All of the above is true, however... the new driver may not work for everyone.
Alan Cox, for example, tried them out and found they didn't work on some of his machines, so he didn't include the new driver in the -ac patches.
So, it seems they work great for some people but not others. Linus decided to keep them in because the new driver is being maintained, and the old one wasn't - a buggy driver that's being worked on is better than a buggy driver that's being ignored... and the new drivers have much better support for the features of the SBLive.
Anyway, if you upgrade and find that your sound card quits working, you have several choices - revert to your previous working version, or try Alan Cox'es patched version of the kernel (look under/people/alan on the kernel mirrors) or, if you know what you're doing, just take the relevant emu10k files from 2.4.7 or so and patch them into your tree. Alan Cox's kernels have lots and lots of other changes besides the SBLive driver, of course.
A lot of work is being done right now on getting the VM (Virtual Memory) balanced right. It works great for a lot of people, but depending on the workload, it can start to thrash really bad and swap itself to death. (kswapd will run, taking all the CPU time and making no progress.) If you run into this problem or others, read the documentation on reporting bugs (see the/linux/REPORTING-BUGS file) and send a mail to the list. Don't hold your breath waiting for a reply, but it will be read, and it might help the developers figure out the pattern of what situations cause problems and what works well.
(If you report bugs on the Linux Kernel Mailing list, be sure to note what kernel you're running and any non-standard changes you've made.)
Oh, and if you are not subscribed to the list and want to be cc'ed on replies, say so at the top of your email.
(I'm not really a kernel developer, yet, but I hack around in the code and read the list, so I know a little about what's going on.)
I know what would get worms back into the media for a long time - a Warhol Worm. You want to read something scary about worms, go read that. Be sure to read the section "A Worst Case Warhol Worm". It gives me the shivers to think about it.
From the article: "A worst case Warhol Worm is truly frightening, capable of doing many billions of dollars in real damage and disruption. Since it can achieve complete spread in well under an hour, and could begin doing damage immediately on infecting a machine, human mediated responses offer almost no hope of stopping it. "
Complete spread in under an hour! Total destruction of infected servers!
Whee!
Watch for one of these coming out with the next major IIS exploit.
Re:analog; hacking cuts both ways
on
Taming the Web
·
· Score: 2
Nitpick: there are methods of public key encryption that don't depend on the difficulty of factoring products of large primes for their security. Elliptical curves, for instance.
And of course, public key / private key is a major convenience, but all you really need is old school secret key cryptography to make a private, secure network.
Finally, there are encryption methods that rely on the conjecture that P != NP - that is, breaking the encryption would answer the most significant unanswered question in mathematics of the last hundred years. I feel pretty safe about that. Oh yeah, and quantum cryptography too.
But anyway, yes, it is a good thing that there are some things that nobody knows how to break.
Actually, you can (still?) buy "pro" audio DA / AD gear with no restrictions. For example, check out M-audio's pro-grade sound cards. There's even Linux drivers.
And they aren't too expensive. The basic "Audiophile 2496" card has 24 bit, 96 Khz DA and AD converters, SPDIF digital IO, and MIDI. Even better, the SPDIF connectors on that board give the user full control over the SCMS (aka "scums") serial copy management system. The
Sound Blaster Live's SPDIF, on the other hand, won't let you copy SCMS-protected content, even if you own the DVD it came from. Apparently this is normal for "consumer" grade stuff.
Get affordable "pro" gear while you still can!
Re:Some interesting implications
on
Share The Pi!
·
· Score: 2
Turing machines with attached "oracles" are used for proofs in theoretical computer science, but it's important to keep in mind that they are purely theoretical - no one will ever build an oracle. You don't think about how an "oracle" works. It's just a concept, like imagining that magic works. So why is this useful?
Well... For instance, you might be able to prove that such-and-such a problem with input size "n" could be solved in polynomial time (i.e. "fast") if you just had a magical oracle to supply you with only log(n)correct, one-bit answers.
The point of the proof would be that you don't need more than log(n) magic bits from the oracle. So what good is that? Well... If you can get the number of magic bits small enough, while still keeping the algorithm fast, it may provide a way to do a randomized algorithm where instead of using the magical oracle (which doesn't exist, remember?) you just use a random number generator, or maybe just try everything. Since a random number generator isn't as accurate as a magical oracle, you run the algorithm a lot of times with different random bits. Maybe you'll be lucky soon, and maybe that's good enough on average.
Theoretical computer science is fun. It's a little crazy and non-intuitive somtimes.:-)
Yes! "Advanced Programming in the Unix Environment", which is part of the same series from Richard Stevens, is my trusted companion as a C programmer making the escape from Microsoft to Linux.
And it's not Linux specific - it's pretty much POSIX, and does talk about some of the differences between implementations where they matter.
I memorized pi to 50 places the summer between grade 10 and 11 just so I could be a smartass in math and physics classes the next year.
