It's a very simple disconnect. In the first case, the GPL gives you the freedom to charge money for the program - this is what allows vendors like CheapBytes to sell Linux distribution CDs composed entirely of free software and turn a profit.
What is disallowed is a viral licensing fee, like that used by shareware authors ("You can try this program and redistribute it, but if you use it for more than 30 days, you owe me $20."). You can't require that anyone who receives the software (not necessarily from you!) has to pay you money. The difference is that you might agree only to send the software to someone if they pay you money (CheapBytes would quickly go out of business if they were required to give away CDs for free), but you can't control what they do with the software afterwards. Imagine what would happen if CheapBytes required everyone who redistributed code from the CDs they sell to pay them a fee for the privlege of doing so!
The GPL does not enforce that software be available at no cost, but the market forces inherent in free software do (would you buy a Red Hat CD for $50, without any manuals or support, if you could get the same for about five bucks with shipping or download it off the Net for pennies?)
The GPL contains language designed to address this concern:
5. You are not required to accept this License, since you have not
signed it. However, nothing else grants you permission to modify or
distribute the Program or its derivative works. These actions are
prohibited by law if you do not accept this License. Therefore, by
modifying or distributing the Program (or any work based on the
Program), you indicate your acceptance of this License to do so, and
all its terms and conditions for copying, distributing or modifying
the Program or works based on it.
The GPL isn't a license to *use* software, it's a license to modify and redistribute software. So using GPL'd software doesn't commit you to following the GPL, because the GPL doesn't cover the act of running the program. If, however, you copy or modify GPL'd code, the only way you can legally do so is by following the demands of the GPL.
Of course, this only applies to licenses that are a grant of copyright rights like the GPL. Licenses that take away rights (such as the "no reverse engineering" clause in nearly every piece of commercial software) are another story; this is why the industry is trying so hard to get UCITA passed.
I'm in the process of moving to a new apartment right now, and I've been setting up utilities. Phone and power went like a breeze. The customer service reps (with whom I was able to speak to nearly immediately) took down my name, address, social security number, asked me for a date of activation, and then thanked me and hung up.
That was easy. The local cable modem company (RR) is that easy, too; I called to ask how long an installation took and the rep seemed surprised that I didn't want an install in the next couple of days. This is service, people.
I decided to go with DSL, instead, because there was *one* ISP doing business in town that I could get a static IP from for a reasonable price (Southwestern Bell will give me five static IPs for $80/mo, or one dynamic for $40/mo, but nothing in between - does this make any sense?)
I'm paying the cost not in money, but in time. A month without broadband will be a long time; if it weren't for the compelling feature set and the fact that I know I'll be stuck with whatever I'm getting now for a year, I'd be jumping ship to cable in a heartbeat.
Why on earth does DSL take so long to set up? Is there an order backlog a mile long? Does the phone company introduce some sort of mandatory delay into the process?
This just happened about an hour and a half ago.
I'm sitting in Rice's COMP 314 programming class, taught by Dan Wallach, one of the authors of the paper. He's spent the first half of class giving us the rundown on his predicament, and moves on to the lecture topic for the second half of class.
In the middle of the lecture, something like this transpires (paraphrased):
"And so you see that there can be occurences when...oh, here's an occurence. My phone's ringing."
[answers his phone] "Hello?"
"Actually, I can't talk right now. I'm sort of teaching a class."
[class laughs] "Yeah, you can hear them laughing in the background?"
"Okay, I'll be in my office around 4."
[hangs up phone] [to class] "That was John Markoff from the New York Times. He wants to have a chat with me."
[resumes lecture seamlessly]
Speaking as a Matrox owner: At least on the G400, the HAL library is not required for 3D acceleration. It is required for using the dualhead features of the card.
It's possible that HALlib is required for the G450, but I'm not sure.
That's not true at all - the "Open Source" movement was largely intended to put a happy corporate face on free software. The two represent basically the same idea; the reason you see RMS griping about the difference is likely twofold:
He pioneered the idea of free software, under the moniker of "free software," and very likely feels upset that those ideals have been usurped by this new movement with a new name (this motivation to have his contributions recognized in name can also be found in his urgings that the OS is called "GNU/Linux," not just "Linux");
He fears that, by removing the word "free" from the name, we're also removing the concept of freedom from our thoughts about the subject. Freedom of software is important to him and he wants the users of free software to be reminded that they are excercising their freedom by using it. "Open Source," in his mind, dilutes that association in users' minds.
