--- Day changed Sat Dec 29 2018
00:09 -!- Comstock [Comstock@gateway/vpn/privateinternetaccess/comstock] has quit [Read error: Connection reset by peer]     
00:42 < waxwing> belcher, yeah interesting. but "only applies to p2ep things", i think it's just, only applies to non-standard payments. there's bound to be various other scenarios where it happens right, like someone sending out payouts to lots of recipients.
03:17 < belcher> yes, for batched payments its completely different
03:20 < waxwing> oh and re: UIH , thanks for clearing that up; it's a good name. i wonder, did meiklejohn's paper mention something similar, too? i remember there was a whole middle section about change output identification, wondering whether that was one of them.
03:26 < belcher> i havent read that paper yet
03:28 < waxwing> fistful of bitcoins
03:29 < waxwing> https://cseweb.ucsd.edu/~smeiklejohn/files/imc13.pdf
03:33 < waxwing> had a quick read again, nothing about that.
03:34 < waxwing> basically their heuristic 2 was a complete botch, they start by assuming that change addresses are defined by one-time-use(!), but insisting that only one such address exists in the output, but that still didn't work and they added other caveats. it's all a bit of a mess, but basically they massaged the data to get reasonable clustering coming out.
03:35 < belcher> from other summaries of that paper iv read, the thing they most advance with is the CIOH
03:37 < waxwing> sure; but without some kind of change identification i don't think they could get their clustering going. of course the fun part of the paper is where they bought stuff of silk road et al :)
03:38 < waxwing> oh, maybe they didn't actually do silk road actually, but various merchants/sites
03:38 < waxwing> pretty sure some other researchers did though, at one point
03:47 < belcher> yes, change address identification has to go with CIOH
03:48 < belcher> if the change address is always kept secret, and addresses arent reused, then CIOH doesnt help much
03:51 -!- asymptotically [~asymptoti@gateway/tor-sasl/asymptotically] has joined #joinmarket     
04:02 -!- GitHub42 [GitHub42@gateway/service/github.com/x-ijcokzbcokojbpqe] has joined #joinmarket     
04:02 < GitHub42> [joinmarket-clientserver] AdamISZ closed pull request #271: Improve manually selected tx fee documentation (master...patch-1) https://git.io/fhIaq
04:02 -!- GitHub42 [GitHub42@gateway/service/github.com/x-ijcokzbcokojbpqe] has left #joinmarket []     
04:02 -!- GitHub1 [GitHub1@gateway/service/github.com/x-awfdxjchcpgpeqft] has joined #joinmarket     
04:02 < GitHub1> [joinmarket-clientserver] AdamISZ pushed 2 new commits to master: https://git.io/fhI9d
04:02 < GitHub1> joinmarket-clientserver/master d238172 AlexCato: Improve manually selected tx fee documentation
04:02 < GitHub1> joinmarket-clientserver/master 4fedcdb AdamISZ: Merge #271: Improve manually selected tx fee documentation...
04:02 -!- GitHub1 [GitHub1@gateway/service/github.com/x-awfdxjchcpgpeqft] has left #joinmarket []     
05:26 -!- undeath [~undeath@hashcat/team/undeath] has joined #joinmarket     
06:50 < waxwing> noting this down in case i forget, since doing other stuff but: the whole UIH2 thing seems to imply that alice should use a greedy or similar algo, so that it's more likely she contributes no too-big utxo, then it'll make it more likely that bob succeeds in avoiding triggering that heuristic (he won't need as big of a utxo).
06:50 < waxwing> it's like the more greedily the two parties are sweeping utxos up, the better, i guess
07:27 < waxwing> undeath, ping, put some feedback on #246
07:27 < undeath> waxwing: thank you!
07:28 < undeath> sorry about the screw-up in the list/zip fix
07:29 < waxwing> i guess there's no last() equiv of next() .. guess it wouldn't make sense either. and you can't do next(next()) :)
07:30 < undeath> you can use islice()
07:30 < undeath> you can use islice(iter, start, end)
07:30 < undeath> it's in the itertools recipies
07:31 < waxwing> k, yeah i tried that, see it works. but is there any difference between that and just using list()?
07:32 < waxwing> i guess more efficient with big datasets
07:32 < waxwing> to use iterators
07:32 < undeath> list will create a new list obj
07:32 < undeath> exactly
07:32 < waxwing> k     
07:32 < undeath> using list() will have to touch each single element once
07:33 < waxwing> also i kind of forgot as i got to the last comment, what do you think about changing 'deposit' to 'change' for internal?
