I've recently come to an understanding about the proper standards for binary prefixes for quantities of data - i.e. There are Decimal Prefixes that are commonly used: KB (Kilobyte) meaning 1,000 bytes, MB (Megabyte) meaning 1,000,000 bytes, etc.
And there are the SI Binary prefixes measuring quantities in two to powers of multiples of ten: KiB (Kibibyte meaning 1024 bytes), MiB (Mebibyte meaning 1,048,576 bytes). Historically manufacturers have always used the decimal system, and operating systems have used the binary system, but with the same prefixes, which, as of 1999, is no longer correct by the IEC standards.
I was wondering if it is plausible that Directory Opus could have an option to use the correct prefixes (with the "i")for quantities, as defined by IEC 60027-2. Maybe someday even Microsoft will implement this properly into their operating system, even!
I can't say I really see the point - unless you're seeing the actual byte size of a file it's an approximation anyway, so whether "1.5MB" means ~1.510241024 or ~1.510001000 wouldn't really make much difference would it?
Agreed, and as far as I know/remember, KB, MB etc. have always meant multiples of 1024 in computer-software contexts. It seems strange to now try and redefine what KB, MB mean and introduce the new (and IMO ugly) KiB and MiB that mean what KB and MB used to mean. That doesn't resolve confusion, it creates more of it.
I wouldn't be against new things that represented multiples of 1000 since to me they have always been the odd/exceptional case with 1024 the norm, but maybe that reflects that I'm a software person and not a hardware person and people on the other side would feel the reverse. Either way, redefining KB and MB in the software context after so many years doesn't seem like the right thing to do, even if KB and MB were badly chosen due to their inconsistency with other things like km (kilometres). If the definition of a word or symbol is wrong for 30 years it becomes right and trying to change it after so long is confusing.
If we really must have new symbols then then I think we should have new symbols to represent both 1024 and 1000 multiples and do away with the old, ambiguous suffixes entirely.
Windows itself still uses the KB/MB/GB suffixes to represent 1024 multiples, as does pretty much everything else I've encountered, so it seems the software industry has rejected the SI's proposal (or is unaware of it, but it's been there since 1999).
All of that said, I'm not against the idea of an option in Opus for people who feel differently. This is more my reaction to the idea of KiB and MiB in general than the idea of them being supported by Opus. (Gah, you can't even type those suffixes easily... What were they thinking? ...I bet it's all a marketing ploy for Men In Black.)
There's really good point in the Wikipedia discussion pages among the heated arguments:
HDD manufacturers report sizes in 1000 multiples and use a footnote to point this out to avoid confusion, because people are used to KB being 1024. (Or at least are used to what Windows reports and thus implicitly used to KB beign 1024 even if they don't know it.) They get sued sometimes when they don't make it clear enough enough, yet no footnote or lawsuit is needed when people say KB and mean 1024 multiples.
In addition, while HDD manufacturers quote HDD size in 1000 multiples with footnote, when they're talking about the size of the HDD cache they report it in 1024 multiples without a footnote.
Heh... in a way it's not any easier for the lay person, but at least now with the new SI prefixes, it is very specific and certain that you are talking about 2^10's rather than 10^3s. I can guess that the idea behind this was that since the kilo, Mega, Giga prefixes are powers of ten, then to be consistent and straightforward, 2^10 multiples should have their own prefixes.
I don't know if the world outside of the computer science community will ever fully embrace this system, but it is a pretty common question I get from people... "How come my 200 GB hard drive only shows up in Windows as 186 GB?
With consumer disk drives exceeding 1TB capacities, that discrepancy will become significantly more noticeable than it was with 100 MB drives.
100,000,000 bytes = 93 MiB - a 7 MiB discrepancy
1 TB = 0.909 TiB - a "loss" of almost 100 Gigs!
As for Dopus, it'd be neat to be able to select the units you want. Maybe, as with one of the number formatting packages for Perl, you could arbitrarily specify your own unites for any unit you like... Probably not useful unless it's really easy to do, but an idea.
I hate to bump such an old post but this seems EXACTLY what I was looking for. I assume this is why Dopus is saying my brand new 4TB Hard Drive is 1% full.
It was added a long time ago: Preferences / Miscellaneous / Advanced: file_size_units
(But changing which units are used to display sizes will not affect percentage values at all, for reasons that should be obvious. A 1% full drive will still be 1% full no matter which convention is used, and the raw free space statistics come from Windows so you should see the same percentage (subject to differences in how things rounding values up or down to the nearest whole number) in all programs.)
Well, better yet "omg I need more storage space" before you really need it is better than "omg I thought I had more room on here" when your drive hits 0 bytes free.
