Robert Sanderson writes:
>>> While we're changing relation semantics, I'd like to propose that
>>> <> be deprecated in favour of 'ne' for not equal to, to be
>>> consistent with 'eq' for equality.
>>
>> I am not so fond of this; can you say more in favour of it?
> 
> It seems strange to me that equality is 'eq' (a string) but
> inequality is a combination of the special characters < and >.
> 
> More consistent to me would be == and !=, == and <>, or 'eq' and 'ne'.
> 
> But 'eq' and '<>' seems similar to having '>' and 'lessThan'...
> understandable but inconsistent.

I agree that if we were designing this now, we might choose "ne"
rather than "<>"; but I don't think it's sufficiently preferable to
merit a backwards-compatible change.  (Although the "="-becomes-
server-choice change is also technically backwards-incompatible, in
practice nearly all old-style queries will continue to be interpreted
as before.)

 _/|_	 ___________________________________________________________________
/o ) \/  Mike Taylor  <[log in to unmask]>  http://www.miketaylor.org.uk
)_v__/\  "I've made my bed; and now I've got to weasel out of it" --
	 Bart Simpson.