[00:00:15] it's easily overridden by accident. Which is a yay for ResourceLoader since the options will be in a more secured Map [00:00:22] mw.options.get('skin') [00:06:11] <^demon> kaldari: The FundraiserPortal extension isn't used anymore, right? [00:06:42] not currently, but it might be one day [00:07:26] I think the only wiki it's turn on in currently is test.wikipedia [00:07:31] turned [00:07:34] <^demon> aw wanted the fixme to go away :( [00:16:15] easier AIO query is easier [00:16:16] select cr_status, count(*) from code_rev WHERE cr_repo_id = 1 AND cr_id < 77974 AND cr_status IN ( 'fixme', 'new' ) GROUP BY cr_status; [00:37:18] robla: What's the situation with the Bugmeister? [00:54:24] OK, if someone beside me can review r79696, I'll mark the 3 dependant fixmes as resolved. http://www.mediawiki.org/wiki/Special:Code/MediaWiki/79696 [00:55:17] this is just removing some wikimedia-specific hacks to CentralNotice that we utilized during the fundraiser, but won't need anymore [00:57:17] 5 o'clock, I guess everyone went home ;) [00:58:11] kaldari, looks like no one has moved :P [00:58:56] I guess they just went home mentally :) [01:13:32] kaldari: I think we'd mark the three dependent ones as new.... [01:13:59] (I'm saying "I think" because I'm not sure and I'm waiting for someone to jump in and either correct or confirm) [01:18:42] Shirley: we're planning on shuffling our hiring order, so Bugmeister will get postponed. [01:24:08] kaldari, yeah, feel free to mark as new [01:24:23] cool [01:27:40] robla: Oh? Is the new hiring order public? [01:32:14] Shirley: still getting approvals [01:32:24] All right. [01:37:28] *jorm approves. [01:40:06] watch as i delete the "history" link from posts in LQT discussions! abracadabra, it's MAGIC! [01:40:53] brion, join in with CR. Go on, you know you want to ;) [01:44:20] pssh [01:44:37] get us some git lovin' and CR's gonna be way friendlier [01:44:54] less need to shove crap into trunk before it's working just to share it [01:45:28] brion, have you ever worked with perforce? [01:48:11] not i [01:48:29] it can't be worse than the stories my dad tells of clearcase [01:48:46] is perforce like perchance? [01:48:58] to sleep, perforce to dream [01:48:59] not that i'm awareof. [01:49:04] i've used clearcase. it was hell. [01:49:17] also that microsoft thing, too. i forget what it was called. [01:49:28] sourcesafe? or was that another one [01:49:32] perforce is nice because you can make nigh infinite branches and it's really slick at merging. [01:49:35] *Reedy barfs [01:49:37] yes. sourcesafe. [01:49:41] hehe [01:49:42] TFS isn't much better [01:49:44] man. i have a sourcesafe nightmare story. [01:49:45] brion, I think git is a case of "yes, just not now" [01:49:47] nigh! [01:49:53] jorm, it lost your data? #trory [01:50:05] yeah, being very branch-and-merge friendly is one of my main 'yay's on git [01:50:09] we're using a lot of old-fashioned english [01:50:13] a long long time ago, in a galaxy far, far away, i was on a team to rebuild starbucks.com [01:50:19] most of the dvcs's are broadly similar in that regard though i'm sure folks have their favorites [01:50:22] brion, local stashing would be quite nice [01:50:30] now, microsoft was just getting into the web game, right? so they paid for the rebuild. [01:50:40] suckas [01:50:53] they gave us a million dollars to rebuild starbucks.com on the condition that we ONLY used microsoft software to do so. [01:50:56] soup-to-nuts. [01:51:05] which included sourcesafe. [01:51:08] oh god [01:51:14] FOUR TIMES. FOUR FUCKING TIMES. [01:51:21] sourcesafe corrupted on us. [01:51:25] Ouch [01:51:40] we lost EVERYTHING except for the code deployed on the test server, or what we had locally, FOUR FUCKING TIMES. [01:52:08] oooooouch [01:52:11] that was in addition to having to use their shitty visual editor and .asp and everything else. [01:52:11] "they gave us a million dollars to rebuild (a high traffic website) on the condition that we ONLY used microsoft software to do so."....that sounds like a really bad reality tv show [01:52:12] no backups? [01:52:16] or... no useful backups? :) [01:52:21] no useful backups. [01:52:38] it kept everything in binary files. because, you know, who would ever think those could go bad? [01:52:53] this was maybe 1998. [01:52:57] i like Visual Studio [01:53:02] for coding [01:53:02] we got drunk a lot. [01:53:39] oh, and we got in serious fucking trouble, too. [01:53:42] you are always getting drunk [01:53:49] and getting into serious effing trouble [01:53:58] at work [01:54:09] one of the developers wrote a comment on the homepage: [01:54:26] nah. never drunk at work, me. [01:54:38] the MS site comments of like "stupid fucking ie6 hack" [01:54:40] that wasn't my comment, either. [01:55:25] Ah, the good ol' days [01:55:28] a friend of mine got fired once because he used a bunch of test data in a comments form that talked shit about the client (macy's, iirc), and they saw it. [01:55:29] "sun fucking blows goats" :D [01:55:49] i used to love browsing code and seeing comments like this: // drunk; fix later. [01:56:01] or // i have no idea what the fuck this does now that i'm sober [01:56:26] haha [01:56:32] jwz once told me that when they released mozilla, they had to spend days going through the codebase and removing "fuck" from the comments. [01:56:39] something like 6,000 instances of it. [01:56:47] cursing wants to be free [01:56:54] 6,000 instances of "fuck" [01:56:57] they defucktified it. [01:57:00] we had to do that for Helix too [01:57:15] I don't remember if it was 6000, but it was way more than a couple [01:57:22] my favorite comment of all time was in xterm.c, about ANSI rendering. [01:57:33] // HERE THERE BE DRAGONS [01:58:02] it may still be there, for all i know. [01:58:32] http://www.youtube.com/watch?v=o22eIJDtKho Johnny Cash singing NIN [01:58:44] is that the hurt cover? [01:58:49] yes [01:58:51] one of my favorite cash tracks. [01:58:51] yeah, has to be [01:59:13] that whole album is fantastic [01:59:13] the way he does it is awesome. it's just the following chords: C, D, A over and over again. [01:59:25] but he stutter-strokes them. [01:59:48] like, the C chord is five strings. but he does them like 1, 2, rest of the chord. [02:00:33] this is my current favorite cover: http://www.youtube.com/watch?v=lyJeC99QO8A [02:00:47] so you get that da-da-DAH ..... da-da-DAH, da-da-DUM, da-da-DAH thing. [02:00:51] Jonathan Coulton's "Baby Got Back" [02:01:22] (it's a couple years old, but I only just discovered it) [02:01:55] you only like it 'cause sir mix-a-lot's posse is on broadway. [02:02:53] i'm a huge johnny cash fan. [02:03:27] this is in my house: http://www.facebook.com/photo.php?pid=1779474&l=9b58d8b91a&id=1066421084 [02:10:56] can someone see what $wgCentralPagePath is set to in the cluster config? Not sure if it's shared or not [02:17:01] kaldari, http://noc.wikimedia.org/conf/highlight.php?file=CommonSettings.php [02:17:05] $wgCentralPagePath = 'http://meta.wikimedia.org/w/index.php'; [02:17:18] cool, thanks! [02:34:07] robla, still about? [02:34:17] Reedy: yup, what's up? [02:34:22] Just was gonna say [02:34:28] I'm going through the fixme's and stuff [02:34:35] Seems quite a few were already resolved/could be marked new [02:34:38] Which is nice [02:35:38] ah...that's why the number was going up :) [02:36:25] | fixme | 79 | [02:36:25] | new | 454 | [02:36:26] Yeah [02:46:32] no more fixmes for me :) [02:46:43] time to go home [02:50:59] my only fixme is an extension which i think may be abandoned. [02:58:20] 77, 455 [03:00:47] jorm: I suspect we can mark it as deferred then [03:01:04] *jorm DEFERRED [03:02:15] Indeed [03:02:18] Or delete the extension [03:02:19] and reverted [03:02:58] no, deleting the extension may irritate Erik. [03:04:49] feel free to mark the revs as deferred [03:06:09] More fixmes that can be marked new [03:15:04] reedy, are you still at the office? [03:15:11] Yup [03:18:14] | fixme | 73 | [03:18:14] | new | 457 | [03:33:05] | fixme | 70 | [03:33:05] | new | 458 | [03:40:06] Dude. Go get some food and go home. [03:40:10] who else is there? [03:42:37] jorm, no one [03:42:45] I've been here 8-9 or so most nights [04:02:36] Hmm [04:02:43] I'm probably somewhat done for today now [04:02:50] | fixme | 69 | [04:02:51] | new | 458 | [04:24:12] | fixme | 69 | [04:24:12] | new | 445 | [07:33:15] TradeTrack? [12:48:23] Hello. How the permission 'disableaccount' works? I've noticed that we have it avalaible at http://meta.wikimedia.org/wiki/Special:GlobalGroupPermissions/steward [12:49:19] I don't now, ask werdna [12:49:21] He wrote it [12:50:15] Thank you. [17:05:59] hrn. my old game made metafilter today. [19:08:59] hi hexmode, check out http://eiximenis.wikimedia.org/1-17 [19:09:16] clicking [19:09:26] I'm hoping you can help out with the assignment task we're going through [19:10:20] so, here's the idea: [19:11:24] we'd like to make sure that all /trunk/phase3 revisions prior to 77974 that are marked "new" have a tag associated with one of the reviewers [19:12:15] the query at the top of that doc ( https://secure.wikimedia.org/wikipedia/mediawiki/w/index.php?limit=500&title=Special:Code/MediaWiki/status/new&offset=77974&path=/trunk/phase3 ) should work for getting the list [19:12:33] so, "hexmode" will be my tag... since I'm so hexxy here [19:14:04] Marked the ones I had open under includes/parser [19:14:22] (EtherPad needs an Emacs client ...) [19:14:38] *hexmode puts it on his TODO list [19:15:03] actually, Etherpad really doesn't need an Emacs client. Emacs might need it, but Etherpad doesn't :-P [19:15:12] :P [19:15:25] Nothing is complete until it works with Emacs [19:15:45] Some people think they want their software to run on Linux [19:15:54] I want it to work with Emacs [19:18:52] m-x do-what-i-want [19:19:03] robla: could you in the future send a note to the mailing list when you are making plans about code review and such? [19:19:14] that way volunteers could have an idea what is going on as well [19:20:54] hexmode, there's a few testy related revisions from memory [19:21:01] Bryan, it's actually Roans plan :P [19:21:38] or Roan [19:21:49] Reedy: thanks for the reminder. [19:22:18] Bryan: no philosophical objection to discussing on the list; just have a backlog of writing I need to do [19:23:38] not saying that you should discuss this on the mailing list, but a mail afterwards "we made a plan and it can be found here" would be nice [19:25:47] Bryan: sure....I was already planning on putting this inforamtion here: http://www.mediawiki.org/wiki/MediaWiki_roadmap/1.17 , so I'll send out mail after I've updated it [19:26:05] ok :) [19:37:35] I gave up my +staff rights today. [19:37:51] jorm, why that? [19:38:00] because i don't need them. [19:38:02] at all. [19:38:12] you'll get the status again shortly [19:38:14] but with no rights ;) [19:38:25] i don't like having checkuser. [19:38:43] That's one of the reasons why we're going to remove most of the rights from the flag. [19:39:34] i kept thinking to myself, "well, this right would be useful for me to do my job" but then i'd think, "no, my job is design; doing vandalism checking or other wikignome crap is the job of the admins of the specific wiki." [19:39:57] so in the end i'm just a civilian with a loaded gun and no need to use it so what's the point? [19:40:20] if i ever *do* need access (say, to edit central notice banners), I can get it from a local admin. [19:41:03] i was going to drop my +reviewer on enwiki. i picked it up for help with my designing bits of pending changes. i would drop it, but it's possible that it comes back. [19:41:19] hrn. [19:41:26] a status might be interesting, but that's also part of it. [19:41:56] having (WMF) in my username is kind of weird; i understand the need for separation, but it seems to give undue weight to my opinions in places. [19:42:07] jorm, see http://office.wikimedia.org/wiki/Office_IT/SUL_Permission_Changes_Proposal [19:42:12] now, there are places where i *should* have extra weight. but not day today. [19:42:54] reboot, brb [19:42:59] Reedy: the only pre-branch rev in /tests is r61911 which I can't review. [19:43:07] ohh, thought there was more [19:43:09] Oh well :) [19:44:00] there's lots since the branch, so its good for me to keep in mind [19:55:56] hexmode, you've got 2 fixme's pre branch though [19:55:56] https://secure.wikimedia.org/wikipedia/mediawiki/w/index.php?title=Special:Code/MediaWiki/status/fixme&author=mah [19:56:30] Seemingly no comment from yourself though [19:56:39] dang, was hoping the magic pixie dust would remove them. [19:56:54] haha [19:57:05] If you think they're fixed/whatever, make a note, and set them back new at least [20:09:51] Reedy, mostly I did just forget about them, which is a shame since *I* marked one of them fixme. [20:13:18] robla, do you need communications / community help with the 1.17 release? I could give a hand if you think I can be helpful. [20:15:31] guillom: ooh, yeah, actually that would be helpful if you've got time for it [20:15:38] down to ~450 unreviewed pre-branch rev [20:15:56] robla, I do (have time for it). But I'd need an idea of what you need. [20:16:32] robla, maybe you can make a short list and email it to me before next week? [20:16:38] and we can discuss it next week [20:17:40] sure! there's some immediate stuff I'm thinking of that I need to get done this week, but I'm sure there is more needed next week :) [20:18:37] if you send me stuff today I can try to get to it tomorrow while you sleep [20:18:43] if it's specific enough [20:19:48] guillom: there's a lot of general copyediting that needs to happen on this: http://www.mediawiki.org/wiki/WMF_Engineering_Overview_January_2011 [20:20:05] ok [20:20:09] *guillom adds to todo list [20:20:43] guillom: according to your thing, the engineering rights are all i'd need. [20:20:55] jorm, good! [20:21:28] guillom: I should have invited you to our drafting session for it this time. do you want me to put you on the invite list on future drafting sessions? [20:21:50] ugh, writing this email to zexley and