Re: NIT now vs Intel's solutions later (MIDs)?
 

I don't think the applications are meant to be ported directly across platforms. What can and should be is technologies like dbus, ssh, samba and nfs. Those combined with unique (fitting the platform) software that bothers to be developed properly (in the way that ensures integration, which is conveniently what dbus is all about) means nothing is missed, even if the smaller gadget doesn't run the same web browser as a desktop computer.

The NITs allow that just fine, which is why I am concerned Intel may be taking this in the wrong direction. As JohnX mentioned, it's not going to work as a tiny desktop, so why push so hard in the direction of performance?