Megabytes (MB) or Mebibytes (MiB)? 437
An anonymous reader says: "KernelTrap has an interesting story about megabytes versus mebibytes. Though the article refers to Linux, the topic is applicable to all computers. Will there be a time when all computer users will talk about adding mibibytes of RAM, rather than a megabytes?
From the article: '[the kernel patch] changes references from the familiar MB (megabyte) and GB (gigabyte) to the NIST standard MiB (mebibyte) and GiB (gibibyte). According to these standards, technically a megabyte (MB) is a power of ten, while a mebibyte (MiB) is a power of two, appropriate for binary machines. A megabyte is then 1,000,000 bytes. A mebibyte is the actual 1,048,576 bytes that most intend.'"
Pronounciation (Score:5, Funny)
Re:Pronounciation (Score:4, Funny)
Its a lousy goddamn word (Score:5, Insightful)
And will hard drive manufacturors decide to stop lying about the size of their drives? Magic 8 ball says doubtful.
ostiguy
Re:Its a lousy goddamn word (Score:5, Insightful)
In addition, since "80 long GB" sounds like you're getting more (and in fact, you ARE getting more) it might encourage HD makers to switch over to the same measure that RAM makers use, and thereby end all this confusion once and for all.
Re:Its a lousy goddamn word (Score:2, Funny)
Amen, brother. Besides, the standard is power-of-two multiples. So, if people want to deviate from te standard, let them use a different word. I propose Mediabytes (1,000,000 bytes) and Greediabytes (a billion).
Likewise I would like a penny for every faulty URL, per copy. That would probably teach everyone that www.somewhe.re in itself is meaningless, or at least not an URL.
And will hard drive manufacturors decide to stop lying ...?
Wrong question: when will corporations stop lying? Answer: when it stops being (seemingly) lucrative.
Stefan.
Re:Its a lousy goddamn word (Score:3, Insightful)
Re:Its a lousy goddamn word (Score:2)
Re:Its a lousy goddamn word (Score:3, Insightful)
When was it defined as a power of two? The meaning of mega, kilo, deca, hecto, pico, nano, femto etc. was defined as powers of two back in 1960, when the SI system of units [nist.gov] was defined as a standard in the scientific world [nist.gov].
I think it's a kludge that kilo means 1024 in kilobyte, while it means 1000 in kilometer, kilogram, kilonewton, kilovolt etc. It's time we give it up, and accept correct terms. I agree kibi sounds ridiculous, but that's just a matter of habit. We'll get used to it.
Besides, not everywhere in CS kilo is 1024 and mega is 1024*1024: in datacommunications, the correct numbers are used. 10 mbit/s is 10000 bits per second.
Re:Its a lousy goddamn word (Score:3, Insightful)
I think it's a kludge that kilo means 1024 in kilobyte, while it means 1000 in kilometer, kilogram, kilonewton, kilovolt etc. It's time we give it up, and accept correct terms. I agree kibi sounds ridiculous, but that's just a matter of habit. We'll get used to it.
I think it's a kludge that hacker means criminal in the media, while it has very good connotations in our circles. It's time we give it up and accept correct terms. I agree cracker sounds ridiculous, but that's just a matter of habit. We'll get used to it.
on weirdness [Re: Its a lousy goddamn word] (Score:2, Insightful)
But think about this: outside of geek circles, *everything* we say sounds weird, and many things have yet to reach wide acceptance. If you tried to explain to someone that
then "mebibytes" would not be the only term they would not understand. Inside geek circles, on the other hand, if you say "mebibytes," people will know exactly what you mean.
Precision in speech and writing is a virtue. In my mind, if this eliminates a little ambiguity in documentation, I think it's a suitable win.
Regarding how the words sound, I happen to like them. They're cute. "Hey there, wittlwe mebibyte... don't be shy..." Perfect for use when talking to an iMac. Hey, wait a minute...
Re:on weirdness [Re: Its a lousy goddamn word] (Score:2)
So you say. There is a lot of disagreement about that on linux-kernel, and hard drives aren't the only problem. (BTW, I think you mean "2^20". If you think a Megabyte is 2e30 (2x10^30) bytes, then we've got bigger problems than I thought.)
Basically the only thing safe to say is that "in the computer industry, megabyte sometimes means 2^20 bytes, sometimes means 10^6 bytes, and sometimes means something else like 10^3 * 2^10 bytes." How can one call that "precise"? It would be nice if we could say that "in the Linux kernel documentation, a MiB is 2^20 bytes, and a MB is 10^6 bytes." I don't think ESR wants to change the world, just to make CML2 is as unambiguous as possible. That seems to me a good goal.
