--- Day changed Fri Jun 28 2019
00:37 -!- midnightmagic [~midnightm@unaffiliated/midnightmagic] has quit [Ping timeout: 264 seconds]     
00:45 -!- reallll is now known as belcher     
00:51 -!- midnightmagic [~midnightm@unaffiliated/midnightmagic] has joined #joinmarket     
02:00 -!- undeath [~undeath@hashcat/team/undeath] has joined #joinmarket     
02:27 -!- belcher [~belcher@unaffiliated/belcher] has quit [Quit: Leaving]     
03:50 -!- lnostdal [~lnostdal@77.70.119.51] has joined #joinmarket     
04:12 < AgoraRelay> [agora-irc/CgRelayBot] [cgan/AlexCato] two reddit questions about the tumbler, to which I do not know the answer and currenlty lack the time to dig through the code. Maybe someone else can chime in:
04:13 < AgoraRelay> [agora-irc/CgRelayBot] [cgan/AlexCato] https://old.reddit.com/r/joinmarket/comments/c2y7um/tumble_to_internal_not_working/ : someone trying to tumble to a destination address, which is within the same joinmarket wallet. Definitely not intended to use the tumbler that way...
04:14 < AgoraRelay> [agora-irc/CgRelayBot] [cgan/AlexCato] https://old.reddit.com/r/joinmarket/comments/c69njt/is_it_ok_to_add_new_funds_to_depth_0_while_the/ : OK to add funds to mixdepth 0 while the tumbler is running? Funds will ofc not be lost, but unsure what the tumbler's behavior is in that case
04:45 -!- harding [~harding@c-73-112-147-20.hsd1.nj.comcast.net] has quit [Quit: leaving]     
05:00 < waxwing> alexcato answered the first one. will do the second one now.
05:11 < waxwing> also done.
05:34 < AgoraRelay> [agora-irc/CgRelayBot] [cgan/AlexCato] thanks!
06:07 -!- belcher [~belcher@unaffiliated/belcher] has joined #joinmarket     
06:20 < belcher> "Yours is probably set to 3 or so, which currently would try to pay more than 100,000 sat/byte for the transaction." <--- you mean 100,000 sat/kilobyte, AlexCato?
08:30 -!- harding [~harding@c-73-196-98-36.hsd1.nj.comcast.net] has joined #joinmarket     
09:26 < waxwing> huh, showerthought about coinjoinxt: you remember that the reason i was proposing a lightning "endpoint" in the ("PTG", proposed transaction graph) was to break subset sum. another idea might be to deliberately create fake subset sum completions with the ancestors of the funding transaction, and/or the promise utxos.
09:27 < waxwing> so an analyst who crunched the entire blockchain graph over the last month, say, might find 20 different sets of txs with subset sums adding up, but 10 of them were designed to be fake, i.e. he sees the wrong boundaries.
09:28 < waxwing> hmm now i write this out, it sounds completely implausible. oh well :)
09:28 < nothingmuch> more like payjoin or more like knapsack mixing?
09:28 < nothingmuch> and why completely implausible?
09:40 < waxwing> i think completely implausible just because if you're trying to "hide in the crowd" like that it's going to be super complex not to drop some fingerprint, but ... yeah no i'm being too pessimistic
09:40 < waxwing> writing software like that is going to be fairly complex anyway, that perhaps isn't so much more complicated.
09:42 < waxwing> cjxt doesn't suffer from having to do multiple rounds with cross-block-interactions i.e. waiting around. only thing is client does hvae to remember state for a while after the single round of interaction. but that's nbd compared to some other things that bitcoin software does.
09:49 < nothingmuch> fwiw i think it's worth exploring such ideas because composition of different tech has the potential to have compounding wins for privacy as well as the underlying complexity (potential for reuse)
09:52 < nothingmuch> e.g. i can imagine at some point joinmarket's p2p IRC comms being replaced with lightning messaging given some sort of general pay-for-relay onion routing, but right now that's obviously insane from a maintenance or impl complexity POV
09:55 < nothingmuch> a few months ago when i was still in the UK i wanted to discuss this stuff with you, but never made it... basically assuming no hard forks, what might the tx graph end up looking like assuming an idealized integration of the various L2 techniques, and what the implications are for privacy
10:42 -!- MaxSan [~four@87.101.92.171] has joined #joinmarket     
11:02 < waxwing> nothingmuch, yes good though, i guess, to consider it at the 50k ft level 'what would the tx graph end up looking like'
11:02 < waxwing> good though*t*
11:21 -!- kristapsk [~KK@gateway/tor-sasl/kristapsk] has joined #joinmarket     
11:27 -!- MaxSan [~four@87.101.92.171] has quit [Ping timeout: 258 seconds]     
12:18 < nothingmuch> waxwing: yep, the tradeoffs of impossibility of hidden inflation, scaling limitations and privacy seem complementary if extrapolating far enough into the future
12:19 < AgoraRelay> [agora-irc/CgRelayBot] [cgan/AlexCato] @belcher: correct, i'll fix my reddit post. It's 100,000 sat/kbyte of course
12:23 < AgoraRelay> [agora-irc/CgRelayBot] [cgan/AlexCato] about Joinmarket on a mobile phone: this could work right now on any phone which can run python3 and SSH. Thats actually how i have set it up: bitcoin core & TOR node one one machine and joinmarket on another one (which is completely restricted, can only talk to the bitcoin core/tor machine via SSH and disallows every other
12:23 < AgoraRelay> communication to
12:23 < AgoraRelay> [agora-irc/CgRelayBot] protect the private keys).
