--- Log opened Sat Aug 07 00:00:31 2021
00:32 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has quit [Remote host closed the connection]
00:33 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has joined #joinmarket
00:36 -!- NorrinRadd [~username@154.6.20.108] has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
01:02 -!- jonatack [~jonatack@user/jonatack] has joined #joinmarket
01:05 -!- Michail1 [~Michail@michail.com] has quit [Ping timeout: 245 seconds]
01:08 -!- Michail1 [~Michail@michail.com] has joined #joinmarket
01:24 -!- jonatack [~jonatack@user/jonatack] has quit [Ping timeout: 246 seconds]
01:29 -!- JMBridged [~CoinJoins@gateway/tor-sasl/jmbridge] has joined #joinmarket
01:29 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has quit [Quit: Bridge selfdestructed.]
01:29 -!- JMBridged is now known as JMBridge
01:31 -!- jonatack [~jonatack@user/jonatack] has joined #joinmarket
01:56 < waxwing> i believe #969 and #965 are good to go. as mentioned yesterday, these are the crucial ones as they fix up maker and taker behaviour wrt timelocked outputs.
01:57 < waxwing> so double check them asap if you want to look at them. once they're done, we're going to make the bugfix release soon. if there are specific PRs that you think really should be included, please say.
02:15 -!- undeath [~undeath@user/undeath] has joined #joinmarket
02:55 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has quit [Remote host closed the connection]
02:55 -!- JMBridged [~CoinJoins@gateway/tor-sasl/jmbridge] has joined #joinmarket
02:56 -!- JMBridged is now known as JMBridge
03:04 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has quit [Remote host closed the connection]
03:05 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has joined #joinmarket
03:19 < waxwing> undeath, i think i'm going to merge them now so there's a good few days to see if something's wrong.
03:47 -!- undeath [~undeath@user/undeath] has quit [Quit: WeeChat 3.1]
04:24 <+JoinMarketRelay> [hackint/JJXXZZ] 0.9.0 is safe to be used? any guidance on fidelity bond size & duration? or still to be determined by the market?
04:26 < waxwing> JJXXZZ yes, re: bugs mentioned above, they are around the spending of unlocked timelock coins. they are spendable, but only in certain ways, not in others (eg. you can't spend them as part of yieldgenerator).
04:26 < waxwing> re: guidance, i personally would recommend keeping times shorter for now (like, a month or two), and easing into it, but i think belcher_ views it a little differently.
04:27 <+JoinMarketRelay> [hackint/JJXXZZ] so should it be a percentage of total "AUM"? or does the absolute value matter?
04:28 <+JoinMarketRelay> [hackint/JJXXZZ] wait, I think there was a formula in the new README
04:29 < waxwing> yeah see fidelity-bonds.md in docs/ , good idea to read it in detail
04:29 <+JoinMarketRelay> [hackint/JJXXZZ] bond_value = (locked_coins * exp(interest_rate * locktime))^2
04:29 < waxwing> check it in master, because there was a typo in the formula originally iirc but it's fixed now
04:30 <+JoinMarketRelay> [hackint/JJXXZZ] bond_value = (locked_coins * (exp(interest_rate * locktime) - 1))^2
04:31 <+JoinMarketRelay> [hackint/s45rutz7e] Need study some math
04:32 <+JoinMarketRelay> [hackint/JJXXZZ] I think if you replace all the complicated operators with a "+", that's still directionally right enough
04:37 <+JoinMarketRelay> [hackint/JJXXZZ] curious, do makers somehow announce/leak their JM version? do we know from past upgrades how long roughly it takes these upgrades to propagate?
04:41 <+JoinMarketRelay> [hackint/JJXXZZ] also, is there an estimate how real the sybil attacks are, e.g.., % (range) of malicious makers?
04:46 <+JoinMarketRelay> [hackint/s45rutz7e] No version. the nick prefix is J5* ... for backward compatible versions I think. would be nice to know versions. if there is a bad version such as the 0xdeadbeef leaks you may like to ignore such outdated counterpartys. on the other side, the running maker do not notify you of new jm version. you would need signup github release notify to get
04:46 <+JoinMarketRelay> any notice.
04:47 <+JoinMarketRelay> [hackint/s45rutz7e] Why you suspect of any malicious makers?
04:47 <+JoinMarketRelay> [hackint/JJXXZZ] one more ... so, I get that the quadratic part is supposed to incentivize running just 1 bot instead of multiple.... but doesn't it also mean that the largest fish (who can afford the biggest bonds) will capture an even bigger share of the joins going forward?
