More on the length of sequences
The shortcoming of our experiments was that they demanded too much. In our early experiments, we often went to almost unbelievable lengths to get some new process to be implemented, and to get it to work. But the amount of effort we had to make to get it to work — the very source of our success — was also the weakness of what we achieved. In too many cases, the magnitude of special effort that had to be made to shore up a new process was massive — too great, to be easily or reasonably copied.
We succeeded because we replaced an existing system with a large system in which every aspects of procedure, process, attitude, rules, were changed: it worked. But such an inhuman effort was not easily repeatable. It was, indeed, almost impossible to repeat, and unreasonable to expect that others would repeat the nearly superhuman effort needed to make the process work.
Toward the end or our first generation of experiments, even we ourselves could scarcely muster the special energy needed — emotionally, politically, financially — to sustain our effort.
Often a problem with software, too: software is often designed to replace too big a process and then doesn’t get adopted.
Our new or revised living processes were too “large” to be widely copied. Some path-breaking process models take a very great deal of time to think out, simulate, test in real circumstances. Ours often had too many new components. They required too much to change. It was therefore hard to adopt them in their entirety, and therefore also unlikely that they would be spread widely, or be copied widely.
This inspires me to look at the product/app design process from a perspective that asks, “What’s the minimal change to the operation of the existing system — which could be just the operating system — that can be made to repair something that is broken (= improve something)?”
Nevertheless, in spite of its limitations, the co-housing process has been copied all over the United States because it is compatible with 20th-century professional definitions of architect, contractor, and so forth, and precisely because the deep roots of sequence change that would be needed to achieve a fully living process are not required by it.
When models of a new process are too intricate, too complete, too indivisible, they require specially trained people to carry them out, they put unworkable demands on the practical social system in which the innovation occurs. Also, people who can act according to the new process which has been worked out, are very highly skilled, hard to train, and above all rare. There are not many of these people around.
So, for that reason, too, it is extremely hard for the new process, even when it has been shown to work, to spread — because it is too hard to do.
#book/The Nature of Order/2 The process of creating life/20 The spread of living processes throughout society#