🚀 go-pugleaf

In article <1992Nov10.190554.11097@sfu.ca> palmer@sfu.ca (Leigh Palmer) writes:
>In article <ETHANB.92Nov9155340@ptolemy.astro.washington.edu>
>ethanb@ptolemy.astro.washington.edu (Ethan Bradford) writes:
>>The simple explaination of why yellow monochromatic light looks the
>>same as equal combinations of green and red light is that the color
>>receptor systems in the eye (the cones) have sensitivity curves which
>>trail off (I remember them as being roughly gaussian).  A signal
>>halfway between the two peaks excites the two receptors equally, as do
>>two signals right on each peak.
>
>This article piqued my old annoyance with the free use of the term "gaussian"
>which is so common among us. In the case mentioned above it is the case that
>none of the three visual response functions looks gaussian to me. I have even
>heard Lorentzian curves described as gaussian by colleagues in an unthinking
>moment.
>
>I believe it is time that we had another term for curves which have
humps.

(Picturesque solutions deleted.)  I agree with your annoyance
completely.  Gaussians are very useful but pretending that every
molehill or mountain is a Gaussian can lead to lots of errors - so it's
good that mathematicians have a technical term for the kind of function
you are referring to.  It's called a "bump function", or "bump".