If I want the price to be free for my über game, and Amazon decides it is worth $5? They get $4 per sale and i get $1 (that I didn't ask for).
Wrong. The formula (at least the one given in the summary is max(0.2*(suggested price), 0.7*(real price)). So if your suggested price is $0, but the real price (set by Amazon) is $5, then you get max(0.2*$0,0.7*$5) = $3.50 -- indeed, if you look at the formula, you never get less than 70% of the actual price, however you'll get at least 20% of the suggested price even if that is more than 70% of the actual retail price. To make an extreme case: Say you suggested $5, but Amazon decides to give it for free (in order to draw people to the shop), then you'll still get $1 per copy.
Even better: I say my app is worth $2 and Amazon agrees. They keep 80%.
No. They keep 30%.
I don't like that manipulation room. Make the developers cut 70% regardless of how the price was set.
You don't like the option to get more than 70% if Amazon sells for much less than you suggested?
Unlike Apple Store for iAnything, Amazon's app store is not the only place where you can trade Android apps. If you don't like Amazon's conditions, you can simply use another one, or even open your own.
Simply suggest as price five time the amount you'd like to get as your share. Probably it will be too high, so Amazon will lower the price. However, since you are entitled to 20% of your suggested price, you'll get what you wanted.
In the German Wikipedia, someone added yet another name [link target in German] to the long list of names of Karl-Theodor zu Guttenberg, a German politician. Afterwards, some newspapers copied the changed name from Wikipedia (without giving the source). Then someone at Wikipedia reverted that change, only to get the revert reverted again, citing one of those newspapers as source.
They were averaging over many measurement values. Maybe you read up about error propagation again and consider what happens for an average of about 1000 values.
Maybe the whole strategy of explaining was wrong. Instead of telling the scientific facts, they should have claimed: "God doesn't want us to burn so much fossil fuel. If we don't stop, He will send us a new flood." Make up an "explanation" from the bible. Find a charismatic person to tell that, and you might get enough followers to make a difference.:-)
This isn't spin, it's established science. So seeing fear, anti-government sentiment, and a parroting of the Glenn Beckesque rhetoric that's unfortunately a large part of the news here in the US right now doesn't surprise me one bit.
Conclusion: In the age of three years, we were all conservatives. So where's the conservative campaign to allow three-year olds to vote?:-)
In case you didn't notice: The subject here isn't whether the uploader or downloader does copyright infringment (of course he does). It's about the responsibility of the hoster. And it's obvious that a hoster cannot be required for each file someone uploads to search for another file which, if combined with it, would happen to give a copyrighted work he's not authorized to distribute.
What's the problem that your solution is intended to address?
If you had read the complete thread (or ad least the last few parents) you would have seen that someone claimed that distributing partial files was forbidden because you could still get something watchable/listenable from such partial file. Therefore I thought of a method which would make the single partial file completely useless (splitting by bits of the byte, instead of splitting the file in several chunks). If that claim is true, then distributing any of the files separately would not violate copyright (of course a person uploading or downloading all 8 files would violate the copyright in any case).
In any case, nobody can reasonably be expected to check if for a given file there exist seven other files somewhere on the net which together with that file might give a copyrighted work for which they don't have the license to distribute, therefore anyone hosting it should be secure (unless it's a service intended specifically for such partial files).
What if you split the file in 8 files, the first one containing the highest significant bits of all bytes, up to the 8th one containing the lowest significant bits from all bytes? I doubt that you could do much with any single file in that case.
What about hosting random data? I guess no one can sue you for that. So if someone encrypts a file with a one-time pad, stores the encrypted file at provider A and the one-time pad at provider B, both individually only get random data. You have to XOR both to get the original file. And it's not even possible to tell which one is the encrypted file and which one the pad!
Or would they demand that random (or seemingly random) data may not be stored? But then, what if someone wants to share his newly generated high-quality random numbers, for use in Monte Carlo simulations?
Both wildly successfull, huh ? I guess you see PNG used these days but how long did it take to become moderately popular and today does your camera save PNG's or still those nasty encumbered JPG's ?
Actually, contrary to what the parent poster claimed, PNG was only designed to replace GIF. It doesn't have lossy compression like JPEG. Which also explains why your camera saves JPEG. You wouldn't get as many PNGs (or GIFs, for that matter) on the chip.
This is off-topic from this particular thread of the discussion, but I also remembered the other objection to C++ that's developed: language interop (using C++ stuff in other languages) is usually nigh impossible. From what I can tell the best way to do language interop to C++ is to write a C wrapper for the C++ stuff, then interface to the C wrapper.
That's true of pretty much any language out there. How do you interface Ruby to Python? Perl to Java? Haskell to Eiffel? Fortran to Prolog?
Some implementations (e.g. g++) give a warning if you use NULL in a non-pointer context. This is not possible with the literal 0 (because the compiler has no way to know that this specific instance of the literal 0 was meant to be a null pointer, and not a zero).
Slashdot Mirror
Wrong. The formula (at least the one given in the summary is max(0.2*(suggested price), 0.7*(real price)).
