For American numbers

[Erik Max Francis]

Re: For American numbers

Dan Bishop wrote:
[color=blue]
> They must have gotten the idea from floppy disks, which also use a
> 1024000-byte "megabyte".[/color]

It's pretty common industry-wide. Memory is measured in binary prefixes
(x 1024), but disk space and bandwidth are measured in decimal prefixes
(x 1000).

--
Erik Max Francis && max@alcyone.com && http://www.alcyone.com/max/
San Jose, CA, USA && 37 20 N 121 53 W && AIM erikmaxfrancis
I'm not worried about anything. I'm not fearing any man.
-- Dr. Martin Luther King, Jr.

[Mike Meyer]

Re: For American numbers

Nick Coghlan <ncoghlan@iinet .net.au> writes:
[color=blue]
> Peter Hansen wrote:[color=green]
>> Only for hard drive manufacturers, perhaps.
>> For the rest of the computer world, unless I've missed
>> a changing of the guard or something, "kilo" is 1024
>> and "mega" is 1024*1024 and so forth...[/color]
>
> Given that there are perfectly good ISO prefixes for the multiples of
> 2**10, I don't see any reason to continue misusing the 10**3 prefixes
> for the purpose.[/color]

From what I found <URL: http://physics.nist.gov/cuu/Units/binary.html[color=blue]
>, it's not clear those are ISO prefixes yet - but they have been[/color]
adapted by some standards agencies.

Possibly you have better references?

<mike
--
Mike Meyer <mwm@mired.or g> http://www.mired.org/home/mwm/
Independent WWW/Perforce/FreeBSD/Unix consultant, email for more information.

[Nick Coghlan]

Re: For American numbers

Mike Meyer wrote:[color=blue][color=green]
>>From what I found <URL: http://physics.nist.gov/cuu/Units/binary.html[/color]
>[color=green]
>>, it's not clear those are ISO prefixes yet - but they have been[/color]
>
> adapted by some standards agencies.
>
> Possibly you have better references?[/color]

My mistake - IEC, not ISO :)

And I did get one wrong in my sample code - it's 'exbi' not 'ebi':

http://en.wikipedia.org/wiki/Binary_prefix#IEC_standard_prefixes

Cheers,
Nick.

--
Nick Coghlan | ncoghlan@email. com | Brisbane, Australia
---------------------------------------------------------------

http://boredomandlaziness.skystorm.net

[Peter Hansen]

Re: For American numbers

Alan Kennedy wrote:[color=blue]
> [Peter Hansen][color=green]
>> For the rest of the computer world, unless I've missed
>> a changing of the guard or something, "kilo" is 1024
>> and "mega" is 1024*1024 and so forth...[/color]
>
> Maybe you missed these?
>
> http://en.wikipedia.org/wiki/Kibibyte
> http://en.wikipedia.org/wiki/Mebibyte
> http://en.wikipedia.org/wiki/Gibibyte[/color]

Definitely missed them, in the sense of "didn't see them
yet". I must say I'm somewhat astounded that anyone
bothered to do this.

Don't miss them at all, in the sense of "have no plans
to use them and am not stressed in the least over the fact
that they are lacking in my life."
[color=blue]
> kilo-mega-giga-etc-should-be-powers-of-10-ly y'rs,[/color]

Maybe, but they aren't always... that's just the way
it is.

I take a descriptive view rather than a prescriptive one.
Google for the usage of the above and you'll find the
following ratios for the numbers of pages that use
the two forms:

kibibyte vs. kilobyte: 1:60 (impressively high, I admit)
mebibyte vs. megabyte: 1:234
gibibyte vs. gigabyte: 1:2082

I strongly suspect that the vast majority of the former
set of pages are of the form "use this instead of kilo!",
but perhaps I really have been asleep while much of the
computing world converted.

I'll be one of the last holdouts, too... it's really not
so hard to work in powers of two if you try...

