--- Log opened Wed Aug 11 00:00:35 2021
00:55 -!- k3tan [~pi@user/k3tan] has quit [Ping timeout: 245 seconds]
01:05 -!- k3tan [~pi@user/k3tan] has joined #joinmarket
01:58 -!- undeath [~undeath@user/undeath] has joined #joinmarket
02:37 < waxwing> davterra, i guess you're responding to the idea of a ring sig. in the gist i discussed amount ranges. it would indeed be entirely uninteresting with a published amount.
02:38 < waxwing> in retrospect, fixed amounts probably make more sense but whatever, that's not the hard part.
04:35 < JoinMarketRelay> [hackint/s45rutz7e] If the yg would just restarts rigth after the CJ successfully but not confirmed yet, he will NOT offer the same maxsize again, because until the coins unconfirmed, it is not advertised.
04:36 < JoinMarketRelay> [hackint/s45rutz7e] Re: blchr
04:42 < waxwing> well yeah for sure. there's a lot of details in there. for example, when *privacyenhanced was made (like uh 5 years ago :) ), one of the main ideas was to avoid re-announcement as much as possible (by not choosing the max size depth when avoidable), but that has been superseded by the concentration algo
04:43 < waxwing> and then there's the idea that when it *does* reannounce, again, after the confirmation, then it's arguable whether it matters
04:43 < waxwing> and then there's the randomization etc etc
04:49 -!- Zenton [~user@user/zenton] has quit [Remote host closed the connection]
05:07 < JoinMarketRelay> [hackint/s45rutz7e] After tx confirm the yg reannounce offer yes. so you could probably find a match to previous nickname again, yes. if randomisation does not cover behind other makers enough.
05:09 < JoinMarketRelay> [hackint/s45rutz7e] Actually the makers do send the !hp2 commitments through other peers that publish them in the irc message channel for them to not make obvisoubly who take part in which coinjointx, but afterwards all makers that took part reannounce there offers at same blocktime anyway. so it is easy to tell which maker took part and was chooosen by taker in
05:09 < JoinMarketRelay> the last cjtx.
05:10 < waxwing> but notice there are two quite separate issues here: (1) does the bot reannounce after a tx, thus indirectly revealing that it took part in that join , and (2) persistence of nicks or FBs over periods of multiple joins
05:10 < waxwing> well, re the !hp2 re-routing, that's directly related to the above point. if they don't re-announce, then it's not obvious.
05:10 < JoinMarketRelay> [hackint/s45rutz7e] Yes, well I did not mention this with FB in mind.
05:10 < waxwing> they wouldn't reannounce if their maxsize didn't change, which it doesn't, if the join was source from the non-max-mixdepth
05:11 < JoinMarketRelay> [hackint/s45rutz7e] Ok got you
05:11 < waxwing> what they could, and should do, most likely, is randomized large delay before reannounce
05:12 < waxwing> technically it's economically non-optimal
05:12 < JoinMarketRelay> [hackint/s45rutz7e] So this will only be the clear case for new maker with only mixdepth 0 funded or as you said any maxsize max-mixdepth usage in future. just thought about the single mixdepth case
05:13 < JoinMarketRelay> [hackint/s45rutz7e] You may miss some fill orders of taker, but user may could want the yg do that if he uses the privacyenhanced yg.
05:14 < waxwing> yes for the very first join of a maker in that case, it's guaranteed they have to reannounce, true
05:14 < JoinMarketRelay> [hackint/s45rutz7e] The yg-privacyenhanced.py already does obscure your maxsize by randomising it below your max threshhold. so in theory you could already miss some large cj with cjamount near your real maxsize, all for privacyenhanced.
05:15 < waxwing> also true
05:15 < waxwing> it's usually pretty small though
05:15 < JoinMarketRelay> [hackint/s45rutz7e] Teh defaults yes
05:16 < JoinMarketRelay> [hackint/s45rutz7e] If the only disadvantage is economic for some random short time period, the privacyenhanced should prefer enhanced privacy instead ecomy?
05:18 < JoinMarketRelay> [hackint/s45rutz7e] Combined with nickname change and combined with tor circuit stream isolation, maker could theoretically become new idendity after each chose by taker
05:18 < waxwing> but if they do that then they have to just reannounce on restart anyway?
05:19 < waxwing> so yeah if you imagine massive randomization then i guess.
05:19 < waxwing> but again it comes back to: are you analyzing global or targeted, active attack?
05:19 < waxwing> for the latter there is utxo probing too to consider (and the podle defence)
05:19 < JoinMarketRelay> [hackint/s45rutz7e] At some point they need reannounce a randomized maxsize again, to take part again
05:22 < JoinMarketRelay> [hackint/s45rutz7e] It would make it harder, not impossible. this wouldnt defend against utxo snooping takers.
05:24 < JoinMarketRelay> [hackint/s45rutz7e] Personally I do not care at all about any taker may know my utxo. But I do not like permanent identities. If that is a problem and any you need privacy 1st, switch roles to taker with tumbler please.
05:25 < JoinMarketRelay> [hackint/s45rutz7e] For "yield" there is a little tradeoff between economically & privacy
05:27 < JoinMarketRelay> [hackint/s45rutz7e] The more randomisation for amounts and nicks could help avoid linkage probability % of targeted active attacks I could imagine.
