In Denmark (I expect that the company has a clausule on wanting every dispute settled in a Danish court)
If they copied the code, then they committed a crime. Criminals don't get to chose which court they get tried in. They could be sued anywhere they sell their product. In some places, like the US, mplayer developers have to dot their i's first, but they could bring suit in the US if they so chose.
Take a few classes in German and you'll get an idea of just how horrible English is.
Right, because it's much harder to have the word "book" and figure out "the books" then it is to remember what gender Buch is and then figure out how to pluralize that particular word. (Not that English plurals are entirely regular, but the vast majority of words you just add an s (or es if it ends in s or z), and the exceptions generally date back 1,500 years, when the language was supposedly "purer".)
some people are pompous and like to pretend they are correctly applying something you don't know, so they use them anyway.
Some other people like to be pompous and insult other people for having fun. This isn't about being pompous; it's about having fun with language.
And this is a way that languages change and grow. You incorrectly pluralize ox as oxes, because that's familiar to you from other words in the language. Other people pluralize box as boxen, because of its similarity to ox and oxen. People generalize rules in language; from ox, oxen, comes a more general rule, *ox, *oxen. (The * in this case is the regex *.) You apply the rule that *x, *xes. (Actually, *s, *ses phonologically.) Same phenomenon, different directions.
all the legacy code using 1-byte characters has to be revised.
But only once for Japanese, Chinese, Korean or Unicode.
Chinese text also requires a full Latin set
Multiple copies of which are tucked away in most Chinese character sets.
right-left, left-right and up-down orientations.
For the most part, you can get by with left to right. Top to bottom is needed mainly for desktop publishing applications. I have never seen right-to-left Chinese, and the majority of Chinese systems can't handle it (in Unicode systems, it comes for free with Hebrew/Arabic support), so it's not a big worry in Chinese enabling.
And UNIX has worked for people with weird (read: non-ASCII) writing systems for AGES!
Windows has always worked with non-ASCII writing systems. The latest version of Windows has usually handled more writing systems and more complex writing systems then Unix; for example, only recent versions of Gnome handle Hindi, and even they won't print Hindi. Windows has handled Hindi for a while.
There have been Chinese versions of MS Office for almost 10 years. That's a lot harder than alphabetic scripts like Hebrew.
No, it's not. Chinese is about the simplest language to enable a computer for, in theory; it's a series of constant-width characters that don't word-wrap and don't combine in any way, all going right-to-left. The fact that there's 10,000 of them in modern use just dictates the size of ints you use to store characters. Hebrew gets complex because in modern use, they mix right-to-left (Hebrew) and left-to-right characters (Latin and numbers), as well as have the voweling system that needs to combine with the base system, yet be ignored in spell-checks and in searches. It's much more complex.
There are a crazy number of hoops that need to be gone through to get unclassified info off of a classified system. They can't have people encoding stuff in pictures of Barney then walking away with it.
Step number one is, even if it looks innoculous, don't let it through. Nobody is going to let you email or floppy a picture of Barney out of a classifed system, because there's no reason to, and it might contain classified information. It doesn't matter what the stegnography filter says, it won't go.
I've actually scanned a bill before. I ran across a bill with a message written on it in Arabic and put it online for a few friends to look at. (It said "I love you, Fatima!", or words to that effect.) Of course, I cropped and false colored it (to highlight the writing) first.
The world thinks you're wackos, yes, but not for the reason you think. It's not because you don't back down -- that's an admirable trait -- but because you do things for three reasons only: Cash, Moolah and Dough.
That's bullshit. Look at Vietnam for one extreme example. Look at us putting a man on the moon. Look at how we basically gave Britain ships prior to us entering WWII. Look at us releasing control over the Philippines and Japan.
On the flip, look at English support for the installation of a dictator in Iran because the Iranian democrats wanted to see a cut of Iranian oil money. Americans are hardly the only people in the world who do things based on money.
You're splitting the hair pretty thin in order to redefine the english language in the image you want.
You chose the definition.
I could split the hair just as much and call this text you're reading lossy.
