As hardware gets cheaper and more powerful and becomes
a commodity, Apple is likely to have an increasingly difficult time selling its own line of expensive machines.
With the Mac OS now a layer on top of Unix, I wouldn't be surprised if Apple eventually gives up reserving its software for its own hardware and begins to sell Mac OS as a GUI and software bundle on top of Linux, essentially a commercial counterpart to Gnome or KDE.
btw: who told you that clerical employees don't have aspirations?
Fair enough, up to a point. I really meant to say two separate things. One was that programmers often like programming, so they'll still do it, either purely as a hobby or by starting their own business, even if they no longer need to support themselves. To my knowledge there aren't an awful lot of people who do clerical work because that's what they like to do.
Secondly, lots of people, including programmers and clerical workers, have other interests and/or financial aspirations, and freed of the need to support themselves will occupy have no trouble occupying themselves.
I'm surprised at this. I'd understand it if we were talking about people who weren't particularly interested in what they were doing and just did it to earn a living, say clerical workers and management types, but I would have expected programmers to have their own interests to which they would be happy to be able to devote their time once freed of the need to support themselves. Most programmers I know have a project they'd like to work on, a business they'd like to start, or strong non-programming interests, rock climbing, music, what have you.
I have never entered into a contract with Microsoft.
Indeed, the last few machines I have bought had MS Windows pre-installed, so I never even had to click on "I accept" to install it. Under the doctrine of First Sale, in the absence of a contract I can do what I please with the goods that I purchase. Can someone explain to me how Microsoft's wishes could possibly be binding on me?
For me this is a purely hypothetical question since I have no interest in running MS Windows,
but I am perplexed by the idea that Microsoft's
EULA's can be binding on people who either never saw them until after they purchased the software or on people who have never even looked at the EULA. Haven't the courts ruled that such "shrinkwrap licenses" are invalid?
Of course the abstract doesn't matter. That's why I said to go to the claims. The claims in this patent are typical in that they start out with a very broad characterization of the system and then narrow down to what the invention really consists of. These are what are called "dependent claims", which you can tell by the references back to previous claims.
They are not completely independent monopolies. The key thing is that the early, very broad claims are almost surely invalid, since as the original poster pointed out, the prior art goes back to the 1950s. That leaves only the much more specific subsequent claims as likely having any validity. The upshot is that the only people who really need to worry about this patent are people implementing memory management systems for parallelizing supercomputers that avoid swapping.
If I understand it correctly, this patent is much more specific than parent suggests. The abstract and initial claims make it sound like the standard general memory allocation system, but if you read further it turns out to be intended for parallelized FORTRAN scientific computing. The purpose is to ensure that all of the necessary data is actually in memory and prevent swapping.
Speech recognition is not really a solved problem. For some applications it works adequately, but if you take a look at the error rates for the Sphinx system to which the post links, you'll see that the Word Error Rate for large vocabulary is over 18%. Even for 5,000 words it is 7%. For many applications that is unacceptable.
A second factor is that these statistical speech recognition systems require extensive data for their language model. Building such a system requires recording real speech, segmenting it and creating a set of examples from which to compute the probabilities, which requires some knowledge of acoustic phonetics, and doing the computation for the model.
This is time-consuming.
Speech recognition technology isn't a dark secret, but it isn't trivial to create a system with good performance either.
Its pretty obvious that Microsoft wants to preserve their right to sue, but I'm surprised that the deal with sun provides an exception. If the two agree that there is a real prospect of Microsoft winning a
suit over OpenOffice.org, I would think that Sun would want to be protected and would have held out until Microsoft agreed. And I would think that Microsoft would have agreed, if they really wanted the deal with Sun, perhaps with the likely damages subtracted from Microsoft's payment to Sun. So what this suggests to me is that Microsoft's legal strategy involves a suit in which for some reason it would be difficult to avoid naming Sun as a defendant but in which Sun does not anticipate actually have to pay any significant damages or litigation expenses. I'm not sure how this situation would arise - maybe a situation in which MS has to name everybody initially but would then drop Sun as a defendant after discovery?
