13:59:41 #startmeeting metrics team 13:59:41 Meeting started Thu May 5 13:59:41 2016 UTC. The chair is karsten. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:59:41 Useful Commands: #action #agreed #help #info #idea #link #topic. 14:00:15 I guess this will be a smaller meeting, with oma being on vacation and Letty possibly distracted by other things as I hear. 14:00:32 but that could mean we'll have more time per topic. 14:00:39 https://pad.riseup.net/p/zUNzEIFRq5S4 is the meeting pad. 14:00:44 hi iwakeh! 14:00:45 hi there! 14:01:04 please add topics to that pad. 14:01:35 or remove topics if they contain your name and don't make much sense to you. this was just me making some suggestions. 14:04:04 is virgil around? 14:04:24 if not, I wonder if I should just start with my topic, and then we have the rest of the meeting for collector and team documentation. 14:04:39 that'll be fine 14:04:59 ok! 14:05:00 * Discontinuing Globe (karsten) 14:05:25 the latest update there is that we're trying to find a time for a meeting to go through trac tickets. 14:05:36 globe and atlas tickets, that is. 14:05:49 ah, who we? 14:06:13 and then we decide which of them we want to keep and reassign to atlas, which of them need to be completed before shutting down globe, and which ones we can close. 14:06:16 https://lists.torproject.org/pipermail/tor-project/2016-April/000299.html 14:06:48 we = phw, isis, and I, but whoever else wants to attend is welcome to come. 14:07:14 my original plan was to meet in 1 week from now using this time slot, but I think phw cannot make that. 14:07:26 so, I might start a doodle later today or tomorrow. 14:08:24 after that meeting we'll need to implement the remaining tickets. 14:08:39 hopefully with help of friendly volunteers. 14:09:08 I don't expect that to be many tickets. right now there are 3. there shouldn't be more than twice as many. 14:09:11 I hope. 14:09:33 okay, that's all I wanted to say here. 14:09:48 happy to move on if there are no questions. 14:10:16 ok! 14:10:18 * Team documentation wiki page (iwakeh) 14:10:35 yes, we finished the first collection 14:10:42 and got a pretty good 14:10:51 selection of questions. 14:10:56 agreed! 14:11:00 These need to be answered now. 14:11:03 And, 14:11:26 as a next step I would like to names 14:11:42 of people who anser specific topics 14:12:20 where is the wiki page again? ... looking. 14:12:45 https://trac.torproject.org/projects/tor/wiki/org/teams/MetricsTeam/Documentation 14:12:54 maybe, we should link it soon to the team page :-) 14:13:06 right. can you do that? 14:13:22 yes, I'll look for a good place. 14:13:53 i think 14:14:08 operation and coding answers 14:14:20 can be provided from the guide docs. 14:14:37 open are the team and funding etc questions. 14:15:29 I guess I should try to answer them. or rather, write a first draft of an answer that we can then improve. 14:15:41 good! 14:15:56 that's a start. 14:16:02 isabela: I wonder if you'd enjoy answering the funding questions together with me? 14:16:11 https://trac.torproject.org/projects/tor/wiki/org/teams/MetricsTeam/Documentation#Funding 14:16:57 iwakeh: okay, so we put my name next to "About the Metrics-Team", "Road-map, ...", and "Funding", 14:17:04 karsten: yes 14:17:11 i love this faq :) 14:17:23 will pass this over to other teams to do something like that 14:17:28 isabela: perfect! much appreciated. 14:18:02 iwakeh: and yours next to "Contribution and Operation"? or what was the plan? 14:18:07 yes, 14:18:14 and guide documents. 14:18:25 right. sounds great! 14:18:41 I'll add this to the doc. 14:18:46 later. 14:18:59 thanks! and where do we write answers? below questions in the same document? 14:19:12 I thought about that too 14:19:15 maybe, 14:19:25 we could also use a pad. 14:19:32 and copy results back to the wiki later. 14:19:38 rather a wiki page. 14:19:51 works for me. 14:19:54 maybe, have one wiki page 14:20:08 per person answering and/or topic 14:20:18 I don't 14:20:29 know if trac has concurrency issues ;-) 14:21:07 in case we edit at the same time. 14:21:16 hmm, what we can do to avoid those is pre-write answers in a local tool and copy them over to the wiki when we're done. 14:21:32 if we divide the work by topic we shouldn't run into major issues. 14:21:47 that would work, too. 14:21:55 so, same doc? 14:22:02 btw, always keep a copy of your edits locally before submitting. 14:22:06 sure, same doc is fine. 14:22:13 should we have a length limit? 14:22:15 we had cases of trac eating your changes. 14:22:22 and that hurts. 14:22:31 huh, probably. 