--- Day changed Mon May 14 2018
00:54 -!- lnostdal [~lnostdal@77.70.119.51] has quit [Ping timeout: 264 seconds]     
01:25 -!- Bailey39Lubowitz [~Bailey39L@ns334669.ip-5-196-64.eu] has quit [Remote host closed the connection]     
01:52 -!- lnostdal [~lnostdal@77-85-34-21.ip.btc-net.bg] has joined #joinmarket     
01:53 -!- lnostdal [~lnostdal@77-85-34-21.ip.btc-net.bg] has quit [Max SendQ exceeded]     
01:53 -!- lnostdal [~lnostdal@77-85-34-21.ip.btc-net.bg] has joined #joinmarket     
02:02 -!- liead [~adlai@unaffiliated/adlai] has joined #joinmarket     
02:03 -!- adlai [~adlai@unaffiliated/adlai] has quit [Ping timeout: 248 seconds]     
02:38 -!- lnostdal [~lnostdal@77-85-34-21.ip.btc-net.bg] has quit [Ping timeout: 268 seconds]     
02:59 -!- lnostdal [~lnostdal@85-118-74-155.mtel.net] has joined #joinmarket     
03:39 -!- asymptotically [~asymptoti@gateway/tor-sasl/asymptotically] has joined #joinmarket     
05:31 -!- quitobro [~quitobro@pool-108-41-0-186.nycmny.fios.verizon.net] has joined #joinmarket     
05:35 -!- quitobro [~quitobro@pool-108-41-0-186.nycmny.fios.verizon.net] has quit [Ping timeout: 260 seconds]     
05:39 -!- quitobro [quitobro@gateway/vpn/privateinternetaccess/quitobro] has joined #joinmarket     
06:47 -!- lnostdal [~lnostdal@85-118-74-155.mtel.net] has quit [Ping timeout: 256 seconds]     
06:57 < waxwing> arubi, so i'm merging that right? seems like a pretty big change but cool, thanks for doing it. do i take it those tests run in parallel? so it's about 10 minutes overall?
06:57 < waxwing> (also i worry about using tons and tons of travis resources? never really been sure about that)
07:02 -!- lnostdal [~lnostdal@77.70.119.51] has joined #joinmarket     
07:11 < waxwing> oh, you said: 2-3 minutes expected due to caching. hmm.
07:25 < arubi> waxwing, 2-3 minutes less than what's currently shown*
07:26 < arubi> it's running a batch of 5 in parallel I think.  then the rest
07:27 < arubi> let's do it this way, don't merge yet.  I'll keep it updated with current master for a bit and we'll see if there aren't any surprises
07:29 < waxwing> ok     
08:31 -!- GitHub150 [GitHub150@gateway/service/github.com/x-cjtuwbszivqitjep] has joined #joinmarket     
08:31 < GitHub150> [joinmarket-clientserver] AdamISZ opened pull request #151: Ensure sweep fraction in tumbler is greater than a nominal amount (5%). (master...tumble-sweep-min) https://git.io/vpHxD
08:31 -!- GitHub150 [GitHub150@gateway/service/github.com/x-cjtuwbszivqitjep] has left #joinmarket []     
08:33 < waxwing> that fixes the nondeterministic test failure, but more detail in #150 for the zero people who are interested :)
08:45 < arubi> :)  gonna go over it soon
08:53 < waxwing> well you can if you want, but it's not super interesting. there are other conceivable tweaks to address the same problem, but i did what's the simplest and it should be fine in pretty much any scenario.
08:54 < waxwing> only possible scenario it'd be bad is if you were doing like more than 20 transactions per mixdepth (a lot more, really), and nobody's going to do that really.
09:12 < arubi> I was curious about that python-y while loop really.  python notations are still weird for me so when there's a real world example I try to follow.  a while ago this kind of thing : 'amount_fractions = [1.0 - x for x in amount_fractions]' would really confuse me :)
09:13 < arubi> alright, when you merge that, it would be a nice opportunity to rebase the dockers PR.  as it's running many instances, it had that much higher chance to hit the 0.9998.. thing
09:18 < arubi> am I correct that "sum(amount_fractions)" would be run for every element in amount_fractions ?  probably meh as it's not a huge array but would be prettier to compute that once out of the loop and use the sum as a variable?
