[07:16:03] * Ironholds whistles as he works [16:33:55] Standup? [16:34:05] Ironholds [16:34:10] DarTar... [16:34:13] Leila... [16:36:51] Hallo? [17:23:48] yuvipanda, ping [17:24:02] the UUIDs on apps, what are the min/max values for how long it can be? [17:24:16] also, new schema: when someone does something fingerprinty they have to document it. [17:24:21] *scheme [17:29:41] Ironholds: they're just UUIDs [17:29:55] https://en.wikipedia.org/wiki/Universally_unique_identifier [17:30:59] V1? 2? 3? ;p [17:34:43] Ironholds: does it matter? :P [17:34:56] Ironholds: http://docs.oracle.com/javase/7/docs/api/java/util/UUID.html [17:35:18] > Static factory to retrieve a type 4 (pseudo randomly generated) UUID. The UUID is generated using a cryptographically strong pseudo random number generator. [17:35:21] Ironholds: so type 54 [17:35:22] err [17:35:22] 4 [17:36:00] ta [17:36:08] oh joy, there are apparently URLs with multiple ones [17:36:28] multiple whats? [17:37:06] UUIDs [17:37:09] that is, a URL query path with multiple things prefaced by appInstallID [17:37:21] seriously, can you guys write this stuff down? [17:44:00] can you give me an example? [17:44:05] that sounds like a bug. [17:44:21] Ironholds: you should also bug the new android tech lead for these things! dbrant on #wikimedia-mobile [17:44:52] I will, and I'll just work around it [17:44:52] in reverse order [17:45:08] :) ok [17:46:49] Hey Ironholds, hows sshing to stat1003 the past couple of hours? [17:46:55] I'm having issues. [17:47:01] Issues with bast1001 too. [17:47:05] Wondering if it is me. [17:47:14] OH THANK GOD [17:47:29] halfak, lemme guess, really slow handshake on the initial connect, and then total lack of responsiveness to the keyboard? [17:47:38] Yup. Period drops [17:47:42] *periodc [17:47:43] i [17:47:44] like, weirdly variable, going from "slightly bad" to "I've been sitting here for five minutes and nothing is happening to what I typed"? [17:47:54] yup [17:48:06] it's on stat1002 too and it's frustrated me so much I was about to phone comcast. Voluntarily, I was going to talk to comcast. [17:48:10] thought it was my tubes so I rebooted everything. [17:48:20] But yeah. only getting worse. [17:48:29] who do we shout at about this? [17:48:41] Good Q. We could bother otto and see how that goes. [17:48:49] do I just announce very loudly "I think I should join the largest tables we have. Nobody will miss a production slave, right?" and then DevOps appear behind me? [17:49:00] that works too but is less fun [17:55:35] ping the ops list [17:55:36] Ironholds, did you guys talk about inviting the PLoS guys to talk at the last R&D meeting? [17:56:56] halfak, I don't think so. When were we meant to? [17:57:07] oh, sorry, misread [17:57:15] I don't recall that but my internet was patchy for a good two weeks [17:57:20] the new place has 100mb/s down. woooo. [17:57:21] Not sure. It looks like DarTar has them scheduled. [17:57:41] Sounds OK to me. Just seems like this is something we should have discussed. [17:58:30] yeah [18:00:35] okay, I am popping out for five minutes to go buy some plates so I have something to eat lunch off [18:00:39] ##newplaceproblems [18:01:25] analytics showcase starting [18:01:45] yuvipanda, any chance there was a spike in signups to quarry last week? [18:01:58] halfak: I could find out, but I monitor recent queries and no new names... [18:02:02] except theDJ [18:02:07] but he's not really new [18:02:10] halfak, yuvipanda ^^ [18:02:17] DarTar, see scrollback for discussion of PLoS @ showcase. [18:02:19] DarTar: oh? maybe I could pop in. [18:02:25] do you guys have a link? [18:02:49] https://plus.google.com/hangouts/_/calendar/d2lraW1lZGlhLm9yZ19jYjM3bXU0OGNuaHRkN2hybmE4czI3b25hb0Bncm91cC5jYWxlbmRhci5nb29nbGUuY29t.97lu2jl487aan6llad1s2no234 [18:03:33] DarTar: invite me? can't join [18:04:24] kevin just sent an invite [18:04:49] yuvipanda: ^ [18:05:00] DarTar: halfak ah, yeah, am there [18:05:01] ty [18:42:37] halfak, IH-away: ellery will join us for staff at 2.30 PT [18:43:29] I’ll