Talk:Pi: Difference between revisions
Content added Content deleted
(→Time to delete: ah, minor edits.) |
|||
| Line 71: | Line 71: | ||
:::::::No, it's not an argument on the merits. The best thing I've read on the merits of the subject is [http://esr.ibiblio.org/?p=3481 here], and all that tells me is that we don't know which is really better than the other. The trouble with pi vs tau on RC is that your average ''non''-mathematician isn't yet likely to be familiar with tau, and so using tau in tasks is very likely to confuse what should be a simple subject; to resolve the confusion, use of tau would need annotations like "tau is 2*pi", and that would strike me as too trivial to warrant further complicating the task description. In short, even if we posit tau to be a more elegant symbol than pi, right ''now'', it's not a more elegant way to write task descriptions. --[[User:Short Circuit|Michael Mol]] 21:07, 1 August 2011 (UTC) |
:::::::No, it's not an argument on the merits. The best thing I've read on the merits of the subject is [http://esr.ibiblio.org/?p=3481 here], and all that tells me is that we don't know which is really better than the other. The trouble with pi vs tau on RC is that your average ''non''-mathematician isn't yet likely to be familiar with tau, and so using tau in tasks is very likely to confuse what should be a simple subject; to resolve the confusion, use of tau would need annotations like "tau is 2*pi", and that would strike me as too trivial to warrant further complicating the task description. In short, even if we posit tau to be a more elegant symbol than pi, right ''now'', it's not a more elegant way to write task descriptions. --[[User:Short Circuit|Michael Mol]] 21:07, 1 August 2011 (UTC) |
||
::::::: But there's no exact arguments on merits. Tau vs Pi is really a matter of radius vs diameter of a circle, and you can't argue which is of more merit than the other. A well defined constant should convey most symmetry or invariance of a system, where radius is arguably better because one end of r is always at the origin--but in real world diameters are almost always easier to measure: try directly tell the radius of a ball bearing with a caliper. In any event, for calculating digits of pi, the tau debate is not even relevant, where the most useful constant is probably Pi/4 any way. --[[User:Ledrug|Ledrug]] 21:55, 1 August 2011 (UTC) |
::::::: But there's no exact arguments on merits. Tau vs Pi is really a matter of radius vs diameter of a circle, and you can't argue which is of more merit than the other. A well defined constant should convey most symmetry or invariance of a system, where radius is arguably better because one end of r is always at the origin--but in real world diameters are almost always easier to measure: try directly tell the radius of a ball bearing with a caliper. In any event, for calculating digits of pi, the tau debate is not even relevant, where the most useful constant is probably Pi/4 any way. --[[User:Ledrug|Ledrug]] 21:55, 1 August 2011 (UTC) |
||
::::::: That http://esr.ibiblio.org/?p=3481 argument (and the pi manifesto it referenses) suggests that this whole thing is just about conciseness, and not about some of the other issues (like constant 1/2 in simple r^2 equations and 1/6 in simple r^3 equations). I can understand that conciseness has advantages, and I also agree that pi's familiarity/popularity can be a major advantage in a cookbook equation context. Anyways, the "pi is wrong" slogan, while catchy and motivating is itself wrong. "Pi is useful, but pi can also occasionally be misleading or confusing" would be a more accurate (though boring) phrasing. There's room in the world for both constants, and I dislike reasoning (even from its advocates) that suggest that there can be only one. And "tau is 2*pi" is simple, but too trivial? We have "Goodbye, world" here. And, 99 bottles of beer -- how can "tau is 2*pi" be too trivial? (Though, ok, I can understand not wanting to like to a "pi is wrong" page.) --[[User:Rdm|Rdm]] 11:48, 2 August 2011 (UTC) |
|||