As a transcription of your voice, it is. But you didn't try to transcribe your voice; you wrote it as proper English text, which is composed of discrete entities (characters). Likewise, I'm not seeing it in the font and size you might recommend, but there's no evidence that you tried to include that; even if it is part of a master copy, there is a level that's just text that is lossless.
Just because CDs are typically mixed from many audio streams (that may include positional data, or higher fidelity) doesn't mean that the final two-channel product lost something. The final stream is simply the finished composition.
How can you take a higher fidelity audio stream and turn it into a lower fidelity stream and say you haven't lost anything? On one hand, there are soundtracks that are designed to have many channels that are lost in conversion to CD; on the other, you can designate a JPG the finished composition; I've seen newsgroups send around md5sums for JPGs to make sure everyone was getting the same high-quality originals. How are those JPGs not finished compositions?
That doesn't make Euros lossy, just a different currency format.
If you are buying and selling in Euros, then they aren't lossy. If you post your prices in Euros and buy and sell in dollars, then Euros are lossy.
MPG changes the picture and audio intentionally, to reduce the number of bits used
A camera changes the picture intentionally, to reduce the number of bits (or in an analog camera, the amount of data), by compressing the spectrum into red, green and blue, or some other set of hues, usually three in color.
Scan a picture and save it as a GIF or BMP (lossless), and then as a series of JPGs (lossy) at increasing levels of compression
I've done that. I've also saved a picture as a B&W PNG and as a greyscale JPGs and found that the "lossy" format better preserved the image.
In fact, as a general rule, if I scan a color picture, it will look better saved as a JPG then as a GIF, which only supports 256 colors.
Yes, there's a difference between "lossy" and "lossless" compression formats as compression formats. But that doesn't mean that a file saved in a "lossless" format will be higher quality then a file saved in a "lossy" format. Just because a compression format is "lossless" doesn't mean that it's not low quality and that stuff converted into it won't be degraded.
BTW, I believe you are alone including analog formats in your definition of 'lossy'.
so analog formats aren't lossy, how could your statement that
You seem to have a definition of 'lossy' that is heavily biased against digital formats
be true?
In any case, Foldoc's definition is biased towards digital techniques, as algorithms imply computers. Note also that it never talks about lossy or lossless formats, it talks about lossy and lossless compression techniques, a whole different matter. As a "compression" technique on most audio formats inside the computer, wavs are lossless (though not if you have multichannel audio, which WAV compresses by deleting or merging channels).
It now sounds like you're trying to say that any method of storing information is 'lossy'. What's an audio or video format that isn't lossy to you?
There is none, unless your information is fundamentally digital, like text, to begin with.
Are DVDs lossy? If they "actually reduce[] the amount of information in the data", then so do stereo (or mono) CDs, because they compress multiple audio streams into two (or one).
The market should be allowed to decide which model should prevail (or if both should coexist), without being tainted by some sort of acquired "morality".
So your morality is that all shall kneel before the all-mighty dollar. Don't try and pass it off as neutral.
Being a human question, morality comes into play, and cooperation is a value that is pounded into us early.
I believe future historians will judge RMS as having done about as much harm as good.
On one hand, he wrote software (GCC) that enabled free software to exist. On the other, he has made some speeches to a generally intellegent and thoughtful bunch of people telling them that sharing is good. I hardly think that he's done a whole lot of harm by doing that.
You seem to have a definition of 'lossy' that is heavily biased against digital formats
I said that people had been using lossy formats for 90 years; I think it fairly obvious that I include analog in my definition of lossy. That makes most of your argument strawman.
You can convert between non-lossy formats, and get back identical data each time.
Only if you have one true model of the data space. You can't convert between YUV and RGB colorspaces and get identical data, but both are generally considered lossless. Likewise, if you renormalized your view of the sound space to make MP3s the one true sound model, WAVs would just be another odd lossy sound model.
its resolution does make sense in the overall scheme of things, since it matches that of its usual destination, a TV.
That could easily be said of MP3s, though; most people play their music through car stereos and basic computer speakers, not anything that would make the difference between MP3s and CD quality stand out. I'm not sure most people could tell the difference with the best quality system.