It's actually not hard to do. Break it up into groups of 5 digits, and memorize 5 digits a day and keep practicing. A few minutes a day, and you can be a loser^H^H^H^H^H, er, cool geek like me too.
I wish I could get into the real story, but it's slashdotted, so I'm posting pointless crap like this. Geez.
One obvious way to do it: Get a really, really good CD player which has top-notch D/A converters, a good transport mechanism, and very high quality analog circuitry. (From what other posters have written here, you _have_ to use the D/A converters in the CD player, as that's the only way to get all the information from the ECC bits.)
Then hook the analog output of that CD player to a decent outboard A/D converter using good cables. The outboard A/D converter feeds the digital result to your computer through an SPDIF cable, and there you go... The only headache is that you don't get the automatic CDDB lookup and track numbering.
Remember, since you're going to be compressing the result to OGG or MP3 anyway, the small loss in sound quality that you have by going through the CD player's D/A and the outboard A/D converter is essentially irrelevant. Especially if you're going to play it through computer speakers, which are all pretty lousy anyway.
In fact, with a good CD player and A/D equipment, the resulting compressed audio would sound better than your average 128 bit Napster crap.
Attention, companies and advertising agencies that might be reading this!
I make a note of companies that have irritating ads, and I will NEVER buy from them!
For example, those damn X10 popup ads which invariably feature women in suggestive poses... I'll never, ever buy anything from that company. And that's despite the fact that I'm a geek with a house full of electronics, and wireless cameras are right up my alley. Forget it. Ain't gonna happen. Same with any other company that gets on my nerves.
If I wanted annoying advertising, I would watch TV. And I gave that up completely three years ago.
I'll have to get an "us" domain. There's too many great possibilities to pass up.
For instance, a great second level domain would be "with.us". Then you could resell all the commercially useful third level domains: bank.with.us, shop.with.us, save.with.us, fly.with.us.
But there's loads of other great possibilities. You read them here first:
No, the GPL does not allow _linking_, but linking is NOT the same thing as communicating via some sort of message passing.
On some architectures, that line can get blurry, but under Windows and Linux it's pretty clear.
It's the difference between calling a function in a DLL or.so (linking) and sending messages down a pipe, or using a shared file.
This is NOT a loophole that you would want to close. Suppoose you did, and you created an operating system and libraries that used your "improved" license. No one would even be able to RUN closed source software on it, because running a program under Unix involves _exactly_ the same mechanisms that you would have "closed". So say goodbye to Oracle or Quake 3 or Wordperfect or ViaVoice... or any of the other closed-source apps that are really nice to have on Linux.
The GPL strikes a very, very fine balance between the rights of users to have control over their software, and the ability of programmers to write closed source software and still run it on an open system.
It is true that things can become blurry. On some "message-passing" style operating systems, there's less of a distinction between a function call and interprocess communication. Luckily, that's not really an issue for any popular OS right now.
Exactly! If someone is going to sell hardware at a loss, that's their problem.
This is about to become very interesting with Microsoft's X-box coming out. Here we have a high-powered PC, with top-notch graphics, a P-III 733Mhz, 64MB of RAM (or is it 128?) plus DVD, 100MBit ethernet, and other useful stuff. It will be sold by Microsoft for at least $100 dollars below cost.
Now, what's to stop people from porting Linux to it and using it as a home computer? Or a cheap network server? Maybe it doesn't have VGA output?
Maybe Microsoft will try to write the contract / purchase agreement to make that illegal. I doubt that will work, just have a minor buy the hardware, the contract will not be legally binding then.
But since Microsoft intends to make the money back by licensing software development kits, they have to maintain firm control over who can create software for the box.
My expectation is that the BIOS of the X-Box will refuse to boot from any device which does not have a Microsoft-copyrighted signature in the boot sector. That doesn't mean that people won't be able to hack it, but it does mean that anyone that *distributes* software for it without a license from Microsoft can be sued for copyright violation.
It will keep hacks private and non-commercial at least. Will that be enough? After all, if someone puts an ISO image of an X-Box bootable Linux distro on the net, missing only the Microsoft signature, and the same web page has instructions on how to copy the Microsoft signature off your copy of "Halo" to make the Linux disk bootable... I'm sure Microsoft will send out the attack dog lawyers.
Getting back on point.... Netpliance could have done the same things to make sure the device only was used the way they want. But they didn't. Tough for them.
Well, if they did this, I might buy a couple hundred dollars worth. Not that I would expect to make money anytime soon, or maybe ever, but because I like Mandrake.
The reason it's different from donations is that presumably, as a stockholder, I would have some input into the direction of the company. Maybe they could have shareholder voting on company direction questions over the internet - one vote per share... like, should we make Mozilla the default browser in Mandrake 8.1? Should we continue shipping proprietary software in the boxed version? Should we hire more developers? What should the developers work on? Of course, as a private company the CEO is ultimately responsible, and shareholders can really just vote to replace the CEO and board. Or something, I'm no expert on corparate law, especially for a French company.
