--- Day changed Tue Oct 25 2016
00:02 -!- RedEmerald [~RedEmeral@unaffiliated/redemerald] has joined #joinmarket     
00:03 < RedEmerald> is the pit down? my yg-pe is saying Socks5Error: 'Host unreachable' for both onion links
00:06 < RedEmerald> https://cyberguerrilla.info/ isn’t loading for me either
00:08 -!- RedEmerald [~RedEmeral@unaffiliated/redemerald] has quit [Quit: RedEmerald]     
00:40 < waxwing> the pit is still up on agora, unfortunately most users don't know about it
00:40 < waxwing> oh he's gone
01:25 < belcher> waxwing i think its a good solution fwiw
01:26 < belcher> that or adlai's idea of !fill'ing more makers than you need
01:27 < waxwing> belcher: thanks, me too, i ran a tumbler test ok too last night, but i need to go back and check the case of the restart when a user chooses to disable this feature.
01:27 < waxwing> belcher: yes as i was at pains to point out in the PR, this is *not* intended as a replacement for that, not at all
01:27 < belcher> yes     
01:28 < waxwing> i think fundamentally it's about re-thinking what was previously an excellent solution, the recovery code, with something that works at least reasonably well in an environment where requests are not free
01:30  * waxwing afk for a couple of hrs
04:26 -!- Giszmo [~leo@pc-40-227-45-190.cm.vtr.net] has joined #joinmarket     
05:47 < waxwing> tweaked 647 a bit, see comment, seems to be testing out well
06:12 < GithubBot5678> [joinmarket] AdamISZ closed pull request #641: Modify default fee to 0.02% (develop...default_fee_change) https://git.io/vP90O
06:36 -!- instagibbs [~instagibb@pool-100-15-114-3.washdc.fios.verizon.net] has joined #joinmarket     
06:41 -!- Socket_0x03 [~Socket_0x@c-66-176-87-156.hsd1.fl.comcast.net] has joined #joinmarket     
06:53 -!- freekevin [freekevin@gateway/shell/xshellz/x-jojylhgjykdgzeku] has quit [Ping timeout: 250 seconds]     
07:03 -!- Socket_0x03 [~Socket_0x@c-66-176-87-156.hsd1.fl.comcast.net] has quit [Quit: Leaving]     
07:04 -!- Socket_0x03 [~Socket_0x@c-66-176-87-156.hsd1.fl.comcast.net] has joined #joinmarket     
07:17 -!- Socket_0x03 [~Socket_0x@c-66-176-87-156.hsd1.fl.comcast.net] has quit [Remote host closed the connection]     
07:25 -!- Cory [~Cory@unaffiliated/cory] has quit [Ping timeout: 256 seconds]     
07:27 -!- Cory [~Cory@unaffiliated/cory] has joined #joinmarket     
07:28 -!- owowo [ovovo@gateway/vpn/mullvad/x-ilfadhhwzpaybxtz] has quit [Ping timeout: 244 seconds]     
08:55 -!- shinobimonkey [~vagabond@108.59.8.142] has quit [Ping timeout: 252 seconds]     
09:51 < JM-IRCRelay> [AlexCato] people are cloning the master branch by default and running into the usual problems (e.g. IndexError, new post here: https://www.reddit.com/r/joinmarket/comments/59btis/that_indexerror/ ). Would love to see a working master branch because thats what new users download
10:00 < waxwing> yes, the indexerror one was significant enough for a new release imo, but i wanted to wait until we had an amelioration of the unresponsive bots
10:02 < JM-IRCRelay> [AlexCato] yeah, understandable to fix that right away as well
10:03 < waxwing> that's pretty interesting: https://twitter.com/jp_koning/status/790938295215718400
10:09 < arubi> wow pretty cool.  also 'kidnapers', guess they needed less P's back then :)
10:11 < JM-IRCRelay> [AlexCato] indeed... though that must have been a lot of work back in the day
10:13 < JM-IRCRelay> [AlexCato] btw, I can help with updating the readme for a new version 0.2.2 (i guess?) if you like, so that links & mentioned versions are in sync with code
10:13 < waxwing> alexcato , i usually do that right at the end, but it wouldn't hurt, thanks
10:17 < JM-IRCRelay> [AlexCato] it's a bad use of your time to do such menial work :)
10:18 < waxwing> i appreciate the sentiment, but that part is easy compared to doing the release notes, as i'm sure belcher will confirm :)
10:19 < waxwing> most helpful thing right now is testing #641 either on regtest or mainnet, to sanity check my tests. running more tests of the tumbler with it now.
