(What follows is MediaWiki community stuff, not technical discussion, offered in the interests of transparency and collaborative planning.)
TL;DR version: I don't think we can do Google Code-In well, so I don't think we should apply to participate this year.
Since MediaWiki has participated in the Google Summer of Code mentorship program, we have also received an offer to apply to participate in Google Code-In, which runs Nov. 21 2011-Jan. 16 2011.
"Google Code-in is a contest for pre-university students (e.g., high school and secondary school students) with the aim of encouraging young people to participate in open source. We work with open source organizations, each of whom will provide a list of tasks to be completed by student contestants. Tasks can be anything a project needs help with, from bug fixes to writing documentation to user experience research."
http://www.google-melange.com/gci/document/show/gci_program/google/gci2011/f...
https://code.google.com/p/google-code-in/wiki/GCIAdminMentorInformation
I have now gotten more information about what it's like to participate in GCI, from organizations that have taken part in the past. And it sounds like we just do not have the community capacity to do GCI this year.
* I can't take the time to develop the task lists or wrangle others to do so by the deadlines (applying by 1 November, creating big task lists by 21 November), due to other commitments (another volunteer development/mentoring program, India hackathon 18-19 November, mentoring existing new contributors). And I do not believe anyone else in the MediaWiki community has the capacity to administer our participation between now and mid-January, either. Tell me if I'm wrong!
* We'd need enough mentors on call to review the teenagers' assignments as soon as they're submitted, so they aren't stuck waiting around before they can grab a new task. This is for the whole two-month period, including any winter holidays. Right now I do not think we can satisfactorily guarantee that. We all have too much other work that takes priority.
So: it's a cool idea, but I don't think we can do it well in the given time period, so I'm turning it down. (Unless you want to run it, and can guarantee some mentors' attention! In which case, tell me ASAP so we can get an application in by 1 November.)
BUT: the number of participants we're getting in our Coding Challenge (thanks, Greg!) means that if we want to do something like this *ourselves* next year, on our timeline, we could probably get some pretty good participation rates -- especially if we partner with Wikimedia Foundation's Global Education Program. So let's come back to this idea, perhaps sometime in the spring.