[05:36:07] https://en.wikipedia.org/wiki/Wikipedia:Village_pump_%28technical%29#Extracting_PMIDs ""Do you know if it is possible to extract [all] PubMed ID (PMID) or PMCIDs from Wiki references? Furthermore, could you dump those IDs out into a list for analysis?"" [05:36:16] I assume someone in here will know. :) [15:17:02] * Ironholds yawns [15:17:03] mornings [15:20:24] oh holy shit [15:20:26] sweet NIBBLETS [15:20:31] halfak, halfak! [15:20:41] R6 classes allow for variable storage within the class definition! [15:20:43] Hey Ironholds [15:20:45] This is the best day! [15:20:51] God bless Winston Chang. [15:21:03] :P R's finally catching up 'eh? [15:21:15] yup! It's like we've hit the eighties! [15:21:27] 80s is pretty recent for programming languages. [15:21:58] "Python was conceived in the late 1980s[23] and its implementation was started in December 1989[24]" [15:22:00] https://en.wikipedia.org/wiki/Python_(programming_language) [15:23:54] sounds about right [15:24:01] on that note, I finally discovered the origins of <- [15:24:03] it's an APL hangover. [15:24:12] The predecessor of the predecessor of R. [15:24:22] * Ironholds shakes head [15:34:48] halfak, would NavigationTiming_10076863 only contain completed requests or do I need to do filtering if I want to get that out of it? [15:35:26] Ironholds, not sure. I've not made much use of that particular schema. When I did the module storage work, we used a different schema [15:35:33] * Ironholds nods [15:35:35] thankee! [15:35:42] o/ [15:36:04] what are you using for the HHVM testing? [15:36:06] hey Nettrom :) [15:36:18] hey Ironholds :) [15:37:04] This morning has Fun With R6 Classes. You'd enjoy them, I think. [15:37:20] although this is reminding me why OO always gives me a headache; making the call as tow hether things need to be distinct methods. [16:32:56] halfak, etc, can't make standup. Hangouts has decided to crash again :/ [16:33:55] * Ironholds tries to rejigger [17:11:00] DarTar, halfak, Ironholds: how do you like Research-and-Data for our Phabricator project name? [17:11:07] the name can't have spaces [17:11:15] R&D! [17:11:23] I have no opinion; phabricator confuses the hell out of me anyhoo ;p [17:11:24] * halfak might have already made one [17:11:26] * halfak checks [17:11:31] and they suggest using dashes to separate and also have it easily searchable [17:11:39] Nope. i didn't [17:11:41] (sooo, R&D may not be easily found?) [17:11:51] yeah [17:13:35] Ironholds, don't worry about that part now. We just want to have a placeholder there until the ScrumMaster arrives. Then, he/she can take care of things on Phabricator. [17:14:13] uhm. [17:14:28] if we're talking about it as a trello replacement, it's absolutely going to be all of ours' problem ;p [17:14:49] I don't think I've ever worked on a team where the expectation is that the SM will do all the completion/timekeeping/tasking notation, just lead it. [17:15:12] it will be a Trello replacement eventually, everyone is moving to Phabricator, so we should move sooner or later. [17:15:50] oh, agreed [17:15:57] I just disagree with your assurance about not having to worry ;p [17:16:07] I'm saying you don't have to worry right now [17:16:08] :D [17:16:18] in the long run, you should always worry. :p [17:17:25] thank you for that ;p [17:33:00] hokay! Research-and-Data it is. [18:06:49] leila, I just registered "halfak" on phab. Can you add me as an admin for the project? [18:06:58] adding you in a sec [18:07:17] you're in halfak [18:07:22] thanks! [18:07:25] np1 [18:07:27] ! [18:29:24] tnegrin, was it August 2014 with the hinky PV counts? [18:29:55] with the robots? [18:29:58] yup [18:30:11] yeah -- but it's pretty linear -- I wouldn't call it hinky [18:31:29] gotcha [18:49:07] halfak, I see that you're living the Phab. [18:49:24] Totesally phabulous :D [18:49:30] :D [18:55:41] say, halfak, DarTar: did we ever come up with any agreement on analytics ethical data usage guidelines? [18:55:53] I have a request I find frankly creepy that's juuust legally permissible but that makes me want to forkbomb stat2. [18:57:03] Hmm... That's a broad question. [18:57:10] I think we have come to many agreements. [18:58:10] fair point [18:58:20] I will start a thread if the request is explicitly invoked [19:08:29] I just got a data request email with this: "We are looking for data to show that retention of new editors is not happening. We've seen older articles alluding to this statement