If you want to plan such a course, here are some interesting, readable books that would be useful:
For the history of Mathematics, invention of zero, etc. read: "Mathematics: Queen & Servant of Science" by Eric Temple Bell.
Before the internet and the personal computer, two of the major uses and research topics for computers were Cryptography, and Artificial Intelligence. Of course plain old number crunching has always been important, but I don't know of any books on that.
But for the early history and development of cryptography, check out: "The Code Book: the evolution of secrecy from Mary, Queen of Scots to quantum cryptography" by Simon Singh
For a read on the early history of AI from a nay-sayers perspective, check out "What Computers Still Can't Do: A Critique of Artificial Reason" by Hubert L. Dreyfus.
Hmmm. Possible course outline schedule:
1. History of mathematics.
2. Pre-history of computers: Abacus, mechanical machines, Pascal, Von Neumann, etc.
4. First major use of computers: number crunching. The development of mathematical algorithms.
5. The Artificial Intelligence hype of the 60's and 70's.
6. Cryptography and computers.
7. Theoretical results: Turing, the incomputability theorems, equivalence of artificial languages, "All computer languages are the same", define and describe the P=NP question. "All the interesting questions are too hard".
8. The rise of the personal computer. I think "Fire in the Valley" is supposed to be good for this but I haven't read it?
9. The rise of the internet...
Maybe about a week on each of those... you would have to move pretty quick and just hit the high points, but it would be a pretty good "tour" of computer science.
Speaking of tours, another really interesting book is: "The Turing Omnibus: 61 Excursions in Computer Science" by A. K. Dewdney. It has standalone, easy to read chapters on topics like Algorithms, Finite Automata, Simulation, Godel's theorem, The Chomsky Hierarchy, Random Numbers, Error correcting codes, Boolean Logic, Time and Space Complexity, Recursion, Neural Nets, The Fast Fourier Transform, Public Key Cryptography, Number Systems for Computing, Parallel Computing, Logic Programming, Church's Thesis, Relational Databases...
Heck, you could teach the course entirely out of "Mathematics, queen and servant of science" followed by the Turing Omnibus. That would cover everything important...
I think the main point of having a really short algorithm is that it would be easy to memorize it, put it on T-shirts, and otherwise distribute it in so many ways that the MPAA will drive themselves crazy trying to stomp it out.
However, I think what would be more useful than the ultra-short C version would be the shortest readable version. That would be easier to understand, and easier to memorize.
That way, if ever dragged into court, you could tell the judge "My brain is an illegal circumvention device. So what are you going to do, put me in jail for knowing something?"
Exactly... look out, or the Turing theory of machine equivalences will become restricted information under the DMCA! (Any Turing-complete computer can emulate any other Turing-complete computer.) Heh. Teaching theoretical computer science will become illegal! Really, that's the logical end result of the DMCA.
That would be the obvious way to break this thing... Use Wine, or VMWare, or whatever to emulate a regular Windows machine so completely that the software running on it can't tell it isn't talking to the hardware.
Then your "virtual video card" can make copies of anything, and your "virtual sound card" can save everything to disk, and the pathetic copy management software running in the emulator doesn't know and can't stop it.
Of course, it might be difficult to write a good enough emulator. One obvious challenge would be for the copy management software to only allow playback/display on devices with digitally signed drivers. If I understand how VMWare works, that would be a problem because VMWare uses special Windows video and sound drivers that interface to the VMware virtual machine. But that can be solved as well, by emulating the video and sound hardware and running signed drivers on it.
The only way this stuff could ever be somewhat secure is if the software runs on sealed-box, tamper-proof, non-upgradable, un-documented hardware. That would make writing an emulator so difficult that most people wouldn't bother.
These companies should stop wasting everyone's time and just change their business models. I, for one, would be happy to pay for a music downloading service that reliably supplied me with top-quality, high bitrate MP3s, or even better, Vorbis Ogg files. Of course, the price should be fair (i.e. low), I should be able to get just one or two songs without having to buy the whole album, and most of the money should go to the artist. One dollar per song would be acceptable to me, and the artists could make more money that way.
But that destroys the business model of the big record labels, so they will fight it to the death... their business death or the death of our freedom, whichever is the weakest.
In no way does the GPL "play fair." As Stallman himself says, the GPL is "not Mr. Nice Guy." It's a Faustian bargain designed to destroy programmers' livelihoods.
What is unfair about it? The reason it is not "Mr Nice Guy" is that it disallows freeloading. That is, it PROTECTS the original programmer or programming community from being exploited. It has some teeth in it, backed up by copyright law, to prevent this exploitation.
Please explain why you think this is unfair. What's the Faustian bargin? If you don't like it, don't use it... no one puts a gun to a programmer's head, saying "We will make you use this GPLed code. And then all your code are belong to us, bwa ha ha ha!"
It's not like there aren't other options... if you don't want to deal with the GPL, there are hundreds of companies willing to sell you programming tools, software toolkits, and anything else you want. Or you can write it yourself.