So if your suggested price is $0, but the real price (set by Amazon) is $5, then you get max(0.2*$0,0.7*$5) = $3.50 -- indeed, if you look at the formula, you never get less than 70% of the actual price, however you'll get at least 20% of the suggested price even if that is more than 70% of the actual retail price. To make an extreme case: Say you suggested $5, but Amazon decides to give it for free (in order to draw people to the shop), then you'll still get $1 per copy.
No. They keep 30%.
You don't like the option to get more than 70% if Amazon sells for much less than you suggested?
Unlike Apple Store for iAnything, Amazon's app store is not the only place where you can trade Android apps. If you don't like Amazon's conditions, you can simply use another one, or even open your own.
The Kindle has the revoke feature already built-in. If it's technically possible, I'm sure they'll build it into their Android store as well.
Simply suggest as price five time the amount you'd like to get as your share. Probably it will be too high, so Amazon will lower the price. However, since you are entitled to 20% of your suggested price, you'll get what you wanted.
In the German Wikipedia, someone added yet another name [link target in German] to the long list of names of Karl-Theodor zu Guttenberg, a German politician. Afterwards, some newspapers copied the changed name from Wikipedia (without giving the source). Then someone at Wikipedia reverted that change, only to get the revert reverted again, citing one of those newspapers as source.
Actually his sentence should have been longer by an "a" and a space.
Citation needed.
They were averaging over many measurement values. Maybe you read up about error propagation again and consider what happens for an average of about 1000 values.
Maybe the whole strategy of explaining was wrong. Instead of telling the scientific facts, they should have claimed: "God doesn't want us to burn so much fossil fuel. If we don't stop, He will send us a new flood." Make up an "explanation" from the bible. Find a charismatic person to tell that, and you might get enough followers to make a difference. :-)
The fact that we are humans. Dolphins would care more about dolphins, of course.
And millions of species have died in the past. Often due to climate change.
Conservatives, scientifically, are more scared of loud noises and scary pictures, were described as being frightened and easily offended as three year olds, and have a larger 'fear' center and smaller 'anticipation and decision-making' center
This isn't spin, it's established science. So seeing fear, anti-government sentiment, and a parroting of the Glenn Beckesque rhetoric that's unfortunately a large part of the news here in the US right now doesn't surprise me one bit.
Conclusion: In the age of three years, we were all conservatives. :-)
So where's the conservative campaign to allow three-year olds to vote?
In case you didn't notice: The subject here isn't whether the uploader or downloader does copyright infringment (of course he does). It's about the responsibility of the hoster. And it's obvious that a hoster cannot be required for each file someone uploads to search for another file which, if combined with it, would happen to give a copyrighted work he's not authorized to distribute.
If you had read the complete thread (or ad least the last few parents) you would have seen that someone claimed that distributing partial files was forbidden because you could still get something watchable/listenable from such partial file. Therefore I thought of a method which would make the single partial file completely useless (splitting by bits of the byte, instead of splitting the file in several chunks). If that claim is true, then distributing any of the files separately would not violate copyright (of course a person uploading or downloading all 8 files would violate the copyright in any case).
In any case, nobody can reasonably be expected to check if for a given file there exist seven other files somewhere on the net which together with that file might give a copyrighted work for which they don't have the license to distribute, therefore anyone hosting it should be secure (unless it's a service intended specifically for such partial files).
What if you split the file in 8 files, the first one containing the highest significant bits of all bytes, up to the 8th one containing the lowest significant bits from all bytes? I doubt that you could do much with any single file in that case.
What about hosting random data? I guess no one can sue you for that. So if someone encrypts a file with a one-time pad, stores the encrypted file at provider A and the one-time pad at provider B, both individually only get random data. You have to XOR both to get the original file. And it's not even possible to tell which one is the encrypted file and which one the pad!
Or would they demand that random (or seemingly random) data may not be stored? But then, what if someone wants to share his newly generated high-quality random numbers, for use in Monte Carlo simulations?
Given that Apple holds some h.264 patents, I'd not be surprised if at some time the fine print said you may not use the codec to encode porn ...
Actually, contrary to what the parent poster claimed, PNG was only designed to replace GIF. It doesn't have lossy compression like JPEG. Which also explains why your camera saves JPEG. You wouldn't get as many PNGs (or GIFs, for that matter) on the chip.
Adobe's Flash player isn't Free in the sense of Free Software. Flash just doesn't cost the end-user anything.
If they included it in their browser, they'd either violate the patent licensing conditions, or the code licensing conditions.
It would be indeed very strange if in 2020 audio is encoded with a video codec :-)
That's true of pretty much any language out there. How do you interface Ruby to Python? Perl to Java? Haskell to Eiffel? Fortran to Prolog?
No, the world was hacked together in Perl.
Some implementations (e.g. g++) give a warning if you use NULL in a non-pointer context. This is not possible with the literal 0 (because the compiler has no way to know that this specific instance of the literal 0 was meant to be a null pointer, and not a zero).
A conforming implementation can align the fields in the struct however it wants, though.
That's true in C as well.
What is sigma?