[16:11:38] dsaez: o/ [16:31:25] hi leila [16:31:58] hi dsaez. [16:32:13] dsaez: I don't know how to respond to a proposed change in meeting time. do you know? [16:32:14] : [16:32:15] :) [16:32:37] eeh... accept ? [16:33:57] dsaez: ;) [16:34:06] dsaez: I have to discuss it with others. I wonder if there is a way for that? [16:34:22] dsaez: or that has to happen on a different thread and the result be applied to the suggestion? [16:34:22] no idea, did you receive an email? [16:34:46] dsaez: I received an email which is in the form of an update to the event and says what you wanted to say. :) [16:35:14] dsaez: but that email has only me as a received so I can't discuss it with others directly. Okay. worry not. At least there is no obvious way you know. ;) [16:40:11] To be honest I have no idea, this what google offers when you are not allowed to move a meeting [16:51:34] leila, are collaboration person-based? I thought that was by institution [16:52:17] dsaez: no. they are per person. [16:52:57] dsaez: check https://www.mediawiki.org/wiki/Wikimedia_Research/Formal_collaborations#Current_collaborations . For every person who is a formal collaborator, we will need specific info. [16:52:59] and process [16:57:29] I see, sounds painful [17:23:15] ciao bmansurov! Question: what is the best way to test that dom_interactive_time works correctly? [17:23:17] Thanks :D [17:24:57] miriam: o/ set a timer right before loading the page, once the page becomes usable, stop the timer. Then compare its value with domInteractiveTime. [17:26:07] dsaez: did you create that phab task for dataset competition? if not, no worries, i'll just go ahead and do it [17:26:22] thanks bmansurov! Another quesiton sorry: did we remove the citation identifier field? [17:26:26] for this round? [17:26:31] miriam: yes [17:26:40] ok bmansurov, sorry I missed that :) [17:26:48] isaacj, not yet, super bussy monday :S [17:26:55] sorry for that [17:27:00] no problemo - i'll create it and subscribe you to it [17:27:31] miriam: T212937#4906463 [17:27:32] T212937: Citation Usage instrumentation issues - https://phabricator.wikimedia.org/T212937 [17:28:03] https://phabricator.wikimedia.org/T212937#4906463 [17:28:24] stashbot needs to handle hash values!!! [17:28:24] See https://wikitech.wikimedia.org/wiki/Tool:Stashbot for help. [17:28:58] oki, great, thanks! [17:29:08] bmansurov, thanks [17:29:57] poor stashbot :) doing a great job! [17:30:23] not so great, apparently ;) [18:15:31] thanks isaacj! [18:16:10] welcome! [18:41:34] 10Quarry: Show query run date above outputs section - https://phabricator.wikimedia.org/T215831 (10Framawiki) [18:41:40] 10Quarry: Show query run date above outputs section - https://phabricator.wikimedia.org/T215831 (10Framawiki) p:05Triage→03High [19:29:22] isaacj: I have a few min delay. sorry. Let's start at 35 [19:29:31] leila: sounds good [19:53:58] https://xkcd.com/2110/ [20:58:24] haha guillom: unfortunately no simple english version of https://en.wikipedia.org/wiki/Bayesian_inference [22:47:21] leila: regarding DNT [22:48:54] nuria: in a meeting now. will read later. [23:01:49] leila: k let me know when you are out [23:08:39] leila: as far as i know, webrequest does not have any info about dnt , let me know if you were thinking about a different table [23:21:17] nuria: the info gets captured as part of the varnish. Check T127571#2047300 for example (I will close that task shortly, side-note). It's captured as part of x_forwarded_for or x_analytics I think. need to double-check. [23:21:17] T127571: Percentage of users with DNT on - https://phabricator.wikimedia.org/T127571 [23:23:20] leila: it is on varnish, yes [23:24:18] leila: but we never populated that on webrequest headers, let me recheck varnish export [23:26:06] leila: let me find varnish code that does that , we have an outstanding request for dnt headers on webrequest , i think we never got to do it [23:29:08] nuria: I think the task you have in mind is T98831 ? [23:29:08] T98831: Honor DNT header for access logs & varnish logs - https://phabricator.wikimedia.org/T98831 [23:30:41] leila: ah ya, that is one, see: (I think this is the right line) https://github.com/wikimedia/puppet/blob/production/modules/varnishkafka/manifests/instance.pp#L113 [23:31:59] leila: the dnt header -as far as i know - was never moved from varnish-> varnishkafka -> webrequest [23:32:34] leila: we recently added accept headers to that line (per services team request) but that's teh only modification i can think that happen as of recent [23:33:14] *the [23:34:18] leila: let me triple check template [23:36:06] nuria: I may be blanking. checking on my end, too. [23:36:31] leila: ya, this is the other place where thsi is at: https://github.com/wikimedia/puppet/blob/production/modules/profile/manifests/cache/kafka/webrequest.pp#L141 [23:38:01] leila: so, confirming, dnt is not there unless there was a mixxup i do not know of [23:39:44] leila: DNT was honored on eventlogging since a change ori did back in the pleistocenic period [23:42:12] nuria: yup, re eventlogging. for webrequest logs, my bad. we may have read it in the past using varnish and that's what I remembered. [23:43:14] leila: the work that ori did was a looonggg time ago to estimate it, it took 15 minutes and a raw dump of varnish requests, we never used it other than to give bulk estimates [23:43:48] nuria: we on our end did. :) [23:44:06] nuria: outside of the estimation work by ori. but again, you're right. it's not part of webrequest logs. [23:45:00] leila: ok, nor have we been able to use DNT to discard user data [23:45:12] Research folks: please pardon the noise while I'm closing some open phab tasks that are resolved. [23:45:29] nuria: who is we? :) [23:45:32] ;) [23:45:43] leila: we -> the data users [23:46:23] leila: meaning that there is no way in our systems to " analyze on data excluding from users with DNT " [23:47:10] nuria: in EL is possible, and you already mentioned that. [23:47:23] leila: no, in EL the data never gets to us [23:47:47] leila: so we do not get it if DNT is on, [23:47:58] leila: javascript prevents teh data from being sent [23:48:11] leila: makes sense? [23:48:23] nuria: yes. you're right. basically it never makes it to EL.