I am not sure I am happy about the idea of adding a multitouch layer to Mac OS X. This isn't because I think multitouch (or other direct manipulation UI) is a bad idea, but because I think we need a rethink about the way we construct software systems in the first place.
For many years, I have operated under the technical philosophy of 'technology should work for people, not people work for technology.' For me, the ideal technology is one that allows people to apply their brains and bodies to 'higher-level' concerns without adding a new lower level concern. We are finite creatures, and we only have so much time in this world to spend. Why spend substantial portions of that time learning how to operate a machine, when it is possible to reduce that and free up the time for other pursuits?
That said, I believe other changes are necessary before we'll reach that stage. Yes, the UI is vitally important but there are important reorientations necessary at the controller (currently, application) and model (currently, file system) levels too (there is a wealth of research into these ideas in the literature that is worth understanding if you want to know what options are on the horizon).
We'll see what happens over the next few years, but I for one won't mourn WIMP UIs if they fade from common use.
For many years, I have operated under the technical philosophy of 'technology should work for people, not people work for technology.' For me, the ideal technology is one that allows people to apply their brains and bodies to 'higher-level' concerns without adding a new lower level concern. We are finite creatures, and we only have so much time in this world to spend. Why spend substantial portions of that time learning how to operate a machine, when it is possible to reduce that and free up the time for other pursuits?
That said, I believe other changes are necessary before we'll reach that stage. Yes, the UI is vitally important but there are important reorientations necessary at the controller (currently, application) and model (currently, file system) levels too (there is a wealth of research into these ideas in the literature that is worth understanding if you want to know what options are on the horizon).
We'll see what happens over the next few years, but I for one won't mourn WIMP UIs if they fade from common use.