[00:06:32] <werdna>	TrevorParscal: sweet. Maybe I should talk to somebody to figure out what my schedule is :p
[00:06:44] <werdna>	TrevorParscal: sit QUIETLY may be difficult.
[00:07:32] <guillom>	hahaha
[00:07:34] <guillom>	ok
[00:07:40] <guillom>	it's been a long day
[00:07:42] <werdna>	TrevorParscal: Option 2 is obviously better
[00:07:43] <guillom>	see you later folks
[00:08:00] <werdna>	for versioning
[00:10:44] <TrevorParscal>	...
[00:12:03] <TrevorParscal>	werdna: yeah, although we're still unsure how we plan to "detect" when a file has changed other than checking all the mtime stats each load (or every so many loads)
[00:12:42] <TrevorParscal>	a cron job is cool, but it needs to work with or without cron jobs (3rd party instlalls)
[00:13:26] <werdna>	You should ask Tim or Domas how expensive those stat() calls would be.
[00:13:32] <TrevorParscal>	RoanKattouw: so, we should play with distributed pair programming again sometime...
[00:13:52] <werdna>	Another option is invalidating on sync
[00:14:02] <RoanKattouw>	TrevorParscal: Yeah I was thinking about that
[00:14:15] <TrevorParscal>	sync?
[00:14:19] <TrevorParscal>	like, svn up?
[00:14:30] <RoanKattouw>	We have that Eclipse plugin
[00:14:46] <RoanKattouw>	But then didn't we decide I'd be writing the server and you'd be writing the client code? Although they kinda do touch
[00:15:02] <RoanKattouw>	Or  I guess I could /drive/ while writing the server code and vice versa
[00:15:09] <TrevorParscal>	right
[00:15:27] <TrevorParscal>	we don't have to always use it, but we should explore using it
[00:15:42] <TrevorParscal>	I think we might be quite productive using dpp
[00:16:16] <TrevorParscal>	i also think if it works well for us, it might be an interesting trend to start at WMF
[00:16:19] <werdna>	TrevorParscal: no, like sync-file
[00:16:35] <TrevorParscal>	that's a custom WMF cluster thing right?
[00:16:41] <werdna>	TrevorParscal: of course, that's wmf-specific, but you could just fall back to stat()s on other sites
[00:16:47] <TrevorParscal>	right
[00:17:08] <werdna>	basically I would think that using stat() could be the default, and then you could turn that off and invalidate manually with custom scripts if you liek performance
[00:17:37] <TrevorParscal>	right, and... we like performance... :)
[00:18:30] <TrevorParscal>	so, how do we handle making sure the dynamic loader requests have the right versions? we send a version number with the loader code?
[00:18:36] <TrevorParscal>	seems simple enough...
[00:19:45] <TrevorParscal>	RoanKattouw: we also have not speced out how the loader handles modules who's source is in the database, like user scripts, gadgets and site scripts.... and how they localize
[00:19:52] <TrevorParscal>	we might want to spec that out
[00:20:07] <werdna>	TrevorParscal: what do you mean?
[00:20:16] <werdna>	oh, right
[00:20:34] <TrevorParscal>	how do those modules get registered
[00:20:36] <werdna>	TrevorParscal: problem with this is that if you put the version number in the HTML, you have to invalidate the HTML to bump the version number
[00:20:47] <TrevorParscal>	is one issue
[00:21:05] <TrevorParscal>	werdna: yes, that's why option 1 is ... an option
[00:21:27] <TrevorParscal>	it solves that.. but it has it's own drawbacks
[00:22:08] <RoanKattouw>	TrevorParscal: Have an option to specify a callback instead of a file as a script source?
[00:22:10] <werdna>	like waiting 8 hours for your JS to go live
[00:22:32] <RoanKattouw>	Tim was talking about 1-hour expiries
[00:22:39] <werdna>	well, whatever
[00:22:40] <TrevorParscal>	RoanKattouw: what does the callb
[00:22:49] <TrevorParscal>	*callback do?
[00:22:51] <werdna>	why do you need a version number in the html anyways?
[00:22:55] <werdna>	oh right, because of client side caches
[00:22:56] <RoanKattouw>	TrevorParscal: Return JS
[00:23:04] <werdna>	server-side caches can just be purged, of course
[00:23:12] <RoanKattouw>	Yes
[00:23:17] <werdna>	you could give the client-side cache a ten-minute expiry and purge the server-side cache
[00:23:23] *werdna 	shrugs
[00:23:25] <TrevorParscal>	why not just allow the path to be int:User:myScriptIsCool.js or something
[00:23:35] <TrevorParscal>	if it starts with int: it's in the db, not the fs
[00:23:38] <RoanKattouw>	Well 10 mins is very short, but yeah, we have Squid to take a lot of work out of our hands
[00:24:02] <RoanKattouw>	TrevorParscal: array( 'file' => 'somefile.js' ) vs array ('foo' => 'bar' ) ?
[00:24:17] <TrevorParscal>	then we need to introduce things like MediaWiki:Loader.js and User:Jo
[00:24:30] <TrevorParscal>	User:JohnDoe/loader.js
[00:24:32] <TrevorParscal>	or something
[00:25:01] <TrevorParscal>	RoanKattouw: yes, we can vary the key, not encode the value
[00:25:06] <TrevorParscal>	that makes sense
[00:25:18] <TrevorParscal>	but we'd need to check some well-known places for loader scripts
[00:25:48] <TrevorParscal>	like the 2 places I said, plus somehow hook in gadgets...
[00:27:39] <TrevorParscal>	dude, where are gadgets stored?
[00:27:50] <RoanKattouw>	MediaWiki:Gadget-Foo.js I think
[00:27:57] <TrevorParscal>	hmmm
[00:28:16] <TrevorParscal>	so, we can lump them in with site scripts
[00:28:22] <TrevorParscal>	they are just fancy site-scripts
[00:30:57] <TrevorParscal>	hmm, the MediaWiki:Loader.js and User:JohnDoe/loader.js only help with dependencies...
[00:31:04] <TrevorParscal>	we still need registration...
[00:34:32] <TrevorParscal>	or maybe not...
[00:41:01] <TrevorParscal>	hi Ryan_Lane
[00:41:13] <Ryan_Lane>	TrevorParscal: hi there
[00:41:26] <TrevorParscal>	brb - getting wife from downstairs
[00:41:31] <Ryan_Lane>	ok
[00:45:21] *werdna 	can't find his keys
[00:45:24] <werdna>	hey Ryan_Lane
[00:45:36] <Ryan_Lane>	werdna: howdy
[00:47:18] <Ryan_Lane>	werdna: got your trip all set?
[00:47:22] <werdna>	yeah, finally
[00:47:27] <werdna>	except somewhere to sleep in SF
[00:47:32] <werdna>	but worst case i hear the office has comfy couches
[00:48:09] <Ryan_Lane>	the streets apparently aren't bad either.
[00:48:33] <Ryan_Lane>	you going to SF before wikimania?
[00:48:37] <werdna>	depends, i'd need cardboard for a sign
[00:48:38] <werdna>	yeah
[00:48:40] <werdna>	brb, getting noms
[00:48:42] <Ryan_Lane>	k
[00:48:46] <werdna>	assuming i can find my damn keys
[00:57:59] <RoanKattouw>	OK I guess Tim is not coming
[00:58:06] <RoanKattouw>	And maybe I shouldn't have stayed up till 3
[00:58:10] <RoanKattouw>	Good night all
[02:01:26] *werdna 	returns
[02:01:29] *werdna 	rewaves Ryan_Lane 
[03:11:20] <atelaes>	Hey, can I report a bug in the Vector code here?
[03:11:38] <atelaes>	...or am I totally in the wrong place?
[03:13:31] <Amgine>	Unfortunately, it's after working hours in North America, Europe. I dunno about werdna or others who may be in Australia.
[03:16:15] <werdna>	wazzap?
[03:16:20] <werdna>	but I'm not a usability person
[03:17:23] <atelaes>	Well, the toolbars on the left aren't collapsible for me, though I know they were before.
[03:17:23] <Amgine>	Atalaes is finding an issue with the navigation not being collapsible, Ubuntu, FF, 3.6.3
[03:17:38] <atelaes>	And something in Vector keeps throwing an error.
[03:17:51] <atelaes>	Error: missing ] after element list
[03:17:52] <atelaes>	Source File: http://bits.wikimedia.org/w/extensions/UsabilityInitiative/Vector/Vector.combined.min.js?282f
[03:17:54] <atelaes>	Line: 3, Column: 41
[03:17:55] <atelaes>	Source Code:
[03:17:57] <atelaes>	var mod={'browsers':{'ltr':{'msie':[['>='.7]],'opera':[['>=',9.6]],'konqueror':[['>=',4.0]],'blackberry':false,'ipod':false,'iphone':false,'ps3':false},'rtl':{'msie':[['>=',7]],'opera':[['>=',9.6]],'konqueror':[['>=',4.0]],'blackberry':false,'ipod':false,
[03:18:08] <atelaes>	I wondered if the two might be related.
[03:18:23] <atelaes>	The msie bit seems to be missing a comma.
[03:18:35] <Amgine>	<blames, uhm, AryehGregor, or Guillom...>
[03:19:19] <Amgine>	No particular reason, those names just happened to be there.
[03:19:23] <guillom>	what am I blamed for, now?
[03:19:24] <guillom>	ah
[03:19:29] <guillom>	good, so I don't have to care
[03:19:33] <Amgine>	<laughs>
[03:20:10] <atelaes>	....bearing in mind that I'm a technoretard, so I could be way off here.
