stars sequences

[John Conway]

On Thu, 18 Sep 2003, Ralf Stephan wrote:

> There are two sequences related to these astronomical objects
> 
> %N A072171 Number of stars of visual magnitude n.
> %N A053406 Number of stars brighter than visual magnitude n-1.
> 
> One would expect the latter being the (shifted) partial sums
> of the former but this is not even remotely so:

   [values snipped]

> Can someone clarify?

    Yes.  They're just taken from different sources.  These numbers are 
not at all well-defined, and so should definitely NOT be in the EIS.  

   There are many reasons for the discrepancies, perhaps the most
obvious one being that there are plenty of stars whose magnitude varies
(for instance Mira's has been up and down between "1" and 9).  Others
are that the magnitudes reported by different observatories can
differ quite substantially.  

   The latter problem is particularly intense for the so-called "visual" 
magnitudes.  They are supposed to be what a typical human observer
would report, but in fact were computed as a weighted average from 
photographs taken in different colors.  Probably that averaging is
now done automatically from CCD's instead of photographs, introducing
another source of discrepancy.

   As I said, such meaningless numbers shouldn't be in the EIS.  Or
if they are, you should add some more, like "weights of US Senators,
in order", to go with them.

    John Conway