1. Seth Finkelstein on May 26, 2004 7:57 PM writes...

Absolutely.

The richest people are not necessarily the most famous.

The most famous people are not necessily the most rich.

One TV show was titled "Lifestyles of the Rich AND Famous", not "Lifestyles of the Rich OR Famous".

But in any case, there's only a relatively few slots for each.

2. Christina on May 27, 2004 11:22 AM writes...

I think the point is perhaps something different: authors add links to their blogroll because the authors feel that the sources are important (e.g., I read the NY Times and WSJ every day), but are actually citing the granular bit of a blog that is relevant to the comment. Maybe someone who has historically had interesting things to say vs. a good/bad article. How does this differ from what's found in citation analysis/bibliometrics: every new article on subject a starts with a citation to important article b in the literature review section, but then continues with citations to c, d, e that are on point for the specific topic?

3. Adina Levin on May 29, 2004 6:30 PM writes...

Another difference is intertia.

People link very often, and update their blogrolls much less often. Therefore a blogroll will be out of date.

4. Lawrence Krubner on May 31, 2004 11:19 PM writes...

The real issue is the extent to which each power curve influences the other. We know that people with fame but no money can sometimes turn their fame into money. We know that people with money but no fame can sometimes, with effort, turn their money into fame.

When it comes to weblogs, the first task is to map out all of the power curves that are relevant to people's viewing habits. But then the second, important step is to map how each power curve influences the other. Only once we see the interactions between the many, many power curves do we begin to have a model that is as dynamic as we know the situation to be. Our experience of every form of social competition, online and off, is fraught with the tension between the inertia of the status quo and the explosiveness that arises from everyone's aspirations. There are allinaces, groupings, factions, strategies, individuals forming cliques forming cells forming territories forming tendencies forming movements.

Specifically of blogrolls and linking, the one is clearly a lagging indicator, the other a leading indicator. Someone writes well for awhile then falls off, people are slow to remove them from their blogrolls. Hell, think of how many well-linked people stop writing for awhile - their dead sites are still well-linked for awhile. Blogrolls are a lagging indicator, they tell you who was doing well 6 months ago. Links is weblogs are a leading indicator. If someone is on a tear, people will link to them, eventually they'll find their way into blogrolls.

5. Adina Levin on June 1, 2004 9:30 AM writes...

(standard comment follows) The focus on power laws grafts our cultural habits from the age of mass media onto internet media. Or, more charitably, we focus on the metric that's easiest to measure, rather than the metrics that are most illuminating.

The interesting thing about weblogs is not that at their most popular, they start to approach mass media, but that they support a flourishing range of sub-communities in geographies, language groups, friendship groups, and sub-topics; and that individuals participate in multiple sub-communities at once. This is harder to measure but at least as important.