onlynet=tor without addnode or connect
391 2014-01-31 23:10:12 <jcorgan> i've got a long-lived, well-connected EC2-based hidden node that I'd be happy to contribute in some way