DVDs can encode movies using progressively-scanned frames
Having watched DVDs on my computer, which drops frames rather heavily, I can pick out which DVDs in my collection are progressively-scanned, now that you point it out. So why is so much stuff interlaced? Do they really throw out the film for TV shows and TV movies and keep only the interlaced stuff?
It requires an above-average amount of foolish short-sightedness for a person to be willing to buy in a lossy format,
There's no such thing as a non-lossy digital encoding of analog data. You have to start throwing away data that comes below a certain threshold. CDs are just a lossy format which isn't well tuned to what humans actually hear, so there's a lot of room to throw away data.
In any case, people bought seriously lossy formats for the first 90 years of music, and they still buy a lossy format for video. (Not only does DVDs use MPEG4 and sample at a resolution way below film, it stores data at interlaced TV rates instead of what was actually filmed.)
When cops are ticketing bimbos for flashing thier tits at a bar, they aren't out ticketing drunk drivers or doing something USEFUL with the public's money.
The public passed the law, so obviously they think it's useful. How long did it take to arrest her? They sat down and wrote out a police report. They didn't have to stake out a strech of highway for thirty-four hours and then have twelve cop cars chase him down. I'd say that's proportional to the crime.
Are you fucking kidding me? Their job? Maybe their job should be looking out for people who are actually doing some damage.
Maybe it should. Perhaps you should run for office to change the job. But right now, their job is to enforce the law as it is.
The society of Lincoln, Nebraska, like every society, has certain rules of decorum. Do whatever you want in the privacy of your own home, but when you are dealing with other people you have to deal with their collective social rules.
Honestly, as far as the computer is concerned, what difference if characters being displayed are displayed to be read left to right or right to left?
Because it's not that simple. You need to mix Hebrew and English, and even if you don't, modern Hebrew uses European digits, which are treated as left to right in most systems. The Unicode standard spends 14 pages on this, with a fairly complex algorithm that must be used to display right to left text.
A new implementation could replace that code with simple, generic implementations
So we should take fast, working code and completely rewrite it to be slow? At least most of the time when people suggest a complete rewrite, it's too write faster code.
most of whom don't even bother to take the trouble to learn how to say "please" and "thank you" in the language of the nation they're in at the moment.
And other nations would never do that. Oh no, all the English in India when it was a colony all spoke Hindi and whatever languages were appropriate for the area they lived in. And all the French in Vietnam spoke fluent Vietnamese.
Europeans often are insular and parochial. Look how much of fight that goes on for the Basque to just use their own language. Look at the French who pass innumberal laws to stop normal language evolution. I was once flamed by a Luxemberger because Americans don't know where Luxemberg, a country smaller then every American state, probably smaller then every Chinese state, smaller then thousands of cities. But it is vitally important that every one know exactly where it is.
This is pure cultural arrogance.
Or it's simple cost-effect balance. It's clear he doesn't need to learn Spanish, so he doesn't. This is a rule that has been used in Europe for centuries, that only the languages of the noble important countries need to be learned, not that of the peasants or the heathens. Russian rulers sometimes didn't know Russian
Of course, if it was chickens, we wouldn't care so much, would we? After all, we kill over 2 billion chickens yearly for food, and somehow that doesn't seem to concern too many people.
No, because we raise those chickens for food. The birds being killed are wild birds, which means that this is affecting the environment, and it is reasonable to be concerned about how we're affecting the environment (even if this case is moot.)
it says KFC kills 750 million birds per year. Add that to McDonalds and you have over a billion/year. Puts some perspective on the turbines...
And if KFC and McDonalds didn't serve bird, those birds wouldn't have come into existance at all. Those companies have no effect on the wild bird population.
Slashdot Mirror
In Denmark (I expect that the company has a clausule on wanting every dispute settled in a Danish court)
If they copied the code, then they committed a crime. Criminals don't get to chose which court they get tried in. They could be sued anywhere they sell their product. In some places, like the US, mplayer developers have to dot their i's first, but they could bring suit in the US if they so chose.
Take a few classes in German and you'll get an idea of just how horrible English is.
