1 2015-10-06 01:58:33 <jcorgan> waiting for bitcoin-dev list to break out into the intellectual heights of "Is not!" "Is too!" argumentation. Oh wait...
2 2015-10-06 01:59:12 <alpalp> screw you guys, im going home!
3 2015-10-06 02:27:04 <instagibbs> Time like these when batch mode isn't enough to keep sanity
4 2015-10-06 02:29:44 <floopflorg> if I have a bitcoind instance running with -disablewallet enabled, what is the best way to see incoming transactions? Is polling the rawmempool RPC call the best way? On that note, are there any good alternative bitcoind implementations better suited for running for a service?
5 2015-10-06 02:31:05 <jimpo> I belive there is a ZMQ socket with transaction notifications https://github.com/bitcoin/bitcoin/blob/master/doc/zmq.md
6 2015-10-06 02:34:09 <jcorgan> yes, this was recently merged into master, however
7 2015-10-06 02:46:48 <floopflorg> jimpo: That is exactly what I was looking for. jcorgan: indeed, it looks like this was added just 20 days ago.
8 2015-10-06 02:47:45 <floopflorg> jcorgan: if it is in master, will it be included in the next release?
9 2015-10-06 02:48:14 <jcorgan> yes
10 2015-10-06 02:48:41 <floopflorg> meanin 0.12 or 0.11.1?
11 2015-10-06 02:48:45 <jcorgan> 0.12
12 2015-10-06 02:49:06 <jcorgan> which i think is tentatively planned around end of year, but i'm not positive on that
13 2015-10-06 02:49:50 <floopflorg> ok, I imagine this project would be in dev till at least then anyway. It would be feasible to run with zmq compiled from master on testnet right?
14 2015-10-06 02:51:23 <jcorgan> sure
15 2015-10-06 03:12:22 <floopflorg> jcorgan: Thanks for the help, and the work, I see you started that
16 2015-10-06 03:12:55 <jcorgan> yeah, a long time ago. jgarzik improved it substantially.
17 2015-10-06 03:13:41 <floopflorg> well, there will be at least one person banging on that code soon
18 2015-10-06 03:14:36 <jcorgan> it existed as a PR for a very long time, i think some others have been using it in production
19 2015-10-06 03:19:32 <floopflorg> well good, hopefully I will not have to bug you or jgarzik about it too much then
20 2015-10-06 05:42:32 <hardyred> Hey quick question - if I am playing around with the bitcoin RPC interface what is the best method for finding the balance of a random account's address? Should I just sum the spendable UTXO's or is there a better way?
21 2015-10-06 05:43:53 <Luke-Jr> hardyred: bitcoin addresses do not have balances
22 2015-10-06 05:44:01 <Luke-Jr> despite the misinformation various notorious websites show
23 2015-10-06 05:45:00 <hardyred> I meant the keys associated with the address Luke-jr
24 2015-10-06 05:45:06 <hardyred> Poor phrasing sorry
25 2015-10-06 05:47:17 <Luke-Jr> hardyred: keys also do not have balances; I know what number you are asking about, but my point is that number is meaningless
26 2015-10-06 05:48:36 <hardyred> Luke-Jr: Well.. Luke I am playing around with Bitcoind for the first time here so bear with me. I am trying to figure out how to reflect ACCOUNT BALANCES - in your opinion what would be the most efficient way to calculate this?
27 2015-10-06 05:49:13 <Luke-Jr> account balances can be seen with the getbalance RPC, but accounts are deprecated
28 2015-10-06 05:49:33 <hardyred> I mean for accounts that are not your own
29 2015-10-06 05:49:35 <Luke-Jr> if you need accounts, you should probably implement them yourself in some SQL database
30 2015-10-06 05:49:50 <hardyred> What is used now, wallets?
31 2015-10-06 05:49:58 <Luke-Jr> hardyred: uh, ask whoever runs the accounts?
32 2015-10-06 05:50:05 <Luke-Jr> the blockchain contains ZERO information about accounts
33 2015-10-06 05:50:20 <Luke-Jr> that is entirely local to the PC running them
34 2015-10-06 05:50:38 <hardyred> My terminology is poor Luke-Jr I really do get that and I apologize
35 2015-10-06 05:51:27 <Luke-Jr> to get information about someone else's wallet, you also need (some) private information from their wallet
36 2015-10-06 05:52:53 <hardyred> So let's take a very basic "regular" user site like (don't kill me for saying this Luke-Jr, Blockchain.info or blockexplorer.com) - do they just calculate the addresses by using a database to import the blocks
37 2015-10-06 05:53:15 <hardyred> and then just recalculate based on transactions to and from each individual address?