04:47 <+JoinMarketRelay> [hackint/s45rutz7e] Yes
04:48 <+JoinMarketRelay> [hackint/s45rutz7e] The biggest will get most cj with his biggest fb he will suck up all takers to his own advantage
04:48 <+JoinMarketRelay> [hackint/s45rutz7e] In the >v0.9.0 default takers maker selection algorithm
04:49 <+JoinMarketRelay> [hackint/s45rutz7e] Weigthed by fb value
04:51 <+JoinMarketRelay> [hackint/JJXXZZ] so that's going to become an interesting optimization excercise ... larger bond for smaller but more frequent joins? smaller bond for bigger but less frequent joins?
04:54 <+JoinMarketRelay> [hackint/JJXXZZ] (and then you also have fees to play with... which could suddenly become more relevant...)
04:59 <+JoinMarketRelay> [hackint/JJXXZZ] (as in, will the big bond guys be smart enough to raise their fees, now that they have less competition? and will the smaller bond guys be required to discount their fees?)
05:22 <+JoinMarketRelay> [hackint/s45rutz7e] Fb release raised default fees 4x to compensate a little the locked funds of fb. actually the more you lock, the less available for joins of course. need to find some balance between. but all depends on current market competition?
05:56 -!- NorrinRadd [~username@154.6.20.108] has joined #joinmarket
07:04 < waxwing> you can set bondless_makers_allowance if you want to de-prioritise fidelity bonds. default is 0.125 i.e. 1/8 chance for each selection, of choosing a non-FB
07:04 < waxwing> you can also change your interest rate.
07:53 <+JoinMarketRelay> [hackint/cryptool] Imagine someone sends 100 BTC into JM and peels it for about two weeks and someone took just one minute on block explorers to look at it can connect them? in my humble opinion those have high probability of being related utxo path ? found those two tx from snickers finder tool. ab93c88f29fa9ad66679d2046c6f59f6631c11b681b491d461c62290785bb053 =>
07:53 <+JoinMarketRelay> fd88c3a03710da7f8ffbf84aee41fabd855f787c315d54faa5fa3e6f20a0d8ec
08:10 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has quit [Remote host closed the connection]
08:10 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has joined #joinmarket
09:01 < waxwing> you can definitely connect the change outputs, assuming you mean 100BTC in 1 utxo. connecting what goes into cj outs is different.
09:02 < waxwing> you might find some of this interesting: https://joinmarket.me/blog/blog/the-445-btc-gridchain-case/
09:16 -!- JMBridged [~CoinJoins@gateway/tor-sasl/jmbridge] has joined #joinmarket
09:16 -!- JMBridge [~CoinJoins@gateway/tor-sasl/jmbridge] has quit [Ping timeout: 244 seconds]
09:16 -!- JMBridged is now known as JMBridge
09:25 <+JoinMarketRelay> [hackint/cryptool] Thx for links. yes single 100BTC utxo 78e80cc49e157df595d97fa38f870558152f68f9d843118078243ed5dc974cdf:1 https://kycp.org/#/ab93c88f29fa9ad66679d2046c6f59f6631c11b681b491d461c62290785bb053
09:39 <+JoinMarketRelay> [hackint/cryptool] You are rigth. i was following the change outputs but can't tell the cj outputs of it.
10:07 -!- NorrinRadd [~username@154.6.20.108] has quit [Ping timeout: 240 seconds]
10:18 <+JoinMarketRelay> [hackint/curious] Hi, I have a Fidelity Bonds question: I read the bond_value formula but I don't understand it, could anybody please tell me in *simple words* in what ratio locktime and locked_coins affect the resulting bond_value? E.g. "10 bitcoins for 2 years would give the same bond value as 5 bitcoins for 4 years"? Is it this simple and they both affect
10:18 <+JoinMarketRelay> bond_value equally inversely
10:18 <+JoinMarketRelay> [hackint/curious] proportionally Or one of them weights more, e.g. "half the bitcoins would require thrice the locktime to get the same bond value"?
10:18 -!- NorrinRadd [~username@173.239.197.60] has joined #joinmarket
11:15 <+JoinMarketRelay> [hackint/curious] I can answer my own question - it's almost 1:1 with time being tiny bit more valuable. If I lock bitcoins for twice as long I will need only 49.85% of them to get the same bond_value. If I lock twice as many bitcoins I will need to lock them for 50.15% of the duration to get the same bond_value.
11:18 <+JoinMarketRelay> [hackint/Wukong1971] Hi, I'm interested in implementing https://github.com/JoinMarket-Org/joinmarket-clientserver/issues/792, "YG coinjoin out to xpub":
11:19 <+JoinMarketRelay> [hackint/Wukong1971] I agree with kristapsk that adding the ability of tracking the addresses of an external xpub is a good first step.