give him a round of intros in the office over lunch and make sure he’s invited to all the relevant research/analytics events [18:43:43] Sounds good. [18:43:47] Is he on IRC yet? [18:43:52] He's not real until he's on IRC. [18:43:52] no, not yet [18:44:08] he’s with HR/IT at the moment [18:45:09] is this a new researcher? [18:45:25] Yup [18:45:34] yuvipanda: yes, working primarily with fundraising [18:45:35] He'll be working with Fundraising to start off. [18:45:40] :P [18:45:40] aaah nice [18:48:32] wow, I had no idea about the ongoing debate about notifications, halfak: you should check it out (see wikitech + ee lists) [18:50:09] halfak: btw, do we have any research, etc on blocks? [18:50:22] about 200-300 people who created accounts in the app are being blocked on enwiki per day, apparently [18:55:27] DarTar, cool! [18:58:01] DarTar, 262 days [18:58:05] in 10 hous [18:58:07] *r [18:58:10] can I code or can I code? :D [18:58:20] Ironholds, ^ see yuvipanda's question about blocks above. [18:58:35] yuvipanda, yes, yes we do! [18:58:35] DarTar, yeah saw that conversation. I haven't had the time to engage though. [18:59:01] Oh [18:59:04] Link? [18:59:07] yuvipanda, have you tried just...hand-coding some of the blocks? [18:59:15] (I've done some work with blocks cross-language too, but Ironholds has done more than me) [18:59:27] It's not really documented. But I did a project that categorised blocks based on regexes over their rationale, and it came up with some interesting stuff [18:59:32] I was going to do that next actually [18:59:37] Oooooh [18:59:55] And? [18:59:57] I'd suggest starting there then. I mean, it should be a relatively simple SELECT over ipblocks or logging [19:00:09] oh. Vandalism blocks have been going down since 2009 almost continuously [19:00:14] with a direct correlation to abusefilter hits [19:00:24] which suggests that the abusefilters are doing their job and relieving kind of a lot of work [19:00:41] I wanted to make it sciencey and write it up but never got around to it because some assholes hired me to do research full time and gave me interesting datasets. [19:04:09] yuvipanda: sorry, I was AFK – I’m not surprised about blocks (considering the crazy signup rate) [19:04:42] Heh [19:04:46] but boy, we need to have these numbers generated by Vital Signs [19:04:51] Hehe [19:04:57] Indeed indeed [19:05:49] there’s no way for person X to be aware of a spike on Y on project Z from source site/method W unless we easily produce this data [19:07:24] curation/governance/moderation signals is something I am hoping we can officially tackle in the next quarter [19:15:40] There was yet again a webstatscollector issue. [19:15:52] I do not want to spam the public analytics list, [19:16:01] but then ... webstatscollector data gets used widely. [19:16:32] What do you think ... are such announcements still on-topic there, or should I ask people [19:16:49] that are interested in such announcements to follow webstatscollector's wikipage? [19:17:00] qchris, I think it would be publicly interesting [19:17:06] I think we should announce it on the list as we've been doing [19:17:17] as long as you specify "webstatscollector, which generates the data at stats.grok.se" to put it in context for lay followers [19:17:22] Ok. Glad to hear that. Then I'll spam along :-D [20:06:25] okay [20:06:29] when we implement UUIDs [20:06:33] they are going to be in their own field [20:06:34] that is all [21:27:23] Ironholds: I saw it, thank you [21:30:08] Ironholds, halfak: service announcement, tnegrin is stuck in a conversation with terry, allow a few minutes before we start [21:30:14] kk [21:30:15] prodding him now [21:30:19] getting chayed [21:30:55] Ironholds: have you seen the playing cards with terry's face on them? [21:31:58] Ironholds: https://fbcdn-sphotos-g-a.akamaihd.net/hphotos-ak-xfa1/t1.0-9/p417x417/10615568_1545929262302196_4815564426308172737_n.jpg [21:36:57] TerrifiedPanda, yep! [22:45:46] Ironholds: yt? [23:02:51] TerrifiedPanda, wanna collab on something?