38 2015-10-06 05:53:47 <Luke-Jr> they might as well make up the misinformation shown
39 2015-10-06 05:54:05 <Luke-Jr> transactions are not from an address ever
40 2015-10-06 05:54:07 <hardyred> I understand that their method is not inherently how Bitcoin works and is not very helpful if you understand the backend. I am just trying to learn what they do in order to get a better understanding
41 2015-10-06 05:54:37 <Luke-Jr> hardyred: everything they do is probably home-grown, which is why it malfunctions constantly
42 2015-10-06 05:54:47 <Luke-Jr> you can find addresses they will claim have negative balances
43 2015-10-06 05:55:26 <hardyred> Luke-Jr: Well I'd be interested to hear your opinion on how a blockchain explorer would function better
44 2015-10-06 05:55:39 <hardyred> Would you not have it display balances at all?
45 2015-10-06 05:57:39 <Luke-Jr> probably correct
46 2015-10-06 05:58:46 <Luke-Jr> yes, I can't think of any blockchain-visible balances to show
47 2015-10-06 06:01:18 <jcorgan> hardyred: an accurate way to present information would be to answer, "what are all the UTXOs I can spend based on the credentials i have stored locally?", or a summary of that
48 2015-10-06 06:02:05 <hardyred> jcorgan: thank you, I am working on a web block explorer as a small project to get familiar with bitcoind/bitcoin in general
49 2015-10-06 06:02:50 <jcorgan> please, please, please, don't ever think or display things in terms of accounts or balances
50 2015-10-06 06:03:02 <hardyred> So if I wanted to make a developers's life any easier what would you display on an explorer then? Information on TXID's, new blocks, old blocks?
51 2015-10-06 06:03:14 <hardyred> Visualization tools?
52 2015-10-06 06:03:16 <jcorgan> all that exists in in bitcoin are UTXOs, some subset of which a user has the ability to spend
53 2015-10-06 06:03:40 <hardyred> jcorgan: Right, I was curious on how the popular blockchain explorers that many users utilized function
54 2015-10-06 06:04:00 <jcorgan> it sounds crazy, but every single one of them get this point wrong
55 2015-10-06 06:04:07 <hardyred> jcorgan: from what I have seen from Blockexplorer is that they just import all the blocks and calculate them from there
56 2015-10-06 06:04:09 <jcorgan> be the first to do it right
57 2015-10-06 06:04:50 <hardyred> jcorgan: I think the reason they do it is because it makes it easier for users to visualize unfortunately
58 2015-10-06 06:05:38 <hardyred> jcorgan: So if I would do it right, do you agree with Luke-Jr and basically never display any kind of balance considering it can be wildly inaccurate?
59 2015-10-06 06:06:30 <jcorgan> the term "balance" carries with it many assumptions and expections of behavior, which draw users toward actions which can cause them to lose bitcoin
60 2015-10-06 06:07:25 <hardyred> jcorgan: So what would you display if you had to think of an ideal site for an explorer (which I am interested in working on to get a better understanding of working with Bitcoin)?
61 2015-10-06 06:07:50 <hardyred> jcorgan: And thank you for answering my questions (I'm sure it can be very annoying)
62 2015-10-06 06:07:58 <hardyred> You too Luke-Jr
63 2015-10-06 06:08:06 <jcorgan> no, it's just a pet peeve of mine
64 2015-10-06 06:09:01 <jcorgan> all the data in the blockchain is there to display. there are blocks and transactions, and changes in the spent vs. unspent transaction set.
65 2015-10-06 06:09:35 <Luke-Jr> hardyred: for a txid, you'd show the input UTXO ids, input scriptSig, and for outputs both addresses as well as UTXO indexes
66 2015-10-06 06:10:30 <Luke-Jr> for an address, you would show the associated txid (if any); if it received more than once, maybe some general idea of which coins are properly confirmed, etc
67 2015-10-06 06:10:41 <Luke-Jr> (note addresses should not receive more than once)
68 2015-10-06 06:11:20 <hardyred> Luke-Jr: They shouldn't but they often do
69 2015-10-06 06:11:35 <Luke-Jr> hardyred: yes, but you wouldn't want to design the page expecting that
70 2015-10-06 06:11:47 <hardyred> Luke-Jr: Blockchain.info, still an incredibly popular web wallet does not even try and change regularly
71 2015-10-06 06:11:50 <Luke-Jr> ie, it should display the info, but look best with the expected use
72 2015-10-06 06:12:13 <Luke-Jr> hardyred: broken webwallets are okay looking weird in a proper block explorer :p
73 2015-10-06 06:12:15 <hardyred> Luke-Jr: Ok great thank you, you gave a perfect description of useful transaction information
74 2015-10-06 06:12:51 <hardyred> Luke-Jr: Is all of that information obtainable through the Bitcoin CMD?
