Computers on the cheap

I'm not sure if you've noticed recently, but computers have gotten pretty darn cheap.  Like really cheap.  Maybe I'm just getting old, but back in my day, you couldn't go to your local discount store and pick up a decent computer for $300.  I remember when, aside from walking uphill both ways to school in the snow, computers with this much horsepower weren't available outside of the NSA, let alone at your friendly neighborhood Wal-Mart.
So what exactly does $300 buy you?  And should a cash-starved educator like me run out and spend my meager budget on a bunch of these little suckers?  It's tempting when the computers I have are, by and large, dinosaurs from the era during which I was walking to school myself (uphill both ways in the snow, of course).  Let's tackle the first question.  $300 dollars doesn't get you a monitor, although most of us probably have more than a few 15"-17" monitors sitting around.  It does, however, get you a 1.8 GHz AMD Sempron processor, a quarter gig of RAM, an 80GB hard drive, built-in Ethernet, and a CD-burner.  Heck, you even get a wheelie mouse!  Seems like pretty good stuff to me.  More than enough to keep my students writing term papers, cranking out PowerPoint shows, and otherwise trying to circumvent my Internet content filters.
What about the second question?  Should I buy a whole boatload?  Sempron's all around? While it's true that we as educators (at least in too many public schools) largely need to take what we can get, we still need to keep in mind lifecycle management.  What are our requirements?  What plans do we have for an eventual  tech refresh?  If we need several low-end computers for a basic academic lab (the above-mentioned Word, PowerPoint, and content filter circumvention), then maybe 25-30 of these represent a good deal.  On the other hand, what if a computer science lab is aging?  We're talking CAD, Mathematica, and Visual Basic, rather than Firefox.  What if administrators with critical computing needs have machines in need of replacement?  What if what you really need is a thin client setup with a serious server, or you need a new web server, or, or, or...
Chances are you need all of these things.  I know I do.  I also know that I can rarely afford the big ticket purchases in a given year and, more importantly, if I don't spend my entire budget within the first few months of the school year, it will probably be frozen or spent on heating oil instead of technology.  
Wouldn't it be nice if we could save some money towards these larger purchases? Keep in mind that low end computing needs can be addressed in the short term with donations, government surplus, etc. (more on this in a later blog).
Which means that our role as educational technologists is not just to run around putting out brushfires and keeping aging computers running.  We also need to be lobbyists - it's time that we convinced our administrators that long-term planning cannot just be applied to facilities, or we will continue to limp along, overrun with Semprons instead of building a manageable, diverse infrastructure serving a variety of needs.
The moral of the story here, folks, is that realistic IT lifecycle planning and management doesn't go away, just because we measure our budgets in thousands instead of millions of dollars.  We simply need to be more cautious about determining requirements vs. wants and resist, whenever possible, the call of an Ace bandage when a full hip replacement may be looming in our future.