Linked by Thom Holwerda on Wed 30th Jan 2013 00:38 UTC
Hardware, Embedded Systems Marco Arment: "Everyone should play by the same rules. A proposal: storage capacities referenced or implied in the names or advertisements for personal computers, tablets, and smartphones should not exceed the amount of space available for end-user installation of third-party applications and data, after enough software has been installed to enable all commonly advertised functionality. With today's OSes, iPads could advertise capacities no larger than 12, 28, 60, and 124 GB and the Surface Pros could be named 23 and 83 GB." Wholly agreed. When I buy a box of 100 staples, I expect it to contain ~100 staples - not 50 because the other 50 are holding the box together.
Thread beginning with comment 550838
To view parent comment, click here.
To read all comments associated with this story, please click here.
ssokolow
Member since:
2010-01-21

The "xiB" nomenclature was invented by HDD manufacturers just so they could lie about their capacities.


I don't know what input any hard drive manufacturers who had members in these organizations might have had might have had, but the binary prefixes (kibi/gibi/tebi/etc.) were developed by the IEC with support from the IEEE and CIPM (International Committee for Weights and Measures) due to concerns that the old "everybody knows kilo- means 1024 in the context of computers" attitude was proving problematic as computers became a more and more generally-used tool.

It's actually meant to SOLVE the problem you're complaining about... though more in the context of standards, research, and business. (People in related but different fields both saying 5GB but meaning different things)

When I studied, 1GB meant 1073741824 bytes.


...because people in computer science decided "Ehh... 1000? 1024? Close enough. Use the SI prefixes."

Personally, I think it'd make more sense to complain about wood measurements like 2x4 referring to the size they're cut before the wood is dried out.



Just because Google Calculator is still non-compliant with a definition standards bodies agreed on over a decade ago doesn't mean that should continue to be the case.

GiB is unambiguous. GB will become unambiguous when we stop abusing it to refer to powers of 1024.

Complain to the hard drive manufacturers (Good. Ditch GB) or to Microsoft (Not something I want. Keep GiB.) to get them to agree on a set of units.

Here's the NIST page on the topic:

http://physics.nist.gov/cuu/Units/binary.html

Edited 2013-01-30 05:42 UTC

Reply Parent Score: 5

HappyGod Member since:
2005-10-19

The problem is that the number of people who do not comply with the standard is larger than the people who do.

The situation is way more confusing now than it was before the standards bodies got involved. Because now you buy a 2TB HDD, and don't (apparently) get anywhere near that much free space when Windows tells you your drives capacity. I wonder how many calls HDD manufacturers get per day because of that issue alone.

due to concerns that the old "everybody knows kilo- means 1024 in the context of computers" attitude was proving problematic as computers became a more and more generally-used tool.


This is total bull$hit. There is almost no situation where knowing the number of bytes in a MB/GB is required in order to successfully use a computer.

People simply understood that they were copying 5MB, and the computer says they had 6MB free. No problems. In fact that's still how it works because Windows still uses the old definition of a GB.

Standards bodies, I think, overstepped their bounds. While they are responsible for setting standards, they should have sought the opinion of the major stakeholders before making the change they did. And if it transpired they could not at least get a majority YES vote, then it should not have happened.

Edited 2013-01-30 08:25 UTC

Reply Parent Score: 3

Brendan Member since:
2005-11-16

Hi,

The problem is that the number of people who do not comply with the standard is larger than the people who do.


Yes, these uneducated people still need to be educated. Sadly, some of them are teachers - I've even seen university professors use "Gb" when they meant "GiB".

There is almost no situation where knowing the number of bytes in a MB/GB is required in order to successfully use a computer.


Let's try this. If you've got a 2 MB disk and you're downloading data at a rate of 8 KB per second; how long until you run out of space to store received data?

Possible answers include:
a) 2*1024*1024/(8*1024) = 256 seconds
b) 2*1000*1024/(8*1024) = 250 seconds
c) 2*1000*1000/(8*1024) = 244.14 seconds
d) 2*1024*1024/(8*1000) = 262.144 seconds
e) 2*1000*1024/(8*1000) = 256 seconds
f) 2*1000*1000/(8*1000) = 250 seconds

Note: Networking hardware typically uses "K = 1000" and hard drive manufacturers have a nasty habit of using "M = 1000*1024", so (e) is potentially the most likely answer, unless the disk is SSD or USB flash where (a) might be more likely, or the prefixes comply with international standards and (f) is the only right answer.


Standards bodies, I think, overstepped their bounds. While they are responsible for setting standards, they should have sought the opinion of the major stakeholders before making the change they did. And if it transpired they could not at least get a majority YES vote, then it should not have happened.


No. The use of "K = 1024" was always wrong and never complied with any standard (despite common usage). The common usage of "wrong" is likely to have been caused by laziness/convenience (e.g. it's easier to say "1 KB of RAM" and be wrong, and harder to say "1.024 KB of RAM" and be right).

The only thing the standards bodies did was create a more convenient alternative that is right (e.g. it's easy to "1 KiB of RAM" and be right and harder to say "1.024 KB of RAM" and be right).

But yes, some people aren't educated and prefer to remain wrong.

- Brendan

Reply Parent Score: 4

ssokolow Member since:
2010-01-21

"due to concerns that the old "everybody knows kilo- means 1024 in the context of computers" attitude was proving problematic as computers became a more and more generally-used tool.


This is total bull$hit. There is almost no situation where knowing the number of bytes in a MB/GB is required in order to successfully use a computer.
"

I already said they were focusing on business, research, and standards. Situations where a confusion in terminology could be expensive. (eg. If you pay for a supercomputer or space probe to be built and get less memory/storage/bandwidth than you intended or you're forced to go over budget because they specced more than you intended.)

Standards bodies, I think, overstepped their bounds. While they are responsible for setting standards, they should have sought the opinion of the major stakeholders before making the change they did. And if it transpired they could not at least get a majority YES vote, then it should not have happened.


Computer scientists overstepped their bounds in arbitrarily re-defining the meanings of the SI prefixes. This is just fixing the problem.

Reply Parent Score: 4

TemporalBeing Member since:
2007-08-22

" The "xiB" nomenclature was invented by HDD manufacturers just so they could lie about their capacities.


I don't know what input any hard drive manufacturers who had members in these organizations might have had might have had, but the binary prefixes (kibi/gibi/tebi/etc.) were developed by the IEC with support from the IEEE and CIPM (International Committee for Weights and Measures) due to concerns that the old "everybody knows kilo- means 1024 in the context of computers" attitude was proving problematic as computers became a more and more generally-used tool.

It's actually meant to SOLVE the problem you're complaining about... though more in the context of standards, research, and business. (People in related but different fields both saying 5GB but meaning different things)

When I studied, 1GB meant 1073741824 bytes.


...because people in computer science decided "Ehh... 1000? 1024? Close enough. Use the SI prefixes."
"

Personally the whole XiB vs XB nomenclature is just confusing, and doesn't resolve the issue at all. Rather is just makes you have to pay attention to the fine print or lack thereof.

And yes, 1GB ought to simply mean its historical definition of 1024MB or 1024*1024 KB or 1024*1024*1024 bytes.

Simply put - it was for a long time evident that in communications the numbers were base-10 - 1000 bytes = 1 kB, and in computers the numbers were base-2 - 1024 bytes = 1KB. Adoption of the HDD manufacturers using the communications variant is what setup the whole mess, and the supposed solution doesn't resolve anything, just makes it worse especially as the computer definition - which all programmers are use to, and the vast majority programs are written to display - was redefined to the silly XiB nomenclature.

Reply Parent Score: 2