[00:01:08] (03CR) 10MaxSem: [C: 032 V: 032] Update ConfirmEdit to wmf/1.23wmf14 tip [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/114648 (owner: 10MaxSem) [00:01:25] (03CR) 10TTO: "> Why does it have to go in the "other" section instead of in the regular iwm?" [extensions/WikimediaMaintenance] - 10https://gerrit.wikimedia.org/r/112920 (owner: 10Gerrit Patch Uploader) [00:01:33] (03CR) 10MaxSem: [C: 032 V: 032] Update MobileFrontend to wmf/1.23wmf15 tip [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/114646 (owner: 10MaxSem) [00:02:08] (03PS22) 10Inchikutty: Embedding leaflet maps based on OSM data to UploadWizard [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/109263 [00:02:11] (03CR) 10MaxSem: [C: 032 V: 032] Make ApiCreateAccount return camelcase statuses [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/114483 (owner: 10Yuvipanda) [00:04:58] (03PS1) 10Krinkle: Only run *-ruby1.9.3lint jobs when commit touches .rb files [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/114662 [00:05:12] (03CR) 10Krinkle: [C: 032] Only run *-ruby1.9.3lint jobs when commit touches .rb files [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/114662 (owner: 10Krinkle) [00:05:17] (03Merged) 10jenkins-bot: Only run *-ruby1.9.3lint jobs when commit touches .rb files [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/114662 (owner: 10Krinkle) [00:06:15] (03PS1) 10Krinkle: Revert "Enable VisualEditor-jsduck" [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/114663 [00:06:27] (03CR) 10Krinkle: [C: 032] Revert "Enable VisualEditor-jsduck" [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/114663 (owner: 10Krinkle) [00:09:39] (03PS2) 10Krinkle: Revert "Enable VisualEditor-jsduck" [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/114663 [00:09:57] (03Abandoned) 10Krinkle: Revert "Enable VisualEditor-jsduck" [integration/zuul-config] - 10https://gerrit.wikimedia.org/r/114663 (owner: 10Krinkle) [00:11:48] (03PS1) 10EBernhardson: Show most recent items for watchlist but recentchanges [extensions/Flow] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/114665 [00:14:33] (03CR) 10EBernhardson: [C: 032] Show most recent items for watchlist but recentchanges [extensions/Flow] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/114665 (owner: 10EBernhardson) [00:14:41] (03PS1) 10EBernhardson: Show most recent items for watchlist but recentchanges [extensions/Flow] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/114667 [00:14:44] (03Merged) 10jenkins-bot: Show most recent items for watchlist but recentchanges [extensions/Flow] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/114665 (owner: 10EBernhardson) [00:15:02] (03CR) 10EBernhardson: [C: 032] Show most recent items for watchlist but recentchanges [extensions/Flow] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/114667 (owner: 10EBernhardson) [00:15:12] (03Merged) 10jenkins-bot: Show most recent items for watchlist but recentchanges [extensions/Flow] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/114667 (owner: 10EBernhardson) [00:15:52] (03PS15) 10Inchikutty: UploadWizard:OSM embedding- Geocoding, reverse geocoding [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/112342 [00:16:30] (03PS1) 10EBernhardson: Update Flow in 1.23wmf14 [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/114668 [00:16:42] (03CR) 10EBernhardson: [C: 032] Update Flow in 1.23wmf14 [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/114668 (owner: 10EBernhardson) [00:17:36] (03PS1) 10EBernhardson: Update Flow in 1.23wmf15 [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/114669 [00:20:40] (03CR) 10EBernhardson: [C: 032] Update Flow in 1.23wmf15 [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/114669 (owner: 10EBernhardson) [00:20:44] (03Merged) 10jenkins-bot: Update Flow in 1.23wmf14 [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/114668 (owner: 10EBernhardson) [00:23:02] (03PS16) 10Inchikutty: [WIP]UploadWizard:OSM embedding- Geocoding, reverse geocoding [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/112342 [00:24:24] (03Merged) 10jenkins-bot: Update Flow in 1.23wmf15 [core] (wmf/1.23wmf15) - 10https://gerrit.wikimedia.org/r/114669 (owner: 10EBernhardson) [00:24:34] (03CR) 10Legoktm: "I'm not very familiar with the mw.hook stuff, but is it possible to add something here so extensions like Titleblacklist or AntiSpoof can " [core] - 10https://gerrit.wikimedia.org/r/114400 (owner: 10Bartosz Dziewoński) [00:45:15] (03CR) 10Isarra: "Aiight, so it does seem this needs more to be properly effective:" [core] - 10https://gerrit.wikimedia.org/r/112066 (owner: 10Isarra) [09:46:16] (03PS1) 10Ori.livneh: Update MultimediaViewer to master for I41b6e9753 [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/114715 [09:46:57] (03CR) 10Ori.livneh: [C: 032] Update MultimediaViewer to master for I41b6e9753 [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/114715 (owner: 10Ori.livneh) [09:47:38] The holy ghost descended upon me and I had an epiphany [09:47:53] perl 6? [09:48:34] The answer was hidden because it was in front of everyone [09:48:50] Nemo_bis: do it on labsssss [09:48:52] "The Analytics team empowers and supports data informed decision making" first ten words http://www.mediawiki.org/wiki/Analytics#Mission [09:49:21] that sentence... needs at least one hyphen, possibly two [09:49:58] WMF's analytics are designed for decisions; community analytics are for serendipity [09:50:52] (03Merged) 10jenkins-bot: Update MultimediaViewer to master for I41b6e9753 [core] (wmf/1.23wmf14) - 10https://gerrit.wikimedia.org/r/114715 (owner: 10Ori.livneh) [09:52:40] Now I need a second revelation on how to reconcile the two... But I only know one archangel, Mikael aka Nikerabbit [09:54:12] (03CR) 10Addshore: "I guess once upon a time something somewhere depended on it? :>" [core] - 10https://gerrit.wikimedia.org/r/110879 (owner: 10Hashar) [09:54:15] Nemo_bis: Nikerabbit is analytics' angel? [09:54:40] ori: Part of the reason I'm anti-private mailing lists at the moment, is that there seems to be a lot of "lets take this conversation to private space" comments recently [09:54:54] yuvipanda: this is what 5 minutes of looking at perl does to someone [09:55:34] The Flow bug, the CSS font thread, recent bug requesting new private mailing lists for multimedia. [09:56:11] what was this flow bug again? [09:56:56] comments like "Personally I find impersonal communication methods like bugzilla a passive aggressive way to air issues without actually talking to anyone directly." [09:57:10] mostly this bug https://bugzilla.wikimedia.org/show_bug.cgi?id=57659 [09:57:27] bawolff: mostly side effect of concentration on SF office, I think [09:58:01] "In the future, let's not rely on passive Bugzilla communication around things you feel are serious blockers; let's just talk to each other directly" [09:58:34] bawolff: what multimedia ML? [09:59:06] (03PS1) 10Aude: Add visibility to User functions [core] - 10https://gerrit.wikimedia.org/r/114716 [10:00:04] bug 61731 is requesting new private mailing list for multimedia for unknown reasons (The existing mailing list is totally unused, so I'm not sure what the point is) That when could be me being crotchetey though, there could potentially be legitament reasons (If they want to discuss hiring or something) [10:00:38] Every team has its own private mailing list I think. [10:00:45] (03CR) 10jenkins-bot: [V: 04-1] Add visibility to User functions [core] - 10https://gerrit.wikimedia.org/r/114716 (owner: 10Aude) [10:00:47] (03PS2) 10Aude: Add visibility to User functions [core] - 10https://gerrit.wikimedia.org/r/114716 [10:01:17] And often private chats too, in addition to the actual meetings or office chat if they are in the office [10:01:21] Anyways, I know there's going to be lots of offline communication when everyone is in the same place, and that's not a bad thing. But people shouldn't be discoraged from saying things publically [10:01:22] bawolff: link to the mediawiki-config thing that broke betalabs? perhaps we can get ori to revert it [10:01:32] i am fixing it [10:01:36] sorry about that [10:06:24] beta is back [10:06:44] ooo [10:06:45] woo [10:07:29] Anyways, I'm not advocating something crazy like http://lists.wikimedia.org/pipermail/wikitech-l/2010-September/049168.html - All I'm saying is that things should be public where it is technically possible for them (e.g. on a mailing list where we can just as easily have it public) unless there is an absolute demonstrable need to have them private (security, hiring decesions), and if there... [10:07:30] ...is a choice between doing something privately or publically, the official line should be that public is the right answer. [10:08:00] bawolff: yeah, i agree with that [10:08:31] it takes some reminding but i think on the whole things are okay [10:09:11] oh yes, things could certainly be much much worse [10:09:20] they always can [10:09:50] prtksxna: https://bugzilla.wikimedia.org/show_bug.cgi?id=57891 [10:13:27] bawolff: hey, thanks for linking to that post. it's actually really well-argued. i think i agree with it. what do you find so crazy? [10:13:39] (03PS1) 10Prtksxna: Increase the space between the link and the popup [extensions/Popups] - 10https://gerrit.wikimedia.org/r/114720 [10:13:58] ori: Mostly the last paragraph I think is too extreme to be realistic [10:14:01] (03CR) 10Yuvipanda: [C: 032 V: 032] Increase the space between the link and the popup [extensions/Popups] - 10https://gerrit.wikimedia.org/r/114720 (owner: 10Prtksxna) [10:14:25] e.g. "Don't conduct teleconferences about development, ever" [10:15:24] And well he makes a good argument for "Stop concentrating tech employees in San Francisco", I think that's also rather unrealistic, and there are important benefits to having lots of the developers together [10:17:33] well, i happen to find teleconferences awkward and unproductive, so i'm only too eager to find some moral grounds for my preference :) but yeah, it'd be a hard sell. not it's not crazy, imo. [10:17:44] *but it's not [10:17:56] I think private vs. public is mostly a false dichotomy. Look at Wikimedia projects, communityLore can all be public but if I have to read 2 MB of talk archives in order to find out how to place a comma "transparency" has not helped me much. [10:18:40] Nemo_bis: that's interesting; i hadn't thought of that [10:19:18] are you saying that the degree of investment and experience you need to have to participate meaningfully in these discussions makes them exclusive even if they're technically out in the open? [10:19:19] Nemo_bis: True that, although hidden in 2 MB is a lot better than hidden by virtue of being non-public [10:19:35] * yuvipanda invites ori to a midnight meeting with 8 participants of which only 3 shall speak [10:19:49] yuvipanda: don't remind me [10:19:58] heh [10:20:23] bawolff: not necessarily; a private conversation with expected public outcomes forces people to produce a summary, if they care in the least; while if everything is public some may wrongly assume there is nothing more to do [10:20:43] yuvipanda: The sacrafice a virgin on the alter of Greg so that your extension will get deployed, meeting? [10:21:00] bawolff: why would that need 3 persons taling? [10:21:22] bawolff: heh, no :) [10:21:29] bawolff: I've gotten an extension deployed, I didn't have to sacrifice much [10:21:44] although it was not really an 'extension' as a bunch of RL modules so maybe that [10:22:15] a cock to asclepius [10:22:15] Nemo_bis: prime number. Occult developer meetings like prime numbers [10:22:48] the irc diaspora is pretty bad [10:23:01] bawolff: you can only factorise projects in a limited number of them? [10:23:04] i've probably done my share to contribute [10:23:11] Agreed, I used to hate #wikimedia-dev [10:23:29] but compared to #wikimedia-, its not too bad [10:23:44] yeah, but #mediawiki is half-dead [10:23:54] and half-bot [10:23:57] basically, it's robocop. [10:24:20] #wikimedia-mobile, #wikimedia-growth, #mediawiki-core, #wikimedia-dev, #wikimedia-design (which is active now!!1), #wikimedia-labs, #wikimedia-operations, #wikimeda-opearations, #wikimedia-corefeatures, #wikimedia-qa, #mediawiki, #wikimedia-analytics [10:24:39] add in an average of 3 cabal channels [10:24:58] The fun part is, that whenever I go to a random one of these channels, Gloria is almost always there, no matter how obscure it is [10:25:06] oh, #mediawiki-i18n, but that's almost all bot anyway (that team uses Skype, I believe, instead of IRC) [10:25:33] mediawiki-i18n is kind of a community channel for translatewiki.net from what I understand [10:25:35] yeah, but attempts to consolidate things get mired in details [10:25:53] should we appoint a small committee of users to make some binding decisions about irc channels? [10:25:58] i think that might be away to be productive [10:26:13] That wouldn't be too bad if only on could safely guess where all the conversations about a certain topic might have happened, like a sorting by topic of IRC chats. But most channels are created with significant overlaps. [10:26:20] yeah, and the commitee should include at least one person who is exclusively on each channel. that should help [10:26:30] oh, there's also #wikimedia-multimedia [10:26:31] I think [10:27:11] i use #mediawiki-core to hide sometimes but usually from other wmfers :P [10:27:38] heh [10:27:56] ori: I can't see why you would want to hide at all ;) [10:28:23] bawolff: #mediawiki-i18n has existed since 2006 [10:28:57] it was originally created to bootstrap some huge renovation efforts of l10n files [10:29:18] then there's #mediawiki-feed, which has nobody I think [10:29:25] from what I gathered in public information ;) [10:29:41] yuvipanda: I think historically that had the cia bot in it, which was used to feed another bot for the other channel [10:29:43] yuvipanda: there is wm-bot, nobody else needed [10:30:31] so, having #mediawiki is uncontroversial, i presume; ascending the controversy ladder -- but only by a single rung -- is the requirement for a developer- (rather than user-) focused channel. is there anyone who disagrees with that? [10:30:42] wm-bot: oh, you're there now; it would've been fun to keep a conversation about private rooms without our logging bot in room :) [10:30:43] Hi Nemo_bis, there