But if you want to use the output of the free software community, you have to join the community. That sounds perfectly fair to me. I'm a software programmer, and the GPL isn't destroying my livelihood. In fact, I get paid very well to work on GPL'ed software, adapting it, fixing it, and configuring it to suit my company's needs. I love every minute of it. No more fighting with undocumented APIs, no more reverse-engineering buggy code to try to work around problems I can't fix...
Let me tell you: Working on GPL'ed code is what every programmer should dream for. I'm never going back to a closed source platform, now that I've had a taste of freedom.
Finally, just because I work on GPL'ed code and use a GPL'ed platform doesn't mean I, or anyone else, is forced to GPL all the code I write. I don't. My company has proprietary code where that makes sense. So... I really can't see why you think the GPL is so unethical. You still haven't explained why you think so.
Oops. I was wrong. Thank you for the gracious correction.
But... what about stuff like the Westlaw databases? That's a collection of public-domain legal information, and IIRC, all they did was add references and page numbers. But that's definitely copyrighted... perhaps that the example I should have used, rather than the phone book.
So, is the rule for copyrighting databases something like, "if you add anything significant, like an index, it can be copyrighted"?
If so, I suppose that CDDB can probably claim their database is copyrighted, but would not be able to prevent people from doing queries against it and building a new database from the results...
RMS's embrace and extend strategy is really quite a lot like Microsoft's.
Except that Microsoft's strategy is designed to bring them maximum profits, at the expense of the freedom of their users. But the GPL was specifically designed to provide maximum freedom to users, treating that as more important than profits.
And Microsoft leverages monopolies in one area to reduce customer choices in other areas, while GPL'ed software aims to be standards-compliant, thereby maximizing customer choices.
So, yes, they are "quite a lot like" if you define that to mean "totally different".
... BSD-licensed code, which (unlike GPLed code) is licensed ethically. Unaffected by the GPL's poison pill, it would have a much better chance of survival.
"Unethical"? "Poison pill"? Give me a break. Don't you understand the basic concept of the GPL?
It is: "I will share my code with you, but you have to share your improvements back to me. That way we all benefit."
And you call that UNETHICAL? You think that playing fair deserves the label "poison pill"? What kind of twisted, messed up ethics do you have? The license you prefer allows people to freeload - to take code, change it, and make it proprietary, refusing to share the improvements with the community that wrote the software in the first place. Is that the kind of ethical behaviour you like?
If you compile a database of pretty much anything, you can copyright it. If I compile a database containing only content owned by other people, such as say, an O'Reilly Nutshell book, can I still copyright it? Where do you draw the line? Just because you re-type something doesn't make it your own.
Are you a copyright lawyer? I don't think so! Neither am I, but I am sure that people can and have copyrighted collections (databases) of information that they put together - even if the information itself is free. The classic example, was phone books. If I recall correctly, the legal case that set the precedent was a company that was scanning, copying and selling phonebooks, because "all the information in them was public domain". They were sued and lost. You are not allowed to do that anymore. The copyright theory has something to do with a legal concept called "sweat of the brow" - if you put a lot of work into collecting information together, you can copyright the collection, even if all the data in it is public domain.
Read up on it. Copyright law is stranger than your intuitive idea about what is wright and wrong. Of course, there is a fuzzy line in there somewhere - a few pages of data probably wouldn't qualify. As they say, Ask A Lawyer.
And of course, retyping a book is a completely different thing. Don't confuse the issue. That would not be a collection of data, that's just a copy of information taken from ONE PLACE.
I have a 12 disk cd changer in the trunk, which is a relatively high risk of being stolen. There's no way I want to lose 12 CD's along with the hardware. Insurance doesn't cover CD's, either.
Especially since some of those CD's are out of print and are irreplacable. And many other CD's have only a few good tracks. And most music CD's only have about 45 minutes of music, even though I can record at least 70 minutes onto a CDR.
There are lots and lots of excellent, completely legal reasons to make CDR copies and compilations of music CDs. If they take away my ability to do that, I will either work (hack) around it, or refuse to buy such CDs. Prediction: If this catches on, the market for used, unprotected CD's will explode.
Hmmm. I wonder if the music world will eventually split into a "closed", copy-protected, mass-marketed, Limp Britney Eminem Metallica Bizcuit Big Record Company world, vs. an open, non-copy-protected, listener-supported, word-of-mouth marketing Internet music world. I just hope the bands I listen to end up on the good side. (U2 is the only "popular" band I really like, so I might be lucky).
Maybe someday we'll see "open music" and "free music" like open source and free software - and the free music bands will make their money selling T-shirts and concert tickets.
Somewhat more on topic, Shannon is mentioned directly in the Jargon file, where we find that Claude Shannon first used the word "bit" in print in the computer science sense.
Also, I just read the book "Crypto" by Steven Levy and was surprised to find out how critical Shannon's work on information theory was to early non-government crypto research. If it wasn't for his papers, it is likely that the US government would still have the only really strong cryptography in the world.
