The contest system isn't perfect. For instance, I wrote a log4net article on December 29th. It didn't even make it into the contest, I believe, yet if you search Google for log4net, the first non-apache link is the link to my article. Obviously it was/is popular but it didn't get the recognition necessary to enter the contest (and I cannot re-enter it into another contest). Also, there are a lot of articles that won't make it in at all because of their code language. For example, SQL articles don't have a category to be voted on. Finally, depending on the month you enter, a great article might not win because of an even better article whereas the next month, a very good (but not great) article might win because of the lack of true competition in that category that month. However, even though there are obviously gaps/issues/etc., I think the contest is fine the way it is. If you expanded it to updated articles, new articles would probably not have a shot since the updated articles have a following already. Also, it would greatly increase the number of articles that should be in the contest. Finally, determining how to place the article into the contest in the first place would be difficult. Would it be number of votes that month? If so, those who saw the article earlier wouldn't get to vote again. If it is traffic, the established article would have more traction. I think in the end, we might need to look at tweaks to the contest, but not major changes (like this one). At the end of the day, the contest is just a cool bonus. It won't ever be fair, so we should treat it as good fun but not put so much stock into the results that we overlook the fact that we have excellent authors posting amazing articles for free to benefit the community every day.
