15:00:00 <karsten> #startmeeting metrics team 15:00:00 <MeetBot> Meeting started Thu Dec 22 15:00:00 2016 UTC. The chair is karsten. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:00 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic. 15:00:01 <asn> (btw karsten will you be atccc?) 15:00:11 <asn> (btw linda will you be at ccc?) 15:00:22 <karsten> asn: still uncertain. :/ 15:00:25 <asn> hello meeting 15:00:29 <dgoulet> karsten: :( 15:00:40 <linda> asn: I wont be. :( 15:00:56 <karsten> https://pad.riseup.net/p/3M7VyrTVgjlF <- please add agenda items. 15:00:56 <asn> ooh 15:01:01 <linda> asn: that's christmas time with the family, so I don't know if I can ever go (without making my family very upset). 15:01:35 <asn> makes sense :) 15:01:40 <asn> at least some sense 15:01:42 <hiro> yeah timing is a bit unfortunate for CCC 15:01:44 <linda> asn: hehe 15:02:01 <linda> hiro: agreed. 15:02:13 <linda> but maybe hackers don't generally celerbrate christmas?... 15:02:17 <linda> .... that was my inference. 15:02:30 <karsten> linda: do you want to lead the UX/website topic, or shall I? 15:02:42 <hiro> linda: I don't but my parents do 15:02:50 <linda> karsten: :3 you can lead! 15:03:00 <karsten> ok. 15:03:07 <karsten> any other topics? 15:03:19 <linda> karsten: I feel like we'd be talking more about stuff that affects you, and not too much on the design alterations. 15:03:24 <iwakeh> all fine. 15:03:47 <linda> I'm good too. 15:03:53 <karsten> okay! 15:03:57 <karsten> http://107.23.39.12/ 15:04:00 <karsten> let's talk about it. 15:04:11 <linda> is rafe not here? :o 15:04:15 <karsten> we already have feedback from RaBe. 15:04:21 <RaBe> i'm here :D 15:04:42 <karsten> on the pad now. 15:04:52 <karsten> so, adding downloads to the start page sounds good to me. 15:05:03 <linda> RaBe: hi! 15:05:04 <karsten> one minor change would be that I'd want to rename it to "Applications". 15:05:09 <RaBe> hi! 15:05:16 <karsten> because otherwise people will ask what they can download there. 15:05:45 <linda> karsten: I think as long as the sentence explains something, I'm fine with it. 15:06:20 <linda> karsten: sometimes, the top horizontal bar disappears. (i.e.:http://107.23.39.12/userstats-censorship-events.html) 15:06:33 <linda> I don't think that was intentional. 15:06:46 <Sebastian> atagar: do you think there would be a way to run the stem tests that would have caught the spec compliance issue automatically? Has that been tried at all? 15:06:56 <karsten> linda: ah, on table pages. will fix after the meeting. 15:07:16 <linda> karsten: :) 15:07:35 <linda> I also am mildly in favor of getting rid of the icons on top, since they seem distracting. 15:07:46 <karsten> awwwwww :( 15:07:47 <linda> But I'm not going to fight it. 15:07:53 <iwakeh> oh, there nice ! 15:07:58 <iwakeh> they are 15:08:02 <linda> ehhehe okay then 15:08:14 <karsten> we could take them out on small resolution. 15:08:22 <karsten> like we're taking out the ones in the primary navbar. 15:08:24 <RaBe> that's already happening 15:08:41 <iwakeh> well, they won't be visible in 3pt anyway ;-) 15:08:53 <karsten> only in the primary navbar. 15:08:54 <linda> oh, then I'm more okay with them. 15:09:06 <linda> another peice of feedback is that I wish there was a horizontal line in some tabs. 15:09:10 <karsten> try to resize your browser and see how they go away. 15:09:11 <linda> (i.e. http://107.23.39.12/userstats-censorship-events.html) 15:09:33 <linda> I wish there was a horizontal line around the tab. I don't know if I am communicating right 15:09:45 <iwakeh> there is. 15:09:50 <karsten> more generally, I wonder if we can pick shorter names. 15:09:52 <karsten> tab names. 15:10:01 <karsten> to avoid having four rows of them. 15:10:02 <iwakeh> yes, that'll be useful. 