I have a ton of respect for Aza Raskin, who has pushed this, and I respect that infinite scrolling was a neat solution for some problems of the 2004 web. But I'

I have a ton of respect for Aza Raskin, who has pushed this, and I respect that infinite scrolling was a neat solution for some problems of the 2004 web. But I'm not sure those problems are the biggest problems today, and I mostly I feel like "fixing scrolling" is working on the wrong problem, when we could instead be working on "ditching scrolling." The addendum to Fitt's Law is key here: The current location of the pointer is an infinite target. Or, to paraphrase, dragging is usability hell.* If I'm flipping through pages, I want to click or tap one thing, I don't want to have to manipulate things by clicking one spot, holding a button, finding another target and releasing. I know clicking and dragging is intuitive and easy for all of us, but compared to one simple click at the current mouse location, it's a usability eternity. My approach would be to display all search results from the first page of Google cleanly on one screen. Maybe two columns wide if necessary, or fit to the screen in a tumblr style way. Not infinite scrolling, not any scrolling. I can keep my mouse on top of a little set of forward and back arrow buttons in the corner and flip through results that way. Or I could just use arrow keys or taps on a mobile device. * Sure, dragging is the only sensible approach in some cases, maps and canvases and 3D object manipulation, sure. But if it can be designed away. So I could see click and drag if your search results were not linear, but a two dimensional graph. Maybe you get that graph by... going right or left puts more priority on different terms in the search box, or going up and down adds different commonly included words to your search, I don't know.