It all started with a desk. My lady needed a new one, and we were looking at a number of possibilities. She had found what looked like the perfect computer desk: the center area was just the right size for a 21″ display; the shelves were exactly the size of CD cases; and there was a specialized nook just for storing the printer. I looked at it for a bit, and got suspicious: “What happens when you go to a dual monitor configuration? And what if you get a printer that’s two inches taller than your current one.” We didn’t have a good answer. It didn’t take much looking to realize that there were a lot more computer configuations that would fail to fit in the desk than there were configurations that would fit. With a bit more searching, we found one that didn’t look like it was custom made for her current configuration, but would work just fine anyway, and would also work for her next 3 configurations. This started me thinking about the dangers of over-specialization.
However, the story doesn’t end here. The desk was for a new house, and this house’s kitchen came equipped with an electric motor. 
Built into the counter is a speed dial and a rotating spindle, to which you can theoretically attach whatever accessory you might like. It was made by NuTone, and you can judge the success of the concept by the fact that there are absolutely no references to the concept anywhere on the company’s site. Our collection of accessories includes a mixing bowl, a blender, and a juicer, but we don’t have (and wouldn’t know where to get) the food processor, meat grinder, or knife sharpener. We’ve never used it. Why would we, when we can find 7 blenders with more features for under $20 at Kmart. In this case, we find that what seems like a wonderfully useful device is woefully underspecialized.
So what’s the deal here? Someone designs the perfect desk for the job, and I want a general tool I can adapt. Someone designs a general, adaptable tool for the kitchen, and I insist on buying the perfect tool for the job. Am I just impossible to please? Yep — you got it on the first guess. However, I think that there are some principles that can be applied to determine when specialization is appropriate:
- The target task is well-defined and generally agreed upon.
- Either the task’s frequency or the incremental benefit of specialization is high.
- The cost of ownership is low.
Point one is fairly important. Blenders have become so much a part of our lifestyle that everyone knows exactly what to expect, and what you can do with them. You’d be hard pressed to find a cookbook that doesn’t tell you to use a blender for some task. Desk’s are likewise well understood, but only up to a point: almost every desk will have a flat surface suitable for writing or supporting objects, and will have at least one drawer. Computer desks will typically have facilities for a keyboard in the center. However, there is no agreement upon exactly what standard variety of objects will be placed upon a desk, or how they will be arranged. Individual tastes vary, as do varieties of office equipment. This places a limit on the degree of specialization which is appropriate.
Point two is equally important, but perhaps less obvious. You may not need a corkscrew more than once a year, and may never need a fire extinguisher, but when you do need it, nothing else will do the job right. On the other hand, there are many things that can substitute for a cigarette lighter — but if you are a smoker you will use it so often that you’ll want it anyway. Many bachelors will survive for years without a blender, but the first time they try crushing ice for daquiris without one, they may find that it’s a necessity after all. (It has come to my attention that my mother sometimes reads this blog. I should note, for her benefit, that neither of the preceding sentences should be taken as indications of my lifestyle.)
The third point ties closely to the second. The nutmeg grater may provide you with only a very small benefit, but it’s easy to tuck into a drawer and forget. The wine press and the butter churn, on the other hand, tend to get in the way, so you’d better use them a lot. (Note to Mom. See above.)
After all of this build-up, I can finally get around to the title of the post. (You probably thought that I forgot, didn’t you.) The TiVo is an excellent example of a specialized device which has found the perfect balance. In fact, it bears exactly the same relationship to a personal computer that a blender bears to the NuTone kitchen appliance. Under the hood, it is a full-fledged Linux workstation pretending to be a VCR. (Looked at in this light, a blender might be considered an electric motor that has forgotten how to sharpen knives or slice carrots.) Let us see how it satisfies our three rules:
- Everybody knows what a VCR does, and TiVo takes that as a starting point. You can record shows and watch them later; you can pause, rewind, and fast-forward. It then mixes in other fairly well understood tasks: organize saved shows; search for new programs; etc.
- The average household spends a lot of time watching TV. Optimizing and improving this common activity produces a large payoff for this average household. For the occasional TV family, the benefit can still be high, since it increases the likelihood that the content they see will be both high-quality and easily viewed.
- Initially, the initial cash outlay for TiVo was prohibitively high, and limited it to only those who would gain a serious benefit. This has been reduced greatly, and it is now within the reach of most households. The other costs are low, as the form-factor of a TiVo is well suited to the average TV cabinet, and can easily augment or replace standard VCRs. By simplifying the programming process, TiVo actually bears lower mental costs then the VCRs which preceded it.
That’s it for now, but I’ll probably return to this theme every now and then. If you spot any particularly good examples of over-specialization, under-specialization or “Goldilock’s approved” just-right-specialization, drop me a line. I’m always keeping an eye out.