The fact remains that for a very long time before computers because significant, kilo- meant 10^3 and mega- meant 10^6, and giga- meant 10^9. To many people not well-versed in the mess that is computer terminology, these remain their only meanings. When I first became involved with computers (yes, some of us dinosaurs were not born in the age of the PC), I had to learn about all the weird exceptions, and so, I imagine, does everyone else.
Just because hard drive manufacturers use the deceitful practice of quoting disk capacity in units of 1e6 doesn't mean we should all change our ways to match.
Remind me to yell at my old physics professors for using those deceitful SI units. Seriously, the deceit is in taking a pre-existing unit prefix with a long- and well-established meaning and changing that meaning to suit one's purposes. Just because the small portion of our society that is the computer industry has been engaging in the deceitful practice of calling kilo- 2^10 when it suits them doesn't mean we all have to go along with it.
Also, NIST doesn't have any pull with the industry on this - they're a National institute, but they can't mandate usage.
Who said they could? I never said they could. Alan Cox simply calls the prefixes "recommended." Where is this talk about a "mandate" coming from? Do you think the NIST called up ESR and threatened him with incarceration or bodily harm if he didn't make the switch?
There are perfectly rational reasons to go with the NIST's recommendation (reasons which I have touched upon elsewhere). These reasons, in my mind, outweigh the unfamiliarity of the prefixes or the fact that some puerile Anonymous Cowards think they sound "gay".
Re:on weirdness [Re: Its a lousy goddamn word] (Score:2)
A megabyte is 2*10^30?! News to me. (n)e(x) means n*10^x, not n^x. And to top it off, you use the syntax in one way for 2e30 and the other way for 1e6 right after that! I'm only nitpicking because this thread is about precision.
how to use i.e. and e.g. (Score:2, Informative)
It would help (Score:2, Interesting)
Re:It would help (Score:2)
-
Decimal aka Metric: 1 MB = 10^6 or 1,000,000 bytes
-
Binary : 1 mb = 2^20 or 1,048,576 bytes. Note the lower case distinction here. SI prefixes for large units always use capital letters. OK, so I stole this from 4DOS. So shoot me.
- Mixed: a meg (Mb or mB) is one K k, or one k K, which is 10^3 x 2^10, or 1,024,000. And I've actually seen this used a few places.
The problem is that memory chips will always be measured in powers of 2, and marketdroids will always want to use the powers of 10 to make their hard drives look larger, with the fact of SI prefixes to back them up. (Things such as disk cluster/block sizes will tend to be binary, however.) So I always try to say "Binary gigabyte" when I'm talking about memory, if there's any question which I mean.Re:It would help (Score:2, Insightful)
And no it's not false advertising
Yes it is. 1GB = 2^30 bytes, not 1e9. Drive manufacturers use the smaller unit so you'll think that their drives are bigger than they are. That's deceptive
Re:It would help (Score:2)
That's deceptive
Look carefully, you'll find the disclaimer in the fine print...
I heard this years ago (Score:3, Insightful)
Re:I heard this years ago (Score:2)
Agreed. How many people here know that "gigabytes" is actually supposed to be pronounced "jigabytes"? Many I would presume, but that would be 0.0001% of the world. "a hundred jigs" just sounds so ridiculous now.
Furthermore, the poster of the article couldn't even be consistent in *spelling* "mebibytes" or "mibibytes".
Re:I heard this years ago (Score:2)
Conflict (Score:4, Funny)
Good Lord (Score:3, Insightful)
Hm... on second thought, maybe not. I'm not getting much productive work done lately and I still don't care about either of the above...
Metric Revolution (Score:4, Troll)
Most users don't know how many bytes are in a megabyte or a kilobyte, or think (naturally) 1000 rather than 1024.
However, hard drive manufacturers [ibm.com] already use Megabyte to specifically mean 1,000,000 bytes, Before long computer OS's and RAM manufacturers will use the same definition.
Why come up with a new 'Mebibyte' system? What does 'kilo-' and 'mega-' actually mean? Answer: 1000 and 1,000,000, not the perversion of the computer scientists.
Now that computers are becoming more popular, the meaning of the terms megabyte and kilobyte are shifting back to compatibility with normal English usage.
There is no need for new terms at all, IMHO.