-Peter

[Peter Hansen]

Re: For American numbers

Martin v. Löwis wrote:[color=blue]
> Peter Hansen wrote:
>[color=green]
>> For the rest of the computer world, unless I've missed
>> a changing of the guard or something, "kilo" is 1024
>> and "mega" is 1024*1024 and so forth...[/color]
>
>
> In case this isn't clear yet: you have missed a changing
> of the guard or something. "kibi" is 1024, "mebi" is
> 1024*1024 and so forth. "kilo" is 1000.[/color]

Yeah, saw the link. The IEC doesn't speak for me, I'm afraid.
And as the wikipedia notes, "As of 2005 this naming convention
has not gained widespread use."

I suspect I and many others will go to our graves not being
comfortable mebbling and gibbling over our bytes, though
we'll probably spend a lot of time kibbling over the issue...

-Peter

[Erik Max Francis]

Re: For American numbers

Peter Hansen wrote:
[color=blue]
> I'll be one of the last holdouts, too... it's really not
> so hard to work in powers of two if you try...[/color]

The difficulty isn't with working in powers of 1024, it's that the terms
are used inconsistently even within the computing industry. Memory is
measured in kibibytes, but disk space is measured in kilobytes.
Something as basic as "1 meg" has different numeric meanings depending
on whether you're talking about memory or disk space or metered
bandwidth usage. And a 1.44 MB isn't 1000^2 bytes or 1024^2 bytes, but
rather 1024*1000 bytes!

--
Erik Max Francis && max@alcyone.com && http://www.alcyone.com/max/
San Jose, CA, USA && 37 20 N 121 53 W && AIM erikmaxfrancis
The decree is the Sultan's; the mountains are ours.
-- Dadaloglu

[Michael Hoffman]

Re: For American numbers

Nick Coghlan wrote:
[color=blue]
> My mistake - IEC, not ISO :)[/color]

For all intents and purposes an IEC standard should be as good as an
ISO one. They usually develop standards for different areas, or jointly
if it is an overlapping area (but ISO/IEC standards are usually referred
to as "ISO standards").
--
Michael Hoffman

[JanC]

Re: For American numbers

Peter Hansen schreef:
[color=blue]
> Given the clear "units='byt es'" default above, and my restricting
> my comments to "the rest of the computer world", it should be
> clear I was talking about a very limited subset of the planet.
>
> A subset, however, which has an extremely strong attachment to
> 1024 instead of 1000 (for very good reasons), and which is
> less likely to abandon backwards compatibility and widely accept
> 1000 than the US is likely to adopt metric widely in the near
> future...[/color]

The problem is that that 'subset' uses kilo for both 1000 (at least hard
disks & some network transmission speeds) and 1024 in computer science.


--
JanC

"Be strict when sending and tolerant when receiving."
RFC 1958 - Architectural Principles of the Internet - section 3.9

[Dave Brueck]

Re: For American numbers

Peter Hansen wrote:[color=blue]
> Martin v. Löwis wrote:
>[color=green]
>> Peter Hansen wrote:
>>[color=darkred]
>>> For the rest of the computer world, unless I've missed
>>> a changing of the guard or something, "kilo" is 1024
>>> and "mega" is 1024*1024 and so forth...[/color]
>>
>> In case this isn't clear yet: you have missed a changing
>> of the guard or something. "kibi" is 1024, "mebi" is
>> 1024*1024 and so forth. "kilo" is 1000.[/color]
>
> Yeah, saw the link. The IEC doesn't speak for me, I'm afraid.
> And as the wikipedia notes, "As of 2005 this naming convention
> has not gained widespread use."
>
> I suspect I and many others will go to our graves not being
> comfortable mebbling and gibbling over our bytes, though
> we'll probably spend a lot of time kibbling over the issue...[/color]

Indeed - it's a little early to say that we've missed a changing of the guard! :)

Multiple definitions aside, "kilo" and "mega" are far too entrenched - even if I
could manage to say "kibibyte" with a straight face, I'd get nothing but blank
stares in return.

