[seqfan] Re: A033473

[M. F. Hasler]

On Mon, Feb 17, 2014 at 3:37 AM, <israel at math.ubc.ca> wrote:

> I have added the denominator sequence (A238015), but I see Maximilian has
> already proposed changes to A033473 using rounded values.
>

I make this proposal right after your first message, before the discussion
came up. I don't object of my proposal being reverted to the "numerator"
version if this is favoured by others.

I suggest the discussion take place at https://oeis.org/draft/A033473


 One problem with "Nearest integer to ..." is that many of the denominators
> that are not 1 are 2, and "nearest" leaves ambiguous whether you round up
> or down.


I agree, but  actually this is quite frequent when round(.) is used on OEIS.
I believe that rounding up 0.5 to the next larger integer is the default
(almost "universally" agreed upon), unless "rounded down" (= truncated) is
specified explicitely. (At least for positive numbers...)

Maximilian