So, Andrei, where do we go from here?
My paper on the taxonomy of Web search talks about three generations of web search. I believe that we are now entering an entirely new phase. I call this next phase “search without a box”. Search today is confined to putting in something and getting something back, a pull model. The next stage is for information to come in a context without actively searching, a push model. My favorite example is GPS. Instead of looking up your way on a paper map, you are in your car, and your GPS navigator gives you directions, shows gas stations near you, and so on. A year or two from now perhaps it will show you where those gas stations are, but only when you are low on gas. So you get information on an “as needed, when needed” basis without explicitly asking for it. In the same vein, we will move from information retrieval to information supply.

Exactly. This is not about removing control, but about elimination of repeating obvious steps. This is the direction most AI technologies head, and Information Retrieval is part of it.

When building technologies, we often get the interfaces wrong. Because it is an iterative process; we can't skip the steps until we've worked them out. My browser now has a search bar, because it turns out that the majority of my interface with the web is through search. My browser now has tabs, because that's a more effective way for me to surf (before I used to open windows by the dozen). In the future, that may not be the case; flock is making interesting noise about the way a blogging friendly browser should operate. And the general search tool may be less relevant as well. When I'm working on code, the reference pages I look at should be more accessible. Debugging -- give me more information. And give me a small distraction tool; maybe tied into the period of my debugging cycle.

I appreciated that my old cell phone gave me the option of exposing my E911 required GPS information to the mobile web browser; but too little used it. Advertisers have gotten very aggressive about using IP based locality information... so should the good services. Memestreams should work differently when I am in my office than at home, and maybe I shouldn't have to tell it to. When I am working at work, or just killing time. Etc. Firefox knows, or could.