How do you measure "comment quality"? Average comment karma per user? Average votes per comment? Some ratio of upvotes to downvotes?
If it's any of these I see a problem similar to what I think has happened to Stackoverflow. I've answered a lot of questions on SO (although virtually none in the last year) and started doing so a few months after it launched. In that time I noticed two things:
1. The low-hanging fruit got answered (generally speaking most of the highest voted questions and answers are early); and, this is the important one
2. The volume of questions and answers was higher. One impact of this (IMHO) is that people see each question for a shorter period before it drops off the front page. It's that period on the front page where the bulk of votes come from (although there is a significant long tail, particularly on questions that surface on the front page of Google a lot).
HN is of course a news site so the "low hanging fruit" doesn't equally apply but there is some relationship. Each article is, to a varying degree, part of a larger debate whether it be about free software, user rights, the future of the computer/phone/tablet/whatever, etc.
In the early days of HN such debates probably had livelier discussion due to previously unsatisfied demand. Now though I think there is a lot of rehashing of the same point. Early users probably have reduced interest in this.
To (2), if there are more active users commenting then it stands to reason that each user sees less comments overall. Depending on how this scales you may in fact have less people overall reading and voting on a particular comment.
If it's any of these I see a problem similar to what I think has happened to Stackoverflow. I've answered a lot of questions on SO (although virtually none in the last year) and started doing so a few months after it launched. In that time I noticed two things:
1. The low-hanging fruit got answered (generally speaking most of the highest voted questions and answers are early); and, this is the important one
2. The volume of questions and answers was higher. One impact of this (IMHO) is that people see each question for a shorter period before it drops off the front page. It's that period on the front page where the bulk of votes come from (although there is a significant long tail, particularly on questions that surface on the front page of Google a lot).
HN is of course a news site so the "low hanging fruit" doesn't equally apply but there is some relationship. Each article is, to a varying degree, part of a larger debate whether it be about free software, user rights, the future of the computer/phone/tablet/whatever, etc.
In the early days of HN such debates probably had livelier discussion due to previously unsatisfied demand. Now though I think there is a lot of rehashing of the same point. Early users probably have reduced interest in this.
To (2), if there are more active users commenting then it stands to reason that each user sees less comments overall. Depending on how this scales you may in fact have less people overall reading and voting on a particular comment.