15:10:10 <iwakeh> but hard to come up with. 15:10:15 <linda> karsten: +1 15:10:19 <karsten> not trivial, indeed. 15:10:21 <RaBe> i'll add some lines and send them to karsten :) 15:10:28 <karsten> RaBe: sounds good. 15:10:46 <linda> RaBe: you got what I said? I explained it badly.. haha thanks. 15:11:13 <RaBe> linda: yeah, i'll try to make inactive and multi line tabs more visible... 15:11:45 <linda> RaBe: :D 15:11:57 <karsten> RaBe: but regarding icons going away, I think only the ones in the primary navbar go away. should we do the same in the secondary navbar? 15:12:07 <karsten> or does that not really matter? 15:12:17 <linda> karsten: I would argue that both should go and stay at the same time. 15:12:18 <RaBe> karsten: oh, absolutely, yes. 15:12:34 <RaBe> karsten: i'll send you something... 15:12:34 <karsten> ok. can you send me a patch for that? 15:12:37 <karsten> thanks. :) 15:12:55 <linda> on smaller screens, it seems like there is no distinction between the primary and secondary navigation. 15:13:16 <karsten> oh, right. 15:13:16 <RaBe> that is correct. we just have one mobile navigation 15:13:38 <karsten> I briefly tried adding a <hr class="visible...something>, but that didn't look great. 15:13:44 <linda> Hmm. I wish the secondary seemed more secondary. 15:14:09 <linda> Or it's not clear to me that operation isn't "operation metrics" and that servers are "server metrics" 15:14:09 <iwakeh> smaller? 15:14:24 <karsten> ah, navigation group? 15:14:40 <linda> I don't really have a set solution in mind. 15:14:59 <RaBe> i don't want to get rid of the icons, but it might help... 15:15:19 <karsten> in mobile view? why? 15:15:42 <RaBe> are we talking about mobile view or smaller views like tablets? 15:16:05 <karsten> I think mobile. 15:16:22 <linda> I was just looking at it on a smaller window on my desktop.. 15:16:30 <karsten> smallest possible? 15:16:50 <hiro> I like the icons actually 15:16:56 <linda> I would prefer something like listing the primary 6 when you click the hamburger icon 15:17:04 <RaBe> i think linda is talking about the 2nd nav bar in desktop mode on small devices looks as big as the main nav 15:17:06 <karsten> in the smallest possible view I have 12 entries with no distinction between primary and secondary. 15:17:11 <linda> and having a "more" section or something and requiring an extra click for the news, sources, etc. 15:17:21 <linda> karsten: yes, that's what I am talking about 15:17:30 <RaBe> linda: ah ok :) 15:17:33 <karsten> I wonder if we can add a horizontal line there. 15:17:39 <RaBe> karsten: sure 15:17:55 <karsten> or two small groups "Metrics" and "Documentation" or whatever. 15:18:02 <RaBe> karsten: i could also increase the first items size to make them a bit more "important" 15:18:15 <linda> I don't want to add additional font weights. :( 15:18:24 <linda> But I think the horizontal lines and sectioning would help. 15:18:34 <linda> Although i still would like them more hidden 15:18:59 <RaBe> linda: ok, i'll keep the size, but i'll send karsten an update for some kind of sectioning... 15:19:25 <ryonaloli> dgoulet: is there no way for the syscall() wrapper in torsocks to not use a whitelist? 15:19:33 <ryonaloli> it's causing so many problems. 15:19:53 <dgoulet> ryonaloli: #tor-project (meeting happening) 15:19:58 <dgoulet> here* 15:20:02 <ryonaloli> oops, sorry 15:20:20 <karsten> sounds good, RaBe. 15:20:29 <karsten> any feedback on the pad we did not address? 15:20:41 <karsten> any further ideas what needs tweaking? 15:20:55 <linda> isa did not like the icons.. lol 15:21:02 <linda> but that's all she really said 15:21:12 <karsten> and she likes other icons more? :D 15:21:26 <linda> karsten: the icons on the secondary bar 15:21:32 <karsten> ah ok. 15:21:45 <linda> Maybe i'd add some content to the lonely pages (dev and operation) 15:21:56 <linda> but I don't see anything else I would like to change. 