is some error, I am a stupid bot and I am not intelligent enough to hold a conversation with you :-) [10:31:06] that's adorable [10:31:34] ori: I think -dev is ok. [10:31:46] I've grown to accept -dev [10:32:05] but -tech and -dev overlap, and -tech is dead [10:32:17] -tech is linked to from our 500 pages or somesuhc [10:32:19] *somesuch [10:32:29] -tech used to be ops, which is why I don't idle there as I don't care about ops :) [10:32:32] because -tech was spinned off to -operations which was spinned off to _security [10:32:47] There's a _security channel! [10:32:48] and the wikimedia error never linked -tech [10:33:10] I thought it did in the early days, but changed to #wikipedia after it annoyed too many people or something [10:33:12] let's take the site down for a moment to check [10:33:16] what if we shuttered -tech and forwarded it to -dev? [10:33:33] -tech is commonly used for users asking WMF tech questions [10:33:36] freenode's chanserv is capable of an irc equivalent of an http 301 [10:33:47] yeah, but they might as well ask them here, no? [10:34:00] I'd say let's try it and see what happens [10:34:04] so in many ways its closer to overlapping with #mediawiki, which tends to be third party users asking same kind of question [10:34:23] yeah but -tech has questions like 'this template isn't working the way i expect, wattodo?' [10:35:03] that question wouldn't be horrendously out of place here [10:35:24] indeed [10:35:41] this channel is completely useless, it should never have existed [10:36:14] Lets forward #wikimedia-dev to #mediawiki, how many angry people do you think that would get? [10:36:14] yeah, but it took off [10:36:21] and eclipsed -tech [10:36:21] it was useless in all its former lives too [10:36:37] not really, it cannibalised #mediawiki [10:37:01] i would support forwarding it to #mediawiki [10:37:14] ++++ [10:37:21] * yuvipanda would be ok with merging everything except -ops and -security too [10:38:08] who has the credentials and clout to pull that off? [10:38:23] an RFC might be the way to go [10:39:03] credentials is easy, clout is harder [10:39:08] yeah, +1 to RfC [10:39:11] Yeah, you might need an rfc to pull it off [10:39:32] * ori isn't volunteering, I am already under-delivering with prose [10:40:11] Historically some rather high clout people have said negative things about this channel, but didn't manage to get it shut down, so this isn't the first attempt [10:40:24] (03CR) 10Addshore: [C: 032] Add visibility to User functions [core] - 10https://gerrit.wikimedia.org/r/114716 (owner: 10Aude) [10:40:41] hehe [10:40:57] * yuvipanda pokes ori about sleep and relative time [10:41:12] Time is an illusion. Bed time doubly so [10:41:14] w00t! [10:42:51] bawolff: I thought that was lunch time [10:43:06] that too [10:43:43] all time is cubic, of course [10:43:51] which actually mean 4 rather than 3 or 6, clearly [10:45:14] (03Merged) 10jenkins-bot: Add visibility to User functions [core] - 10https://gerrit.wikimedia.org/r/114716 (owner: 10Aude) [10:45:49] yuvipanda: it'll be used to enhance absolute timestamps, presumably [10:45:56] ori: yes. [10:46:02] so that means it'll have to run at the bottom of at the earliest [10:46:02] so no js gives you abs times. [10:46:14] $(document).ready? [10:46:25] bottom of comes sooner [10:46:30] hmm,right [10:46:30] which means it will cause a FOUC [10:46:34] yup [10:46:37] I mentioned it in the bug [10:46:40] and it might even cause relayout [10:46:43] since the messages will be different [10:46:44] and a particularly nasty species of foucs [10:46:46] right [10:46:46] and so lengths etc [10:46:52] make the actual text shift [10:46:54] yup [10:47:02] link targets can change [10:47:12] I *hate* when websites move link targets around [10:47:28] yes, and your click registers on a different element than the one you clicked on [10:47:31] yes [10:48:05] espncricinfo does it by lazy loading a banner on top asking me to install their chrome extension. shifts everything by about 150px vertically. ANNOYING. [10:48:11] i think we should do it server-side [10:48:17] ori: but caching? [10:48:28] we have to kick that addiction [10:48:44] ori: agreed, we can't hide behind caching all the time to hide the fact that our perf actually sucks balls [10:48:47] hhvm might be a unique opportunity [10:49:09] hmm, right [10:49:26] I don't know how plausible it is, though [10:49:29] server-side is probably the only really sensible way to do this; i'd imagine that's what facebook does [10:49:47] ori: at least github autoupdates, so it is possibly server side + clientside [10:50:04] but yeah, no relayouts/content flashing [10:50:25] if you design for it from the get-go you can preallocate a fixed space for timestamps in your layout [10:50:32] ori: if we make the