Message: 6 Date: Sat, 2 Apr 2011 08:23:36 +0000 (UTC) From: Marcin Cieslak saper@saper.info Subject: Re: [Wikitech-l] Focus on sister projects To: wikitech-l@lists.wikimedia.org Message-ID: slrnipdn7t.2e75.saper@saper.info Content-Type: text/plain; charset=us-ascii
MZMcBride z@mzmcbride.com wrote:
Ryan Kaldari wrote:
Yeah, the local CSS/JS cruft is definitely a problem. I've tried doing clean-up on a few wikis, but I usually just get chewed out by the local admins for not discussing every change in detail (which obviously doesn't scale for fixing 200+ wikis). I would love to hear ideas for how to address this problem.
This caught my eye as Wikimedia has far more than 200 wikis. There seems to be a shift happening within the Wikimedia Foundation. The sister projects have routinely been ignored in the past, but things seem to be going further lately....
The good thing about forgotten/abandoned/unloved/etc. projects is that they probably don't have lots of cruft accumulated in the global CSS/JS files (as they require quite lively tech-savvy community to maintain them).
So those sites will not probably require any changes and will survive HTML5 migration without any problems.
//Marcin
On the contrary, I find the small language wiki projects to be in much worse shape. Often they have syntax errors in their js files, breaking _all_ js. Other times they have stuff that is just plain wrong. (Like anyone remember the toolserver tool to gather stats before stuff was available at http://dammit.lt/wikistats/ - where projects would put js that pinged the toolserver once every 100 hits. That code is still in many small projects [usually with the project code field set to the wrong project]. I've even seen that code in wikis that were created after said toolserver tool stopped working). On the other hand they probably won't complain, as the js is already fairly broken ;)
-bawolff
bawolff wrote:
The good thing about forgotten/abandoned/unloved/etc. projects is that they probably don't have lots of cruft accumulated in the global CSS/JS files (as they require quite lively tech-savvy community to maintain them).
So those sites will not probably require any changes and will survive HTML5 migration without any problems.
//Marcin
On the contrary, I find the small language wiki projects to be in much worse shape. Often they have syntax errors in their js files, breaking
(..)
-bawolff
I just want to leave a quick reply emphasizing what bawolff is saying.
Because of my "2011 Resource Walker"[1] edition of the Tour de Wiki[2] I can only say that it's very true. More often than not, the smaller the wiki – the worse the local site wide resources are.
Blindly copied mess from one wiki to another, in some cases (the better ones) messed with until they work. Other administrators have copied dozens of scripts just to get 1 certain functionality going (they dont know how it works, they just know that wiki Foo has it and they don't, so if you copy all global css/js...., it should work, right ?)
In many cases the contrary was the result, even then the functionality would still not work (ie. because it was comparing wgPageName to a hardcoded pagename (such as "Special:Watchlist") which will never return true on a non-English wiki. Or something else that is dependant on something. In the end the script would just rot.
A few other random examples: * Loading scripts (such as pcount.wikimedia.org) that haven't been around for a long time * "Temporary" hacks. (ie. someone at en.wikipedia tries something for a day, which happends to be the day another wiki copies everything. Then another wiki copies it from there etc.) * Useless CSS (I've seen styles like #enWikiMP_FooBar on my wiki's site wide stylesheets, even though it's useless bandwidth for all of them) * Redundant JS (Declerations of importScript, window.ta = {}, etc. you name it) * And more, much more.
Although it requires integrity and a fair bit of javascript experience, I can only welcome more people to join the tour. And to just take 1 wiki from the list and: * Mark it in the table as "checking" and ~~~ * go through the checklist [2] * go through the migration guide [3] and js deprecation list [4] * for [Vector|Common|Monobook].css/js * Check if there's any MediaWiki:Gadget* or MediaWiki:J[q|Q]uery* prefixes wich may have to be replaced with mw.loader.load/using [4]
On average it's about 30 minutes for wiki. Some wikis dont have any global js/css at all, others are a big fat mess.
-- Krinkle
[1] http://meta.wikimedia.org/wiki/User:Krinkle/Le_Tour_de_Wik%C3%AD [2] http://meta.wikimedia.org/wiki/User:Krinkle/Le_Tour_de_Wik%C3%AD/1-17-allwik... [3] http://www.mediawiki.org/wiki/RL/MGU [4] http://www.mediawiki.org/wiki/RL/JD [5] http://www.mediawiki.org/wiki/RL/MGU#Keep_gadgets_central
wikitech-l@lists.wikimedia.org