Slashdot Mirror


Windows 7 Multitouch Demonstration

Starturtle writes "Bill Gates and Steve Ballmer have shown a small snippet of the upcoming Windows 7 at Walt Mossberg's D: All Things Digital conference. It seems like the Windows team have switched their focus for inspiration from Mac OS X to the iPhone OS. Multitouch is the biggest addition, and will appear system-wide, usable anywhere. The most interesting part of the touch UI is not the eye candy, it's the Task Bar, which seems to have morphed into a pie menu."

2 of 329 comments (clear)

  1. Re:Alias/Wavefront the patent holder? by jcr · · Score: 4, Informative

    A/W may have gotten a patent on a particular form of pie menu, but Don Hopkins invented it.

    -jcr

    --
    The only title of honor that a tyrant can grant is "Enemy of the State."
  2. Re:Drivers by dgatwood · · Score: 4, Informative

    I'm losing my chance to moderate so I can reply to this. Yes, it is an OS feature. Simple gesture support for devices is easy to do in a driver, but is nowhere near what you really want out of multitouch. An ideal implementation should allow applications to deal with multiple simultaneous touches, drag events, etc. simultaneously. For example, an audio editor application should allow me to use three fingers to push three sliders simultaneously up and ride them while a finger on my other hand touches a mute button on channel 3 to pull it out of the mix because I'm planning to cut that 30 seconds out but haven't had a chance to do it yet.

    To handle such things, the application must be able to simultaneously get multiple touch events at different locations that indicate that a finger has gone down at a particular spot and now is moving in a particular manner. These finger events must then remain individually trackable. To handle this correctly requires significant extensions to the event system of the host OS, probably on an opt-in basis to avoid confusing applications that only support simple events like click/drag or lightweight touch events like zoom in/zoom out. Therefore, it pretty much has to be an OS feature.

    The only way I can think of to do this without OS changes would be to allow an application to capture the device and take exclusive control and communicate with it directly outside of normal OS channels (e.g. a user client). Those sorts of designs are okay for specialized devices like tablets that only one or two apps will ever care about, but they are hardly ideal for input devices that are intended to be general purpose.

    --

    Check out my sci-fi/humor trilogy at PatriotsBooks.