[03:21:00] <guillom>	I'm not a developer, and I don't even have commit access, so I'm arguably one of the people who can help you the least
[03:21:39] <Amgine>	Hey, random choices of names in channel. Almost never works, but when it does it's really cool.
[03:25:41] <werdna>	try Ryan_Lane maybe
[04:17:25] <Ryan_Lane>	werdna: eh? try me for what?
[04:19:31] <Ryan_Lane>	ah. for the error in vector... I didn't do any frontend work :)
[04:37:40] <werdna>	hey TrevorParscal
[04:37:44] <TrevorParscal>	hi
[04:37:54] <werdna>	TrevorParscal: there was some bug report in channel earlier
[04:38:02] <TrevorParscal>	oh yeah?
[04:38:05] <TrevorParscal>	i can scan the backlog
[04:38:06] <werdna>	a stray comma or something
[04:38:12] <TrevorParscal>	hmm
[04:38:13] <TrevorParscal>	k
[04:38:18] <werdna>	03:18 < Amgine> Atalaes is finding an issue with the navigation not being  collapsible, Ubuntu, FF, 3.6.3
[04:38:21] <werdna>	03:18 < atelaes> And something in Vector keeps throwing an error.
[04:38:24] <werdna>	03:18 < atelaes> Error: missing ] after element list
[04:38:27] <werdna>	03:18 < atelaes> Line: 3, Column: 41
[04:38:28] <werdna>	03:18 < atelaes> Source Code:
[04:38:28] <werdna>	03:18 < atelaes> var
[04:38:28] <werdna>	mod={'browsers':{'ltr':{'msie':[['>='.7]],'opera':[['>=',9.6]],'konqueror':[['>=',4.0]],'blackberry':false,'ipod':false,'iphone':false,'ps3':false},'rtl':{'msie':[['>=',7]],'opera':[['>=',9.6]],'konqueror':[['>=',4.0]],'blackberry':false,'ipod':false,
[04:38:33] <TrevorParscal>	yeah
[04:38:40] <TrevorParscal>	Roan patched in the msie part
[04:38:47] <TrevorParscal>	and screwed up the comma
[04:38:52] <TrevorParscal>	and I thought he fixed it..
[04:39:01] <TrevorParscal>	vecause someone noticed it like, right away
[04:39:10] <TrevorParscal>	that was around noon or something
[04:40:16] <TrevorParscal>	http://prototype.wikimedia.org/logs/%23wikipedia_usability/20100617.txt at [21:19:35]
[04:40:52] <TrevorParscal>	did roan not patch that?
[04:40:58] <TrevorParscal>	cause if he didn't, maybe you could?
[04:41:02] <TrevorParscal>	it's a production hack
[04:41:19] <TrevorParscal>	the SVN has code that reorgs the borwser testing code
[04:41:30] <TrevorParscal>	so, he avoided doing a more complex merge and just hacked it
[04:41:34] <werdna>	what do I have to do?
[04:41:50] <werdna>	do you have commit access to branches/wmf?
[04:42:27] <TrevorParscal>	well, that file http://bits.wikimedia.org/w/extensions/UsabilityInitiative/Vector/Vector.combined.min.js?somecrazyversionstring
[04:42:35] <TrevorParscal>	it looks fixed
[04:42:49] <werdna>	maybe he didn't get the version bump
[04:42:52] <TrevorParscal>	so it's a caching issue
[04:43:01] <TrevorParscal>	with bits, we have no purge
[04:43:07] <TrevorParscal>	you just have to version bump
[04:43:12] <TrevorParscal>	so, yeah - that must be why
[04:43:13] <werdna>	I can do that
[04:43:18] <TrevorParscal>	thanks man!
[04:44:02] <werdna>	done
[04:44:07] <TrevorParscal>	werdna: i'm writing a proof of concept, for.. mediawiki 2.0 let's call it..
[04:44:17] *werdna 	eyes TrevorParscal suspiciously
[04:46:17] <TrevorParscal>	i have things divided up into systems, which are essentially admin stuff (command line interface) a front-end system (rendering html pages and including all that css and js stuff) and a data access system which can be called via HTTP (like an API) or just directly via PHP
[04:46:48] <TrevorParscal>	it's a layered approach, lots of experiements, some great, some strange, whatever, it's a big science experiment
[04:46:51] <TrevorParscal>	so, the point here is
[04:47:28] <TrevorParscal>	I'm thinking about how to name these different systems... or entry points... or.. layers.. or whatever they are..
[04:47:58] <TrevorParscal>	i have admin, client, server and core, which has all the shared code (most of it).
[04:48:09] <TrevorParscal>	but the client server thing is a little... misleading probably
[04:48:17] <werdna>	This is why you're never allowed to work on LiquidThreads.
[04:48:25] <TrevorParscal>	ha ha ha
[04:48:29] <werdna>	It's so poorly factored that your first action would be to rewrite it entirely.
[04:48:38] <TrevorParscal>	true
[04:48:41] <TrevorParscal>	I will admit that
[04:49:16] <TrevorParscal>	so, do you think admin, view, query, core would be more logical?
[04:49:27] <TrevorParscal>	view being the front end stuff, query being the data access stuff
[04:49:51] <werdna>	hmm, seems reasonable i guess
[04:50:33] <TrevorParscal>	am I asking a person with a high tollerance for poorly factored code and a low tollerence for fussing about with naming things?
[04:50:46] <werdna>	You're asking a distracted person
[04:50:50] <TrevorParscal>	no worries
[04:50:55] <werdna>	I'm doing some consulting stuff :)
[04:51:01] <TrevorParscal>	I'm throwing allot at you...
[04:51:06] <TrevorParscal>	sorry..
[04:51:14] <werdna>	it's cool.
[04:51:14] <TrevorParscal>	thanks for listening to my madness though
[04:51:28] <werdna>	hehe, does that mean I don't have to say anything?
[04:51:31] <werdna>	in which case, go right ahead
[04:51:39] <TrevorParscal>	ha ha...
[04:51:54] <TrevorParscal>	you need to say "wow, that's brilliant", or "I love that!"
[04:52:00] <TrevorParscal>	about every 15 lines
[04:52:07] <TrevorParscal>	you can also just write a bot to do that
[04:52:13] <TrevorParscal>	SelfEsteemBot is born
[04:52:21] <werdna>	;-)
[04:52:31] <TrevorParscal>	i will let you go
[04:52:50] <TrevorParscal>	i just like to type out loud i think... (broken metaphors are fun)
[04:53:08] <TrevorParscal>	thanks again for the patch in deployment
[04:53:33] <werdna>	no problem
[04:53:38] <werdna>	I didn't mean to dismiss you entirely ;-)
[04:56:59] <TrevorParscal>	dude, no worries
[04:57:17] <TrevorParscal>	usually I bounce things off of Roan, barring that... my 3 year old daughter..
[04:57:21] <werdna>	I've been examming for about two weeks, it's been fun :)
[04:57:23] <werdna>	hahaha
[04:57:25] <TrevorParscal>	at least I get the feeling you understand me :)
[04:57:27] <werdna>	poor girl
[04:57:39] <werdna>	her first words will be "better factored"
[04:57:46] <werdna>	well, she probably already talks
[04:57:48] <werdna>	but still
[04:58:05] <TrevorParscal>	it's probably leaving her with scars that will never heal
[04:58:32] <TrevorParscal>	my wife often tells me "Stop refactoring that", and I'm not writing code... i will be like... cooking or cleaning..
[04:58:47] <TrevorParscal>	usually organizing things... to no end
[04:59:11] <TrevorParscal>	so, you are in SF for like.. 1 or 2 weeks?
[04:59:39] <werdna>	TrevorParscal: 23rd to the 2nd
[04:59:54] <werdna>	disappearing to london at 6:55 on the 2nd, arrive in at 9am on the 23rd
[05:00:01] <werdna>	18:55, that is.
[05:00:03] <werdna>	not 06:55
[05:00:45] <TrevorParscal>	right on
[05:00:53] <TrevorParscal>	well, we're excited to have you
[05:00:58] <werdna>	I'm excited to be back
[05:02:48] *TrevorParscal 	goes back to writing code that thinks about how to write code that writes templates and code that uses those templates
[05:03:11] <TrevorParscal>	(just kidding...)
[05:03:46] <TrevorParscal>	omg - my wife made the best flan ever
[05:06:09] *TrevorParscal 	becomes uncomfortable with view and query being verbs.... tries to think of nouns better than client and server
[05:06:35] <werdna>	flan?
[05:07:28] <TrevorParscal>	yes, it's a spanish desert custard covered in a carmel sauce
[05:07:44] <TrevorParscal>	http://en.wikipedia.org/wiki/Flan
[05:08:28] <TrevorParscal>	not popular in your neck of the woods?
[05:09:16] <werdna>	is carmel a typo or a spelling difference?
[05:09:20] <werdna>	out of interest.
[05:09:44] <werdna>	possible, I'm not familiar with it
[05:09:48] *werdna 	shrugs
[05:10:10] <TrevorParscal>	typo
[05:13:31] <TrevorParscal>	k
[05:13:35] <TrevorParscal>	going to play with kids
[05:13:38] <TrevorParscal>	thanks for the chat
[05:13:43] <TrevorParscal>	have a good ... day?
[05:13:47] <TrevorParscal>	afternnoon there yes/
[05:14:02] <TrevorParscal>	anyhoo...
[05:14:04] <TrevorParscal>	enjoy!
