Talk:Witch of Agnesi

Special nature
What's so special about this curve?--Anakata 18:28, 25 June 2006 (UTC)


 * The Witch of Agnesi is functionally equivalent to the Cauchy-Lorentz Distribution, which has many uses in mathematics and physics. The one with which I am most familiar is the Lorentzian profile, or "natural broadening", of atomic emission and absorption spectral lines. I have added a link to the Cauchy Distribution page under "See Also".  AmberRobot 21:04, 21 February 2007 (UTC)
 * I would say that the special nature is due to the geometric definition which was added a few days after Anakata's question. Note that the curve is not unilaterally scaleable: stretching it in (only) the horizontal or vertical direction will change the curve into a non-witch (unlike, say, an ellipse, which can be scaled arbitrarily).  So only specific Cauchy-Lorentz distributions are actually witches. Cheers, Doctormatt 00:31, 22 February 2007 (UTC)

Fermat
This page states that Fermat studied this curve in 1666, but Fermat's page states that he died in 1665. This combination seems improbable, as zombies generally prefer to eat brains rather than do mathematical research. Is there a source for this claim? Sean Patrick Santos (talk) 03:18, 27 August 2009 (UTC)
 * The date should be 1630, Fermat was in his late 20's. The article has been fixed.--RDBury (talk) 05:39, 7 September 2009 (UTC)
 * I took out the date because I couldn't find a proper reference to it and the references I did find simply say "before 1666" (duh). —David Eppstein (talk) 05:30, 18 May 2018 (UTC)

Error regarding centroid
As this curve is infinite in width and asymptotic with the x-axis (according to the illustration), the centroid of the curve should be (0,0). The centroid of the area between the curve and its asymptote is (0,a/2). Additionally, the centroid of the generating circle is the center of the circle at (0,a).

Versicra
A yet another example of internet proparating the nonsence, a modern mondegreen so to say. Even Britannica Online (as of today) writes that 'versicra' means 'witch' in Italian, which is nonsense, as easily verifiable from modern and old dictionaries. There was/is no such word.

In fact I suspect this is a modern artifact of OCR publishing. If you search google books for 'versicra' you will find some, but if you look into original images of the corresponding books, you will see it is 'versiera', and 'versicra' is just sloppy job of subsequent readers. Max Longint (talk) 17:54, 7 October 2011 (UTC)

interesting links:

Max Longint (talk) 18:29, 7 October 2011 (UTC)
 * 
 * 
 * 
 * Where and in which version of the article do you see "Versicra"? I've searched back to May of last year and cannot find it.--RDBury (talk) 06:00, 8 October 2011 (UTC)
 * I was writing about "intertet propagating nonsense". I am glad that wikipedia was not fooled with the magic word "Britannica".
 * Sorry I didn't make it clear that this post was a warning not to add "versicra" to the article, despite references floating around. Max Longint (talk) 19:01, 9 October 2011 (UTC)
 * Thanks for the clarification. We have a non-OCR version of the 1911 Britannica article on the subject at WikiSource. It doesn't mention "versiera" but it does state that Agnesi invented the curve when she only wrote about it. There are a number of such factoids that have been propagated as historical truth; the internet is not a requirement though it does speed up the process a bit.--RDBury (talk) 04:37, 10 October 2011 (UTC)

Centroid
I am questioning this article's value of the centroid's x-coordinate at Reference desk/Mathematics. Comments are invited.--Jasper Deng (talk) 04:48, 8 September 2014 (UTC)
 * I took this out for now. If there is a centroid its x-coordinate must be 0 (by symmetry) and its y-coordinate is necessarily 0 (because of the heavy tails) but there are competing claims in the literature whether that really is the centroid or whether the centroid is undefined (because the integral that would define it diverges). The trouble is that "the integral that would define it" can mean any of several different things. Really, it's more a statement about the Cauchy distribution and its (lack of an) expected value, than about this curve itself, so I think it's ok to omit it. But we can put it back in if we find a source that goes into this issue in more depth than just the obvious considerations above, and handles it as a property of this curve rather than of the associated probability distribution. —David Eppstein (talk) 01:37, 18 May 2018 (UTC)

Capital "W"?

 * In mathematics, the Witch of Agnesi is ...
 * In mathematics, the witch of Agnesi is ...

This article currently consistently uses a capital "W". Should it? Michael Hardy (talk) 18:33, 1 June 2018 (UTC)