10:20 < waxwing> oh nice another one finished, 10% maliciousness and completed OK.
10:21 < waxwing> sorry #647 i meant
10:44 < waxwing> of vague interest: https://www.reddit.com/r/Bitcoin/comments/59c1kg/expect_no_privacy_from_shapeshift/
10:49 < arubi> "I have been harassing..." - with warrants? :P
10:49 < instagibbs> good example of why "mixing" via exchanges is a bad idea
10:50 < waxwing> it's kinda cool being able to see all the transactions. in a way :)
10:51 < waxwing> probably you can feed the API into a "shitcoin sentiment analysis" neural net for your high frequency trading needs :)
10:52 < arubi> most of these shitcoins probably have just about zero transactions a day
10:52 < arubi> not too hard tracking around someone and their funds
10:53 < instagibbs> I'm sure they're already all parked at exchanges :)
10:54 < instagibbs> from what I've been told, most exchanges are a few nodes run by devs and an exchange or two
10:54 < waxwing> yeah but you need shapeshift to move them from one exchange to the other :)
10:54 < waxwing> "node"? what's that? :)
10:54 < instagibbs> it's the thing that makes sure that the blocks are signed correctly by the dev :)
10:55 < arubi> it's that 100gb rock that I keep carrying around since years now :P
10:57 < waxwing> you are both wrong. node is what you write your beautiful ICO webpage in :)
10:57  * arubi is not a brogrammer :(
10:59 -!- owowo [~ovovo@unaffiliated/ovovo] has joined #joinmarket     
11:12 < JM-IRCRelay> [AlexCato] from the readme.md:
11:12 < JM-IRCRelay> [AlexCato] It's aim is to improve the confidentiality and privacy of bitcoin transactions, as well as improve the capacity of the blockchain therefore reduce costs.
11:12 < JM-IRCRelay> [AlexCato] uh, a JM aim is to reduce blockchain capacity? With all the maker's internal adresses, this seems to have the opposite effect
11:18 < arubi> well if makers are also patient takers, maybe a single jm can be useful as actual payment for multiple parties in a single tx?
11:18 < arubi> a single cj*
11:20 < JM-IRCRelay> [AlexCato] ic, true in concept!
11:20 < JM-IRCRelay> [AlexCato] now the patientsendpayment script just needs to work :)
11:20 < arubi> I actually always wondered why there's a difference between a maker and a taker, and really a maker only exist because there's value in it.  if it was all "free", then maybe there will only be takers in a rush, and patient takers :)
11:20 < arubi> hehe :)
11:23 < arubi> patient taker is a really cool concept in jm, and it's been around since forever..  do I mind paying a fee and wait 2-6 blocks, or do I just wait 2-6 blocks and cj for free? :)  I always liked it
11:23 < JM-IRCRelay> [AlexCato] the release-notes-0.x.x.md files: pure manual work or is there some kind of automation to automatically add the commits & commit-messages?
11:28 < JM-IRCRelay> [AlexCato] this error is reported to be happening on the current develop branch: https://paste.ubuntu.com/23380018/ . Any ideas?
