I'm an O'Reilly author. They've been completely respectful about this -- no pressure at all, except to respond one way or the other.
emergent properties make this moot
on
AI in Sci-Fi
·
· Score: 1
There are emergent properties of programs that don't have to be programmed in. Internet weather, Gnutella topology, and mob development, for example, don't follow a pre-determined path. If machines don't have to be programmed to do every task, then the rest of your argument doesn't hold.
I don't mean to argue that emergence is the same as intelligence, and I agree with your general point that hand-crafted programs can't ever become intelligent.
If you've ever worked on shrinkwrapped software, you know that he's (mostly) right. Users don't buy bug fixes, they buy features.
Shrinkwrapped software makers have lower and lower returns on each bug. Once the number of bugs goes below the threshold of what's acceptable, it's not profitable or even sensible to fix them.
Sad but true. Bugs don't get fixed because customers don't buy bug fixes. Microsoft may be evil, but it's not dumb.
The article contradicts the spec, which clearly states that at least part of the purpose is to put a tracker in MP3s.
From the spec at: http://www.cidf.org/japanese/english/docs/gen /cidf -gen-en-79.pdf "This Identifier can also be used for usage surveillance services called Net-Police, which uses web spider to search out sites that might potentially be infringing content."
The pdf at that URL is *much* better documentation than the news article.
Jay is an extremely clueful guy with great personal authority to speak on this topic. His long involvement with Linux-related activism in New York is second to none, and anyone here would tell you the same. I understand that you worked hard and got the best deal you could, but there is still room for dissent.
You need to give us more detail about the tradeoff, Bruce. What options do patent holders have? Why is it necessary to accept any patented techniques at all, given that they defy the point of standard?
I don't think the data belongs to the songwriters or the record companies. The data is facts about the album ("this is the list of song titles" vs "this list of song titles is a creative work"). Facts don't belong to anyone.
I have an Archos Studio 20 and wouldn't recommend it. Intensely buggy -- constant power problems and disk errors. I'll probably chuck it and buy an iPod when iPod Linux support is mature.
KaZaA is promoting something an "Intellectual Property Use Fee". This is heavily flawed but still the only proposal aside from suggesting the labels change their business models.
See http://www.oreillynet.com/pub/wlg/1187 (my own writing) for a quick overview and endless related discussion on the [Decentralization] list, threads beginning at http://groups.yahoo.com/group/decentralization/mes sage/5337
The reason this can't happen is that the wireless service providers are in complete control of what goes over their airwaves. The wireless world does not use the end-to-end principle.
That may change at some point, but it hasn't happened yet.
That's a way of thinking that's really about filesharing applications, which were just the launching point for P2P. In the long run P2P is a bag of techniques (like location independent naming) that are being absorbed into the web. Asking who pays for those techniques is like asking who pays for tuplespaces -- the people that pay for P2P are the employers of the programmers who use P2P techniques to solve business problems.
A certificate signer like Thawte is explicitly in the business of managing reputation. Their certificates are as valuable as their certification process. If they see existence of a formalized organization as a good predictor of future behavior, then they should incorporate that information into their ruleset.
The amount of reputation information in a certificate from a high volume provider like Thawte is pretty small, and anything Thawte does to reduce the volume increases the amount of information. What they have done is a legal and completely relevant technique.
What they haven't done is provide more contextual information about the meaning of their certificate. What they _sell_ is a statement of reputability. What they _ship_ is a statement of having matched the claimed identity with a civil identity. Those are different things. In the long run I can't imagine that certificates won't become more closely matched with context.
We are indeed at the moment when the political implications of the internet can't be ignored. There are many latent ambiguities in the US constitution regarding these new questions, so Americans won't be able to fall back on precedent. There will be a long process to establish the rules of this new world.
It is a mistake to assume that there is no jurisdiction on the internet. To handle extraterritorial jurisdiction issues we will simply see treaties. There will be holes, and underground sites, but these will be only as accessible as the underground in meatspace.
A major battleground will be code. New technologies and new controlling legislation will be in an evolutionary race. Decentralized services make things more free, centralized services allow more control.
If you're a hacker who wants to be on the side of freedom, build highly distributed architectures. Use the GPL. Build privacy protecting tools to keep pace with new privacy invading tools.
just blowing some karma to say the posts in this thread are truly high quality/.: vintage in the making. Goes to show that for all the noise, signal still gets through. Thanks to all for contributing.
