- Feature Articles
- CodeSOD
- Error'd
-
Forums
-
Other Articles
- Random Article
- Alex's Soapbox
- Announcements
- Best of…
- Best of Email
- Best of the Sidebar
- Bring Your Own Code
- Coded Smorgasbord
- Mandatory Fun Day
- Off Topic
- Representative Line
- News Roundup
- Editor's Soapbox
- Software on the Rocks
- Souvenir Potpourri
- Sponsor Post
- Tales from the Interview
- The Daily WTF: Live
- Virtudyne
Admin
Yeah. Can we see some NaN's instead?
Admin
I'd have to say why are you using Safari then?? S*&ts all over IE5.5
Captcha: Darwin, because sometimes I start to believe in the big bang..
Admin
[quote user="AdT"][quote user="fuzz"]Why the F is kilogram considered the base unit of mass instead of gram? [/quote]Because of the French Revolution. It was supposed to be the grave, but after the French government became unpopular, so did the unit.
By the way, there's a simple way of dealing with the whole KB vs. KiB mess - stop using nonstandard prefixes. Suck it up, nobody outside IT cares whether powers of 2 are easy to calculate; "kilo" means "one thousand", period. Don't want to count in powers of ten? Then don't make it look like you do.
I do understand base-2 prefixes, but I make a point of writing them with the 'i' (even if I do use the ambiguous term in speech) - there's no reason at all to be ambiguous on purpose and, quite frankly, it's pretty immature to cry over hurt pride because the rest of the world doesn't want to confuse your nonstandard prefixes with real ones.
By the way, I have heard "binary <prefix>byte" as an alternative that is standard-consistent and doesn't sound stupid. Yes, it's less efficient, but so is all language. Use this only to clarify which kind of prefix you're talking about, though; otherwise some smartass might remark that bytes tend to be binary.
Admin
Admin
cool! Penta! I used to play that in the Arcade
Admin
Oh, I see. Historical inconsistency is ok when it comes to SI units, but when it comes to the long time de-facto standard in IT, the excuse doesn't wash. Double standards for sure.
If you had read my post - all of it - you would have noticed that mega meaning 1 million, giga meaning 1 billion etc. are only standard in science. And information theory is not science, period. It belongs to mathematics (where "kilo", "mega" etc. had no meaning previously).
Although there hasn't been a de jure standard in IT, the de facto standard was using powers of 2, which is much more, not less consistent when talking about computers that entirely revolve around the binary system. Just because SI used these prefixes for powers of 10 - for scientific units only - nobody may use them in another way outside of science?
And I don't care what kind of words anybody outside IT wants me to use. Suck it up!
I am mainly hesitating to use these prefixes unless anybody might think I endorse the farcical "kibibyte" baby language. My pride is not hurt, but it will surely be if I start babbling like a newborn.
I'd say that if and only if the precision is actually required and the person I'm talking to cannot be assumed to know what I mean otherwise. When talking to a non-techie, this is most likely totally pointless - (s)he will just scratch his head whether I'm saying "kibibyte" or "binary (anything)", and ask "Ok, so how many MP3 songs is that?"
Another reason not to use powers-of-ten prefixes in this context. And you can call me smartass till you bleed.
By the way, science nowadays makes massive use of binary computers. Plus: The binary system is used because almost all computers use it. The decimal system is only used because some neanderthal noticed that he has ten fingers. Isn't it horribly inefficient and inaccurate to use decimal numbers and decimal powers with binary computers? Why don't we redefine SI prefixes to mean 10^1010, 10^10100, 10^11110 etc. (all in binary)? Why don't we all use binary? Think about how easy it would be to learn the multiplication table! There you have it: Unprecedented consistency. Eat your own medicine, dogs!
(Seriously, the binary system is cumbersome to use in talk and writing but I wouldn't mind if the decimal system gets replaced by sexadecimal*.)
*) The consistent version of "hexadecimal".
Admin
By the way, this describes how the kilogram was standardized in 1875, but it doesn't explain why SI considers it the base unit of mass since 1960.
Admin
I'm not decended from a Neanderthal. You're probably not either (we're decended from Cro-Magnon beasts). Our number system (base 10) came from India centuries ago. At which point all Neanderthals were extinct.
Ther are tribes (in south America, I think) that count in binary.
For an interesting look at where our number system comes from read Zero: Biography of a dangerous idea
And to understand why this kibi/kilo argument is pointless try The Language Instnict
In the end the most popular one will win, and the least popular one will disappear. Reasoning that one reduces confusion or the other is outside the scope of standard scientific notation is futile (or whatever your argument).
Personally I hope that kibi disappears, I just don't like it.
Admin
"Science" doesn't need to justify having a consistent set of unit prefixes. (Incidentally, SI units are also used extensively in engineering - and computing falls under both umbrellas, at least in part). The fact is that giga-, mega-, and kilo- have consistent meanings except when discussing binary multiples. The mistake was made by early computer scientists; they should never have introduced the terminology which is now sadly incumbent. Confusion can, and does, arise. Just look at how the gigabytes one gets in one's web quota are most likely different to the gigabytes your BitTorrent client downloads... even setting aside the question of protocol overhead.
If the same technical term is trying to mean two different things, it makes sense for the older and more widely used definition to take precedence. I'm not really trying to argue in favour of the "mebi" system, so much as against overloading "mega" in such a way. If it's possible to get an accepted alternative for binary exponents (which is looking doubtful), I think we should certainly do so.
Furthermore, I put it to you that the "kibi" prefixes are in fact no more "silly" or "baby-talkish" than SI prefixes. There's no objective difference; it's purely unfamiliarity that makes it sound odd to you.
