Tablets? Notebooks? PDAs? Mac Minis? Supercomputing grids? What should you buy your users?

Every user is different and brings a host of requirements to the table.  I received a fair amount of mixed feedback to my post "Should I have bought those Mac Minis?"  A few readers applauded my understanding of my Mac-loving users, while others told me I should stick to my guns and standardize, standardize, standardize.  This, of course, begs the question, what should you actually buy for your users and when does satisfying requirements become pandering?

Obviously, I need to set policy, but as I've said before, I also need to give a very heterogeneous group of users the best tools I can afford.  Just as obvious is my tongue-in-cheek tone above.  My social studies teachers, though highly qualified and incredibly smart, have no use for a supercomputing grid, even though the capability is built into most modern Macs and a lot of Linux distros. I'm sure that a lot of teachers could find a use for tablets (Windows or otherwise) with some modicum of training.  They already use notebooks quite deftly and the English department head just about had a stroke when he saw his Mac Mini lab set up.  Students have already begun laying out the next edition of the student newspaper on the new machines.

I still haven't made believers out of all of the staff at school as we get ready to roll out thin clients in three labs.  They understand when I explain that since there is only one of me, something that gives more kids access to computing but reduces the actual number of computers I need to keep running is a good thing.  However, when I start talking computing clusters for redundancy and fault-tolerance and have to explain why they won't be running Vista, eyes tend to glaze over.  Better yet is the look I get from the 30-year veteran whose 25 PCs are being replaced by said thin clients.  She didn't care for the transition from IBM Selectrics, either, though, so I guess I need to take it with a grain of salt.

My point is that our users often don't know what they want or need.  It makes sense that we should all have brand-spankin' new computers of some flavor on our desks or our laps.  It makes sense at first blush that every student should have a laptop, too.  However, we have yet to see any real benefit or an appropriate return on investment for one-to-one computing.  Us IT-types also know that 90% of our users couldn't tell the difference between the performance of a Sempron and a Xeon. 

Which means that we can define a reasonable set of criteria for most of our students and staff and provide satisfactory computing facilities (IMHO, best via relatively low-cost Windows or Linux PCs or, depending upon your network infrastructure, Terminal Service or LTSP-based thin clients) that satisfy most of their requirements quite handily, even if they don't know what those requirements are.  It's the other 10% of the users that make this a bit more complicated.  These are usually the folks who will most be able to exploit the equipment we give them, as well, and their ability to maximize utilization must be considered.

This isn't to say that you should blow your budget on a lab of Mac Pros so a teacher who videotapes weddings on the weekends can teach a class in professional video editing.  However, I stand by the principle that some well-placed specialized resources (tablets, for example, or even a Mac or 2) can benefit both teachers and students.