You seem to be stuck in an overly compartmentalized universe, where this one area of technology has every right to redefine a term that every other field of human endeavor defines differently. As pointed out above, grams, meters, seconds, and every other quantity with something resembling widespread use define it in the same way, as a power of 10. The term has been defined the same way everywhere else, therefore, it's stupid to redefine it for the purposes of convenience.
Why rewrite all software, and god forbid, patch all old software going back however many DECADES into the past to implement this change, when harddrive manufacturers could simply start labelling their drives correctly?
Because that software is INCORRECT. Should we have just decided that time was circular instead of fixing the Y2K bugs? It would have been easier!
Besides, when you buy a gigabyte of ram, are you really getting 1 billion bytes? or 1073741824 bytes? You tell me:)
Your premise is false. You actually bought a gibibyte of RAM. You ought to be pissed that they mislabeled the product.
Drawing a distinction between computing prefixes and SI prefixes, when they share the same name, violates the principle of least surprise. As software guys, we should be receptive to that.
How is the base in which the computer counts relevant? The computer outputs in (bitmap representations of) base 10.
Do you also expect your 1 GB/s network connection to transfer the contents of your ram in 1 second? Because it won't be able to. What it boils down to is that your initial assumption is incorrect. You have 1 GiB of RAM. You decided to make a bad approximation of how much RAM you have because you like round numbers but hate pedantry. You are now angry that your approximation has led to imprecision. That is stupid.
Computers also use ASCII, not bitmaps. However, we convert from one to the other when outputting information for human use. Just as we now have graphics hardware that can do the conversion, we also have ALUs that can divide by 10.
Yes, for example networking hardware uses 10^x prefixes for bytes. And SI prefixes are used to measure energy, which is quantized. And yes, you can have 1.3932 bits. That's about what the entropy per character of English is, for example. Finally, your argument against the pedantry of gibi- is that it mixes latin and greek? Can you not see the irony in that?
So if I dream up some new type of unit, am I free to use powers of 3*pi for it? Or would that be stupid? Is this less stupid because it's a better approximation?
"Computers" includes networking, where they use SI prefixes. Which means that your hard drive fills up at only 1 kB/s when you're downloading at 1.024 kB/s. How is that not retarded?
There's technical merit behind using powers of two. No shit. The technical merit is lacking in the practice of abusing existing terminology that refers to powers of 10. We should have made up our own words from the start. We fucked up. Thus we don't have a right to whine.
Nobody's saying mebi- is in general use, but that it ought to be. The SI- prefixes were here first. Just because the computer industry initially misused them doesn't mean that use is correct.
Just because you can't convert between different units doesn't mean we should use a different base for different units. Especially if we use the same prefixes. That's stupid. I like powers of two as much as the next guy, but you can't deny they really ought to have different prefixes. If only kibi- and mebi didn't sound so stupid.
According to the back of this envelope, on an unformatted floppy disk you can fit about 100 bits on an area the size of the period at the end of this sentence. No wonder the damn things never work. Heh, they were great for getting out of homework back in the day.
Actually, the patent describes using "-/" to denote a URL that is not a search string. (I can't place where, but I swear I've seen URLs like that before.)
When I first saw it I thought to myself "Huh, that's pretty nifty, I wonder how they did it?"
I'm fairly sure even Ted Stevens could have told you that after the URL comes out of the tubes, the server figures out whether it's a search string, and if so, sends you an internet with the results.
They also tend to say things like "just because it is obvious to *you* doesn't mean it is. Please don't worry about whether it is obvious. Let us do that".
Yeah, just because it's obvious to you, who are skilled in the art, doesn't mean it's obvious to those ski... oh wait...
The patent doesn't specify that it needs to be a full text search. Wikipedia does normalize (fix caps, remove spaces, add parenthesis) the article names, and it's certainly using a database of some kind. Claim 8 requires "a query server that processes search strings from users to identify responsive items within a database" and that the query server will "execute a search in which the character string is used in its entirety as a search string to search the database of items". In this case the items are article titles.
Anyway, Wikipedia certainly isn't the only example of prior art. Reading Slashdot so far I've seen php.net and answers.com. (Dunno if they're prior.)
Slashdot Mirror
Um, to one decimal place, 1048576 bytes = 1.0 MiB = 1.0(48576) MB.