Re:Metric Revolution (Score:2)
8 bits make 1 byte. 1024 bytes is 1K. 1 Gigabyte is 1024 Megabytes. 1 Megabyte is 1024 Kilobytes. 1 Kilobyte is 1024 bytes.
and
1Kbps is 1024bps. 56kbps is 57344 bytes per second, about 5K per second. A 128Kbps is really 13K per second of bandwidth. A 768K is 78K per second of bandwidth. A Megabyte is 1024K not 1000K.
Check here [jamesshuggins.com] for a good table.
-
Any fool can tell the truth, but it requires a man of some sense to know how to lie well. - Samuel Butler (1835 - 1902)
Re:Metric Revolution (Score:2)
1Kbps is 1000 bps. The communications industry follows the standard SI prefixes.
And bandwidth should be measured in Hz, KHz, and MHz, not bps, Kbps, and Mbps. Bandwidth is the width of a frequency band, i.e. the difference between the highest frequency of the band and the lowest.
Re:Metric Revolution (Score:2)
HH
Re:Metric Revolution (Score:2)
Re:Metric Revolution (Score:3, Funny)
ugly (Score:4, Insightful)
That's right: ugly it just doesn't sound right, but it is a more accurate description. I don't see the computer world moving away from MB and GB anytime soon though.
Why not use the real numbers? (Score:3, Insightful)
10E9
10E12
and
2E15
2E20
and so on.... No confusion
After all our car tires can work with sizes like 175-70 R13 and son on and we do not mess them up.
Re:Why not use the real numbers? (Score:3)
"2E15" means 2*10^15. What you want to say is 2^15, or 2**15, or 2<sup>15</sup>.
Re:Why not use the real numbers? [offtopic] (Score:2)
Unix printf(1) command:
The C programming language:
The Perl programming language:
And no, this is not an Americanism. It is standard engineering notation, used worldwide.
Umm... (Score:2)
As if there isn't geeks are made fun of...
Scientific notation? (Score:2, Insightful)
1KiB = 1024 bytes = 1x2^10 bytes
1MiB = 1048576 bytes = 1x2^20 bytes
...
or maybe even:
1KiB = 1024 bytes = 1x1024^1 bytes
1MiB = 1048576 bytes = 1x1024^2 bytes
...
or we could abolish bytes too and just say everything in bits:
1KiB = 8192 bits = 1x2^13 bits
But then again, "Hey, I just got 1x2^31 bits of RAM!" just doesn't have the same ring to it...
language confusion? and for what?! (Score:3, Funny)
I already have a hard time convincing other people of the distinction between hacker and cracker.
to them: hacker is a criminal, cracker is a southern, white, klansman-criminal.
I give up, and try and express that every computer hacker is not a criminal, they are all computer science researchers, doing sometimes unpopular work.
Now you want to change the terms for measuring storage? The normal aim for changing terms is to clarify the matter, but this is just obfuscation for 99% of the people in the world, who already suffer at understanding the difference between 1024 and 1000. Please, do not do this.
Use Attacker (Score:3, Insightful)
In the world of hackers, there are attackers, and there are defenders. It's easy to attack. It's much more interesting and important to defend.
--Dan
Comment removed (Score:5, Funny)
Re:And here's the mainstream news version... (Score:2, Interesting)
Mibi? Why not just call them "bibbity-bobbity-boo" bytes?
When I think measurement of data (Meg, Gig) I think powers of 1024.
When I think of measurements of physical objects (water, distance) I think powers of 10.
Now some physicists are saying "no, no, use powers of 2"...STFU.
What really gets me is that the same thing is going on with former @home-ers (charter/att or whatever you got switched to) where the phone monkeys here in Ga are saying kbits and the techs are saying Kbytes. Ok, which is it?
Granted all this is freaking meaningless (check your cable modem settings by pointing your browser to 192.168.100.1) mine at the moment is set to 700000bits when during the switchover it was 128000bits--up and down, btw.
The thing I got on to the tech about: if you are telling me bytes/s then why is my modem set to 128000...you are missing a zero, dude.
(heh, scary that I knew more about DOCIS (sp?) compliance than the tech I was talking to, just a little more than he...but at least he saw my point of view.
Anywho, don't wanna deal with it on vacation, so letting it drop for now...at least disconnections are getting fewer and further between.
At least the are using a decent news server..but that download *speed* cap is annoying, and only 2 connects at once...they need to cache the downloads localy {if possible, mind you} to speed things up and decrease traffic outbound, IMO.
Just my mild rant + $.02.
.