11:22 <+JoinMarketRelay> [hackint/Wukong1971] I'm new to the development of JoinMarket, so I'd like to get feedback on my ideas. I'm currently thinking about creating a new wallet type, YGOutputWatchonlyWallet, in WALLET_IMPLEMENTATIONS.
11:25 < xyy> <waxwing> "you can set bondless_makers_allo..." <- I’d say realistically most people are going to keep the defaults 
11:27 < xyy> So the concerns around “whales” being able to scoop up more of the joins AND being able to charge higher proportional fees are probably not irrational?
11:29 <+JoinMarketRelay> [hackint/Wukong1971] And I'm thinking about adding a new field, `b'associated_output_wallet'` in the storage. If that is set, then we monitor the transactions of this additional wallet at the same time. Does this make sense? Thanks.
11:39 <+JoinMarketRelay> [hackint/curious] Oh, I see from the logs that you were just talking about this formula earlier today, what a funny coincidence.
11:46 <+JoinMarketRelay> [hackint/JJXXZZ] it's the new hot sh*t !
11:46 <+JoinMarketRelay> [hackint/curious] I am also concerned that the big whales that can afford big amount AND a long time period for the fidelity bond will completely overpower the JoinMarket :( Little makers will have no chance.
11:47 <+JoinMarketRelay> [hackint/curious] I am afraid that little makers in pure desperation will be locking their bonds for *decades* to at least little compete with the whales.
11:50 <+JoinMarketRelay> [hackint/Wukong1971] > concerned that the big whales that can afford big amount AND a long time period for the fidelity bond will completely overpower the JoinMarket < I think when we have more and more makers, we can increase the default CJ count
11:51 <+JoinMarketRelay> [hackint/Wukong1971] Today, by default, takers only does a CJ with about 8~10 counter parties. This is very small comparing to the 100 CJ parties Wasabi does. And to combat this, it is encouraged to use multiple CJ, i.e., tumbler, to get better privacy.
11:52 -!- belcher_ is now known as belcher
11:52 <+JoinMarketRelay> [hackint/curious] re: interest rate - from my calculations the default 1.5% has basically no effect. If I lock bitcoins for 5x as long time I will need 19.75% of them to have the same bond_value, still it's nearly 1:1.
11:52 < belcher> if someone is asking for guidance and therefore is unsure, then they should definitely do smaller amounts and shorter locktimes
11:54 < belcher> makers with no bonds at all still get a 1-in-8 chance of being chosen for coinjoins by default
11:55 <+JoinMarketRelay> [hackint/curious] Now I am wondering is there an upper limit for the lock period? Maybe the small makers will try to lock their small bonds for hundreds of years to increase their bond_value?
11:55 < belcher> i understand the concern that fidelity bonds make being a yield-generator less egalitarian, but my view is the point of joinmarket is to improve bitcoin privacy not to improve equality
11:55 < belcher> curious the upper limit is when the time-value-of-money is the same as burning the coins
11:57 <+JoinMarketRelay> [hackint/curious] I mean in simple words. I will try to play with the calculate_timelocked_fidelity_bond_value() to find out when the bond_value stops increasing.
11:57 < belcher> so less-rich makers max out at just burning the coins, i.e. if they have 3 btc they could get a fidelity bond worth 9 btc^2
11:57 <+JoinMarketRelay> [hackint/curious] I just want to find out what's the limit, 10 years, 100 years, etc..
11:58 < belcher> after about 46 years the time-value-of-money is the same as burning
11:58 < belcher> assuming interest rate of 1.5% (the default)
11:58 < belcher> so theres no point locking longer than about that duration
11:59 <+JoinMarketRelay> [hackint/curious] Thanks, I will try to verify.
