I'm comfortable with command line interface (CLI), as many of us who have been computing since the early days are. From CP/M to DOS and into the various Unix/Linux branches, I've learned most of the basic and some advanced commands to get the job done.
Today, I still use CLI, usually on a couple of my rather vintage machines (my Tandy 102 and 200 and my Atari Portfolio) as well as my 1993 ThinkPad 500, which happily runs DR-DOS and BasicLinux. I like these operating systems because they are light, nimble and offer you an extraordinary amount of control over not only the software but the computer itself.
That I am comfortable doing so is a testimony to the fact that I have spent years studying and playing with these things, and to a certain extent hesitate to give the older ones up. It allows me to use computers that would otherwise be thought of as obsolete.
Not everyone shares my enthusiasm for the arcane, of course.
Nowhere is this a greater concern than in the task of trying to help disadvantaged families gain access to the Internet. Yes, they can use Windows or Macintosh, but a simpler approach is needed. What must be remembered here is that the primary task is helping the studen's and their families get this access. First and foremost, that is the goal.
Let's keep that in mind while we look at the subject of interfaces and operating systems.
What is really needed is an operating system that has a very small learning curve. While it is true that both of the major OS's in the world of home computers, Windows and Macintosh, are fairly intuitive, the simple fact remains that some skills are still necessary to do some of the more complex tasks.
What is needed is the approach that mobile computing has taken. Apple's iOS and Google's Android get out of the user's way once they are running. Both are somewhat light with system resources and very intuitive. However, both are limited, and in the case of iOS, proprietary. Windows Mobile platform supposedly offers the same, but they appear to be headed in the same direction as their Windows CE operating system, which held so much promise but failed to live up to it.
Sugar, the desktop environment that was initially developed for the OLPC XO-1 laptop, holds more promise than ever in providing an environment that stays out of the way of the user. While you can still learn how to customize and modify Sugar, it is not necessary to know how to do so out-of-the-box, so to speak; anyone can pick up an XO-1 and pretty much figure how everything is supposed to work.
Closer to the goal, but still not there. Perhaps it is too daunting.
The perfect system would run so far in the background as to be invisible to the user; they don't know, nor do they care. Or even need to.
It would be intuitive, and evolve with the user; as their skills grow, it allows more access.
They would truly own it. It would not be proprietary nor licensed. Once they have it, they have it. Would support be available? Yes. How would that be done? I haven't a clue at this point.
But that is the direction we need to be heading in. It has to be about what the computer is being used for and not about the computer itself.