DBZ (Score:3, Funny)
Unfortunately Dragonball Z (Funimation) has already licensed the term "bibbity-bobbity-boo" from Disney for the naming of several bad guys.
.
Re:And here's the mainstream news version... (Score:2)
BTW - the other thing the kernel patch should have fixed is to write 'byte' instead of 'B', since B often stands for bel as in decibel (and 'b' stands for bit). Sometimes in information theory you might be talking about bels and bits in the same sentence, so there is some scope for confusion. Alternatively, make it implicit that all memory sizes are in terms of the machine's addressing capability, and just say '40505 free' instead of '40505 bytes free'.
Re:And here's the mainstream news version... (Score:2)
Violence is destroying our future... and our boxen...
Why not? (Score:2, Insightful)
I always find it odd when people resist what is otherwise a good (if initially less comfortable) idea.
It is inevitable that as computers get more advanced, our technical terms will eventually fall out of use. For example, we don't often talk about "words" of memory anymore often even in the technical mainstream (outside of assembly language and code or storage optimization), because all the manufacturers eventually went to an eight-bit byte as a standard. (Eventually, the byte will become less meaningful to the mainstream, and will eventually shift to the "character" as we head towards the Unicode standard. Eventually, we'll be switching /.ebibytes/ for /.ebichars/, and the complaining will probably begin anew.)
Resisting inevitable changes like this just hinders Linux (and *BSD) from making steps towards the mainstream and maintains the perception that it's only suited to technogeeks.
While "mebibyte" sounds too close to "maybebyte" for my tastes, it does make sense to meld "mega" with "binary" in this way. I wish they'd gone farther; I could have dealt more easily with "mibyte" (pronounced either /mee-bite/ or /mih-bite/) rather than "mebibyte." Perhaps that will become the natural phonetic erosion as such terms get adopted, but that's hard to count on.
On a personal level, clearing up the distinction would at least make things less annoying as far as my life goes. My mother still doesn't understand this whole powers-of-two thing, or even the concept of bits versus bytes, and I don't expect she ever will ("But the modem is 56K, and I'm only transferring at 5K!"). I don't know why I bought a 75GB disk six months ago (75 GiB, to be precise), and then bought an 80GB disk from the same manufacturer last month at about the same price to find myself with exactly the same amount of storage as last time (75 GiB). That ticks me off---I could have used the extra five /Gi?B/. It's really going to tick me off if memory manufacturers start playing similar games. At least unifying this usage will reduce the confusion in the marketplace. (I'd also quit wondering whether a transfer rate of "49K per second" meant 382,812, 384,000, 392,000, 393,216 or 401,408 bits per second. Fortunately, I don't wonder that often, but still.)
I say, let's adopt /[KMGTPA]iB/ as a standard, call 'em /kib/, /mib/, /gib/, /tib/, /pib/ and, uh, /eyeb/, and be done with it. Maybe if we do that, we'll be one step closer to adopting the metric system as well.
something of a linguistic perspective.. (Score:5, Insightful)
The same thing happened with the word Judaism. It's supposed to be pronounced jew-DUH-ism, but in America we call it jew-DEE-ism. While it's not technically correct, everyone knows what you're talking about, and it's the standard, accepted way to talk about the Jewish faith.
Basically this is an effort to reverse linguistic evolution. The current terminology isn't broken for the public which understands gigabyte and megabyte, so don't fix it.
Big deal (Score:2)
I think it's just another geek attempt to confuse the general population as a whole to confuse the general population by adding even more minimally worthwhile techno-jargon.
No. (Score:2)
In a word, NO!
Man I hate dumb sounding words.....
Octets (Score:2)
Re:Octets (Score:2)
hmm (Score:2)
This is bound to fail (Score:2, Interesting)
The same goes with many things in spoken language : for example, the official translation for "email" in France is "mél", but nobody ever says or write that apart from people in the administration (i.e. look for "mél" in a letter and you know it's from the government).
Finally, there is a small argument in favor of keeping Kb, Mb and Gb around : these units are not 1000, 1M or 1G, therefore they are confusing, therefore they constitute in themselves another way for CS teachers to weed out students who have no talent for CS : I used to teach C, and within a week of being told a what a Kb was, I could tell which of my 1st year students were going to struggle and/or not going to make it if they didn't handle the 1000/1024 distinction like they were breathing.