What discussions, like this one, of UI standards usually miss is that UI standards benefit experienced users as much as newbies, if not more. The point of these guidelines and standards is to allow people to reuse their knowledge. It's a major time saver to not have to figure out new stuff constantly.
Think of the confusing array of copy keystrokes you may use in a day: esc w (emacs) ctrl k ctrl u(pico) ctrl x (windows) ctrl insert (secureCRT under windows)
the attention you have to spend tracking this stuff would be better spent elsewhere.
That is the relevance of a book like this. However, the reviewer doesn't discuss the quality of the standards given by the books, which makes the review much less useful.
This post strikes me as kinda wacko, but I have to agree that having to notify the government in any way is a free speech issue.
To me the core issue is still unresolved, and hasn't even been addressed: to what degree is code speech? Information floats through different formats completely indifferent to the format. Laws regarding encryption treat it as an object, while it is just as much pure information, to which different laws apply.
My point is that there needs to be a major legal precedent clarifying the relationship between information in a platonic state, its instantiation in different passive formats (for example as text), and embodiments of the information within machines.
Imagine a hardware system (for example, an alarm clock) where legal restrictions apply - product liability laws. Replace the hardware system with a digital controller + physical noise maker. Out of all the steps to create the object - conceptualizing, designing logic flow, implementing source code for a particular digital context, compiling the source, running the executable, steps taken within the executable itself, interfacing between the executable and the physical world (EG by assisting the program in sending i/o to devices like a bell) - which one "created" it? The implication being that, at some point, responsibility has been taken on for product flaws.
The law right now is very different for objects and ideas. But digital versions of physical machines enable us to effectively replicate the machines; disassembly converts the machine back to information. The object itself is just another, potentially less liquid, form of information.
If a programmer speaks a program into a computer which then runs the program, do speech or object laws apply?
Re:WebMacro, Java servlets, and other comments
on
Java Success Stories
·
· Score: 1
As a result of your plugs here I went and checked out WebMacro. Your analysis of the problem is spot on and your design for the solution is elegant. Way cool.
However I notice that the last release was May 3rd, 1999, and that it's still beta. Also, going more than 6 months between releases - particularly for such early stage software - doesn't indicate a thriving project. What's the status of current development?
Can any WebMacro users out there comment on stability/robustness/completeness etc?
Learning LISP takes well over a half hour, particularly because using elisp depends on discovering a large number of API functions; ie dinking around for a long time while you learn the miscellaneous hacks that have come into use.
This is not an argument over the technical merits of LISP. This is an argument over means vs. ends. If the end goal is to make software more free, LISP was a bad choice because it discourages user contributions.
Requiring contributors to have learned programming in college - as they would have to to know LISP - eliminates the large number of programmers who learned on the job, or by hacking other free software. Emacs is the only major project that uses LISP, which is the single best reason why I am right and you are wrong.
Incidentally, you're stupid. How old are you, little fellow?
I don't see why dynamic extension should be the litmus test. If freedom is best served by enabling the user to change the source, then the popularity of C + the high power of C would have made it a better choice.
My point is that Lisp is and was such a non-standard that it works against the goal of enabling users to work on the software.
Discussion about LISP's technical value is beside the point, because the use of LISP in a Free Software project is counter to the ideology of Free Software.
Free Software in the RMS sense is about having the users be able to modify the code to fit their needs. But requiring knowledge of a little known language like LISP defeats this purpose. It keeps the majority of programmers from being able to do their own code modifications, since that would require them to learn a highly non-standard new language.
This is a perfect illustration of RMS' large blind spot. He is for a democracy of the elite, where the elite is programmers in academia. The value of usability - in GUIs (viz emacs totally wierd layout), programming languages (using LISP despite its unusability to the vast majority of programmers), and verbal communications (even the phrase "free software" requires some secret background knowledge) - to his larger goal of a more free society escapes him. If you are interested in devoting yourself wholly to GNU, then you are welcome; otherwise you're second class.
Serious revenue comes from those other, higher priced, versions. Keep in mind that almost all unix software is that expensive - it's the legacy price level.
Slashdot Mirror
What's the difference between Lustre and InterMezzo? Any idea?
I'm an O'Reilly author. They've been completely respectful about this -- no pressure at all, except to respond one way or the other.