15:22:01 <linda> It's so beautiful. *-* 15:22:12 <karsten> it is! 15:22:41 <iwakeh> yes :-) 15:23:19 <karsten> I also have a small list of things here that I want to fix. 15:23:25 <linda> ? 15:23:40 <karsten> - Organize operation and development pages. 15:23:45 <karsten> - add some way to link to #anchor. see https://collector.torproject.org/#available-descriptors 15:23:56 <linda> +1, +1 15:23:58 <karsten> - Should postpone tooltips for glossary items, which requires a fair bit of hacking and which we might have to re-implement after switching the web framework. 15:24:08 <karsten> - Update news.json with whatever changes have been made to the wiki page after the initial import. 15:24:17 <linda> oh, i see. 15:24:18 <linda> +1 15:24:56 <karsten> I guess those are the main ones. 15:25:14 <RaBe> do you need my help on any of that things on your list? 15:25:30 <karsten> on the anchors maybe? 15:25:54 <RaBe> what's the problem there? :) 15:26:03 <karsten> go to https://collector.torproject.org/#available-descriptors 15:26:12 <karsten> when you hover over "Available Descriptors", 15:26:16 <karsten> there's a #. 15:26:23 <karsten> that you can click and bookmark. 15:26:32 <karsten> no idea how mobile friendly that is. 15:26:46 <karsten> but I find that quite useful for linking to some part of a page. 15:26:56 <RaBe> it won't show up on mobile devices... I ll send you something for the desktop :) 15:27:01 <karsten> cool! 15:27:33 <karsten> regarding tooltips, how about we put the glossary back for now? ... 15:27:46 <karsten> or should we put in a dirty hack? 15:28:04 <iwakeh> no, rather wait. 15:28:12 <linda> I think we should put the glossary back. 15:28:24 <linda> As long as the tooltips happen eventually, I'd be happy. 15:28:25 <karsten> separate page under About/Glossary, or on Glossary? 15:28:33 <karsten> errr, 15:28:37 <karsten> separate page under About/Glossary, or on About? 15:28:52 <linda> I don't understand what you are saying? 15:29:00 <karsten> ah, sorry. 15:29:12 * iwakeh neither 15:29:16 <karsten> we could add a new section to http://107.23.39.12/about.html with all glossary items. 15:29:36 <karsten> or we could add a 1-sentence section there saying "Here's our <a href="glossary.html">glossary</a>. 15:29:50 <karsten> with a separate page for glossary.html, breadcrumbs Home > About > Glossary. 15:30:03 <iwakeh> this seems better. 15:30:03 <RaBe> i'd vote for a separate page... 15:30:11 <karsten> okay! 15:30:18 <linda> I like the idea of a separate page too. 15:30:22 <hiro> are we going to have a lot of temrs in glossary? 15:30:28 <hiro> s/temrs/terms 15:30:30 <karsten> 12? 15? 15:30:39 <hiro> ah yeah then it's better kept separated 15:30:41 <karsten> https://metrics.torproject.org/about.html 15:31:19 <karsten> alright. 15:31:39 <karsten> I'm counting 12 issues/suggestions on the pad. 15:31:46 <karsten> most of these seem easy. 15:31:54 <karsten> should we talk about next steps? 15:31:56 <linda> \^~^/ 15:32:37 <karsten> I briefly mentioned this plan before: how about we ask folks on tor-project@ to give us feedback? 15:32:54 <karsten> we could set up a pad to collect feedback. 15:33:25 <karsten> and I said earlier that we could set up a DNS redirect for metrics-staging.torproject.org, but maybe that's not worth the effort. 15:33:43 <linda> I thik that what you have set up is fine. 15:33:48 <karsten> b/c I'm not sure whether we need to migrate the current ec2 instance to a tpo instance. 15:33:50 <linda> And it would be nice to get some feedback. 15:34:05 <linda> (with parameters) 15:34:19 <linda> Like give people a week max, and maybe we can direct their feedback. 