Re:This is bound to fail (Score:2)
You can find the english/french correspondance for some things at [gouv.qc.ca]
I Knew this blind-enforcement of gotta-be-French was braindead Quebec Separatist groupthink. I'm not accusing you of being one of those head-up-their-arse separatists, but the first post I read about enforcing French words no matter how permeated the English term was sounded like something to come out of la belle province's brainiac political leaders.
May seem silly now, but... (Score:3, Insightful)
Re:May seem silly now, but... (Score:2)
We're working on it (Score:2)
How its been done for years ... (Score:2)
This aside, powers of 2 is the only metric that is usefull (IMHO) on computers -- and if it weren't for HD manufacturers thinking they are tricking people (I wager they're just pissing people off), we wouldn't be having this conversation.
Re:How its been done for years ... (Score:3, Funny)
Also, computers need not be based on powers of 2 anyway, so chaining everything to 2 is losing a level of abstraction (eg. quantum computers, ternary computers, EBCDIC computers (9 bits per byte), etc.)
What abbreviation do you use for "million bytes"?
It sounds stupid! (Score:2)
Usage defines meaning (Score:2)
Phillip.
Standards wars... (Score:2)
Will there be a time when all computer networks are based around OSI stacks, instead of TCP/IP?
No.
When it comes to a battle between de facto standards and de jure standards, the de facto standard always wins.
Better solution: switch to base-8 everywhere! (Score:5, Funny)
If we used base 8 like God intended (after all, He gave us 8 finger and 2 thumbs, not ten fingers!) this wouldn't be an issue.
As an extra benefit, the sudden conversion of account balances from decimal to octal numbers will be much need shot in the arm economically. Everyone will be richer! (or owe more money, but we can't all be winners unless we're competing in the Special Olympics.)
Beldar said it best... (Score:2)
Seriously, this is going to be about as successful as the United States' official switchover to the metric system in 1980. The old ways are too ingrained, and the new ones, no matter how much more appropriate they are, will never catch on. The people who really need to know that 1KB=1024B and not 1000KB, already know, and without some fancy-schmancy new nomenclature to tell us.
And manufacturers won't try to force everyone to use the new naming, because the vast majority of their customers can't even be bothered to learn the current terminology-- ever hear someone in CompUSA asking a salesperson how many RAMs or Megahertzes is in the computer they're looking at? I know I sure have.
~Philly
Re:Beldar said it best... (Score:2)
See, in 1980 we 'ignorant Americans' made a choice: we chose to get busy spending our defense budget in case we had to save your smarmy, unappreciative European asses for a THIRD time, instead of making sure McDonald's started selling the "One Hundred Thirteen Grammer" instead of the "Quarter Pounder."
~Philly
Re:Beldar said it best... (Score:2)
Well, if your reading comprehension is for shit, I'm sorry, but what I meant was, worldwide, this will be about as successful as America's switch to the metric system. Nobody else seemed to misunderstand.
You don't think that everyone in the world is an American -- you just think that the rest of the world are all Europeans!
No, but it's a fairly valid assumption on Slashdot that someone posting disdainfully about us 'ignorant Americans,' from behind the 'Anonymous Coward' moniker is European, probably French.
A quandary for dictionary makers... (Score:3, Interesting)
The American Heritage dictionary can't decide:
gigabyte (jg-bt, gg-) n.
1. A unit of computer memory or data storage capacity equal to 1,024 megabytes (230 bytes).
2. One billion bytes.
Princeton University's WordNet decided to decide:
gigabyte n : a unit of information equal to one billion (1,000,000,000) bytes or one thousand megabytes.
The Free On-Line Dictionary of Computing chose the power or two, but went "outside the box" when it came to a definition:
2^30 = 1,073,741,824 bytes = 1024 megabytes.
Roughly the amount of data required to encode a human gene sequence (including all the redundant codons).
I still like the Homer Simpson version... (Score:3, Funny)
Giga-ma-bytes
Saxa-ma-phone
etc.
Damn! (Score:2)
Thanks alot!
My Vote: (Score:5, Funny)
1st: 'Bitches', this refers to 8 bits.
2nd: 'MegaBitches', Obviously, in oldschool terms this would be a Megabyte.
3rd: 'GigaBitches', following the entire byte-to-bitches theme, this would previously have been a Gigabyte.
Some suggested slang based on 'bits-n-bitches':
'Slap'N'ThemBitches', this is what you do when you add any amount of space (memory or harddrive) to your computer.
'StankBitches', bad RAM or a crappy harddrive.
'BadAssMofoBitches', this is any amount of space greater than what you have.
'UglyBitches', this is typically an embarrassingly small amount of space, so much so that you don't tell anyone that's how much you have.
