--- Day changed Fri Oct 28 2016
01:00 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has joined #joinmarket     
01:05 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has quit [Ping timeout: 260 seconds]     
01:46 -!- lnostdal [~lnostdal@90.149.73.62] has quit [Read error: Connection reset by peer]     
01:46 -!- lnostdal [~lnostdal@62.90-149-73.nextgentel.com] has joined #joinmarket     
02:51 -!- imposter [uid57046@gateway/web/irccloud.com/x-iupwxmbgmfgwohdv] has joined #joinmarket     
02:58 -!- mkarrer [~mkarrer@128.red-83-47-106.dynamicip.rima-tde.net] has joined #joinmarket     
02:59 -!- mkarrer_ [~mkarrer@7.red-83-47-85.dynamicip.rima-tde.net] has quit [Ping timeout: 252 seconds]     
03:01 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has joined #joinmarket     
03:06 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has quit [Ping timeout: 250 seconds]     
03:40 -!- molz [~molly@unaffiliated/molly] has quit [Ping timeout: 256 seconds]     
04:17 -!- Cory [~Cory@unaffiliated/cory] has joined #joinmarket     
04:19 < waxwing> adlai: can i convince you about the above? the purpose of the minmakers var is to have a fallback that means txs go through the large majority of the time. switching 2 to 3 is i think going to be quite meaningfully reduce the success of that strategy, in an environment where people are not that keen on >5 counterparties due to cost, also worth mentioning that liquidity is thin at some amounts and so 6 and above may not be that practical.
04:20 < waxwing> yes, 2 counterparties is poor privacy, but it would only be the worst case scenario, and it's a hell of a lot better than getting stuck unable to do a transaction imo.
04:43 -!- Numin0us [~Numin0us@cpe-74-64-92-78.hvc.res.rr.com] has joined #joinmarket     
04:43 -!- Numin0us [~Numin0us@cpe-74-64-92-78.hvc.res.rr.com] has quit [Changing host]     
04:43 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has joined #joinmarket     
04:52 < waxwing> belcher: opinion? sorry i know it seems bikesheddy but i feel it's important practically
05:00 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has quit [Ping timeout: 260 seconds]     
05:09 -!- gijensen92 [~gijensen@gijensen.xyz] has joined #joinmarket     
05:10 -!- gijensen [~gijensen@gijensen.xyz] has quit [Ping timeout: 276 seconds]     
05:10 -!- gijensen92 is now known as gijensen     
05:20 -!- Netsplit *.net <-> *.split quits: Pilate, roasbeef
05:35 -!- roasbeef [~root@104.131.26.124] has joined #joinmarket     
05:36 -!- Pilate [~pilate@pilate.io] has joined #joinmarket     
05:36 -!- Pilate is now known as Guest35871     
05:45 < waxwing> ok, release notes ready. just waiting on that 1 point.
05:52 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has joined #joinmarket     
05:55 < waxwing> adlai: belcher ping
05:56 < waxwing> i'd like to do the release now, but i'd like a response on the above ^
06:26 < adlai> ugh.
06:27 < adlai> so, thinking about it pragmatically... maybe we should reduce to 2, but include a warning that you should always check to see how many counterparties were used, and if it was too few, you should consider doing more joins
06:28 < waxwing> well consider 2 outcomes: 1/ people get less counterparties than they were expecting, ask what's going on, we tell them to bump the value 2/the default is 3, they choose -N 4 and ask why the upgrade still leaves them with stuck coins all the time?
06:28 < waxwing> i want outcome 1 over outcome 2, it is far less disruptive and not that big of a deal
06:29 < waxwing> well "we tell them to bump the value" could even be we tell them to set it to zero and not use the feature, of course, minor detail
06:34 -!- Giszmo [~leo@pc-40-227-45-190.cm.vtr.net] has joined #joinmarket     
06:38 < waxwing> btw there is already a warning/note before you choose 'y' when the feature is enabled. for sendpayment anyway. the short version is i don't want defaults leading to 'stuck' condition, as it so often does now, this a very bad thing (TM). if next release enables a smarter solution, great.
