Computer proverbs for Fred Shapiro

[James A. Landau]

"Big-Endian versus Little Endian"

This phrase (if not the exact words) goes back to the Lilliput section of Gulliver's Travels, where Swift satirizes the religious struggles of England in the 1600's by implying they were over the question of which end of an egg to crack ("How many Angles can dance on the end of an egg?").  Dr. Seuss used the same idea in a satire on "buttered side up" versus "buttered side down" (I don't remember the book title, but the same satire appears in Seussical).  I don't know whether Seuss got the idea on his own or was imitating Swift.

The computer community picked up "Big-endian" and "Little-Endian" as the nomenclature for a very real computer problem.  A computer can work with numbers stored small-end-first or stored large-end-first, and it is up to the "architect" (designer) of a computer which technique to use.  This means that when "porting" software from one computer to another, if the two computers differ on which end is in front, you may have a big headache.

Another term for this problem is "NUXI", because if "UNIX" is stored as two two-byte numbers and you transfer it from a Big-Endian computer to a Little-Endian or vice versa, the U and N get scrambled, and the I and X get scrambled separately.

---------------------------------------------------------

"The key, the whole key, and nothing but the key, so help me Codd"

This odd phrase is neither a satire nor a spoof; it is of all things a mnemonic.  Back in the 1970's Edgar F. Codd of IBM worked out the theory of relational databases.  Part of his theory was a set of three rules for the structure of the "key field", that is, the data item that uniquely identifies each data record.  The above quote provides a surprisingly near-exact description of Codd's three rules.

If you want to know what the three rules are, rather than ask me to explain you should check any good textbook on database management systems.

Nowadays database systems like Microsoft Access and its competitors use Codd's rules but bury them so deeply under a GUI interface that only serious computer scientists delve into the gooey details.  It was not always so.  As late as 1984 (when I was in systems procurement) there was considerable debate on whether Codd's ideas were worth implementing.

I learned the mnemonic from my boss in 1988, when I was taking a graduate class in database theory.

     - James A. Landau
       systems engineer