06:14 -!- jonatack [~jonatack@user/jonatack] has quit [Quit: Client closed]
06:31 -!- jonatack [~jonatack@user/jonatack] has joined #joinmarket
08:26 -!- technonerd [~techno@user/technonerd] has quit [Remote host closed the connection]
08:26 -!- technonerd [~techno@user/technonerd] has joined #joinmarket
09:39 -!- ementar4729 [~ementar47@189.122.121.102] has quit [Ping timeout: 268 seconds]
09:40 -!- NorrinRadd [~username@102.67.16.112] has joined #joinmarket
09:44 < JoinMarketRelay> [hackint/the-eht] so I went through the joinmarket.conf again yesterday, as I was questioning my previous choices during the 0.9.0 upgrade... given the any knobs, there's probably a big risk of ending up with a unique fingerprint if one tweaks too much? Has anyone tried to research that?
09:44 < JoinMarketRelay> [hackint/the-eht] the *many knobs
09:45 < JoinMarketRelay> [hackint/the-eht] if true, maybe a disclaimer somewhere would help, because I totally didn't think about that when I first set up the config
09:47 < JoinMarketRelay> [hackint/the-eht] I haven't done anything in that direction, but I wouldn't be shocked if you could at least cluster the makers (one of these 4 makers is X) if you tried to
09:47 < JoinMarketRelay> [hackint/the-eht] am I paranoid?
09:48 < JoinMarketRelay> [hackint/the-eht] maybe there's much less leaked/published than I assume.
09:48 -!- NorrinRadd [~username@102.67.16.112] has quit [Ping timeout: 245 seconds]
09:50 -!- NorrinRadd [~username@173.239.197.64] has joined #joinmarket
09:50 < JoinMarketRelay> [hackint/the-eht] also, how much of a problem is it if you can assing a unique ID to a maker? isn't that what the fidelity bonds do anyway?
09:54 < JoinMarketRelay> [hackint/the-eht] (yup they do, the unique ID would be the UTXO/redeem script/etc.)
10:28 -!- ghost43 [~ghost43@gateway/tor-sasl/ghost43] has joined #joinmarket
10:37 -!- greypw [~greypw@user/greypw] has quit [Quit: irc_error]
10:37 -!- greypw [~greypw@user/greypw] has joined #joinmarket
11:18 -!- undeath [~undeath@user/undeath] has quit [Quit: WeeChat 3.1]
12:47 < JoinMarketRelay> [hackint/Wukong] Hi, I just noticed the topic of this channel is still showing "Latest release 0.8.3", we should probably update it to "0.9.0"?
13:09 < JoinMarketRelay> [hackint/the-eht] true.dat .... needs to be updated
13:10 < JoinMarketRelay> [hackint/the-eht] I have another basic question about joinmarket ... guess I could figure it out by looking at the code but maybe someone knows ...
13:11 < JoinMarketRelay> [hackint/the-eht] 1) when the yg-privacyenhanced.py script starts, it says that it's listening on some port ... what's that? thought the communication protocol was IRC?
13:12 < JoinMarketRelay> [hackint/the-eht] 2) is the idea that each bots redundantly sends out all messages to all connected networks/channels? and then the other participants collect messages on a best effort basis from whatever networks they are connected to?
13:13 < JoinMarketRelay> [hackint/the-eht] "[INFO]  Joinmarket daemon listening on port 27183"
13:14 < JoinMarketRelay> [hackint/the-eht] I guess someone like @belcher or @waxwing might know :) :)
13:15 < JoinMarketRelay> [hackint/the-eht] or @kristapsk ?
14:09 < JoinMarketRelay> [hackint/s45rutz7e] You could run the yg and jm daemon on seperate proccess
14:10 < JoinMarketRelay> [hackint/s45rutz7e] #set to 0 if the daemon is run separately (using script joinmarketd.py) try: no_daemon = 0
14:11 < JoinMarketRelay> [hackint/s45rutz7e] Communication protocol is IRC
14:12 < JoinMarketRelay> [hackint/s45rutz7e] Read some of the files in /docs directory instead pining all devs? :/ you may not even want ot run it without inproccess deamon in almost all case
14:21 -!- Lightsword [~Lightswor@user/lightsword] has quit [Quit: ZNC]
14:31 < JoinMarketRelay> [hackint/the-eht] maybe the devs love my attention
14:32 < JoinMarketRelay> [hackint/the-eht] thank you the daemon part makes sense
14:32 < JoinMarketRelay> [hackint/the-eht] what about the comms part ?
14:58 -!- Lightsword [~Lightswor@user/lightsword] has joined #joinmarket
15:01 < JoinMarketRelay> [hackint/kristapsk] the-eht, it's the port used to communicate between client and daemon parts internally inside JM, you should not worry about that; IRC is used to communicate with other JM peers
16:57 -!- JMBridged [~CoinJoins@gateway/tor-sasl/jmbridge] has joined #joinmarket
16:58 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has quit [Ping timeout: 244 seconds]
16:58 -!- JMBridged is now known as JMBridge
17:46 -!- jonatack [~jonatack@user/jonatack] has quit [Quit: Client closed]
18:53 < JoinMarketRelay> [hackint/the-eht] thank you. any thoughts on the second part of the IRC question? "2) is the idea that each bots redundantly sends out all messages to all connected networks/channels? and then the other participants collect messages on a best effort basis from whatever networks they are connected to?"
20:36 -!- ghost43 [~ghost43@gateway/tor-sasl/ghost43] has quit [Remote host closed the connection]
20:36 -!- ghost43 [~ghost43@gateway/tor-sasl/ghost43] has joined #joinmarket
21:40 -!- belcher_ [~belcher@user/belcher] has joined #joinmarket
21:44 -!- belcher [~belcher@user/belcher] has quit [Ping timeout: 248 seconds]
23:01 -!- belcher_ is now known as belcher
23:46 -!- JMBridged [~CoinJoins@gateway/tor-sasl/jmbridge] has joined #joinmarket
23:48 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has quit [Ping timeout: 244 seconds]
23:48 -!- JMBridged is now known as JMBridge
--- Log closed Thu Aug 12 00:00:36 2021