12:01 < belcher> these graphs here hopefully provide some intuitive feel for how the value function behaves https://gist.github.com/chris-belcher/87ebbcbb639686057a389acb9ab3e25b#graphs
12:01 < belcher> you can see on the top graph the line maxes out at 1
12:01 <+JoinMarketRelay> [hackint/curious] But thinking more about it I am afraid this won't help the small makers either because if they can lock their small bonds for 46 years so can the big whales their bigger bonds :(
12:02 < belcher> what do you mean by helping small makers
12:03 <+JoinMarketRelay> [hackint/curious] to compete with the whales
12:03 < belcher> if they have less money available then competition will be harder
12:04 < belcher> the biggest help for small-valued makers is the bondless makers allowance (which is 1-in-8 by default)
12:04 <+JoinMarketRelay> [hackint/curious] Yes, that's why I said that they maybe in pure desperation will lock their bond for the max time (46 years as you said). But that will not help them because the whales can do the same thing :(
12:06 < belcher> from the point of view of the customers (i.e. takers), whales provide more resistance to sybil attacks, its much more expensive to overpower a whale's fidelity bond than to overpower a small maker's
12:07 <+JoinMarketRelay> [hackint/JJXXZZ] there was never a ton of money to be made on JoinMarket, but I somehow assumed that was just a temporary thing while adoption grows and the economics are figured out ... the bonds more or less destroy that perspective for smaller makers
12:07 <+JoinMarketRelay> [hackint/JJXXZZ] but they can still be there for the "free conjoins", like today
12:07 <+JoinMarketRelay> [hackint/curious] 7-in-8 coinjoins will be fully controlled by whales who will make "a lot of" money, i.e. the already rich will get even richer.
12:07 <+JoinMarketRelay> [hackint/JJXXZZ] btw, I would remove the word "burn" from the docs, it's too scary, not everyone will get the concept
12:08 <+JoinMarketRelay> [hackint/JJXXZZ] @curious: I think realistically nobody is getting rich, but the normal people just went from almost nothing to nothing :P
12:09 <+JoinMarketRelay> [hackint/curious] It's not really funny, though
12:09 <+JoinMarketRelay> [hackint/JJXXZZ] yea I will probably move funds to my LN node
12:10 <+JoinMarketRelay> [hackint/JJXXZZ] but let's see, I think it will take a while until enough people use the bonds
12:10 <+JoinMarketRelay> [hackint/Wukong1971] I think the best hope for small makers to make some money is by takers coin join with more makers, and thus, the smaller makers will also be able to participate when the number of makers needed is large enough.
12:10 < belcher> i saw for years how many people actually run multiple yieldgenerators, you could see because they'd all connect/disconnect at the same time, and also some people admitted it on reddit... doing that really damages joinmarket's privacy, but that party is over now because of fidelity bonds
12:11 < belcher> they'll always be a limit to joinmarket or coinjoin growth because the transactions are just bigger and cost more miner fees
12:11 <+JoinMarketRelay> [hackint/cryptool] But your Ln does not yield at all? :o
12:11 <+JoinMarketRelay> [hackint/JJXXZZ] at least it's easier to be a whale in LN :P
12:11 <+JoinMarketRelay> [hackint/Wukong1971] > But your Ln does not yield at all? :o < I lost tens of thousands of sats running my LN node X)
12:12 <+JoinMarketRelay> [hackint/curious] Anybody will 25+ btc is already a dollar millionaire. I mean these people really don't *need* to make more money with coinjoins :( Small guys do but they will now have 8x smaller chance.
12:12 <+JoinMarketRelay> [hackint/curious] with*
12:12 <+JoinMarketRelay> [hackint/curious] You can make money with LN node?
12:13 < belcher> its the same with any investment really, if the price of some ticker goes up 5% then a small guy with $100 invested gets $5 extra, a whale with $100million invested gets $5million extra
12:13 <+JoinMarketRelay> [hackint/JJXXZZ] but listen to Belcher, if there's actual evidence that there was a ton of Sybil activity then maybe it's needed .. I asked the question yesterday and got no response ... the question no wis though, can the parameters still be tweaked
12:13 <+JoinMarketRelay> [hackint/Wukong1971] > You can make money with LN node? < If you run a routing node and lots of people connect to you, yes.
12:13 < belcher> the rich-get-richer is nothing new here
12:13 < belcher> joinmarket is not a project aiming for greater wealth equality
12:13 < belcher> JJXXZZ which parameters?
12:14 <+JoinMarketRelay> [hackint/JJXXZZ] the formula, to make the small guy a bit happier
12:14 <+JoinMarketRelay> [hackint/JJXXZZ] defuse that ^2 a little bit or sth like that
12:14 <+JoinMarketRelay> [hackint/JJXXZZ] not sure how much is needed to stop the Sybil
12:15 < belcher> its open source so anyone could fork it or change it, however remember that you need to get takers to run it
12:15 < belcher> i really doubt that takers will run something that makes their privacy worse only to help "small makers"
12:16 < belcher> remember your customers are the takers, the takers pay your coinjoin fees, everything is done for them
12:17 < belcher> (at least in my worldview of joinmarket, i know other people see maker's privacy as important in-itself rather than important just to keep takers private)
12:18 <+JoinMarketRelay> [hackint/curious] I don't want to start any argument but I am afraid FB will not really prevent Sybil because the adversaries (the governments) are the very ones who have access to the biggest amounts of bitcoins, if not directly then indirectly via their infinite legal power.