time at the bottom of the page every response relative, for example - that'll kill caching completely. [10:50:46] ori: true. but the timestamp itself would change, and pre-allocating across languages is a pain [10:51:34] ori: ESI is a potential solution, but we don't have anything for that I suspect [10:52:53] (03CR) 10Springle: "user_password_expires has been added to all wikis." [core] - 10https://gerrit.wikimedia.org/r/92037 (owner: 10CSteipp) [10:53:19] plop [10:53:30] +1 to less irc channels btw [10:53:58] Nikerabbit: now you need to repeat that in the other 50 channels [10:55:20] I can't say how much I hate it when people ask me to join channel X because they are having discussion there which is relevant to me, but as I was not in the channel I do not have the log of the discussion [11:48:09] legoktm: You know about the abuse filter! want to answer https://commons.wikimedia.org/wiki/Commons:Administrators%27_noticeboard#Spam_filter_on_talk_pages [11:53:27] grrrit-wm: are you alive [11:53:57] Nemo_bis: possibly. I just got a ping in -mobile to a patchset i pushed a few mins ago [14:36:27] (03CR) 10Anomie: [C: 032] "Ok, let's go for it" [core] - 10https://gerrit.wikimedia.org/r/110210 (owner: 10Aaron Schulz) [14:39:50] (03CR) 10Prtksxna: "Works properly on my instance. I think this works very well. I would have liked it if the "errorbox" came below the input field right now " (031 comment) [core] - 10https://gerrit.wikimedia.org/r/114400 (owner: 10Bartosz Dziewoński) [14:40:15] (03Merged) 10jenkins-bot: UploadStash us_props fix for Postgres [core] - 10https://gerrit.wikimedia.org/r/110210 (owner: 10Aaron Schulz) [14:43:21] (03PS1) 10Zfilipin: Updated to the latest version of mediawiki_selenium Ruby gem [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/114743 [14:43:49] (03CR) 10Manybubbles: [C: 032] Updated to the latest version of mediawiki_selenium Ruby gem [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/114743 (owner: 10Zfilipin) [14:43:54] (03Merged) 10jenkins-bot: Updated to the latest version of mediawiki_selenium Ruby gem [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/114743 (owner: 10Zfilipin) [14:54:56] (03PS1) 10BryanDavis: Cleanup scap.ssh integration [tools/scap] - 10https://gerrit.wikimedia.org/r/114744 [14:55:43] (03CR) 10BryanDavis: [C: 04-1] "Needs more thought." [tools/scap] - 10https://gerrit.wikimedia.org/r/114692 (owner: 10BryanDavis) [15:02:05] werdna: do you know about hide vs. suppress vs. delete in Flow? https://translatewiki.net/wiki/Thread:Support/About_MediaWiki:Flow-suppress-post-content/sv [15:02:45] It would realy be very very useful if qqq docs clarified something and linked the corresponding messages in core to ensure some consistency. [15:03:34] maybe quiddity understands this new level of terminological complexity [15:04:19] Nemo_bis: AFAIK: Hide is like "collapse per default" (but everyone can still unhide), delete and suppress is like with usual MediaWiki revisions [15:04:25] maybe a flow person can confirm [15:05:11] ugh [15:05:48] can't it just be called "collapse" if that's what it does, sob sob sob [15:07:29] One day I'll just sit down for a hour and rewrite the terminology of revdelete, oversight and all silly imitators [15:13:50] (03CR) 10Edokter: [C: 031] "Go ahaead with this one; the primary reason for my previous patch was to fix wrapping behaviour, which has been resolved by disabling nowr" [core] - 10https://gerrit.wikimedia.org/r/94443 (owner: 10Cscott) [15:13:52] Now how to file a bug about this without having to follow it [15:14:25] Nemo_bis: file a bug and then remove yourself from cc? [15:15:15] again this myth [15:15:21] IT DOESN'T WORK [15:15:27] heh [15:15:30] oh [15:15:31] wait [15:15:34] you can't remove as reporter [15:18:13] Does someone know if dan-nl is still active? [15:18:24] I've not seen him around for weeks [15:20:11] me neither [15:20:50] hoo: does your capiunto experience also make you able to answer this? https://translatewiki.net/wiki/Thread:Support/About_MediaWiki:Scribunto-luastandalone-proc-error-proc-open/hi [15:22:51] Nemo_bis: These are PHP functions, so there's *no* localization [15:23:18] hoo: fantastic, can you say so on thread [15:23:42] plop [15:24:19] hi Nikerabbit [15:24:27] ug [15:24:49] Nikerabbit: I also have something for you as ti sei palesato [15:25:22] Nemo_bis: tu sei? [15:25:38] Nikerabbit: nope, "ti sei" reflexive [15:25:40] https://translatewiki.net/wiki/Thread:Support/About_MediaWiki:Logentry-translatorsandbox-rejected/en [15:25:49] mihi tibi etc. [15:26:04] is that a bunny language? [15:26:19] it's latin [15:26:28] keep the dative idea and drop the [hb]i [15:26:57] whatever, what was the point you were making? [15:27:18] (03PS2) 10Phuedx: Center GettingStarted modal in IE7+ [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/114347 [15:27:30] that as you're here you can open that thread [15:28:29] Nemo_bis: open? [15:28:43] it is marked as open [15:28:58] and it says only Nike (no idea who he is) can resolve it [15:29:49] I warn you, I bite [15:30:08] I warn you, my blood is poisonous [15:30:58] I'm mitridatizzato [15:44:56] (03CR) 10Nemo bis: Add messages for Trademark Licensing form for foundationwiki (031 comment) [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/110243 (owner: 10Reedy) [15:46:17] (03CR) 10Nemo bis: "ugh. So basically we're going to translate the content of foundationwiki in WikimediaMessages extension? O_O" [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/110243 (owner: 10Reedy) [15:48:51] (03CR) 10Reedy: "No. This is for 2 contact forms for legal usage (see also https://gerrit.wikimedia.org/r/113883 and previous commits)" [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/110243 (owner: 10Reedy) [15:49:46] (03CR) 10Reedy: "And https://meta.wikimedia.org/wiki/Trademark/Request_a_license_form and https://meta.wikimedia.org/wiki/Trademark/Report_abuse" [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/110243 (owner: 10Reedy) [16:17:48] (03PS1) 10Zfilipin: Updated to the latest released version of mediawiki_selenium Ruby gem [extensions/Flow] - 10https://gerrit.wikimedia.org/r/114755 [16:18:57] (03CR) 10Cmcmahon: [C: 032] Updated to the latest released version of mediawiki_selenium Ruby gem [extensions/Flow] - 10https://gerrit.wikimedia.org/r/114755 (owner: 10Zfilipin) [16:20:31] (03CR) 10Phuedx: [C: 032] Use VisualEditor configuration for appendices (e.g. beta). [extensions/GuidedTour] - 10https://gerrit.wikimedia.org/r/114698 (owner: 10Mattflaschen) [16:50:15] reminder: RFC review meeting in 10 min in #wikimedia-office [16:50:54] https://www.mediawiki.org/wiki/Architecture_meetings/RFC_review_2014-02-21 [16:53:11] (03CR) 10Phuedx: [C: 031] "+1 until we've finished discussing it on Trello (https://trello.com/c/c45DXjZF/288-updates-for-gettingstartednosuggestedarticle)." [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/114704 (owner: 10Mattflaschen) [16:59:32] (03PS2) 10EBernhardson: Makefile command to run phpstorm inspections [extensions/Flow] - 10https://gerrit.wikimedia.org/r/114701 [16:59:36] (03PS2) 10EBernhardson: pre-review script to run check-i18n and analyze-phpstorm when appropriate [extensions/Flow] - 10https://gerrit.wikimedia.org/r/114703 [16:59:38] (03PS2) 10EBernhardson: Makefile command to check i18n for matching en and qqq keys [extensions/Flow] - 10https://gerrit.wikimedia.org/r/114702 [17:08:22] (03CR) 10Siebrand: [C: 04-1] "Any reason this isn't in a separate i18n (JSON) file?" [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/110243 (owner: 10Reedy) [17:32:47] DanielK_WMDE_: http://bots.wmflabs.org/~wm-bot/logs/%23wikimedia-office/20140221.txt has the log of the rfc meeting up till now - in #wikimedia-office [17:33:28] doh, I missed it :P [17:34:06] hoo: it's still happening now [17:53:42] ^d: DanielK_WMDE_ csteipp anomie yurik - any RFCs you want to have discussed, get feedback, etc. in next week's RFC meeting? [17:53:58] Tim was out this week but will be back next week [17:54:10] <^d> I won't be around most of next week anyway, vacation. [17:54:13] ok [17:54:27] I'm offline most of next week [17:54:30] DanielK_WMDE_: csteipp - I already have TitleValue & passwords down for 5 March https://www.mediawiki.org/wiki/Architecture_meetings/RFC_review_2014-03-05 [17:54:32] sumanah: None for me, thanks [17:54:32] ok [17:56:28] MaxSem: ori Krenair yuvipanda StevenW and anyone else: same question. any RFCs you have authored or commented on and want to have discussed at next week's RFC meeting, get feedback & decisions? [17:59:27] sumanah, Inline diffs, probably there will be a complete implementation for DataStore [18:00:54] (03CR) 10Krinkle: [C: 031] "No idea how this works, but this seems to be the only mention of it. The property is never set to anything else by UploadWizard, and is in" [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/114697 (owner: 1001tonythomas) [18:01:12] thx MaxSem - is your GCal accurate so I can schedule partially based on when you are available? [18:01:24] sumanah, yes [18:04:57] gwicke: not sure if anyone will, just curious if I was missing it [18:06:04] greg-g: we'll see [18:06:28] To be honest, the service stuff was the thing I was interested in seeing be discussed, and we didn't really get to it [18:06:48] bawolff, we discussed that during the architecture summit [18:06:57] and pretty much everyone agreed [18:07:28] I mean, I was interested to see if progress had been made towards it actually happening [18:07:31] packaging is one of the things that are needed to make services