--- Day changed Sun Oct 28 2018 01:08 -!- Lightsword_ [~Lightswor@107.170.253.193] has joined #joinmarket 01:08 -!- Lightsword [~Lightswor@107.170.253.193] has quit [Ping timeout: 244 seconds] 01:08 -!- M1 [~Michail@michail.com] has quit [Ping timeout: 244 seconds] 01:09 -!- Lightsword_ is now known as Lightsword 01:14 -!- M1 [~Michail@michail.com] has joined #joinmarket 03:27 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Excess Flood] 03:27 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket 03:30 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Excess Flood] 03:31 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket 04:32 -!- undeath [~undeath@hashcat/team/undeath] has joined #joinmarket 06:17 -!- stoner19 [stoner19@gateway/vpn/privateinternetaccess/stoner19] has quit [Excess Flood] 08:01 < waxwing> undeath, do you consider #197 finished? i can do a couple of quick tests if so and then ack. 08:02 < undeath> I _think_ it should work in every way intended now 08:02 < undeath> but I haven't had any time to actually test it with ygrunner 08:03 < waxwing> oh i see, you mean referring to test setups. yes i can check that. 08:41 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Excess Flood] 08:43 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket 08:44 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Excess Flood] 08:44 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket 09:11 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Excess Flood] 09:12 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket 09:19 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Excess Flood] 09:20 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket 09:21 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Excess Flood] 09:22 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket 09:23 < undeath> waxwing: if you have any suggestions for an improved help message? Would you like to include the actual effect for each possible command? 09:24 < waxwing> yeah i think along those lines. hang on, i have to push a fix, because i notice the rpcwallet thing created a bug 09:30 < waxwing> ok all should be fine now. 09:31 < undeath> thank you once again for all your testing! 09:33 < waxwing> not at all. it looks like you've finally fixed that mixdepth nastiness. i remember thinking at a few points that you really need two 'pointers' like that to address it fully, but got lost under everything else i guess. 09:34 < waxwing> but yeah cleaning up the help message would be nice. get it printed nicer, and state in a more exact way how the syntax for each method works. in some cases it's almost a hint rather than clear instructions. 09:34 < waxwing> anyway not a high prio. 09:35 < undeath> true, some of the usage is more a kind of trial and error 09:36 < undeath> i think it could be implemented much nicer with argparse, but that'd probably be a weekend's work 09:41 < undeath> in argparse you can define subcommands, so you can actually have different options for each command 12:16 -!- reallll [~belcher@unaffiliated/belcher] has joined #joinmarket 12:19 -!- belcher_ [~belcher@unaffiliated/belcher] has quit [Ping timeout: 264 seconds] 12:27 -!- lukedashjr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket 12:28 -!- lukedashjr [~luke-jr@unaffiliated/luke-jr] has quit [Excess Flood] 12:29 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Ping timeout: 244 seconds] 12:30 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket 12:35 < waxwing> shall i set -r here? 12:35 < waxwing> also belcher is/are the relay bot(s) functional? 12:36 < waxwing> bot seems to be up on agora 12:37 < waxwing> is alexcato around anywhere? i guess if i want to talk to him i'll have to use github 12:38 < waxwing> i'd really like to herd the metaphorical cats of #166 #172 #174 so we can finally make a release and get some rather needed fixes out there. 12:38 < undeath> -r should be ok, haven't seen any spam in probably a week 12:39 -!- mode/#joinmarket [+o waxwing] by ChanServ 12:39 <@waxwing> damn how do you do it again? it's not "/set" 12:39 < undeath> /mode #joinmarket -r 12:39 < undeath> i think 12:39 <@waxwing> oh mode, that was it, thanks 12:40 -!- mode/#joinmarket [-r] by waxwing 12:40 -!- mode/#joinmarket [-o waxwing] by ChanServ 12:42 < undeath> should I rebase #166? 12:42 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Excess Flood] 12:42 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket 12:45 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Excess Flood] 12:46 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket 12:50 < waxwing> i dunno, guess can't hurt, i need to start reading. istr a can of worms being kind of opened :) 12:58 -!- reallll is now known as belcher_ 12:58 < belcher_> alexcato can also be reached on reddit PM 13:33 < waxwing> undeath, ok i've read through and it makes sense to me (again). one thing, i think you can remove this TODO as it's addressed, right: https://github.com/JoinMarket-Org/joinmarket-clientserver/pull/166/files#diff-cc68ef76b7c622c229f24ede1540d990R9 13:34 < undeath> will do 13:34 < undeath> just finished rebasing 13:35 < waxwing> second, i don't really think we can set it as default, unless we finesse. i know this was already discussed but: if we set it as default and everyone's using the same default value, it's too much of a clustering effect imo and too gameable. my feeling is: have it as non-default to start with, then let people try it over time and thus give us a chance to assess. 