But anyway, my point is that it would not be just a donation to Mandrake.
If this is really successfull, Mandrake might end up kind of like an incorporated version of Debian, just as user/shareholder-driven, but with a better installer and less politics. (/me grins, ducks flames...)
First of all, to all the people whining about the GPL: No one is making you use GPL'ed code. If you don't like it, use BSD-licensed code, or buy proprietary software!
In fact its is more demanding than any closed source licence.
Bwa ha ha ha. With the GPL you can view the code, use the code any way you want, and change it in any way you want, and redistribute it freely.
And that's "more demanding" than, say, Microsoft's license, where you can't view the code, can't use the source code, can't even use the binaries in some ways(!), can't change the code, and can't redistribute it freely?
Or if you are part of the "shared source" club, maybe you can view the code. But you still can't change it, use it, or redistribute it.
Or maybe you sit down and negotiate a source code license from some software company with a proprietary license. (That might take a month or two.) Now you can view it and use it and maybe change it, in some limited and restricted ways. You will have to pay, of course... probably a lot.
"But", you cry, "Now I have the freedom to keep my changes secret!" Well guess what, you can do exactly the same thing with GPL'ed code: Go negotiate a secondary, non-GPL source code license with the developer! Exactly like you would with a closed-source license!
"But", you whine, "The developers of the GPL'ed code don't want to do that, or they want a lot of money for an alternate license!"
Well boo hoo! If you don't like the terms, write it yourself! Or go talk to Microsoft and see what cheap, easy, friendly terms they give you for source code access. Don't be ridiculous. "More demanding than any closed source license". Yeah right!
>I'm all for the creators of software controling >what happens with it but the GPL rubs me the >wrong way when it seeks to control other works >that the original author did not create.
That is a self-contradictory statement. Creators of software who release it under the GPL are simply exerting some minimal control over what happens to it, which you say you have no problem with. Now, since the GPL does not seek to control _independent_ works that have nothing to do with the GPL'ed code, when you say that "the GPL rubs you the wrong way" you must not like that it has some minimal control over changes to the original work by the author. The only control is that the modified source must continue to be available. But that's the WHOLE POINT of the GPL, and that's why the authors released it that way... which you say you don't have a problem with. So which is it?
To summarize: Everyone who complains about the GPL either:
1. Does not understand it, or,
2. Wants to leech from the Free Software community by taking and not giving.
You are correct - the reason that the kernel is not specified is that applications do not (normally) depend on the kernel. Perhaps a better name for the Linux Standard Base would be the "Gnu + X11 + Filesystem Layout Standard Base" but lets be realistic, it's going to be Linux kernels in there for a while. On the other hand, a specification like this should make it easier for alternate kernels to be accepted - if the Hurd can be used to make an LSB-compliant system, that should accelerate it's acceptance.
The point of the LSB is so that big companies can write applications and installers that will run and install on any LSB-compliant system. This is a good thing, otherwise we will continue to get applications that are "supported on Red Hat Linux 6.2" only.
The key parts to the specification seem to be the Gnu C Library (although a different C library could probably be made to work) and X11. I am happy to see that an OpenGL library is part of the spec! This should make it much easier for 3D games to work on any Linux system without all the messing around that you have to go through now.
Hopefully in the next version they will specify:
1. Fonts - how to use them, where to keep them.
2. X Extensions - X Render, for example
3. Audio
And maybe in future versions they can nail down the locations of configuration files. Beware of the inevitable holy war over init scripts... Probably the best thing about this spec is that it will make management happier about Linux, and make it harder for Microsoft to FUD about Linux fragementing.
Imagine that the EFF and 2600 lose the DeCSS case all the way up to the supreme court. Could happen. But since DeCSS is everywhere, it can't be stomped out. Imagine that very similar things happen with CD audio, SDMI, etc.
Further imagine that some serious disasters happen due to software bugs. As software gets to be in more and more places, there will be more and more ways for software bugs to kill and hurt people. What if some congress person stood up in 2003 and said "Bad software is killing people, and illegal software like deCSS is hurting the economy. We have to do something! Think of the children!"
Government and MS will propose a solution: To write software, you will have to get a license. You have to join an association of software developers, and get liability insurance, so that if something goes wrong with your software, you can be sued. This is very similar to what doctors and lawyers have to do.
That would hurt development of free software in the US, but other countries would still continue. So they would have to make a law that you are only allowed to run software developed by a licensed, insured software practicioner. Note that they could also use technological means to achieve this - what if new computers started coming out that used encryption techniques so that they could only boot operating systems that were digitally signed with a "Licensed Software Developer" key. Of course the DMCA would make it illegal to hack around this. Red Hat could get Linux signed, but continued development would come to a screeching halt because all the evening and weekend coders wouldn't be able to make patches and test them.