I have another experience with "processing". I have worn glasses since the fourth grade. A few years ago, I got my eyes re-examined, and got a set of contact lenses. This was after wearing the same set of glasses for about 5 years - I see very poorly without glasses. I still remember walking out of the opthamologist's office, wearing those contacts for the first time.
Not only was my visual clarity about twice as sharp as it had been with the old glasses, there was the significant difference that my peripheral vision was in focus, unlike peripheral vision with glasses.
The visual clarity was overwhelming - I was noticing so much - the edges of leaves on trees that had previously been sort of an indistinct green blur, details of people's hair across the street... so much sharp detail on the complex, organic stuff in the real world.
It gave me a headache within a couple of hours.
My brain was not used to getting so much visual information. The "software" for visual processing was suddenly having to deal with twice as much resolution, and it was constantly "getting behind" processing details that I didn't actually need. Over the next week, the sensation and problems went away as I got used to it - the brain is very adaptive.
I hope the "OpenPGP consortium" doesn't make it their objective to write yet another version of PGP. We all ready have one - Gnu Privacy Guard - which is both open source and RFC-whatever compliant. Plus it's fully scriptable, so it's easy to hook it up to other programs. And the documentation is even good.
On the other hand, if the OpenPGP consortium works with Hushmail, Zero Knowlege, and all the other companies out there to try to make secure email interoperable, that would be very, very nice.
I'm sure the NSA,CIA,FBI, and others get the giggles every day they decrypt email and think "Damn, these people are dumb! PGP has been out there in the world for years now, and almost nobody uses it!"
But frankly, it's a pain to use because it isn't integrated into enough software. For example, it would be nice if you could attach an OpenPGP signature to the text you put into an on-line form in Mozilla - like I am right now. Then we could have secure-signed Slashdot postings. Why? It's not like Slashdot's cookie-based login system is very secure - not that it was ever claimed to be - but if hacked into Slashdot (again) and managed to steal some username/password combinations, they could do a lot of damage to some people's reputations. I'm not talking about karma loss here - what if posts under your userID started showing up badmouthing the company you work for, and praising kiddie porn, and threatening to kill the president? You would have a rough time fixing that. GPG signatures would make it easy to prove you didn't do it.
And if my W2K box at work supported OpenPGP in Outlook, that would be nice too. So, I wish the best to Phil Katz and the OpenPGP consortium, as long as they don't bother to reinvent the Gnu Privacy Guard wheel. Look for innovative ways to add Open-PGP signatures to everything!
Regarding the relative quality of the Hollywood card and the ATI card... I have to say I had the opposite experience.
I have a P2-300 machine with both an ATI All-In-Wonder Rage 128 and a Sigma Designs / RealMagic Hollywood card. Under Windows 2000, the RealMagic card puts out a noticably better picture than the ATI does. The Sigma Designs Netstream 2000 should be even better.
Torrey Hoffman (Azog)
Re:The natural evolution of this...
on
License to Sit
·
· Score: 2
Hah! When I was in Morocco, in the main bus station in Fes, I paid to use a bathroom. "So what?", you ask...
This bathroom had squat toilets only. It wasn't clean. There was no toilet paper. There was only cold water. There was no soap. While I was squatting, my wallet fell out of my back pocket... into the hole. I had to wonder... what am I paying for, here?
Two weeks travelling in Morocco for cheap is enough to make you appreciate western style plumbing. And I mean North American - your average European bathroom is nowhere near as nice as an average US/Canadian bathroom. (Heh. How's that for flamebait?)
Well... you are being awfully paranoid. If I use the free version of Kylix to write a GPL'ed program, what possible danger could there be in anyone using it? All the libraries that program needs will be GPL'ed. So why shouldn't those libraries be included in distributions?
The old, thankfully resolved KDE license was very different. There were real legal issues with distributing KDE binaries that mixed QT and GPL licensed code.
There would be no such issue with GPL'ed programs written in Kylix. The only limitation is that the tool used to compile the program is not, in itself, GPLed. But it's a free download, so users would always have the right and the capability to modify the program they are using.
So... as long as that free download can't become "un-free" I don't think there's much to worry about. I will be carefully checking the click-through license on it though. I could imagine a scenario where Borland goes out of business, and you can't get the free version of Kylix from them any more, but the license makes it illegal for anyone else to distribute it.
That would be a real problem, because people using the GPL'ed software would not be able to exercise their right to modify it, because they wouldn't be able to get the tools to do so.
I suspect that it won't be an issue though, because it seems that Borland has thought this through pretty well. But that's hypothetical, and until we see the details of the license on the free version of Kylix it's not worth getting all stressed out about.
1. There is a free download version. You can only write GPL'ed programs with it because it links in code released under the GPL - even though the Kylix platform is not all GPL. I have no problem with that, I applaud Borland for coming up with an interesting way to support free software development while still maintaining some intellectual property. I know I'll give the free version a try when it comes out.