09:41 < waxwing> arubi, on the confusing syntax: they're termed "list comprehensions"; they are extremely weird looking at first, but nice for compactness.
09:41 < arubi> yea, I'm really starting to see that
09:42 < waxwing> on repeated calculation of sum: you're right, but yeah it's completely trivial here.
09:43 < arubi> alright, gtk.  I'll get pinged when it's merged and rebase the dockers
09:47 < waxwing> arubi, don't you think list comprehensions read like English "this = 10 times that for (every) that in some-list"
09:48 < waxwing> the real confusions lie with stuff like flattened lists: flat_list = [item for sublist in l for item in sublist]
09:48 < waxwing> although iirc there's another way to do that that's maybe even more compact (but doesn't look as crazy)
09:51 < arubi> yes, but eg the 'amount_fractions = [1.0 - x for x in amount_fractions]', I wouldn't be sure at first about the evaluation order, so I would be confused about how many times left hand side is being assigned the right hand side.  so in that example, I wouldn't be sure if the left hand side changes at each iteration of the loop and affects the next one or not, but now it's clear that the rhs is done entirely and only then the lhs is assigned the 
09:51 < arubi> list
09:54 < arubi> I wonder if there's conway's game of life written only using these list comprehensions, that would be fun to follow :)
09:57 < arubi> heh, https://stackoverflow.com/a/40575047/9068872
10:09 -!- asymptotically [~asymptoti@gateway/tor-sasl/asymptotically] has quit [Remote host closed the connection]     
10:21 < waxwing> yeah rhs all first before assignment. it would seem a bit mad (to me) to do it otherwise.
10:25 -!- asymptotically [~asymptoti@gateway/tor-sasl/asymptotically] has joined #joinmarket     
10:29 -!- GitHub70 [GitHub70@gateway/service/github.com/x-gfjtrgklannaedau] has joined #joinmarket     
10:29 < GitHub70> [joinmarket-clientserver] AdamISZ pushed 1 new commit to master: https://git.io/vpQYe
10:29 < GitHub70> joinmarket-clientserver/master 5b30404 AdamISZ: Merge #151: Ensure sweep fraction in tumbler is greater than a nominal amount (5%)....
10:29 -!- GitHub70 [GitHub70@gateway/service/github.com/x-gfjtrgklannaedau] has left #joinmarket []     
10:29 -!- GitHub161 [GitHub161@gateway/service/github.com/x-vqgffufrnsifsplb] has joined #joinmarket     
10:29 < GitHub161> [joinmarket-clientserver] AdamISZ closed pull request #151: Ensure sweep fraction in tumbler is greater than a nominal amount (5%). (master...tumble-sweep-min) https://git.io/vpHxD
10:29 -!- GitHub161 [GitHub161@gateway/service/github.com/x-vqgffufrnsifsplb] has left #joinmarket []     
10:30 < arubi> right in retrospect it seems obvious, but I come from bash so expansion of a word's value is always a literal ${} wrapped around the word, and pretty much no other language does that I was a bit confused (the exotic list comprehension syntax only added to the confusion :) ).  I'm really getting into python at work now so a lot of it is cleared
10:32 < arubi> alright, let's see if travis is smart enough to make use of the cache even though I force pushed a rebase
10:33 < arubi> yep, it is
10:41 -!- quitobro [quitobro@gateway/vpn/privateinternetaccess/quitobro] has quit [Quit: quitobro]     
10:52 < arubi> sweet, all of them took 2-3 minutes less each except for centos7 for some reason which actually took a few seconds longer even though it did use the cache.  prolly a fluke but will be good to keep an eye on (for me).  16:37 total build time vs. 19:33 with no cache
10:54 < arubi> tbh I come from a place where we're using IBM's LSF.  travis is so much better.  I'm amazed at how they offer this service for free for open source projects
10:57 -!- a87ry5 [~a87ry5@cpe-158-222-196-182.nyc.res.rr.com] has joined #joinmarket     
10:58 < arubi> there's openlava\tephra which is an open source fork of LSF made by some of the folks who wrote LSF, but IBM told them to stop whatever they're doing there.  shame.  it has a bunch of useful features and supposed to have a better scheduler.  oh well
11:01 < waxwing> IBM has to be slower because of the "revenue code" ;)
11:06 < arubi> hehe, hopefully they'll become irrelevant as people start ditching all the weird proprietary hardware which is pretty much the only reason to use their stuff since it actually does work absolutely anywhere.  we had rakes and rakes of Sun hardware, IBM bladecenters..  at some point the customers just straight told us ".. can't you just install ubuntu somewhere?", and now we only have a few
11:07 < arubi> now ARM is starting to take the place of those, and some of the dell intel's are probably next
11:13 < waxwing> right, network effects matter over time. makes you think about current multi-coin trends, it's maybe similar to how other previous tech was in the early days; lots of niche stuff, but over time certain "generally agreed" solutions win out.