feasible for third parties [18:07:34] or otherwise what the status of it is [18:08:03] bawolff, there are several projects ongoing in that direction [18:08:25] parsoid is an obvious one, but there's also Rashomon/Storoid, Mathoid, the PDF renderer etc [18:09:26] mmm, all nodejs [18:09:27] there was fairly strong consensus that having a storage abstraction is a good candidate for a service [18:09:52] AaronSchulz has been working on a nice PHP interface for services [18:10:15] https://www.mediawiki.org/wiki/Requests_for_comment/PHP_Virtual_REST_Service [18:10:32] WIP at https://gerrit.wikimedia.org/r/#/c/110129/ [18:12:31] bawolff, it is a gradual process in any case; the main point was to agree on a direction we want to move in [18:13:16] Ok. Thanks for the update (I haven't really been following things, so its nice to hear what's up with things) [18:18:17] (03PS1) 10BryanDavis: Fix TypeError exception in scap.tasks.scap.run_on_cluster [tools/scap] - 10https://gerrit.wikimedia.org/r/114760 [18:19:49] Can review a 3-character change in a python file for scap? ^^ [18:20:11] I'll take a +1 or even a sure what could that hurt review [18:22:11] (03CR) 10BryanDavis: Fix TypeError exception in scap.tasks.scap.run_on_cluster (031 comment) [tools/scap] - 10https://gerrit.wikimedia.org/r/114760 (owner: 10BryanDavis) [18:22:30] gwicke: How about you? https://gerrit.wikimedia.org/r/#/c/114760/ [18:22:47] Inline comment shows the error being corrected [18:24:57] (03CR) 10GWicke: [C: 031] "LGTM" [tools/scap] - 10https://gerrit.wikimedia.org/r/114760 (owner: 10BryanDavis) [18:27:40] gwicke: Thanks [18:29:52] (03CR) 10BryanDavis: [C: 04-1] "Needs rebased on I7ea1bb5cb8. Don't mix pretty with critical." [tools/scap] - 10https://gerrit.wikimedia.org/r/114744 (owner: 10BryanDavis) [18:30:10] (03CR) 10Manybubbles: [C: 032] "You can't len an int." [tools/scap] - 10https://gerrit.wikimedia.org/r/114760 (owner: 10BryanDavis) [18:30:16] (03Merged) 10jenkins-bot: Fix TypeError exception in scap.tasks.scap.run_on_cluster [tools/scap] - 10https://gerrit.wikimedia.org/r/114760 (owner: 10BryanDavis) [18:36:13] greg-g: Is it OK if I touch a file to work around some code being stubbornly cached in localStorage for some people? [18:36:24] (Where "touch" literally means touch(1)) [18:37:45] Has anyone raised the issue about meetbot topics in a bug somewhere? [18:41:09] (03PS2) 10Krinkle: jquery.tablesorter.test: Re-organise variables [core] - 10https://gerrit.wikimedia.org/r/111962 [18:42:06] (03PS3) 10EBernhardson: Makefile command to run phpstorm inspections [extensions/Flow] - 10https://gerrit.wikimedia.org/r/114701 [18:42:09] (03PS3) 10EBernhardson: pre-review script to run check-i18n and analyze-phpstorm when appropriate [extensions/Flow] - 10https://gerrit.wikimedia.org/r/114703 [18:42:12] (03PS3) 10EBernhardson: Makefile command to check i18n for matching en and qqq keys [extensions/Flow] - 10https://gerrit.wikimedia.org/r/114702 [18:42:15] (03PS1) 10EBernhardson: share code between pre-commit and pre-review hooks [extensions/Flow] - 10https://gerrit.wikimedia.org/r/114764 [18:42:29] (03CR) 10jenkins-bot: [V: 04-1] pre-review script to run check-i18n and analyze-phpstorm when appropriate [extensions/Flow] - 10https://gerrit.wikimedia.org/r/114703 (owner: 10EBernhardson) [18:42:48] (03CR) 10jenkins-bot: [V: 04-1] share code between pre-commit and pre-review hooks [extensions/Flow] - 10https://gerrit.wikimedia.org/r/114764 (owner: 10EBernhardson) [18:44:54] (03PS4) 10EBernhardson: pre-review script to run check-i18n and analyze-phpstorm when appropriate [extensions/Flow] - 10https://gerrit.wikimedia.org/r/114703 [18:45:13] (03PS2) 10BryanDavis: Cleanup scap.ssh integration [tools/scap] - 10https://gerrit.wikimedia.org/r/114744 [18:45:44] (03PS8) 10Prtksxna: Add Triangle/Pokey to the NavigationPopups [extensions/Popups] - 10https://gerrit.wikimedia.org/r/113951 [18:46:56] (03PS2) 10EBernhardson: share code between pre-commit and pre-review hooks [extensions/Flow] - 10https://gerrit.wikimedia.org/r/114764 [18:47:35] (03CR) 10Ori.livneh: [C: 032] Cleanup scap.ssh integration [tools/scap] - 10https://gerrit.wikimedia.org/r/114744 (owner: 10BryanDavis) [18:47:37] (03Merged) 10jenkins-bot: Cleanup scap.ssh integration [tools/scap] - 10https://gerrit.wikimedia.org/r/114744 (owner: 10BryanDavis) [18:48:31] ori: thanks [18:54:12] RoanKattouw: yep, go ahead, bd808 is scaping right now though [18:54:21] OK [18:54:24] bd808: Have you started yet? [18:54:34] Almost done. [18:54:39] Oh OK [18:54:42] and done [18:54:45] Then I'll wait [18:54:48] Oh hah OK [18:54:51] I'll go now then [18:56:18] RoanKattouw: All clear [18:56:24] Going [19:03:20] ori: still lots of fixes needed to the job patch (e.g. not pretending that the write happens in the background