Rob Oakes – Oak-Tree Blog » Reduction and Selective Exposure

But even though reduction is essential in the testing and presentation of scientific thought, I’m starting to realize that it is often far-removed from the processes that generates, incubates and prepares it. It is simply too neat and tidy. And discovery and innovation is anything but tidy.

Robin Sloan of Snarkmarket provides a rather neat summary of my thinking (I wish I’d have written it myself):

[Ideas] come from weird combinations, of unexpected collisions that reveal new and interesting relationships. [To produce something innovative] you need to pile it all into your brain and wait for interesting things to happen, not necessarily on the conscious level! [You throw] it all in the brain pot and allow mysterious reactions to occur.

To anyone who is engaged in research, design or science (or training those who will be), it seems like we should be encouraging mysterious reactions. When conducting an experiment, a scientist or thinker may draw inspiration from many sources. She may seek out statistical advice to determine the size of a trial, related fields to see if a more efficient technique exists, or unrelated fields to seek new thrills. Casting a large net in the interest of filling the pot is a spectacularly good thing. It increases the probability that something new and interesting might emerge.

(I think it also requires more than the lip service that we pay to “cross disciplinary collaboration”. The experience of others is simply no substitute for personal familiarity, even if it is relatively superficial. After all, subject experts are most useful when you know the right sort of question to ask, and formulating the right sorts of questions requires knowledge.)

To see why esoteric knowledge can be valuable, consider how high quality typography made its way to the the personal computer. During his time at Reed College, Steve Jobs took a course in caligraphy. This one course sparked an interest in handwriting and typesetting and years later, while working on the Macintosh Project, he hired a team of typesetting experts to create a high quality type and font engine for his new operating system. This group took many of design conventions inherent to high quality typesetting and strove to bring them to desktop computing.

There are countless other stories of designers, writers, scientistis and engineers deriving inspiration from information that was (at best) peripherally related to the question that they were pondering. In some cases, the connection might have only existed in the mind of the individual that made it. When Philo T. Farnsworth, for example, was working on the technology for television broadcast, he solved an important technical problem by understanding how a continuous signal could be broken into uniform sets of information. Where did the inspiration for this technical advance come from?

From his childhood on a farm.

It turns out that there is a great deal in common between the way that a tractor plows a field without ever needing to raise its blade and techniques used to break an image into a single signal of information [1, p. 17-19]. Of course, after the relationship is explained, it becomes obvious. But isn’t that the way of most brilliant insights?

Now what happens if the right ingredients don’t get added to the mix? Or worse, if you deliberately limit your exposure out of the interest of specialization? (Increased depth of knowledge versus breadth of knowledge, better use of constrained time, etc.) Might it be that casting a narrow net actually has a negative effect? Could it stunt the types of questions that can be asked or the variety of ideas that can be pondered?

Whenever I confront the increased emphasis on specialization and the need to seek “selective exposure”, I find myself raising these sorts of questions. Reducation and specialization certainly have their places, but they aren’t at the beginning of the scientific process. When you’re still trying to figure out what questions you’d like to ask, that’s hardly the time to start throwing options away.

I’m willing to bet that the father of reduction himself might even be willing to agree.