On Sun, Nov 14, 2010 at 7:04 PM, Platonides <Platonides(a)gmail.com> wrote: Agreed. And deferred doesn't mean that nobody cares like people keep saying. It means that the code reviewer in question doesn't care /right now/. This happens for one of the following reasons 1) i18n commits: we almost never review these in full because they're huge and it's a very tested process (exporting and checking in the changes) 2) It's a non-WMF extension or other piece of code (eg: WikiWord) 3) Your own code from quite some time ago, and nobody has bothered reviewing it (the newer 'old' status might be better for this, however) Deferred isn't a graveyard, anything can be pulled out of it for review. FWIW, I just marked all the SocialProfile deferred commits back to new since he wants it reviewed. Generally speaking, I review the following (in no particular order, and (b) overlaps with (a) and (c) sometimes) a) Core changes b) Things I am interested in c) WMF extensions I work on, eg: FlaggedRevs & CodeReview Most extensions don't fall into this category for me, so I don't bother looking at them. I'm perfectly willing to leave SocialProfile as new if someone indeed will take up review of it. My main issue is leaving things "new" forever that nobody actually reviews. -Chad

Hoi, Do consider that we promote the use of the Wikimedia SVN for MediaWiki software. With extensions in SVN it is easier to provide localisation support for any and all extensions. When an installation has the LocalisationUpdate extension running, new relevant localisations will become available. The idea of a "dontcare" status is demotivating and it is not at all in line with best practice. Thanks, GerardM On 15 November 2010 15:08, Chad <innocentkiller(a)gmail.com> wrote:

"Gerard Meijssen" <gerard.meijssen(a)gmail.com> wrote in message news:AANLkTimP5=DRht3bjsuZuZOE8gmFzAr3cEVA7qjqFXgQ@mail.gmail.com... It may be demotivating, but it is essentially just calling a spade a spade. The difference between "won't be reviewed" and "will be reviewed at the end of the world" is purely semantic. "Deferred" does currently contain a mixture of "do honestly intend to get round to eventually" and "don't honestly intend to get round to"; but separating them only exposes the question of which types of commit fall (or should fall) into each category, it doesn't answer it. --HM

On Sun, Nov 14, 2010 at 4:35 PM, Tim Starling &lt;tstarling(a)wikimedia.org&gt; wrote: I didn't read Jack's comment as a complaint about anyone's generosity (though admittedly, it's a little tough to parse that sentence). I read it as acknowledging the fact(?) that the review queue is too long to get through without marking some extensions/etc as "deferred". After discussing this with Jack on IRC[1], I encouraged him to send this email. This seems like a real problem. If Jack wants someone to review his code, should he simply move it back from "deferred" to "new"? What criteria are used to mark something as "deferred" in the first place? I'm assuming it's currently standard practice to mark extensions that aren't slated for deployment by WMF as "deferred"; do I have that right? Rob [1] http://toolserver.org/~mwbot/logs/%23mediawiki/20101112.txt ...starting at "[17:47:55] <ashley> speaking of things in SVN vs. things used by WMF..."