07:33 < undeath> that's not accurate for external deposits
07:33 < waxwing> on reflection i think users will indeed get confused to see 'deposit' for change outputs from payments they made to destination outside the wallet.
07:33 < undeath> you wouldn't see anything about addresses anywhere outside the wallet
07:34 < undeath> sending from one internal address to another seems like a very rare edge case to me
07:34 < waxwing> scenario is: i pay 1btc to something outside the wallet, and i get 0.5 change back. it goes into the internal branch of course, but it's marked as 'deposit'.
07:34 < undeath> oh, is it?
07:34 < waxwing> yeah, forget that. focusing specifically on paying to a destination outside the wallet.
07:35 < waxwing> see last paragraph of https://github.com/JoinMarket-Org/joinmarket-clientserver/pull/246#issuecomment-450496142
07:36 < waxwing> we could count it as 'change' purely based on it being the internal branch, deposits are to go to external. another alternative would be to say "we can't know everything" and revert to using the label "used" when we're not sure.
07:38 < undeath> ah, you mean in case you do a direct send (non-cj) with a real output and change back to a internal address
07:39 < waxwing> undeath, yes just direct send, sorry i should have disambiguated 
07:39 < undeath> technically that's a deposit imo, because it doesn't originate from a coinjoin
07:39 < undeath> so it's not really a change output
07:44 < undeath> some kind of 'internal-tx' would be a little more helpful here, but I'm not sure how to accurately identify that
07:44 < undeath> unless we really just check if it's an internal/external address
08:03 < waxwing> undeath, i think you're going to find that all users would interpret a payment back to the wallet as a change output
08:03 < waxwing> a deposit means something coming from outside the wallet.
08:04 < waxwing> that's what "change" means for all other non-coinjoin bitcoin wallets
08:04 < waxwing> and when we use direct send, that's how our wallet is behaving, as an ordinary bitcoin wallet
08:08 < undeath> yeah, I've already added a commit that uses a different label now :)
09:20 < waxwing> undeath, you forgot to import islice from itertools i think?
09:27 < waxwing> also i think i prefer cj-change (to go with cj-out) and just 'change' for the other one. non-cj-change is getting too long.
09:27 < waxwing> (but that second point is not something that bothers me really)
09:45 -!- GitHub186 [GitHub186@gateway/service/github.com/x-wfbiswclhkwmwocz] has joined #joinmarket     
09:45 < GitHub186> [joinmarket-clientserver] AdamISZ pushed 2 new commits to master: https://git.io/fhIpE
09:45 < GitHub186> joinmarket-clientserver/master 596261a undeath: implement wallet-tools display[all] extended usage status
09:45 < GitHub186> joinmarket-clientserver/master 106e384 AdamISZ: Merge #246: display[all] extended usage status...
09:45 -!- GitHub186 [GitHub186@gateway/service/github.com/x-wfbiswclhkwmwocz] has left #joinmarket []     
09:45 -!- GitHub61 [GitHub61@gateway/service/github.com/x-oavjtfuyllpesacq] has joined #joinmarket     
09:45 < GitHub61> [joinmarket-clientserver] AdamISZ closed pull request #246: display[all] extended usage status (master...display-addr-use) https://git.io/fpKiW
09:45 -!- GitHub61 [GitHub61@gateway/service/github.com/x-oavjtfuyllpesacq] has left #joinmarket []     
09:46 -!- GitHub184 [GitHub184@gateway/service/github.com/x-gfyabdmjnixlwbkk] has joined #joinmarket     
09:46 < GitHub184> [joinmarket-clientserver] AdamISZ pushed 1 new commit to master: https://git.io/fhIpu
09:46 < GitHub184> joinmarket-clientserver/master 481b688 AdamISZ: add missing islice import
09:46 -!- GitHub184 [GitHub184@gateway/service/github.com/x-gfyabdmjnixlwbkk] has left #joinmarket []     
09:50 < undeath> that's what you get for not testing your code >.> sorry about the import
10:05 -!- nothingmuch [~user@109.202.107.10] has joined #joinmarket     
10:07 < nothingmuch> re payjoin, ephemeral nickname or payment information - might be worth looking into magic-wormhole
10:08 < nothingmuch> it generates an ephemeral, low entropy passphrase for SPAKE, to exchange information, implemented in python, and i suspect not hard to make the server based stuff work over IRC instead
10:08 < nothingmuch> (not related to the actual proposal, just the ease of use comments)
10:28 < waxwing> nothingmuch, well i mean we already have ephemeral identities for each session and do ecdh. not sure we'd need to do anything else?
