MOO-cows Mailing List Archive
[Prev][Next][Index][Thread]

Re: implied ANY

Followup to:  <3201107E@msgate.litc.lockheed.com>
By author:    "Adams, Charles" <cadams@msgate.litc.lockheed.com>
In newsgroup: local.moo
> 
> I agree with Pavel's assessment of the wrongness of catching ANY, but I also 
> have to agree with Quinn.  Let's have the rope to hang ourselves with.  :)
> 

On that same note, I would like that asinh(), acosh() and atanh() are
reinstated.  Pavel rejected these three functions based on the
argument that since these are not standard C functions, they will be
lower quality on certain machines (I had implemented them by having
autoconf look for them in the math library and use a substitute
function if not present.)

This is true, but they *do* exist on all BSD-, SVID- or Linux-based
UNIX systems (that should account for 98% or more), and as I
discovered when looking for a better algorithm, there *is* no "decent
substitute" (better than the obvious formula) that can be constructed
using only the standard library without knowing the format of fp
numbers.  Therefore, people needing these functions will have to do
in-db what I did in the server, but this will affect *all* users, not
only those on brain-damaged systems.

It doesn't seem worthwhile to me to penalize all users for the
existence of a few idiotic/obsolete systems.

This is not meant as a flame: Pavel and I had different views on some
of the fp stuff, and I think it was for the better in most cases.  The
above is the only one where I still have to disagree with Pavel on
implementation.

	-hpa

-- 
PGP public key available - finger hpa@zytor.com
Key fingerprint: 2047/2A960705 BA 03 D3 2C 14 A8 A8 BD  1E DF FE 69 EE 35 BD 74
"The earth is but one country, and mankind its citizens."  --  Bahá'u'lláh
Just Say No to Morden * The Shadows were defeated -- Babylon 5 is renewed!!
References:
Home | Subject Index | Thread Index