Pi is wrong...according to the proponents of Tau Day (http://tauday.com/). And no, its not about legislating Pi to be three in accordance with the Bible, but a quite sensible proposal:


The Tau Manifesto is dedicated to the proposition that the proper response to “π is
wrong” is “No, really.” And the true circle constant deserves a proper name. As you may have guessed by now, The Tau Manifesto proposes that this name should be the Greek letter τ (tau):

τ≡ C/r =6.283185307179586…

Throughout the rest of this manifesto, we will see that the number τ is the correct
choice, and we will show through usage (Section 2 and Section 3) and by direct argumentation (Section 4) that the letter τ is a natural choice as well.


They argue, that Pi is the wrong choice for the circle constant and that the true circle constant should be what is currently known as 2π. After all, the fact that people have to use 2π all over the place (in sine and cosine functions, in Fourier transforms, in Gaussian distributions, in polar coordinates, ...) is a sign, that Pi might be a poor choice as a constant.

But because redefining conventions usually leads to a mess (I am looking at you, electric charge constant!), they propose to introduce a new constant τ, which equals exactly 2π. So a full circle would be τ radians, a half circle τ/2, a quarter circle τ/4 and so on.

And I really like this proposal, because implementing it would save me from having to use 2π over and over and over, especially when having to convert between frequencies and angular frequencies (which I have to do all the time).

And why this thread today? In the strange American way to write dates, today is 6/28, which matches the first three digits of τ, so today is Tau Day.

So what do you think of this proposal?

Tau is pretty heavily used as it is in science. Not in favor of the proposal.

Tau is pretty heavily used as it is in science. Not in favor of the proposal.

In most cases, Tau is used as a replacement for t if there are more than one t in the equation. It would not be that hard to use another letter in these cases (and most letters have multiple meanings in physics anyway, even Pi).

I actually hate, hate, hate using tau like that as a dummy variable in integration, gets frustrating the way the symbols look, so that's a strong reason I'd support this proposal.

However it would likely be rather problematic for people to adjust to and you'd still find situations with non-integer fractions or even powers and roots of pi and tau anyway. And we'd still have uppity mathematicians doing Fourier transforms backwards (calling the inverse the regular one and vice versa) and the like. On the fence here really.

Think of all the pi jokes that can be made, now try and think of a Tau one.
I think we have our answer

I know someone who would disagree with this:
http://images.wikia.com/simpsons/images/e/e4/Simple_Simpson_promo.jpg

Is 2π really such a burden? :confused:

To put things in perspective, it's much less of a big deal than, say, the cgs mess with universal constants.

It is really ubiquitous though and kinda silly to have the 2π everywhere, which it really is, so replacing it with tau gets rid of an unecessary factor, and there may be some hypothetical benefit to teaching younger students that would help them learn geometry and trig, but there's also the disadvantages to changing everything up.

Is 2π really such a burden? :confused:

Mathematicians (and derivative disciplines such as physics and computer science) are inherently "lazy". Hence other peculiarities, such as the existence of natural log (otherwise known as log base e).

Is 2π really such a burden? :confused:

It is not so much a burden as an annoyance.

For example it is common in my field to give frequencies as Ω / 2π = ... MHz to avoid confusion with angular frequencies. However this is slightly ambiguous as the strict mathematical reading would be (Ω / 2) * π, which is not what is intended. One could avoid this by adding parentheses and write Ω / (2π), but this makes the expression even more unwieldy. Writing Ω / τ would save characters and eliminate this ambiguity.

It is not so much a burden as an annoyance.

For example it is common in my field to give frequencies as Ω / 2π = ... MHz to avoid confusion with angular frequencies. However this is slightly ambiguous as the strict mathematical reading would be (Ω / 2) * π, which is not what is intended. One could avoid this by adding parentheses and write Ω / (2π), but this makes the expression even more unwieldy. Writing Ω / τ would save characters and eliminate this ambiguity.

I'm sorry, but 48/2(9+3) = 2. Always.