2. Given that a good developer costs far more than $1000 a week in salary, then this software is worth getting even if it saves less than a week of development time. If I like the free version of this software and find a use for it, I'm sure my company will buy me a copy.
Exactly. Prior art is what this site is about, and that's great.
But the biggest problem with patents today is the "obvious" factor. Apparently, patent examiners think everything is non-obvious. And there's no really good way to prove something was obvious years after the fact.
For example, web searching - sure, it's obvious. But how could you, in 2001, prove to a judge that the concept of a search engine was obvious whenever that patent was granted? Well, it was obvious as soon as the web existed, because archie was doing the a similar thing for FTP sites, and veronica was doing exactly the same thing for gopher sites.
Prior art is just the easiest way to get patents invalidated, even if they should never have been granted in the first place.
No! You missed the technology preview at the beginning! That 3D stuff was real camera footage - all of it. They have several hundred cameras set up around the rim of the statium, all controlled by custom software that can coordinate them. They all feed into "custom" hard drive video recorders and there's software that can basically do "matrix" style video tricks in near real time.
Frankly, seeing how they did that that was the coolest part of the whole 4 hours of TV. Although the halftime show with AeroSync was amusing.
I agree with your thought that technology is sometimes making it harder to tell what's real and what's "reconstructed", but in this case, all the "EyeVision" stuff was real, and based on switching very quickly between hundreds of cameras.
Same reason the waiters and waitresses at restauraunts always "sing" or shout some stupid "yeah, it's your birthday! Your birthday! your birthday! Yeah! Yeah! Whooooo!" song instead of Happy Birthday.
The restaurants would get sued otherwise. Evil, evil, evil.
On a related topic - you know Disney will pay whatever it takes - millions of dollars if neccessary - to bribe Congress to extend copyright again. They will never allow Mickey Mouse to enter the public domain.
Whoa there. When you talk about Linux becoming "less sophisticated and more user friendly" that really has almost nothing to do with the internals of the Linux kernel. Also, to become more user friendly, the kernel will probably continue to become more sophisticated, not less - things like hot swapping PCMCIA cards and USB devices require a very sophisticated kernel to be user friendly.
But in general, user friendliness is more about the graphical desktop and applications: Gnome, KDE, KOffice, StarOffice, Evolution, etc. There's a lot of differences at that level, so it would be impossible to have a single user reference for the notional "Linux User Interface". But the GUI is not strongly tied to the kernel.
I think that it should be possible to have good referenece documentation for the kernel, probably as a two volume set - one for the networking stack, and one for everything else. Just because you think the kernel is too complex to understand doesn't mean that everyone else agrees... And if we don't try to document it and understand it, we will end up in a situation where a very small number of people have the technical skills to work on the kernel, and that would be bad for the Linux community. One of the things that keeps Linux on course is that people have the option to fork it - either because they want to experiment with something, or because they have a different problem to solve (RTLinux), or they think they really can do it better than Linus. If the kernel is too hard for anyone but Linus, Alan Cox, and 30 other people in the world to understand, then that option to fork isn't really there anymore, and that would be bad for the future of Linux.
User documentation, on how to use Linux, is also important but is a completely separate issue and is probably best done by documenting distributions. Like "How To Use Red Hat 7". Of course that book doesn't need to say anything about how the kernel works - users don't care, and they shouldn't have to.
Myrio (www.myrio.com) has been doing something like this for a while. In Livingston, Texas, there are now paying customers who have TV over DSL. The Myrio system also provides video on demand, which is like having a virtual DVD player - you can rent movies over the DSL, and get full capability to pause, fast forward, rewind, and play it over again for the length of the rental period. The movie content is streamed off big NCube servers at the telco head end.
A little more info: Myrio's system is based on streaming MPEG-2 through full rate DSL. That's typically about 8 Mbps, and is enough bandwidth for two set top boxes to watch two different TV channels (or movies) simultaneously. There's no real limit to the number of TV channels the system can handle. The video quality is very high - movies are very close to DVD, and TV is better than regular cable.
Full disclosure - I work for Myrio. I don't know anything about MPhase.
What you said may very well be true for OpenBSD vs. Linux 2.2.x.
I don't think Linux 2.4.x has been out there long enough for anyone to be making absolute statements about which TCP stack is more robust or fast... but remember, the Linux 2.4 TCP/IP was specifically rewritten to be very fast and highly deserialized (i.e. SMP-capable) so even if [Open,Free]BSD TCP/IP was faster and better, it might not be now.
Especially on SMP hardware, which current BSD's don't support. And even though I'm not flaming you, I should mention that I have used both Linux and BSD for firewalls. Currently my firewall is OpenBSD, and I'm not in any rush to change it - not much need for speed on a DSL connection.
Slashdot Mirror
If you want to plan such a course, here are some interesting, readable books that would be useful:
For the history of Mathematics, invention of zero, etc. read: "Mathematics: Queen & Servant of Science" by Eric Temple Bell.
