The interesting thing is that people end up with hierarchical models even in the area of free software development. See for example:

http://web.mit.edu/ghudson/thoughts/bitkeeper.whynot

Although in this case, I believe it's little to do with control, more with hiding inefficiencies across the hierarchy.

Florian: James Surowiecki makes the same point in The Wisdom of Crowds, which I cited in part 5.

Seth: Thanks for your support and the original inspiration for this series. I hope it's useful to someone. If I think back to a year ago (when I had merely an inkling that blogs were inadequate), I couldn't possibly have come up with the New Gatekeepers theory then. It took a number of months of research on this. To what ends of course-- it doesn't matter a lick until I get the modules written.

For readers waiting for us to actually disagree on something, here's an attempt. I don't see the difficulty being in the technology-- but the decision to go ahead. It's like when my Mom discovered that AT&T customer service used web-based IM to respond to her, and that was the first she had encountered it. Wow! She asked me, how did they do that? And I explained that IM was a fairly old technology; the big work was in getting the CIO to sign off on it. (especially tricky considering there haven't been, to my knowledge, conferences on "IM, Telephones, and Credibility."

It seems like you felt that the idea TrustRank was a good step in the direction of trying to meet these sorts of needs. I'm happy with working towards half-solutions, which are better than the non-solutions we currently have.

"I'm happy with working towards half-solutions, which are better than the non-solutions we currently have."

Ah, but it's more like one-googol-th (not a typo, a pun) solutions. And even those are very hard.

I'm not *against* them, or even against trying - it's just that I think they're much, much harder than might be apparent on first glance.