No, he's got a point. New members tend to just be going through stuff that older members figured out long ago. This causes reposts or beginner material to be posted and talked about more often, leaving the more experienced members unsatisfied. Eventually, the experienced members leave for greener pastures and bring a lot of knowledge with them. Worse, this accelerates the spiral and degrades the quality of the site further.

Hacker News helps mitigate this through the continual involvement of PG and YC. Because there are a group of experienced technical people who are grounded to the site, it helps keep quality up. While I have noticed a slight decline in the site, it's far less than I noticed with Digg, Slashdot, or Reddit.

So people who want to learn from HN should not?

From what you two are saying a person like myself, who is just starting out and therefore not writing much in the way of code or blogposts, should not be reading HN because I can't join the discussion with anything but questions?

What about likeminded people discussing a topic that interests them? Is there no room for that?

I think the idea they're trying to convey can be summarized thusly:

lurk moar.

Of course, some elite will outgrow any community save for the few right, private channels in IRC (as with reading the newspaper, anyone will frown at the oft-jornalistic reporting of whatever they know too well). Conversely, some newcomers should really be lurking more, or at least asking around at StackExchange before asserting anything here.

But since the point of HN seems to be "sponge learning" at large (of course, different levels of sponge learning), I don't see any sane way of moderating what should be worthy of HN from what should be asked or posted elsewhere, so long as it has the feel and common interests of the community, and "gratifies one's intellectual curiosity."

Nope, it's just an inevitable quality of any site like this. Keep doing what you're doing.

It's not the reading or even the submitting and commenting that worries me. Reading affects no one but the reader, and HN has pretty effective self-regulatory mechanisms for maintaining the quality of submissions and comments.

What worries me is the voting, which is anonymous and hard to self-regulate. Take a recent, heavily-technical article that made the front-page: http://news.ycombinator.com/item?id=2730286 How qualified is someone (who just learned that something like Coffeescript is even possible) to judge the technical content of a blog post about how GCC shuffles references between the stack and registers, or the discussion around said blog post?

Probably he/she isn't. I personally don't upvote comments that I do not understand.