Before the internet and the personal computer, two of the major uses and research topics for computers were Cryptography, and Artificial Intelligence. Of course plain old number crunching has always been important, but I don't know of any books on that.
But for the early history and development of cryptography, check out: "The Code Book: the evolution of secrecy from Mary, Queen of Scots to quantum cryptography" by Simon Singh
For a read on the early history of AI from a nay-sayers perspective, check out "What Computers Still Can't Do: A Critique of Artificial Reason" by Hubert L. Dreyfus.
Hmmm. Possible course outline schedule:
1. History of mathematics.
2. Pre-history of computers: Abacus, mechanical machines, Pascal, Von Neumann, etc.
4. First major use of computers: number crunching. The development of mathematical algorithms.
5. The Artificial Intelligence hype of the 60's and 70's.
6. Cryptography and computers.
7. Theoretical results: Turing, the incomputability theorems, equivalence of artificial languages, "All computer languages are the same", define and describe the P=NP question. "All the interesting questions are too hard".
8. The rise of the personal computer. I think "Fire in the Valley" is supposed to be good for this but I haven't read it?
9. The rise of the internet...
Maybe about a week on each of those... you would have to move pretty quick and just hit the high points, but it would be a pretty good "tour" of computer science.
Speaking of tours, another really interesting book is: "The Turing Omnibus: 61 Excursions in Computer Science" by A. K. Dewdney. It has standalone, easy to read chapters on topics like Algorithms, Finite Automata, Simulation, Godel's theorem, The Chomsky Hierarchy, Random Numbers, Error correcting codes, Boolean Logic, Time and Space Complexity, Recursion, Neural Nets, The Fast Fourier Transform, Public Key Cryptography, Number Systems for Computing, Parallel Computing, Logic Programming, Church's Thesis, Relational Databases...
Heck, you could teach the course entirely out of "Mathematics, queen and servant of science" followed by the Turing Omnibus. That would cover everything important...
Torrey Hoffman (Azog)
I think the main point of having a really short algorithm is that it would be easy to memorize it, put it on T-shirts, and otherwise distribute it in so many ways that the MPAA will drive themselves crazy trying to stomp it out.
However, I think what would be more useful than the ultra-short C version would be the shortest readable version. That would be easier to understand, and easier to memorize.
That way, if ever dragged into court, you could tell the judge "My brain is an illegal circumvention device. So what are you going to do, put me in jail for knowing something?"
Heh.
Torrey Hoffman (Azog)
Exactly... look out, or the Turing theory of machine equivalences will become restricted information under the DMCA! (Any Turing-complete computer can emulate any other Turing-complete computer.) Heh. Teaching theoretical computer science will become illegal! Really, that's the logical end result of the DMCA.
That would be the obvious way to break this thing... Use Wine, or VMWare, or whatever to emulate a regular Windows machine so completely that the software running on it can't tell it isn't talking to the hardware.
Then your "virtual video card" can make copies of anything, and your "virtual sound card" can save everything to disk, and the pathetic copy management software running in the emulator doesn't know and can't stop it.
Of course, it might be difficult to write a good enough emulator. One obvious challenge would be for the copy management software to only allow playback/display on devices with digitally signed drivers. If I understand how VMWare works, that would be a problem because VMWare uses special Windows video and sound drivers that interface to the VMware virtual machine. But that can be solved as well, by emulating the video and sound hardware and running signed drivers on it.
The only way this stuff could ever be somewhat secure is if the software runs on sealed-box, tamper-proof, non-upgradable, un-documented hardware. That would make writing an emulator so difficult that most people wouldn't bother.
These companies should stop wasting everyone's time and just change their business models. I, for one, would be happy to pay for a music downloading service that reliably supplied me with top-quality, high bitrate MP3s, or even better, Vorbis Ogg files. Of course, the price should be fair (i.e. low), I should be able to get just one or two songs without having to buy the whole album, and most of the money should go to the artist. One dollar per song would be acceptable to me, and the artists could make more money that way.
But that destroys the business model of the big record labels, so they will fight it to the death... their business death or the death of our freedom, whichever is the weakest.
Torrey Hoffman (Azog)
What is unfair about it? The reason it is not "Mr Nice Guy" is that it disallows freeloading. That is, it PROTECTS the original programmer or programming community from being exploited. It has some teeth in it, backed up by copyright law, to prevent this exploitation.
Please explain why you think this is unfair. What's the Faustian bargin? If you don't like it, don't use it... no one puts a gun to a programmer's head, saying "We will make you use this GPLed code. And then all your code are belong to us, bwa ha ha ha!"
It's not like there aren't other options... if you don't want to deal with the GPL, there are hundreds of companies willing to sell you programming tools, software toolkits, and anything else you want. Or you can write it yourself.
But if you want to use the output of the free software community, you have to join the community. That sounds perfectly fair to me. I'm a software programmer, and the GPL isn't destroying my livelihood. In fact, I get paid very well to work on GPL'ed software, adapting it, fixing it, and configuring it to suit my company's needs. I love every minute of it. No more fighting with undocumented APIs, no more reverse-engineering buggy code to try to work around problems I can't fix...
