(copied from a reddit post I made a minute ago)
So, currently, the best argument I see for each side is:
π: It would be confusing to change.
One might argue that this is not a good argument, but we don't usually change notation, even though it would make sense. π is already widely known and we use it a lot.
τ: τ is more natural.
This argument is also good. I personally believe in discarding legacy, but I can see the other side too. I think this page is good at explaining why τ is more natural.
In my opinion, the question is whether to discard badly chosen notation or not, and I think I like another idea better: instead of discarding π, we just write 2π, even in cases of the area of a circle (½ 2π r2 ). This way, people will understand it without knowing τ (you can add a footnote if you want to make it as clear as possible) while still using the (at least in my opinion) better constant.
Meant r2, not r2.
Oh, and the page I was talking about, which obviously does not get copied, was https://sites.google.com/site/taubeforeitwascool/
One reason to pick π: e^i*π + 1 = 0. The version with τ is not as elegant or good.
e^iτ = 1 seems to be more elegant to me.
Umm... where's the zero? e^i*π + 1 = 0 is the best because it puts together the fundamental "figures" of mathematics: e, i, π, 1, 0, +, *, ^, =. DUH
So e^iτ-1=0. Or if you prefer + to - for some reason, e^iτ=0+1. Or you could make it look nice and symmetric with the ^s and write e^iτ=1^0.
If you want to act very naturally,
e^(i tau) = cos tau + i sin tau = 1 + 0i
There's all of them : e, i, tau, 1 and 0.
I don't like tau...
You do not like it.
So you say.
Try it! Try it!
And you may.
Try it and you may, I say.
I do not like Green Eggs in Tau.
Why would does everyone act as if adopting tau would necessarily mean discarding pi? Is it not true that, even with the use of tau, there are places where the use of pi would still make perfectly good sense?
Tau is wrong because it can not be represented as accurately as a date as pi can (since 44/7 is not a date, I doubt any date-like approximation of tau could be as accurate as 22/7)
Also, it's not nearly as delicious when multiplied by e.
That's only a problem until we introduce our new calendar system.
Watch for the roll-out of my new website, "Time Before It Was Cool".
The best "date approximation" of tau is 25/4, with an error of ~0.03. 30 times worse than 22/7 is for pi. :)
Who uses 22/7? :p
355/113 doesn't look much like a date.
Ah right. Sorry about that. :D
Pie is a delicious food!
So just when
was tau discovered?
Am I the only one that noticed how weird it is that pi is 1/2 of tau, but tau looks like 1/2 of pi?
The denominator is half!
And it's funny because he has a birthday and no one's there because no one knows about it, or if they do, they don't care. Tau is sad.
Pi spent 42 years in total obscurity after William Jones first used it to represent the number 3.14... Then Euler adopted it, everyone else followed, and pi became famous. This is only tau's second birthday, if you count from when Michael Hartl released The Tau Manifesto two years ago. Even if you count from the publication of Bob Palais' excellent (but with a symbol that most people couldn't use) paper, it's only been 11 years. At most, you could count 20 years, back to my crappy little tau essay and the handful of people who read it. Even that's less than half as long as pi took to START its rise to fame. So don't write this two-year-old off so quickly. Speaking of which... Hey Mike, you put too many candles on tau's birthday cake!
I didn't realize that I was writing tau off. But now that you mention it ... It'll never catch on at the high school level, because most students will never need it at the high school level and no teacher will ever use it, either. So it'll be one more thing to learn in college ... if they get that far in math, which most students will not. So, no, I really don't think it'll catch on outside a circle of academics. If it ever showed up on a game show, it would leave the contestants scratching their heads.
my argument for Pi:
What is easier to write (and looks better): 2π or τ/2 ?
As long as there are formulas where you only have π (and not 2π) i will stick to using π.
By that same logic, you should favor switching to η = π/2
What is easier to write (and looks better): 2η or π/2 ?
As long as there are formulas where you only have η (and not 2η) i will stick to using η.
And so it came to pass that pi converged to 0...
η looks like n, which would be very confusing to almost everyone. π does not look like anything else.
Sooooo... I think everyone should search youtube for "what tau sounds like"
Hehe! "Tau is wrong because it can not be represented as accurately as a date as pi can (since 44/7 is not a date, I doubt any date-like approximation of tau could be as accurate as 22/7)" - You're right!
What, no post for happy 22/7 day?
I don't like tau...it makes things more complicated rather than simple...
Well, what about radians.
Yet if that still doesn't convince you, I understand.
Why can't we use both anyways?