You seem to be stuck in an overly compartmentalized universe, where this one area of technology has every right to redefine a term that every other field of human endeavor defines differently. As pointed out above, grams, meters, seconds, and every other quantity with something resembling widespread use define it in the same way, as a power of 10. The term has been defined the same way everywhere else, therefore, it's stupid to redefine it for the purposes of convenience.
[citation needed]
Because that software is INCORRECT. Should we have just decided that time was circular instead of fixing the Y2K bugs? It would have been easier!
Your premise is false. You actually bought a gibibyte of RAM. You ought to be pissed that they mislabeled the product.
Easier still would be to remember that they have 1 GiB of RAM.
Drawing a distinction between computing prefixes and SI prefixes, when they share the same name, violates the principle of least surprise. As software guys, we should be receptive to that.
:-)
This is a really fun flamewar.
How is the base in which the computer counts relevant? The computer outputs in (bitmap representations of) base 10.
Do you also expect your 1 GB/s network connection to transfer the contents of your ram in 1 second? Because it won't be able to. What it boils down to is that your initial assumption is incorrect. You have 1 GiB of RAM. You decided to make a bad approximation of how much RAM you have because you like round numbers but hate pedantry. You are now angry that your approximation has led to imprecision. That is stupid.
Computers also use ASCII, not bitmaps. However, we convert from one to the other when outputting information for human use. Just as we now have graphics hardware that can do the conversion, we also have ALUs that can divide by 10.
Yes, for example networking hardware uses 10^x prefixes for bytes. And SI prefixes are used to measure energy, which is quantized. And yes, you can have 1.3932 bits. That's about what the entropy per character of English is, for example. Finally, your argument against the pedantry of gibi- is that it mixes latin and greek? Can you not see the irony in that?
So if I dream up some new type of unit, am I free to use powers of 3*pi for it? Or would that be stupid? Is this less stupid because it's a better approximation?
"Computers" includes networking, where they use SI prefixes. Which means that your hard drive fills up at only 1 kB/s when you're downloading at 1.024 kB/s. How is that not retarded?
There's technical merit behind using powers of two. No shit. The technical merit is lacking in the practice of abusing existing terminology that refers to powers of 10. We should have made up our own words from the start. We fucked up. Thus we don't have a right to whine.
Nobody's saying mebi- is in general use, but that it ought to be. The SI- prefixes were here first. Just because the computer industry initially misused them doesn't mean that use is correct.
Just because you can't convert between different units doesn't mean we should use a different base for different units. Especially if we use the same prefixes. That's stupid. I like powers of two as much as the next guy, but you can't deny they really ought to have different prefixes. If only kibi- and mebi didn't sound so stupid.
I just bought 2 gigs of RAM, but when Vista booted, I ended up with only like 1.7. The nerve!
Interestingly, according to Wikipedia, the formatted capacity of a floppy is actually 1.44 kibi-kilobytes.
I'm adding "kill the first person to use SI prefixes for powers of 2^10" to my list of things to do when I get my time machine.
Don't the taxpayers fund the BBC? Sorta throws a wrench in the "if you don't like it, take your business elsewhere" angle.
According to the back of this envelope, on an unformatted floppy disk you can fit about 100 bits on an area the size of the period at the end of this sentence. No wonder the damn things never work. Heh, they were great for getting out of homework back in the day.
Are AD/HD and autism really on the rise, or are we just diagnosing them more?
That's better than the people who don't admit they don't have a clue, and get modded up by mods who think it sounds truthy.
Actually, the patent describes using "-/" to denote a URL that is not a search string. (I can't place where, but I swear I've seen URLs like that before.)
Algorithms are math. They are also thought processes. Both are unpatentable.
I'm fairly sure even Ted Stevens could have told you that after the URL comes out of the tubes, the server figures out whether it's a search string, and if so, sends you an internet with the results.
Yeah, just because it's obvious to you, who are skilled in the art, doesn't mean it's obvious to those ski... oh wait...
The patent doesn't specify that it needs to be a full text search. Wikipedia does normalize (fix caps, remove spaces, add parenthesis) the article names, and it's certainly using a database of some kind. Claim 8 requires "a query server that processes search strings from users to identify responsive items within a database" and that the query server will "execute a search in which the character string is used in its entirety as a search string to search the database of items". In this case the items are article titles.
Anyway, Wikipedia certainly isn't the only example of prior art. Reading Slashdot so far I've seen php.net and answers.com. (Dunno if they're prior.)
Absolutely. One only needs to read half the replies to this topic to see that the illusion of control is everything to people's perception of safety.