Hacker News new | ask | show | jobs
by johnnyb_61820 2620 days ago
Well, I guess that's debatable. However, the way I see it, the old version was presented as a fraction, but couldn't be used as a fraction. The new version is also presented as a fraction and can be used as a fraction.

Perhaps it is due to my old-school nature, but if I wrote a fraction in school that didn't work as a fraction, that would be a mistake, would it not? Especially if it was possible to write it correctly? I couldn't go to my professor and just say, "well, actually, while it looks like a fraction, it was actually just a piece of notation that needed future clarification."
1 comments
EdwardCoffin 2619 days ago
> Well, I guess that's debatable.

What part of it is debatable? The article clearly states that it is just a notational improvement: there was really no flaw in calculus, it was more a wart in notation. This change would not result in different answers unless the original calculations were done incorrectly.

I'd regard the current notation more like an irregular verb, or something that is spelled differently from how it is pronounced: it's inconvenient to deal with, but everyone adept at calculus knew about it and how to deal with it. This article is just describing a notational improvement that would make it easier to get right.

link
johnnyb_61820 2618 days ago
The fact that there was a well-known kludge to get around the problems inherent in the notation doesn't make the notation right. As the paper pointed out, the new notation didn't drop magically from the sky, it is literally the application of the quotient rule to the derivative. The reason that no one noticed this notation before, is that they didn't think to apply the quotient rule to the quotient dy/dx.

If I had a problem, and my solution involved forgetting to apply the quotient rule to a prominent quotient, it would still be wrong, even if I also came out with a set of kludges that allowed me to get right answers in the most common cases.

link
yters 2618 days ago
It seems similar to Roman numerals vs Arabic numerals. We could stick with Roman numerals, but mathematics would be pretty much at a standstill if we had to use such an unwieldy system.

A major improvement in notation is not insignificant.

link