Let me tell you: Working on GPL'ed code is what every programmer should dream for. I'm never going back to a closed source platform, now that I've had a taste of freedom.
Finally, just because I work on GPL'ed code and use a GPL'ed platform doesn't mean I, or anyone else, is forced to GPL all the code I write. I don't. My company has proprietary code where that makes sense. So... I really can't see why you think the GPL is so unethical. You still haven't explained why you think so.
Torrey Hoffman (Azog)
hey, you stole my sig!
Or do you read alan's on-line diary too?
Torrey Hoffman (Azog)
Oops. I was wrong. Thank you for the gracious correction.
But... what about stuff like the Westlaw databases? That's a collection of public-domain legal information, and IIRC, all they did was add references and page numbers. But that's definitely copyrighted... perhaps that the example I should have used, rather than the phone book.
So, is the rule for copyrighting databases something like, "if you add anything significant, like an index, it can be copyrighted"?
If so, I suppose that CDDB can probably claim their database is copyrighted, but would not be able to prevent people from doing queries against it and building a new database from the results...
Hmm.
Torrey Hoffman (Azog)
And Microsoft leverages monopolies in one area to reduce customer choices in other areas, while GPL'ed software aims to be standards-compliant, thereby maximizing customer choices.
So, yes, they are "quite a lot like" if you define that to mean "totally different".
Torrey Hoffman (Azog)
It is: "I will share my code with you, but you have to share your improvements back to me. That way we all benefit."
And you call that UNETHICAL? You think that playing fair deserves the label "poison pill"? What kind of twisted, messed up ethics do you have? The license you prefer allows people to freeload - to take code, change it, and make it proprietary, refusing to share the improvements with the community that wrote the software in the first place. Is that the kind of ethical behaviour you like?
I sincerely pity you.
Torrey Hoffman (Azog)
Are you a copyright lawyer? I don't think so! Neither am I, but I am sure that people can and have copyrighted collections (databases) of information that they put together - even if the information itself is free. The classic example, was phone books. If I recall correctly, the legal case that set the precedent was a company that was scanning, copying and selling phonebooks, because "all the information in them was public domain". They were sued and lost. You are not allowed to do that anymore. The copyright theory has something to do with a legal concept called "sweat of the brow" - if you put a lot of work into collecting information together, you can copyright the collection, even if all the data in it is public domain.
Read up on it. Copyright law is stranger than your intuitive idea about what is wright and wrong. Of course, there is a fuzzy line in there somewhere - a few pages of data probably wouldn't qualify. As they say, Ask A Lawyer.
And of course, retyping a book is a completely different thing. Don't confuse the issue. That would not be a collection of data, that's just a copy of information taken from ONE PLACE.
Better check your own assumptions.
Torrey Hoffman (Azog)
Of course I make copies of CDs. All the time.
I have a 12 disk cd changer in the trunk, which is a relatively high risk of being stolen. There's no way I want to lose 12 CD's along with the hardware. Insurance doesn't cover CD's, either.
Especially since some of those CD's are out of print and are irreplacable. And many other CD's have only a few good tracks. And most music CD's only have about 45 minutes of music, even though I can record at least 70 minutes onto a CDR.
There are lots and lots of excellent, completely legal reasons to make CDR copies and compilations of music CDs. If they take away my ability to do that, I will either work (hack) around it, or refuse to buy such CDs. Prediction: If this catches on, the market for used, unprotected CD's will explode.
Hmmm. I wonder if the music world will eventually split into a "closed", copy-protected, mass-marketed, Limp Britney Eminem Metallica Bizcuit Big Record Company world, vs. an open, non-copy-protected, listener-supported, word-of-mouth marketing Internet music world. I just hope the bands I listen to end up on the good side. (U2 is the only "popular" band I really like, so I might be lucky).
Maybe someday we'll see "open music" and "free music" like open source and free software - and the free music bands will make their money selling T-shirts and concert tickets.
Torrey Hoffman (Azog)
Actually, LISP's funny interpretation is "Lots of Irritating Superfluous Parentheses".
Check it out at the Jargon File:LISP
Somewhat more on topic, Shannon is mentioned directly in the Jargon file, where we find that
Claude Shannon first used the word "bit" in print in the computer science sense.
Also, I just read the book "Crypto" by Steven Levy and was surprised to find out how critical Shannon's work on information theory was to early non-government crypto research. If it wasn't for his papers, it is likely that the US government would still have the only really strong cryptography in the world.
He will be remembered.
Torrey Hoffman (Azog)
I have another experience with "processing". I have worn glasses since the fourth grade. A few years ago, I got my eyes re-examined, and got a set of contact lenses. This was after wearing the same set of glasses for about 5 years - I see very poorly without glasses. I still remember walking out of the opthamologist's office, wearing those contacts for the first time.