Right, because it's much harder to have the word "book" and figure out "the books" then it is to remember what gender Buch is and then figure out how to pluralize that particular word. (Not that English plurals are entirely regular, but the vast majority of words you just add an s (or es if it ends in s or z), and the exceptions generally date back 1,500 years, when the language was supposedly "purer".)
And child, children.
some people are pompous and like to pretend they are correctly applying something you don't know, so they use them anyway.
Some other people like to be pompous and insult other people for having fun. This isn't about being pompous; it's about having fun with language.
And this is a way that languages change and grow. You incorrectly pluralize ox as oxes, because that's familiar to you from other words in the language. Other people pluralize box as boxen, because of its similarity to ox and oxen. People generalize rules in language; from ox, oxen, comes a more general rule, *ox, *oxen. (The * in this case is the regex *.) You apply the rule that *x, *xes. (Actually, *s, *ses phonologically.) Same phenomenon, different directions.
all the legacy code using 1-byte characters has to be revised.
But only once for Japanese, Chinese, Korean or Unicode.
Chinese text also requires a full Latin set
Multiple copies of which are tucked away in most Chinese character sets.
right-left, left-right and up-down orientations.
For the most part, you can get by with left to right. Top to bottom is needed mainly for desktop publishing applications. I have never seen right-to-left Chinese, and the majority of Chinese systems can't handle it (in Unicode systems, it comes for free with Hebrew/Arabic support), so it's not a big worry in Chinese enabling.
You can convert between BMP, TGA or PNG (and other lossless formats), and not lose a single pixel of detail.
Provided you start with 24 bit color. If you start with color that is 16-bit per channel, TGA loses information.
by your definition of 'lossy', this text is a lossy approximation of my thoughts before I even touch the keyboard.
As a representation of thoughts, text is indeed a lossy format. However, text is not usually thought of in those terms.
Name a video or audio format that you consider lossless
There is no such format. According to Heisenberg, there can be no such format. There are only successively better approximations.
And UNIX has worked for people with weird (read: non-ASCII) writing systems for AGES!
Windows has always worked with non-ASCII writing systems. The latest version of Windows has usually handled more writing systems and more complex writing systems then Unix; for example, only recent versions of Gnome handle Hindi, and even they won't print Hindi. Windows has handled Hindi for a while.
There have been Chinese versions of MS Office for almost 10 years. That's a lot harder than alphabetic scripts like Hebrew.
No, it's not. Chinese is about the simplest language to enable a computer for, in theory; it's a series of constant-width characters that don't word-wrap and don't combine in any way, all going right-to-left. The fact that there's 10,000 of them in modern use just dictates the size of ints you use to store characters. Hebrew gets complex because in modern use, they mix right-to-left (Hebrew) and left-to-right characters (Latin and numbers), as well as have the voweling system that needs to combine with the base system, yet be ignored in spell-checks and in searches. It's much more complex.
There are a crazy number of hoops that need to be gone through to get unclassified info off of a classified system. They can't have people encoding stuff in pictures of Barney then walking away with it.
Step number one is, even if it looks innoculous, don't let it through. Nobody is going to let you email or floppy a picture of Barney out of a classifed system, because there's no reason to, and it might contain classified information. It doesn't matter what the stegnography filter says, it won't go.
I've actually scanned a bill before. I ran across a bill with a message written on it in Arabic and put it online for a few friends to look at. (It said "I love you, Fatima!", or words to that effect.) Of course, I cropped and false colored it (to highlight the writing) first.
The world thinks you're wackos, yes, but not for the reason you think. It's not because you don't back down -- that's an admirable trait -- but because you do things for three reasons only: Cash, Moolah and Dough.
That's bullshit. Look at Vietnam for one extreme example. Look at us putting a man on the moon. Look at how we basically gave Britain ships prior to us entering WWII. Look at us releasing control over the Philippines and Japan.
On the flip, look at English support for the installation of a dictator in Iran because the Iranian democrats wanted to see a cut of Iranian oil money. Americans are hardly the only people in the world who do things based on money.
You're splitting the hair pretty thin in order to redefine the english language in the image you want.
You chose the definition.
I could split the hair just as much and call this text you're reading lossy.