06:41 < adlai> i guess disabling the feature gets you stuck in the current situation, due to the flakes
06:42 < waxwing> yes, as of today, i wouldn't recommend disabling it at all. i used it yesterday/day before and will for sure continue to whenever i want to use joinmarket to send coins.
06:43 < waxwing> but then my use case is that kind of wishy-washy 'send a single payment as a join' which isn't what really counts.
06:48 < waxwing> well, meh, maybe that downplays the value/importance of single sendpayments. if nothing else they're a natural starting point.
06:49 < waxwing> i'll be back in a couple of hours; is it ok to revert? i even think backtracking on (5,7) to (4,6) is logical, but i don't think that's really important at all, just the min 2 is what i'd like to switch back, is that OK?
06:56 -!- molz [~molly@unaffiliated/molly] has joined #joinmarket     
07:00 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has quit [Ping timeout: 245 seconds]     
07:00 < adlai> yes, i guess if the minimum is 2, the other increases are not necessary. oh well... this does basically mean that flooding the marking with flakes reduces the anonymity set people get, at ~no cost
07:16 < waxwing> at the moment flooding the market with fakes reduces the anonymity set much more profoundly - it prevents joins happening at all.
07:33 -!- DeathShadow--666 [~IDSE@S0106a84e3f595813.vc.shawcable.net] has quit [Read error: Connection reset by peer]     
07:35 -!- DeathShadow--666 [~IDSE@S0106a84e3f595813.vc.shawcable.net] has joined #joinmarket     
08:51 < waxwing> adlai: so is it OK? I understand you don't like this solution, but it is a solution, and it works, and i want to release it. default can be changed, but i want the default not to fail often.
09:01 -!- Guest35871 [~pilate@pilate.io] has quit [Changing host]     
09:01 -!- Guest35871 [~pilate@infoforcefeed/pilate] has joined #joinmarket     
09:01 -!- Guest35871 is now known as Pilate     
09:02 < waxwing> i'm going to go ahead and release with default 2 unless i hear otherwise belcher adlai 
09:29 -!- coins123 [~coins123@unaffiliated/coins123] has quit []     
09:40 -!- iinaj [sid110431@gateway/web/irccloud.com/x-hwbkleenswwvgdwu] has quit [Write error: Connection reset by peer]     
09:40 -!- btcdrak [uid165369@gateway/web/irccloud.com/x-rciknzszcjtggjyd] has quit [Read error: Connection reset by peer]     
09:41 -!- iinaj [sid110431@gateway/web/irccloud.com/x-yvmglbposjvmolyc] has joined #joinmarket     
09:41 -!- btcdrak [uid165369@gateway/web/irccloud.com/x-mstznvfiweecveal] has joined #joinmarket     
10:16 < GithubBot5678> [joinmarket] AdamISZ pushed 1 new commit to develop: https://git.io/vXkFO
10:16 < GithubBot5678> joinmarket/develop 78470c3 Adam Gibson: finalize release notes 0.2.2...
10:17 < waxwing> please review release notes ^ if you have a chance, set minimum to 2 also in that, but did not change anything else (think 5,7 is fine), i will do the tag/release later tonight if all OK.
10:44 < adlai> so to recap: POLICY.minimum_makers will default to 2, so joins complete as often as possible (with more than one counterparty). and defaults will be [5,7] for sendpayment, and mean=6 sigma=1 min=4 for tumbler, so that there's a good chance users will still get reasonable counterparty counts in each join
10:45 < waxwing> yeah, i think that's fine
11:02 -!- chuckymcgee [443460bf@gateway/web/freenode/ip.68.52.96.191] has joined #joinmarket     
11:14 < chuckymcgee> So as far as moving joinmarket along, I'd like to work on a quick Youtube clip + infographic, however, I think there needs to be an easier outlet for takers to get involved. What would be the types of technical and political hurdles needed to integrate joinmarket into, say, Electrum or OpenBazaar to support joinmarket?