12:19 < belcher> forcing sybil attackers to pay loads of money is the only tool we really have... its how bitcoin mining works after all
12:19 <+JoinMarketRelay> [hackint/curious] But locking bitcoins does not really mean paying.
12:19 < belcher> some people have said the whole PoW mining thing wont work because governments have loads of money, and that might turn out to be true, but many people think its worth trying anyway as the alternative is the governments can censor transactions or inflate currencies for free
12:20 < belcher> locking bitcoins means paying because the locker sacrifices time-value-of-money
12:21 <+JoinMarketRelay> [hackint/curious] That is true only if you see any value in bitcoins. Governments see no value in bitcoins. They seized some 70 thousand bitcoins from darkmarket, they have no problem with locking all of them for 46 years.
12:23 <+JoinMarketRelay> [hackint/curious] They don't even need to own any bitcoins, they can legally compel exchanges to lock their (exchanges') cold wallets and give them (the governments) the signatures needed to prove FB.
12:25 < belcher> this is possible yes, but compare to the previous situation where the attacker doesnt have to pay anything
12:25 < belcher> exchange's cold wallets being locked will cause them problems though, it opens up a bank run type situation
12:26 <+JoinMarketRelay> [hackint/curious] The legislative frameworks around the world is aiming at centralizing the crypto custody and making self-custody illegal. I can very well imagine that in the future the custody will be legally only allowed by an "approved" entity (like central banks) which will be controlled by the governments. Then the governments will have all the access they
12:26 <+JoinMarketRelay> need to get the timelocked
12:26 <+JoinMarketRelay> [hackint/curious] signatures for FB
12:27 < belcher> governments could also use their vast wealth and power to mine themselves, and remove bitcoin's useful properties that way
12:28 <+JoinMarketRelay> [hackint/curious] Well if the exchanges are legally forced to timelock their cold wallets then they will have to limit withdrawals to prevent bank runs. I can even imagine that they (both governments and exchanges) will present this step as a good thing because timelocking cold wallets will prevent a theft. They can spin anything any way they see fit.
12:29 <+JoinMarketRelay> [hackint/curious] But I don
12:29 < belcher> looks like the relay bot bugged out, i just saw "<JoinMarketRelay> [hackint/curious] But I don"
12:29 <+JoinMarketRelay> [hackint/curious] t want to start any argument, I am sure a lot of thinking and effort was put into creating FB.
12:29 <+JoinMarketRelay> [hackint/curious] sorry too early pressed enter
12:30 < belcher> ah
12:30 < belcher> yeah i appreciate the thoughts
12:31 < belcher> and you are right maybe our side will just lose the fight, though its clear fidelity bonds makes the situation better, without them an attacker could deanonymize joinmarket just by running many many yieldgenerator bots and it wouldnt cost them much at all
12:31 <+JoinMarketRelay> [hackint/curious] I would summarize my "argument" like this: any condition you can put on (honest) makers to prevent Sybil, the adversaries can fulfil 10000 times better.
12:33 <+JoinMarketRelay> [hackint/curious] The adversaries are more powerful and incomparable superior in any way, shape or form, compared to honest makers.
12:34 < belcher> the system is designed so that honest makers are more effective, thats what the ^2 term does
12:34 <+JoinMarketRelay> [hackint/curious] So in my humble opinion any limitations/conditions you put on makers will in the end harm only the honest makers, not the adversaries.
12:34 < belcher> thats just not true
12:35 <+JoinMarketRelay> [hackint/curious] But there is nothing that honest makers can do that the adversaries cannot do better. Think about it. They have direct or indirect access to any amount of bitcoins and they don't can about locking them for any amount of time since they don't see any economic value in them.
12:35 <+JoinMarketRelay> [hackint/curious] don't care*
12:35 < belcher> an honest maker is one who doesnt want to spy on anyone, they put their available coin into just one fidelity bond.... the dishonest maker wants to spy and has to put his coins into multiple fidelity bonds, the ^2 term means the honest maker has a massive advantage
12:36 <+JoinMarketRelay> [hackint/curious] Yes but governments can get direct or indirect access to more bitcoins than all honest makers combined.
12:36 <+JoinMarketRelay> [hackint/curious] And they don't even need to own the bitcoins, they just need the signature.
12:37 < belcher> they need to lock up the coins
12:37 < belcher> so someone will lose access to them
12:38 <+JoinMarketRelay> [hackint/curious] yes, if it's not their own bitcoins they will legally compel the owner, presumably the cold wallets' owners, presumanly the exchanges and investment funds etc.
