Using standardized SI prefixes

Christof Krüger ubuntu at
Tue Jun 12 07:37:44 UTC 2007

On Mon, 2007-06-11 at 19:56 -0500, Mark Reitblatt wrote:
> On 6/11/07, Alex Jones <alex at> wrote:
> > Fine. Stick with Kilobytes, but strictly define it as 10^3 bytes. Just
> > choose one over the other and be consistent.
> That's not "consistent". Kilobyte has always meant 2^10 bytes. "kilo"
> in "kilobyte" is not an SI prefix. SI prefixes only apply to SI
> measurements, of which "byte" is not a member. There is no confusion;
> the only place where a kilobyte != 2^10 bytes is in hard drive
> manufacturer's advertising materials. This is the way it has been for
> decades, and it is a perfectly acceptable and desirable standard.

It's all not as simple as you write it.

Bit rates have been usually measured in 10^(3x) bit per second, e.g.
kbps or kbit/s. So when talking about transfer rates, kilo meant
thousand. However, when talking about file/memory sizes, kilo meant
1024. But then again, a lot of people aren't aware of this difference
and there are a lot of programs which present e.g. download speeds in
2^(10x) bytes per second (_bits_ per second are more common to be used
in "lower levels", but there are also programs which use 2^10 _bits_ per
second as transfer speed units).

Another "historic" example is a floppy-MB:
A 1.44MB floppy disc can store 1,474,560 Bytes, that is 1440 KiB and
1.40625 MiB or approximately 1475KB or 1.48MB with kilo=10^3 and
However, these floppies were known as "1.44MB"-floppies. (MB meaning
1000 times 1024 bytes). Very consistent!

Your example of hard drive manufacturers is a another good example why
we actually SHOULD have unambiguous prefixes. Advertising always tends
to abuse ambiguities. When SI prefixes were used consistently, it would
have been clear from start that you cannot fit 100 GiB of data on a hard
drive advertised as having 100GB free space available.

Just because something has been done wrong for a long time doesn't make
it right. People who know the inconsistencies get used to them and do
not want to change it because it may be inconvenient for them or it
simply sounds stupid to them (what an argument!).
However, this means that _every_ new generation of students and
hobbyists has to go through learning the inconsistencies if we change
nothing. Hooray, confusion till the end of times!

But if we pushed the use of SI-prefixes, the computer-gurus would have
to get used to the new system but following generations would profit
from having a consistent unit system. In my opinion this is something
that is worth the effort. The problem with such big changes is that a
critical mass is needed to benefit from this new system and the faster
it is achieved, the shorter the confusion-period will be. 
I think that the open source community should participate since
consistent and unambiguous conventions are a good thing (TM).

Christof Krüger

More information about the Ubuntu-devel-discuss mailing list