As a transcription of your voice, it is. But you didn't try to transcribe your voice; you wrote it as proper English text, which is composed of discrete entities (characters). Likewise, I'm not seeing it in the font and size you might recommend, but there's no evidence that you tried to include that; even if it is part of a master copy, there is a level that's just text that is lossless.
Just because CDs are typically mixed from many audio streams (that may include positional data, or higher fidelity) doesn't mean that the final two-channel product lost something. The final stream is simply the finished composition.
How can you take a higher fidelity audio stream and turn it into a lower fidelity stream and say you haven't lost anything? On one hand, there are soundtracks that are designed to have many channels that are lost in conversion to CD; on the other, you can designate a JPG the finished composition; I've seen newsgroups send around md5sums for JPGs to make sure everyone was getting the same high-quality originals. How are those JPGs not finished compositions?
That doesn't make Euros lossy, just a different currency format.
If you are buying and selling in Euros, then they aren't lossy. If you post your prices in Euros and buy and sell in dollars, then Euros are lossy.
MPG changes the picture and audio intentionally, to reduce the number of bits used
A camera changes the picture intentionally, to reduce the number of bits (or in an analog camera, the amount of data), by compressing the spectrum into red, green and blue, or some other set of hues, usually three in color.
Scan a picture and save it as a GIF or BMP (lossless), and then as a series of JPGs (lossy) at increasing levels of compression
I've done that. I've also saved a picture as a B&W PNG and as a greyscale JPGs and found that the "lossy" format better preserved the image.
In fact, as a general rule, if I scan a color picture, it will look better saved as a JPG then as a GIF, which only supports 256 colors.
Yes, there's a difference between "lossy" and "lossless" compression formats as compression formats. But that doesn't mean that a file saved in a "lossless" format will be higher quality then a file saved in a "lossy" format. Just because a compression format is "lossless" doesn't mean that it's not low quality and that stuff converted into it won't be degraded.
BTW, I believe you are alone including analog formats in your definition of 'lossy'.
so analog formats aren't lossy, how could your statement that
You seem to have a definition of 'lossy' that is heavily biased against digital formats
be true?
In any case, Foldoc's definition is biased towards
digital techniques, as algorithms imply computers.
Note also that it never talks about lossy or lossless formats, it talks about lossy and lossless compression techniques, a whole different matter. As a "compression" technique on most audio formats inside the computer, wavs are lossless (though not if you have multichannel audio, which WAV compresses by deleting or merging channels).
It now sounds like you're trying to say that any method of storing information is 'lossy'. What's an audio or video format that isn't lossy to you?
There is none, unless your information is fundamentally digital, like text, to begin with.
Are DVDs lossy? If they "actually reduce[] the amount of information in the data", then so do stereo (or mono) CDs, because they compress multiple audio streams into two (or one).
The market should be allowed to decide which model should prevail (or if both should coexist), without being tainted by some sort of acquired "morality".
So your morality is that all shall kneel before the all-mighty dollar. Don't try and pass it off as neutral.
Being a human question, morality comes into play, and cooperation is a value that is pounded into us early.
I believe future historians will judge RMS as having done about as much harm as good.
On one hand, he wrote software (GCC) that enabled free software to exist. On the other, he has made some speeches to a generally intellegent and thoughtful bunch of people telling them that sharing is good. I hardly think that he's done a whole lot of harm by doing that.
You seem to have a definition of 'lossy' that is heavily biased against digital formats
I said that people had been using lossy formats for 90 years; I think it fairly obvious that I include analog in my definition of lossy. That makes most of your argument strawman.
You can convert between non-lossy formats, and get back identical data each time.
Only if you have one true model of the data space. You can't convert between YUV and RGB colorspaces and get identical data, but both are generally considered lossless. Likewise, if you renormalized your view of the sound space to make MP3s the one true sound model, WAVs would just be another odd lossy sound model.
All spam blocks, except for a couple uber-expensive systems, will make false positives.
Yes, but the ratio of false positives to false negatives is tunable.
You know enough about spam blocks to know that there are shitloads of keywords they're looking for
It has little to do with keywords. The things that catch most people is broad block lists.