12:28 < waxwing> alexcato there seems to be no orders; might be a case of no coins, or misconfigured settings so an order isn't generated or something? would need to know which script etc
12:29 < waxwing> alexcato w.r.t. the blockchain capacity comment, i vaguely remember reading that and thinking that i didn't agree with that, i don't think it should be there
12:30 < waxwing> in theory it can be a small saving in capacity, in practice peole get fungibility to whatever extent they get it, by creating a lot of extra ones; that's how it is for now, anyway
12:31 < waxwing> release notes are manual for now, yeah (well save some trivial find and replace scripting maybe), i don't think we are near a situation where we can automate it yet
12:52 < belcher> yes thank god for the patientsendpayment script, if only to remind people
12:53 < belcher> today i was actually thinking how comparatively easy it would be to add something like it to the tumbler, an option to be a maker for a while to reduce fees
12:55 < waxwing> yes, interesting point there (switch roles in tumbler). would go a way to addressing concerns in my tumbler_privacy doc. at some point when things are more stable that could be a natural point of focus for the project - how to incentivize more role-switching behaviour
12:57  * waxwing plays coveralls roulette and comes up green
12:58 < JM-IRCRelay> [AlexCato] ...and since the tumbler has to wait for at least 5 confirmations for each recently moved utxo anyways (but the restriction does not apply to being a maker), this is actually a pretty smart idea
12:58 < waxwing> yes that's a good point but consider that if you do a maker-side join you create *new* utxos :)
12:58 < waxwing> and consume old ones
12:59 < waxwing> but in practice it might work out as you say
13:00 < belcher> and the tumbler doesnt care too much about the actual coinjoin amount
13:02 < waxwing> yeah great idea i expect .. certainly for the more patient cases
13:04 < waxwing> belcher: probably just expand patientsendpayment into one super-bot that does every conceivable thing and delete the rest :)
13:09 < JM-IRCRelay> [AlexCato] PR with updated readme & suggestion for release notes: not gonna be done today, i'll hopefully do that tomorrow
13:17 < waxwing> alexcato thanks :)
13:17 < waxwing> belcher: this is a colourful read, thought you might find it interesting: https://www.reddit.com/r/Bitcoin/comments/59btk4/bitcoin_seminars_lectures_in_london/d97ac9t/
13:17 < belcher> ty     
13:18 < waxwing> lol "come check out IBM fumbling around naked without a clue"
13:18 < waxwing> actually that paragraph only gets better :)
14:08 < waxwing> 137 in the pit, 75 counterparties. and our 7.49 is not there, either (at least not with that equal amt)
15:33 -!- owowo [~ovovo@unaffiliated/ovovo] has quit [Read error: Connection reset by peer]     
15:50 -!- owowo [~ovovo@unaffiliated/ovovo] has joined #joinmarket     
16:20 -!- viasil_ is now known as viasil     
16:30 < belcher> " It's the heartbeat of London's crypto community still relatively untapped by company people, which just goes to show you how uninterested companies are in getting to grips with 'blockchain tech'" <---- i have the opposite experiance fwiw
16:31 < belcher> its right near the city so a few finance types attend, i remember getting a somewhat suspicious reception when i said i was working on something to make bitcoin more anonymous
16:32 < belcher> on the other hand i havent been that much, other times people were very interested
17:06 < belcher> a taker who used my maker seems to have constructed a tx with a low fee rate
17:07 < belcher> ~6000sat/kb and ~4000sat/kb, those wont confirm until something like sunday night 
17:22 < belcher> there is certainly some competition for block space now, the 1 block confirm target rate is up ~97000 sat/kb, i never seen it go above 100k wonder if it will this time
17:47 -!- Giszmo [~leo@pc-40-227-45-190.cm.vtr.net] has quit [Quit: Leaving.]     
18:23 -!- Taek [~quassel@2001:41d0:1:472e::] has quit [Ping timeout: 250 seconds]     
18:25 -!- Taek [~quassel@2001:41d0:1:472e::] has joined #joinmarket     
18:42 -!- moli [~molly@unaffiliated/molly] has quit [Read error: Connection reset by peer]     
19:08 -!- pigeons [~pigeons@94.242.209.214] has joined #joinmarket     
19:23 -!- moli [~molly@unaffiliated/molly] has joined #joinmarket     
19:46 < pigeons> how long do the blacklists last?
20:53 -!- deafboy_ is now known as deafboy     
22:54 < waxwing> pigeons: eternity, but see https://github.com/JoinMarket-Org/joinmarket/issues/620
22:58 < pigeons> between that and all the "Authorisation failed" i had to just give up 
22:59 < waxwing> pigeons: you mean give up doing a sendpayment, or?
22:59 < pigeons> yes     
22:59 < pigeons> imported the key into bitcoind to spend
23:00 < waxwing> right, note a few things: first, there is now a direct send so you don't have to import private keys
23:00 < waxwing> second, there is a PR 647 which should fix a very common case causing blocking
23:00 < waxwing> third, it also does depend on how you did it
23:01 < waxwing> did you (a) send into your wallet with multiple utxos (sendtomany), not critical but useful, did you wait 5 blocks before starting (that'll cause a lot of authorisation failed if you don't)
23:02 < waxwing> also until 647 is merged (should be very soon), it's a good idea to use -P and pick. i always do that.
