has pointed me to this article on power laws by Jakob Nielsen
, which has also appeared in Slashdot
It deals with the known fact that incoming links follow a power law in the web at large, and also in the blogosphere at large. However, he says:
Six years later the same power law was shown to hold for the genre of websites known as weblogs. This isn't surprising, since when you subset a Zipf distribution you get a new one.
That is plainly not true. He follows up supposing that blogs have an even distribution in the ranking of all possible websites, but that is not true. You can create subsets on any criteria (say, webpages on golf) and you could get a power law or not (and that is a meme also given to me by the aforementioned rvr). Indeed, the spanish blogosphere does not follow a power law
, and the paper in english it points to is here
). We still don't know the reason for this, but it might be because it's not self-contained. Or something completely different.