There are emergent properties of programs that don't have to be programmed in. Internet weather, Gnutella topology, and mob development, for example, don't follow a pre-determined path. If machines don't have to be programmed to do every task, then the rest of your argument doesn't hold.
I don't mean to argue that emergence is the same as intelligence, and I agree with your general point that hand-crafted programs can't ever become intelligent.
If you've ever worked on shrinkwrapped software, you know that he's (mostly) right. Users don't buy bug fixes, they buy features.
Shrinkwrapped software makers have lower and lower returns on each bug. Once the number of bugs goes below the threshold of what's acceptable, it's not profitable or even sensible to fix them.
Sad but true. Bugs don't get fixed because customers don't buy bug fixes. Microsoft may be evil, but it's not dumb.
The article contradicts the spec, which clearly states that at least part of the purpose is to put a tracker in MP3s.
n /cidf -gen-en-79.pdf
From the spec at:
http://www.cidf.org/japanese/english/docs/ge
"This Identifier can also be used for usage surveillance services called Net-Police, which uses web spider to search out sites that might potentially be infringing content."
The pdf at that URL is *much* better documentation than the news article.
Jay is an extremely clueful guy with great personal authority to speak on this topic. His long involvement with Linux-related activism in New York is second to none, and anyone here would tell you the same. I understand that you worked hard and got the best deal you could, but there is still room for dissent.
You need to give us more detail about the tradeoff, Bruce. What options do patent holders have? Why is it necessary to accept any patented techniques at all, given that they defy the point of standard?
I don't think the data belongs to the songwriters or the record companies. The data is facts about the album ("this is the list of song titles" vs "this list of song titles is a creative work"). Facts don't belong to anyone.
Once you get used to a portable HD packed with many gigs of MP3s you never go back. CDs aren't even close.
I have an Archos Studio 20 and wouldn't recommend it. Intensely buggy -- constant power problems and disk errors. I'll probably chuck it and buy an iPod when iPod Linux support is mature.
To be fair, the FSF's contribution was the bootstrap for all the rest.
KaZaA is promoting something an "Intellectual Property Use Fee". This is heavily flawed but still the only proposal aside from suggesting the labels change their business models.
s sage/5337
See http://www.oreillynet.com/pub/wlg/1187 (my own writing) for a quick overview and endless related discussion on the [Decentralization] list, threads beginning at http://groups.yahoo.com/group/decentralization/me
The reason this can't happen is that the wireless service providers are in complete control of what goes over their airwaves. The wireless world does not use the end-to-end principle.
That may change at some point, but it hasn't happened yet.
That's a way of thinking that's really about filesharing applications, which were just the launching point for P2P. In the long run P2P is a bag of techniques (like location independent naming) that are being absorbed into the web. Asking who pays for those techniques is like asking who pays for tuplespaces -- the people that pay for P2P are the employers of the programmers who use P2P techniques to solve business problems.
A certificate signer like Thawte is explicitly in the business of managing reputation. Their certificates are as valuable as their certification process. If they see existence of a formalized organization as a good predictor of future behavior, then they should incorporate that information into their ruleset.
The amount of reputation information in a certificate from a high volume provider like Thawte is pretty small, and anything Thawte does to reduce the volume increases the amount of information. What they have done is a legal and completely relevant technique.
What they haven't done is provide more contextual information about the meaning of their certificate. What they _sell_ is a statement of reputability. What they _ship_ is a statement of having matched the claimed identity with a civil identity. Those are different things. In the long run I can't imagine that certificates won't become more closely matched with context.
We are indeed at the moment when the political implications of the internet can't be ignored. There are many latent ambiguities in the US constitution regarding these new questions, so Americans won't be able to fall back on precedent. There will be a long process to establish the rules of this new world.
It is a mistake to assume that there is no jurisdiction on the internet. To handle extraterritorial jurisdiction issues we will simply see treaties. There will be holes, and underground sites, but these will be only as accessible as the underground in meatspace.
A major battleground will be code. New technologies and new controlling legislation will be in an evolutionary race. Decentralized services make things more free, centralized services allow more control.
If you're a hacker who wants to be on the side of freedom, build highly distributed architectures. Use the GPL. Build privacy protecting tools to keep pace with new privacy invading tools.
