I do not understand. As far as I can see, several big companies that are oriented towards profit, such as IBM and H-P, are doing little to improve their incomes. Consider IBM. The company sells hardware and consulting. For it, the GNU/Linux operating system is an inexpensive `complementary product'.
So why do people still find some GNU/Linux applications hard to configure? Why do some applications lack good documentation? Why do some applications come without an easy-to-use `advanced configuration' option for those who want a bit more than the standard customizations? (These latter people are often opinion leaders; they are respected for their knowledge; people listen when they talk.)
In terms of business, IBM is like an airline that carries passengers to Miami. The airline wants Miami hotels to charge less. This is for good reason: when hotels charge less, an airline sells more tickets. For the airline, the hotel rooms are a `complementary product'.
The `complementary product' is not gratis: for example, airlines advertise destinations such as Miami; but for airlines, the product is inexpensive.
Conversely, for hotels, airline seats are a `complementary product'. When the cost of travel comes down, more tourists visit Miami. When they visit, they purchase more hotel rooms.
Similarly, when a company sells computer hardware and consulting, it can expect to sell more when its `complementary product' becomes less expensive.
By expense, I mean costly as a customer measures costly. A difficult-to-configure program will be costly to a customer who values his or her time. If a distant company wastes its customer's time, the customer may not return.
I can understand programmers who develop software that is awkward to configure. Eric Raymond wrote a horror story about his experience with one such program.
Programmers are not regular users. What is important or obvious to them is different from what is important or obvious to ordinary `users'.
But IBM does not hire only programmers. It hires accountants. It has corporate executives. Both gain when their company gains. From their point of view, they have a motive to better their company's complementary product.
Perhaps the people at IBM figure it sells only to sophisticated purchasers who use IBM equipment and services for servers. Perhaps those at IBM exclude corporate purchases of `desktop' machines. Or perhaps the decision makers figure that `desktop' purchases will be small, and as Christensen describes 1, exclude such possibilities.
Either way, I think the managers in IBM (and at other similar companies, such as H-P) have made a major mistake.
I hope that rather than misunderstanding, I am simply out of touch. I hope that indeed many large companies (and trade organizations, too) are undertaking `human factors' studies. My hope is that these studies have not yet been described on the various free software sites that I frequent, such as Slashdot and Debian.
That notion is not impossible. It took me awhile to become aware of the study that Sun undertook a while back. (Unfortunately, it looks to me that Sun did not follow up well. If they had followed up successfully, I would be praising them. And, in any case, one study is insufficient.)
Return to: Notions
Or return to: Rattlesnake Home Page