Okay, so I really botched a few facts there.
its resolution does make sense in the overall scheme of things, since it matches that of its usual destination, a TV.
That could easily be said of MP3s, though; most people play their music through car stereos and basic computer speakers, not anything that would make the difference between MP3s and CD quality stand out. I'm not sure most people could tell the difference with the best quality system.
DVDs can encode movies using progressively-scanned frames
Having watched DVDs on my computer, which drops frames rather heavily, I can pick out which DVDs in my collection are progressively-scanned, now that you point it out. So why is so much stuff interlaced? Do they really throw out the film for TV shows and TV movies and keep only the interlaced stuff?
It requires an above-average amount of foolish short-sightedness for a person to be willing to buy in a lossy format,
There's no such thing as a non-lossy digital encoding of analog data. You have to start throwing away data that comes below a certain threshold. CDs are just a lossy format which isn't well tuned to what humans actually hear, so there's a lot of room to throw away data.
In any case, people bought seriously lossy formats for the first 90 years of music, and they still buy a lossy format for video. (Not only does DVDs use MPEG4 and sample at a resolution way below film, it stores data at interlaced TV rates instead of what was actually filmed.)
Especially since they're a huge ISP who has to be conservative with their spam blocking techniques.
What makes you think that? AOL tends to have a lot of false positives when blocking spam.
When cops are ticketing bimbos for flashing thier tits at a bar, they aren't out ticketing drunk drivers or doing something USEFUL with the public's money.
The public passed the law, so obviously they think it's useful. How long did it take to arrest her? They sat down and wrote out a police report. They didn't have to stake out a strech of highway for thirty-four hours and then have twelve cop cars chase him down. I'd say that's proportional to the crime.
Are you fucking kidding me? Their job? Maybe their job should be looking out for people who are actually doing some damage.
Maybe it should. Perhaps you should run for office to change the job. But right now, their job is to enforce the law as it is.
The society of Lincoln, Nebraska, like every society, has certain rules of decorum. Do whatever you want in the privacy of your own home, but when you are dealing with other people you have to deal with their collective social rules.
Honestly, as far as the computer is concerned, what difference if characters being displayed are displayed to be read left to right or right to left?
Because it's not that simple. You need to mix Hebrew and English, and even if you don't, modern Hebrew uses European digits, which are treated as left to right in most systems. The Unicode standard spends 14 pages on this, with a fairly complex algorithm that must be used to display right to left text.
A new implementation could replace that code with simple, generic implementations
So we should take fast, working code and completely rewrite it to be slow? At least most of the time when people suggest a complete rewrite, it's too write faster code.
most of whom don't even bother to take the trouble to learn how to say "please" and "thank you" in the language of the nation they're in at the moment.
And other nations would never do that. Oh no, all the English in India when it was a colony all spoke Hindi and whatever languages were appropriate for the area they lived in. And all the French in Vietnam spoke fluent Vietnamese.
Europeans often are insular and parochial. Look how much of fight that goes on for the Basque to just use their own language. Look at the French who pass innumberal laws to stop normal language evolution. I was once flamed by a Luxemberger because Americans don't know where Luxemberg, a country smaller then every American state, probably smaller then every Chinese state, smaller then thousands of cities. But it is vitally important that every one know exactly where it is.
This is pure cultural arrogance.
Or it's simple cost-effect balance. It's clear he doesn't need to learn Spanish, so he doesn't. This is a rule that has been used in Europe for centuries, that only the languages of the noble important countries need to be learned, not that of the peasants or the heathens. Russian rulers sometimes didn't know Russian
Of course, if it was chickens, we wouldn't care so much, would we? After all, we kill over 2 billion chickens yearly for food, and somehow that doesn't seem to concern too many people.
No, because we raise those chickens for food. The birds being killed are wild birds, which means that this is affecting the environment, and it is reasonable to be concerned about how we're affecting the environment (even if this case is moot.)
it says KFC kills 750 million birds per year. Add that to McDonalds and you have over a billion/year. Puts some perspective on the turbines...
And if KFC and McDonalds didn't serve bird, those birds wouldn't have come into existance at all. Those companies have no effect on the wild bird population.