What forced revisionism?

Things like hard drives often used decimal/metric sizing from the start. Because their capacity has always been based on physical platter size and density, not powers of two the way memory is.

So this confusion has been with computing since the beginning. The attempt to introduce units like KiB isn't revisionism, it's an attempt at clarity around something that has always been ambiguous.

And obviously, if you need two separate prefixes, you're going to change the one whose unit of measurement differs from all the rest of science and technology.

▲

kazinator a day ago | parent [-]

> The attempt to introduce units like KiB isn't revisionism

Yes it is; it is literally asking people who call 1024 bytes "kilobyte" to stop doing that and say "kibibyte" instead, and to revise the meaning of "kilobyte" to 1000 bytes.

Some people have not stopped doing that, so there is more confusion now. You no longer know whether a fellow engineer is using powers of 1000 or powers of 1024 when using kilobyte, megabyte or gigabyte; it depends on whether they took the red pill or the blue pill.

▲

crazygringo a day ago | parent [-]

It's a partial renaming but it's not revisionism.

> You no longer know whether a fellow engineer is using powers of 1000 or powers of 1024 when using kilobyte, megabyte or gigabyte

You never knew this, that's the point. You didn't know it in e.g. 1990, before KiB was introduced in 1998. People didn't only start using powers of 10 once KiB was formally introduced. They'd always used them, but conventions around powers of 10 vs 2 depended greatly on the computing context, and were frequently confusing.

There isn't more confusion now. Fortunately, places that explicitly state KiB result in less confusion because, at least in that case, you know for sure what it is.

Unfortunately, a lot of people won't get on board with it, so the confusion persists.

And frankly, I don't care what you call it when you're speaking, as long as you just use the right label in software and in tech specs.

▲

kazinator 21 hours ago | parent [-]

> You didn't know it in e.g. 1990

False: source, I was there. Kilobyte and megabyte were powers of 1024, except in well-delineated circumstances (mass storage devices).

The size labeling of mass storage devices was widely reviled due to using a weasly definition of terms that everyone normally undestood to be powers of 1024.

> a lot of people won't get on board with it, so the confusion persists.

The idea that people refusing to change their behavior according to someone's wishes are causing confusion is fallacious.

Of course it's those introducing change that are introducing confusion.

The kibi-mebi people failed to predict human behavior; that they cannot just roll out a vocabulary change to all of humanity the way you roll out a new kernel throughout a machine cluster.

The irony is that you can even find people who were not born at the time, who are using kilobyte to mean 1024 bytes.

	▲	crazygringo 21 hours ago \| parent [-]
		I was there too. What you call being "widely reviled" is just another way of saying "in common usage". Maybe you "reviled" it, but it was just convention. So it's not false, and it wasn't well delineated. Just like it continues not to be. Why don't you take a look at Wikipedia which clearly describes the many, many, many places in which powers-of-10 is used, and then also has a section on powers-of-2: https://en.wikipedia.org/wiki/Byte#Units_based_on_powers_of_... Remember, it wasn't just hard drives either. It's been data transfer speeds, network speeds, tape capacities, etc. There's an awful lot of stuff in computing that doesn't inherently depend on powers of 2 for its scaling. And so as long as we have both units and will always have both units, it makes sense to give them different names. And, obviously, the one that matches the SI system should have the same name as it. Can you seriously disagree? Again, I don't care what you say in conversation. But in labels and specifications, how can you argue against it?