Reading
this entry in Joho's blog, where he speculates about the future of blogging, and says:
8. The distinction between the big, high-traffic blogs and the rest of the world of blogging will be increasingly sharply etched. The "tail" will gain more and more value as the number of high-traffic blogs necessarily grows much more slowly
I sit down to think a bit about the future of blog
incoming hyperlinks power laws (as shown by
Kottke and
Kevin Mark). We have all heard about this, often in the same phrase as "rich get richer", and stuff like that, implying that, in the future, A-list blogs will get even more incoming links at the expense of the tail, those at the far left of the graph. This is probably also Dave Weinberger point in that phrase, although it's not too clear, since the second phrase seems to contradict the first one.
This is what I try to sketch in this plot: different scenarios of evolution of the power law (which is a log-log plot of different ax
b functions,
x being the blog rank, and the dependent variable the number of incoming links, changing
a and
b). The first phrase ("... increasingly sharply etched") might look like exp=2.5, with the rich effectively getting richer. The second, more like exp=1.9, with the tail gaining more value, little by little, and.
What is the most like scenario? My hunch is that it might look more like
exp=2.2, that is, the exponent of the power law will change ever so slightly, maybe 0,1 up or down. The explanation is that link decay only happens in the long run, so links are always added, they never decrease; with an ever-increasing number of incoming links, that's how things will look like.
But that's only part of the picture. My opinion is that it's an error to plot incoming links for
all eternity; time frames should be taken into account. The web is never a static thing, and those popular now might not be so in the future. If we consider a 1-year moving window, and take into account only links created during that period, what will happen?
It's anybody's guess. Mine is that the situation will not be too different: the independent term will increase while the number of blogs continues to grow, while the exponent will not change much.