but we're interested in more recent data regarding these statements." [19:09:05] My response: Here's a metric shit ton of data showing that retention is dropping. Shall we talk about next steps? [19:15:33] halfak: source? [19:15:44] :P [19:15:58] I mean, source of the request [19:17:04] Oh! https://twitter.com/linda_setch [19:17:15] I haven't figured out her institution. [19:17:26] ha [19:17:32] tnegrin, you've got mail [19:17:34] Looks like it is a class project. [19:17:37] she’s a strategist [19:17:41] also, this problem is why I hate stacked bar charts so much [19:17:45] they're a pie chart by any other name [19:18:03] But they are immensely better than pie charts. [19:18:13] Ironholds: FYI I pitched the cube data to folks in mobile, fasten your seat belts ;) [19:18:14] they'd struggle to be quite as bad, it's true [19:18:24] but they're still visually awkward and they make it hard to actually estimate deltas [19:18:25] ha ha -- another q -- why is the number so high? [19:18:32] halfak: unrelated – do you have any experience/examples of EL schemas logging float data? [19:18:38] now /that/ I'm digging into! [19:18:52] "float" data? [19:18:54] ta [19:19:12] context: we’re planning on measuring offsets expressed in multiples of viewports as a unit [19:19:28] for reasons that I’ll dig into this Thursday at RG [19:20:03] and I wonder if we have best practices in EL about storing numerical values with decimals at a given precision [19:20:21] I guess I should know since I was the de facto PM for EL *cough* [19:20:38] brb [19:20:39] Gotcha. Yeah, I haven't put a float into an EL schema yet. [19:21:06] kk I’ll poke Ori or Matt [19:24:24] halfak: “Do you wiki” 8-o https://twitter.com/linda_setch/status/523114860629229568 [19:24:46] * DarTar goes off to get lunch [19:39:18] * halfak continues to be berated for improperly using phab by quim [19:41:49] * Ironholds blinks [19:41:58] well, I guess we can not experiment with it and thus not know how to use it. [19:42:06] or we could not experiment with it and instead use Trello [19:42:16] but I suspect neither outcome there would meet his goals, so. [20:34:36] halfak, ping [20:34:44] Hey dude. [20:34:58] staff yo [20:36:09] halfak, staff meeting y'all! [20:36:18] Woops [20:37:02] Having issues. [20:37:34] * halfak reboots browser [20:38:04] "There is a problem connecting to this video call. Try again in a few minutes." [20:38:26] * halfak boots up firefox [21:52:46] Ironholds, are the 301 "undefined" requests all followed up with a 200 "Undefined" request? [21:53:15] that is, the HTTPS ones? [21:53:28] Sure. Whatever. Any of them. [21:53:35] * Ironholds thinks [21:53:52] yes! Except for the sampled log data, which doesn't get any data from our SSL machines directly: there it'd just have 200s. [21:54:08] Ahh. so you have checked? [21:55:12] not directly? This is based on a convo I had with Ottomata/Yuvi about how our redirection scheme displays [21:55:19] I can if you'd like! Investigating a theory? [21:56:06] Well.. I was imagining that, if the request is coming from JS, then it might not be following the redirect -- and since you filter out non-200 requests, it might squash the issue. [21:56:13] ooh [21:56:22] oh! You mean, it wouldn't have a 200 at all? [21:57:09] Bah. Nevermind. It looks like the spec tells XMLHTTPRequest to transparently follow redirects. [21:57:16] Nevermind Ironholds, not worth checking. [21:57:39] ahh [21:57:41] ah well :( [22:03:02] leila: yt? [22:03:09] yes, DarTar [22:03:21] let's meet whenever you and tnegrin have time [22:03:21] toby and I are looking for you [22:03:32] mmm [22:26:01] I'm an engineer from Sweden doing some research together with the Swedish chapter of WMF. I've run a few queries with the database at quarry.wmflabs.org, and I now need to run some bigger queries that are timing out. I talked to Jan Ainali at WMF Sweden and he said you might be able to help me here? [22:30:36] Hi Icebear_! [22:30:53] So, you have some good options. The best one is to get set up with a Wikimedia Labs account. [22:30:57] How is your UNIX? [22:34:59] Ironholds, halfak, one of you might be able to help ocaasi with https://en.wikipedia.org/wiki/Wikipedia:Village_pump_%28technical%29#Extracting_PMIDs (or know where to point him) [22:35:34] Thanks quiddity. Will respond. [22:48:19] Hmm.. My UNIX is not the best I guess, I can manage basic things. What do I need to do? [23:55:20] Hey Icebear_ sorry to walk away. Are you still around?