No, I was entirely right. The point I made was that there were big differences in the spread of the use of an innovation in software today and in areas such as mechanics in the past. These differences lie in the costs of implementation and in the rapidity of communication. In both cases, I was talking about what happens after a new idea is discovered. Whether or not the average programmer is able to make such discoveries is irrelevant to this point.
In any case, although some innovations are unlikely to be made other than by computer scientists, many potentially patentable innovations are made by ordinary programmers.
There isn't a clearcut division between computer scientists who discover things and programmers who implement them.
While RMS isn't very explicit about it in the Newsforge piece, one distinction between software patents and mechanical patents to which he alludes is that, arguably, a piece of software of any complexity is likely to involve many more potentially patentable components than a comparable mechanical device. To the extent this is true, it means that it is much more difficult to know when one is infringing a patent when writing software and that it would be much more difficult to set up a system for paying royalties.
It's true that patents don't seem to have prevented the Industrial Revolution, but there may be some critical differences. One is that, it seems to me, patents didn't come to be widely used until a great many fundamentals were already in the public domain. That meant that everybody had a large base of ideas that they were free to work with. Where very basic ideas were patented,
those patents did indeed pose a danger to progress. An example is the AT&T patent on the transistor, which the US government forced AT&T effectively to give up precisely because it was such a basic thing that it would have given AT&T a stranglehold on the semiconductor industry.
The other factor is that for much of the Industrial Revolution there were generally fairly large costs and/or specialized skills needed to implement a new idea, and the means of communication were relatively slow. As a result, the duration of a patent was relatively short in comparison to the time needed for ideas to diffuse. In contrast, implementing a new idea in software costs very little and requires no skills beyond those of the average programmer, and communication is very fast. As a result, people can adopt a new idea very quickly. The time for ideas to diffuse is small in relationship to the duration of patent, so patents become a bottleneck.
If this latter idea is correct, it means that the problems with software patents should arise in other areas in which costs of adoption are low and communication rapid. I wonder if genetic technology is not coming to be similar to software in this respect.
According to the article, the ring of which he was allegedly a member made use of machines at MIT. If true, although he didn't physically set foot in the US, he did indeed commit crimes in the US. Moreover, copyright is protected in Australia and most other countries and by international agreements. This doesn't seem to be a case of unreasonably applying local laws to someone elsewhere who doesn't know about them or who has no reason to believe that they are relevant to him.
There are some kinds of net activity that
present real jurisdictional problems, e.g. kinds of speech (such as insulting Islam) that are legal in some places but not in others, where an activity that is legal in one place spreads to a place where it is illegal by the normal operation of the internet. As far as I can see, this case doesn't fall into that category. If I sit at my terminal in the US and break into a computer in Australia and do mischief there, I know perfectly well that what I am doing is wrong and I have made an explicit decision to do it. It didn't just happen in the course of the normal operation of the net. Why shouldn't I be subject to prosecution in Australia?
Why is it unfortunate that he will probably be convicted? If he did what he is accused of doing, why shouldn't he be convicted? Unauthorized distribution of copyrighted music and software is indeed illegal, and its hard to believe he didn't know it. While many of us, including myself, favor free software, that doesn't make it right to violate current law, on which many people's business models and livelihoods are based. Indeed, licenses such as the GPL are based on the very same copyright law. As far as I can see, this guy and people like him are common criminals, not visionaries, rebels,or heroes.
I'm stunned by the idea that you can't bring your laptop on campus at all. Is there really such a rule? What justification do they give for it? I'm hard pressed to believe that a public institution could get away with such a thing. I've heard of banning guns, but laptops?
It seems to me that, like most other advertising, this is a waste of time and money. Persuading people to buy more than they need or to buy a certain product for a silly reason is not a social good. I wish business people would devote more of their effort to providing useful information about their products. It is still often quite difficult to find out what you really want to know about a product or to compare products effectively. Indeed, in some ways the situation is getting worse. Sony Stores, for instance, no longer have spec sheets. The manager of one store told me that only the engineering side of the company has them.
My impression is that the Wikipedia is pretty accurate in areas that attract people with real expertise. Even if some contributors have a bias or are ignorant or mistaken on certain points, after a while the article gets to be pretty good through collaborative editing. So it tends to be good on subjects that techies find interesting and are knowledgable about. The problematic areas are ones in which the contributors have an interest but lack real expertise. The collaborative editing process doesn't work very well here because there is no one involved who actually knows the subject, or the real experts are a small minority among the contributors and are not able to have much influence. Topics that are particularly likely to be problematic are those about which some geeks are enthusiastic but not truly knowledgable.