Seriously, say "gigajoule" a few times and tell me it doesn't sound like baby talk. "Gigajoo" sounds like exactly the sort of thing one would say bouncing a baby up and down!
But truth be told, although I do use "KiB" in written documents, I don't usually pronounce the binary prefixes in speech. They're not widely enough known at the moment.
Admin
You are correct in all particulars (though I don't share your distaste for kibi). However, the implications of your comments are noteworthy.
It's true that the most popular terminology will survive. However, there is only one group of people that can possibly determine the fate of the "binary prefix" concept: the computer & IT industry, since that comprises more than 99% of the potential users. If it is useful enough for disambiguation, it will survive. If not it will go the way of many other neologisms in search of a user-base.
That's why forums such as this are probably the only place where it's worth discussing the binary prefixes: these are places where the potential users are found. If I talk to a bloke in the pub about it, I've just wasted time. But in a place like this, it's actually possible to have some impact. Naturally, a new term will not be taken up unless there are reasons for doing so, since there is a cost involved (having to explain to people what the new prefixes mean and why it's worth using them). So, it is worth talking about the reasons behind the "kibi" notation; if no one talks about the reasons for them, then it's certain that the terms won't be adopted.
(By the way, the story behind the terms "big-endian" and "little-endian" is another interesting overlap between IT and linguistics - the terms would probably be barely known today outside of their IT context).
Admin
TRWTF is that people are writing whole assays about the evils of the new SI prefixes.
I, for one, like them. They resolve some lingering confusion I always have when I'm working with disk-space requirements and specifications.
Admin
Actually, "giga" is supposed to be pronounced as "ji-ga". If you watched the Back to the Future movies, the flux capacitor takes 1.21GW (gigawatt), using the more traditional pronunciation. The way everyone says it a la "gigabyte" is actually a corruption caused by computer scientists.
It's only used so prevalently that both meanings are actually acceptable. Which is OK since they mean the same thing (1 billion) regardless of pronunciation.
Anyhow, I suppose part of the reluctance to use the binary prefixes is because it's admitting that marketing people were right for their trickery, and everyone gets offended when that's the case.
Admin
Maybe, and I'm just throwing this out there, he really meant cenobite?
Admin
BTW, doesn't everyone think we should have just one popup, every day? Shouldn't we have some sort of discussion about this???
Admin
I agree that dicussing the terms will promote the use of the kibi. I disagree in the need for disambiguation.
There is no need for a term for 1000 bytes. The measure is totally pointless, as bytes are counted in powers of two. So, what people are really asking for is the removal of the term kilobyte (etc.) altogether. Yes, it is a term bourne out of convenience (oh! 1024 bytes is almost 1000 bytes, lets call it a kilobyte) but so is the whole of language.
The only real way the term kibi would be adopted over kilo is if it's adopted at the level of hardware vendors and picked up by magazine within reviews (not adverts) and other places, such as schools. Trying to argue with people stuck in their ways (like me) isn't going to work. We'll just say something like "Why standardise this part of language, when no one else is standardising theirs? Citizens of the USA and the British still measure everything in miles, that's far more stupid."
Admin
Interesting how most of the posts here have nothing to do with the article...
BTW - for all those involved in unit prefixes, here are some questions for you to answer. Assume that the context is related only to computing:
I have 4 GB of data, how much HDD space do I need to store it (not including file system overhead)?
I have 4 GB of data. How long will it take for me to transfer it over a connection that transfers 256 Mb/s (again, assume no overhead for simplicuty)?
How long MP3 audio recording at 128 Kb/s CBR can I store in 128 MB available space?
How many KB do I need to store 10 seconds of uncompressed 16 b stereo (total 32 b) audio sampled at 44.1 kHz?
If I place n KB of memory that dissipates some thermal energy E for each bit it stores in 1 L of clean air-filled space having temperature of 295 K, how long will it take to heat that space by 5 K? Can you express that "heating speed" as "Kelvins per Kilobyte second"?
What the heck did I do I my first 4.5 questions? Is "KB" a Kelvin Bel or something else? Was I supposed to be using kB (kilobel) and kb instead?
'nough... you can all get any points you like. Whether current system is good or not is up for debate, but "context" can easily be ambiguous. This exists elsewhere as well:
How many liters are in gallon? (UK or US)?
How many millions are in a billion? (in various languages different amounts, sometimes million, other times a thousand; some languages have a special name for a "thousand million", e.g. http://en.wikipedia.org/wiki/Milliard).
...
Admin
One small correction to above ... didn't mean 256 MB/s but 256 KB/s. This may be very important to those who pay attention to the "context".
Admin
Admin
[quote user="NeoMojo"][quote="AdT"]In the end the most popular one will win, and the least popular one will disappear.[/quote]That therefore debating about them and reasoning about then is pointless does not follow, because debates and arguments can influence popularity.
Admin
[quote user="Shinobu"][quote user="NeoMojo"][quote="AdT"]In the end the most popular one will win, and the least popular one will disappear.[/quote]That therefore debating about them and reasoning about then is pointless does not follow, because debates and arguments can influence popularity.[/quote]
It's as influential as a room full of people shouting the term. Reasoned argument is just hyperbole when it comes to the uptake of new language.
Admin
Admin
Fair enough, though, I do get your point - "mega", etc, had no scientific meaning as prefixes until SI came along, but some of these prefixes (or variants of them) were already in use in words derived from Ancient Greek, such as "millipede", "decagon"* and "chiliad".