--- Day changed Wed Dec 12 2018
00:33 -!- belcher_ [~user@unaffiliated/belcher] has quit [Ping timeout: 246 seconds]     
00:49 -!- belcher_ [~user@unaffiliated/belcher] has joined #joinmarket     
02:45 -!- GitHub167 [GitHub167@gateway/service/github.com/x-wafvagxracwgdpfn] has joined #joinmarket     
02:45 < GitHub167> [joinmarket-clientserver] AdamISZ closed pull request #248: Convert JoinmarketQt to PySide2, qt5reactor (master...PyQt5) https://git.io/fp9W4
02:45 -!- GitHub167 [GitHub167@gateway/service/github.com/x-wafvagxracwgdpfn] has left #joinmarket []     
02:45 -!- GitHub37 [GitHub37@gateway/service/github.com/x-khuwhovgpfnzvkks] has joined #joinmarket     
02:45 < GitHub37> [joinmarket-clientserver] AdamISZ pushed 1 new commit to master: https://git.io/fp5wY
02:45 < GitHub37> joinmarket-clientserver/master 5907cc2 AdamISZ: Merge #248: Convert JoinmarketQt to PySide2, qt5reactor...
02:45 -!- GitHub37 [GitHub37@gateway/service/github.com/x-khuwhovgpfnzvkks] has left #joinmarket []     
02:50 < waxwing> is there any good reason not to make --develop the default installation mode?
02:53 -!- GitHub141 [GitHub141@gateway/service/github.com/x-ptrmniocgflvjytb] has joined #joinmarket     
02:53 < GitHub141> [joinmarket-clientserver] AdamISZ pushed 1 new commit to master: https://git.io/fp5rL
02:53 < GitHub141> joinmarket-clientserver/master 9fe7fd8 AdamISZ: update install instructions in README for new JoinmarketQT install
02:53 -!- GitHub141 [GitHub141@gateway/service/github.com/x-ptrmniocgflvjytb] has left #joinmarket []     
04:24 < waxwing> looks like it's likely worth a watch (haven't done yet) https://www.youtube.com/watch?v=9s3EbSKDA3o
04:25 < waxwing> met this guy last year, i believe he does a lot of work/research on zcash.
04:30 < belcher> watching now
04:31 < belcher> says that bitcoin is completely nonprivate, but if thats true then what about all the stolen bitcoins where the thief was never found
04:31 < belcher> its sad that the nuance is not mentioned in some talks like this
04:35 < waxwing> ofc, you should have fully expected the negative narrative :) but, i also expect him to say some interesting things about deanon. techniques. if not, i apologise :)
04:49 < belcher> ok i watched it
04:50 < belcher> it talks about the problems of "decoy systems" which work by adding decoy spent coins, for example coinjoin and monero
04:50 < belcher> theres a bunch of attacks that work when repeated payments happen
04:51 < belcher> like if you spend a coin and include 7 decoys then the chance of a spy choosing the right coin is 1/7, but if theres repeated payments then you can overlap all the probabilities and the chance becomes something like 1/7 + 1/7 + 1/7 + ...
04:52 < belcher> zcash doesnt use decoys instead it entirely hides the tx graph so the above attack cant be used
04:53 < belcher> then he mentions usability and UX and how that is a problem
04:54 < belcher> but doesnt go into detail, but in my point of view thats the really important factor; you can build any privacy system but if it uses too much resources so people dont use it then it wont provide much privacy
04:54 < belcher> i tweeted this thing a couple of days ago https://twitter.com/chris_belcher_/status/1065985329826541568 "If you had to choose, what's more important to bitcoin?" "Security from full nodes OR privacy and fungibility"
04:55 < belcher> people overwhelmingly chose security over privacy.... indirectly monero/zcash are choosing privacy over security, as their full nodes are more resource-intensive to run
04:57 < belcher> so its like theres a tradeoff between privacy and resource usage (which is security)...
04:57 < belcher> the best privacy solution would avoid this tradeoff by completely removing the tx graph from public view rather than adding decoys, so better privacy actually becomes more efficient
04:58 < belcher> thats what LN does
05:01 < belcher> theres a similar talk at scalingbitcoin which also talks about the privacy problems of using decoys to hide, its called "How Much Privacy is Enough? Threats, Scaling, and Trade-offs in Blockchain Privacy Protocols"
05:01 < belcher> also from a zcash guy
05:02 < belcher> coinswap would also be method that answers these criticisms, because it doesnt rely on decoys either (except if you combine it with p2ep, but thats different)
05:04 < waxwing> belcher, thanks for your thoughts, i'll probably think about it a bit on the train :)
05:04 < waxwing> talking of which are you coming tonight?
