To add to Freddie's comments:
Like many IT Projects of this nature, our journey has been one of evolution. We have learned through hard knocks as well as trial and error what works and what doesn't.
To my mind our current iteration is the best of all worlds.
We are purposely NOT using legacy PCs - which we found to be problematic (because they were old to start off and inevitably the hardware failed through no general fault of the equipment - it was just old). Consequently, people were sometimes equating the hardware failures to the Linux systems, which was unfair, but that is the reality. By using good quality "diskless" clients we have achieved a high degree of confidence in the hardware as well as the Linux system - this I believe is of paramount importance.
As well by using good "diskless" clients we are now able to offer accelerated graphics which supports streaming video and the desktop environment Beryl3D. When kids and staff get on the systems now they are not only "satisfied" by how reliable it is, but stunned by the desktop - KDE + Beryl3D. It not only IS state-of-the-art, it looks state-of-the-art and easily rivals Vista and, we believe, greatly improves upon it.
As well by incorporating other features such as:
- Multilingual support when staff and students login (all menus and programs are automatically translated to the language of choice)
- Secure, remote access using FreeNX
Students are not only NOT receiving a diminished set of tools but getting something much beyond anything we could have given them using commercial tools - at any price.
Although, I am the IT Manager of this school district, I am truly fortunate to work with a very talented and amazing development staff. Under the leadership our Systems Analyst, John Cuzzola and his very talented team (Freddie being one of them. with Dean Montgomery), we have been able to prove conclusively that not only can you provide high quality educational tools using open source software, but improve on what is available commercially at a fraction of the cost.
Finally School Districts which are hoping to introduce this type of system into their districts require the following:
A Visionary - someone high enough up the administrative chain who can see the efficacy of the system and drive it
An Expert - someone who has a high degree of Linux/OS skills who can make it work and support it
A Champion - an educator - school-based administrator who will be able to "sell" and provide the school vision
It's definitely not an easy road, particularly in this world of Micro$oft FUD and other fallacies but it definitely can be done.