On Monday, Tim O'Reilly demonstrated that the Web 2.0 world has so wholeheartedly disappeared up its own arse it has now reappeared at the other end. Only a bit grubbier.
In a post that actually decries an environment he cheerleaded for the last few years, O'Reilly takes aim at herd behaviour. It's not hard to find examples of herd behaviour on the Interweb, although his prime example is from the place where all the best people like to flock: finance.
O'Reilly notices, thanks to a bit of research posted on another blog, that quantitative hedge funds didn't actually do all that much hedging in practice. Or, in other words, it's hard to bet against the market when everybody else decides to do the same thing as you. This, apparently, is a bad thing. Is it? Only if you invested in one of these funds. A bunch of other people did very nicely thank you out of not getting involved in stupid trades and, in effect, picked over the bones of the quants.
Given that hedge funds were a minority sport, with arguably greater influence over the whole finance space than they should have had, the herd here was pretty limited. The real market fallout came not from an inability to read the tea leaves in market data but from the laziness of banks when it came to computing their actual risks when taking on other people's debt. You don't need a big computer to do that, just a little care and attention and the willingness to ask exactly what is something that is no more than a repackaged bundle of IOUs.
In true Thought for the Day style ("A pint of milk turned up on my doorstep this morning. I like milk and I like having it delivered. In some ways, Jesus is like a pint of milk..."), O'Reilly's attention turns to his real target: Techmeme. And the dastardly Techmeme leaderboard.
Apparently, this is doing nothing but encouraging herd behaviour. Well it is called Techmeme. I'd say a core part of its design is tracking herd behaviour on the Internet. The leaderboard just lets you know who the top herdspeople are.
"You always see this amazing pile-on effect. I'm not sure it's healthy," complains O'Reilly about what happens on Techmeme.
There are two parts to that. One is that people only notice the subjects where there has been a pile-on. Those that don't attract a lot of blog posts just drop quickly off the bottom. Second, as with finance, there is a kind of reward mechanism going on which is only partly fuelled by Techmeme.
What seems to happen is that subjects that get the pile-on effect are often those that will drive significant traffic independent of Techmeme itself. Did all the bloggers commenting on Leopard suddenly think: "better get with that Techmeme pile-on?" Or is it down to the knowledge that writing about Apple in general nets more hits? You only have to scroll down the comments from Apple fans on any mildly critical piece to see how much traffic that can generate. These are network effects in action. And, in the case of Techmeme, like hedge funds, they aren't exactly large herds.
But, the thing that made me do a double-take is that O'Reilly was one of the people who, at the beginning, was encouraging people to comment, to write, to blog on whatever they wanted. Anything they produced would enrich the collective whole according to Web 2.0 canon. Now, apparently, they are only allowed to comment in the slightly world-weary environment that O'Reilly now recognises if they bring "fresh inputs". Nothing else is good enough for the future.
However, in accusing people involved in the pile-ons of not looking for fresh inputs, O'Reilly reveals his own failings here. He has assumed that is the case without apparently checking. Does he know that there were no cross-connections to other subjects? He does not seem to have checked. That's not unusual for a good many bloggers, but it's not a good place to start with criticism of an activity you've promoted for a number of years.