A year or two ago my boss was on a kick to put some pressure on Microsoft in terms of how they were licensing their Office Suite to Higher Ed. At the time, Sun Microsystems was pushing their Star Office as a viable alternative to the Microsoft version. We had more than one discussion at our Cabinet about how we might implement on campus. For my part, I went so far as to buy myself a very inexpensive Linspire (formerly Lindows, a linux OS that has a user interface linke Windows) computer and try out that OS. My argument was that anybody who was working on a Windows box would use Microsoft Office products. But if somebody was operating in a Linux environment, then using open alternatives would seem reasonable.
The initiative failed miserably. The story I was told is that Sun didn't follow through. But the demographics didn't make sense, in my opinion, so it was doomed from the start. My guess is that the people on campus most likely to try non-Microsoft offerings are Mac users. But Star Office doesn't have a Mac version. We should have tried with Open Office instead. (Sun has played a background role in Open Office but Star Office and Open Office are not the same.) If we did that, however, we would be on our own. Sun wouldn't have been involved.
I did learn some things from this experiment. Failures do have their lessons. I spoke to quite a few people in CITES about providing lab computers cheaply. Most of them (and these folks had real knowledge about support) were dead against things like Sun Ray terminals, which like Citrix, run desktop software off a server. They said this stuff is too slow or too unreliable. Also, they would really like to buy their computers from a known vendor - Dell has been the standard for quite a while in the CITES computing labs. The feeling was that the Linspire computers which one can buy at Walmart actually have a higher total cost of ownership or their performance is much lower.
I would concur with the latter verdict. These boxes used chips from AMD rather than from Intel and the box I got was from a company I had never heard of. I don't know what Microsoft gets from Dell or other vendors for the Windows software, but just checking the Dell site this morning there is a starter desktop system that one can get for $300 (with rebate). This is a full system that comes with XP home edition. From the Linspire site, I found this very cheapie box but note that it doesn't come with a mouse, keyboard, monitor, or speakers. If you compare the two processors, the Dell box is 2.4 GHZ with a Celeron processor. The cheapie box is 1.5 GHZ with an AMD processor. I've got nothing against cheapie, but this is an apples and oranges comparison.
So how does any of this apply to K-12 and High Ed outreach there? The answer is simple. Reliability is worth a lot. The approach that works must have computers that are reliable and that can be maintained by whatever staff the schools have. At present it is my belief that Apple still has the lion's share of the market in the schools and that is because of its historic give away programs. If that is right, that market share itself makes Mac the low cost product because that is easiest for the schools to support. In recent years when Apple was in trouble, some of the schools may have switched to PC in which case they probably have Dells or Gateways. Those are the environments that are there and those are what we should design to now, if we're doing outreach.
The society as a whole must embrace open software first, before it makes sense to use that approach in the schools. Will that happen? I have no idea. There is such an installed base for the Microsoft stuff, I have my doubts. But most kids don't learn Excel before coming to college. (They do learn PowerPoint and Word.) If somebody in the open source universe cleverly de-bundled the spreadsheet and database pieces, they might make some more penetration. Here's betting that is not likely to happen.