11:15 < chuckymcgee> I see some work on the Electrum plugin on the joinmarket github
11:15 < waxwing> chuckymcgee: there was a GUI interface for sendpayment, and an electrum plugin in protean form that i did, but there are some steps between here and re-initiating those.
11:16 < waxwing> after releasing 0.2.2 in a few hours, i plan to going back to look at refactoring to help that side of things. sorry it's a bit complicated to explain
11:17 < waxwing> this was working before the new version: https://github.com/AdamISZ/electrum-joinmarket-plugin, but only in as much as a few people tested it, it still needed a lot of work
11:18 < waxwing> and this i marked as "in limbo" since upgrade after 0.2 is difficult, the problem is that the core code needs refactoring beforehand imo https://github.com/JoinMarket-Org/JMBinary
11:18 < waxwing> although nothing stopping other people working on it
11:18 < waxwing> also there were serious efforts to build a Bitcoin Core integration
11:18 < waxwing> https://github.com/JoinMarket-Org/joinmarket/tree/bitcoind-int
11:19 < waxwing> but like i say, personally i think the core of joinmarket should be refactored first, although it's really quite debatable exactly what should be done.
11:19 < waxwing> joinparty has also built a codebase for taker side, plugged into libbitcoin
11:20 < waxwing> afk for an hr or two
11:25 < chuckymcgee> Gotcha. And what do you mean when you say "refactoring the core" of joinmarket?
11:27 -!- imposter [uid57046@gateway/web/irccloud.com/x-iupwxmbgmfgwohdv] has quit [Quit: Connection closed for inactivity]     
12:05 < adlai> chuckymcgee: restructuring so that it's easier to invoke the primitive operations from other software
12:16 < chuckymcgee> What would that restructuring process involve, exactly?
12:33 < waxwing> chuckymcgee: making it have properly separated layers, examples might be: having the basic set of functions installable as a package, removing unnecessary code duplication, more modular so easier to set up tests, and perhaps certain architectural changes with a view to what adlai said
12:34 < waxwing> take as an example, the joinmarket-qt (in the JMBinary repo) which quite a few people were using for quick-installable joinmarket, even on tails, i had to do some really hairy copy-paste stuff every time there was a new release.
12:35 < waxwing> i could have just mindlessly continued doing that, but it was getting even harder with 0.2, so i set it to one side until some work went into refactoring (and maybe go with a different "easy" installation interface like electrum, not sure). but hands totally full until now just making 0.2 work, it was (as fully expected) a significant disruption.
12:35 < waxwing> anyway will tag now
12:43 < GithubBot5678> [joinmarket] AdamISZ merged develop into master: https://git.io/vXIYn
12:44 < chuckymcgee> And so who is capable of executing this restructuring?
12:45 < waxwing> anyone in principle. anyone is also capable for example of updating the existing Joinmarket-Qt; as i said in the notes, i'd be welcome to help anyone who has questions.
12:46 < waxwing> as you can see ^ release is done, now will do pit, bitcointalk and reddit. belcher pls tweet when convenient
12:46 < waxwing> err "i'd be welcome"? heh :)
13:11 < chuckymcgee> How many hours would you estimate it would take a talented programmer to progress here? What's the ultimate rate limiting step here?
13:12 < waxwing> done another test with -N 5, got 3 counterparties.
13:12 < waxwing> (went through ofc). just sanity checking from master.
13:13 < waxwing> chuckymcgee: no idea really, lots of different ways to look at it. i think it might help get more people involved if we refactored it. but it's kind of a big topic.
13:15 < chuckymcgee> I'm just trying to wrap my head around the scope of this as someone with an extremely limited programming background. Thanks
13:16 < waxwing> as i think i said above, now that update is done, i'd personally like to have a crack at it myself. i've made a start but put it to one side to implement the new things you see in the release notes. adlai is looking at making a more sophisticated fix to our main current issue (nonresponsive makers) also.