23:03 < waxwing> the pit is actually very active right now, seems some people are figuring it out, don't ask me who. it is indeed too fragile for casual use, but it rather depends on if we have malicious bots around
23:03 < pigeons> the payment to my joinmarket wallet was 2 outputs, one was change
23:03 < waxwing> if we don't, you should be able to do it, and if you don't know why it failed after reading commitments_debug.txt (please tell me you read that pigeons!), i'd like to know what happened
23:04 < pigeons> yes it was more than 5 confirmations
23:04 < waxwing> pigeons: yeah that should be fine really
23:04 < pigeons> yes it got blacklisted because it was seen too many times
23:04 < pigeons> so the real question is why was it ailing 5 times 
23:05 < waxwing> when was this? last 12-24 hours or before?
23:05 < pigeons> yes last within the last 18 hours
23:05 < pigeons> i never had any problems before
23:05 < waxwing> this is why 647 is so urgent imo, it should fix exactly this case which people have been complaining about for weeks, where even one non-responsive bot will fail and burn a commitment
23:06 < waxwing> as soon as it is merged i will make a release, there are plenty of other somewhat important fixes/features but that's crucial
23:06 < pigeons> ok cool 
23:07 < waxwing> just to be sure, you didn't edit the joinmarket.cfg fields about this right
23:07 < pigeons> yes one sec i edit one field that something told me too, and it didnt help
23:07 < pigeons> yes i tried again with a higher taker_utxo_retries 
23:08 < waxwing> well, you shouldn't do that, but i'm guessing you tried that *after* you'd already had the message saying you'd used up all your tries, in which case it's academic
23:08 < pigeons> correct. I think it was the commitments_debug.txt that seemed to suggest it
23:09 < waxwing> i think it said it's highly inadvisable or something :) i mentioned it there because in some future there might be some debate about what the value should be, for now it's too hard to change it globally
23:10 < pigeons> oh now i see it does 20 lines later at the bottom which i didnt read that time
23:10 < waxwing> fwiw there is a lot of activity at the moment, someone has figured this out ...
23:11 < waxwing> my guess is that a certain subset of users have just given up, after experiencing this kind of things, others more motivated have figured out what they need to. but no idea of course.
23:11 < waxwing> only thing is, last 24 hours, i think the main set of malicious bots dropped out, but if you were trying and just getting unresponsive makers, then it's likely that
23:11 < pigeons> the first line is (see taker_utxo_retries in the config), and the next 20 lines were things that werent wrong, utxcos that were well over 5 confirmations, so i didnt see the part about it being not advised :)
23:12 < pigeons> yes it seemed like just unresponsive makers
23:12 < waxwing> in the "sourcing commitments for joins" page which everyone must read, it puts in bold what you should/should not do
23:12 < waxwing> everyone who is a Taker must read, i should say
23:13 < waxwing> i said at the top that if you didn't have the patience to read it all, at least read the bold parts, that's essential
23:19 < JM-IRCRelay> [AlexCato] sure that the malicious bots are gone? They're not on the OB watcher, but my YG is missing there as well; so that might not be a totally reliable source. I also see 4 failed join attempts in exactly the malicious bots' range
23:21 < JM-IRCRelay> [AlexCato] btw, if we assume 1 non-responding maker on average, the random maker number range shouldnt be decreased just at this time; unless the initial idea of asking N+x number of makers is following in the future. Though even then i'd move that 4-6 to 3-5 back to then
23:30 < waxwing> it's academic really because using equal amts as a marker is silly, but pretty sure (ob-w is at least accurate at start up) as of some time yesterday, don't know right now.
23:31 < waxwing> re 4-6 3-5 yeah i can just revert that. probably best. i don't understand your last sentence.
23:31 < waxwing> afk     
23:33 < JM-IRCRelay> [AlexCato] what i mean by the last sentence: the commit, that implements sending !fill requests to N+x makers should be combined with the change to 3-5. Because at that point it doesnt really matter any more if there is 1 nonresponsive bot or not