Not only was my visual clarity about twice as sharp as it had been with the old glasses, there was the significant difference that my peripheral vision was in focus, unlike peripheral vision with glasses.
The visual clarity was overwhelming - I was noticing so much - the edges of leaves on trees that had previously been sort of an indistinct green blur, details of people's hair across the street... so much sharp detail on the complex, organic stuff in the real world.
It gave me a headache within a couple of hours.
My brain was not used to getting so much visual information. The "software" for visual processing was suddenly having to deal with twice as much resolution, and it was constantly "getting behind" processing details that I didn't actually need. Over the next week, the sensation and problems went away as I got used to it - the brain is very adaptive.
Torrey Hoffman (Azog)
I hope the "OpenPGP consortium" doesn't make it their objective to write yet another version of PGP. We all ready have one - Gnu Privacy Guard - which is both open source and RFC-whatever compliant. Plus it's fully scriptable, so it's easy to hook it up to other programs. And the documentation is even good.
On the other hand, if the OpenPGP consortium works with Hushmail, Zero Knowlege, and all the other companies out there to try to make secure email interoperable, that would be very, very nice.
I'm sure the NSA,CIA,FBI, and others get the giggles every day they decrypt email and think "Damn, these people are dumb! PGP has been out there in the world for years now, and almost nobody uses it!"
But frankly, it's a pain to use because it isn't integrated into enough software. For example, it would be nice if you could attach an OpenPGP signature to the text you put into an on-line form in Mozilla - like I am right now. Then we could have secure-signed Slashdot postings. Why? It's not like Slashdot's cookie-based login system is very secure - not that it was ever claimed to be - but if hacked into Slashdot (again) and managed to steal some username/password combinations, they could do a lot of damage to some people's reputations. I'm not talking about karma loss here - what if posts under your userID started showing up badmouthing the company you work for, and praising kiddie porn, and threatening to kill the president? You would have a rough time fixing that. GPG signatures would make it easy to prove you didn't do it.
And if my W2K box at work supported OpenPGP in Outlook, that would be nice too. So, I wish the best to Phil Katz and the OpenPGP consortium, as long as they don't bother to reinvent the Gnu Privacy Guard wheel. Look for innovative ways to add Open-PGP signatures to everything!
Torrey Hoffman (Azog)
Regarding the relative quality of the Hollywood card and the ATI card... I have to say I had the opposite experience.
I have a P2-300 machine with both an ATI All-In-Wonder Rage 128 and a Sigma Designs / RealMagic Hollywood card. Under Windows 2000, the RealMagic card puts out a noticably better picture than the ATI does. The Sigma Designs Netstream 2000 should be even better.
Torrey Hoffman (Azog)
Hah! When I was in Morocco, in the main bus station in Fes, I paid to use a bathroom. "So what?", you ask...
This bathroom had squat toilets only. It wasn't clean. There was no toilet paper. There was only cold water. There was no soap. While I was squatting, my wallet fell out of my back pocket... into the hole. I had to wonder... what am I paying for, here?
Two weeks travelling in Morocco for cheap is enough to make you appreciate western style plumbing. And I mean North American - your average European bathroom is nowhere near as nice as an average US/Canadian bathroom. (Heh. How's that for flamebait?)
Torrey Hoffman (Azog)
Well... you are being awfully paranoid. If I use the free version of Kylix to write a GPL'ed program, what possible danger could there be in anyone using it? All the libraries that program needs will be GPL'ed. So why shouldn't those libraries be included in distributions?
The old, thankfully resolved KDE license was very different. There were real legal issues with distributing KDE binaries that mixed QT and GPL licensed code.
There would be no such issue with GPL'ed programs written in Kylix. The only limitation is that the tool used to compile the program is not, in itself, GPLed. But it's a free download, so users would always have the right and the capability to modify the program they are using.
So... as long as that free download can't become "un-free" I don't think there's much to worry about. I will be carefully checking the click-through license on it though. I could imagine a scenario where Borland goes out of business, and you can't get the free version of Kylix from them any more, but the license makes it illegal for anyone else to distribute it.
That would be a real problem, because people using the GPL'ed software would not be able to exercise their right to modify it, because they wouldn't be able to get the tools to do so.
I suspect that it won't be an issue though, because it seems that Borland has thought this through pretty well. But that's hypothetical, and until we see the details of the license on the free version of Kylix it's not worth getting all stressed out about.
Torrey Hoffman (Azog)
Two points in response:
1. There is a free download version. You can only write GPL'ed programs with it because it links in code released under the GPL - even though the Kylix platform is not all GPL. I have no problem with that, I applaud Borland for coming up with an interesting way to support free software development while still maintaining some intellectual property. I know I'll give the free version a try when it comes out.
2. Given that a good developer costs far more than $1000 a week in salary, then this software is worth getting even if it saves less than a week of development time. If I like the free version of this software and find a use for it, I'm sure my company will buy me a copy.
Torrey Hoffman (Azog)
Exactly. Prior art is what this site is about, and that's great.
