If you want relevant content, write it and submit it.
If anyone actually thinks this isn't so, think about the mental effort and learning required to write a good article about monads vs the effort required to rant about Bush or Obama or post a cool/funny pic.
("good" and "bad" are simplistic. I actually thought this article was kind of interesting; I just don't think it belongs here)
Not that I want HN to become a joke-a-day site but accidental-hacker-humor like this is a unusual enough find that I can deal with the rare exception.
This is perhaps a point of overlap between reddit and HN and really has very little direct relevance to startups, hacking, programming.
The piece in question here was not particularly deep but made for pleasant lunchtime thoughts about recursion and that fun little Linux fractal viewer that I havent played with in ages (xaos). I wondered how the guy figured this out and whether or not the Capian Sea were really the largest lake... Then when I looked at the comments I saw a few good points and this metacomment.
Not terribly deep thoughts. But I am not so sure that articles like this are irrelevant. It is good to have ludic content that generates seemingly random thoughts.
I used to follow a site called everything2.com - which i imagine lots of people remember (or still use).a simple Perl driven site that if I remember well was related to the slashdot engine(slashcode).
Spending an hour on the site following tangential entries could be really fun but beyond that could be a way to get unstuck when in a mental rut.
HN has that ability at times. In recent months articles on airplane safety; playing GTA with one's 4 yr old; pneumatic brass tubes at Stanford; Feynman on Science and religion have had that affect. None of these articles were particularly 'relevant'. But I am glad to have stumbled across them.