Thanks to our so hip words, now your everyday average IT guy can have a conversation like this with his boss:
"Yo man, yesterday I found some UglyBitches over at the office, and yo, some of them were some StankBitches, yo! So I got rid of them StankBitches and got me some BadAssMofoBitches, and yo, I slap'n'themBitches early this morning. That shit was shweet!"
My $0.02 (Score:2)
Continue to use "megabyte" for 10^6. Use "binary megabyte" for 2^20. If people see "mebibyte" they will think it's a typo.
Advertisers can continue to use "megabyte" in large type without fear since it has a clear-cut definition, even though it does lead to values that are somewhat inflated. The masses probably don't care about this. Geeks can either look for the binary megabyte number in the fine print, or guesstimate it themselves.
Re:My $0.02 (Score:2)
Use "binary megabyte"
Thank you. That is the first term i've seen on the subject (aside from the real stuff like Megabyte) that doesn't make me want to break things. Perhaps i drink too much coffee...
The confusion was ALWAYS there BEFORE... (Score:2, Insightful)
RAM sizes, since they are relevant to hardware binary addressing logic, have always been in sized in power of two. It makes no sense to manufacturer or design for a RAM (or magnetic core array) with 1000 or 1000,000 or 1,000,000,000 bytes or words of memory.
Clock speeds and communications speeds have always been decimal. Or do you think a 1 GHz Pentium has a clock speed of 1,073,741,824 Hz?
Disks are a mess. The total amount of disk storage is continuously variable, and is rarely an exact power of two. On the other hand, the amount stored per sector is related to RAM considerations and is often 512 bytes or somesuch. Disk capacities are sometimes quoted in powers of ten, sometimes in powers of two, and I have even seen "mixed systems" in which 1 "megabyte" of disk space meant 2000 (decimal) sectors of 512 bytes each, i.e. one "disk megabyte" was 1,024,000 bytes.
Nobody has any idea what the current terminology really means. At the one gig level, the discrepancy, 7%, is starting to be annoying. When we get to terabyte disks, which can't be far off, the discrepancy will be 10%. Let's start using terms that have well-defined meanings.
NIST recommends "kibi" rather than "kilo", but... (Score:2)
We all know that KIBO is alive and well in our computers.
Jar-Jar Speak (Score:2, Funny)
Origins... (Score:2, Insightful)
Uhm, mebibyte is the incorrect term... (Score:2)
A megabyte has *always* been 1024k, which is in turn 1024 bytes. Hence a megabyte is 1,048,576 bytes. Always has been, always will be. Giving it a new name simply validates the mistakes of the mail-in MCSEs who've never had to key anything in in hexadecimal.
That's like saying "we're gonna start measuring network bandwidth in megabytes per second cause people are too dense to realize it's actually megabits". It's still wrong.
Re:Uhm, mebibyte is the incorrect term... (Score:2)
mega = 10^6 _everywhere_ with an exception made for bytes, and only a mild exception. You'll notice that harddrives come with packaging that states "1GB = 1000000B" to avoid confusion.
Non issue (Score:2, Informative)
BTW, if you can't remember what the prefix is, remember: first two letters of the SI unit, then 'bi' for binary. A kilo-binary-byte = Kibibyte.
Re:Non issue (Score:2, Interesting)
Brilliant. Now SI includes inherently ambiguous prefixes: what will "debi" mean? Will it be "deci-binary" or "deka-binary"?
MiB and GiB prevents miscommunication (Score:3, Interesting)
For some reason, people who grew up in router land use GB to mean 10^6, while most software developers use GB to mean 2^20.
To resolve this, my group prepared a document that explains the use of the binary nomemclature and we refer readers to this base document in all of our prepared documentation. The document also explicitly states what the accepted abbreviations are (KiB, MiB, GiB, etc.). We also explicity define the capital B to mean byte, while a lower case 'b' is a bit. Therefore, Mib means mebibit.
This has reduced confusion to a great amount and now various groups looking at our performance testing results can make an accurate assesment.
This should catch on ... (Score:2, Funny)
problems, current and future (Score:2)
I can see the motivation for this, but I would be much more impressed if they could get people to properly distinguish between bits and bytes as in Mbps (Mega bits per second) vs. MBps (Mega bytes per second). That's a FAR greater difference (800%) than the 4.8% difference between the proposed megabyte/mebibyte.