15:34:20 <iwakeh> do we have some questions regarding the feedback we want? 15:34:39 <linda> i.e. "is there any information you find superflouous?" 15:34:44 <iwakeh> I mean, we don't want to discuss colors and the like. 15:35:01 <linda> iwakeh: yes, I think that giving people a limited time with a set of questions might be helpful. 15:35:01 <karsten> good point. 15:35:17 <linda> And we can tailor it to what we want feedback on. 15:36:01 <karsten> can we ask them tomorrow and give them two weeks, until our next team meeting? 15:36:06 <iwakeh> Maybe, even ask: how fast can you find a term in the glossary? 15:36:08 <karsten> maybe with a reminder in a week from now? 15:36:44 <linda> I like the idea of sending it after the tor break, but I don't mind if it goes out now. 15:36:53 <karsten> tor break? 15:36:57 <iwakeh> ? 15:37:15 <iwakeh> x-mas break ;-) 15:37:36 <karsten> ah! can we still send it in december, though? 15:37:56 <karsten> (so that it can go into the monthly report, which would look quite sad otherwise.) 15:38:04 <linda> (oh, I see.) 15:38:09 <iwakeh> sure. 15:38:10 <hiro> ah yeah if it is sent this week some people might be out already 15:38:15 <karsten> right. 15:38:31 <hiro> maybe we should just give people more time to participate ? 15:38:32 <karsten> iwakeh, linda: is that something you'd want to prepare? 15:39:24 <linda> Sure. 15:39:40 <linda> I think it would be simple, like 3 (5 max) questions. 15:39:55 <linda> mine would be: 15:40:00 <linda> -does the grouping make sense 15:40:12 <linda> -do you find anything frustrating or confusing 15:40:22 <linda> - is the added information helpful 15:40:25 <linda> o_o 15:40:31 <iwakeh> that's it :-) 15:41:01 <karsten> cool! 15:41:33 <karsten> who sends this message after the break? 15:41:42 <linda> I can do it. 15:41:51 <karsten> perfect. thanks! 15:41:53 <iwakeh> fine. 15:41:56 <RaBe> perhaps we could ask the "how does it work on your mobile device" question, just to mention we're mobile now? :) 15:42:04 <iwakeh> true. 15:42:04 <karsten> hehe 15:42:11 <linda> RaBe: good idea! 15:42:25 <iwakeh> timing? 15:42:37 <karsten> so, our next meeting is on jan 5. 15:42:38 <iwakeh> remnder date and final date? 15:43:09 <linda> I can send it the monday of that week 15:43:14 <linda> and give people until the meeting. 15:43:17 <linda> Is 3 days too short? 15:43:22 <linda> *4 days 15:43:26 <karsten> that's not in december anymore. :( 15:43:26 <iwakeh> too short. 15:43:32 <linda> oh yeah 15:43:37 <linda> I can send it out tomorrow. 15:43:48 <iwakeh> we have all the questions already. 15:43:51 <linda> (or when the fixes have been made?) 15:43:51 <hiro> how about we send it this week but we give people till the first week of january to reply? 15:43:56 <iwakeh> that'll be fine. 15:43:58 <karsten> sounds good to me. 15:44:07 <linda> I would prefer to have the fixes done before we send it out though 15:44:09 <karsten> linda: I'll ping you when the fixes are in. 15:44:14 <linda> okay 15:44:37 <RaBe> karsten: i'll send my updates today 15:45:00 <karsten> RaBe: neat! 15:45:18 <linda> :D 15:45:21 <karsten> alright, quite a plan. 15:45:23 <karsten> blog post! 15:45:35 <karsten> we said last time we'd send linda our input for the blog post. 15:45:36 <linda> The outline of the blog post is that 15:45:40 <linda> I'll talk about 15:45:40 <linda> addressed pain points in the interface 15:45:40 <linda> added additional content to the website, and 15:45:40 <linda> gave it a style makeover. 15:45:54 <linda> oops, I copy and pasted 3 bullets. it looks odd on irc. 15:45:57 <iwakeh> and introduced mobile. 15:46:02 <linda> oh, right. 15:46:36 <linda> I guess I lumped that into the styling. 