as it can't in PHP, and fixing a race condition in ignore_user_abort()) [19:03:58] AaronSchulz: yeah, i know. it was a "yes, this is the right direction" +1 [19:04:09] like if I do sleep(5) in front of the ignore_user_abort(), it doesn't run [19:04:12] AaronSchulz: do you think it would be feasible to make it use api.php rather than a novel entrypoint? [19:05:15] not sure how to fully deal with that, I've been messing with sending STARTED\n and flushing that write after the ignore_user_abort() call, then blocking until that is read in wiki.php [19:07:18] the non-blocking write wasn't so you could write in the backround when the socket's buffer has space, it was so you could abort and move on [19:07:42] it's simpler to just use stream_set_timeout() [19:07:43] it's probably safe to assume that in the vast majority of cases you would be able to write "GET /w/chron.php?..." in one go [19:07:57] either way you end up waiting until you timeout [19:08:07] I think 2 seconds should be fine [19:08:08] but if the socket would block, just forget it, move on and try again in another request [19:08:31] maybe 1 second [19:08:47] * AaronSchulz goes with 1 for now [19:09:58] why not leave it for another request? [19:11:17] try a nonblocking write of the full request to the socket, if EWOULDBLOCK then move on [19:12:14] how is that different than a very low timeout? [19:12:57] does the 1 second hang the original request? [19:13:02] either way I need to wait on the response to know that disconnect is safe [19:13:29] which means it has to wait till the write finishes [19:14:13] why wouldn't it be safe to disconnect? [19:14:32] csteipp: so you can remove the -2 from https://gerrit.wikimedia.org/r/#/c/92037/ afaik [19:14:58] ori: if you disconnect before chron reaches ignore_user_abort() then it will stop execution and do nothing [19:14:59] (03CR) 10CSteipp: "Schema change done" [core] - 10https://gerrit.wikimedia.org/r/92037 (owner: 10CSteipp) [19:15:17] AaronSchulz: Thanks! Forgot to do that.. [19:29:40] legoktm: Gloria can either of you +1? https://gerrit.wikimedia.org/r/#/c/114719/ [19:33:03] ori: just made a new version, I think it should be fine now...not sure about whether it can use the API or not [19:33:07] patches welcome ;) [19:33:24] ugh, slow gerrit [19:42:01] (03CR) 10Reza: "First posts here also worth to read: https://groups.google.com/forum/#!topic/persian-typography/ayPZiw32YGc" [extensions/Collection/OfflineContentGenerator/latex_renderer] - 10https://gerrit.wikimedia.org/r/110174 (owner: 10Reza) [19:57:01] yuvipanda: looking [19:59:02] legoktm: ok [20:15:52] (03CR) 10CSteipp: [C: 04-1] "Minor stuff inline. Only thing that needs to change is the maintenance script." (033 comments) [core] - 10https://gerrit.wikimedia.org/r/77645 (owner: 10Parent5446) [20:24:58] (03CR) 10Aaron Schulz: [C: 032] Password Expiration [core] - 10https://gerrit.wikimedia.org/r/92037 (owner: 10CSteipp) [20:25:48] Does that really belong in core...? [20:25:58] Yep :P [20:26:31] I wanted to even have it per user group (in core), but well [20:26:39] heh [20:28:32] (03Merged) 10jenkins-bot: Password Expiration [core] - 10https://gerrit.wikimedia.org/r/92037 (owner: 10CSteipp) [20:28:52] (03CR) 10Aaron Schulz: "I think as long as the change and "rollback" strategy is mentioned in the release notes, then it's fine to change the default." [core] - 10https://gerrit.wikimedia.org/r/77645 (owner: 10Parent5446) [20:32:37] (03PS1) 10Hashar: Fix doctest for utils.find_nearest_host() [tools/scap] - 10https://gerrit.wikimedia.org/r/114771 [20:37:10] csteipp: https://en.wikipedia.org/wiki/Iff ;) [20:39:40] (03CR) 10CSteipp: "Ah, talked with Aaron about the IFF comment. His issue was with the "if" not the "only if". In that two users with the same password will " [core] - 10https://gerrit.wikimedia.org/r/77645 (owner: 10Parent5446) [20:51:32] (03PS2) 10BryanDavis: Fix doctest for utils.find_nearest_host() [tools/scap] - 10https://gerrit.wikimedia.org/r/114771 (owner: 10Hashar) [20:52:35] (03CR) 10BryanDavis: [C: 031] Fix doctest for utils.find_nearest_host() [tools/scap] - 10https://gerrit.wikimedia.org/r/114771 (owner: 10Hashar) [20:53:07] bd808: and I am going to add some tox env finally :-] [20:53:11] bd808: but without setup.py! [20:53:21] hashar: sweet! [20:53:27] (03PS1) 10Hashar: Tests require coverage [tools/scap] - 10https://gerrit.wikimedia.org/r/114774 [20:54:22] (03CR) 10BryanDavis: [C: 032] Tests require coverage [tools/scap] - 10https://gerrit.wikimedia.org/r/114774 (owner: 10Hashar) [20:54:24] (03Merged) 10jenkins-bot: Tests require coverage [tools/scap] - 10https://gerrit.wikimedia.org/r/114774 (owner: 10Hashar) [20:55:30] hashar: If the tweaks I made to your unit test look ok to you I can +2 [20:56:50] bd808: looking [20:58:10]