13:17 < waxwing> chuckymcgee: well, consider that this started as a small project by belcher with a set of scripts to do a market for coinjoins. it has grown, and naturally people want easier interfaces to it, because it isn't easy to use. some people don't even agree that that should be a focus, and for sure it's not the only issue: the other issue is making it reliable in the face of attackers.
13:18 < waxwing> wow, that 1 confirmed fast, you're welcome to some satoshis makers :)
13:20 < chuckymcgee> And that attacker issue was partially addressed in 0.2.0 deterring snoopers, what are the other issues with attackers? Information-leaking makers?
13:21 < waxwing> well the "concern du jour" is makers who simply don't respond midway through. that's been blocking a fair few transactions last few weeks.
13:21 < waxwing> so in the just-released release, you can now complete with less than the specified number of makers (as i just did; got 3 instead of the requested 5)
13:21 < waxwing> this is obviously a partial solution, but it may be a good fix for now, we'll see.
13:22 < waxwing> see previously, in 0.1, the solution was always "just keep trying" which worked well. unfortunately snoopers used this window to request endlessly, gathering a full picture of owned utxos.
13:22 < waxwing> thus making joinmarket fail to do its job, at least, against them.
13:23 < waxwing> actually i exaggerate, they didn't gather *all* utxos, just a lot of them. probably enough to figure out who owns what.
13:24 < waxwing> so 0.2 in its bare form solved one problem, but not another: it stopped people from endlessly requesting, gathering utxo information, but since requests are now not free, we cannot avoid sybil attacks by just trying as many times as needed any more.
13:24 < waxwing> to be fair, i did provide other avenues in 0.2 to re-try more often: you can add external utxos, but it seems that people haven't delved into that (or if they have they haven't talked about it)
13:25 < waxwing> i'd hoped that the prescribed advice: to fund with 3 utxos instead of 1, giving *9* tries available to start with, would be enough, but that worked less well than intended because (a) people didn't follow that advice and (b)makers started popping up that seem to be deliberately blocking transactions.
13:26 < waxwing> and hence the above new feature
13:26 < chuckymcgee> Would it be possible or at all helpful to levy a small fee on the taker/maker to attempt a transaction? A satoshi paid to the server? That may be stupid.
13:27 < waxwing> chuckymcgee: there were various solutions proposed to solve the taker-side spying issue (endless requests), all with a similar theme to that, but not a literal payment to anyone
13:27 < waxwing> system isn't supposed to be centralized, so it wouldn't literally be a payment to a server
13:28 < waxwing> one idea was a "pay to contract hash", kind of like a fidelity bond. what was implemented (podle) was an even less-intrusive-to-user thing.
13:28 < chuckymcgee> I guess whatever IRC chat server the transaction is arranged on?
13:28 < waxwing> podle effectively rate-limits use based on owned utxos, without having to spend them
13:29 < waxwing> or reveal them, except to your counterparties
13:29 < adlai> P2CH isn't a bond though, it's a fee
13:29 < waxwing> oh? i must have forgotten, i thought it simply locked it up
13:29 < adlai> and it also rate limits thanks to requiring bitcoin confirmations. i guess podle rate-limits like this too
13:29 < chuckymcgee> I'm not familiar with podle
13:30 < adlai> waxwing has a blog post explaining it, on http://joinmarket.me/blog
13:31 < waxwing> yeah website down, dang, just a sec
13:31 < adlai> chuckymcgee: https://joinmarket.me/blog/blog/poodle/
13:32 < waxwing> oh; is it not down? maybe my connection?
13:32 < adlai> waxwing: interesting discussion on the post!
13:32 < waxwing> which post?
13:33 < adlai> https://joinmarket.me/blog/blog/poodle/#comments
13:33 < adlai> looks like it's time for pay-to-comment-hash!
13:33 < waxwing> damn, so it's really not down? what is up with my browser. i'm getting nginx page ?!
13:33 < waxwing> must have been last few days; spammers?
13:33  * adlai typed the first link manually, typed 'http' out of laziness; looks like https is required
13:34 < waxwing> they probably hacked all my bitcoinz
13:34 < adlai> seems to all be from the last 25 hours
13:35 < waxwing> oh bugger. yes, rate-limited comments plz.