10:29 < waxwing> although i don't doubt there are schemes that might give extra, useful properties
10:34 < nothingmuch> the only benefit is reducing the size of the out of band message
10:37 < waxwing> nothingmuch, i see, thanks
10:38 < nothingmuch> also it's just a cool project to know about, git-annex uses it for pairing of sync over tor, pretty slick
11:19 -!- Comstock [Comstock@gateway/vpn/privateinternetaccess/comstock] has joined #joinmarket     
11:56 -!- GitHub46 [GitHub46@gateway/service/github.com/x-uugxsxggsqmfuqyy] has joined #joinmarket     
11:56 < GitHub46> [joinmarket-clientserver] AdamISZ pushed 1 new commit to master: https://git.io/fhLfb
11:56 < GitHub46> joinmarket-clientserver/master eaac4be AdamISZ: Fix bug in wallet generate in Qt when cancelled...
11:56 -!- GitHub46 [GitHub46@gateway/service/github.com/x-uugxsxggsqmfuqyy] has left #joinmarket []     
12:05 -!- GitHub170 [GitHub170@gateway/service/github.com/x-tokqrnzqgowwsoyv] has joined #joinmarket     
12:05 < GitHub170> [joinmarket-clientserver] AdamISZ pushed 2 new commits to master: https://git.io/fhLJU
12:05 < GitHub170> joinmarket-clientserver/master 5d66d94 Kristaps Kaupe: Check for empty password and cancellation in JoinMarket-Qt wallet generation
12:05 < GitHub170> joinmarket-clientserver/master d3db7dc AdamISZ: Merge #266: Check for empty password and cancellation in JoinMarket-Qt wallet generation...
12:05 -!- GitHub170 [GitHub170@gateway/service/github.com/x-tokqrnzqgowwsoyv] has left #joinmarket []     
12:05 -!- GitHub54 [GitHub54@gateway/service/github.com/x-omojigvpsrprnxdg] has joined #joinmarket     
12:05 < GitHub54> [joinmarket-clientserver] AdamISZ closed pull request #266: Check for empty password and cancellation in JoinMarket-Qt wallet generation (master...qtgeneratepass) https://git.io/fhvGL
12:05 -!- GitHub54 [GitHub54@gateway/service/github.com/x-omojigvpsrprnxdg] has left #joinmarket []     
12:16 < waxwing> just want to mention that #268 needs some review, i won't be able to finish the payjoin thing before that one goes in.
12:32 < arubi> commented my 2c, but will be reading through more tomorrow :)
13:24 -!- raedah [~x@192.30.89.142] has joined #joinmarket     
13:34 -!- Comstock [Comstock@gateway/vpn/privateinternetaccess/comstock] has quit [Quit: Leaving]     
13:39 -!- Comstock [Comstock@gateway/vpn/privateinternetaccess/comstock] has joined #joinmarket     
15:25 -!- undeath [~undeath@hashcat/team/undeath] has quit [Quit: WeeChat 2.3]     
16:00 -!- asymptotically [~asymptoti@gateway/tor-sasl/asymptotically] has quit [Quit: Leaving]     
16:30 -!- GitHub48 [GitHub48@gateway/service/github.com/x-nxnghyigxhdrnqpt] has joined #joinmarket     
16:30 < GitHub48> [joinmarket-clientserver] kristapsk opened pull request #273: Handle already existing wallet file on generate (master...wallet-generate-exists) https://git.io/fhLqA
16:30 -!- GitHub48 [GitHub48@gateway/service/github.com/x-nxnghyigxhdrnqpt] has left #joinmarket []     
17:54 -!- AgoraRelay [~jmrelayfn@p54866B2E.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds]     
18:07 -!- AgoraRelay [~jmrelayfn@p54866B8B.dip0.t-ipconnect.de] has joined #joinmarket     
21:42 -!- nothingmuch [~user@109.202.107.10] has quit [Quit: leaving]     
22:13 -!- Comstock_ [Comstock@gateway/vpn/privateinternetaccess/comstock] has joined #joinmarket     
22:13 -!- Comstock_ [Comstock@gateway/vpn/privateinternetaccess/comstock] has quit [Remote host closed the connection]     
22:14 -!- Comstock_ [Comstock@gateway/vpn/privateinternetaccess/comstock] has joined #joinmarket     
22:14 -!- Comstock [Comstock@gateway/vpn/privateinternetaccess/comstock] has quit [Disconnected by services]     
22:15 -!- Comstock_ is now known as Comstock