12:38 < belcher> even if they do, thats still a better situation than we had pre-fidelity-bonds
12:38 <+JoinMarketRelay> [hackint/JJXXZZ] was afk, but I think it's not without risk .. what if some resourceful power started hitting up the bots with large balances and offered to pay 20x the rewards if data is shared
12:39 <+JoinMarketRelay> [hackint/JJXXZZ] have to make sure we still have enough active makers
12:39 <+JoinMarketRelay> [hackint/curious] belcher: I am not sure how you mean, maybe you are right, I will have to ponder on that.
12:40 < belcher> curious, pre-fidelity-bonds governments could spy on joinmarket users without needing to force exchanges or investment funds to do much
12:40 < belcher> JJXXZZ you're right that's a possible attack, though its also possible pre-fidelity-bonds
12:41 <+JoinMarketRelay> [hackint/curious] belcher: I think I see your point, you are looking at it comparatively between now and pre-FB, I am looking just at the FB situation.
12:45 <+JoinMarketRelay> [hackint/curious] belcher: So if I understand you correcty basically I am saying that *maybe* FBs won't make it any better, and you are saying maybe I am right maybe not but FBs definitely won't make it worse, so overall FB is better than no FB.
12:45 < belcher> no, im saying fidelity bonds do make the situation better, but they dont make the situation perfect
12:46 <+JoinMarketRelay> [hackint/JJXXZZ] (whatever we do, let's please try not to fork ;-)
12:46 <+JoinMarketRelay> [hackint/curious] I think the crucial question is whether FB can make it worse. If it can't then there is no argument against FB.
12:46 < belcher> i could put it another way, what would you say if someone said all of bitcoin is useless because governments can just buy or manufacture loads of mining hardware and fire it up, then they could do a 51% attack which only lets transactions confirm which have passed their stringent KYC agreements
12:46 < belcher> after all, governments do have lots of wealth and power, they could do this, same as they could get loads of fidelity bonds
12:47 <+JoinMarketRelay> [hackint/curious] Well for that they would have to spend "real" money. To legally compel bitcoin owner to timelock and surrender the signature will cost nobody anything.
12:49 < belcher> it would cost something, it would cost them the time-value-of-money
12:49 < belcher> consider this: whats more valuable to you? 1 bitcoin now or 1 bitcoin in 1 year
12:49 <+JoinMarketRelay> [hackint/curious] Come on, you know we are talking about cold wallets that are not touched anyway, so locking them will not cost anything, in fact it will increase their protection because it will prevent crypto theft.
12:50 < belcher> nobody knows what the future holds, maybe they will be touched
12:50 < belcher> bank runs have happened in history all the time
12:50 < belcher> also, today we see interest rates as a thing
12:50 < belcher> interest rates being a thing means money has a time-value
12:51 <+JoinMarketRelay> [hackint/curious] Again (not trying to argue, but) come on, they can simply timelock only some portion of cold wallets, not all
12:51 < belcher> however, if you can think of a way where fidelity bonds makes joinmarket *worse* then definitely tell us because that would be a bug
12:52 <+JoinMarketRelay> [hackint/curious] yes I think that would be kind of a "dealbreaker", or at least a serious issue
12:52 < belcher> i wonder what the exchange's customers will think of that, "hey this exchange is holding my funds but if everyone tried to withdraw at once they would become insolvent, and because bitcoin is capped at 21m its not possible to print more to fund a bailout"
12:53 <+JoinMarketRelay> [hackint/curious] well if that will be a legal requirement then that's that, everyone will have to live with that
12:53 < belcher> not everyone, people who hold coins in their own control wont have to live with it
12:53 <+JoinMarketRelay> [hackint/curious] they might be breaking the law just by holding coins in self-custody
12:54 < belcher> i remember when nigeria banned bitcoin exchanges a few years ago, of course nobody listened and carried on just trading OTC
12:54 <+JoinMarketRelay> [hackint/curious] the governments can create any law with terrorism/money laundering excuse
12:54 < belcher> just earlier i read a thing about people in cuba using bitcoin, its likely illegal there too
12:55 < belcher> (fwiw nigeria eventually unbanned bitcoin, presumably they thought it was more valuable for them to be able to see all flows in kyc-exchanges rather than see nothing in otc exchange)
12:56 <+JoinMarketRelay> [hackint/curious] but whether individuals won't comply with the law will make no difference on exchanges. the exchanges cannot not comply if they will be obliged to timelock portions of their cold wallets and surrender the signatures to governments so they can take over JoinMarket
12:57 < belcher> you might be right, that is a possible attack, perhaps bitcoin's future depends on as many people as possible holding coins in their own custody
12:58 < belcher> thats not too different from what happened with gold, banks and governments took people's gold and gave them paper, and eventually cut the link between the paper and gold, and now they have all the gold
12:58 <+JoinMarketRelay> [hackint/curious] but that will only make the self-custody legally illegal even more likely
12:58 < belcher> if you have any answers im all ears
12:58 < belcher> i dont see how fidelity bonds makes that situation any worse though
12:59 <+JoinMarketRelay> [hackint/curious] yes I agree that's the crucial question
13:01 < belcher> id be interested to hear what other devs and users think of this topic too, obviously its an important question, and kind of deep and philosophical
13:01 <+JoinMarketRelay> [hackint/curious] Maybe look at it from the point of view of honest makers - the FB requirement will make it "worse" (more difficult) for them. Since the governments are superior then for them it will be not-that-much-more-difficult.
