Extended real arithmetic (was Re: inf - inf = 0 ??)




On Sat, 11 Sep 2010, Andreas Eder wrote:

< 
< Richard Hennessy wrote:
< 
< "Clearly limit(sin(x),x,inf)=>IND and IND-IND must be IND"
< 
< The first statement is clear, the second is not.  Can you prove
< it?
< 
< Well, that statement is clearly wrong, because for any expr with
< limit(expr(x),x,a) => IND we  obviously have  limit(expr(x),x,a) -
< limit(expr(x),x,a)  = limit(expr(x)-expr(x),x,a) = limit(0,x,a) =
< 0.
 
That is an embarassing mis-application of the additivity rule
for limits. I can only guess that you intended to tweak someone's
pique.

Leo

-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.