Relativity and why we still need new computers

One reader of my last post on the saturation of the PC market in Japan (and developed markets in general) took exception to my language regarding extremely inexpensive PCs and their utility for ed tech. He wrote:

I think we need to be realistic about what both "extremely inexpensive" and filling the needs of students "remarkably well" mean. For example, computer which cannot make it through the school day on battery power does not in fact meet students' needs remarkably well. A computer which cannot be purchased and maintained for every student, indefinitely, within the district's operating budget is not extremely inexpensive.

I thought that his comments were actually worth a post of their own, as he brings up an important point and clarification. I want to make it clear that I'm not speaking against lifecycle funding of equipment, advocating for students to just keep using old computers because they are "good enough", or buying the cheapest of the cheap computers because they are, well, cheap. I have lived in a district that has only recently begun to retreat from this mindset and wouldn't wish it on any student or IT admin.

However, my takehome message from the original article was that we are seeing a fundamental shift in the way students as consumers (and we as technology administrators) can purchase equipment. Many students even no longer see the need for bleeding edge hardware, especially as computers are increasingly used for communication (or are replaced by convergence devices). While students are happy to eek out an extra year or two from the computers they have, most of which are remarkably powerful relative to the state of the art even 5 years ago, IT folks have a different set of concerns (but can still benefit from this shift).

As IT staff, our job is to make sure that equipment meets the needs of students and staff reliably and consistently. For us, eeking out another year of life from a lab full of computers beyond a 3 or 4 year lifecycle means increased downtime and maintenance headaches. Our users' requirements aren't met. As the reader points out above, computers can't be maintained indefinitely.

For us, the commoditization of parts simply means that as we can justify and fund more easily than ever real lifecycle management of our hardware. For 95% of the settings in which students and staff use computers, new computers are a necessity to ensure reliability, not because they really need dual quad-core processors. Advances in power management and reasonable increases in speed to support new features are great, but, again, relatively speaking, can be had cheaply.

Relatively speaking, computers that meet student and staff needs very well are extremely cheap, as are innovative solutions like thin clients and various appliances for accessing Web applications. Should we buy our users junk or try to get a 5th year out of a lab? Of course not. Should everyone have high end, underutilized hardware? I don't think so. My recent series of articles on the Intel Classmate suggests just how much computing benefit can be had from hardware well-suited to an application, at costs that many of us can stomach.