[jerf]

I read that article three times to be sure, and there isn't a damn thing in it I haven't seen at least twice (and for most things, more) in the open source world since about 2005. Including "activity-centric" interface. (Not counting shrinking the desktop so you can put the menu up without covering anything, but "meh".)

Personally my opinion is that until something fundamental changes in the IO devices or the intelligence level of the thing reading the IO devices, we've long since grabbed the low-hanging fruit and quite a bit of the not-so-low-hanging fruit. And most of the changes I see proposed for IO changes are steps back, not forward. (Like a "true 3D desktop", which is a recipe for getting the user lost in new and exciting ways, not wonderful new interface empowerment.) We've grunted what we can grunt in our point & grunt interfaces, we need either real words or more real language.

(Since people may not know this: When I say "point & grunt", it is not intended as a derogatory term per se. What it is intended to illustrate is that the rate at which we can communicate information with a computer is very limited. Point at your computer screen and pretend your finger is the mouse cursor. You can point at anything you want on the screen. Now, you have 1, 2, 3, or at any rate a small number of mouse buttons. That is, you have a grunt, and a different grunt, and on those sophisticated verbose UNIX systems, a third type of grunt, the point being that your mouse click doesn't carry all that much information. Information in the information-theoretic sense; at full blast, I'm lucky to get 10 bits a second into my computer. Try communicating with your coworkers that way all day, and you'll grow to appreciate the limitations. Computers get a lot of out it since we've had decades to learn how to cleverly prompt the user's grunts in such a way to get the most information out of it, but there's just so much we can pull, no matter how clever we are with our menus and UIs, from pointing & grunting.)

[trop]

Speaking of point & grunt, bear in mind that in the early '70s Alan Kay and Adele Goldberg were building by testing it in the Palo Alto schools. Now we grown-ups have adopted a paradigm which designed to be understood by young children.

(Though when I go back and reread Kay on this, I'm actually struck by the incredible sophistication he expected of the 12-year-olds, far from "point-and-grunt".)