All that has to happen for this to become possible is for the courts to continue to decide that source code is NOT speech, it is a device, and can be restricted, and that reverse engineering can be eliminated because of the DMCA. After all, they will say, we have restrictions on all sorts of other devices - cars and drivers have licenses, you can't build your own car or airplane and fly it without getting licenses, why shouldn't you have to get a license to write and run software?
I do pray that it never comes to this, or at least that there would be some free country I could move too. I actually believe there would be a revolt before it got that bad, but who knows? Too many people are like sheep.
The "meaning" of "Information wants to be free" is really just an observation that it is easy to spread and copy information, and hard to prevent the spread of information.
It's really hard to keep a secret, or to prevent people from copying MP3's, or to keep people from mirroring DeCSS, or to put the Hacking SDMI paper by Felten and the rest of 'em back in the bottle now that it's out there.
Some people think that's an over optimistic view of things though, and that if governments and big companies really wanted to, they could clamp down pretty well - firewall every ISP, require a license for an internet connection, invent stuff like SDMI and CSS, and then pass laws like the DMCA to back it up. Heaven forbid.
It is fitting that from now on, Greylock and General Atlantic will be known as "Oh yeah, you're the venture capitalists that took Ars Digita, a successful, popular, profitable company and wrecked it in just one year!"
What a way to get a reputation.
I've been a fan of Philip Greenspun since way back in '98 when I read his guide to web publishing. That convinced me I had better learn server-side web programming, and I'm so glad I did, although I've never used the ACS system.
"Microsoft Windows Update has detected unsupported operating system partitions on your hard disk. As unsupported operating systems are known to be "lousy" and "anti-american" and "destroy intellectual property", it will be upgraded to NTFS, giving you 34 GB of free hard disk space! Please wait....".
Upgrade complete! Click OK to reboot your computer...
Hey, I agree with you... I think all this stuff is horrible and everyone involved should be ashamed of themselves. Governments should make it illegal to patent genes, and illegal to make sterile plants.
But, I was just talking about the legal issue regarding Monsanto and that Canadian farmer. It's ridiculous - imagine if Monsanto (secretly) flew a plane all over Saskatchewan, scattering just enough of their seeds out the back to contaminate every field they crossed.
Next year, they go after every farmer in the province. Pay up or else! And the farmers might just decide, "Hell, if I'm paying for the stuff anyway, I might as well grow it and get the (minor) benefits".
And so Monsanto takes over the market. It's awful. Those people should be dragged out into the street and shot.
But at LEAST they should be told: If you don't like it, make your plants sterile. That way they can't play the game both ways, ripping off people whos crops get contaminated.
The argument is, if Monsanto wants to have absolute control over who gets to grow the stuff, they should make the plants sterile so they cannot contaminate other people's crops.
This is my worry as well - it is clear that a major "feature" (cough, spit) of Windows XP is that it integrates copy protection technology directly into the operating system.
I don't think we will ever reach the point where you cannot buy a computer that can run Linux, but maybe, I worry, we might end up in a situation where "Linux" compatible computers are expensive, server type machines, and all the cheaper consumer hardware is locked down with Windows-only BIOSes. Linux might still be able to run on the machine, but only by violating the DMCA or something. Even worse, since Linux-compatible machines would be expensive, rare and "dangerous", the government might start requiring a license for them like cars or guns.
You know it's time to leave the US when...
In that nightmare version of the world, all the inexpensive computers are basically turned into non-programmable appliances. That is happening a little bit already anyway - computer hardware is becoming so cheap that it is being used in appliances like set top boxes, and it soon won't be unusual for people to own several single purpose machines... a set top box that plays digital TV, an "email and web surfing" machine that only can email and surf the web, a "secure music" CD player that hooks up to your stereo system... and all of them, inside, will be pretty much normal PC hardware just because that's the cheapest way to build them.
Personally, I have been buying a lot more CD's recently... I want to build up my collection to the point where if the industry stops selling regular CD's I can just turn my back on them and say "Screw you, RIAA, I already have ALL the music I'll ever want, and I never need to buy another CD again."
Slashdot Mirror
All of the above is true, however... the new driver may not work for everyone.
/people/alan on the kernel mirrors) or, if you know what you're doing, just take the relevant emu10k files from 2.4.7 or so and patch them into your tree. Alan Cox's kernels have lots and lots of other changes besides the SBLive driver, of course.
/linux/REPORTING-BUGS file) and send a mail to the list. Don't hold your breath waiting for a reply, but it will be read, and it might help the developers figure out the pattern of what situations cause problems and what works well.
Alan Cox, for example, tried them out and found they didn't work on some of his machines, so he didn't include the new driver in the -ac patches.
So, it seems they work great for some people but not others. Linus decided to keep them in because the new driver is being maintained, and the old one wasn't - a buggy driver that's being worked on is better than a buggy driver that's being ignored... and the new drivers have much better support for the features of the SBLive.