In my own area of linguistics, for example, I find that articles on formal topics, e.g. "context-free grammar", are generally good, while articles on historical linguistics are often pretty bad. This reflects the fact that techies tend to have real knowledge in areas related to formal linguistics, e.g. mathematics and computer science, while historical linguistics is a subject that lots of people find interesting but few really know much about.
Terpstra's comparison of FLOSS to the Enlightenment and Reformation and of the negative reaction on the part of parts of the software and media industries to the reaction of the Church is interesting and I think well taken. We should remember that some countries in effect have still to undergo the Englightenment, that even in countries that did, many people remain who hold irrational and antiscientific views, and that in some countries the Counter-reformation was successful. The FLOSS movement is important for freedom and for technological progress, but precisely because it is important, its opponents will work hard to suppress it, and they may succeed.
The trouble is, Pat Leahy is a terrific Senator, and it is very likely that his opponent will take the same position on this issue. A better approach would be to try to get Leahy to change his position.
COPS is in pretty bad taste, but most of it probably requires no permission because it takes place in public. You can film anything you want in a public place.
Actually, it is different from having the press present when celebrities are booked. According to the article, the cameras show such things as strip searches and the women's restroom. That's a gross invasion of privacy.
This might be a valid point if we were talking about convicts, but if you read the article, it makes it clear that detainees, that is, people who have been arrested but not convicted of anything, are being filmed. Since they haven't been convicted of anything, it is wrong to punish them or humiliate them. They're innocent until proven guilty.
of course all broadcasters already keep recordings of all their output
I always assumed this until I got to know someone who worked for a radio station (CBC radio in Prince Rupert, British Columbia). To my surprise I learned that they do not tape their broadcasts. Maybe other countries are different.
It's true that the fact that the movie industry is immensely profitable doesn't justify improper copying and downloading, but for me at least that isn't the point. The point is that the MPAA and RIAA and their ilk are asking for extraordinary measures on the grounds that without them their industries are in trouble. They want to eliminate fair use, eliminate time-shifting, outlaw technology could be used to infringe their copyrights, and make it impossible for anyone outside of a small number of companies to create software and hardware. They also want to bypass important legal procedures and protections of privacy, such as the need to get a court to issue a subpoena.
These are extraordinary demands, which should be met, if at all, only for a very good reason. The argument that they make is that without these measures they will be unable to stay in business or at least unable to produce the same quality and quantity of material. The fact that they are actually making money hand over fist shows that
this argument is false. Even if they are losing a lot of sales due to illegal copying, the industry isn't in danger and there is no justification for meeting their demands. They're crying wolf.
The privileges given to an industry always have to belanced against the public good. Consider the parallel case of old-fashioned printed books.
Publishers no doubt lose sales because people can
obtain books from libraries and read them there or borrow them. In spite of this, we wouldn't, I hope, give in to a demand by the publishing industry that libraries be outlawed so that they can maximize their profits. The public good of having libraries outweighs the desire of publishers to be more profitable. Now, if it were
the case that publishers couldn't stay in business without some additional revenue, we might change our position. We might, for example, agree to a system that paid publishers each time a library patron checked out a book or even (here's a use
for RFID, I guess) every time a patrol used a book
in the library. But in the absence of an economic
crisis for the publishing industry we probably wouldn't do this. Our attitude is that publishers make a reasonable amount of money the way things are and that it is just tough that libraries cut into their potential income.
So what the profits for the movie industry tell me is that nothing needs to be changed. They're welcome to enforce their copyrights by existing means. If they can track which Academy members leak films and go after them, fine by me. But since they're making a reasonable profit, there's no reason to give in to their extraordinary demands. It ain't broke, so we don't need to fix it.
Yes, that's why I said "people concerned about audio quality". For those who can tell the difference, or think they can, there's no reason to mess with lossy compression. The great majority of people are in fact perfectly happy with MP3.