co, i am reminded of how great the ea jobs site could have been, and how crappy i was forced to make it. [20:22:04] consider: http://elohim.gaijin.com/jobs-mockup-black-2.jpg vs. http://jobs.ea.com [20:22:07] robla, dunno how that will work remotely, but you can invite me and we'll see [20:22:45] guillom: speaking as someone who was remote for the last one, it worked fine...the only hitch may be scheduling [20:22:57] Mark B was also remote [20:23:24] anyway, I'll take a loot tomorrow [20:23:49] that's assuming you have a decent softphone setup so that you aren't paying an arm and a leg in long distance [20:26:04] I don't have a phone yet, and no obvious solution to get one. I'm still working on getting the microphone on my laptop to work tbh. But I can make myself useful on a wiki without a phone; I'm a Wikipedian ;) [20:29:51] guillom: do you have a usb headset? [20:30:02] no [20:30:19] but I understand how it could be a solution. Good idea. [20:30:46] that can help a lot, and we're already getting everyone headsets. you can ask Carrie for one, I think [20:31:03] ok [20:31:05] thanks for the info [20:31:29] np, I'll send you mail with some more info [20:31:35] thx [20:35:14] http://rim.jobs/ [20:45:40] oh, a site for working on Blackberry phones....of course! [20:49:42] *robla is reminded of the "Corporate Beavis" concept: http://reverendted.wordpress.com/2006/08/30/products-and-projects-whats-in-a-name/ [20:55:57] robla, i'm seriously struggling with that same concept. [20:56:09] we need to ditch "Liquid Threads" as a name [20:56:25] because a) it's poisoned; b) it's not accurate; c) it's user-unfriendly [20:56:32] and i want to go the "Van Halen" route. [20:57:37] Eddie's original name for the band was something stupid like "Rock Guns" or summat. and david lee roth convinced him to go with "Van Halen" because then he wasn't going to be tied to any specific genre and could just play what he wanted as he evolved. [20:58:42] jorm: so we call it "Harris"? :) [20:59:03] no, maybe Garrett! [20:59:05] wait. the original name was "Mammoth" [20:59:09] heh. [20:59:17] no, i'm thinking something more generic. [20:59:46] i thought "echo" at first, because it's ethereal, but then thought better of the "ethereal" part. i think something more solid. [21:00:26] careful what you wish for in a renaming exercise [21:01:06] "Pending Changes" was pretty brutal, and you just trade one form of grousing for another [21:01:42] you can tell WMF cares about a project again when it's name suddenly changes [21:03:12] heh. [21:03:28] FacebookThreads [21:03:57] oh, yeah. [21:04:01] you guys wanna see it? [21:08:21] Now with 50% less beef. [21:08:31] The name is rarely the issue. [21:08:49] FluidThreads [21:09:33] And let's not pretend that a rename doesn't come with its own set of high costs. Sure, "LiquidThreads" might be poisoned, but at least people have some idea what the hell you're talking about. It's an established brand. [21:09:45] Renaming just muddies the waters. [21:09:54] Hi Roan. [21:10:14] Hi Shirley [21:10:15] I think we're just gonna have to disagree. [21:10:24] Okay. [21:10:36] i agree with you to a point but i've moved beyond a simple discussion system. [21:10:55] MEDIAWIKI 2.0 [21:11:06] You think "LiquidThreads" implies a simple discussion system? [21:11:20] Threads can have features. Like Facebook "Like" buttons and other useless shit???. [21:11:48] *bawolff votes for GelThreads! [21:12:09] The worse named extension is AbuseFilter. [21:12:13] Can we have pokes? [21:12:17] jorm [21:12:18] jorm [21:12:19] jorm [21:12:21] You'd like that, wouldn't you? [21:12:21] Can we? [21:12:23] lol [21:12:26] *Reedy pokes Shirley [21:12:37] I don't think "LiquidThreads" is bad. [21:12:39] yes? [21:12:51] you won't need to have pokes. [21:12:54] Pfft. [21:12:57] Why bother then? :D [21:13:05] SuperPokes then. check [21:13:08] "You have been poked by Jorm (WMF). (poke back | contribs | block immediately)" [21:13:10] <^demon> I'm with Shirley on this one. I don't think LQT is a bad name, and I don't really think it's poisoned either. [21:13:17] <^demon> It hasn't seen enough widespread deployment to be hated. [21:13:52] that's not what i'm hearing from the community side. [21:13:57] but "threads" is a shit word for users. [21:14:40] jorm: how much exposure does the name get in the UI? [21:14:52] indeterminate. [21:15:08] <^demon> It doesn't have to get any exposure other than on Special:Version [21:15:17] nikerabbit and siebrand are gonna hate me, though, since i'm changing all the messages. [21:15:25] Very little [21:15:36] <^demon> eg: the only mention of "FlaggedRevs" on enwiki is the extension name itself on the Special:Version page iirc [21:15:49] Yup [21:16:03] on the EN LQT i18n, there is no mention of liquid threads [21:16:07] just threading/similar [21:16:52] I think it's perfectly reasonable to put on your UX hat and nuke the word "thread" [21:17:00] Liquid? [21:17:03] (from the UI that is) [21:17:14] <^demon> Right, I can agree with removing it from the UI. [21:17:32] i am going to nuke the word thread. [21:17:36] we now have "topics" [21:17:40] possibly "discussions" [21:17:47] here is the big change: [21:17:48] http://elohim.gaijin.com/LQT-v2.5-subscriptions.png [21:17:54] that's your new "subscriptions" page. [21:18:06] you will no longer need to patroll 400 talk pages. [21:18:19] everything you watch will show up in your own, local lqt. [21:18:28] What the hell? [21:18:34] and you can change the view - from "full thread" to a collapsed "only replies" [21:18:37] How does a "UX hat" relate to an extension's name? [21:18:40] The subject you entered is invalid. [21:18:40] God. [21:18:40] It may: [21:18:40] * Be too long, or [21:18:40] * Conflict with interwiki prefixes or namespace names. [21:18:42] Ugh [21:18:45] That message is bad [21:18:56] Why not just tell them what is invalid [21:18:56] That bullshit is back? Make arbitrary aesthetic changes and claim it's "usability!!" so nobody can object? [21:19:14] That was at least half of the Wikipedia Usability Initiative. [21:19:55] i'm so not going to have that argument with you except to say that someone like my father doesn't know what a "thread" is but *does* understand a "topic" [21:19:56] Reedy: Don't allow overly long input; and what's wrong with interwiki prefixes... [21:20:06] linguistic changes are very significant UX changes. [21:20:12] Shirley, I didn't add it. [21:20:27] the extension's name falls into that because UX is a holistic principe. [21:20:32] err, principle. [21:20:37] I doubt your father is going to give a shit what that it's called LiquidThreads. [21:20:42] He's going to click "reply" and type. [21:20:56] UX stands for user experience, you realize? [21:20:57] yes. but it is an issue when doing meta-level discussions. [21:21:11] The users don't care about the holistic aspects of an extension name that isn't in the interface. [21:21:13] what i am designing is far removed from what LQ is today. [21:21:23] Shirley: I was referring to the UI, not the extension name. I think I agree with you that the name of the extension doesn't need to change [21:21:31] at any rate, you're bikeshedding on a pointless problem. [21:21:34] <^demon> +1 [21:21:54] there's another reason, which is that i think we'll end up breaking backcompat. [21:21:58] note that there is a PF #useliquidthreads [21:22:01] It isn't bikeshedding. Renaming has real, actual costs. [21:22:15] Yes, please don't rename deployed extensions again [21:22:16] well, to whom? [21:22:24] To the poor souls that have to deploy it [21:22:26] <^demon> Deployment becomes annoying. [21:22:31] yup [21:22:34] And I don't think repeating the FlaggedRevs clusterfuck should be a goal. [21:22:36] <^demon> Translatewiki has to change their settings as well. [21:22:39] that's my point. i think this would be a fresh deployment. [21:22:44] and not only for WMF, but also for 3rd party users [21:22:52] jorm: are you talking about renaming, or forking? [21:23:05] forking, more like. [21:23:06] jorm: What's the benefit to developers/users/outside parties to a fork? [21:23:16] of a fork * [21:23:28] Shirley: being able to break backwards compatibility is a huge advantage [21:23:29] I understand clean slates. I also understand why they're usually bad. [21:23:35] a new version implies that the old version can be upgraded. i'm not entirely sure yet that will be possible. [21:23:44] Are you gonna be changing the backend significantly? [21:24:06] Also it'd suck if existing LQT installs can't be upgraded [21:24:09] i don't know yet. how i'm thinking says it may be required, if only for performance reasons. [21:24:17] i am aware. trust me, i'm aware. [21:24:36] <^demon> Performance may just mean slapping some extra indexes on :) [21:24:47] And setting $wgMiserMode! [21:25:01] Performance can also mean redesigning the DB schema [21:25:04] <^demon> Miser mode sucks because it disables features :( [21:25:06] Once you end up having to do that, I'm sold [21:25:15] jorm: I'm worried that you're going to invest a lot of time and energy into this, get ignored/have a lot of objections raised, and then get really frustrated. [21:25:16] Most things short of that, please don't rename [21:26:21] RoanKattouw: my hope is that we can write decent upgrade scripts. [21:26:41] a lot of this is going to be new tables, i think. but existing tables may require reworking. [21:27:08] <^demon> That's