While I'm at it, I'd like to see them also straighten out those people who write mbps (which actually means millibits per second; i.e. 1/1000 bits per second!) :^)
Hard to say (Score:3, Funny)
Change was incomplete (Score:2)
Stephen quoted a piece from the diff, showing how Eric Raymond had changed the Configure.help. You'll notice this in incomplete, as I've highlighted below:
An ugly committee hack, we can do better (Score:2)
The thing that really annoys me about the whole Megabyte/Mebibyte thing is that the entire standard nomenclature is an ill-considered, quick, dirty, and above all ugly hack addressing an admittedly legitimate problem (Mega meaning 1^06 or 2^20).
Their hack addresses only powers of 10 and powers of 2, which are a subset of a larger question: nomenclature for abitrary (integer) bases. Worse, it mixes the two together in a misguided effort to get one base's representation to approximate the others, despite the fact that the two bases are in fact quite different!
Why is this so stupid? Well, aside from the internal lack of logic (and, I have to say, profound lack of elegance), let's suppose, for example, that in ten years we begin finding more widespread use of balanced trinary systems , or some other hitherto unforseen base. Where's our nomencalture now? Of completely no use, and requiring us to invent a new wheel, yet again.
A far more reasonable approach would have been a subscript denoting the base, with the default being base ten if no subscript is present (i.e. defaulting to standard metric nomeclature). E.g. M(sub)2Byte would be 2^6 Bytes while M(sub)10Byte = MByte = 10^6 bytes. A M(sub)3trit would be 3^6 trits, and so on.
One will immediately notice that what we consider a (base-2) Megabyte is not 2^6, but rather 2^20 Bytes, or 64 vs 1048576 bytes. Well, they want us to learn a different nomenclature anyway, so why not at least make it logical. If Mega always means to the power of six, regardless of base, then we have a rational basis for our nomenclature. Yes, it would take some getting used to, but I would argue it would be far less painful getting used to something this logical than to adopt the use of "mebibyte" in our daily language. YMMV of course.
This ugly hybridization of base-10 nomenclature with base two numerology they are intending to replace (admittedly equally bad) common usage with is both illogical and unnecessarilly specific to one problem set. If we're going to be making up new (and apparently stupid) terms like mebibyte, then lets at least define mebi to represent a power of 20, or better yet 21 as it would then follow exa by an additional power of three, as every other prefix above kilo (and below milli) does. Or better yet, pick a name that doesn't start with the already (overused) 'M'.
Does anyone else see the advantage of this? We have just extended our available nonemclature for all measures, in any base, in a rational, extensible, and fairly scalable approach. Yes, to our base-10 minds we may feel uncomfortable with the small size a Megabyte really is, but I would submit that that is no greater a psychological barrior to overcome than the use of really stupid, childish, and annoying terms such as "mebi," and a heck of a lot more rational to boot.
Of course, this idea came from one person, spontaneously, on a Sunday morning, who (at the time) hadn't even had his coffee yet. Give a self-appointed committee time enough to dumb it down and who knows what hideous form it would then take...
Do we really need the base-2 units? (Score:2)
With computers moving further and further into the daily lives of common (that is, non-CS) people, I suspect we're fighting against a horde of people who already believe that a megabyte is 10^6 bytes.
I agree with some of the others here that "mebibyte" is a very clumsy word.
But I'm wondering, what's with the attachment to the whole base-2 system anyway? I mean, I'm a CSist, and *I* don't know how many bytes are in a gibibyte. I have to run to my calculator.
I'm probably speaking too late to be heard, anyway, but I say ditch the whole base-2 thing and start saying 42.9 gigabytes instead of 40 gigabytes. If you want the exact amount, read the documentation, where (one hopes) the size will be expressed precisely in bytes.
I would like to think that of all the groups of people in the world, computer scientists would have the cool rationality to be able to let go of a misguided standard.
But I guess that, so long as there are those who feel that it measures their penis size, they won't be able to let it go.
Mega also 10240000 (Score:2)
Cylinders * heads * Sector * allocations
= 80 * 2 * 18 * 512
= 1474560 = 1.44 * 1024
Seems the actual style is to use k indifferently for 1000 and 1024, as the end suits the need.
Pointless, Revisionist, And Stupid. (Score:3, Informative)
1. A megabyte is 1,048,576 bytes. Always has been, always will be. It is a unit of measurement specific to its science, like the Mole is to Chemistry, and like the Newton is to Physics. It is not meant to be a general metric measurement, in other words, a megabyte isn't 1,000,000 anymore than a "dozen" is 10. Ever since the term was coined, it has meant that value, specifically. Anyone who tells you otherwise is trying to sell you something, namely marketing idiots who are responsible for great advancements in human culture like MTV and the Home Shopping Network.
