PDA

View Full Version : Bad math in article about a new record pi calculation



jfribrg
2010-Jan-10, 01:25 PM
I saw this article (http://news.yahoo.com/s/afp/sciencemathematicsfranceoffbeat) about a new record for calculating 1.7 trillion digits of pi. Some of the wording in the article seemed to be a bit unusual, perhaps due to the use of an automated translation software. However, near the bottom of the article, I saw the following:


Pi, the ratio of a circle's circumference to its diameter, kicks off with 3.14159... in a string whose digits are believed never to repeat or end.

Emphasis mine. The word "believed" is misleading. The correct word is "proven". It has been well over a century since the transcendence of pi was proven. The irrationality of pi was known before that. And before that is several thousand years of people trying (and failing) to find the rational representation of pi in an impossible quest known as "squaring the circle".

Usually you can blame the reporter for this type of error, but I'm not sure in this case. The programmer who did these calculations is not a mathmetician. He could have used one of the hundreds of formulae for pi without ever knowing any more about the history or properties of pi.

WaxRubiks
2010-Jan-10, 02:47 PM
well, you have to believe in the maths of that proof. :p

which is true, I think...

tdvance
2010-Jan-10, 04:20 PM
I saw this article (http://news.yahoo.com/s/afp/sciencemathematicsfranceoffbeat) about a new record for calculating 1.7 trillion digits of pi. Some of the wording in the article seemed to be a bit unusual, perhaps due to the use of an automated translation software. However, near the bottom of the article, I saw the following:



Emphasis mine. The word "believed" is misleading. The correct word is "proven". It has been well over a century since the transcendence of pi was proven. The irrationality of pi was known before that. And before that is several thousand years of people trying (and failing) to find the rational representation of pi in an impossible quest known as "squaring the circle".

Usually you can blame the reporter for this type of error, but I'm not sure in this case. The programmer who did these calculations is not a mathmetician. He could have used one of the hundreds of formulae for pi without ever knowing any more about the history or properties of pi.

If it's the person I'm thinking of, I believe he IS a mathematician. He's actually an officer of the Mathematical Association of America (which unlike the American Mathematical Society, focuses on teaching math at the university level, grad and undergrad, more than math research, but still is full of mathematicians).


Todd

tdvance
2010-Jan-10, 04:22 PM
Ok--it's NOT the person I'm thinking of....another person just announced a record pi computation, on a desktop!!!! using a new algorithm.

Jens
2010-Jan-12, 05:23 AM
I think it's probably just the academic tendency to use "weasel words." "It can be stated that..." instead of "I state that," "it is commonly accepted..." It probably wasn't so well thought out.