So people are deciding that the odds that something unknown could turn out to be good are better than the odds that something known could turn out to have been misjudged by their network of contacts and info sources.
A secondary point to be made is that simply knowing of something doesn't mean that you know it. You are probably aware of at least a dozen old great books and haven't read them.
I see his point, but he's basically just lamenting human nature.
Furthermore, there are so many benefits to publishing frequently that it's hard to argue against it. For instance, if you publish frequently, you will have a better Google ranking, be seen as an "active" member of the community, get more links on social news sites that link submissions to URLs, constantly show up in peoples' RSS readers, etc. Also, if you write a long, substantive article, many people will not take the time to read it. Most will just skim to see if there's anything interesting, and if they don't find anything, they will leave.
Also, if you want to make money on the Internet, it's something that you need to do daily, or at least very regularly. Dollars follows peoples' eyeballs, and peoples' eyeballs follow constantly updated content. I can't think of an example off the top of my head of a blogger, videoblogger, etc. making a living by posting infrequent posts but I can list a couple dozen people that make very good money with either gawker-like blogs on certain topics, a regularly updated promotional (video)blog, or through advertising dollars on popular YouTube channels.
I know in my rational mind that it is probably better to read content-rich articles on the web, but my reptilian mind is constantly drawn to the new, the hot, the now. Until somebody finds a solution, there will always be a short battle with myself every time I fire up Google reader or think of checking HN.
Also, the example he gave, Cracked.com is notorious for linkbaiting, which isn't a bad thing. Even if people are only viewing new content when they visit the site, at least the old articles have a higher chance of showing up in the SERPs later on.
There are other ways to coordinate, though. Book clubs, for example, often pick non-new books to read. Maybe solutions should concentrate on the coordination problem. I could imagine a version of Hacker News that bans all material published in the last month, so that it is only for timeless classics.
(I am not claiming that coordination is the only cause of this phenomenon.)
Processing new releases into our culture is the way that we form that culture in the first place and make progress.
If we don't keep an eye out for new voices, we'll be stuck with the old ones forever. It's really hard to break through as a new arrival, why would you want to make it even harder?
http://portal.acm.org/toc.cfm?id=SERIES11430&type=series...
Great thinking is not new.
[1] http://www.techcrunch.com/2009/06/03/ron-conway-to-focus-ang...
[2] twitter.com
Of course, this is bound to degenerate into behaviour of hunting for scoops, in a hope to discover a new trend before it gets popular...