Re: nomenclature for representable entities

[r754@xxxxxxxxxxxxxx]

> Subject: nomenclature for representable entities
> Date: Mon, 28 Aug 2006 13:11:56 -0700
> From: "Ferguson, Warren E" <warren.e.ferguson@xxxxxxxxx>
> To: <stds-754@xxxxxxxxxxx>, <stds-754@xxxxxxxx>
>
>
> At the last style review we were asked to come up with a taxonomy of
> "representable entities". Here's my response:
>
> =20
>
> The "representable entities" include NaNs, so by double negation
> anything not NaN must be "a number".
>
> Numbers fall into two classes: "infinity" and "not infinity =3D finite".
>
> Finites fall into two classes: "zero" and "nonzero".
>
> I guess nonzero includes "normal" and "unnormal" - but I'm not sure if
> this is as useful as it used to be for just 754 binary numbers.
>
> =20
>
> I don't want to restrict unnormal numbers to 754's denormal entities
> because there a denormal had a reserved exponent.
>
> =20
>
> Proposal: zero + nonzero + infinite + NaN =3D representable entitites.
>
> =20
>
> Warren Ferguson

        It is a good list.  It is also close to what we use.  My only
        problem is that the most common name used is the most general
        one which, in this case, would be "representable entities".
        I guess my problem is that it is not what the user is expecting
        to see.  I might change this one name to something like
        "floating-point number" which IS something the user expects.
        But, as happened Thursday, some people have a problem with
        "floating-point number" including NaNs.  Just my opinion.  - Dan