Now that there is a download counter of files, popularity of files is being measured by the total number of downloads a file has obtained. Now, it might just be me, but for new files entering into the archive, they are not measured fairly, as top files will most likely have been archived long before a new file is archived.. This will result in little movement
in the popular files chart, thus basically providing static web page content,
as it will be virtually impossible for new files to surpass older archived files.
Might I suggest defining popularity based on downloads during a specific period of time, thus allowing new files a chance at being declared popular or not... This is how I deal with the fairness of file popularity over at my page, but I realize it might not be everyone's way...
Just thought I'd point out some logical flaws in the current system, and hope they will be of some assistance.
The thing that prompted me to add the download couter was having authors keep asking me how many times their script has been downloaded, while labelling it as a measure of popularity was a bit of an afterthought. I think you have a point, and will keep an eye on it to see how things go over time. But sometimes it just takes care of itself--a bug reset all the counters a few days ago