Replying to this thread as it's related to DiscApp stats...
So, out of boredom, I decided to revamp the Perl program that scrapes the logs to generate the stats. I cleaned it up quite a bit and it now skips most spam urls and also compresses hourly stats into daily and daily stats in to monthly. (My free DB is only 20MB).
Anyway, one change I made about a month ago is to include the query string in the url that's saved. This let me parse the data's 'discussion.cgi' URLs for page views by disc app and article id.
I threw together a quick Perl script that does the above parsing into a csv that the table shows below. (I also lazily made a Perl script that converts csv to an html table tag. :D)
Below are the top article page views for NEMB with page views greater than 10 for the past ~30 days. It has the Perl DiscApp URL due to me using that to scrape to get the article information instead of connecting directly to the production database from my script or calling the actual nediscapp.com which would cause additional page views in the gathered stats while I'm generating the output.
Some of the counts I'm assuming are search engine bots following the link from the NE news page. There are some odd ones though like '*Nice' and an old Beta release thread from years ago.