It outlines how Largo, in Florida, uses KDE 2.1.1 on 400 NCD thin clients to cater for 800 users. Centered around a single dual 933Mhz processor server, that has 3GB of memory, 18GB of hard disk and redundant power supplies.
While developing wayV (a free software project that allows people to interact with X via gestures - http://www.stressbunny.com/wayv/ ) I've tested as many input devices as I can get my hands on. NOTE: this occurred over the last year and a half.
A few interesting things have emerged (and yes, I'm stating the obvious in point 1):
1. people interact with User Interfaces in two primary modes, which they switch between. Note: people could switch modes every 30 seonds while others don't for hours.
The first mode is pure keyboard, people only use the keyboard to navigate around the user interface. The second is pure pointer (mouse), people only use the pointer - even though its command vocabulary changes from application to application, i.e. different menus, icons, different positions of icons and menus, etc.
Yes there is a third mode which is a fusion of both keyboard and pointer but that tends to exist while switching modes, or with people who are just learning about computers.
2. different devices REQUIRE different types of gestures.
The trackball is great for inputing loose shapes and if you want some accuracy use a mouse but for even greater accuracy use a tablet.
I'm the author of wayV http://www.sstressbunny.com/wayv which allows you to control your X applications and window manager via gestures. You can associate gesture with keypresses and other actions, e.g gesture left and switch to a different desktop, draw an O to start Opera, etc.
Gesture based computing has been around for a long while, though previous implementations of it were rather limited and application specific. Black & White isn't the first application to have it, it's merely the first popular application to have it.
With regards to a standard set of gestures. Yes that definitely needs to be decided - so please mail me if you're interested in that.
BTW future steps for gesture computing include inputing gestures via a camera - and on and off I work on that:)
Slashdot Mirror
It outlines how Largo, in Florida, uses KDE 2.1.1 on 400 NCD thin clients to cater for 800 users. Centered around a single dual 933Mhz processor server, that has 3GB of memory, 18GB of hard disk and redundant power supplies.
A few interesting things have emerged (and yes, I'm stating the obvious in point 1):
1. people interact with User Interfaces in two primary modes, which they switch between. Note: people could switch modes every 30 seonds while others don't for hours.
The first mode is pure keyboard, people only use the keyboard to navigate around the user interface. The second is pure pointer (mouse), people only use the pointer - even though its command vocabulary changes from application to application, i.e. different menus, icons, different positions of icons and menus, etc.
Yes there is a third mode which is a fusion of both keyboard and pointer but that tends to exist while switching modes, or with people who are just learning about computers.
2. different devices REQUIRE different types of gestures.
The trackball is great for inputing loose shapes and if you want some accuracy use a mouse but for even greater accuracy use a tablet.
BTW I used a trackball the whole time.
Ok - thats removed - clearly some people aren't used to the concept slagging yourself off :)
Happy now?
With regards to a standard set of gestures. Yes that definitely needs to be decided - so please mail me if you're interested in that.
BTW future steps for gesture computing include inputing gestures via a camera - and on and off I work on that :)