13:35 < waxwing> :(     
13:36 < waxwing> i'll just enable approvals or whatever.
13:39 < chuckymcgee> Lol chrome warns me about unsafe scripts on your blog
13:49 < waxwing> i've been attacked with spam comments i'm afraid, fixing it now
14:04 -!- Numin0us [~Numin0us@cpe-74-64-92-78.hvc.res.rr.com] has joined #joinmarket     
14:04 -!- Numin0us [~Numin0us@cpe-74-64-92-78.hvc.res.rr.com] has quit [Changing host]     
14:04 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has joined #joinmarket     
14:10 < waxwing> ok it's back to normal now https://joinmarket.me/blog/blog/poodle/ sheesh. anyway off to bed :)
14:46 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has quit [Ping timeout: 250 seconds]     
14:49 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has joined #joinmarket     
14:54 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has quit [Ping timeout: 244 seconds]     
14:55 < belcher> goodnight waxwing 
14:55 < belcher> hello chuckymcgee 
14:55 < belcher> just got back, read all of today's talk
14:55 < chuckymcgee> howdy
15:01 < belcher> theres also #bitcoin and other channels on freenode
15:23 < belcher> heres the tweet for 0.2.2 https://twitter.com/joinmarket/status/792129520920850432
16:00 -!- sn0wmonster [~yeti@taskhive/developer/sn0wmonster] has quit [Ping timeout: 260 seconds]     
16:09 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has joined #joinmarket     
16:10 -!- sn0wmonster [~yeti@taskhive/developer/sn0wmonster] has joined #joinmarket     
16:20 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has quit [Ping timeout: 250 seconds]     
16:30 -!- pero [~pero@unaffiliated/pero] has joined #joinmarket     
16:30 -!- pero [~pero@unaffiliated/pero] has left #joinmarket ["Leaving"]     
16:42 -!- Giszmo [~leo@pc-40-227-45-190.cm.vtr.net] has quit [Quit: Leaving.]     
16:44 -!- Giszmo [~leo@pc-40-227-45-190.cm.vtr.net] has joined #joinmarket     
17:29 < belcher> adlai i got the same bug as you in wallet tool history
17:29 < belcher> ill try debugging
18:46 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has joined #joinmarket     
19:27 < belcher> adlai looks like the range wasnt big enough
19:27 < belcher> it seems to be that iv made ~13% per annum, im not sure i believe this number
19:28 < belcher> change this line: r = brentq(f, a=0.5, b=-0.5, args=(deposits, deposit_times, now, balance)), note the 0.5 -0.5 instead of 0.1 -0.1
19:35 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has quit [Ping timeout: 245 seconds]     
19:47 < belcher> iv done some calculations manually on the data and it all gives 2-3%, nothing near 10%
19:47 < belcher> really strange, no idea why
20:07 -!- btcdrak [uid165369@gateway/web/irccloud.com/x-mstznvfiweecveal] has quit [Quit: Connection closed for inactivity]     
21:31 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has joined #joinmarket     
21:35 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has quit [Ping timeout: 245 seconds]     
21:58 -!- Giszmo [~leo@pc-40-227-45-190.cm.vtr.net] has quit [Quit: Leaving.]     
22:02 -!- Socket_0x03 [~Socket_0x@c-66-176-87-156.hsd1.fl.comcast.net] has joined #joinmarket     
22:08 -!- Socket_0x03 [~Socket_0x@c-66-176-87-156.hsd1.fl.comcast.net] has quit [Quit: Leaving]     
22:59 -!- btcdrak [uid165369@gateway/web/irccloud.com/x-wahkdadkfteibdna] has joined #joinmarket     
23:24 < chuckymcgee> belcher: maybe it's doing something strange with the dates certain deposits were made into the wallet? Perhaps a weighted average is inflating yield?
23:32 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has joined #joinmarket     
23:37 -!- Numin0us [~Numin0us@unaffiliated/numin0us] has quit [Ping timeout: 260 seconds]