Hey Tyler, Many of these issues have already been discussed on this mailing list. Read Rob and Tim's emails from last week to start with. As explained in the previous emails, the extension being deployed is Scribunto. Regarding performance testing, Rob said this would be done once the extension was deployed to mediawiki.org: "From there, we'll need some time figuring out the performance characteristics of this (making sure we're actually coming out ahead) as well as converting some key templates over." Many of the other questions are covered at https://www.mediawiki.org/wiki/Extension:Scribunto.
And not to be overly dismissive, but the idea that Tim needs to prove that en.wiki wants this feature is absurd. The template system on Wikipedia is BROKEN. It takes over 30 seconds for the parser to render large articles, and articles with a really large number of citation templates can't render at all, they simply error with a timeout. The only reason Lua/Scribunto was developed is because the en.wiki community has been vocally complaining about this problem FOR 3 YEARS. Check out https://bugzilla.wikimedia.org/show_bug.cgi?id=19262 for example.
Ryan Kaldari
On Wed, Aug 22, 2012 at 8:08 AM, Tyler Romeo tylerromeo@gmail.com wrote:
I think Yury has a point. Now would be a good time to maybe discuss exactly what's going on. As exciting a feature it may be, we cannot just "deploy next week" and then have "the schedule for deployment not yet decided". Stuff like this should have a legitimate plan. Furthermore, in alignment with the previous thread on feature development, is there any hard discussion on enwiki, etc. showing the users want this feature? I know sure as hell that I'd love using this feature, but I don't represent all template developers everywhere.
Some good questions we should probably answer (if they haven't been answered already):
- Is Extension:Lua the extension being deployed? If so, why is it still
in Subversion and why is it marked experimental?
- What QA has been done on this extension? How many test cases have been
implemented?
- What are the performance impacts of using this v. regular parser
functions? (Also, what is faster, PECL or external interpreter?)
- Do global variables persist outside of an individual script, i.e., can
one global variable be used in multiple <lua> tags in the same template?
- Has there been any consideration of implementing a "standard library"?
For example, functions that will allow the creation of wikitables and other mediawiki syntax.
- What values for the various wgLuaMax* variables are we planning on
using on WMF wikis? Has there been testing done to determine what a reasonable maximum call time is?
I probably should have looked into this more earlier, but it's been a busy week for me and I haven't had much time.
*--* *Tyler Romeo* Stevens Institute of Technology, Class of 2015 Major in Computer Science www.whizkidztech.com | tylerromeo@gmail.com
On Wed, Aug 22, 2012 at 8:27 AM, Tobias church.of.emacs.ml@googlemail.comwrote:
On 08/22/2012 12:18 PM, Tim Starling wrote:
So my inclination is to push for deployment with a minimum of additional development work. But I'm not the target audience; my inclinations have to be weighed against the needs of the users.
in the name of countless Wikipedians, who are struggeling with that horrible Template/Magic word/ParserFunctions syntax, I say: thank you :)
This page is dedicated to its victims: http://meta.wikimedia.org/wiki/User:Church_of_emacs/Template_love
Cheers, Tobias
Wikitech-l mailing list Wikitech-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-l
Wikitech-l mailing list Wikitech-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-l