Cook’s Collaborative Edge

on my list:

Chris S. The information available on the Internet is a vast ocean. Google and the other search engines tell us, in effect, “here are the general coordinates of where there might be buried treasure.” What I’m talking about is hiring mercenary treasure hunters to really get down there and find it.

Doc: The opportunity consists in faith that somebody will pay for the service.

So far, the largest market (a private one) for this kind of thing is AMT: Amazon Mechanical Turk:

Questions are HITs: Human Intelligence Tasks. Amazingly, lots of people will work for remarkably close to nothing.

I know a lot, because I’ve been on the planet for 62.5 years. I also don’t work cheap, so — even though I’ve registered with AMT — I haven’t answered any questions. Draw (or re-draw) your own conclusions.

But enough people do work for close to nothing that it is possible to use AMT as both a pool of researchers and a pool of subjects.

Here at the Berkman Center, we have become quite good at using AMT for research (mostly through HITs for subjects). In fact we plan to use it for my own ProjectVRM:

Berkman’s (and Harvard Law’s) Jonathan Zittrain compared AMT to a sweat shop, when he was on stage at Supernova earlier this month. The recorded streams are still up at USTREAM: Look for “Supernova Conference, Live Stream, 2nd Hour”: It starts with Danah Boyd and continues with Amazon’s Werner Vogels, followed by Professor Zittrain at about 29 minutes in. AMT comes up at about 50 minutes in.

I had hoped, back when Technorati started during a project David Sifry and I were doing in 2003, that his new company would provide a better way to do research than the likes of Google had allowed. Toward that end (among others0 Technorati created what was for most of the next six years the best “live web” search engine, even though it barely touched its database, of every blog — then every source of RSS and Atom feeds — on the Web.

Alas, what happened to Technorati recently is what has happened to many companies, and it is discouraging.

First, it got hooked on advertising as a business model.

Second, it jiggered the service for maximized traffic.

Third, it finally realized that the best clickers on advertising weren’t bloggers. On the whole they’re downscale popular-culture types. the most down-scale: people obsessed with celebrities and sex, for example. The awful truth that will come home to roost at Google eventually is that most self-informing smart people don’t click on ads most of the time. Because they don’t need to. This fact (and others around it) leave s”search” short of “find” in way too many cases, nearly all non-remunerative for the “engines.”

This is not to say that Google’s business model is doomed, by the way, but rather that advertising-funded search as we know it has outer dimensions of success.

Ask yourself how much Google’s search has improved since it settled on advertising as a business model. The answer is, not much. Sure, it has been good enough to kill off competition, but then its competition wasn’t very imaginative. Still isn’t, Bing included.

Chris S: In the beginning I suggested that Technorati try a pay-for-research model, because I believed there should be a second tier of deeper search that cost small amounts of money. And, from the beginning, I was told in response that users expect search to be free. And that was that.

Doc: So, a couple months back, Technorati deep-sixed its enormous database — a huge loss, imho (it wasn’t the Library of Alexandria, but the analogy applies) — and dropped its old search engine as well. It even killed its highly useful API (we had used it at Berkman for our own research, and it was terrific), and became was seems to be an advertising play. I’m not sure. Whatever it is, it ain’t what it was. (By the way, I believe is now the best live web search engine.)

Still I take the long view: that we are still in the Net’s early paleozoic. Even Google, for all its wonders, is a trilobite. In its day, which lasted dozens of millions of years, trilobites ruled the Earth. They were the best evolution could do. But they were not the end of life’s line.

All species get stuck. (That’s how they become species.) The same goes for business as for life forms.

Right now Google is stuck in a search model built around PageRank, and a revenue model built around advertising. It’s working fine, but it has limits. We’re still inside the envelope of those limits, so we take them for granted. Google’s view still equals the world. But the world is much bigger than that. And deeper. There is so much more research we could do — research that goes far beyond “search” alone.

The odd irony is that we had higher life forms in research before Google showed up. We also had better ways of organizing files than the haystack the Web became. Those life forms are still around — just not on the Web. You described them well at the start of this thread, Chris. Your mom made a living modeling them. The problem is that nobody has figured a good way to make money with those models on the Net. And the biggest reason is that the giant trilobites don’t use them, and that they are therefore unproven.

So far.

Chris S.: Again, what my mom used to do, back in the day, except something like 20 cycles of Moore’s law later…

Doc: Alas, Moore’s Law applies to chip design and manufacture. Not to business innovation.

In fact, I suspect that one of Newton’s law regarding inertia — “a body at rest tends to stay at rest” — applies profoundly to business as usual.

Trackback URI | Comments RSS