75 2015-10-06 06:13:05 <Luke-Jr> No.
76 2015-10-06 06:13:16 <Luke-Jr> address lookups are not possible with bitcoind
77 2015-10-06 06:13:17 <hardyred> Luke-Jr: Right now I wrote a basic JSON-RPC link between bitcoind and node.js server
78 2015-10-06 06:13:38 <Luke-Jr> unless you use btcdrak's fork, but be aware it is known to be broken
79 2015-10-06 06:13:52 <hardyred> Luke-Jr: So I would have to import that in to a db and then just query whenever a user searches
80 2015-10-06 06:13:58 <Luke-Jr> probably.
81 2015-10-06 06:13:59 <hardyred> Hopefully use some kind of caching :p
82 2015-10-06 06:14:29 <Luke-Jr> I have a custom-made software to act as an external address index, but it's got sharp edges and I do not plan to finish it
83 2015-10-06 06:14:52 <hardyred> What language is it written in if you don't mind me asking?
84 2015-10-06 06:14:54 <Luke-Jr> C
85 2015-10-06 06:15:14 <hardyred> Fun fun fun
86 2015-10-06 06:15:18 <Luke-Jr> 1224 LOC total
87 2015-10-06 06:15:35 <hardyred> I haven't touched C/C++ in a few years, been mostly web dev so I will have to refresh my knowledge on that as well
88 2015-10-06 06:15:49 <hardyred> By chance have you read Andreas's book on Bitcoin?
89 2015-10-06 06:15:50 <Luke-Jr> total about 2 GB of index data, as of 1 year ago :p
90 2015-10-06 06:16:15 <hardyred> Asking whether you would recommend it as a refresher/book for technical information
91 2015-10-06 06:16:21 <Luke-Jr> no, he asked me to go over it and make corrections a while ago, but I never found time
92 2015-10-06 06:16:36 <Luke-Jr> I would recommend the developer docs on bitcoin.org, and the Bitcoin Wiki
93 2015-10-06 06:16:41 <hardyred> Gotcha
94 2015-10-06 06:17:04 <hardyred> So one last question, I can check but unfortunately I have to leave very soon
95 2015-10-06 06:17:28 <hardyred> Earlier you said it would be helpful to display "input UTXO ids, input scriptSig, and for outputs both addresses as well as UTXO indexes" - would this information all be accessible through bitcoind?
96 2015-10-06 06:17:45 <Luke-Jr> if you enable txindex, yes
97 2015-10-06 06:17:46 <hardyred> getrawtransaction
98 2015-10-06 06:17:54 <Luke-Jr> right
99 2015-10-06 06:18:16 <hardyred> Ok perfect, thank you a ton! I will be playing around for the next few days with this
100 2015-10-06 06:18:19 <Luke-Jr> np
101 2015-10-06 06:18:38 <hardyred> Gonna get off here, I believe you are in NA so have a good rest of the evening
102 2015-10-06 06:19:18 <jcorgan> a good block explorer would help people understand the changes to the subset of the UTXO set they care about, as each block arrived
103 2015-10-06 06:19:20 <Luke-Jr> I am in Florida, USA, where it is 2:19 AM :p
104 2015-10-06 07:18:39 <btcdrak> warren: any news on the new list?
105 2015-10-06 07:39:14 <Luke-Jr> ^
106 2015-10-06 07:48:42 <btcdrak> warren: it's really important we get that setup sooner than later. The noise level is ramping up again.
107 2015-10-06 08:24:22 <heretolearn> anyone has advice on how wallets should handle deposits/withdrawals properly in light of latest malleability nuisance
108 2015-10-06 08:25:21 <CodeShark> don't rely on the signed hash for tracking them
109 2015-10-06 08:25:27 <CodeShark> instead, use a normalized unsigned hash
110 2015-10-06 08:25:37 <Luke-Jr> â¦
111 2015-10-06 08:25:50 <Luke-Jr> heretolearn: #bitcoin
112 2015-10-06 08:26:04 <CodeShark> oh, I thought this was #bitcoin, Luke-Jr :p
113 2015-10-06 08:26:19 <heretolearn> Luke-Jr there, thanks
114 2015-10-06 08:26:29 <CodeShark> wrong window...