Anyway, if you upgrade and find that your sound card quits working, you have several choices - revert to your previous working version, or try Alan Cox'es patched version of the kernel (look under
A lot of work is being done right now on getting the VM (Virtual Memory) balanced right. It works great for a lot of people, but depending on the workload, it can start to thrash really bad and swap itself to death. (kswapd will run, taking all the CPU time and making no progress.) If you run into this problem or others, read the documentation on reporting bugs (see the
(If you report bugs on the Linux Kernel Mailing list, be sure to note what kernel you're running and any non-standard changes you've made.)
Oh, and if you are not subscribed to the list and want to be cc'ed on replies, say so at the top of your email.
(I'm not really a kernel developer, yet, but I hack around in the code and read the list, so I know a little about what's going on.)
I know what would get worms back into the media for a long time - a Warhol Worm. You want to read something scary about worms, go read that. Be sure to read the section "A Worst Case Warhol Worm". It gives me the shivers to think about it.
From the article: "A worst case Warhol Worm is truly frightening, capable of doing many billions of dollars in real damage and disruption. Since it can achieve complete spread in well under an hour, and could begin doing damage immediately on infecting a machine, human mediated responses offer almost no hope of stopping it. "
Complete spread in under an hour! Total destruction of infected servers!
Whee!
Watch for one of these coming out with the next major IIS exploit.
Nitpick: there are methods of public key encryption that don't depend on the difficulty of factoring products of large primes for their security. Elliptical curves, for instance.
And of course, public key / private key is a major convenience, but all you really need is old school secret key cryptography to make a private, secure network.
Finally, there are encryption methods that rely on the conjecture that P != NP - that is, breaking the encryption would answer the most significant unanswered question in mathematics of the last hundred years. I feel pretty safe about that. Oh yeah, and quantum cryptography too.
But anyway, yes, it is a good thing that there are some things that nobody knows how to break.
Actually, you can (still?) buy "pro" audio DA / AD gear with no restrictions. For example, check out M-audio's pro-grade sound cards. There's even Linux drivers.
And they aren't too expensive. The basic "Audiophile 2496" card has 24 bit, 96 Khz DA and AD converters, SPDIF digital IO, and MIDI. Even better, the SPDIF connectors on that board give the user full control over the SCMS (aka "scums") serial copy management system. The
Sound Blaster Live's SPDIF, on the other hand, won't let you copy SCMS-protected content, even if you own the DVD it came from. Apparently this is normal for "consumer" grade stuff.
Get affordable "pro" gear while you still can!
Turing machines with attached "oracles" are used for proofs in theoretical computer science, but it's important to keep in mind that they are purely theoretical - no one will ever build an oracle. You don't think about how an "oracle" works. It's just a concept, like imagining that magic works. So why is this useful?
:-)
Well... For instance, you might be able to prove that such-and-such a problem with input size "n" could be solved in polynomial time (i.e. "fast") if you just had a magical oracle to supply you with only log(n)correct, one-bit answers.
The point of the proof would be that you don't need more than log(n) magic bits from the oracle. So what good is that? Well... If you can get the number of magic bits small enough, while still keeping the algorithm fast, it may provide a way to do a randomized algorithm where instead of using the magical oracle (which doesn't exist, remember?) you just use a random number generator, or maybe just try everything. Since a random number generator isn't as accurate as a magical oracle, you run the algorithm a lot of times with different random bits. Maybe you'll be lucky soon, and maybe that's good enough on average.
Theoretical computer science is fun. It's a little crazy and non-intuitive somtimes.
Torrey Hoffman (Azog)
Yes! "Advanced Programming in the Unix Environment", which is part of the same series from Richard Stevens, is my trusted companion as a C programmer making the escape from Microsoft to Linux.
And it's not Linux specific - it's pretty much POSIX, and does talk about some of the differences between implementations where they matter.
An excellent and comprehensive set of books.
Torrey Hoffman (Azog)
I memorized pi to 50 places the summer between grade 10 and 11 just so I could be a smartass in math and physics classes the next year.
It's actually not hard to do. Break it up into groups of 5 digits, and memorize 5 digits a day and keep practicing. A few minutes a day, and you can be a loser^H^H^H^H^H, er, cool geek like me too.
I wish I could get into the real story, but it's slashdotted, so I'm posting pointless crap like this. Geez.
Torrey Hoffman (Azog)
One obvious way to do it: Get a really, really good CD player which has top-notch D/A converters, a good transport mechanism, and very high quality analog circuitry. (From what other posters have written here, you _have_ to use the D/A converters in the CD player, as that's the only way to get all the information from the ECC bits.)
Then hook the analog output of that CD player to a decent outboard A/D converter using good cables. The outboard A/D converter feeds the digital result to your computer through an SPDIF cable, and there you go... The only headache is that you don't get the automatic CDDB lookup and track numbering.
