1. Nope. I think there comes a point in every game where one can say "I understand what this game has to offer". In some games (usually particularly flat, terrible or one trick pony games), that point comes very soon. In others, not so much. I think reviewers should play as much as they think they should play to be able to accurately judge the game. No need to set a certain number on it, just make the reviewer state how much of the game they played somewhere in the review.
2. Ha, nope. I often say "the purpose of a review is to inform the consumer"...so what about consumers who are new to a genre? If someone hasn't played many of a particular genre, then their review is more useful for those who haven't played many games in that genre, and less useful for those who have. That opinion is still perfectly worthy of being heard and perfectly valid, assuming it is a well written review.
Additionally, bias towards a genre can be just as strong as bias against a genre. This system just removes half of the complete picture, giving an overly rose tinted view of games.
3. I think you contradicted yourself on this point, which just goes to show how divisive this can be. You say that they should review a game based on its launch day condition, yet you then say that they should review the game based on its condition a full two weeks after launch? This is just ridiculous. I don't think there should be any "rule" here as here is no right way to do it. Just be clear with what your policy is and that should be just fine.
4. Ha, a contract? With who? Their employer? This is just a bit silly...if a site has such a low standard of ethics that it allows skeevy practices to continue, then simply stop going to that site. A contract is just a silly way of dealing with this issue.
5. Another needless rule. First of all, what sites do you go to that feature people who barely game writing reviews? Actually, thats really all that needs to be said. I consider myself to be a pretty big gamer, yet I often play games fairly slowly. It sometimes takes me literally years to play through a game. Now that wouldn't be very useful for a review site ("YOUR REVIEW IS THREE YEARS LATE!!!"), but I think that my opinion of the game is just as valid as someone who marathons games. There has been a lot of talk about the fact that reviewers are often pressured to rush through games to get through them on time, and that just isn't a natural environment for your average gamer. Once again, as long as the consumer is educated on the conditions of the review, I don't see a problem.
-Overall, all of these problems can be overcome by eliminating review scores. When the Polygon review of Bayo 2 came out, my complaints weren't so much "this review shouldn't exist", as it had every right to exist, and it could be helpful for a certain group of people (prudes). My complaint was "should this really factor into the score". Without a score, this wouldn't be a problem and none of these things would really be much of an issue