12:24 < AgoraRelay> [agora-irc/CgRelayBot] [cgan/AlexCato] that restricted machine can technically also be a mobile phone, but would instantly lose all the privacy advantages
12:24 < AgoraRelay> [agora-irc/CgRelayBot] [cgan/AlexCato] (and esp. the security advantages)
12:24 < nothingmuch> i wodner if termux is up to the task?
12:26 < nothingmuch> and i'm sure electrum's android dist has some useful tricks for something less hacky than termux
12:28 < AgoraRelay> [agora-irc/CgRelayBot] [cgan/AlexCato] you just need SSH with port forwarding (one for talking via RPC to the bitcoin core node, one to talk to the TOR relay); I dont know termux, but the webpage mentions it uses openSSH, so i see no reason why it wouldnt work
12:28 < nothingmuch> termux is a terminal emulator and apt based environment for android, with packages
12:30 < AgoraRelay> [agora-irc/CgRelayBot] [cgan/AlexCato] in general, i'd just be terrified to have private keys of any meaningful amount of bitcoins on a mobile phone :)
12:31 < nothingmuch> i only meant the setup you described for the 2nd machine might even be transferrable without much work by using termux
12:32 < AgoraRelay> [agora-irc/CgRelayBot] [cgan/AlexCato] yes, i believe so
12:33 < nothingmuch> i suppose theother setup could also be replicated, apparently orbot supports hidden services
12:34 < nothingmuch> which actually makes me think maybe it's better if there's something like socat running on the server to build a reverse tunnel? instead of the phone initiating the connection
12:59 < undeath> I wouldn't trust a phone as far as I can throw it. Those things are the most vulnerable devices out there. It's bad enough you have to store contacts and conversations on there.
13:04 < nothingmuch> +1.. and actually, yeah - what is the motivation behind moving JM to a phone? esp. if you have a node it can connect to? i guess payjoin, and coinjoin spends?
13:05 < AgoraRelay> [agora-irc/CgRelayBot] [cgan/AlexCato] nope, he explicitly had being a maker in mind: https://old.reddit.com/r/joinmarket/comments/c4iuvi/join_market_ui_for_ios_and_android/
13:08 < nothingmuch> ah... although that sounds a bit like an x y problem.
13:09 < nothingmuch> AlexCato: i thought JM can be used w/ a pruned node? and i imagine -blocksonly wouldn't be too hard with custom tx broadcast?
13:16 < AgoraRelay> [agora-irc/CgRelayBot] [cgan/AlexCato] just checked old posts about that, conclusion: yep, pruned works. Blocks only will not work for makers, they need to know when a TX has been broadcast so they can update their offers (even before they confirm)
13:20 < nothingmuch> oh right!
14:06 -!- MaxSan [~four@2a00:1678:2470:50:4fbf:b745:133a:e951] has joined #joinmarket     
14:35 -!- MaxSan [~four@2a00:1678:2470:50:4fbf:b745:133a:e951] has quit [Quit: Leaving.]     
14:43 -!- kristapsk [~KK@gateway/tor-sasl/kristapsk] has quit [Remote host closed the connection]     
14:43 -!- kristapsk [~KK@gateway/tor-sasl/kristapsk] has joined #joinmarket     
15:39 -!- harding [~harding@c-73-196-98-36.hsd1.nj.comcast.net] has quit [Ping timeout: 258 seconds]     
15:41 -!- harding [~harding@c-73-196-98-36.hsd1.nj.comcast.net] has joined #joinmarket     
15:44 -!- undeath [~undeath@hashcat/team/undeath] has quit [Quit: WeeChat 2.5]     
17:20 -!- AgoraRelay [~jmrelayfn@p5DE4ABEF.dip0.t-ipconnect.de] has quit [Ping timeout: 272 seconds]     
17:29 -!- AgoraRelay [~jmrelayfn@p54866096.dip0.t-ipconnect.de] has joined #joinmarket     
18:10 -!- reallll [~belcher@unaffiliated/belcher] has joined #joinmarket     
18:14 -!- belcher [~belcher@unaffiliated/belcher] has quit [Ping timeout: 244 seconds]     
19:45 -!- kristapsk [~KK@gateway/tor-sasl/kristapsk] has quit [Remote host closed the connection]     
19:45 -!- kristapsk [~KK@gateway/tor-sasl/kristapsk] has joined #joinmarket     
19:54 -!- MaxSan [~four@213.152.161.25] has joined #joinmarket     
21:05 -!- arubi_ [~ese168@gateway/tor-sasl/ese168] has quit [Remote host closed the connection]     
21:06 -!- arubi [~ese168@gateway/tor-sasl/ese168] has joined #joinmarket     
22:38 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has quit [Ping timeout: 248 seconds]     
22:39 -!- luke-jr [~luke-jr@unaffiliated/luke-jr] has joined #joinmarket