7 ms·
Contrary to what this article claims, I would say that the motivation for such a switch has never been weaker. - The switch to metric has already taken place
by bostonpete 11y ago
Contrary to what this article claims, I would say that the motivation for such a switch has never been weaker.
- The switch to metric has already taken place in science and engineering (where it's most harmful to have 2 systems), largely because of the horror stories from decades past mentioned in the article.
- Packaging and labeling costs associated with having 2 systems is effectively zero because it's all automated.
- Everyone has a computer in their pocket capable of performing a conversion if they really get in a bind (though this has literally never happened to me)
On the other hand, people are generally content with using the imperial system so a switch would be met with a lot of resistance. This resistance, coupled with increasingly vanishing costs associated with having 2 systems leads me to believe that this switch is not going to happen in my lifetime.
TLDR; The metric crusaders need to find a new cause to pour their energies into.
- kelseyfrancis 11y agoAgreed. This topic has always felt to me like bikeshedding at its worst.
- gaur 11y ago> The switch to metric has already taken place in science and engineering (where it's most harmful to have 2 systems), largely because of the horror stories from decades past mentioned in the article. A nontrivial fraction of science and engineering is still done using an unholy mix of US customary and metric units. For example: laser optics frequently have diameters specified in inches and focal lengths in millimeters. Even for science experiments, things like screws and threads are SAE/UTS and not metric. Parts for machining are often specified in US customary units. Torque requirements are often specified in foot pounds. Non-metric units aren't dead in science, but I sure wish they were.