[05:14:08] <werdna>	yes, 3pm
[14:07:35] <lecartia>	Morning guys.
[16:41:51] <TrevorParscal>	RoanKattouw:
[16:41:54] <TrevorParscal>	how's it
[17:05:08] <RoanKattouw>	TrevorParscal: Back now, got time
[17:05:14] <TrevorParscal>	sweet
[17:05:17] <TrevorParscal>	i'm at work now
[17:05:18] <RoanKattouw>	Laptop shut down during dinner
[17:05:21] <TrevorParscal>	so, i'm less distracted
[17:05:23] <TrevorParscal>	no worries
[17:09:09] <TrevorParscal>	so, what did you get out of your conversation with Tim?
[17:09:45] <RoanKattouw>	Well it turns out I largely misunderstood his proposal
[17:09:50] <TrevorParscal>	k
[17:10:09] <RoanKattouw>	What he's proposing is that we still use style versions and month-long client-side expiry times, like we do now
[17:10:20] *TrevorParscal 	doesn't remember asking for proposals... only requirements... :|
[17:10:29] <RoanKattouw>	Except that the style versions are timestamp-based (so no manual updates required)
[17:10:43] <RoanKattouw>	And that hour-long expiry he mentions is on the Squid side
[17:11:02] <TrevorParscal>	ok
[17:11:02] <RoanKattouw>	Squid would revalidate URLs from the loader every hour, where the loader may respond with a 304
[17:11:13] <TrevorParscal>	so, what do you think of that?
[17:11:24] <RoanKattouw>	TrevorParscal: I know, but on balance I'm happy that we got implementation proposals despite not asking for them
[17:11:33] <RoanKattouw>	Well
[17:11:40] <TrevorParscal>	I think if that's something that is feesable to develop and makes our ops people happy, I'm cool with it, I don't have lots of passion in this particular area
[17:12:16] <RoanKattouw>	It's not as nice as I would've liked it to be, but any such system would ease the developers' lives at the expense of pretty much everything else that's good in this world
[17:12:46] <Eloquence>	RoanKattouw: I've seen a few reports that the foreground color is not defined. this seems to be fixed in SVN but not deployed to http://bits.wikimedia.org/skins-1.5/vector/main-ltr.css ?
[17:13:15] <Eloquence>	eek.
[17:13:45] <TrevorParscal>	looking
[17:14:14] <RoanKattouw>	Seriously, I want programs to have feelings so I can hurt my copy of Firefox
[17:14:26] *RoanKattouw 	makes mental note to look into upgrading
[17:14:39] <RoanKattouw>	Eloquence: Seen those reports. Not aware of fix in SVN though
[17:14:51] <Eloquence>	https://bugzilla.wikimedia.org/show_bug.cgi?id=20737
[17:15:59] <RoanKattouw>	!r 65380
[17:15:59] --elephant--	http://www.mediawiki.org/wiki/Special:Code/MediaWiki/65380
[17:16:12] <RoanKattouw>	Looks like that never got deployed
[17:16:21] *RoanKattouw 	deploys
[17:17:15] <Platonides>	how can a two-months old revision never get deployed, when more recent ones have?
[17:17:21] <Platonides>	that's asking for problems
[17:17:54] <RoanKattouw>	I deploy UI trunk regularly
[17:18:03] <RoanKattouw>	Intending to that again today after I've reviewed recent revs
[17:18:12] <RoanKattouw>	But Vector skin = core = very infrequent updates
[17:18:19] <RoanKattouw>	If you don't like the latter, complain to Tim ;)
[17:18:25] <Platonides>	RoanKattouw, and it's a good thing you do so
[17:18:37] <RoanKattouw>	I do deploy Vector fixes when I'm aware of their need
[17:18:41] <RoanKattouw>	But sometimes stuff slips through
[17:19:08] <RoanKattouw>	Crap
[17:19:09] <RoanKattouw>	I need a root
[17:19:10] <Platonides>	I think I sounded too harsh
[17:19:15] <^demon>	Speaking of Vector, I kind of don't like r68177 (SimpleSearch disable pref). You defined a preference in an extension but implemented it in core :\
[17:19:19] <^demon>	RoanKattouw: ^^^
[17:19:34] <RoanKattouw>	^demon: mhhh, yes
[17:19:37] <RoanKattouw>	Good point
[17:19:47] <RoanKattouw>	Could you fixme+comment that?
[17:20:06] <Ryan_Lane>	are these extensions going to get moved to core at some point?
[17:20:28] <TrevorParscal>	^demon: simple search is not only an extension
[17:20:41] <TrevorParscal>	you misunderstand the feature - no offense
[17:20:52] <TrevorParscal>	it's actually a part of the vector skin
[17:21:06] <TrevorParscal>	but, the Vector extension includes a module that improves the simplified search
[17:21:10] <Platonides>	then it shouldn't be defined in an extension :)
[17:21:20] <TrevorParscal>	with a suggestions interface and a label that dissapears
[17:21:26] <TrevorParscal>	Platonides: why not?
[17:21:36] <^demon>	TrevorParscal: That's not what I was saying....I was just saying that the implementation of the "disable simple search" was part in core and part in an extension. And that's icky.
[17:21:52] <TrevorParscal>	the skin can render with simplified search, with or without the extension which enahnces it
[17:22:08] <TrevorParscal>	^demon: agreed
[17:22:12] <TrevorParscal>	I have not reviewed that yet
[17:22:18] <RoanKattouw>	^demon: Yes, it's reverse crossover, will fix
[17:22:29] <TrevorParscal>	^demon: thanks for pointing that out
[17:22:32] <TrevorParscal>	how's it going btw?
[17:22:33] <^demon>	No problem :)
[17:22:46] <^demon>	Doing fine. Left Capital One early today.
[17:22:56] <^demon>	Crappy 9am meeting on a Friday...should've have had to go in at all today
[17:23:38] <^demon>	Who schedules meetings for 9am on Fridays? Seriously :)
[17:25:29] <Platonides>	seems the div adding revision wasn't merged either
[17:25:34] <Ryan_Lane>	^demon: at my day job, people don't schedule meetings on fridays :). compressed schedules make everything better
[17:25:48] <TrevorParscal>	^demon: evil people
[17:26:18] <RoanKattouw>	Platonides: Which one is that?
[17:26:33] <TrevorParscal>	hi Ryan_Lane
[17:26:43] <Ryan_Lane>	TrevorParscal: hi there
[17:29:46] <TrevorParscal>	we're working on creating a resource loading system
[17:29:46] <Platonides>	RoanKattouw, the footer part of r66646 seems to have not been merged
[17:29:50] <TrevorParscal>	that doesn't suck
[17:29:54] <TrevorParscal>	we hope
[17:30:10] <TrevorParscal>	for javascript and css
[17:30:18] <Platonides>	seems the upper part was merged, though :/
[17:30:36] <TrevorParscal>	Ryan_Lane: what are you up to?
[17:30:54] <Ryan_Lane>	TrevorParscal: making vms for analytics and donation development
[17:30:59] <RoanKattouw>	Platonides: What is the upper and what is the lower part?
[17:31:09] <Ryan_Lane>	TrevorParscal: why's that? need something?
[17:31:51] <TrevorParscal>	Ryan_Lane: no, just curious
[17:32:23] <Ryan_Lane>	I need to fix a couple selenium things soon too :)
[17:32:41] <RoanKattouw>	adam_miller: OMG bolding in SimpleSearch works on trunk!
[17:32:47] <RoanKattouw>	AWESOME
[17:33:10] <adam_miller>	i got it working last week, i think?
[17:33:26] <RoanKattouw>	I know, I'm slow
[17:33:38] <RoanKattouw>	I'm gonna review all UsabilityInitiative revs today and hopefully deploy trunk
[17:33:48] <RoanKattouw>	To keep the difference between deployment and trunk small
[17:34:03] <RoanKattouw>	Especially useful now that Trevor and I are about to embark on this resource loader adventure
[17:35:01] <adam_miller>	i committed my idea of how the scrolling through suggestions should work today too
[17:35:22] <RoanKattouw>	Yeah I saw that in the commit log
[17:35:25] <RoanKattouw>	Haven't svn up'ed it yet
[17:35:28] <adam_miller>	actually i should get that on a sandbox and email people to make sure everyone else likes my way of doing things
[17:36:09] <RoanKattouw>	^demon: r68224
[17:36:52] <RoanKattouw>	adam_miller: You have the required access and skill to make that happen, right? If so, go for it
[17:37:21] <adam_miller>	i do
[17:37:38] <adam_miller>	that was more of a thinking out loud kinda thing
[17:39:50] <^demon>	RoanKattouw: Looks good thanks :)
[17:47:04] <vvv>	Hah
[17:47:53] <vvv>	Our trolls claim that line between W, omega and И in new logo is a swastika
[17:48:10] <TrevorParscal>	wow
[17:48:32] <Eloquence>	that finally proves that the usability initiative is worse than hitler
[17:48:50] <RoanKattouw>	OK I *kinda* see that just a little bit
[17:49:37] <Platonides>	I'm still waiting to see the original file for the new logo
[17:50:23] <Platonides>	and mmh... the swastika has the arms in the other direction
[17:50:45] <Platonides>	oh, there are swastikas in both sense
[17:52:11] <Platonides>	but nazi's ones go in opposite to the logo
[17:52:16] <TrevorParscal>	Platonides: original file? it's not hidden
[17:52:22] <TrevorParscal>	the SVG is on commons
[17:53:57] <Platonides>	but the svg is derived from another file
[17:54:17] <Platonides>	thus, changes shouldn't go to the svg but to the first one
[17:54:32] <Platonides>	which isn't even in an open format, afaik
[17:55:41] <Platonides>	it wouldn't be strange that in a acouple of years that file got lost too, like nohat's original one
[17:56:21] <TrevorParscal>	it's all SVG
[17:56:25] <TrevorParscal>	you mean the 3D file?
[17:57:23] <Platonides>	yes
[17:57:41] <Platonides>	isn't the svg an export of the 3D?
[17:58:39] <TrevorParscal>	i'm unsure if it was a 2D vector rendering of 3D graphics, or done by hand based on a raster rendering of 3D graphics.
[17:58:52] <TrevorParscal>	but the original 3D model (which I have a copy of) is in Maya format
[17:59:17] <TrevorParscal>	the reason I have a copy, is because I'm trying to get it into blender format
[17:59:22] <TrevorParscal>	in my spare time
[17:59:40] <TrevorParscal>	and it's proving to be a pita
[18:00:40] <vvv>	Is it OK if I store a preference in serialize()'d string array form?
[18:01:01] <RoanKattouw>	Ouch
[18:01:02] <RoanKattouw>	Why?
[18:01:15] <Platonides>	it would be nice having it in blender, Trevor
[18:01:34] <RoanKattouw>	You do realize the entire preferences system was rewritten exactly because it was storing stuff as serialized blobs?
[18:02:06] <TrevorParscal>	vvv: why would you need to do that?
[18:02:08] <vvv>	RoanKattouw: well, I intend to store user-configured toolbar in multidimensional arrays. I have no better idea of doing it
[18:02:09] <Platonides>	vvv, do you have a good excuse for doing that ?
[18:02:24] <Platonides>	user-configured toolbar?
[18:02:26] <RoanKattouw>	Right
[18:02:33] <Platonides>	wouldn't that be a javascript array string
[18:02:35] <Platonides>	?
[18:02:48] <RoanKattouw>	I kinda see how you don't want to store that as a gazillion prefs
[18:02:54] <TrevorParscal>	toolbar confiurations should probably always exist as a json-ish thing
[18:03:14] <TrevorParscal>	which i guess is a form of serialization
[18:03:29] <^demon>	Toolbar configuration shouldn't require an understanding of Javascript ;-)
[18:03:33] <TrevorParscal>	it's a textual representation of a data structure
[18:03:53] <TrevorParscal>	^demon: why?
[18:04:01] <TrevorParscal>	oh...
[18:04:05] <TrevorParscal>	you are joking
[18:04:23] <vvv>	TrevorParscal: I intended to do a dynamical rebuilding of toolbar in PHP script
[18:04:39] <TrevorParscal>	are you serious?
[18:04:40] <^demon>	TrevorParscal: No, I wasn't. I happen to think that ordinary folks should be able to do it.
[18:05:07] <TrevorParscal>	ordinary? the toolbar isn't written in ASM
[18:05:22] *^demon 	facepalms
[18:05:24] <TrevorParscal>	what is the advantage of the toolbar being built in PHP?
[18:05:47] <^demon>	Configuration can move to LocalSettings.php, like everything else.
[18:05:56] <TrevorParscal>	what configuration?
[18:06:00] <^demon>	Configuration of toolbars.
[18:06:12] <Platonides>	that preference modified only by javascript enabled browsers seems sensible
[18:06:16] <TrevorParscal>	what kind of stuff are you trying to configure though?
[18:06:26] <Platonides>	as far as non-js browsers can reset it
[18:06:38] <vvv>	Platonides: toolbar is constructed from JS
[18:06:57] <TrevorParscal>	it's only useful (no matter how it's constructed) to clients with JS
[18:07:02] <RoanKattouw>	vvv: Why not just write JS to [[User:Username/toolbarconfig.js]] and have PHP add a <script src=""> tag for that if it exists?
[18:07:19] <TrevorParscal>	so, by building it out in JS, we have a full API on the client, so userscripts can modify it
[18:07:27] <vvv>	RoanKattouw: because then we'll have to parse it?
[18:07:28] <Platonides>	vvv, then why do you want it in php?
[18:07:28] <^demon>	RoanKattouw: Then nobody could ever have a skin called toolbarconfig
[18:07:37] <vvv>	TrevorParscal: preferences are stored on server anyway, so instead of giving out all available toolbar buttons and list of toolbar button the user use
[18:07:41] <TrevorParscal>	who's more likely to want to tweak the thing, a user writing a userscript, or a sysadmin writing a LocalSettings.php file?
[18:08:18] <^demon>	I think the guy setting up the wiki. And (to me) it makes more sense to put this configuration in LS with all the other settings.
[18:08:33] <^demon>	But that's just me.
[18:09:12] <TrevorParscal>	so, here's the deal.. Mozilla used to be into making "themes", which moved things around, made buttons look different, etc.
[18:09:30] <TrevorParscal>	then one day, they realized that most of the "themes", neigh, all of the "themes" sucked
[18:09:34] <TrevorParscal>	there were 2 issues
[18:10:15] <TrevorParscal>	1, they would spend all their time improving the UI, testing users interacting with it, running statistics on what people clicked, etc., and that data fed into making the default really good
[18:10:41] <TrevorParscal>	any theme some random person cooked up would never live up to that level of attention to detail and testing by fire
[18:11:17] <TrevorParscal>	2, the people making these "themes" were usually doing it to mimic the OS - at least the most popular ones...
[18:11:41] <TrevorParscal>	so, Mozilla said, forget themes, let's use personas instead, which is nothing more than a background image
[18:11:51] <TrevorParscal>	and let's make the browser feel more native on each platform
[18:11:56] <vvv>	Hm, how is it related to the toolbar editing?
[18:12:00] <^demon>	What's the moral of the story?
[18:12:02] <TrevorParscal>	the point
[18:12:18] <TrevorParscal>	We spend allot of money and time on making the toolbar as good as possible
[18:12:35] <TrevorParscal>	adding some things here and there, no harm, that's totally cool
[18:12:40] <TrevorParscal>	that's what the API is for
[18:13:13] <Platonides>	perhaps the custom toolbar should be defined in terms of changes to the default
[18:13:21] <TrevorParscal>	but the idea that a system administrator should choose from some gallery of options and build a PHP array that will affect all users, is more likely to make the toolbar suck than be good
[18:13:31] <TrevorParscal>	Platonides: i agree
[18:13:45] <vvv>	TrevorParscal: tell a guy with no clues about JS (but a lot of FAs on medival culture) that to reorder buttons on the toolbar requires using some API
[18:14:16] <TrevorParscal>	vvv: why does he need to reorder the items on the toolbar?
[18:14:38] <vvv>	TrevorParscal: he feels uncomfortable with them?
[18:14:53] <Platonides>	easier case: the guy creates mathematic articles and wants the math button :)
[18:15:00] <^demon>	TrevorParscal: I'm not saying take away the ability for users to customize it, just that the defaults (and sitewide settings) should be in LS for consistency sake.
[18:15:10] <TrevorParscal>	we should have a gadget that turns on a math button
[18:15:16] <TrevorParscal>	user turns on gadget
[18:15:19] <TrevorParscal>	math button appears
[18:15:23] <TrevorParscal>	everyone moves on
[18:15:57] <vvv>	TrevorParscal: didn't we decide last time that it would be better if user can customize his toolbar in preferences?
[18:16:11] <Platonides>	TrevorParscal, maybe he wants more buttons
[18:16:16] <Platonides>	like... http://commons.wikimedia.org/wiki/File:MS_Toolbar.png
[18:16:32] <TrevorParscal>	^demon: the future of MediaWiki is going to involve more, and more, and more JS... A better approach would probably be to modify those types of things in MediaWIki:Common.js
[18:16:54] <TrevorParscal>	omg - that toolbar is useless
[18:17:03] <TrevorParscal>	we have allot of research that proves that
[18:17:16] <jorm>	wow.
[18:17:16] <vvv>	TrevorParscal: useless to you
[18:17:20] <jorm>	useless to everyone.
[18:17:33] <TrevorParscal>	useless to anyone other than the person who created it
[18:17:40] <TrevorParscal>	and probably not that useful to them either
[18:17:44] <jorm>	with that many tools, you'll need an organizing principle.
[18:17:49] <^demon>	Actually I find all of the toolbars useless and I disable them ;-)
[18:17:53] <jorm>	and then, why have icons?  pulldown menus.
[18:17:59] <^demon>	TrevorParscal: No hard feelings, of course :)
[18:18:10] <TrevorParscal>	^demon: that's what prefs are for
[18:18:12] <TrevorParscal>	:)
[18:18:22] <Platonides>	<Platonides> that's a real extended toolbar added by a gadget <Platonides> I hope you don't want to create a gadget per button ;)
[18:18:47] <^demon>	For awhile I had my edit page really slimmed down. I had disabled all of those annoying notices, the copyright stuff, the character inserter. It was just the textarea, the edit summary box, and submit/preview.
[18:18:50] <Eloquence>	that toolbar looks like something out of dr. strangelove. I'm sure one of these buttons launches nuclear missiles.
[18:18:54] <^demon>	Now that's the way to edit :)
[18:18:58] <TrevorParscal>	what jorm is getting at, is that if users see more than 6 similar items in a group, they will not process them individually, but rather see a pile of crap...