But the biggest problem with patents today is the "obvious" factor. Apparently, patent examiners think everything is non-obvious. And there's no really good way to prove something was obvious years after the fact.
For example, web searching - sure, it's obvious. But how could you, in 2001, prove to a judge that the concept of a search engine was obvious whenever that patent was granted? Well, it was obvious as soon as the web existed, because archie was doing the a similar thing for FTP sites, and veronica was doing exactly the same thing for gopher sites.
Prior art is just the easiest way to get patents invalidated, even if they should never have been granted in the first place.
Torrey Hoffman (Azog)
No! You missed the technology preview at the beginning! That 3D stuff was real camera footage - all of it. They have several hundred cameras set up around the rim of the statium, all controlled by custom software that can coordinate them. They all feed into "custom" hard drive video recorders and there's software that can basically do "matrix" style video tricks in near real time.
Frankly, seeing how they did that that was the coolest part of the whole 4 hours of TV. Although the halftime show with AeroSync was amusing.
I agree with your thought that technology is sometimes making it harder to tell what's real and what's "reconstructed", but in this case, all the "EyeVision" stuff was real, and based on switching very quickly between hundreds of cameras.
Torrey Hoffman (Azog)
But the accent was over the t! That doesn't make sense at all!
Torrey Hoffman (Azog)
Same reason the waiters and waitresses at restauraunts always "sing" or shout some stupid "yeah, it's your birthday! Your birthday! your birthday! Yeah! Yeah! Whooooo!" song instead of Happy Birthday.
The restaurants would get sued otherwise. Evil, evil, evil.
On a related topic - you know Disney will pay whatever it takes - millions of dollars if neccessary - to bribe Congress to extend copyright again. They will never allow Mickey Mouse to enter the public domain.
The only hope is campaign finance reform.
Torrey Hoffman (Azog)
Whoa there. When you talk about Linux becoming "less sophisticated and more user friendly" that really has almost nothing to do with the internals of the Linux kernel. Also, to become more user friendly, the kernel will probably continue to become more sophisticated, not less - things like hot swapping PCMCIA cards and USB devices require a very sophisticated kernel to be user friendly.
But in general, user friendliness is more about the graphical desktop and applications: Gnome, KDE, KOffice, StarOffice, Evolution, etc. There's a lot of differences at that level, so it would be impossible to have a single user reference for the notional "Linux User Interface". But the GUI is not strongly tied to the kernel.
I think that it should be possible to have good referenece documentation for the kernel, probably as a two volume set - one for the networking stack, and one for everything else. Just because you think the kernel is too complex to understand doesn't mean that everyone else agrees... And if we don't try to document it and understand it, we will end up in a situation where a very small number of people have the technical skills to work on the kernel, and that would be bad for the Linux community. One of the things that keeps Linux on course is that people have the option to fork it - either because they want to experiment with something, or because they have a different problem to solve (RTLinux), or they think they really can do it better than Linus. If the kernel is too hard for anyone but Linus, Alan Cox, and 30 other people in the world to understand, then that option to fork isn't really there anymore, and that would be bad for the future of Linux.
User documentation, on how to use Linux, is also important but is a completely separate issue and is probably best done by documenting distributions. Like "How To Use Red Hat 7". Of course that book doesn't need to say anything about how the kernel works - users don't care, and they shouldn't have to.
Torrey Hoffman (Azog)
Myrio (www.myrio.com) has been doing something like this for a while. In Livingston, Texas, there are now paying customers who have TV over DSL. The Myrio system also provides video on demand, which is like having a virtual DVD player - you can rent movies over the DSL, and get full capability to pause, fast forward, rewind, and play it over again for the length of the rental period. The movie content is streamed off big NCube servers at the telco head end.
Check the old press release at Myrio's web site.
A little more info: Myrio's system is based on streaming MPEG-2 through full rate DSL. That's typically about 8 Mbps, and is enough bandwidth for two set top boxes to watch two different TV channels (or movies) simultaneously. There's no real limit to the number of TV channels the system can handle. The video quality is very high - movies are very close to DVD, and TV is better than regular cable.
Full disclosure - I work for Myrio. I don't know anything about MPhase.
Torrey Hoffman (Azog)
What you said may very well be true for OpenBSD vs. Linux 2.2.x.
I don't think Linux 2.4.x has been out there long enough for anyone to be making absolute statements about which TCP stack is more robust or fast... but remember, the Linux 2.4 TCP/IP was specifically rewritten to be very fast and highly deserialized (i.e. SMP-capable) so even if [Open,Free]BSD TCP/IP was faster and better, it might not be now.
Especially on SMP hardware, which current BSD's don't support. And even though I'm not flaming you, I should mention that I have used both Linux and BSD for firewalls. Currently my firewall is OpenBSD, and I'm not in any rush to change it - not much need for speed on a DSL connection.
Torrey Hoffman (Azog)
Wrong. Abrash did not work at all on Quake 3. He worked on Quake 1.
Torrey Hoffman (Azog)