[17:01:47] do we have a phone call with Katie ? [17:05:33] I think she's in the wikia room [17:19:26] she's around somewhere [17:35:26] hi everyone [17:35:36] can anyone tell me how to create a link to a stub? [18:11:22] nimish_g / mdale: did you track down aude ? [18:12:25] ya I think mdale did [18:12:42] yea made it ;) [18:31:40] brion, clusters have all the data in them now (yay) so we can turn clicktracking on (double-yay) [18:38:29] woot [18:59:40] nkomura: ok we meeting you guys at 21st a? [21:12:14] *werdna waves at TrevorParscal, giggles at his pun [21:16:55] :) [21:21:21] whoa, you're editing it [21:21:33] indeed [21:23:08] continue waving - Adam Miller just showed up [21:23:14] well... [21:23:20] maybe I'm not leaving my desk [21:35:36] TrevorParscal: done some wavin' :) [22:55:26] hi brion [22:56:58] *werdna waveth at aude and brion [22:57:05] parutron_: I see you found my Wave :) [22:57:16] hey [22:58:06] s/brion/nkomura/ [22:58:12] werdna: parutron_ is in the meeting [22:58:28] ah, there's a meeting? [22:59:19] she is meeting with the usability firm whom we are working with for next week's study [23:07:44] ah, another usability study [23:14:00] ok where was i [23:14:05] hi nkomura :D [23:14:16] hi brion [23:14:27] nimish_g has an exciting news for us [23:14:37] \o/ [23:14:51] nimish_g? [23:15:42] nkomura: Do I need to mail the invoice and w2 to you guys? they both look like they require signatures [23:15:59] w2 yes [23:16:21] you can just email the invoice to me [23:16:22] brion, I have all the data in the production machines now, do you think we could turn clicktracking on to some (low) threshhold? [23:16:29] ok [23:16:45] (my exciting news is that clicktracking has all the data ready to roll and I now know all sorts of exciting things about servers) [23:17:03] drum rolling [23:19:02] ok lemme poke it over [23:19:52] okie [23:20:02] brion: if we can run it for 10-15 min that'd be great [23:22:11] nimish_g: why does isUserThrottled return a string 'true' or 'false'? [23:22:35] ah i see [23:22:57] for JS. ClickTracking.js uses the 'true' or 'false' for defining the function or not =) [23:23:23] brion: btw, LiquidThreads will hopefully be going live on the strategy wiki on Monday. [23:23:27] opt-in per-page for now [23:23:30] \o/ [23:23:45] I think I've cleared up all the serious bugs [23:24:01] congrats werdna [23:24:18] :) thx [23:24:19] we can start using LTH on usability wiki too! [23:24:31] nkomura: do you want me to stick it there while I'm at it? [23:24:56] there's still some work to be done, some minor adjustments, maybe a UI tweak or wo [23:24:59] werdna: what would happen if we deploy and revert? [23:25:27] nkomura: you'd lose all of the discussions contributed with LiquidThreads, unless you got me to write a script to turn them into the wikitext format [23:25:55] can i get you write that script first ;) [23:26:23] depends, can I steal trevor for a bit longer sometime soon? :) [23:26:27] or parul [23:26:29] nimish_g: can you add some escaping on SpecialClickTracking::getTimeConstraintsStatement? [23:26:47] while we hope that the input values are clean, their source is widely separate from their use and it's hard to tell [23:27:03] makes it much easier on future maintainers to see use of $db->addQuotes($minTime) here [23:27:26] ok. yeah, SpecialClickTracking is still...a work in progress [23:27:36] :D [23:28:11] yeah it's also heavily using subqueries, so won't work reliably in our mysql 4 environment [23:28:35] can we keep that disabled for now? [23:28:41] yes, please do =) [23:28:53] parul will be super busy until the usability study is over [23:29:07] werdna: you should have me soon [23:29:20] I think I will be working a bit with your stuff in the next week [23:29:25] TrevorParscal: awesome :) [23:29:35] durring usability studies I will be poking at CSS and interactive stuffs [23:29:45] brion, is there a quick and easy way to go "this special page is not viewable by anyone at all yet"? [23:29:58] I'm gonna be making sure everything's shiny for the strategy folks tomorrow (working on moderation and such, mostly) [23:30:04] nimish_g: just comment it out rather than adding it to the special pages list :) [23:30:23] but next week I'm all clear for tooling around with the interface [23:33:00] werdna: ok [23:33:34] whee [23:34:57] ok, brion, it's been commented out [23:35:47] excellent :D [23:35:51] was just testing the same :D [23:40:52] Database returned error "1064: You have an error in your SQL syntax. Check the manual that corresponds to your MySQL server version for the right syntax to use near 'CURRENT_TIMESTAMP, [23:40:52] session_id varbinary(255) NOT NULL, [23:40:52] is_log (10.0.6.21)" [23:41:26] nimish_g: looks like you can't do that in mysql 4 either :) [23:41:46] wow [23:41:51] either trust the timestamp column to default to current time (shudder) or pass in wfTimestamp() when you insert the row [23:42:06] oh [23:42:38] explicit time-stamps on queries == fun [23:42:49] I think I pass in time() [23:42:52] don't forget $dbw->timestamp() [23:43:07] our timestamp storage is abstracted away to $dbw->timestamp( wfTimestamp() ) [23:43:21] it's magical mediawiki time [23:43:26] werdna: you can lose that last wfTimestamp(), i think $db->timestamp() will give you current time :D [23:43:30] internally, for mysql at least, that calls wfTimestamp( TS_DB, $ts ), converting it to a special db timestamp [23:43:33] brion: O RLY [23:43:49] *werdna didn't know that [23:43:53] so you can just pass in $dbw->timestamp() and you're good? ok [23:44:00] trackEvent() is not passing anything in, it's letting it autofill [23:44:20] function timestamp( $ts=0 ) { [23:44:21] return wfTimestamp(TS_MW,$ts); [23:44:21] } [23:44:35] ah, by TS_DB I mean TS_MW [23:44:41] so, passing nothing defaults to 0, which passes 0 to wfTimestamp [23:44:58] TrevorParscal: no, he isn't passing that row at all [23:45:00] also i'm not 100% convinced our database boxes have synced timestamps [23:45:03] he's allowing the table default to fill it. [23:45:46] anyway i strongly recommend against TIMESTAMP columns, they're trouble [23:46:05] werdna: understood, I'm confirming brion saying that $db->timestamp() with no params return timestamp for "now" [23:46:14] the first such column auto-updates to the current timestamp whenever you change something else in the row [23:46:19] makes maintenance hell [23:46:21] just use a char(14) :) [23:47:47] does less-than and greater-than logic work if you make it a char(14) tho? like can you go "select where time < 20090815'? [23:48:20] yep [23:48:26] but you really should be quoting that anyway [23:48:45] right, I meant where time < `20090815` [23:49:16] ok [23:49:52] yups [23:49:57] we use those a lot actually :D [23:50:15] without the quotes it's sloooow because it casts them all to ints [23:52:01] ok it's a char(14) now [23:56:34] whee [23:57:39] *brion pokes