Remember, since you're going to be compressing the result to OGG or MP3 anyway, the small loss in sound quality that you have by going through the CD player's D/A and the outboard A/D converter is essentially irrelevant. Especially if you're going to play it through computer speakers, which are all pretty lousy anyway.
In fact, with a good CD player and A/D equipment, the resulting compressed audio would sound better than your average 128 bit Napster crap.
Torrey Hoffman (Azog)
Attention, companies and advertising agencies that might be reading this!
I make a note of companies that have irritating ads, and I will NEVER buy from them!
For example, those damn X10 popup ads which invariably feature women in suggestive poses... I'll never, ever buy anything from that company. And that's despite the fact that I'm a geek with a house full of electronics, and wireless cameras are right up my alley. Forget it. Ain't gonna happen. Same with any other company that gets on my nerves.
If I wanted annoying advertising, I would watch TV. And I gave that up completely three years ago.
Torrey Hoffman (Azog)
I'll have to get an "us" domain. There's too many great possibilities to pass up.
For instance, a great second level domain would be "with.us". Then you could resell all the commercially useful third level domains: bank.with.us, shop.with.us, save.with.us, fly.with.us.
But there's loads of other great possibilities. You read them here first:
x.the.us (rebuild.the.us, destroy.the.us, network.the.us, educate.the.us, subvert.the.us, love.the.us, hate.the.us, fuck.the.us)
x.is.us / x.is.not.us
x.r.us (but toys.r.us will get you slapped for trademark violation.)
x.for.us (linux.for.us, beer.for.us, games.for.us, cars.for.us, software.for.us)
why.not.us, remember.us, its.on.us, kill.us, dont.forget.us, buy.us, fly.us, rent.us, sell.us, silly.us, sucks.to.be.us, computers.for.the.rest.of.us
Quick! Rush out and reserve your us domain today!
But which is better? free.beer.for.us or free.speech.for.us?
Torrey Hoffman (Azog)
No, the GPL does not allow _linking_, but linking is NOT the same thing as communicating via some sort of message passing.
.so (linking) and sending messages down a pipe, or using a shared file.
... or any of the other closed-source apps that are really nice to have on Linux.
On some architectures, that line can get blurry, but under Windows and Linux it's pretty clear.
It's the difference between calling a function in a DLL or
This is NOT a loophole that you would want to close. Suppoose you did, and you created an operating system and libraries that used your "improved" license. No one would even be able to RUN closed source software on it, because running a program under Unix involves _exactly_ the same mechanisms that you would have "closed". So say goodbye to Oracle or Quake 3 or Wordperfect or ViaVoice
The GPL strikes a very, very fine balance between the rights of users to have control over their software, and the ability of programmers to write closed source software and still run it on an open system.
It is true that things can become blurry. On some "message-passing" style operating systems, there's less of a distinction between a function call and interprocess communication. Luckily, that's not really an issue for any popular OS right now.
Torrey Hoffman (Azog)
Exactly! If someone is going to sell hardware at a loss, that's their problem.
This is about to become very interesting with Microsoft's X-box coming out. Here we have a high-powered PC, with top-notch graphics, a P-III 733Mhz, 64MB of RAM (or is it 128?) plus DVD, 100MBit ethernet, and other useful stuff. It will be sold by Microsoft for at least $100 dollars below cost.
Now, what's to stop people from porting Linux to it and using it as a home computer? Or a cheap network server? Maybe it doesn't have VGA output?
Maybe Microsoft will try to write the contract / purchase agreement to make that illegal. I doubt that will work, just have a minor buy the hardware, the contract will not be legally binding then.
But since Microsoft intends to make the money back by licensing software development kits, they have to maintain firm control over who can create software for the box.
My expectation is that the BIOS of the X-Box will refuse to boot from any device which does not have a Microsoft-copyrighted signature in the boot sector. That doesn't mean that people won't be able to hack it, but it does mean that anyone that *distributes* software for it without a license from Microsoft can be sued for copyright violation.
It will keep hacks private and non-commercial at least. Will that be enough? After all, if someone puts an ISO image of an X-Box bootable Linux distro on the net, missing only the Microsoft signature, and the same web page has instructions on how to copy the Microsoft signature off your copy of "Halo" to make the Linux disk bootable... I'm sure Microsoft will send out the attack dog lawyers.
Getting back on point.... Netpliance could have done the same things to make sure the device only was used the way they want. But they didn't. Tough for them.
Torrey Hoffman (Azog)
Or do you get paid to go to university? If so, then it would make more sense that they own (or, at least have rights to) the work you do there.
Torrey Hoffman (Azog)
Yes, but it will be illegal for you to access it. ;-)
Meanwhile, Linus says that "real men don't make backups - they just put their source code out on the internet and let everyone mirror it".
Torrey Hoffman (Azog)
Well, if they did this, I might buy a couple hundred dollars worth. Not that I would expect to make money anytime soon, or maybe ever, but because I like Mandrake.
