> it's easy for non-programing people to bring "visions" regarding > interface design. (and i love do so :) as i know programers, it's quite > hard to establish a new standard. but imho the interface standards > (buttons, dropdown boxes, scrolling, menu-structure, etc.) are now a > couple of years old, and there might be better solutions for specific > tasks. audio seems to me like a good point to start. i wasn't talking about such rudimentary stuff. of course there are alternatives to these basic widgets and several audio applications (even free ones) have begun to support them. the point about a visual interface is that it acts as a "memory buffer" for the user: you do not have to remember much about the structure of the session because the structure is made visible on the screen. can't remember precisely where you put a certain sound? how many copies of the bridge riff did i put in? is the door slam before or after the creak? its all there on the screen, just waiting for you to look at it. as soon as you move away from a visual UI, you have to find some way to avoid requiring the user to remember everything about the session. i think this can be done, but i don't think its trivial or even easy. the visual interface offers another hard-to-replicate feature as well: trivially variable precision. if you try doing cut-n-paste based only on audio feedback, you will find it quite hard/laborious to be as precise as you might want to be. with the visual interface, its much easier to use visual information to get the rough location of an edit and then get to precisely where you want, without many steps. with audio feedback based approaches, i think you will find yourself needing many more iterations through the edit-play-edit-play cycle before you get the location correct. --p