[18:19:02] <vvv>	TrevorParscal: I know many users (including me) who would like to move all "advanced" stuff to the main bar
[18:19:23] <vvv>	And many users who'd like to change things, add some buttons, remove some buttons
[18:19:27] <TrevorParscal>	vvv: just expand it, it will stay open from edit to edit - there's a cookie
[18:19:40] <TrevorParscal>	vvv, that's what the API is for
[18:19:45] <jorm>	right.
[18:20:09] <vvv>	TrevorParscal: usability for *advanced* users means mostly making things customizable as easily as possible
[18:20:26] <TrevorParscal>	and to be fair, the API lets you wipe out the advanced tab and insert all that stuff right into the main toolbar, if that's really what you want
[18:20:47] <TrevorParscal>	vvv: I'm not convinced that's a hard and fast rule
[18:20:57] <jorm>	me neither.
[18:21:08] <TrevorParscal>	advanced users have different goals, and different needs
[18:21:18] <vvv>	Yes
[18:21:25] <vvv>	That's why they need customization
[18:21:29] <Platonides>	wtf my firefox just crashed
[18:21:33] <TrevorParscal>	the tools should ideally be able to satisfy their needs and help them accomplish their goal
[18:21:33] <lecartia>	Usability for advanced users means efficiency.
[18:21:53] <TrevorParscal>	customization is not a need, it's a method of satisfying a need
[18:22:03] <{cary|working}>	you sound like socialists
[18:22:03] <RoanKattouw>	TrevorParscal: ... how does the toolbar API allow you to move stuff currently?
[18:22:06] <TrevorParscal>	a need is "I need to add allot of math code to articles"
[18:22:14] <Platonides>	each user has different needs
[18:22:16] <TrevorParscal>	customization, aka adding a math button, satisfies that need
[18:22:37] <adam_miller>	RoanKattouw: i updated s3 but it's got this 'wgCollapsibleNavForceNewVersion is not defined' js error
[18:22:39] <RoanKattouw>	{cary|working}: I am a socialist, so I guess I'm sounding about right
[18:22:44] <TrevorParscal>	RoanKattouw: well, you can remove things, and add things
[18:22:51] <Eloquence>	I'd love to have little standard "Customize" pop-ups in a few places to allow activation/deactivation of buttons, links, etc.
[18:22:52] <RoanKattouw>	adam_miller: Lemme poke
[18:23:05] <TrevorParscal>	but more improtantly, there's a hook that lets you modify the configuration object before it get's built
[18:23:11] <RoanKattouw>	Yeah
[18:23:13] <TrevorParscal>	and that's where you can splice up the json as you wish
[18:23:20] <Platonides>	TrevorParscal, is all of that documented somewhere?
[18:23:42] <^demon>	s/all of that/anything/
[18:23:47] <TrevorParscal>	probably not well enough
[18:23:56] <TrevorParscal>	but there's plenty of need for that to take place
[18:24:09] <Platonides>	classical toolbar api was easy to understand with a look
[18:24:18] <TrevorParscal>	there's some examples of API calls in the wikiEditor.tests.js file
[18:24:19] <Platonides>	a complex api like this... not so
[18:24:25] <RoanKattouw>	Yes, this API needs some work still
[18:24:39] <TrevorParscal>	i'm not claiming it's perfect, but it does work
[18:24:45] <RoanKattouw>	There are also loading order issues with it
[18:24:50] <TrevorParscal>	and to be fair, it's much more complex because it's much more powerful
[18:24:54] <RoanKattouw>	e.g. addToToolbar needs to queue requests it gets before the toolbar is built
[18:25:02] <TrevorParscal>	the old toolbar API let you add a button, that's it.
[18:25:12] <RoanKattouw>	Yeah the old toolbar was just made of buttons that you could click and that would insert text
[18:25:21] <Platonides>	right
[18:25:22] <RoanKattouw>	So it was icon, alt text, stuff to insert, done
[18:25:27] <Platonides>	and it was good at doing so
[18:25:33] <RoanKattouw>	This toolbar has a range of different widgets with different config options
[18:25:58] <TrevorParscal>	the wikiEditor has more structure, so there are sections (main, advanced, special characters, help)
[18:26:13] <TrevorParscal>	groups (format, heading, insert) with optional labels
[18:26:19] <TrevorParscal>	and tools
[18:26:45] <TrevorParscal>	there's also other types of things, rather than just toolbars such as booklets, which is how special characters and help work
[18:27:02] <TrevorParscal>	anyways, we should document this better, sure
[18:27:08] <RoanKattouw>	Yes
[18:27:13] <TrevorParscal>	we're be preoccupied with getting things working well
[18:27:22] <RoanKattouw>	One of the items on my long TODO list is improve toolbar docs
[18:28:22] *TrevorParscal 	gets back to being productive
[18:31:01] <vvv>	RoanKattouw: are you really socialist?
[18:31:10] <RoanKattouw>	Yes
[18:31:31] <vvv>	I hope not communist?
[18:31:37] <RoanKattouw>	TrevorParscal: Off to watch a movie, ping me if you need me
[18:31:43] <RoanKattouw>	vvv: Noooo, .nl is a sane country :)
[18:31:50] <TrevorParscal>	cool
[18:37:06] <TrevorParscal>	RoanKattouw: when you return, we need to nail down the specification
[18:37:21] <TrevorParscal>	Monday is when we are supposed to begin programming
[18:37:52] <RoanKattouw>	I am here
[18:37:55] <RoanKattouw>	Just watching the movie
[18:38:00] <TrevorParscal>	yeah..
[18:38:03] <RoanKattouw>	How active do you need me to be?
[18:38:16] <TrevorParscal>	well, I need to get Option 3 speced out
[18:38:22] <TrevorParscal>	and I don't quite understand it :(
[18:39:04] <RoanKattouw>	Hmph
[18:39:10] <TrevorParscal>	sorry
[18:39:12] <RoanKattouw>	Can you wait for 2 hours? :)
[18:39:16] <TrevorParscal>	yes
[18:39:23] <TrevorParscal>	i wasn't meaning to distract you from your movie
[18:39:30] <TrevorParscal>	just giving you a feel for the rest of the day
[18:39:34] <TrevorParscal>	:) please, enjoy!
[18:39:40] <TrevorParscal>	I'm going to be getting lunch in a bit anyways
[18:39:45] <TrevorParscal>	now is a good time to start
[18:39:51] <TrevorParscal>	ping me when it's over
[18:39:56] <TrevorParscal>	(what movie is it?)
[18:39:57] <RoanKattouw>	OK
[18:40:39] <TrevorParscal>	.oO(what is he watching?)
[18:41:18] <RoanKattouw>	It's a miniseries called Impact
[18:41:27] <RoanKattouw>	About how the moon is gonna kill us all :P
[18:42:19] *jorm 	shakes his fist at the moon.
[18:48:50] *TrevorParscal 	wonders how the resource loader can make i18n available to gadget developers...
[18:49:21] <TrevorParscal>	while drinking seltzer water from a ridiculously large piece of stemware...
[18:49:26] <RoanKattouw>	loader.php?action=gimmei18n&msgs=foo|bar|baz
[18:49:51] <TrevorParscal>	but where do the messages live?
[18:49:56] <RoanKattouw>	hah
[18:50:03] <RoanKattouw>	Well they can get core+ext msgs that way
[18:50:11] <RoanKattouw>	As for their own, I guess it's DIY
[18:50:12] <TrevorParscal>	sure
[18:50:18] <^demon>	TrevorParscal: Magic!
[18:51:23] <TrevorParscal>	translatewiki.net could build a bridge to JSON i18n data hosted on a wiki....
[18:51:25] <Ryan_Lane>	maybe gadgets should have i18n pages?
[18:51:39] <Ryan_Lane>	and they can be loaded from there, like extensions are?
[18:51:41] <Ryan_Lane>	err
[18:51:44] <Ryan_Lane>	like gadgets are
[18:52:04] <TrevorParscal>	right, but translators need to not have to sift through JSON objects
[18:52:50] <TrevorParscal>	one of the reasons this is sort of fail -> http://www.mediawiki.org/wiki/Extension:TemplateInfo
[18:52:53] <TrevorParscal>	only with XML
[19:28:00] <Ashlee>	You want to pull from translatewiki?
[19:28:36] <Ashlee>	That wiki really ought to be moved to translate.mediawiki.org or something.
[19:29:48] <TrevorParscal>	not sure if the people who run it think so
[19:30:03] <TrevorParscal>	afaik translatewiki also does translations for non wikimedia projects
[19:34:01] <RoanKattouw>	Yes, lots
[19:34:10] <RoanKattouw>	They do StatusNet and lots and lots of other stuff
[19:35:40] <^demon>	There's less incentive to pull them into Wikimedia-land than there was even a year ago :)
[19:35:52] <RoanKattouw>	Didn't WMF offer?
[19:36:14] <TrevorParscal>	RoanKattouw: how was the movie
[19:36:20] <RoanKattouw>	Still going
[19:36:40] <RoanKattouw>	They're doing a cry scene now instead of cool SF stuff
[19:36:45] <RoanKattouw>	So I can afford to chat
[19:37:00] <^demon>	RoanKattouw: I don't know about that. I just know that in the past year they've taken on a lot more non-WMF stuff than before and that gives them less reason to come into the fold I believe.
[19:37:59] <TrevorParscal>	I'm just going over this spec, and making sure we are touching on all of the requriements
[19:39:27] <RoanKattouw>	I'll be back in an hour, tops
[19:43:12] <TrevorParscal>	k
[20:07:15] <thedj>	adam_miller: i think the left side of the magnify icon of search needs a bit more margin. Now the field outline when you have focus partly overlaps the magnify icon.