Slashdot Mirror
As hardware gets cheaper and more powerful and becomes a commodity, Apple is likely to have an increasingly difficult time selling its own line of expensive machines. With the Mac OS now a layer on top of Unix, I wouldn't be surprised if Apple eventually gives up reserving its software for its own hardware and begins to sell Mac OS as a GUI and software bundle on top of Linux, essentially a commercial counterpart to Gnome or KDE.
Fair enough, up to a point. I really meant to say two separate things. One was that programmers often like programming, so they'll still do it, either purely as a hobby or by starting their own business, even if they no longer need to support themselves. To my knowledge there aren't an awful lot of people who do clerical work because that's what they like to do.
Secondly, lots of people, including programmers and clerical workers, have other interests and/or financial aspirations, and freed of the need to support themselves will occupy have no trouble occupying themselves.
I'm surprised at this. I'd understand it if we were talking about people who weren't particularly interested in what they were doing and just did it to earn a living, say clerical workers and management types, but I would have expected programmers to have their own interests to which they would be happy to be able to devote their time once freed of the need to support themselves. Most programmers I know have a project they'd like to work on, a business they'd like to start, or strong non-programming interests, rock climbing, music, what have you.
I have never entered into a contract with Microsoft. Indeed, the last few machines I have bought had MS Windows pre-installed, so I never even had to click on "I accept" to install it. Under the doctrine of First Sale, in the absence of a contract I can do what I please with the goods that I purchase. Can someone explain to me how Microsoft's wishes could possibly be binding on me?
For me this is a purely hypothetical question since I have no interest in running MS Windows, but I am perplexed by the idea that Microsoft's EULA's can be binding on people who either never saw them until after they purchased the software or on people who have never even looked at the EULA. Haven't the courts ruled that such "shrinkwrap licenses" are invalid?
Of course the abstract doesn't matter. That's why I said to go to the claims. The claims in this patent are typical in that they start out with a very broad characterization of the system and then narrow down to what the invention really consists of. These are what are called "dependent claims", which you can tell by the references back to previous claims. They are not completely independent monopolies. The key thing is that the early, very broad claims are almost surely invalid, since as the original poster pointed out, the prior art goes back to the 1950s. That leaves only the much more specific subsequent claims as likely having any validity. The upshot is that the only people who really need to worry about this patent are people implementing memory management systems for parallelizing supercomputers that avoid swapping.
If I understand it correctly, this patent is much more specific than parent suggests. The abstract and initial claims make it sound like the standard general memory allocation system, but if you read further it turns out to be intended for parallelized FORTRAN scientific computing. The purpose is to ensure that all of the necessary data is actually in memory and prevent swapping.
Speech recognition is not really a solved problem. For some applications it works adequately, but if you take a look at the error rates for the Sphinx system to which the post links, you'll see that the Word Error Rate for large vocabulary is over 18%. Even for 5,000 words it is 7%. For many applications that is unacceptable.
A second factor is that these statistical speech recognition systems require extensive data for their language model. Building such a system requires recording real speech, segmenting it and creating a set of examples from which to compute the probabilities, which requires some knowledge of acoustic phonetics, and doing the computation for the model. This is time-consuming.
Speech recognition technology isn't a dark secret, but it isn't trivial to create a system with good performance either.
For those who don't read Japanese, here are specs from the press release:
Does anybody know what the measure of contrast is? That's 450 what?Its pretty obvious that Microsoft wants to preserve their right to sue, but I'm surprised that the deal with sun provides an exception. If the two agree that there is a real prospect of Microsoft winning a suit over OpenOffice.org, I would think that Sun would want to be protected and would have held out until Microsoft agreed. And I would think that Microsoft would have agreed, if they really wanted the deal with Sun, perhaps with the likely damages subtracted from Microsoft's payment to Sun. So what this suggests to me is that Microsoft's legal strategy involves a suit in which for some reason it would be difficult to avoid naming Sun as a defendant but in which Sun does not anticipate actually have to pay any significant damages or litigation expenses. I'm not sure how this situation would arise - maybe a situation in which MS has to name everybody initially but would then drop Sun as a defendant after discovery?
