[16:22:02] *guillom 	arranges a few candles, lights up some incense, and begins to chant to summon robla.
[16:37:01] <Krinkle>	guillom: Hi
[16:37:20] <guillom>	Hello Krinkle.
[16:38:32] <Krinkle>	Just wanted to let you know (in case you didn't know) that the multimedia usability draft mock-up is being used as a starting point for the License intergration project
[16:38:34] <Krinkle>	http://usability.wikimedia.org/w/index.php?title=File:GPaumier_multimedia_usability_draft_mock-ups_Oct09.pdf&page=6
[16:38:48] <Krinkle>	visually speking
[16:38:50] <Krinkle>	speaking*
[16:39:34] <Krinkle>	No promises, but I think that's pretty much what it'll look like.
[16:40:29] <Krinkle>	And the one-click buttons have already been made (albeit as a JavaScript, only on Commons)
[16:49:05] <guillom>	Krinkle, I saw that on your specs pages, and I was very, very happy :)
[16:50:30] <Krinkle>	janpaul123: Hi
[16:50:37] <janpaul123>	Hey!
[17:26:28] <jorm>	^demon: ping
[17:26:39] <^demon>	ohnos!
[17:26:47] <jorm>	installer broke in trunk, mang.
[17:27:20] <^demon>	Broke. Like how?
[17:27:31] <^demon>	Whoops
[17:27:32] <^demon>	Yeah
[17:27:35] <jorm>	broke like it complains about not being able to connect to the database when i haven't even made one
[17:27:47] <^demon>	My fault, fixing.
[17:27:49] <jorm>	on the page after the checks, which it passes.
[17:27:54] <jorm>	k
[17:30:17] <^demon>	The duplication between database constructors is really pissing me off :\
[17:30:25] <jorm>	yes.
[17:34:00] <^demon>	jorm: Pick up r80887
[17:34:07] <jorm>	danke shoen.
[17:35:12] <jorm>	hrm.  still happening.
[17:35:38] <jorm>	sec.  lemme get you a pastie or something.
[17:36:21] <jorm>	^demon: http://pastie.org/1493307
[17:37:47] <^demon>	Slap a if( $server ) around the open in DatabasePostgres' constructor
[17:37:52] <^demon>	I'm working on a real fix.
[17:37:54] <^demon>	This is annoying.
[17:38:10] <jorm>	i can wait.
[17:38:20] <^demon>	I mean seriously.
[17:40:50] <jorm>	i know.  it's kind of what was frustrating me when i was trying to collate the various places where graphical elements get created.
[17:51:40] <^demon>	Ok, well that looks way nicer :) Let's test all 3 db types to make sure I didn't break anything
[18:58:16] <flipzagging>	mediastorage people -- apergos, mark, __nelson, wooster, ryan_lane ?
[18:58:28] <flipzagging>	which local ? I'm in x2003
[18:59:34] <flipzagging>	ok, 2002 then
[18:59:46] <flipzagging>	x2002
[19:44:04] <Krinkle>	RoanKattouw: I'm running into a problem writing the special whatleaveshere-ish. I remember during the Hackathon you mentioned something about it, but not sure what is was (should've written it down). Whatleaveshere: would list internal links (pl, tl, im), external links (el), and categories (cl). How to handle the pagination for the end user ?
[19:44:53] <Krinkle>	I guess three seperate (previous | next) lines (like Categories have now for pages, media and subcategories)
[19:45:10] <Krinkle>	put the problem is that 'internal links' isn't 1 module but three (pl, tl, im).
[19:45:12] <RoanKattouw>	Well in my ideal world this'd use 5 (?) different lists
[19:45:15] <RoanKattouw>	Exactly
[19:45:23] <RoanKattouw>	Rather than trying to jam it all together
[19:45:39] <Krinkle>	whatlinkshere solves this by quering all three with the same limit and then (in most cases) ignoring most results.
[19:46:12] <RoanKattouw>	which is kinda stupid
[19:46:14] <Krinkle>	but the main difference is this: whatlinkshere can use a page if for continue.
[19:46:24] <Krinkle>	whatleaveshere cannot since the target page may not always exist.
[19:46:32] <Krinkle>	it needs an ns|title pair
[19:46:34] <RoanKattouw>	Yeah, so you need to use namespace|title there
[19:46:41] <RoanKattouw>	Not a very big deal
[19:47:07] <Krinkle>	but giving users a like like Special:WhatLeavesHere/Foobar?plfrom=8|Something&tlfrom=&imfrom& etc. isn't very nice.
[19:47:43] <Krinkle>	But there ins't really a better way, is there ?
[19:48:44] <Krinkle>	ohwell, it'll do :)
[19:48:57] <Krinkle>	What're you up to by the way ?
[19:54:06] <RoanKattouw>	Now going through local changes on the cluster
[19:54:15] <RoanKattouw>	Figuring out whether to commit them, remove them, or what
[22:06:22] <Krinkle>	Bryan: What do you think of the idea raised on wikitech-l to keep the tables etc. but storing the input in wikitext for history (rather than versioning ourselfs) ?
[22:07:04] <Krinkle>	Personally I think it has advantages, but one problem I see is that the parserfunctions would have to be directly be on the page, not via a template (since that would be to describe the template not the file page)
[22:07:36] <Bryan>	well, I have replied about why I don't really like it, but I need to read through the follow ups
[22:07:44] <Bryan>	Krinkle: <includeonly> magic
[22:07:58] <Krinkle>	so file_props would become the current representation of the properties. And history in wikitext.
[22:08:16] <Krinkle>	I like it now that I think more about it. but I'll read through follow-ups now too.
[22:08:24] <Bryan>	well, in that case we can just move to page_props altogether
[22:11:08] <Krinkle>	Yeah, I was just thinking the same.
[22:11:27] <Krinkle>	that would solve lots of problems actually.
[22:11:55] <Krinkle>	But one small thing (although no big deal since we're not coding an extension) is to tab into the wikitext before it's been 'template expanded'
[22:12:19] <Krinkle>	ie. removing all {{#fileauthor:}} and {{#filelicences:}}
[22:12:52] <Bryan>	I'm not sure I get you
[22:13:05] <Bryan>	but I'm off to bed
[22:13:11] <Krinkle>	I'll reply on the list :)
[22:13:14] <Bryan>	please leave your comments at wikitech-l
[22:13:17] <Bryan>	yeah :)
[23:50:15] *jorm 	invokes Alolita.
[23:50:35] <werdna>	Alolita: method not found
[23:50:36] *werdna 	runs
[23:53:38] <TimStarling>	are we having a meeting?
[23:54:18] <RoanKattouw>	Oh it's 4
[23:54:31] <RoanKattouw>	Heading over now
[23:57:36] <TimStarling>	sip is broken again, can't connect