13:01 < belcher> that conflates "honest" with "small"
13:01 < belcher> whales can be honest makers too, honest just means they have no interest in spying
13:02 <+JoinMarketRelay> [hackint/curious] But governments are superior even to whales, no?
13:02 <+JoinMarketRelay> [hackint/curious] Governments can be the biggest whales, if not directly then via their ability to legally compel the biggest whales
13:03 < belcher> maybe, maybe not, have you come across this? https://en.wikipedia.org/wiki/Black_Wednesday
13:03 < belcher> a big whale won against the UK government
13:03 < belcher> governments sometimes lose as well
13:03 < belcher> anyway sorry i g2g for now, will bbl
13:04 <+JoinMarketRelay> [hackint/curious] I am afraid today the situation is different than 30 years ago. Today they can hide anything behind the terrorism/money laundering fight.
13:13 <+JoinMarketRelay> [hackint/Wukong1971] > id be interested to hear what other devs and users think of this topic too, obviously its an important question < My concern is if all CJs of all takers would always pick the same few takers, we will have a very small anonymity set.
13:14 <+JoinMarketRelay> [hackint/Wukong1971] We should keep an eye on the makers picking algorithm's weighting on their fidelity bond value.
13:15 <+JoinMarketRelay> [hackint/curious] That depends on the proportions between their bond_value. If there are many makers with similar bond_value then they will be picked randomly.
13:16 <+JoinMarketRelay> [hackint/curious] I think the competition will force makers to be close to each other in their bond_value.
13:16 <+JoinMarketRelay> [hackint/curious] Those who cannot compete with the FB size will try to compensate with the locktime period.
13:19 <+JoinMarketRelay> [hackint/curious] But definitely there will be less of them, but that doesn't necessarily mean the anonymity set will be lower because the low-quality ones (duplicated YGs from the same maker, trolls) will be gone.
13:29 < waxwing> Wukong1971 : re: "We should keep an eye on the makers picking algorithm's weighting on their fidelity bond value." <- yes, while the number is small, people could reset bondless_makers_allowance to create more variety.
13:32 <+JoinMarketRelay> [hackint/curious] Then maybe a fair value would be 50%? Picking makers equally from both worlds.
13:59 <+JoinMarketRelay> [hackint/curious] I can confirm the biggest possible bond_value is at 46.21 years, after that it goes down.
14:04 <+JoinMarketRelay> [hackint/curious] The biggest = the most efficient
14:16 <+JoinMarketRelay> [hackint/curious] Currently the most valuable FB is 12.61058953 btc locked for 3 years. To match this bond_value with the smallest amount of bitcoins would require a FB of 0.58070484 btc locked for 46.21 years.
14:21 <+JoinMarketRelay> [hackint/JJXXZZ] I agree with @curious that the "burn" concept is somewhat alien in Bitcoin given most people just HODL anyway
14:22 <+JoinMarketRelay> [hackint/JJXXZZ] a different question though ... am I right to assume that coins locked in fidelity bonds are just as likely to be "flagged" by the bros as coinjoined funds themselves?
14:23 <+JoinMarketRelay> [hackint/JJXXZZ] so assuming the funds were noKYC before they got locked, the locking would somewhat reduce their value, no?
14:24 <+JoinMarketRelay> [hackint/curious] I think they will be flagged already on the account that you mixed them before using for a FB. Not sure how taproot will affect this.
14:24 <+JoinMarketRelay> [hackint/JJXXZZ] (actually also if they were KYC ... but you wouldn't do that, likely)
14:25 <+JoinMarketRelay> [hackint/curious] Not sure if using a FB that's tied to your identity would make sense.