115 2015-10-06 17:57:41 <jgarzik> warren, aaagh
116 2015-10-06 17:57:49 <jgarzik> warren, The noise on bitcoin-dev is driving me crazy.
117 2015-10-06 17:57:56 <jgarzik> warren, What's the status of that new list?
118 2015-10-06 18:24:53 <stonecoldpat> 30+ emails in 1 day...
119 2015-10-06 18:25:05 <btcdrak> jgarzik: I think we should start spamming warren hourly
120 2015-10-06 18:25:08 <stonecoldpat> does it cost to set up a new list? (ive never done it before
121 2015-10-06 18:25:27 <btcdrak> stonecoldpat: we're waiting for the list to get setup on LF
122 2015-10-06 18:25:42 <jgarzik> btcdrak, ;p
123 2015-10-06 18:25:53 <btcdrak> that last email from V whoever he is is just jawdropping
124 2015-10-06 18:26:08 <stonecoldpat> btcdrak: *fingers crossed* its soon
125 2015-10-06 18:27:46 <moa> mind-numbing
126 2015-10-06 18:52:32 <berndj> this time please don't announce the new list on reddit
127 2015-10-06 19:10:52 <bsm1175321> Hey, look, I can delete a whole thread at once!
128 2015-10-06 19:11:53 <bsm1175321> Here's a funny idea: require a coding puzzle to join the mailing list. "What is the output of the following program"...
129 2015-10-06 19:39:42 <jcorgan> whatever the plan is for the new list, it will attract the same bs if it is not strictly moderated
130 2015-10-06 19:55:29 <jgarzik> jcorgan, A moderation policy will be announced - we are just waiting on the new list to do all that
131 2015-10-06 19:55:56 <jgarzik> jcorgan, (1) announce new policy "stay on topic; off topic goes $here" and (2) create new list $here
132 2015-10-06 19:56:57 <moa> i'm just piping mailing list directly to trash at present
133 2015-10-06 20:42:49 <instagibbs> Piping in here: Don't be afraid to be a tyrant. I know "open list" sounds great, but seriously, just warn people, give them a few strikes, and nuke from orbit.
134 2015-10-06 20:52:49 <jcorgan> +1
135 2015-10-06 21:11:39 <EthanB> Hello I am a university student for software engineering interested in learning as much as I can about bitcoin development to hopefully assist in the near-future; where should I start?
136 2015-10-06 21:14:56 <btcdrak> ethanB: take a look at the github repo.
137 2015-10-06 21:16:00 <bsm1175321> EthanB: I also recommend scouring the bitcoin wiki. http://en.bitcoinwiki.org/
138 2015-10-06 21:16:16 <EthanB> github link?
139 2015-10-06 21:16:53 <btcdrak> github.com/bitcoin/bitcoin
140 2015-10-06 23:10:23 <harding> When signing a raw transaction spending a P2SH output, what data is temporarily placed in the scriptSig field during the signature operation? For P2PKH I know it's the scriptPubKey of the output being spent; but I just read that for P2SH it's the redeemScript. Is that correct?
141 2015-10-06 23:12:44 <phantomcircuit> harding, that doesn't sound right
142 2015-10-06 23:13:22 <harding> phantomcircuit: yeah, that sounded odd to me. I thought we'd just sign OP_HASH160 <Hash160(redeemScript)> OP_EQUAL
143 2015-10-06 23:13:42 <phantomcircuit> yeah im pretty sure we just sign that
144 2015-10-06 23:14:04 <harding> Thanks. I'll check code to confirm.
145 2015-10-06 23:23:41 <SpaceMonkey_> .
146 2015-10-06 23:24:15 <SpaceMonkey_> Bonjour
147 2015-10-06 23:37:15 <harding> Weird, it does seem to sign the serialized redeemscript rather than the scriptPubKey: https://github.com/bitcoin/bitcoin/blob/master/src/script/sign.cpp#L95
148 2015-10-06 23:38:39 <phantomcircuit> harding, uint160
149 2015-10-06 23:38:45 <harding> Oh.
150 2015-10-06 23:39:49 <phantomcircuit> it looks like it's signing just the hash? that seem wrong
151 2015-10-06 23:39:54 <phantomcircuit> no that's right
152 2015-10-06 23:54:47 <warren> btcdrak: the call was rescheduled for Thursday
153 2015-10-06 23:55:35 <warren> Is anyone familiar with a previous analysis of bitcoin core software defect rates, and how it compares to other software?