Re: [gentoo-project] Call for agenda items - Council meeting 2014-04-08

[Ian Stakenvicius <axs@gentoo.org>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On 31/03/14 01:27 PM, Rich Freeman wrote:
> On Mon, Mar 31, 2014 at 11:44 AM, Michał Górny <mgorny@gentoo.org>
> wrote:
>> Dnia 2014-03-31, o godz. 06:56:19 Joshua Kinard
>> <kumba@gentoo.org> napisał(a):
>>>> 
>>> In some respect, if all one cares about is free space on a disk
>>> drive or how fast they can stream a movie, then the KiB/MiB
>>> thing works.  But if you play with bits and bytes from
>>> time-to-time (and worry about byte alignment) or sometimes
>>> fiddle w/ partition tables in a hex editor...you're going to
>>> think in terms of powers of two.
> 
> KiB/MiB ARE powers of two.  It is KB/MB which are powers of ten 
> (depending on who you talk to).
> 
> Drive sizes tend to be reported in MB/GB, and memory tends to be 
> reported in MiB, GiB (though they may or may not use those 
> abbreviations when doing so).
> 

This is very much old "standard" vs new standard in terms of naming.
For those of us that have been around long enough, Mega/Kilo/etc have
always meant 1024 when addressing computational storage, as per for
instance ANSI/IEEE Std 1084-1986.  However, as people know this did
become (or has always been) used ambiguously and so these terms were
apparently deprecated in favour of MiB, KiB etc by the IEC starting at
around 1996 and with formal adoption 1999 with IEC 60027-2 Amendment 2
(and expanded adoption in ISO/IEC IEC 80000-13:2008)

[*] source: http://en.wikipedia.org/wiki/Binary_prefix

+1 for usage of {K,M,G,T,...}iB as per standard.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)

iF4EAREIAAYFAlM5rGkACgkQ2ugaI38ACPAM2QD/eNQIRbUTfrGAolqbCPR6fR83
Is+jgG1MClaBeBki3aoBAJVluP0Gq4fWNVdm7Dl0OoT4tL7JIMjadRCO34mn/mmn
=KM13
-----END PGP SIGNATURE-----