15:47:03 <linda> I planned to make a blog post addressing some of our usability fixes, the new pages, and how we made it mobile friendly and tor-brand-like. 15:47:04 <karsten> sounds good. in a way, you have been writing that blog post for weeks now, on the wiki. :) 15:47:21 <linda> Yeah. I think I'll just condense it a bit more. :D 15:47:26 <karsten> hehe 15:47:47 <linda> That's what I planned to talk about. I wanted to run it by you guys. 15:47:57 <karsten> sure! 15:48:07 <karsten> but this is planned for mid-january or later? 15:48:28 <linda> I was thinking that we could time it with the offical launch of metrics. 15:48:37 <karsten> good idea! 15:48:41 <iwakeh> yes. 15:48:42 <linda> So after we get the feedback, incorporate it, and then launch, I'll relese the post. 15:48:56 <karsten> good plan! 15:49:03 <linda> I'll have a draft for you by the next meeting. 15:49:04 <linda> :) 15:49:24 <karsten> okay, that's all for UX/website for today? 15:49:29 <linda> I think so. 15:50:01 <karsten> awesome! let's move on to the other topics then. 15:50:09 <karsten> dgoulet, asn: still around? 15:50:17 <asn> yes 15:50:22 <dgoulet> o/ 15:50:27 <karsten> - torperf weirdness (karsten) 15:50:45 <karsten> dgoulet: did you look at the graph I posted? 15:50:58 <karsten> https://people.torproject.org/~karsten/volatile/torperf-2016-12-19.pdf 15:51:07 <karsten> iwakeh: did you see it? 15:51:14 <iwakeh> yes. 15:51:30 <dgoulet> karsten: I did but as I said I didn't investiguate much 15:51:42 <dgoulet> asn: context: http://rougmnvswfsmd4dq.onion/torperf.html 15:51:51 <karsten> so, this doesn't look new. 15:51:52 <dgoulet> asn: torperf graph started to do some thigns :) 15:51:56 <karsten> it's been happening in all of 2016. 15:52:16 <karsten> just increasing, which is hardly visible in the graph, but which is how you found it in the torperf graph on metrics. 15:52:44 <karsten> I can provide the logs, if anybody wants to look. 15:52:47 <karsten> tor logs, that is. 15:53:18 <asn> oh wow 15:53:22 <asn> i see the spikes 15:54:13 <karsten> I just wonder, how urgent is this? 15:54:28 <karsten> it's certainly important. but we're running out of days before _the_break_. 15:54:34 <asn> so does this mean that the performance of the network is going worse and worse? 15:54:49 <asn> or could it be a bug in the stats? 15:54:56 <dgoulet> karsten: I don,t think it's that urgent... but I would really like to see the results of OnionPerf and if we have the same, some version of tor introduced something 15:55:15 <karsten> asn: https://people.torproject.org/~karsten/volatile/torperf-2016-12-19.pdf shows raw values, no stats. 15:55:29 <karsten> that is, no aggregation happening there. 15:56:05 <karsten> dgoulet: should I send you the logs? 15:56:12 <karsten> iwakeh: or did you want to investigate further? 15:56:25 <dgoulet> karsten: logs here it's tor logs? 15:56:29 <iwakeh> yes, I could try. 15:56:32 <dgoulet> karsten: arf... debug level or info? 15:56:33 <karsten> yes, tor logs. 15:57:01 <iwakeh> was there any change to the measument? 15:57:12 <iwakeh> this year? 15:57:34 <karsten> not really. but I only plotted 2016. 15:57:42 <karsten> it could have started in 2015, or 2014.. 15:57:45 <iwakeh> at least the 1MB improved in june/july for the bulk. 15:57:48 <asn> btw this pdf shows measurements being the same all along 2016. why do the metrics graphs change only in nov 2016? 15:58:22 <karsten> that's a threshold thing. 15:58:33 <karsten> suddenly, 3/4 of values are larger than some value. 15:59:19 <asn> oook. so the big values did not appear in "1st to 3rd quartile" until very recently? 15:59:58 <karsten> right. 16:00:45 <asn> interesting. 16:00:51 <iwakeh> It's caused by these short darker lines on karsten's plot. 16:00:51 <dgoulet> I'm not sure what to do about it lol 16:01:10 <asn> iwakeh: the one marked by (2) 16:01:21 <iwakeh> yes, roughly. 