I think teaching-wise, tau is better than pi. Something about a circle being tau radians is more intuitive than a circle being two pi radians.

Added bonus - no more stupid math assignments on March 14th! June 28th is summer vacation! :D

but pi r squared!

It's not as bad as the gamma function,

gamma(n) = (n-1)!

What's all that about?

Pi is wrong...according to the proponents of Tau Day (http://tauday.com/). And no, its not about legislating Pi to be three in accordance with the Bible, but a quite sensible proposal:

The bible does not say that Pi=3. http://www.purplemath.com/modules/bibleval.htm

The bible does not say that Pi=3. http://www.purplemath.com/modules/bibleval.htm

I knew when I saw you were last respondent in this thread, this would be what you were posting about.

Getting around the fact you're making assumptions about things not written into the Bible (if I have to pidgin hole an answer in, I'd pick the measurements used on that page as well!), are you claiming that the Bible, the inerrant word of God, says Pi = 3.1395348837...?

but pi r squared!

Having .5 * tau * r^2 puts it in line with the scores of other half-square formulas (e.g. energy stored in a spring, energy stored in a capacitor, etc.)

Is 2π really such a burden? :confused:

Having taken way too much math at university I gotta admit tau would have made calculations and formulas simpler.. but only a bit.

It's not really a huge deal.

I'd come across the issue before - I think they make a good point, and it probably would have been a more natural choice of a constant.

OTOH, I think making the change is more trouble than it's worth, and part of me feels that anyone insisting on using tau is in the same category as someone you meet talking about his "Mibibytes".

It's also not clear to me it's always better - obvious example being the area of a circle (and I'm not persuaded by section 3, yes some physical equations have a half in them, others don't, e.g. e=mc^2).

It is not so much a burden as an annoyance.

For example it is common in my field to give frequencies as Ω / 2π = ... MHz to avoid confusion with angular frequencies. However this is slightly ambiguous as the strict mathematical reading would be (Ω / 2) * π, which is not what is intended. One could avoid this by adding parentheses and write Ω / (2π), but this makes the expression even more unwieldy. Writing Ω / τ would save characters and eliminate this ambiguity.
TBH, it would be easier to just make "2π" a symbol in its own right. I mean, I know when I see "2π", I just treat it as a single character in my head. Join up the "2" and the "π" and call it a new symbol pronouced "two-pi". Not every constant needs a Greek or Latin letter -- we can make our own!!

I knew when I saw you were last respondent in this thread, this would be what you were posting about.

Getting around the fact you're making assumptions about things not written into the Bible (if I have to pidgin hole an answer in, I'd pick the measurements used on that page as well!), are you claiming that the Bible, the inerrant word of God, says Pi = 3.1395348837...?
a cubit is an approximate

There's only one definition for pi. (Not really, but all are equivalent).

It's the smallest x > 0 such that

x = 0.5(pi) => cos(x) = 0

(in radians, of course).

a cubit is an approximate

But the Bible is the word of God, inerrant and definitely not up of approximations. If you allow for approximations here, you open a can of worms.

But the Bible is the word of God, inerrant and definitely not up of approximations. If you allow for approximations here, you open a can of worms.
You are aware that if they didn't approximate you'd have an infinitely long book right? 3.14 is an acceptable approximation

TBH, it would be easier to just make "2π" a symbol in its own right. I mean, I know when I see "2π", I just treat it as a single character in my head. Join up the "2" and the "π" and call it a new symbol pronouced "two-pi". Not every constant needs a Greek or Latin letter -- we can make our own!!

The first one to call Pi wrong (http://www.math.utah.edu/~palais/pi.pdf) actually proposed something like this by joining two π to a new symbol. If someone finds a good new symbol that might work, as it has with h-bar for example. But in the age of computers it would require quite a lot of momentum to make all the relevant software to support this symbol. Momentum that might be difficult to acquire without such support. In contrast, using a Greek letter is much easier.

