Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: Midpoint and unbounded intervals

	(Repost:  The earlier one got lost somehow.  - Dan)

> Date: Wed, 25 Jan 2012 12:31:19 -0800 (PST)
> From: Dmitry Nadezhin <dmitry.nadezhin@xxxxxxxxxx>
> To: <nh@xxxxxxxxxxxxxxxxx>
> Cc: <stds-1788@xxxxxxxxxxxxxxxxx>
> Subject: Re: Midpoint and unbounded intervals
> 
> Looks good.
> 
> However, it seems to me that there is an error in the paragraph
> immediately next after the definition (15). The midpoint of
> [Fmax,+oo[ is not infinity.
> 
> midpoint([Fmax,+oo[)=roundUp((Fmax + Fmax)/2)=roundUp(Fmax)=Fmax .
> 
> If this is correct than simpler definition (15) could be used
> instead of (16) Sorry, if I missed something.
> 
>   -Dima


	Dima,

	When you're right, you're right.

	This is my fault.  An earlier version of midpoint()
	used nextUp & suffered from this problem.  Then Nate
	pointed out that we could use roundUp instead & we
	changed it to that.  But we didn't realized that it
	also cured the problem with the last interval.

	We'll change it as you suggest.

	Thanks,

				Dan