Anybody know what the wireless protocol is for this? WAP/WIP? IP over cell?
Any thoughts on cell phone and other tiny device interfaces? As neat as these things are, is there a utility to those microscopic screens?
just blowing some karma to say the posts in this thread are truly high quality /.: vintage in the making. Goes to show that for all the noise, signal still gets through. Thanks to all for contributing.
What discussions, like this one, of UI standards usually miss is that UI standards benefit experienced users as much as newbies, if not more. The point of these guidelines and standards is to allow people to reuse their knowledge. It's a major time saver to not have to figure out new stuff constantly.
Think of the confusing array of copy keystrokes you may use in a day:
esc w (emacs)
ctrl k ctrl u(pico)
ctrl x (windows)
ctrl insert (secureCRT under windows)
the attention you have to spend tracking this stuff would be better spent elsewhere.
That is the relevance of a book like this. However, the reviewer doesn't discuss the quality of the standards given by the books, which makes the review much less useful.
This post strikes me as kinda wacko, but I have to agree that having to notify the government in any way is a free speech issue.
To me the core issue is still unresolved, and hasn't even been addressed: to what degree is code speech? Information floats through different formats completely indifferent to the format. Laws regarding encryption treat it as an object, while it is just as much pure information, to which different laws apply.
My point is that there needs to be a major legal precedent clarifying the relationship between information in a platonic state, its instantiation in different passive formats (for example as text), and embodiments of the information within machines.
Imagine a hardware system (for example, an alarm clock) where legal restrictions apply - product liability laws. Replace the hardware system with a digital controller + physical noise maker. Out of all the steps to create the object - conceptualizing, designing logic flow, implementing source code for a particular digital context, compiling the source, running the executable, steps taken within the executable itself, interfacing between the executable and the physical world (EG by assisting the program in sending i/o to devices like a bell) - which one "created" it? The implication being that, at some point, responsibility has been taken on for product flaws.
The law right now is very different for objects and ideas. But digital versions of physical machines enable us to effectively replicate the machines; disassembly converts the machine back to information. The object itself is just another, potentially less liquid, form of information.
If a programmer speaks a program into a computer which then runs the program, do speech or object laws apply?
As a result of your plugs here I went and checked out WebMacro. Your analysis of the problem is spot on and your design for the solution is elegant. Way cool.
However I notice that the last release was May 3rd, 1999, and that it's still beta. Also, going more than 6 months between releases - particularly for such early stage software - doesn't indicate a thriving project. What's the status of current development?
Can any WebMacro users out there comment on stability/robustness/completeness etc?
Learning LISP takes well over a half hour, particularly because using elisp depends on discovering a large number of API functions; ie dinking around for a long time while you learn the miscellaneous hacks that have come into use.
This is not an argument over the technical merits of LISP. This is an argument over means vs. ends. If the end goal is to make software more free, LISP was a bad choice because it discourages user contributions.
Requiring contributors to have learned programming in college - as they would have to to know LISP - eliminates the large number of programmers who learned on the job, or by hacking other free software. Emacs is the only major project that uses LISP, which is the single best reason why I am right and you are wrong.
Incidentally, you're stupid. How old are you, little fellow?
I don't see why dynamic extension should be the litmus test. If freedom is best served by enabling the user to change the source, then the popularity of C + the high power of C would have made it a better choice.
My point is that Lisp is and was such a non-standard that it works against the goal of enabling users to work on the software.
Discussion about LISP's technical value is beside the point, because the use of LISP in a Free Software project is counter to the ideology of Free Software.
Free Software in the RMS sense is about having the users be able to modify the code to fit their needs. But requiring knowledge of a little known language like LISP defeats this purpose. It keeps the majority of programmers from being able to do their own code modifications, since that would require them to learn a highly non-standard new language.
This is a perfect illustration of RMS' large blind spot. He is for a democracy of the elite, where the elite is programmers in academia. The value of usability - in GUIs (viz emacs totally wierd layout), programming languages (using LISP despite its unusability to the vast majority of programmers), and verbal communications (even the phrase "free software" requires some secret background knowledge) - to his larger goal of a more free society escapes him. If you are interested in devoting yourself wholly to GNU, then you are welcome; otherwise you're second class.
Serious revenue comes from those other, higher priced, versions. Keep in mind that almost all unix software is that expensive - it's the legacy price level.