14:22:35 100 words, or what's a good limit? 14:22:50 oh, that's a lot. 14:23:03 or? 14:23:06 "What skills are required for participating? What tools are used? What languages are supported (maybe even in addition to the languages used in a project)? " 14:23:09 that's 25 words. 14:23:31 ok, try 100 words. 14:23:33 between 50 and 100, where 50 is better? 14:23:39 yes. 14:23:44 ok. 14:24:12 so, that's it. 14:24:20 great! 14:24:29 moving on.. 14:24:33 * another CollecTor instance (iwakeh) 14:24:38 yep, 14:24:59 I'll soon have access to a server for setting up 14:25:07 another CollecTor instance. 14:25:20 I would like to use it for the first 14:25:47 non-cron-job instance and while setting 14:25:57 it up rewrite the operator guide. 14:26:03 neat! 14:26:11 this means 14:26:11 is the non-cron-job part already written? 14:26:39 first priority is the timed-java changes. 14:26:57 ok. please let me know which tickets that are, and I'll prioritize those. 14:27:12 well, I'll need to write these. 14:27:29 ah, right, I meant prioritize review of those. 14:27:36 ok. 14:27:57 that sort of leads seemlessly 14:28:03 into the next topic 14:28:04 heh 14:28:09 * CollecTor improvement project (iwakeh) 14:28:31 we implemented quite a few things already. 14:29:06 the next steps need prioritization 14:29:23 there is the analysis of 14:29:29 missing descriptors. 14:29:54 the design and implementation and the guide docs. 14:30:41 okay, let's see. missing descriptors analysis is half way done, right? 14:30:49 right. 14:30:53 with the analysis of missing consensuses/votes/etc. not done yet. 14:31:10 though we'll get that almost for free once we have synchronization between collector instances implemented and deployed. 14:31:21 because then we'll see how many descriptors have 1 vs. 2 @source tags. 14:31:43 so, sync-tasks first? 14:31:49 well, assuming that's the design that will be implemented. 14:31:52 together with non-cron? 14:32:01 non-cron seems most pressing. 14:32:06 so that you can set up your instance. 14:32:11 yep. 14:32:26 I'll tackle that 14:32:38 first and at he same time make a better list 14:32:47 for prioritization. 14:33:00 the sync part is important, too, but I think we can also sync offline by copying over tarballs and import them. 14:33:15 I still have the feeling that the analysis 14:33:16 that means, you could set up your instance now, and we sync descriptors in a few weeks. 14:33:31 the descriptors you collect now. 14:34:36 (that the analysis... ?) 14:34:48 will 14:34:59 point to other places 14:35:09 that need care. or 14:35:17 might need improvement? 14:35:20 ah, I think so. 14:35:44 maybe, sync in parallel to second 14:35:46 but if you set up your instance now and start collecting, even if we're missing descriptors, we have a backup. 14:35:57 wrong. non-cron in 14:36:17 I'm also thinking about importing tjr's descriptors from the past months once we have sync in place. 14:36:18 in parallel to second analysis. 14:36:44 (tjr is running an older version of collector since, what, a year or so, and we never exchanged data.) 14:36:58 ok, interesting. 14:37:11 summary: 14:37:22 non-cron for third CollecTor 14:37:33 goes first. 14:37:39 yep 14:37:56 I think the other 14:38:05 task will fall in place once 14:38:14 I'll start on that. 14:38:33 keeping notes for the guide while you're setting things up sounds like a good plan, too. 14:38:42 yes. 14:39:00 shouldn't we update the roadmap? 14:39:11 the team roadmap? 14:39:28 just noticed that it's outdated 14:39:31 partially. 14:40:04 I noticed, too, but I knew that some people would be missing today, so I didn't put it on the agenda. but let's update collector-related tasks if you want? 14:40:29 or before the next meeting would be fine. 14:40:30 should I copy it over to the pad? 14:40:34 sure, or that. 14:40:54 I'll put it on the next agenda. 14:40:58 ok. 14:41:25 well, that's all. 14:41:29 okay, what else can I do than seemingly randomly go through open tickets and respond? 14:41:44 oh, that's fine. 14:41:58 I'll make myself heard when 14:42:04 necessary ;-) 14:42:04 can you maybe set priority to something above medium if you need a response faster? 14:42:09 that's good! 14:42:18 ok, good point 14:42:37 not that we'd use the priority field consistently, but we could start here. 14:42:57 okay, great. anything else to discuss today? 14:43:14 All said. 14:43:21 perfect! 14:43:45 next meeting in 2 weeks, and doodle for the globe meeting some time next week. 14:43:53 fine. 14:44:21 great, let's talk more on tickets! ttyl 14:44:33 #endmeeting