Hi John,

> I think the designers of the SQL Spec picked a bad name in 'NULL'.
> Just looked it up in the dictionary:
>
> "Amounting to nothing : NIL
> 3 : having no value : INSIGNIFICANT
> 4 a : having no elements <null set> b : having zero as a limit <null
> sequence> c of a matrix : having all elements equal to zero
> 5 a : indicating usually by a zero reading on a scale when a given
> quantity (as current or voltage) is zero or when two quantities are
> equal -- used of an instrument b : being or relating to a method of
> measurement in which an unknown quantity (as of electric current) is
> compared with a known quantity of the same kind and found equal by a
> null detector
> 6 : of, being, or relating to zero
> 7 : ZERO 1c "
>
> This confuses loads of SQL novices I know - especially when they're
> programmers too, ('null pointers' etc, and in C# difference
> between 'null' and 'DBNull.Value'!). Personally I always prefer to
> say "if (x is DBNull)", more like sql syntax.
>
> SQL should have chosen UNKNOWN as the name of the state, in my
> opinion.
>
> However, I agree that Firebird is totally correct in it's handling of
> null strings!

And add to the confusion that in Dutch, the word for zero (0)
is pronounced as "null" (but written like "nul"). :-)

With regards,

Martijn Tonies
Database Workbench - developer tool for InterBase, Firebird, MySQL & MS SQL
Server.
Upscene Productions
http://www.upscene.com