what the upgrade hooks are for, so we can make upgrades painless to 3rd parties :) [21:27:28] Exactly [21:27:31] You still won't have to rename [21:27:42] jorm: What is this? [21:27:49] And is it being developed anywhere publicly? [21:28:35] i'm surprised that you'd be worried about my frustration level, actually. [21:28:43] what is what? [21:28:53] > a lot of this is going to be new tables [21:29:01] I assume you're talking about new features. [21:29:06] Are those being discussed anywhere publicly? [21:29:30] yeah, right here right now Shirley :) [21:29:32] Well, as fun as train wrecks are to watch, occasionally I like to shout "incoming!" beforehand. [21:29:47] robla: Really? Where's the discussion about things like a "helpful comment" feature? [21:30:07] Or a "like" button. [21:30:15] not on a wiki yet, no. the decision was made to wait until the new year. there are people prepping for a larger conversation. [21:30:16] Or whatever else is being planned. [21:30:21] <^demon> I've heard him asking for feedback on IRC before. [21:30:34] the "helpful" button is just one mechanism for a reputation system. [21:30:41] it's not a done deal; it's thinking out loud. [21:30:42] Oh, and avatars. [21:30:53] <^demon> Avatars had multiple long IRC discussions. [21:30:53] Some of this seems like a done deal to me. [21:30:56] <^demon> And I agreed to drop it. [21:30:56] you really, truly need to achieve zen acceptance that those are going to happen. [21:31:02] avatars? [21:31:09] really? [21:31:14] if only for a short period of time. [21:31:21] <^demon> Bryan: Drop it, it's not worth arguing :) [21:31:22] "OH NO THE HORROR" [21:31:25] jorm: I'm asking who made that decision and why. And if future decisions _like that_ are being discussed now. [21:31:35] until we get real data that they are a help or hinderance. [21:31:36] <^demon> back in a sec, lots of thoughts on this [21:32:19] if you need someone to blame, you can blame me. [21:32:21] I don't really understand why everything has to be fait accompli. [21:32:49] WE WILL HAVE AVATARS. <-- Says who? And why? And where was this discussed? [21:32:51] hrm. something is going down at the gas station. lots of cop cruisers showing up. [21:33:02] Do you own the gas station? [21:33:14] no; it's across the street from me. [21:33:20] Shirley: I really don't understand why when anyone from WMF states an opinion with any degree of force you go on the attack [21:33:44] robla: You don't understand it because it isn't true. [21:33:52] the avatars and the like button were not a joke? [21:33:54] I object to backroom development of community-facing features. [21:34:00] Bryan: No, not a joke. [21:34:26] really... [21:34:30] robla: When someone comes along and says "we're going to avatars!", I guess I can accept that. My next question is "what else is in the pipeline?" [21:34:41] robla: Which nobody can seem to answer right now. [21:34:42] well, I have to agree with you then [21:34:49] wait, the like button was not a joke? [21:35:14] you're shouting that avatars are going to ruin everything and you don't have any data to back your claim. [21:35:27] jorm: Where did I say they were going to ruin everything? [21:35:35] I asked who decided they _must_ be implemented. [21:35:39] and yet, when we say "we're going to get the data to make the decision" you don't accept that. [21:36:07] Yes. There isn't a infinite amount of developer resources, in case you hadn't noticed. [21:36:20] Me. they must be implemented. We must see if "humanizing" the discussion will help to reverse the rather horrible trend in editor bleed. [21:36:23] Gathering the data about stupid features like avatars has very real costs for other projects. [21:36:31] if we don't stem the tide, and soon, that will be it. [21:36:42] Is that a feature or a bug? [21:36:47] it is a bug. [21:36:59] perhaps you haven't been paying attention to the editor trends. [21:37:03] <^demon> Just as a minor note: just because everything wasn't discussed on the lists or on-wiki does not mean backroom development. Lots of developers do local work and ask for feedback and such on IRC. The code always ends up in the public repository. I think there's some feeling that there's secret meetings to decide things when there really aren't. [21:37:07] Bryan, bawolff: It won't be an actual like button, I don't think. More like "4 people think this comment is helpful". [21:37:11] <^demon> And just because I ask for feedback, doesn't mean I'll use it. [21:37:32] this is not a fucking joke, man. it's a serious fucking problem. and people wrap up their shit in WYSIWYG focus and ignore the very fucking real social problems. [21:37:54] jorm: I responded to that post from