No, I was entirely right. The point I made was that there were big differences in the spread of the use of an innovation in software today and in areas such as mechanics in the past. These differences lie in the costs of implementation and in the rapidity of communication. In both cases, I was talking about what happens after a new idea is discovered. Whether or not the average programmer is able to make such discoveries is irrelevant to this point.
In any case, although some innovations are unlikely to be made other than by computer scientists, many potentially patentable innovations are made by ordinary programmers. There isn't a clearcut division between computer scientists who discover things and programmers who implement them.
While RMS isn't very explicit about it in the Newsforge piece, one distinction between software patents and mechanical patents to which he alludes is that, arguably, a piece of software of any complexity is likely to involve many more potentially patentable components than a comparable mechanical device. To the extent this is true, it means that it is much more difficult to know when one is infringing a patent when writing software and that it would be much more difficult to set up a system for paying royalties.
It's true that patents don't seem to have prevented the Industrial Revolution, but there may be some critical differences. One is that, it seems to me, patents didn't come to be widely used until a great many fundamentals were already in the public domain. That meant that everybody had a large base of ideas that they were free to work with. Where very basic ideas were patented, those patents did indeed pose a danger to progress. An example is the AT&T patent on the transistor, which the US government forced AT&T effectively to give up precisely because it was such a basic thing that it would have given AT&T a stranglehold on the semiconductor industry.
The other factor is that for much of the Industrial Revolution there were generally fairly large costs and/or specialized skills needed to implement a new idea, and the means of communication were relatively slow. As a result, the duration of a patent was relatively short in comparison to the time needed for ideas to diffuse. In contrast, implementing a new idea in software costs very little and requires no skills beyond those of the average programmer, and communication is very fast. As a result, people can adopt a new idea very quickly. The time for ideas to diffuse is small in relationship to the duration of patent, so patents become a bottleneck.
If this latter idea is correct, it means that the problems with software patents should arise in other areas in which costs of adoption are low and communication rapid. I wonder if genetic technology is not coming to be similar to software in this respect.
According to the article, the ring of which he was allegedly a member made use of machines at MIT. If true, although he didn't physically set foot in the US, he did indeed commit crimes in the US. Moreover, copyright is protected in Australia and most other countries and by international agreements. This doesn't seem to be a case of unreasonably applying local laws to someone elsewhere who doesn't know about them or who has no reason to believe that they are relevant to him.
There are some kinds of net activity that present real jurisdictional problems, e.g. kinds of speech (such as insulting Islam) that are legal in some places but not in others, where an activity that is legal in one place spreads to a place where it is illegal by the normal operation of the internet. As far as I can see, this case doesn't fall into that category. If I sit at my terminal in the US and break into a computer in Australia and do mischief there, I know perfectly well that what I am doing is wrong and I have made an explicit decision to do it. It didn't just happen in the course of the normal operation of the net. Why shouldn't I be subject to prosecution in Australia?
Why is it unfortunate that he will probably be convicted? If he did what he is accused of doing, why shouldn't he be convicted? Unauthorized distribution of copyrighted music and software is indeed illegal, and its hard to believe he didn't know it. While many of us, including myself, favor free software, that doesn't make it right to violate current law, on which many people's business models and livelihoods are based. Indeed, licenses such as the GPL are based on the very same copyright law. As far as I can see, this guy and people like him are common criminals, not visionaries, rebels,or heroes.
I'm stunned by the idea that you can't bring your laptop on campus at all. Is there really such a rule? What justification do they give for it? I'm hard pressed to believe that a public institution could get away with such a thing. I've heard of banning guns, but laptops?
It seems to me that, like most other advertising, this is a waste of time and money. Persuading people to buy more than they need or to buy a certain product for a silly reason is not a social good. I wish business people would devote more of their effort to providing useful information about their products. It is still often quite difficult to find out what you really want to know about a product or to compare products effectively. Indeed, in some ways the situation is getting worse. Sony Stores, for instance, no longer have spec sheets. The manager of one store told me that only the engineering side of the company has them.
My impression is that the Wikipedia is pretty accurate in areas that attract people with real expertise. Even if some contributors have a bias or are ignorant or mistaken on certain points, after a while the article gets to be pretty good through collaborative editing. So it tends to be good on subjects that techies find interesting and are knowledgable about. The problematic areas are ones in which the contributors have an interest but lack real expertise. The collaborative editing process doesn't work very well here because there is no one involved who actually knows the subject, or the real experts are a small minority among the contributors and are not able to have much influence. Topics that are particularly likely to be problematic are those about which some geeks are enthusiastic but not truly knowledgable.