-Dave

[Peter Maas]

Re: For American numbers

Dave Brueck schrieb:[color=blue]
> Multiple definitions aside, "kilo" and "mega" are far too entrenched -
> even if I could manage to say "kibibyte" with a straight face, I'd get
> nothing but blank stares in return.[/color]

This kibi-mebi thing will probably fail because very few can manage
to say "kibibyte" with a straight face :)

--
-------------------------------------------------------------------
Peter Maas, M+R Infosysteme, D-52070 Aachen, Tel +49-241-93878-0
E-mail 'cGV0ZXIubWFhc0 BtcGx1c3IuZGU=\ n'.decode('base 64')
-------------------------------------------------------------------

[Michael Hoffman]

Re: For American numbers

Peter Maas wrote:[color=blue]
> This kibi-mebi thing will probably fail because very few can manage
> to say "kibibyte" with a straight face :)[/color]

I agree, I can't do it yet. I can write kiB and MiB though with a
straight face, and find that useful.
--
Michael Hoffman

[Nick Coghlan]

Re: For American numbers

Michael Hoffman wrote:[color=blue]
> Peter Maas wrote:
>[color=green]
>> This kibi-mebi thing will probably fail because very few can manage
>> to say "kibibyte" with a straight face :)[/color]
>
>
> I agree, I can't do it yet. I can write kiB and MiB though with a
> straight face, and find that useful.[/color]

And written communication is where avoiding the ambiguity tends to matter more -
in a conversation, if the difference actually matter, you can just ask. With a
written document, requesting clarification often isn't so simple.

Cheers,
Nick.

--
Nick Coghlan | ncoghlan@email. com | Brisbane, Australia
---------------------------------------------------------------

http://boredomandlaziness.skystorm.net

[Neil Benn]

Re: For American numbers

Scott David Daniels wrote:
[color=blue]
> Kind of fun exercise (no good for British English).
>
> <snip>[/color]

what's American about it? If anything, it's more French than American ;-)

N

--

Neil Benn
Senior Automation Engineer
Cenix BioScience
BioInnovations Zentrum
Tatzberg 46
D-01307
Dresden
Germany

Tel : +49 (0)351 4173 154
e-mail : benn@cenix-bioscience.com
Cenix Website : http://www.cenix-bioscience.com

[Dennis Lee Bieber]

Re: For American numbers

On Mon, 14 Feb 2005 09:47:30 +0000, Michael Hoffman
<cam.ac.uk@mh39 1.invalid> declaimed the following in comp.lang.pytho n:
[color=blue]
> Peter Maas wrote:[color=green]
> > This kibi-mebi thing will probably fail because very few can manage
> > to say "kibibyte" with a straight face :)[/color]
>
> I agree, I can't do it yet. I can write kiB and MiB though with a
> straight face, and find that useful.[/color]

"kibibyte" sounds too much like pet food... "Kibbles and Bits"
anyone...

OTOH, "MiB" has been poisoned by association with a certain Will
Smith movie (or two)... "Men in Black"...

--[color=blue]
> =============== =============== =============== =============== == <
> wlfraed@ix.netc om.com | Wulfraed Dennis Lee Bieber KD6MOG <
> wulfraed@dm.net | Bestiaria Support Staff <
> =============== =============== =============== =============== == <
> Home Page: <http://www.dm.net/~wulfraed/> <
> Overflow Page: <http://wlfraed.home.ne tcom.com/> <[/color]

[Scott David Daniels]

Re: For American numbers

Neil Benn wrote:[color=blue]
> Scott David Daniels wrote:[color=green]
>> Kind of fun exercise (no good for British English).[/color]
> what's American about it? If anything, it's more French than American ;-)[/color]

Well, actually this started with scaling integers, and I was worried
about a billion / billionth (and up / down). In mid-task I switched
to ISO prefix without bothering to update my mental state. So I
warned about a problem that a former version of this code had. As I
hit send, I realized my stupidity, but figured I was better off not
wasting more bandwidth (which I now have done). If I had it to do
over again, I'd have made the default unit something like meters --
again non-British, and it avoids the issue of 1024 / 1000 altogether.

grinning-stupidly,

Scott David Daniels
Scott.Daniels@A cm.Org