How about if we used the @ symbol instead of tau?

[josephlindenberg]

I've never seen an @ symbol used in an equation before, so all the complaints about tau already being used for things like time constants would go away.

It's conveniently right there above the 2 on every keyboard.

The @ symbol sort of looks like you're going once around a circle, although it seems to start from the lower right of the circle instead of the middle right. (OK, in the font this website uses there's an extra squiggle, but if you write your lower case a's the other normal way, there isn't.)

I wonder whether the use of the @ sign for email addresses could cause any problems with equations on web pages. Only if you had characters immediately before and immediately after the @ sign, followed immediately by a period and then more characters?

[xander]

The argument that tau is overloaded is, I think, a red herring. The issue is not whether or not tau is a better symbol than pi, but whether or not the idea that tau is meant to represent is more or less useful than the idea that pi represents. I kind of think that it is unfortunate that there has been as much debate about the symbol as there has been, and so little debate of substance (this forum being one of the few exceptions). It doesn't matter which symbol we use.

That being said, the @ symbol is probably a poor choice. That character has meaning in a lot of programming languages, including TeX and LaTeX, the typesetting language used by many mathematicians. I don't mind overloading a variable, because meaning can be inferred from context. However, computers aren't so good at context, and using a character that is traditionally reserved for some function seems like a bad idea to me. If tau really causes this much consternation, why not (a) go with the three-legged pi thing that Palais suggested, (b) pick a different Greek letter (the capital qoppa might make sense), or (c) pick a symbol from some other alphabet (perhaps the Hewbrew shin or the Russian ж), or (d) invent some new kind of squiggle.

xander

[SpikedMath]

I agree with xander, @ isn't the best choice. Some people think simply writing should be fine.

[xander]

Some people think simply writing should be fine.

:P

xander

[josephlindenberg]

The argument that tau is overloaded is, I think, a red herring. The issue is not whether or not tau is a better symbol than pi, but whether or not the idea that tau is meant to represent is more or less useful than the idea that pi represents.

I disagree.
FIRST, we have to decide whether or not the idea that tau is meant to represent is more or less useful than the idea that pi represents.
SECOND, if it is, then we have to chose a good symbol for it.

It's a two-part question. Both are legitimate issues. And I'm sure a lot of people like myself are convinced of the first part, but aren't so sure about the second part. In fact, the more convinced you are that 2pi is the constant we should be using, the more you want to make sure we pick an acceptable symbol so that other people won't resist switching.

[Mar]

Tau is also already used to represent another irrational number, the golden mean I like your idea for a different symbol.

[xander]

Tau is also already used to represent another irrational number, the golden mean I like your idea for a different symbol.

First off, that is by no means an entirely standard notation. I have seen $\Phi$ and $\phi$ used to represent the golden ratio and mean, respectively. Second, so what? There are tons of symbols that get used all over the place. Personally, I like Palais' three-legged $\pi$, but most packages don't have that symbol. Any symbol that is common to many packages will already be overloaded. It's a bit of a catch-22 if you want a symbol that is accessible and unique.

xander