In my own area of linguistics, for example, I find that articles on formal topics, e.g. "context-free grammar", are generally good, while articles on historical linguistics are often pretty bad. This reflects the fact that techies tend to have real knowledge in areas related to formal linguistics, e.g. mathematics and computer science, while historical linguistics is a subject that lots of people find interesting but few really know much about.
Terpstra's comparison of FLOSS to the Enlightenment and Reformation and of the negative reaction on the part of parts of the software and media industries to the reaction of the Church is interesting and I think well taken. We should remember that some countries in effect have still to undergo the Englightenment, that even in countries that did, many people remain who hold irrational and antiscientific views, and that in some countries the Counter-reformation was successful. The FLOSS movement is important for freedom and for technological progress, but precisely because it is important, its opponents will work hard to suppress it, and they may succeed.
The trouble is, Pat Leahy is a terrific Senator, and it is very likely that his opponent will take the same position on this issue. A better approach would be to try to get Leahy to change his position.
COPS is in pretty bad taste, but most of it probably requires no permission because it takes place in public. You can film anything you want in a public place.
Actually, it is different from having the press present when celebrities are booked. According to the article, the cameras show such things as strip searches and the women's restroom. That's a gross invasion of privacy.
This might be a valid point if we were talking about convicts, but if you read the article, it makes it clear that detainees, that is, people who have been arrested but not convicted of anything, are being filmed. Since they haven't been convicted of anything, it is wrong to punish them or humiliate them. They're innocent until proven guilty.
According to the Nature article, the trout concerned is Oncorhynchus mykiss. The salmon is Oncorhynchus masou. Same genus, different species.
I always assumed this until I got to know someone who worked for a radio station (CBC radio in Prince Rupert, British Columbia). To my surprise I learned that they do not tape their broadcasts. Maybe other countries are different.
It's true that the fact that the movie industry is immensely profitable doesn't justify improper copying and downloading, but for me at least that isn't the point. The point is that the MPAA and RIAA and their ilk are asking for extraordinary measures on the grounds that without them their industries are in trouble. They want to eliminate fair use, eliminate time-shifting, outlaw technology could be used to infringe their copyrights, and make it impossible for anyone outside of a small number of companies to create software and hardware. They also want to bypass important legal procedures and protections of privacy, such as the need to get a court to issue a subpoena.
These are extraordinary demands, which should be met, if at all, only for a very good reason. The argument that they make is that without these measures they will be unable to stay in business or at least unable to produce the same quality and quantity of material. The fact that they are actually making money hand over fist shows that this argument is false. Even if they are losing a lot of sales due to illegal copying, the industry isn't in danger and there is no justification for meeting their demands. They're crying wolf.
The privileges given to an industry always have to belanced against the public good. Consider the parallel case of old-fashioned printed books. Publishers no doubt lose sales because people can obtain books from libraries and read them there or borrow them. In spite of this, we wouldn't, I hope, give in to a demand by the publishing industry that libraries be outlawed so that they can maximize their profits. The public good of having libraries outweighs the desire of publishers to be more profitable. Now, if it were the case that publishers couldn't stay in business without some additional revenue, we might change our position. We might, for example, agree to a system that paid publishers each time a library patron checked out a book or even (here's a use for RFID, I guess) every time a patrol used a book in the library. But in the absence of an economic crisis for the publishing industry we probably wouldn't do this. Our attitude is that publishers make a reasonable amount of money the way things are and that it is just tough that libraries cut into their potential income.
So what the profits for the movie industry tell me is that nothing needs to be changed. They're welcome to enforce their copyrights by existing means. If they can track which Academy members leak films and go after them, fine by me. But since they're making a reasonable profit, there's no reason to give in to their extraordinary demands. It ain't broke, so we don't need to fix it.
Yes, that's why I said "people concerned about audio quality". For those who can tell the difference, or think they can, there's no reason to mess with lossy compression. The great majority of people are in fact perfectly happy with MP3.