13:37 < undeath> removed the todo 13:37 < undeath> the standard/non-standard should probably be discussed in the pr, I'm a bit out of the loop here right now 13:38 < waxwing> apart from that i think i understand the diff. will prob. give it a few tests tomorrow. 13:38 < waxwing> well i think it was, a bit? let me check. 13:39 < undeath> we were afraid makers would just cluster around the upper limit, right? 13:39 < undeath> *default upper limit 13:39 < waxwing> yes, see the last part of belcher_ 's comment here https://github.com/JoinMarket-Org/joinmarket-clientserver/pull/166#issuecomment-410065705 and my response in the next one 13:40 < waxwing> as i see it, it's only bad (and yes i agree it *is* bad) if it's both default and static 13:40 < waxwing> if it's not default, and someone chooses to set it, with their own value, that's fine in my book. a quantization (using multiple levels) is even better, but obv more complex. 13:41 < belcher_> how about randomly chosen per user? 13:41 < belcher_> its more complicated though 13:41 < waxwing> belcher_, the way undeath 's coded it, the user can set it; it's the same kind of setting as for current tumbler (-x). i just think we should not set it as the default order selection algo, so that with no options set, everyone does it, at the same values. 13:42 < waxwing> and also, if it's not the default algo, people will only start using it gradually (and consciously) and that's good so we can assess if there's some unexpected effect (well, somewhat!) 13:42 < belcher_> i see 13:43 < belcher_> that makes sense, and with enough messages so users know it exists 13:50 < waxwing> hmm, here's a thought: the fundamental problem of a system like JM is that identities are not allowed for anonymity; and that means reputation is impossible. but what if you could have identities that asserted they were tied to *some* previous identity, without specifying which one? 13:50 < waxwing> like imagine keeping a record of all the bots that *successfully* completed joins with you, and then later someone could show a token that proved that they were in that set, but in zero knowledge. 13:51 < waxwing> like some variant on a ring signature. 13:51 < waxwing> it doesn't solve someone doing 100s of bots, but it does give you some evidence of non-maliciousness. 13:55 < waxwing> i suppose it could partially address bot duplication too, though. and obv you'd need some kind of policy that allows newcomers, which makes it trickier, but perhaps that's not such a big deal. 13:57 < undeath> it addresses the problem of malicious makers setting a very low fee, so they get favoured by the current algorithm 13:58 < waxwing> undeath, you mean the PR? 13:58 < undeath> yes 13:58 < waxwing> well yes ofc, that's the point :) my little essay above is about something different :) 13:58 < undeath> oh 13:59 < waxwing> apologies for non-sequitur. when i start thinking about JM i always come back to that central problem. 14:03 < waxwing> well i retract 'something different'. it's overlapping of course, but my thoughts above are more focused on the other way sybil manifests (people can sybil to block, instead of sybil to grab more joins with low prices) 14:05 < belcher_> waxwing also an idea iv had is for takers to take into account fidelity bonds made by makers 14:05 < belcher_> so makers lock up some bitcoins in time locked addresses, and that can be a costly identity 14:05 < belcher_> similar to what we were thinking of for coinswap servers 14:06 < belcher_> that allows for some kind of reputation 14:09 < waxwing> yes it's not unreasonable, except it's a footprint on chain ... so obv the advantage is blocking duplication, but i'm hoping you can get somethign that doesn't leave footprints like that. 14:13 < waxwing> iow not 'i am the same bot as you did tx X with' but 'i am one of the bots you did a tx with'. hmm now i put it like that i realise it's always problematic, where you can't really ever prove 'i participated in one jm tx', if it's limited to only txs that the Taker participated in, it's too small a set. hmm. 14:13 < waxwing> and JM txs can ofc always be faked, so doing something more global doesn't really work (if you could even figure out how!) 14:18 -!- GitHub99 [GitHub99@gateway/service/github.com/x-jwqopkptntcuqsdo] has joined #joinmarket 14:18 < GitHub99> [joinmarket-clientserver] undeath opened pull request #202: update travis main ubuntu distribution (master...update-travis-distri) https://git.io/fxQnT 14:18 -!- GitHub99 [GitHub99@gateway/service/github.com/x-jwqopkptntcuqsdo] has left #joinmarket [] 14:18 < undeath> wb github \o/ 14:22 < waxwing> wb? 14:25 < undeath> welcome back 14:26 < waxwing> :) 14:33 -!- dx25 [~dx25@67-3-130-246.omah.qwest.net] has joined #joinmarket 17:18 -!- arubi [~ese168@gateway/tor-sasl/ese168] has quit [Ping timeout: 256 seconds] 17:21 -!- arubi [~ese168@gateway/tor-sasl/ese168] has joined #joinmarket 17:39 -!- undeath [~undeath@hashcat/team/undeath] has quit [Quit: WeeChat 2.2] 18:47 -!- AgoraRelay [~jmrelayfn@p5DE4ADBF.dip0.t-ipconnect.de] has joined #joinmarket 20:45 -!- lukedashjr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket 20:46 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Ping timeout: 252 seconds] 20:50 -!- lukedashjr is now known as luke-jr 21:55 -!- lukedashjr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket 21:56 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Ping timeout: 244 seconds] 22:00 -!- lukedashjr is now known as luke-jr