Revised draft available David Van Horn (31 Dec 2004 04:43 UTC)
Re: Revised draft available campbell@xxxxxx (02 Jan 2005 18:52 UTC)
Re: Revised draft available campbell@xxxxxx (02 Jan 2005 19:27 UTC)

Re: Revised draft available campbell@xxxxxx 02 Jan 2005 20:09 UTC

On Sun, 2 Jan 2005 xxxxxx@autodrip.bloodandcoffee.net wrote:

> Also, one minor, not very significant, comment on the names: perhaps it
> would be better to use, for example, INTEGER16 vs INTEGER16+, to
> indicate that the latter has no implicit sign while the former does.
> (A hyphen between INTEGER & the number could be added as well.)  As it
> is, it looks like it's 'some integer minus sixteen' vs 'some integer
> plus sixteen.'

Sorry, I ought to have been a bit more clear here.  My suggestion is
that, rather than have INTEGER-16 mean 'signed integer' (i.e. there is
a sign stored with the integer) and have INTEGER+16 mean 'unsigned
integer' (i.e. the sign could be considered to always be positive),
there should be instead INTEGER16 or INTEGER-16, for signed sixteen-bit
integers (i.e. integers without a sign automatically assigned to them),
and INTEGER16+ or INTEGER-16+, for unsigned sixteen-bit integers (i.e.
integers with an automatic positive sign).  As it is, INTEGER-16 vs
INTEGER+16 looks like 'an integer minus sixteen' vs 'an integer plus
sixteen.'

(And so on, for all of the signed vs unsigned integer representation
descriptors.)