2. By accepting this "mebibytes" crap, you're allowing marketing people to revise history. The number 1,048,576 is an important value in Computer Science, similar to 8, 256, 1024, and other commonly-used powers of 2. An understanding of the powers of 2 is integral to having an understanding of the underlying principles that form the foundation of this discipline. If you cant think in anything but base 10, you should consider a different line of work, as most computer scientists have no problem thinking in terms of binary, octal, hexidecimal and otherwise. A failure to understand the basic nature of the device you intend on working with for the rest of your career is tantamount to unprofessionalism and neglect. After all, you can't be expected to code competently by using incorrect measurements any more than a carpenter can be expected to build a house competently if his tape measure is made out of elastic rubber.
3. Its just plain stupid. A megabyte is a megabyte. Its not less than a megabyte, or more than a megabyte. If you for some reason feel the need to apply a term to "1,000,000" an essentially meaningless number in terms of the machine, we already have a word for it. Its called "million", as in "a million bytes." Call a spade a spade. A megabyte is 1,048,576. A million bytes is 1,000,000 bytes. They are not equal, and never will be.
Re:Incorrect story quote (Score:2)
Hmm - sounds just like double-speak from 1984
We'll adopt these here in the US just like we've adopted the metric system.
Re:Incorrect story quote (Score:2)
1 kilogram = 1000 grams
1 kilobyte = 1024 bytes
1 kilotonne = 1000 tonnes
1 kilopasical = 1000 pascals
Spot the odd one out?
Have you ever bought a harddrive labelled "40Gb" and gotten home to find it's only 37Gb ?
Re:Incorrect story quote (Score:2, Funny)
No, I've never seen anyone selling 40 gigabit hard drives.
Re:Incorrect story quote (Score:3, Funny)
Oh, how quickly we forget....
Bytes = bits/8 (not accounting for ecc etc)
Equals 5 GigaBytes (in the proper sense of Giga)
5 gig drives were all the rage back in '95
Good thing the marketing droids didn't pick up on the confusion with B and b
Re:Incorrect story quote (Score:2)
The SI prefix Mega has always meant 10e6, which is why HD manufacturers use its true definition to sell their wares.
Oh puh-lease.
HD manufacturers use 10e6 because it makes their capacities sound bigger, just like monitor manufacturers use the tube size instead of the viewable area because it makes their wares sound bigger. Correctness has absolutely nothing to do with it.
Re:Incorrect story quote (Score:2)
used on bytes or collections of them.
Operator overloading (Score:2)
feature of the English language
When I think about the desktop, I don't need to
be bothered that my window manager doesn't actually
have any desk to corrispond to it
The standard is useless!! (Score:2)
A Megabyte === 2^20 or approximately 1 millon bytes, etc. That is the industry standard definition of the term. Even though they are borrowed from SI doesn't mean they need to follow SI. This is a 30 year old term at this point and NIST should have better things to do!
Standard, huh? (Score:2)
If it's the industry standard, why don't HD companies use it to mean 2^20 instead of 10^6 so that there would be no confusion? Then every reference to megabyte would always mean 2^20 and we'd be happy. As it is, I think your standard needs work...
Re:hrm (Score:3, Funny)
MIBs are already taken, too.. by SNMP. Unless you pronounce them "emm eye bees".
Re:hrm (Score:3, Funny)
Quick, call Will Smith and Tommy Lee Jones!
Re:TOO FUCKING LATE! (Score:2)
1 megabyte (8,338,608 bits) != 8 megabit (8,000,000 bits)
Re:TOO FUCKING LATE! (Score:2)
1 megabyte (8,338,608 bits) != 8 megabit (8,000,000 bits)
Uhh, where I come from a megabit is 1048576 bits. i.e. a 1 megabit FLASH or EPROM actually contains 1048576 cells, usually arranged into an 8 x 131072 array, giving you 128 kilobytes.
Hmm, but now that I think of it, the raw throughput of a DS1 is 1.544megabits per second but that is 1544000 bits per second...
Re:Huh? what? (Score:2)
It isn't helped by MS hardcoding the definition of WORD and DWORD to two bytes and four bytes back in the windows 3.1 days (when a 16 bit word made sense).
Re:Dog Food? (Score:2, Funny)
It's ok though, because W. will hold them off using the power of bacon and shrimp chips.