11:14 < waxwing> not to say there isn't still a lot of diversity of course. oh well, it's complicated.
11:20 < arubi> that's it exactly.  everyone are trying to ride some network effect so eventually they become the dominant tech, the slight difference is that in crypto world, one mistake can cost millions for everyone involved and not just some company with its hot new trend or proprietary standard.  the DAO, that ethereum multisig contract..  probably more that of those that I just didn't follow.  eventually we'll have an easy mechanism of using adaptor 
11:20 < arubi> sigs in bitcoin and I think from that point on it's smooth sail.  just need to figure out confidential transactions as a soft fork ;)
12:10 -!- a87ry5 [~a87ry5@cpe-158-222-196-182.nyc.res.rr.com] has quit [Ping timeout: 248 seconds]     
14:23 < waxwing> arubi, this guy built a binary to spend a specific transaction :) https://twitter.com/AlphaJarm/status/996029352478498816
14:31 -!- liead is now known as adlai     
14:34 -!- asymptotically [~asymptoti@gateway/tor-sasl/asymptotically] has quit [Quit: Leaving]     
14:50 -!- belcher [~belcher@unaffiliated/belcher] has joined #joinmarket     
15:03 < AgoraRelay> [agora-irc/Kennu] is joinmarket-pit channel down for 6dvj6v5imhny3anf.onion ?
15:04 < waxwing> is that cgan or agora?
15:05 < waxwing> i think i saw cgan go down and come back up in my logs. not 100% sure.
15:12 < waxwing> hmm looking back at the logs past few hours, i do get the impression maybe cgan is not working. very few privmsgs being sent there from me. almost all to agora.
15:13 < waxwing> joins seem to be working fine though (as you'd expect)
15:14 < AgoraRelay> [agora-irc/Kennu] was mistake on my side ;)
15:14 < waxwing> ok     
16:01 -!- arubi [~ese168@gateway/tor-sasl/ese168] has quit [Remote host closed the connection]     
16:02 -!- arubi [~ese168@gateway/tor-sasl/ese168] has joined #joinmarket     
16:14 -!- lnostdal [~lnostdal@77.70.119.51] has quit [Excess Flood]     
16:14 -!- lnostdal [~lnostdal@77.70.119.51] has joined #joinmarket     
20:47 -!- AgoraRelay [~jmrelayfn@p5DE4A71F.dip0.t-ipconnect.de] has quit [Ping timeout: 260 seconds]     
21:04 -!- AgoraRelay [~jmrelayfn@p5DE4A5BB.dip0.t-ipconnect.de] has joined #joinmarket     
22:37 -!- viasil [~viasil@185.217.171.10] has quit [Read error: Connection reset by peer]     
22:38 -!- viasil [~viasil@185.217.171.10] has joined #joinmarket     
22:40 -!- viasil [~viasil@185.217.171.10] has quit [Read error: Connection reset by peer]     
22:43 -!- viasil [~viasil@185.217.171.10] has joined #joinmarket     
22:49 < arubi> whoa that AlphaJarm guy is really into libbitcoin.  (/work)