The reason it's different from donations is that presumably, as a stockholder, I would have some input into the direction of the company. Maybe they could have shareholder voting on company direction questions over the internet - one vote per share... like, should we make Mozilla the default browser in Mandrake 8.1? Should we continue shipping proprietary software in the boxed version? Should we hire more developers? What should the developers work on? Of course, as a private company the CEO is ultimately responsible, and shareholders can really just vote to replace the CEO and board. Or something, I'm no expert on corparate law, especially for a French company.
But anyway, my point is that it would not be just a donation to Mandrake.
If this is really successfull, Mandrake might end up kind of like an incorporated version of Debian, just as user/shareholder-driven, but with a better installer and less politics. (/me grins, ducks flames...)
Torrey Hoffman (Azog)
Bwa ha ha ha. With the GPL you can view the code, use the code any way you want, and change it in any way you want, and redistribute it freely.
And that's "more demanding" than, say, Microsoft's license, where you can't view the code, can't use the source code, can't even use the binaries in some ways(!), can't change the code, and can't redistribute it freely?
Or if you are part of the "shared source" club, maybe you can view the code. But you still can't change it, use it, or redistribute it.
Or maybe you sit down and negotiate a source code license from some software company with a proprietary license. (That might take a month or two.) Now you can view it and use it and maybe change it, in some limited and restricted ways. You will have to pay, of course... probably a lot.
"But", you cry, "Now I have the freedom to keep my changes secret!" Well guess what, you can do exactly the same thing with GPL'ed code: Go negotiate a secondary, non-GPL source code license with the developer! Exactly like you would with a closed-source license!
"But", you whine, "The developers of the GPL'ed code don't want to do that, or they want a lot of money for an alternate license!"
Well boo hoo! If you don't like the terms, write it yourself! Or go talk to Microsoft and see what cheap, easy, friendly terms they give you for source code access. Don't be ridiculous. "More demanding than any closed source license". Yeah right!
>I'm all for the creators of software controling >what happens with it but the GPL rubs me the >wrong way when it seeks to control other works >that the original author did not create.
That is a self-contradictory statement. Creators of software who release it under the GPL are simply exerting some minimal control over what happens to it, which you say you have no problem with. Now, since the GPL does not seek to control _independent_ works that have nothing to do with the GPL'ed code, when you say that "the GPL rubs you the wrong way" you must not like that it has some minimal control over changes to the original work by the author. The only control is that the modified source must continue to be available. But that's the WHOLE POINT of the GPL, and that's why the authors released it that way... which you say you don't have a problem with. So which is it?
To summarize: Everyone who complains about the GPL either:
1. Does not understand it, or,
2. Wants to leech from the Free Software community by taking and not giving.
I have yet to see a counterexample.
Torrey Hoffman (Azog)
You are correct - the reason that the kernel is not specified is that applications do not (normally) depend on the kernel. Perhaps a better name for the Linux Standard Base would be the "Gnu + X11 + Filesystem Layout Standard Base" but lets be realistic, it's going to be Linux kernels in there for a while. On the other hand, a specification like this should make it easier for alternate kernels to be accepted - if the Hurd can be used to make an LSB-compliant system, that should accelerate it's acceptance.
The point of the LSB is so that big companies can write applications and installers that will run and install on any LSB-compliant system. This is a good thing, otherwise we will continue to get applications that are "supported on Red Hat Linux 6.2" only.
The key parts to the specification seem to be the Gnu C Library (although a different C library could probably be made to work) and X11. I am happy to see that an OpenGL library is part of the spec! This should make it much easier for 3D games to work on any Linux system without all the messing around that you have to go through now.
Hopefully in the next version they will specify:
1. Fonts - how to use them, where to keep them.
2. X Extensions - X Render, for example
3. Audio
And maybe in future versions they can nail down the locations of configuration files. Beware of the inevitable holy war over init scripts... Probably the best thing about this spec is that it will make management happier about Linux, and make it harder for Microsoft to FUD about Linux fragementing.
Torrey Hoffman (Azog)
How can free software be made illegal?
How about this:
Imagine that the EFF and 2600 lose the DeCSS case all the way up to the supreme court. Could happen. But since DeCSS is everywhere, it can't be stomped out. Imagine that very similar things happen with CD audio, SDMI, etc.
Further imagine that some serious disasters happen due to software bugs. As software gets to be in more and more places, there will be more and more ways for software bugs to kill and hurt people. What if some congress person stood up in 2003 and said "Bad software is killing people, and illegal software like deCSS is hurting the economy. We have to do something! Think of the children!"
Government and MS will propose a solution: To write software, you will have to get a license. You have to join an association of software developers, and get liability insurance, so that if something goes wrong with your software, you can be sued. This is very similar to what doctors and lawyers have to do.