[20:07:32] <adam_miller>	thedj: screenshot?
[20:07:46] <adam_miller>	browser?
[20:07:53] <thedj>	saf+macos
[20:08:16] <thedj>	http://imagebin.ca/view/ESlcpn.html
[20:08:20] <adam_miller>	oh i see it
[20:08:23] <adam_miller>	righto
[20:08:29] <adam_miller>	i hate outlines
[20:08:54] <thedj>	well i like them, but then can be annoying at times :D
[20:08:58] <adam_miller>	someone explain to me the point...i feel like i'm an ignorant hater of outlines
[20:09:08] <adam_miller>	why on links firefox??? why?
[20:09:28] <thedj>	yeah that is stupid. but on input fields, i love the visual feedback
[20:10:25] <adam_miller>	so i'm not an idiot for adding a line like  a { outline: none; } to every site i build?
[20:10:37] <adam_miller>	is there truly no point other than to make the link look ugly after you click it?
[20:10:48] <thedj>	the search field on the prototypes has really improved btw. Works very nice.
[20:11:52] <adam_miller>	thanks thanks
[20:12:04] <thedj>	well unless your screen is not that wide. then https://bugzilla.wikimedia.org/show_bug.cgi?id=20234 really kicks in.
[20:22:58] <Eloquence>	tiny bug report that I don't see in bugzilla - link assistant shouldn't identify field contents as external links based on "." in title if namespace prefix is File: namespace - makes it impossible to insert file links
[20:24:59] <RoanKattouw>	I saw that report, but I recall coding it so it didn't do that
[20:25:03] <RoanKattouw>	Lemme check the code
[20:25:05] <RoanKattouw>	TrevorParscal: I am back
[20:25:09] <TrevorParscal>	cool
[20:25:12] <TrevorParscal>	lots to do:)
[20:25:13] <Eloquence>	RoanKattouw: I can give you steps to reproduce as well
[20:25:49] <RoanKattouw>	Not needed
[20:25:53] <RoanKattouw>	I just need to look at the regex
[20:26:03] <RoanKattouw>	Which is somewhere in this ~2,500-line file :P
[20:26:09] <Eloquence>	heh
[20:26:53] <Eloquence>	I should clarify - it's possible to insert wikilinks to images, it just confusingly defaults to external links because of the ".". once you switch to wikilink, it'll work.
[20:28:19] <RoanKattouw>	Yeah I know
[20:28:22] <RoanKattouw>	But it should be possible
[20:28:30] <RoanKattouw>	OK this part looks overinclusive
[20:28:34] <RoanKattouw>	These are the rules for deciding something is an external link
[20:28:47] <RoanKattouw>	1. It starts with http:// , https:// , or anything else in $wgUrlProtocols
[20:28:52] <RoanKattouw>	2. It starts with "www."
[20:29:37] <RoanKattouw>	3. It contains a dot followed by 2 or more lowercase letters followed by either a slash or the end of the string. In regex: \.[a-z]{2,}($|\/)
[20:29:53] <RoanKattouw>	Number three is obviously causing lots of false positives with stuff like File:Example.jph
[20:29:57] <RoanKattouw>	*jpg
[20:34:14] <Eloquence>	mm..you could exempt common file extensions (possibly using mw's list), or just the file namespace from that rule
[20:34:23] <RoanKattouw>	I'm just gonna kill rule #3
[20:34:45] <RoanKattouw>	Then people that enter URLs without a protocol that don't start with www. will have slightly more trouble
[20:35:01] <RoanKattouw>	But experienced editors will probably correct them
[20:35:20] <Platonides>	why would people do that and expect it to work?
[20:35:56] <RoanKattouw>	Not everyone understands the difference between an internal and an external link
[20:36:11] <Eloquence>	and, some parsers like gmail's translate plaintext to hyperlinks even in those circumstances
[20:36:25] <Platonides>	note that we do have articles like [[http://www.google.com]]
[20:36:31] <RoanKattouw>	Yes
[20:36:34] <RoanKattouw>	We know about that
[20:37:17] <RoanKattouw>	By our estimates there are some 6000 enwiki ns0 titles like that
[20:37:18] <TrevorParscal>	thats why the controls for overriding the guess are there
[20:37:38] <RoanKattouw>	Out of 6.9M (at the time of writing, probably 7M+ by now?) that's not too many
[20:38:04] <Eloquence>	but, yeah, killing the rule might be the easiest way to avoid problems. fwiw, when you're in File:, you can also use some parameters which the wizard won't process at all, so there might be a case for special-casing it
[20:38:08] <Platonides>	6000? that's astonishingly high, even if almost all are redirects
[20:38:36] <RoanKattouw>	I've also just killed a rule, lowering that number
[20:40:58] <TrevorParscal>	RoanKattouw: let me know when you are ready to finish this spec
[20:42:18] <RoanKattouw>	Yeah, just a few more minutes
[20:42:37] <TrevorParscal>	no worries
[20:42:55] <Eloquence>	TrevorParscal: expected behavior is for the sig icon to be shown in all namespaces != NS_MAIN, correct?
[20:43:06] <TrevorParscal>	afaik
[20:43:11] *TrevorParscal 	looks at code
[20:43:11] <Eloquence>	k
[20:43:38] <RoanKattouw>	Platonides: With the revised conditions, it's only 218 pages on enwiki
[20:44:49] <TrevorParscal>	yes, as long as the body does not have class="ns-0"
[20:44:53] <Eloquence>	thanks
[20:52:31] <Eloquence>	RoanKattouw: expected behavior confirmed working now
[20:52:38] <RoanKattouw>	Yes, I just deployed it
[20:52:44] <Eloquence>	so I saw :)
[20:52:59] <RoanKattouw>	How? You're not in #wikimedia-tech
[20:53:03] <Eloquence>	identi.ca
[20:55:02] <RoanKattouw>	Ah yes
[20:55:16] <RoanKattouw>	TrevorParscal: I am at your disposal nwo
[20:55:25] <TrevorParscal>	k
[20:55:33] <TrevorParscal>	i'm poking at reoganizing things..
[20:55:41] <TrevorParscal>	but that's just cause I was waiting..
[20:56:07] <RoanKattouw>	OK what are you doing and where?
[20:57:45] <TrevorParscal>	the spec sheet
[20:57:48] <TrevorParscal>	i'm moving things around
[20:57:50] <TrevorParscal>	done for now
[20:57:56] <TrevorParscal>	the requirements sheet
[20:58:03] <TrevorParscal>	I'm marking things as included, if they are
[20:58:08] <TrevorParscal>	you can help
[20:58:23] <RoanKattouw>	OK did you finalize the spec wiki page
[20:58:24] <TrevorParscal>	mostly, you could take a look at some of the holes in the spec
[20:58:24] <RoanKattouw>	?
[20:58:31] <TrevorParscal>	spec wiki page?
[20:58:43] <TrevorParscal>	oh, how tim unilaterally decided to move our work where he wanted it
[20:58:51] <TrevorParscal>	I don't intend on working there
[20:59:07] <TrevorParscal>	we can publish it there when we are done
[20:59:24] <RoanKattouw>	Do you need me to write the caching section?
[20:59:35] <mdale>	TrevorParscal: where was it moved?
[20:59:36] <TrevorParscal>	but I prefer working between you and I in real time to the chaos of wikitext and people commenting inline
[20:59:44] <TrevorParscal>	RoanKattouw: please
[20:59:49] <TrevorParscal>	mdale: somewhere on a wiki...
[20:59:53] <TrevorParscal>	i havn't even looked..
[21:00:06] <mdale>	mediawiki.org?
[21:00:11] <RoanKattouw>	Yes
[21:00:23] <RoanKattouw>	See [[mw:ScriptLoader]] which now redirects to the new place
[21:00:35] <TrevorParscal>	his email says "I've moved it to the wiki with the other design documents"
[21:01:47] <mdale>	this "process" is kind of strange ( to state the obvious )
[21:02:34] <RoanKattouw>	Oh I agree
[21:02:42] <TrevorParscal>	mdale: well, since Roan and I are mediating between multiple parties in disagreement, let's just agree to not dissagree on that too
[21:02:43] <RoanKattouw>	But for most things MediaWiki, there is no process
[21:02:50] <RoanKattouw>	Someone just implements it
[21:03:20] <RoanKattouw>	And if that someone has less that 2-3 years of MW experience, their code is very likely to either get reverted or rot
[21:03:59] <mdale>	and my requirements were fundamentally at odds ... like stand alone mode.
[21:04:04] <TrevorParscal>	RoanKattouw: yeah, please fill in the caching and versioning stuff
[21:04:12] <TrevorParscal>	feel free to remove the diff. options
[21:04:18] <TrevorParscal>	and just explain what we are going to do
[21:06:47] <RoanKattouw>	TrevorParscal: About those 5 module states (loading stages): that's a nice piece of theory but I doubt anyone's ever gonna use them. XHR has a 4- or 5-stage process too and no one cares about any of that except that == 4 means "I'm done" so they just use that
[21:07:33] <mdale>	RoanKattouw: I agree
[21:08:04] <mdale>	you really only want callback to run or not... ~maybe~ a second callback for "fail"
[21:08:17] <Ryan_Lane>	these vms need LDAP and nfs mounted home dirs. it's a PITA to set up accounts a million places
[21:08:53] <TrevorParscal>	the spec says a dependency is either satisfied or not... the other states don't need to be exposed
[21:09:00] <TrevorParscal>	it's not a system you poll or something
[21:09:08] <TrevorParscal>	it specifically says there's a callback when it's done
[21:09:40] <TrevorParscal>	the point of the states is, the system keeps track of what's been requested, arrived but is waiting on more stuff, etc.
