I would like to ask for the opinion of testers and community developers on the super-tester program (this message has also been posted to the testing-squad list, I'll try to bridge ideas should the discussion diverge significantly). The inital rules implemented by Niels are as follows:

===

Currently 3 votes down from the testers will demote an application when the 20 day limit is reached. This should take care of the obviously broken lingering packages.

When 3 votes up and after 20 days -> promotion unlock.

The developers will receive an unlock/reject mail.

The testers will only be able to influence promote/reject when at least 3
votes are the same and less than 3 differ.

===

Now, the tricky part, the admin section. The initial admins are Council members (i.e. Council members can promote/demote someone into a super-tester). Now, for the actual people who would be super-testers. It should be clear that being a super-tester is something that should be a result of a proven testing track record, and that it shouldn't be taken lightly. I also urge future super-testers to use their powers wisely - being a super-tester is not about demonstrating force and vanity - it is about helping stuck packages get to users (or out of testing if they have blockers), but you already know this.

My initial proposition is to bootstrap the list of super-testers with the current top 10 testers (visible at http://maemo.org/profile/list/category/packagetesting/ ), with the option of someone applying for super-tester status when having a *tester* karma of 100 (and probably the recommendation/guarantee of an existing super-tester or Council member). Generally, I don't expect someone to be demoted from super-tester status, but let's have a suggestion there, too - in case of abuse, a super-tester can suggest demotion which will be put in front of the Council/admin body for formal consideration if there are two more super-testers who support the notion.

Now that we have the ball rolling, ideas, comments welcome. Remember, it's going work the way we make it - and the goal is to make things *smoother* and simpler, while keeping in mind the original goals of extras-testing.

Do the rules handle the case where the super-tester is also the package maintainer?

Currently that is a known 'undefined' area. It seemed to me that the current setup ignores tester status in case of maintainership, and unless there are good reasons for changing this, I'm keen on saying we should keep it that way.