Unless a better count system is proposed, I will replace the comma check with a greater-than-zero-size check within twelve hours.
Good idea -- basically if something non-blank exists in the article namespace, it should be an article, even if it is a stub. If we add categories, we may want to exclude all articles with [[Category:Stub]] from the count (or from an alternative count), but that can wait.
We should not count #REDIRECTs. We should trim whitespace (blanks, tabs, newlines) before doing the bytecount.
Regards,
Erik