> I am curious: Can you elaborate on how such analysis is being made?
Try doing analysis on the time at which posters in a thread joined this site. I weigh their dates based on the number of times they comment in a thread. If you do that you'll see a recency bias. The mass of the distribution is clumped more heavily in the last 3 years since the article's submission. If you think about it, that makes a lot of sense. People are probably much more energized to comment when they first join and eventually get bored and stop. You point out a really similar effect out in your own analysis about posters that stop posting after a year.
A poster that joined in the last 2-3 years.
> I am curious: Can you elaborate on how such analysis is being made?
Try doing analysis on the time at which posters in a thread joined this site. I weigh their dates based on the number of times they comment in a thread. If you do that you'll see a recency bias. The mass of the distribution is clumped more heavily in the last 3 years since the article's submission. If you think about it, that makes a lot of sense. People are probably much more energized to comment when they first join and eventually get bored and stop. You point out a really similar effect out in your own analysis about posters that stop posting after a year.