But I agree, that the symbol used for the new constant does not matter as much as convincing people that a new one is needed.

You are aware that if they didn't approximate you'd have an infinitely long book right? 3.14 is an acceptable approximation

Of course, but the Bible doesn't say, "roughly 30 cubits" or "about 10 cubits" it says 10 and 30. Thus the Bible cannot possibly be inerrant. Even when CH wants to pull numbers out and post hoc justify the passage, it still is impossible.

I mean, any normal person would say, "the Bible didn't specifically say roughly, but clearly the intent was just to create a mental image when retelling the story." Yes, I wouldn't argue that. But if you take the position, like some fundamentalist evangelicals, that the Bible is 100% inerrant, then you run into problems.

If someone finds a good new symbol that might work, as it has with h-bar for example. But in the age of computers it would require quite a lot of momentum to make all the relevant software to support this symbol. Momentum that might be difficult to acquire without such support.

Actually not really, that's not a problem for Unicode which is quite robust, even less so when publishing with TeX and whatnot. It's not like pi itself is directly better and we've still got stuff like American keyboards having no easy way to type accented letters at all, way more of a theoretical problem but no one cares. Getting people in general to accept it again would still be more of a problem.

TBH, it would be easier to just make "2π" a symbol in its own right. I mean, I know when I see "2π", I just treat it as a single character in my head. Join up the "2" and the "π" and call it a new symbol pronouced "two-pi". Not every constant needs a Greek or Latin letter -- we can make our own!!

http://img9.imageshack.us/img9/8940/twopi.png

I think that this makes a better 2pi symbol than tao:
http://img856.imageshack.us/img856/7680/2pi.png

Of course, but the Bible doesn't say, "roughly 30 cubits" or "about 10 cubits" it says 10 and 30. Thus the Bible cannot possibly be inerrant. Even when CH wants to pull numbers out and post hoc justify the passage, it still is impossible.

I mean, any normal person would say, "the Bible didn't specifically say roughly, but clearly the intent was just to create a mental image when retelling the story." Yes, I wouldn't argue that. But if you take the position, like some fundamentalist evangelicals, that the Bible is 100% inerrant, then you run into problems.

That's like saying ATM machine, what the heck?

A cubit is a measure of forearm length, the length of a forearm varies from person to person

Poll options are bad. It shouldn't be changed because it makes much more sense, as it makes exactly the same amount of sense.

It should be changed because that would make things more elegant.

It's fine as is. Using tau needlessly complicates things.

e^(i pi) +1 = 0 is so much better than e^(i tau/2)+1=0.

I dunno you get the smallest prime number in that equation as well!

That's like saying ATM machine, what the heck?

A cubit is a measure of forearm length, the length of a forearm varies from person to person

Well, unless the measurement was taken by two different people with two different arm lengths, the ratio should always be 3.14159. If our measurement is my cat's claw or a pill, the ratio isn't going to change.

We're mostly arguing the same thing: that the Bible isn't 100% infallible in every word. It's completely reasonable to expect approximations in measurements like the passage CH quoted. The thing I take issue with is biblical literalism.

e^(i pi) +1 = 0 is so much better than e^(i tau/2)+1=0.But not as good as e^(i*tau)=1!

Yeah but with that example it's like you guys are still arguing over which Dan Brown novel is the best work of all time when other people would mention Shakespeare. It's all far inferior regardless to not use j.

Yeah, maybe it would make for somewhat more convenient and elegant notation but the tau symbol is kind of busy already, if you have to have a replacement constant for 2pi I would suggest you find a different letter to call it by.

But not as good as e^(i*tau)=1!
me gusta

It's missing my second favorite number. :(

Well you can write e^(i*tau)=1^0...

I'm sorry, but 48/2(9+3) = 2. Always.

288. :mad:

Well you can write e^(i*tau)=1^0...

And for those who like an equation containing one each of addition, multiplication, and exponentiation:

e ^ (i * tau) + 0 = 1