| Robert Berry wrote:
| > a 5 letter key would suffice for all the insects in the universe
There are 456,976 possible four-letter codes, which is far more
than the roughly 10,000 bird species, so there is no mathematical
necessity for a six-letter code. Theoretically birds could even
be encoded with three letters (17,576 combinations), but a lot of
them would have codes like XQJ that are not terribly meaningful.
In this day and age, when you can buy an 8-gigabyte disk for a
few hundred dollars, I don't really think that saving memory is
a defensible consideration.
The reason I went to six letters is to make it possible to encode
and decode without (much) reference to printed materials. In my
code system there are currently only 35 collisions, most of which
show up only when you go back to the early part of the century
when there was a vogue for using subspecies names whenever
possible. Omitting ancient subspecies names, there are fewer
than 20 collisions, so one can memorize them or put them on a
small reference card. This allows rapid encoding without
reference to a full listing of codes.
John Shipman, http://www.nmt.edu/~shipman
Zoological Data Processing, 507 Fitch NW, Socorro, NM 87801 USA