James Stanley

Incongruous technologies

Sun 2 July 2023
Tagged: science, futurology, philosophy

Take saws, for example. We have saws for cutting wood, saws for cutting metals, saws optimised for cutting curved shapes, electric saws, computer-controlled saws, and so on. Saws are pretty well explored. Saws are general-purpose. You probably won't discover a new use for saws.

But then there are some technologies that just seem so incredibly far advanced compared to other things that are possible, but also are only applied to incredibly specific applications.

Watchmaking is one example. We can make these unbelievably tiny systems of springs and gears that can keep time quite accurately, and then when you look for other things using the same sort of technology, there's just... nothing. Really? There isn't a single application of unbelievably tiny systems of springs and gears apart from making watches?

Integrated circuits are another example. For some reason we can make ICs with such precision that we're almost placing individual atoms, but it can only be used to make integrated circuits? There's literally nothing else we can make with this technology?

Feynman wrote about "micromachines", where IIRC the idea is you make a machine that can produce slightly-smaller versions of all of its own parts, and then repeat the process until the machine is arbitrarily-small. Obviously it's harder than that makes it sound, because the tolerances have to go down in tandem with the overall dimensions, but I don't see why in principle it wouldn't work.

I think part of the problem is that watchmaking and IC fabrication are both highly specialised fields, and people outside these fields kind of forget, or don't know, that the techniques are possible. If someone outside watchmaking wants to design a small mechanism, and they start imagining gears and springs in their head, the smallest reasonable-seeming scale is still maybe 10x larger than what is typical in watchmaking. You just wouldn't assume that you can pack dozens of gears into an object about the size of a large coin. You wouldn't even consider such a design, it is so obviously outrageous.

Now I don't personally have any designs in mind that can make use of watchmaking-scale gears, or IC fab-scale whatever-sort-of-stuff-it-is-that-ICs-are-made-of, but that's exactly the point: there aren't any other applications because people don't have any other ideas, but I think the reason we don't have any ideas is because we forget that it's possible. So I think it would be worth trying to bear in mind that these things are possible, so that we don't overlook possible inventions that could pop into our heads.

If you like my blog, please consider subscribing to the RSS feed or the mailing list: