Gibibyte

"Ken Blake" wrote in message
...
Everything you say above is correct. But language is not fixed; the
meaning of words changes with time, and the prefixes kilo-, mega-, giga-,
tera-, etc. are an example of this. Today, they represent binary
magnitudes. Should that change have happened? As far as I'm concerned, no.
But it did.

For all practical purposes the prefixes like mebi-, with "b" inserted, are
almost never used, and should be dropped, even though they are
international standards. The only real exception to everyone's using the
decimal terms in a binary way is disk manufacturers. They use the terms in
a decimal way because it makes what offer for sale sound larger. As far as
I'm concerned, it's deceptive advertising. It confuses many people and
should be prohibited. They should be required to state the sizes of their
drives using powers of 2, as everyone else does.

Yes, I'm as heretical as you. I reckon that the meaning of "mega", "giga"
etc should vary with the context in which is is used: for example kilo in
kilometre means 1000, but kilo in kilobyte means 1024.

kibi, mebi sound utterly stupid (like someone whose had his mouth
anaesthetised at the dentist, as Rene Lamontagne put it so eloquently!). I
find it very difficult to use them without an involuntary smirk ;-)

I agree that disk manufacturers should be required to specify their disk
sizes in power-of-two definitions of mega, giga etc - or else to say
*prominently* both sizes (eg "2 TB equivalent to 1.86 TiB") which would make
people stop and think "what does TiB mean - ah it must be the conversion
between 1000- and 1024-based numbering".


But to all intents and purposes, 1 TB is the "same" as 1 TiB (as a good
approximation), in the same way that a bit rate of 10 Mbit/sec is a
data-transfer rate of *approximately* 1 MB/sec [byte], allowing for
overheads etc.

Ken Blake wrote:
If I talk about a kilobyte, I mean 1,024 bytes, and so on.

That is ridiculous. An 'amount' can't be ambiguous. There are 1000
amount kilobytes and 1000000 amount megabytes. So 1024 amount has to be
something else which is kibibyte and 1048576 amount has to be something
else which is mebibyte, which is also 1024 kibibytes.

Saying 'my kilobytes' are different from standard kilobytes is nuts/
misguided.

kilo has *always* been decimal. The introduction of its misuse as
something else is to be condemned. It should have never happened and it
is/ has/ been done away with.


--
Mike Easter

Carlos E.R. wrote:
On 15/11/2019 21.39, Rene Lamontagne wrote:
On 2019-11-15 1:44 p.m., VanguardLH wrote:
NY wrote:

The problem is not with the concept - which is very laudable - but the
names. "Mebibyte", "gibibyte" and "tebibyte" sound like someone with
a speech impediment (*) trying to say "megabyte", "gigabyte" and
"terabyte".
I suspect "bi" was used to denote "binary". It gave a hint to the base.
^^ ^^

Instead of using wholly new magnitude prefixes that everyone would have
to learn and for which adoption would be slow, especially in the
computer industry where change is the norm, they wanted magnitude
prefixes that hinted at the old meanings, so people would instinctively
perceive the magnitude based on the old prefixes, but also hint that the
magnitude was binary instead of decimal.

What would you have come up for magnitude prefixes that represented
magnitudes similar to the decimal-based ones but hint it was a binary
prefix?

I have no idea, just not my forte.

Ah, but the industry does have an idea, and the chose Gibibyte.

https://en.wikipedia.org/wiki/Gibibyte

"The use of gigabyte (GB) to refer to 1000000000 bytes in some contexts
and to 1073741824 bytes in others, sometimes in reference to the same
device, has led to claims of confusion, controversies, and lawsuits.

The IEC created the binary prefixes (kibi, mebi, gibi, etc.) in an
attempt to reduce such confusion."

Example of lawsuit.

https://www.geek.com/news/western-di...r-size-562705/

"Basically, manufacturers use a decimal system whereas operating systems
classify by binary system. The differences between the two methods
of measurement mean that the capacity stated on hard drives is always
going to be different than that shown in the operating system."

Drives are listed in GB, so that's the 1,000,000,000 kind.

And golly, memories are listed in GB too, when they should be
listed in "something-else units" :-) So I guess the IEC rules
only apply to companies that have been sued. OK, I get it now.
It's a pretty clear system. Of weights and measures. Take
your finger off the scale, please. So when I'm buying a
steak tonight, it should be in kibigrams. Got it.

Paul

"Carlos E.R." wrote:

VanguardLH wrote:

Ken Blake wrote:

The only real exception to everyone's using the
decimal terms in a binary way is disk manufacturers.

And memory (all types) since addressing is binary.

Not on a rotating disk...

What does that have to do with kibi, gibi, and other bi prefixes?

Why are you narrowly focused on HDDs? Those aren't the only devices
where capacity is measured.

On Fri, 15 Nov 2019 19:52:40 -0800, Mike Easter wrote:

Ken Blake wrote:
If I talk about a kilobyte, I mean 1,024 bytes, and so on.

That is ridiculous. An 'amount' can't be ambiguous. There are 1000
amount kilobytes and 1000000 amount megabytes. So 1024 amount has to be
something else which is kibibyte and 1048576 amount has to be something
else which is mebibyte, which is also 1024 kibibytes.

Saying 'my kilobytes' are different from standard kilobytes is nuts/
misguided.

kilo has *always* been decimal. The introduction of its misuse as
something else is to be condemned. It should have never happened and it
is/ has/ been done away with.

I've been in telecom / IT / enterprise networking for 35-40 years and have
never run into anyone on the job who uses mebi/kibi etc., so neither do I.
It may be inaccurate, but at least we're all speaking the same language and
that's what communication is all about. 'We know what we mean.'

"Paul" wrote in message
...

Drives are listed in GB, so that's the 1,000,000,000 kind.

This is GB meaning GB - the power-of-10 unit. They are not lying, but they
are trading on the fact that people will think (wrongly) "this is computing,
so I *assume* it means GiB".

And golly, memories are listed in GB too, when they should be
listed in "something-else units" :-)

Presumably with RAM, it's GB meaning GiB - it makes sense for memory to be a
round number in powers-of-2 so that all of it is addressable with a given
number of address bits - so sizes of 1, 2, 4, 8, 16 etc MiB/GiB.

On 2019-11-15 4:52 p.m., NY wrote:
"Ken Blake" wrote in message
...
Everything you say above is correct. But language is not fixed; the
meaning of words changes with time, and the prefixes kilo-, mega-,
giga-, tera-, etc. are an example of this. Today, they represent
binary magnitudes. Should that change have happened? As far as I'm
concerned, no. But it did.

For all practical purposes the prefixes like mebi-, with "b" inserted,
are almost never used, and should be dropped, even though they are
international standards. The only real exception to everyone's using
the decimal terms in a binary way is disk manufacturers. They use the
terms in a decimal way because it makes what offer for sale sound
larger. As far as I'm concerned, it's deceptive advertising. It
confuses many people and should be prohibited. They should be required
to state the sizes of their drives using powers of 2, as everyone else
does.

Yes, I'm as heretical as you. I reckon that the meaning of "mega",
"giga" etc should vary with the context in which is is used: for example
kilo in kilometre means 1000, but kilo in kilobyte means 1024.

kibi, mebi sound utterly stupid (like someone whose had his mouth
anaesthetised at the dentist, as Rene Lamontagne put it so eloquently!).
I find it very difficult to use them without an involuntary smirk ;-)

I agree that disk manufacturers should be required to specify their disk
sizes in power-of-two definitions of mega, giga etc - or else to say
*prominently* both sizes (eg "2 TB equivalent to 1.86 TiB") which would
make people stop and think "what does TiB mean - ah it must be the
conversion between 1000- and 1024-based numbering".


But to all intents and purposes, 1 TB is the "same" as 1 TiB (as a good
approximation), in the same way that a bit rate of 10 Mbit/sec is a
data-transfer rate of *approximately* 1 MB/sec [byte], allowing for
overheads etc.

So what is the i series word for yottabyte?

Rene

Char Jackson wrote:
Mike Easter wrote:
Ken Blake wrote:
If I talk about a kilobyte, I mean 1,024 bytes, and so on.

That is ridiculous. An 'amount' can't be ambiguous. There are 1000
amount kilobytes and 1000000 amount megabytes. So 1024 amount has to be
something else which is kibibyte and 1048576 amount has to be something
else which is mebibyte, which is also 1024 kibibytes.

Saying 'my kilobytes' are different from standard kilobytes is nuts/
misguided.

kilo has *always* been decimal. The introduction of its misuse as
something else is to be condemned. It should have never happened and it
is/ has/ been done away with.

I've been in telecom / IT / enterprise networking for 35-40 years and have
never run into anyone on the job who uses mebi/kibi etc., so neither do I.
It may be inaccurate, but at least we're all speaking the same language and
that's what communication is all about. 'We know what we mean.'

I will grant that there has been 'popularization' of the 'misuse', but
there are numerous instances of the use of 'bad language' which needed
improvement and reform.

There are a few sentences on the wp article on this subject which are
worth mentioning; many of those footnotes cited are from the Wayback
Machine:

// The kilobyte has traditionally been used to refer to 1024 bytes (210
B), a usage still common.[5][6][7] The usage of the metric prefix kilo
for binary multiples arose as a convenience, because 1024 is
approximately 1000.[8] - The binary interpretation of metric prefixes is
still prominently used by the Microsoft Windows operating system,[9] but
is deprecated or obsolete in other operating systems.

In December 1998, the IEC addressed such multiple usages and definitions
by creating prefixes such as kibi, mebi, gibi, etc., to unambiguously
denote powers of 1024.[10] Thus the kibibyte, symbol KiB, represents 210
= 1024 bytes. These prefixes are now part of the International System of
Quantities. The IEC further specified that the kilobyte should only be
used to refer to 1000 bytes.
// https://en.wikipedia.org/wiki/Kilobyte


1998 was over 20 years ago. That is long enough for the workplace to
show reformation from obsolete language.

--
Mike Easter

Rene Lamontagne wrote:
So what is the i series word for yottabyte?

yobi, of course.

--
Mike Easter

On 2019-11-16 10:16 a.m., Mike Easter wrote:
Rene Lamontagne wrote:
So what is the i series word for yottabyte?

yobi, of course.


Thanks, Rene

On Sat, 16 Nov 2019 07:57:28 -0800, Mike Easter wrote:

Char Jackson wrote:
Mike Easter wrote:
Ken Blake wrote:
If I talk about a kilobyte, I mean 1,024 bytes, and so on.

That is ridiculous. An 'amount' can't be ambiguous. There are 1000
amount kilobytes and 1000000 amount megabytes. So 1024 amount has to be
something else which is kibibyte and 1048576 amount has to be something
else which is mebibyte, which is also 1024 kibibytes.

Saying 'my kilobytes' are different from standard kilobytes is nuts/
misguided.

kilo has *always* been decimal. The introduction of its misuse as
something else is to be condemned. It should have never happened and it
is/ has/ been done away with.

I've been in telecom / IT / enterprise networking for 35-40 years and have
never run into anyone on the job who uses mebi/kibi etc., so neither do I.
It may be inaccurate, but at least we're all speaking the same language and
that's what communication is all about. 'We know what we mean.'

I will grant that there has been 'popularization' of the 'misuse', but
there are numerous instances of the use of 'bad language' which needed
improvement and reform.

snip

1998 was over 20 years ago. That is long enough for the workplace to
show reformation from obsolete language.

I think that's your answer. If the 'misuse' hasn't abated by now, it
probably won't ever. That's fine with me.

On 11/15/19 10:25 PM, Paul wrote:

[snip]

So when I'm buying a
steak tonight, it should be in kibigrams. Got it.

Â*Â* Paul

1 kibigram = 2^10 grams = 1024 grams = 1.024 kilograms

--
39 days until the winter celebration (Wed, Dec 25, 2019 12:00:00 AM for
1 day).

"Religion is the original war crime." -- Michelle Malkin (Feb 26, 2005)

Rene Lamontagne wrote:

On 2019-11-15 4:52 p.m., NY wrote:
"Ken Blake" wrote in message
...
Everything you say above is correct. But language is not fixed; the
meaning of words changes with time, and the prefixes kilo-, mega-,
giga-, tera-, etc. are an example of this. Today, they represent
binary magnitudes. Should that change have happened? As far as I'm
concerned, no. But it did.

For all practical purposes the prefixes like mebi-, with "b" inserted,
are almost never used, and should be dropped, even though they are
international standards. The only real exception to everyone's using
the decimal terms in a binary way is disk manufacturers. They use the
terms in a decimal way because it makes what offer for sale sound
larger. As far as I'm concerned, it's deceptive advertising. It
confuses many people and should be prohibited. They should be required
to state the sizes of their drives using powers of 2, as everyone else
does.

Yes, I'm as heretical as you. I reckon that the meaning of "mega",
"giga" etc should vary with the context in which is is used: for example
kilo in kilometre means 1000, but kilo in kilobyte means 1024.

kibi, mebi sound utterly stupid (like someone whose had his mouth
anaesthetised at the dentist, as Rene Lamontagne put it so eloquently!).
I find it very difficult to use them without an involuntary smirk ;-)

I agree that disk manufacturers should be required to specify their disk
sizes in power-of-two definitions of mega, giga etc - or else to say
*prominently* both sizes (eg "2 TB equivalent to 1.86 TiB") which would
make people stop and think "what does TiB mean - ah it must be the
conversion between 1000- and 1024-based numbering".

But to all intents and purposes, 1 TB is the "same" as 1 TiB (as a good
approximation), in the same way that a bit rate of 10 Mbit/sec is a
data-transfer rate of *approximately* 1 MB/sec [byte], allowing for
overheads etc.

So what is the i series word for yottabyte?

See the table at:
https://en.wikipedia.org/wiki/Binary_prefix

Not all decimal magnitude prefixes have a binary counterpart. There are
no "bi" magnitude prefixes for when the exponent is negative, like for
deci, centi, milli, micro, nano, pico, fento, att, zepto, yocto.

https://physics.nist.gov/cuu/Units/binary.html
"It is suggested that in English, the first syllable of the name of the
binary-multiple prefix should be pronounced in the same way as the first
syllable of the name of the corresponding SI prefix, and that the second
syllable should be pronounced as "bee."

I've been pronouncing the "bi" part as "beh", not "bee". Using "bee"
makes it more distinguishible that you're using a binary magnitude
prefix.

https://www.iec.ch/si/binary.htm
"As time has passed, kilobytes have grown into megabytes, then gigabytes
and now terabytes. The problem is that, at the SI tera-scale (1012), the
discrepancy with the binary equivalent (240) is not the 2,4 % at
kilo-scale but rather approaching 10 %. At exa-scale (1018 and 260), it
is nearer 20 %."

The error or difference in misusing decimal prefixes for binary
magnitudes gets worse as capacities grow. The difference was tolerated
when magnitudes were small, but is becoming a problem as they grow.

https://alexey.chernyak.id.au/prefBin.xhtml

The problem with adoption is that the new prefixes have significance to
only a small portion of the population: those involved with computers
(and not just using them as consumers). Adoption of Celsius over
Fahrenheit has been slow (only in those countries that still use
Fahrenheit, like the USA). Global adoption of the binary prefixes will
be even slower because it will slowly leak out from a small community
where the differences matter.

On 16/11/2019 12.26, NY wrote:
"Paul" wrote in message
...

Drives are listed in GB, so that's the 1,000,000,000 kind.

This is GB meaning GB - the power-of-10 unit. They are not lying, but
they are trading on the fact that people will think (wrongly) "this is
computing, so I *assume* it means GiB".

They did this from the start. On a 10 MB disk the difference is small.

10*10^6 = 10000000
10*2²0 = 10485760

Had they changed the method at some time, they would be intentionally
misleading.


And golly, memories are listed in GB too, when they should be
listed in "something-else units" :-)

I have not seen those :-?


Presumably with RAM, it's GB meaning GiB - it makes sense for memory to
be a round number in powers-of-2 so that all of it is addressable with a
given number of address bits - so sizes of 1, 2, 4, 8, 16 etc MiB/GiB.

Right.

Consider than in RAM, the address bus works "binarily", in multiples of
two. But not in disks. They work, currently, in linear sectors, as many
as they can fit on the surface. And previously, in CHS counts. To them,
tying to multiples of two was not necessary/wanted/desired.

Even the sector size is not 512, we forget other metadata that is there,
like checksums. I don't know what the actual figure is.



My take on this was that calling 1024 bytes a kilobyte was an acceptable
approximation at the time, as long as everyone was aware that it was an
approximation. Aprox 1K=1000. As the units got bigger, the approximation
was no longer negligible, so better invent a different unit and avoid
the confussion. Yeah, some people hate change, but this is a needed change.

I also had to change after a lifetime of using the old units.

--
Cheers, Carlos.

On 2019-11-16 12:35 p.m., VanguardLH wrote:
Rene Lamontagne wrote:

On 2019-11-15 4:52 p.m., NY wrote:
"Ken Blake" wrote in message
...
Everything you say above is correct. But language is not fixed; the
meaning of words changes with time, and the prefixes kilo-, mega-,
giga-, tera-, etc. are an example of this. Today, they represent
binary magnitudes. Should that change have happened? As far as I'm
concerned, no. But it did.

For all practical purposes the prefixes like mebi-, with "b" inserted,
are almost never used, and should be dropped, even though they are
international standards. The only real exception to everyone's using
the decimal terms in a binary way is disk manufacturers. They use the
terms in a decimal way because it makes what offer for sale sound
larger. As far as I'm concerned, it's deceptive advertising. It
confuses many people and should be prohibited. They should be required
to state the sizes of their drives using powers of 2, as everyone else
does.

Yes, I'm as heretical as you. I reckon that the meaning of "mega",
"giga" etc should vary with the context in which is is used: for example
kilo in kilometre means 1000, but kilo in kilobyte means 1024.

kibi, mebi sound utterly stupid (like someone whose had his mouth
anaesthetised at the dentist, as Rene Lamontagne put it so eloquently!).
I find it very difficult to use them without an involuntary smirk ;-)

I agree that disk manufacturers should be required to specify their disk
sizes in power-of-two definitions of mega, giga etc - or else to say
*prominently* both sizes (eg "2 TB equivalent to 1.86 TiB") which would
make people stop and think "what does TiB mean - ah it must be the
conversion between 1000- and 1024-based numbering".

But to all intents and purposes, 1 TB is the "same" as 1 TiB (as a good
approximation), in the same way that a bit rate of 10 Mbit/sec is a
data-transfer rate of *approximately* 1 MB/sec [byte], allowing for
overheads etc.

So what is the i series word for yottabyte?

See the table at:
https://en.wikipedia.org/wiki/Binary_prefix

Not all decimal magnitude prefixes have a binary counterpart. There are
no "bi" magnitude prefixes for when the exponent is negative, like for
deci, centi, milli, micro, nano, pico, fento, att, zepto, yocto.

https://physics.nist.gov/cuu/Units/binary.html
"It is suggested that in English, the first syllable of the name of the
binary-multiple prefix should be pronounced in the same way as the first
syllable of the name of the corresponding SI prefix, and that the second
syllable should be pronounced as "bee."

I've been pronouncing the "bi" part as "beh", not "bee". Using "bee"
makes it more distinguishible that you're using a binary magnitude
prefix.

https://www.iec.ch/si/binary.htm
"As time has passed, kilobytes have grown into megabytes, then gigabytes
and now terabytes. The problem is that, at the SI tera-scale (1012), the
discrepancy with the binary equivalent (240) is not the 2,4 % at
kilo-scale but rather approaching 10 %. At exa-scale (1018 and 260), it
is nearer 20 %."

The error or difference in misusing decimal prefixes for binary
magnitudes gets worse as capacities grow. The difference was tolerated
when magnitudes were small, but is becoming a problem as they grow.

https://alexey.chernyak.id.au/prefBin.xhtml

The problem with adoption is that the new prefixes have significance to
only a small portion of the population: those involved with computers
(and not just using them as consumers). Adoption of Celsius over
Fahrenheit has been slow (only in those countries that still use
Fahrenheit, like the USA). Global adoption of the binary prefixes will
be even slower because it will slowly leak out from a small community
where the differences matter.


Thanks for all the info, VanguardLH
But as you say all these larger numbers are of interest to a limited
group of computer experts and people who deal in large system data
farms, etc.
I really as a computer user should not clutter my head with stuff that
is of no real concern or value to me.

Rene