When using my computer, if it requires more than one or two clicks to get what I want, I get frustrated. So my overall goal for my "newi" project (name still in development ha) is to put all of your files/applications/web pages just one click away (or if absolutely necessary two clicks). Of course you could do this right now by moving all of your applications to your desktop, but lets be real no one wants to do that. And in comes my idea for the first cycle of newi.
Without going into too many details, I was picturing a typical "windowed" browsing system but removing the need to click to open a new folder and click to go back a folder. I felt that a user's mouse movement should be enough to tell you what or where they are actually interested in going. By emphasizing this area, the user will be informed of what the newi intends to do. At this point the user can continue the movement to execute or continue the action, or change the movement to suggest a new action.
The action will vary from object to object (and should be customizable by file type). For example, by hovering over a folder it will zoom the folder and expose the folders/files inside. To move back they would move to a small frame that will remain visible. By hovering over an application for x amount of time the application would launch (file browsing within the application would work the same). And by hovering over a file (text document for example) it would launch the appropriate application and open the document.
So this all sounds great, but there are plenty of concerns. Could this create a "too instant" experience? Would users be in and out of folders multiple times because of newi jumping the gun? Of course it would require a lot of tweaking of the sensitivity, but there is bound to be a limit to that.
With my very initial prototypes (and I mean very initial) I already discovered that the operations will be quite math intensive. I didn't experience any issues in Javascript, though I was running Chrome the supposed king of Javascript, though it was with only one window. I imagine multiplying this by at least dozens would have an impact on performance. I also think a compiled application will give better performance.
I'd love to hear any thoughts or suggestions. And check up on my progress here: 2bdev. Time to try out some psuedo-object oriented Javascript!