That would hurt development of free software in the US, but other countries would still continue. So they would have to make a law that you are only allowed to run software developed by a licensed, insured software practicioner. Note that they could also use technological means to achieve this - what if new computers started coming out that used encryption techniques so that they could only boot operating systems that were digitally signed with a "Licensed Software Developer" key. Of course the DMCA would make it illegal to hack around this. Red Hat could get Linux signed, but continued development would come to a screeching halt because all the evening and weekend coders wouldn't be able to make patches and test them.
All that has to happen for this to become possible is for the courts to continue to decide that source code is NOT speech, it is a device, and can be restricted, and that reverse engineering can be eliminated because of the DMCA. After all, they will say, we have restrictions on all sorts of other devices - cars and drivers have licenses, you can't build your own car or airplane and fly it without getting licenses, why shouldn't you have to get a license to write and run software?
I do pray that it never comes to this, or at least that there would be some free country I could move too. I actually believe there would be a revolt before it got that bad, but who knows? Too many people are like sheep.
Torrey Hoffman (Azog)
Just to answer your question...
The "meaning" of "Information wants to be free" is really just an observation that it is easy to spread and copy information, and hard to prevent the spread of information.
It's really hard to keep a secret, or to prevent people from copying MP3's, or to keep people from mirroring DeCSS, or to put the Hacking SDMI paper by Felten and the rest of 'em back in the bottle now that it's out there.
Some people think that's an over optimistic view of things though, and that if governments and big companies really wanted to, they could clamp down pretty well - firewall every ISP, require a license for an internet connection, invent stuff like SDMI and CSS, and then pass laws like the DMCA to back it up. Heaven forbid.
Whatever. That's it's supposed to mean, anyway.
Torrey Hoffman (Azog)
It is fitting that from now on, Greylock and General Atlantic will be known as "Oh yeah, you're the venture capitalists that took Ars Digita, a successful, popular, profitable company and wrecked it in just one year!"
What a way to get a reputation.
I've been a fan of Philip Greenspun since way back in '98 when I read his guide to web publishing. That convinced me I had better learn server-side web programming, and I'm so glad I did, although I've never used the ACS system.
Torrey Hoffman (Azog)
That's a disappointment.
It obviously isn't intelligent life...
Torrey Hoffman (Azog)
Heh...
"Microsoft Windows Update has detected unsupported operating system partitions on your hard disk. As unsupported operating systems are known to be "lousy" and "anti-american" and "destroy intellectual property", it will be upgraded to NTFS, giving you 34 GB of free hard disk space! Please wait....".
Upgrade complete! Click OK to reboot your computer...
Torrey Hoffman (Azog)
Hey, I agree with you... I think all this stuff is horrible and everyone involved should be ashamed of themselves. Governments should make it illegal to patent genes, and illegal to make sterile plants.
But, I was just talking about the legal issue regarding Monsanto and that Canadian farmer. It's ridiculous - imagine if Monsanto (secretly) flew a plane all over Saskatchewan, scattering just enough of their seeds out the back to contaminate every field they crossed.
Next year, they go after every farmer in the province. Pay up or else! And the farmers might just decide, "Hell, if I'm paying for the stuff anyway, I might as well grow it and get the (minor) benefits".
And so Monsanto takes over the market. It's awful. Those people should be dragged out into the street and shot.
But at LEAST they should be told: If you don't like it, make your plants sterile. That way they can't play the game both ways, ripping off people whos crops get contaminated.
Torrey Hoffman (Azog)
Sterile seeds.
The argument is, if Monsanto wants to have absolute control over who gets to grow the stuff, they should make the plants sterile so they cannot contaminate other people's crops.
Torrey Hoffman (Azog)
This is my worry as well - it is clear that a major "feature" (cough, spit) of Windows XP is that it integrates copy protection technology directly into the operating system.
I don't think we will ever reach the point where you cannot buy a computer that can run Linux, but maybe, I worry, we might end up in a situation where "Linux" compatible computers are expensive, server type machines, and all the cheaper consumer hardware is locked down with Windows-only BIOSes. Linux might still be able to run on the machine, but only by violating the DMCA or something. Even worse, since Linux-compatible machines would be expensive, rare and "dangerous", the government might start requiring a license for them like cars or guns.
You know it's time to leave the US when...
In that nightmare version of the world, all the inexpensive computers are basically turned into non-programmable appliances. That is happening a little bit already anyway - computer hardware is becoming so cheap that it is being used in appliances like set top boxes, and it soon won't be unusual for people to own several single purpose machines... a set top box that plays digital TV, an "email and web surfing" machine that only can email and surf the web, a "secure music" CD player that hooks up to your stereo system... and all of them, inside, will be pretty much normal PC hardware just because that's the cheapest way to build them.
Personally, I have been buying a lot more CD's recently... I want to build up my collection to the point where if the industry stops selling regular CD's I can just turn my back on them and say "Screw you, RIAA, I already have ALL the music I'll ever want, and I never need to buy another CD again."
Torrey Hoffman (Azog)