[21:10:00] <TrevorParscal>	so that ongoing requests for modules don't cause a bunch of repeat reqeuests
[21:10:24] <TrevorParscal>	I could replace the section with "the system isn't stupid." if you like..
[21:10:27] <TrevorParscal>	or just delete it
[21:10:49] <TrevorParscal>	the point is, I agree that the interface is as simple as a callback for done, maybe one for error - sounds cool to me
[21:11:04] <TrevorParscal>	but that's not really mutually exclusive to the content I sketched out...
[21:11:15] <TrevorParscal>	obviously it needs to be rewritten though
[21:11:18] <RoanKattouw>	haha OK
[21:11:21] <TrevorParscal>	if it's causing this confusion
[21:11:21] <RoanKattouw>	No the list is fine
[21:11:41] <mdale>	TrevorParscal, RoanKattouw : there is a lot of stuff ~sort of~ missing from the spec... like how one requests a mediaWiki page resource, how the local messages transformations of work, a rough mapping of the core "helper" functions etc. I don't know if its important to map ~all~ this out or just deal with in implementation.
[21:11:47] <RoanKattouw>	As long as it doesn't make people get the impression that all of those states are queryable
[21:12:11] <RoanKattouw>	mdale: Hm it is true we haven't discussed stuff like client-side PLURAL handling yet
[21:12:13] <TrevorParscal>	mdale: no offense, but many of your comments seemed like they were coming from not reading the whole thing - and I don't mind because I think it's disorganized and boring to read, but... we probably agree with you more than you seem to think
[21:13:07] <TrevorParscal>	mdale: there's stuff missing, but the first statement in the thing says "This document specifies areas of the Resource Loader which have specific design requirements."
[21:13:18] <mdale>	oky
[21:13:19] <TrevorParscal>	not "This is everything, including how we are going to do it"
[21:13:30] <mdale>	oky
[21:13:51] <mdale>	but the loader should probably include how we would load mediaWiki resources no ?
[21:13:52] <TrevorParscal>	we've been getting more "do it this way" than "i need this feature" ... so it's slowed us down
[21:14:03] <RoanKattouw>	But some of that has been useful
[21:14:07] <TrevorParscal>	some
[21:14:12] <RoanKattouw>	Yes, some
[21:14:13] <TrevorParscal>	probably all
[21:14:16] <TrevorParscal>	in some way
[21:14:27] <TrevorParscal>	but it's more difficult to parse and turn into a spec
[21:14:30] <TrevorParscal>	that's all I'm saying
[21:14:33] <TrevorParscal>	it slowed us down a bit
[21:14:42] <mdale>	MediaWiki:Loader.js and User:JohnDoe/loader.js if they exist. .... interesting ;)
[21:14:44] <TrevorParscal>	but it's probably good in the long run
[21:15:06] <TrevorParscal>	mdale: like I said, it might be there, but you didn't read it... ;)
[21:15:16] <RoanKattouw>	TrevorParscal: PLURAL handling isn't there though
[21:15:22] <TrevorParscal>	right
[21:15:23] <TrevorParscal>	so,
[21:15:33] <TrevorParscal>	the spreadsheet has stuff in it
[21:15:43] <TrevorParscal>	i'm going down and marking things as included if they are
[21:15:50] <TrevorParscal>	so we can narrow down what else needs to be speced
[21:15:55] <RoanKattouw>	OK
[21:16:06] <TrevorParscal>	some of this stuff can just be mentioned, or just done
[21:16:06] <RoanKattouw>	I'm beefing up the spec doc
[21:16:06] <RoanKattouw>	True
[21:16:18] <RoanKattouw>	From what I know JS2 has a good PLURAL-in-JS implementation
[21:16:23] <TrevorParscal>	this is mostly about controversial points, or important architectural decisions
[21:16:32] <TrevorParscal>	which I hope to copy-paste
[21:16:51] <mdale>	it has pretty good coverage
[21:17:22] <TrevorParscal>	mdale: you know you love the diagram on the document
[21:17:30] <mdale>	I do ;)
[21:17:47] <TrevorParscal>	the url scheme will likely change though
[21:17:52] <mdale>	but as mentioned in the email.. .oky
[21:17:53] <TrevorParscal>	that was just something random
[21:18:25] <TrevorParscal>	i think the important thing here, give allot of attention to big architectural things, and nearly non to small detail things (at this point)
[21:18:35] <mdale>	oky
[21:18:44] <TrevorParscal>	because we will quickly get bogged down and never get anything done if all we do is talk about url schemes
[21:18:57] <TrevorParscal>	and we have ALLOT to do
[21:19:21] <RoanKattouw>	Yes
[21:19:43] <RoanKattouw>	TrevorParscal: I say we test that pair programming in Eclipse thing first thing Monday morning
[21:20:31] <^demon|away>	I just want to put it out there. You three are entirely too excited about the script loader :p
[21:20:32] <TrevorParscal>	yes
[21:20:35] <TrevorParscal>	that would be cool
[21:21:13] <TrevorParscal>	^demon|away: well, it's a chance to build a solid foundation for all client side mediawiki code
[21:21:24] <TrevorParscal>	I'm equally excited about rewriting the backend
[21:21:30] <TrevorParscal>	but we can talk about that some other day
[21:23:06] <^demon|away>	Which parts ;-)
[21:25:01] <TrevorParscal>	^demon|away: too much of it for my own good
[21:26:56] <RoanKattouw>	^demon|away: I am totally excited about this. I finally get to do hard-code backend dev building stuff from the ground up. Awesome!
[21:27:14] <^demon|away>	:D
[21:27:39] <^demon>	Abstract-schema is way cooler, once I finally finish the other things that are a bit more urgent :)
[21:29:40] <RoanKattouw>	Yes, that is also cool
[21:29:41] <mdale>	p.s coverage for js2 plural: http://prototype.wikimedia.org/s-9/extensions/JS2Support/tests/testLang.html
[21:30:34] <^demon>	TrevorParscal: Bringing sanity to client-side resources will be wonderful, I agree :)
[21:30:41] <TrevorParscal>	:)
[21:31:36] *TrevorParscal 	lets RoanKattouw write specifications
[21:31:52] <RoanKattouw>	TrevorParscal: Killed Versioning section, changed "Caching" to "Caching and Versioning" and added content
[21:32:04] <TrevorParscal>	awesome
[21:32:06] *TrevorParscal 	reads
[21:32:34] <TrevorParscal>	perhaps you could identify requirements not in the spec that seem reasonable, add them as a heading, and we can discuss what needs to be speced or not..
[21:33:07] <RoanKattouw>	OK well one thing is we need to support stuff like PLURAL but everybody knows we need that and JS2 already has the code
[21:33:19] <RoanKattouw>	I could add a sentence about that
[21:34:14] <TrevorParscal>	love it
[21:34:19] <TrevorParscal>	just the right level of detail
[21:34:36] <TrevorParscal>	RoanKattouw: yes, that's a good thing to add
[21:34:37] <TrevorParscal>	somewhere...
[21:34:49] <TrevorParscal>	that's like.. Core library or something
[21:35:04] <mdale>	Well I have to head out... but overall looks good... I just hope its not too much work for me migrate the 25k lines of modules ;)
[21:35:06] <mdale>	later
[21:35:10] <RoanKattouw>	heh
[21:35:14] <RoanKattouw>	Later
[21:35:17] <TrevorParscal>	i see where you are going, l10n is a good place
[21:35:30] <RoanKattouw>	|<--	^demon has left freenode (Quit: zzzzzzzzzzz) ---> wtf?
[21:35:40] <RoanKattouw>	It's like 5:30pm over there
[21:35:45] <TrevorParscal>	ha ha... seriously!
[21:35:59] <RoanKattouw>	That seriously scared me for a second
[21:36:13] <{cary}{working}>	It's called an automatic message
[21:36:21] <RoanKattouw>	If people on the East Coast are starting to go to sleep, that pretty much means the sun is about to come up here
[21:36:25] <RoanKattouw>	Yeah, figured
[21:36:30] <RoanKattouw>	Still :P
[21:36:30] <{cary}{working}>	IRC clients have those....
[21:36:59] <TrevorParscal>	{cary}{working}: Roan's point is, at first glance, he thought he'd lost track of time.. by like 6 or 8 hours
[21:37:03] <Prodego>	yes
[21:37:23] <{cary}{working}>	locking yourself in a room and boarding up the windows will do that to you
[21:37:30] <RoanKattouw>	Well it's before midnight, so all is well
[21:37:34] <TrevorParscal>	yes... i've been there...
[21:37:59] <RoanKattouw>	Flying across too many timezones also works
[21:38:14] <TrevorParscal>	we should probably talk about requirement #35
[21:38:20] <TrevorParscal>	jQuery plug-ins should be centrally organized
[21:38:30] <TrevorParscal>	more like..
[21:38:35] <TrevorParscal>	3rd party plugins
[21:38:46] <TrevorParscal>	or.. modules rather
[21:39:13] <TrevorParscal>	if you have some library, you didn't write, but you depend on, someone else might want to use it too, but maybe not the extension you are using for...