16:01:36 <iwakeh> also (3) 16:02:20 <karsten> Nov 07 08:53:32.000 [notice] Tor 0.2.5.0-alpha-dev (git-0a7b3e9645218bb8) opening log file. 16:02:36 <karsten> notice logs only. 16:02:47 <iwakeh> visually torperf is worse all year than moria, siv. 16:02:55 <asn> (what's siv, btw?) 16:03:01 <asn> onionperf logs might also be useful i guess. 16:03:02 <karsten> a machine. 16:03:03 <iwakeh> on karsten's plot. 16:03:12 <asn> dgoulet: there is not much to do apart from checking for bugs i guess. 16:03:19 <asn> dgoulet: but if performance is degrading so *much*, we should be aware of it 16:03:27 <asn> this looks like a very big and gradual degradation if it's real 16:03:35 <dgoulet> yeah I want OnionPerf results also about it 16:03:45 <karsten> I agree that we should look at onionperf results. 16:04:00 <karsten> anyway, let's not debug this now. 16:04:09 <karsten> I can provide more data if somebody wants it. 16:04:24 <dgoulet> cool 16:04:26 <karsten> but it could be an issue with our torperf instances. 16:04:29 <iwakeh> yes. 16:04:35 <karsten> which doesn't matter for other tor clients. 16:04:41 <karsten> alright. 16:04:46 <karsten> - onionperf deliverable for R (karsten) 16:04:56 <karsten> asn: so, my hope there is that dec 31 is a soft deadline. 16:04:58 <iwakeh> maybe, setup other measuring clients? 16:05:02 <asn> btw is this for deliverable or for our own gratification? 16:05:06 <karsten> iwakeh: yes, onionperf clients! 16:05:17 <iwakeh> yep 16:05:21 <karsten> asn: this == ? 16:05:34 <asn> this onionperf work 16:05:48 <asn> because dgoulet wrote a summary to the deliverable ticket, and perhaps we can count this as done? 16:05:53 <karsten> we have sponsor X deliverables for onionperf work. 16:05:55 <asn> and then work on this for our own sake? 16:06:01 <asn> wrt SponsorR I mean. sorry. 16:06:08 <karsten> I don't know. 16:06:14 <asn> dgoulet: ^ ? 16:06:47 <karsten> we didn't make much progress on our onionperf deliverables for sponsor X in 2016 though. 16:06:59 <karsten> but we still have until june 2017. 16:07:01 <dgoulet> asn: wait the summary I wrote is not Onionperf related 16:07:18 * asn can't find the ticket 16:07:27 <asn> ah #13209 16:07:42 <karsten> that's unrelated, right? 16:08:05 <dgoulet> OnionPerf is for R and *ideally* we would have a metric graph at some point, that I think is what we told R 16:08:05 <karsten> so, realistically, we cannot do much with onionperf in 2016 anymore. 16:08:12 <dgoulet> yeah I think it,S fine 16:08:33 <asn> roger told us about #13209 wrt the onionperf deliverable. 16:08:56 <dgoulet> no two different things 16:09:06 <asn> (btw the graphs that we were just discussing are for onion services ,or for general tor?) 16:09:12 <dgoulet> onionperf measures HS but health measurer it something else 16:09:21 <dgoulet> I believe HS for OnionPerf is the goal 16:09:26 <dgoulet> (for R) 16:09:50 <karsten> asn: the graphs we discussed earlier are non-hs-related. 16:09:55 <asn> ack 16:10:08 <karsten> I think rob is running onionperf for a while now. 16:10:14 <karsten> and might have data and graphs. 16:10:55 <asn> dgoulet: our SponsorR milestone is "Run OnionPerf more thoroughly. Maintain 16:10:56 <asn> the hidden service health measurer codebase." 16:11:05 <asn> so closing #13209 definitely handles the second part. 16:11:17 <karsten> "Let me know if February 28 is too late to get this done. In that case 16:11:17 <karsten> I'll see how we can be faster here. Otherwise, I'll let you (and 16:11:17 <karsten> accounting folks) know when we're done." 16:11:27 <karsten> is what I wrote 1 week ago, and then didn't hear back. 16:11:30 <dgoulet> asn: and the first one is already done by Rob but _ideally_ we have it run on tpo infrastructure with metrics graph, that would be the great end result :D 16:11:43 <dgoulet> karsten: and I think it's fine by then as we don't have a choice anyway :) 16:11:44 <asn> karsten: unfortunately, i don't know how soft the deadline is. 16:11:52 <asn> dgoulet: ok great. 