I've done a few dozen more Yahoo vs. Google comparisons. I'm convinced that Yahoo squandered a good chance to overtake Google. Here's my view on how to overtake Google:
Give users access to the variables.
What do I mean? At O'Reilly's ETCon two weeks ago I saw Google's Nelson Minar, Google's senior software engineer speak. He told us that Google tracks 100 variables that they can play with to move around results.
It's clear that Yahoo has something like the same kinds of variables. Yahoo, for instance, clearly ranks individual's sites (er, weblogs) lower than Google does.
What if the user had control of that? I want a search engine that lets me control the variables. I might never want to see any results with webloggers included. I might want to have results that have only webloggers included (yeah, today I could use Feedster for that). But, what about the other variables?
A search engine that would let me control the variables would be instantly the one I'd use. Imagine if you could use such an engine via a web service? I could display results here. That kind of engine would instantly be the geek's favorite. Why? Cause they could tweak it to give better results.
Even better, why not provide hooks into such an engine so we could come up with new variables that would provide even stronger results?
Anyway, Yahoo will keep its current market share (which is sizeable) but won't convince many people to switch with its current engine. Why? Cause it is isn't demonstratably better.
I think it's possible to make a search engine that +is+ demonstratably better. And, no, I haven't seen MSN's engine yet. I doubt it'll give users the kind of control over variables that I'm asking for too.
I think the innovators will be the small engines like Technorati and Feedster. It'll be interesting to see if any of those open up their variable tables for us to play with. [Scobleizer: Microsoft Geek Blogger]