[21:39:42] <TrevorParscal>	if we have a managed namespace, it seems like we can survive lax rules on a folder in the svn where client side libraries like this can live
[21:39:57] <RoanKattouw>	Well
[21:40:07] <RoanKattouw>	What I think Neil (right?) is getting at here
[21:40:12] <TrevorParscal>	yes
[21:40:18] <RoanKattouw>	Is that there's two extensions that need some custom jQuery plugin
[21:40:25] <RoanKattouw>	And one of them ships with v1.5 while the other ships v1.6
[21:40:38] <TrevorParscal>	thats requirement #36
[21:40:40] <RoanKattouw>	Now with our proposed loader system, the first loaded one wins because they control the same global var
[21:40:44] <TrevorParscal>	JS module version conflict management
[21:40:53] <RoanKattouw>	Right
[21:41:05] <TrevorParscal>	what I was telling him, I thought we should do about #36, is nothing.. first come first server
[21:41:08] <TrevorParscal>	*serve
[21:41:12] <RoanKattouw>	But just how does one manage those conflicts?
[21:41:31] <RoanKattouw>	I mean if another extension happens to load an older version of your dependency, you're screwed
[21:41:44] <RoanKattouw>	We could let the highest version win, should be relatively safe
[21:41:59] <RoanKattouw>	Counting on those plugins to maintain decent back compat
[21:41:59] <TrevorParscal>	more likely to be safe
[21:42:00] <TrevorParscal>	right
[21:42:08] <TrevorParscal>	forward compat is less liekly :)
[21:42:11] <RoanKattouw>	Yes
[21:42:29] <TrevorParscal>	so, we do note version numbers?
[21:42:38] <TrevorParscal>	highest wins
[21:42:42] <RoanKattouw>	I say we should
[21:42:50] <TrevorParscal>	what about #35
[21:43:13] <TrevorParscal>	it's still an issue of having 10 copies of the same jquery lib on the server
[21:43:17] <RoanKattouw>	Central place for jQuery plugins in SVN?
[21:43:30] <TrevorParscal>	I think it's a descent idea
[21:43:36] <TrevorParscal>	not sure how that works, permissions wise
[21:43:44] <TrevorParscal>	extension developers need access to it too
[21:43:47] <RoanKattouw>	Yes
[21:43:49] <RoanKattouw>	That can be arranged
[21:43:51] <TrevorParscal>	k
[21:44:05] <RoanKattouw>	I am not completely familiar with the capabilities of authz but I'm sure Tim can work it out
[21:44:34] <TrevorParscal>	so if you have phase3/some_place_where_we_put_the_3rd_party_js_libs/
[21:44:50] <TrevorParscal>	and a standard naming scheme
[21:45:00] <TrevorParscal>	module-version.js or whatever
[21:45:08] <TrevorParscal>	and the namespace is managed on the wiki
[21:45:20] <TrevorParscal>	what about loaders.. are they not needed for these sorts of plugins?
[21:45:27] <TrevorParscal>	they could be optional?
[21:46:06] <TrevorParscal>	coolLibrary-1.2.3.js
[21:46:39] <TrevorParscal>	it could be in the coolLibrary folder and have loader.js inside it..
[21:46:41] <TrevorParscal>	or.. load.js
[21:46:43] <RoanKattouw>	They would need to have a loader I guess
[21:46:44] <TrevorParscal>	or whatever
[21:46:58] <RoanKattouw>	But I reject the idea of not offering the flexibility to use a name other than loader.js
[21:47:12] <RoanKattouw>	Michael wants to just always use loader.js and modulename.i18n.php
[21:47:13] <TrevorParscal>	oh, yeah
[21:47:18] <TrevorParscal>	I think that's silly
[21:47:25] <RoanKattouw>	But whereas I think those should be the defaults, they should be customizable
[21:47:35] <TrevorParscal>	you can use a well-known as a default I guess, and be able to specify otherwise
[21:47:46] <RoanKattouw>	So you could have coolLibrary-1.2.3-loader.js
[21:48:05] <RoanKattouw>	Then we're fine as long as no one's stupid enough to put -loader in their version numbers
[21:48:06] <TrevorParscal>	but I think well-knowns are not worth it
[21:48:20] <TrevorParscal>	they make all these rules have to be known...
[21:48:31] <RoanKattouw>	You're saying always specify explicitly?
[21:48:33] <TrevorParscal>	explicit declaration is good
[21:48:51] <TrevorParscal>	i can be flexible on that, but what well-known things do we do in MW?
[21:48:57] <TrevorParscal>	I don't quite like the idea
[21:49:07] <TrevorParscal>	but I can be pursueded
[21:49:25] <TrevorParscal>	persuaded?
[21:49:35] <TrevorParscal>	i can't spell worth crap... let's just face it
[21:50:13] <RoanKattouw>	Ah
[21:50:18] <RoanKattouw>	I actually don't think we do that anywhere
[21:50:28] <TrevorParscal>	RoanKattouw: so... modules specify what version they are registering, but that's only for the loader's benefit
[21:50:29] <RoanKattouw>	We have conventions but no magic enforcement
[21:50:34] <RoanKattouw>	Yes
[21:50:39] <RoanKattouw>	For figuring out version conflicts
[21:50:40] <TrevorParscal>	the actual request urls need not care or know
[21:50:46] <TrevorParscal>	they will just get the latest version
[21:50:56] <RoanKattouw>	Yes
[21:51:04] *RoanKattouw 	ponders dynamically registered modules
[21:51:43] <RoanKattouw>	For server-side registered modules the server can just do a version_compare() and the client need never know that there were competing versions
[21:52:05] <RoanKattouw>	But what happens when a module tries to register itself while a different version has already registered?
[21:52:14] <RoanKattouw>	Like, a module added on the client side
[21:52:28] <RoanKattouw>	PHP doesn't know it exists
[21:52:47] <RoanKattouw>	Do we care about that?
[21:53:33] <RoanKattouw>	TrevorParscal: Under Packaging: "[some explanation of the general process of responding to a request for modules]" --> what do you want here?
[22:07:06] <RoanKattouw>	TrevorParscal: Poke. I'm going to bed very soon now so if you need anything else from me speak up. Also, yuo have msgs in Google Chat
[22:13:11] <TrevorParscal>	sorry talking to neil
[22:13:11] <TrevorParscal>	..
[22:13:12] <TrevorParscal>	 :)
[22:14:01] <RoanKattouw>	Yeah I heard
[22:14:17] <RoanKattouw>	So 1) please tell me if there's anything you need me to do and 2) please read the stuff I wrote in Google Chat
[22:52:02] <AlexZ_>	Eloquence, hey, thanks for the email. I just got home, so I'm trying to get a few things organized. :)
[22:52:29] <Eloquence>	cool :)
[23:35:00] <guillom>	flipzagging: is XML import enabled on Commons prototype?
[23:35:24] <guillom>	apparently I can only transwiki import
[23:37:51] <guillom>	I fixed the month-showing-weird-things-on-description-pages issue for English
[23:38:12] <guillom>	but to fix it for all languages I need to import ~200 mediawiki pages that I special:exported from Commons
[23:42:55] <flipzagging>	guillom: just a mnute
[23:42:56] <flipzagging>	minute.
[23:43:01] <guillom>	sure
[23:45:50] <flipzagging>	I have never used the Import facility over the web, I enabled it via some shell scripts.
[23:46:06] <flipzagging>	do you know offhand what I would need to do to 'enable' imports?
[23:46:10] <flipzagging>	I see no documentation.
[23:46:25] <flipzagging>	it seems more of a matter of putting users into the right group
[23:47:09] <flipzagging>	(guillom ?)
[23:47:38] <guillom>	flipzagging: I know of http://www.mediawiki.org/wiki/Manual:Importing_XML_dumps & http://meta.wikimedia.org/wiki/Help:Import ; but if it's easier for you to import it yourself, I can just give you the XML dump
[23:48:00] <flipzagging>	guillom assume for the moment that I know exactly as much about MediaWiki as you might pick up in six months
[23:48:16] <flipzagging>	guillom: what I'm saying is, assume ignorance rather than preference
[23:49:25] <flipzagging>	those help pages talk about how "import is usually disabled" without explaining how to enable it.
[23:49:46] <guillom>	right
[23:51:51] <guillom>	oh well, I'll figure that out later
[23:52:03] <guillom>	we have English, it's all we need for now
[23:52:15] <flipzagging>	how did you fix it, btw?
[23:52:26] <guillom>	I created MediaWiki:lang
[23:52:32] <guillom>	and put what was on Commons
[23:52:38] <flipzagging>	that isn't automatically created?
[23:52:44] <flipzagging>	just by installation?
[23:52:50] <guillom>	we need to do the same for the ~200 subpages to make it work for them
[23:52:52] <guillom>	no
[23:52:57] <flipzagging>	that seems wrong somehow
[23:53:04] <guillom>	I think it's a set of pages manually created
[23:53:15] <guillom>	by the Commons folks
[23:53:33] <flipzagging>	hm, they do manage their lang stuff differently.
[23:54:24] <guillom>	flipzagging: try to add the importupload right to sysops?
[23:54:36] <guillom>	importupload 	allows user to import several pages per time from XML files.
[23:54:45] <flipzagging>	guillom: it's already added
[23:54:50] <flipzagging>	and I think I'm a sysop
[23:54:53] <guillom>	ok
[23:54:55] <flipzagging>	checking now
[23:57:38] <flipzagging>	guillom: we're both sysops.
[23:57:44] <flipzagging>	maybe I'm wrong about the config being included
[23:58:10] <guillom>	bah, it's not that important
[23:59:48] <guillom>	and some folks like roan / Ryan_Lane probably know how to do that in an instant
[23:59:54] <flipzagging>	ah, the relevant import of config is commented out... missed that
[23:59:57] <Ryan_Lane>	?