14:28 <+JoinMarketRelay> [hackint/curious] "so assuming the funds were noKYC before they got locked, the locking would somewhat reduce their value, no?" <- what do you mean by reduce their value?
14:28 <+JoinMarketRelay> [hackint/JJXXZZ] of course not ... so you either use a coin that you just cj'd .. that would have been flagged anyway ... or you use a noKYC coin with some history
14:28 <+JoinMarketRelay> [hackint/JJXXZZ] i have those ... nobody else?
14:29 <+JoinMarketRelay> [hackint/JJXXZZ] @curious well, more difficult to sell
14:29 <+JoinMarketRelay> [hackint/curious] noKYC coins you mean coins that don't come from any exchange?
14:30 <+JoinMarketRelay> [hackint/JJXXZZ] coins with enough non-identifying but non-suspicious hops after the exchange
14:30 -!- DeanGuss [~dean@user/deanguss] has quit [Quit: buhbye]
14:31 <+JoinMarketRelay> [hackint/curious] i think you will want to sell noKYC coins p2p
14:31 <+JoinMarketRelay> [hackint/JJXXZZ] it's probably more of a theoretical thought ... aka what else do I sacrifice by locking up coins in FBs
14:31 <+JoinMarketRelay> [hackint/JJXXZZ] I sacrifice some optionality, potentially
14:31 <+JoinMarketRelay> [hackint/curious] if they trace back to an exchange then they are not really noKYC
14:32 <+JoinMarketRelay> [hackint/JJXXZZ] you have plausible deniability depending on the specific history
14:35 <+JoinMarketRelay> [hackint/curious] If the trail to an exchange is unambiguous then they are not really noKYC. you can claim at some point (hop) in the past it wasn't yours but it still is tied to someone's identity. As opposed to mixed coins where a transaction has multiple identical outputs so there you cannot really trace them back to inputs.
14:37 <+JoinMarketRelay> [hackint/curious] But I don't think it would be more difficult to sell. More people will use bitcoins in the future and more of them will not want to KYC so more of them will be happy to buy your nonKYC coins.
14:47 <+JoinMarketRelay> [hackint/curious] I think some portion of your btc stash you will never spend under normal circumstances, so it's beginning to seem reasonable to me to put those as FB for 46.21 years and use the rest of bitcoins to make new bitcoins while as a side-effect having all those bitcoins perpetually mixed for free.
14:53 <+JoinMarketRelay> [hackint/curious] I see a point in locking FB for 1-2 years but if more than that then I would go directly for 46 years and maximize the bond_value. I think the person who locked his FB for 3 years made a mistake.
14:55 <+JoinMarketRelay> [hackint/curious] If he was willing to sacrifice his 12.6 btc for 3 years he may as well went for 46 years and his bond_value would be 159 instead of 0.3372. Now he basically wasted the opportunity.
14:58 -!- DeanGuss [~dean@nonplayercharacter.me] has joined #joinmarket
14:58 -!- DeanGuss [~dean@nonplayercharacter.me] has quit [Changing host]
14:58 -!- DeanGuss [~dean@user/deanguss] has joined #joinmarket
15:38 -!- NorrinRadd [~username@173.239.197.60] has quit [Ping timeout: 256 seconds]
15:42 -!- belcher_ [~belcher@user/belcher] has joined #joinmarket
15:42 -!- belcher [~belcher@user/belcher] has quit [Ping timeout: 258 seconds]
16:18 <+JoinMarketRelay> [hackint/Wukong1971] > If he was willing to sacrifice his 12.6 btc for 3 years he may as well went for 46 years < No, 3 years is very different from 46 years.
17:00 -!- NorrinRadd [~username@173.239.197.66] has joined #joinmarket
17:01 <+JoinMarketRelay> [hackint/curious] Depends how you look at it. In 3 years bitcoin may not exist anymore. In that case had he intead used a 46 years long timelock his bond_value would be significantly higher which would allow him to earn more during those 3 years before bitcoin cease to exist.
17:03 <+JoinMarketRelay> [hackint/curious] Or instead of locking 12.6 btc for 3 years he could have locked only 0.58 btc for 46 years and use those extra 12 btc for more earning.
17:26 <+JoinMarketRelay> [hackint/curious] I just think timelocking 12.6 btc for 3 years only a few days after the FB feature was released was a very bad idea. I think he will regret it, wish he used different amount and/or duration, and since FBs don't stack he will either have to be upset for 3 years or make a new FB and this one will be totally dead unusable money neither earning nor
17:26 <+JoinMarketRelay> mixing.
23:56 -!- belcher_ is now known as belcher
--- Log closed Sun Aug 08 00:00:32 2021