Make vs. Buy

The always prolific Glenn Reynolds had an article up this week on Tech Central Station. Some guy is apparently preaching the new gospel of personal fabrication. I think it’s time for a little reminder in economic realities.

The idea seems to be that in the nebulous but not too distant future, people will be using some unspecified blend of hardware and software to build their own fantastic collections of personal items. I’m dubious, for little reasons like economy of scale and design difficulty, but I’ll let that part of it slide.

The part that kind of amused me is this one:

Gershenfeld writes that it's possible to do a surprising amount of general-purpose personal design and manufacturing work by combining existing off-the-shelf components in new ways, and he spends a lot of time talking about the results of his experiments in that direction. His discussions are very interesting, but to me the most interesting thing was his discovery that lots of people want this kind of capability -- not because they hope to make money out of it, necessarily, but because they want to be able to make things for themselves that they can't buy elsewhere.

There’s a gaping hole in the argument here that should be familiar to everybody that does product design for living. There is an old story about a manager at Black and Decker walking into a conference room and asking the assembled group why people bought drills. A variety of reasons are proposed, but none strike him as correct. The reason people buy drills, he points out, is not because they have an intrinsic need for a drill. They want a hole. A drill, like all tools, is a means to that end.

In general, people do not “want to be able to make things for themselves”. What people want is the actual thing. People want the custom item, not the ability to make it. The actual manufacturing process is profoundly irrelevant to most people. If magical gnomes built it for them, they would be just as happy, provided they got what they wanted.

Can you build exactly what you want in terms of some consumer items rather than buy one off the shelf? Sure. I’ll use furniture here as an example. I can make the perfect bookcase for any room in my house. I have the knowledge, the tools, and the skills to pull this off. My house is full of bookcases that came in a box from IKEA. Kind of odd, isn’t it? The problem comes down to time and the opportunity costs thereof. All things considered, an IKEA bookcase takes about 4 hours out of my life. I’m figuring two hours, round trip, to IKEA. I’m budgeting another hour to wander the giant maze, get the shelves, and get out. Then it takes another hour back at YPS Manor for assembly. At the end, I have one bookcase. I’ve spent a Saturday morning, but I now have a place to keep more books.

Alternatively, I can spend two hours designing a bookcase. Then I can spend a couple of hours procuring the necessary materials. Then I can spend a few more hours cutting, drilling and routing the materials to size and final dimension. I now can spend a few more hours assembling the case. Then I have to worry about finishing the thing, which takes even more time. All told, I’m probably doing good to get one bookshelf built in a weekend. The opportunity cost of a lost weekend for me right now is pretty high. So it actually would take me more like a month and a half, start to finish, because I can't devote a full weekend to it at any given time. It's going to be an hour here, a couple hours there, and finally, I'll have a bookshelf.

Similarly, for most people, the opportunity cost of building things themselves is pretty high. Learning the skills necessary to do their own design/builds is also not trivial and another opportunity cost to consider. Even the dodge about assembling pre-made components has to reflect this. Figuring out how to do that is going to be time-consuming for a lot of people, not to mention the fact that design is a skill in and of itself. All the skills must be learned in addition to the actual fabrication time involved. Until the learning curve and the opportunity costs come down, most people will buy rather than build. I don’t question the assertion that people are interested in customization. I just question the idea that they are eager to do the work themselves.