Tim. I don't think he ever responded back. [21:38:02] no, there aren't any secret meetings, really. [21:38:08] <^demon> And once it's in the public repository, it becomes part of the public review/fix/revert/ok cycle like every body else. [21:38:09] ^demon: When the primary development forums aren't used, you don't think there's a reason to object? [21:38:23] *bawolff is excited for the flame wars that you can like other peoples posts [21:38:28] ^demon: That isn't how it has worked in the past. [21:38:30] <^demon> I think when people have meetings and declare things by fiat that's bad. [21:38:40] ^demon: Which is the story behind avatars. [21:38:45] And other features. [21:39:02] jorm: The meetings aren't secret. They're just private. [21:39:11] And their decisions are apparently final. [21:39:15] <^demon> The usability initiative sucked, yes, I agree. Can we move on from that? [21:39:20] you seem to think this isn't a republic. it is. [21:39:31] and no decision is final. [21:39:39] Except avatars. Those are coming. [21:39:41] this is not a republic [21:39:42] Hell or high water. [21:39:49] we do not vote features [21:40:03] they are coming, if only so we can get enough data to determine that they are not helping. [21:40:11] ^demon: I wasn't really thinkg about the Usbaility Initiative there. [21:40:25] <^demon> The vast majority of features come from the minds of one or two people. [21:40:32] <^demon> Very few get wide discussion before committing. [21:40:37] ^demon: Someone spends several days/weeks working on features under contract for WMF, they get committed, then people object. Is that code/work just going to be thrown out? [21:40:40] Of course not. [21:40:48] bryan: no, voting on features would be a democracy [21:40:51] It's usually changed [21:40:59] <^demon> Yes, it's improved. [21:41:00] <^demon> Refined [21:41:03] Not wholly thrown out [21:41:07] <^demon> Reverted (for now) if broken [21:41:13] <^demon> Just like everybody else. [21:41:24] But if people are ignorant enough to implement something that no one wants, then yes, it'll be reverted or disabled [21:41:26] That assumes the underlying idea has merit. [21:41:32] RoanKattouw: No, it won't. [21:41:39] Not when WMF is paying for it. [21:41:51] I don't know what reality you two are living in... [21:41:59] Try me [21:42:12] At the end of the day, it comes down to "we sign the paychecks, this is going live." [21:42:22] If a staff member puts something idiotic in an area of MW or an ext that I'm familiar with, I will take it out [21:42:45] LQT isn't one of those areas however [21:42:53] you really dont have any idea how the wmf works, i think. [21:43:01] Sorry. I'm new here. [21:43:15] shit, i'm pretty sure that one major feature is about to get scrapped. [21:43:26] Which one is that? [21:43:26] Which feature? Get scrapped by whom? [21:43:40] You talk about "features" as though it's been the subject of conversation in here. [21:43:47] "Oh, y'know, that major one." [21:43:59] I can't be the only one who's completely lost. [21:44:22] <^demon> Oh I know [21:44:26] <^demon> Because of my secret meetings [21:44:29] <^demon> Bwahahahaha [21:44:38] heh. my installer work got reverted, man. [21:44:42] I have an inkling, but I really don't have enough context to know what jorm is referring to [21:44:44] We should have more of those ^demon [21:44:46] this shit happens. [21:44:49] jorm: And you re-reverted it :) [21:44:54] {{BeBold}} [21:44:57] <^demon> And it's better now :) [21:44:59] well, i re-reverted it, fixing it. [21:45:04] Yeah [21:45:19] it still needs an "okay" mark, though. [21:45:24] Rev #? [21:45:27] I seem to remember it being considered an afront that Tim reverted another WMF staffer's work. [21:45:40] That was because jorm missed the CR notif e-mail [21:45:46] So it hit him cold [21:45:47] 77898 ad 77929 [21:45:54] !r 77898 [21:45:54] --elephant-- http://www.mediawiki.org/wiki/Special:Code/MediaWiki/77898 [21:45:57] !r 77929 [21:45:57] --elephant-- http://www.mediawiki.org/wiki/Special:Code/MediaWiki/77929 [21:45:59] yeah, pretty much. [21:46:01] Shirley: you're not the only one who is a bit lost, but you certainly are the only one who seems to think that with this tone people will take your seriously [21:46:03] o_0 [21:46:08] Where did elephant come from? [21:46:10] <^demon> jorm: I'm ok with it conceptually now since the show/hide-ability has been restored [21:46:12] <^demon> And parsing fixed. [21:46:26] Reedy: prototype? It's been here forever? [21:46:33] Bryan: If you think a different approach will work better, feel free to try. But it's mostly shouting at a brick wall. [21:46:37] I've never seen/noticed it before [21:46:43] The elephant in the