05:04 < belcher> yep     
05:05 < waxwing> k, see you then
05:05 < belcher> last time a zcash guy also came, and at one point him and antoine has a bit of a discussion
05:05 < belcher> with some of the same points "plausible deniability is not a defence"
05:06 < belcher> which IS a good point, since much of the problems like targetted advertising dont need 100% confirmation to work
05:39 -!- viasil [~viasil@165.231.102.6] has quit [Ping timeout: 250 seconds]     
05:42 -!- viasil [~viasil@165.231.102.6] has joined #joinmarket     
06:59 -!- nothingmuch [~user@213.152.161.85] has quit [Ping timeout: 240 seconds]     
08:29 < waxwing> yeah good stuff. i mostly agree with your pov on it belcher . his dismissal of plausible deniability doesn't paint the complete picture. the way to cut the gordian knot of scalability vs privacy is to abstract individual actors' actions from global verification, but not to remove the isolated state changes represented by individual payments.
08:30 < waxwing> because without the latter you can't "prune" (talking really vaguely, just mean the global state doesn't have to permanently balloon)
08:30 < waxwing> the security arguments are kinda related but ... dunno, it's maybe more complex.
08:31 < waxwing> but still after having watched it i do think it was a very good summary of the main ideas that people should understand.
08:32 < waxwing> the most important ofc is cluster intersection, to give it its probably most generic name
08:39 < waxwing> going out on a limb, but my bet, 10 years from now: the CT style privacy fix; i.e. there's a coin but you don't know its value, is the one that'll persist. (because it still preserves the crucial scalability feature of isolated state: spent or unspent).
08:40 < waxwing> but - it really is going out on a limb; it seems more common sense to think that ZKP style complete obscuration would win.
08:52 < belcher> im my still-unfinished wiki page iv called that idea "data fusion", which is used in data science a lot
08:52 < belcher> where if the adversary obtains more than one privacy break then they can achieve far better spying if they combine them together
08:53 < belcher> remember that "when the cookie meets the blockchain" paper? that was straightforward data fusion of the tx graph on the blockchain with tracking cookies
09:34 -!- nothingmuch [~user@213.152.162.5] has joined #joinmarket     
10:45 -!- nothingmuch [~user@213.152.162.5] has quit [Ping timeout: 250 seconds]     
10:50 -!- nothingmuch [~user@62.102.148.162] has joined #joinmarket     
14:29 -!- belcher [~belcher@unaffiliated/belcher] has quit [Ping timeout: 250 seconds]     
14:29 -!- nsh [~lol@wikipedia/nsh] has quit [Ping timeout: 250 seconds]     
14:31 -!- belcher [~belcher@unaffiliated/belcher] has joined #joinmarket     
14:36 -!- nsh [~lol@wikipedia/nsh] has joined #joinmarket     
17:47 -!- AgoraRelay [~jmrelayfn@p5DE4AB71.dip0.t-ipconnect.de] has quit [Ping timeout: 244 seconds]     
18:01 -!- AgoraRelay [~jmrelayfn@p5DE4AAAB.dip0.t-ipconnect.de] has joined #joinmarket     
19:56 -!- felix34 [~muh@5.2.76.19] has joined #joinmarket     
19:59 < felix34> are JM transactions vulnerabile to known-plaintext attacks, i.e if Eva knows that Tx1, tx2, ...TxN are Alice sending money to Bob, Charlie... Xenu, then how easy is it for Eva to look at the addresses those transactions have in common and derive information about Alice's wallet and other transactions?
21:55 -!- midnightmagic [~midnightm@unaffiliated/midnightmagic] has quit [Ping timeout: 260 seconds]     
22:12 -!- midnightmagic [~midnightm@unaffiliated/midnightmagic] has joined #joinmarket     
23:01 -!- Cory [~Cory@unaffiliated/cory] has quit [Ping timeout: 246 seconds]     
23:07 -!- Pasha [~Cory@unaffiliated/cory] has joined #joinmarket     
23:10 -!- Pasha is now known as Cory     
23:53 -!- belcher_ [~user@unaffiliated/belcher] has quit [Ping timeout: 246 seconds]     
23:53 -!- belcher_ [~user@unaffiliated/belcher] has joined #joinmarket