16:12:01 <asn> dgoulet: i mean great wrt deadline being soft. 16:12:15 <karsten> okay. 16:12:30 <asn> dgoulet: now wrt running onionperf from tpo infrastructure, i agree it would be a great end result, but it can be lots of work I imagine. 16:12:34 <karsten> we need to move on. we're already 12 minutes over time! :) 16:12:40 <asn> sorry about this karsten ! 16:12:44 <asn> let 16:12:44 <asn> 's move on 16:12:50 <karsten> asn: well, we're planning to do something like that. 16:13:00 <karsten> 2017-02: Replace all existing Torperf instances gathering current Tor 16:13:00 <karsten> network performance measurements with OnionPerf instances. (Sponsor X 16:13:01 <karsten> 5.1. OnionPerf) 16:13:10 <hiro> dgoulet if i can help wrt running onionperf on tpo infr let me know 16:13:13 <karsten> and 2017-04: Develop and deploy at least one more user model in addition 16:13:14 <karsten> to the current model. (Sponsor X 5.2. OnionPerf) 16:13:22 <asn> nice 16:13:26 <dgoulet> hiro: neat! 16:13:28 <karsten> which can involve an onion service, for example. 16:13:42 <karsten> okay, moving on. 16:13:46 <karsten> - #20596: first input (iwakeh) 16:13:49 <karsten> iwakeh: still around? :) 16:13:53 <iwakeh> yes. 16:14:18 <iwakeh> as the comment says thing are more subtle than expected, but 16:14:24 <iwakeh> it's worth the effort. 16:14:42 <iwakeh> I need the filled repo to provide patches for the projects. 16:14:58 <karsten> okay, let me review that and push to the repo. 16:15:04 <karsten> might not happen tonight, but tomorrow. 16:15:11 <karsten> okay? 16:15:19 <iwakeh> (btw: I had to solve several captchas to upload that patch, grrr) 16:15:25 <karsten> ugh. 16:15:31 <iwakeh> well, 16:15:55 <iwakeh> i think it's kind of 16:16:04 <iwakeh> difficult to review half the work. 16:16:28 <iwakeh> Without the other half. 16:16:45 <karsten> hmm, how can I help then? 16:16:55 <iwakeh> So, maybe commit and then review once the other patches are in. 16:17:10 <karsten> ah, commit, so that there's a repository at all? 16:17:16 <iwakeh> The contents of metics-base won't be run alone. 16:17:21 <iwakeh> yes! 16:17:49 <karsten> okay, can do after this meeting. 16:17:52 <iwakeh> of course, the metrics_checks.xml can be reviewed. 16:18:09 <iwakeh> I spent more time solidifying the idea. 16:18:33 <karsten> I really like the idea! 16:18:39 <iwakeh> :-) 16:18:50 <karsten> having several almost-the-same versions of those checks was... disturbing. ;) 16:18:59 <iwakeh> yes. 16:19:06 <karsten> alright. I'll push in a moment. 16:19:10 <karsten> anything else for this meeting? 16:19:14 <iwakeh> It will result in a checkstyle task for metrics-lib tests. 16:19:22 <iwakeh> no other topic. 16:19:29 <iwakeh> a maybe timing. 16:19:36 <karsten> ah, we didn't have style checks there yet?.... 16:19:45 <iwakeh> or 'the break'? 16:19:49 <karsten> right. 16:19:52 <karsten> no meeting next week, right? 16:20:09 <iwakeh> right, but I might want to do the 16:20:22 <iwakeh> patches for the streamlining this year. 16:20:30 <karsten> okay. 16:20:34 <iwakeh> as there will be a metrics-lib release in january. 16:20:45 <karsten> .oO( monthly report! ) 16:20:48 <iwakeh> another topic for the report. 16:20:54 <iwakeh> yes :-) 16:21:03 <karsten> how about I read trac email, in particular for that ticket? 16:21:09 <karsten> and we talk on trac? 16:21:16 <iwakeh> should be sufficient. 16:21:21 <karsten> great! 16:21:25 <iwakeh> fine. 16:21:38 <iwakeh> 27min late. 16:21:44 <karsten> okay then, let's end this meeting. next one starts in 8 minutes. :) 16:21:59 <karsten> are you sure your clock is correct? :) 16:22:05 <iwakeh> ok, back to work :-) 16:22:07 <karsten> anyway, thanks, everyone! bye! 16:22:13 <karsten> #endmeeting