room. [21:46:44] Reedy: Actually if you count the -usability channel, it's been here for well over a year [21:46:49] It's the logging bot [21:46:53] It's really just an mwbot instance :) [21:47:03] I presumed that :P [21:47:25] It doesn't have a full copy of the bot brain, I initially just set it up for logging [21:47:34] Overkill maybe but meh, it does what it needs to do [21:47:37] jorm: You don't seem to have answered the question about which major feature. [21:48:02] <^demon> I'm +1 on that, without context I'm lost. [21:48:17] RoanKattouw, stuff like !r in here is useful [21:48:23] Totally [21:48:25] And !b [21:48:27] we don't need most of the other stuff hte other one has [21:48:30] So those I did put in [21:48:48] I don't feel at liberty to talk about it yet. [21:48:58] i probably shouldn't have mentioned it, tbh. [21:49:06] Open development, not so much. [21:49:07] <^demon> ...I'm not sure I'm ok with that. [21:49:19] <^demon> Planning to remove a "major feature" [21:49:25] <^demon> Yeah, I'd like context on that. [21:49:26] Has the decision even been made? [21:49:45] RoanKattouw, would you mind dropping the caching time of the MW CR stats on mediawiki wiki? [21:49:50] What decision? [21:49:54] <^demon> ...context, anyone. Please. [21:49:55] Reedy: Which $wg var and to what value? [21:50:00] Shirley: To scrap it [21:50:04] RoanKattouw: What's it? [21:50:15] ^demon: Not so fun to be in the dark, is it... [21:50:28] well. "major feature" is possibly a stretch term. [21:50:29] RoanKattouw, $wgCodeReviewRepoStatsCacheTime = 6 * 60 * 60; // 6 Hours [21:50:32] Is the current [21:50:33] Brandon told me in private, but he said 'probably' [21:50:44] RoanKattouw: No, it hasn't been made. [21:50:50] Right [21:50:56] RoanKattouw, 60 * 60 for 1 hour would be better [21:51:00] jorm: The meetings aren't secret. They're just private. [21:51:12] Right-o... [21:51:17] this wasn't a meeting. it was analysis of data. [21:51:19] I personally don't see why we wouldn't just say it's being considered, but I'll leave that up to Brandon [21:51:29] eh, fine. [21:51:41] RoanKattouw: It's much more fun to pull teeth. [21:51:50] <^demon> I understand pondering removal of a major feature in your head and then asking for feedback. [21:51:57] <^demon> But to hint at it, and not give any context. [21:51:58] AFT isn't giving us different data than we expected. it's being used different than expected, and people want different things from it. [21:51:59] <^demon> Yeah no. [21:52:09] jorm: to be honest, with remarks like "I don't feel at liberty to talk about", you're not dispelling rumours about secrecy [21:52:29] my personal opinion is that we didn't run it on enough pages. [21:53:01] well, it wasn't even a discussion; it was my interpretation of the data and the analysis of it. [21:53:08] <^demon> Being ignored is fun! [21:53:12] WHAT data? [21:53:31] "this clearly isn't doing what it was supposed to, so maybe we need a different thing entirely." [21:53:32] I'm glad it's not just my frustration level that has risen. It's comforting. [21:53:48] there's a ton of data we have, it just doesn't make a lot of sense. doesn't prove much. [21:53:56] And it hasn't been made public? [21:54:04] jorm: Are you talking about turning off the ArticleFeedback tool? [21:54:11] are you deliberately ignoring every request for clarification :/ [21:54:14] (Not that I care about AF data at all, I stick to programming generally. But it sounds useful to publicize) [21:54:17] it's public, roan. [21:54:22] Oh [21:54:23] OK [21:54:23] Public where? [21:54:31] howie posts it on mw.org [21:54:32] Bryan, Shirley: Yes, this is about the ArticleFeedback tool [21:54:40] AFT [21:54:50] <^demon|away> So "removal" means the tool? [21:54:55] <^demon|away> See how much time we could've saved. [21:55:05] disabling, modifying. [21:55:07] <^demon|away> "Removing a major feature" could mean "removing ability to categorize pages" for all I knew. [21:55:20] Well, who would've thought such a tool wouldn't work out. [21:55:22] users don't really care about rating; they care about giving feedback. "this article sucks because x, y, z." [21:55:39] Speaking of poorly named extensions. [21:55:42] :-) [21:55:51] i personally think that the ratings aren't of much use but the tool's possible use as an on-ramp would be valuable. [21:56:09] http://www.mediawiki.org/wiki/Special:Contributions/Howief [21:56:29] http://www.mediawiki.org/wiki/Article_feedback/Public_Policy_Pilot/Early_Data [21:58:39] Whatever provide the greatest opportunity for narcissism is the new measure for success in this Facebook world [22:00:02] oh brother. [22:07:48] hexmode, http://www.mediawiki.org/wiki/Special:Code/MediaWiki/72858