There isn't much difference, beyond the symbolic, between free and open source software. Microsoft's latest overture to allow its biggest customers to view the source code to some of its products does not qualify as open source by any stretch of the imagination - they're forbidding modification, redistribution, even compilation of the revealed code. And open source software cannot be bound by "oppressive EULA's" - see Term 7 of the Open Source Definition, which forbids the execution of additional licenses. Keep in mind that the OSD derives, nearly directly, from the
Debian Free Software Guidelines.
The difference is a difference of name; a kernel by any other name would smell just as sweet.
If I recall correctly, the timespan of a copyrighted work originated by a corporation (a "work for hire" like a Hollywood movie) is now 90 years, recently extended by the Sonny Bono Act from 70. The fate of the company has no bearing on the duration of copyright.
One of the
Debian Free Software Guidelines is "License Must Not Be Specific to Debian." So, no, this piece of software will not be allowed into the Debian system unless and until SuSe decides to relicense it in a distribution-agnostic manner. Remember, though, that the Debian mirrors contain a 'non-free' branch which, while not technically part of the Debian system, allows Debian users to easily install such packages. The net result is that Debian users who are willing to accept the license will probably have an easy way to install the software at the end of February, but it won't be on any "official Debian" CDs for a while after that.
This release has simply blown me away. It's fast - almost as fast as Netscape 4, and completely tolerable on this pII/233 (I can't wait to try it out on my new dual 733...)
And I noticed one sort of odd thing: it hasn't crashed yet on me. When I started trying Mozilla the thing blew up all the time; now I feel completely comfortable with the idea of Mozilla as my primary browser, particularly with the integrated (and free!) crypto.
The only edge Konqueror has over Mozilla now, in my opinion, is being based on QT - which is the only toolkit so far that's been patched to use the excellent Xft
library for antialiased fonts under XFree86 4.0.2 and later. I can't wait for Mozilla to pick up support for this thing.
I certainly could be misinformed. I remember reading an explanation of the RSA patent that went something like that, but it was written by a cryptographer, not a legal expert.
So what is the status of a US patent internationally? Are there any treaties that give patents international scope?
Normally, U.S. patents are valid in most other countries due to international treaty. The RSA patent was a special exception: the creators of the algorithm published before filing for a patent. In most countries (this could either be due to common patent law or to the terms of the treaty; IANAL), that creates prior art and invalidates the patent, but U.S. patent law allows a one-year grace period after first publishing to apply for a patent. When the patent was awared, it became valid in the United States, but not in other countries.
Most patents awarded in the US are recognized by other countries.
She refused to extend the turnin deadline at the
request of Palm Beach County; the county had not
completed its recount by 5 p.m. EST today, which
was the deadline set by the Florida Supreme Court
when it ruled for Gore last week. Palm Beach's
hand recount updates were, therefore, not included
in the final Florida tally, but a running total
showed Gore having picked up only fifty-odd votes there.
There are still plenty of issues outstanding, though:
The United States Supreme Court hears the case about hand recounts on Friday; they could rule that the deadline be extended and Palm Beach be allowed to continue counting, they could rule that the hand recounts as performed were illegal, or they could affirm the Florida high court's decision. (I personally think the last is their most likely course; elections are traditionally the jurisdiction of the states.)
There are still legal cases brewing about the confusing "butterfly ballots" which probably caused several thousand elderly Palm Beach residents to cast their vote for Buchanan, when they likely intended to vote for Gore. This case is a wild card; a liberal ruling here could well tip the election to Gore.
Gore's campaign can (and probably will) officially contest the election in Florida. The reason the Supreme Court set the cutoff date for hand recounts so early was to give the contesting procedure time to go through. They've got several angles to go by here, including the Palm Beach deadline and the decision by Miami/Dade to stop recounting.
Re:quantum security and the new elite
on
Quantum Security
·
· Score: 3
It isn't as bad as all that. According to the article, a quantum codebreaking machine will have to perform a computation of order O(sqrt(n)), where n is the number of possible keys, in order to solve the problem. Classical brute-force searches are, of course, O(n).
The net effect is that a quantum computer in the hands of an eavesdropper halves the effective keylength - a 128-bit key is reduced to 64 bits of effectiveness. 64 bits is, of course, not enough security to defend against government-level surveillance resources, but all that has to be done to solve the problem is to increase the keylength to 256 bits.
One of the requirements for the AES candidates was that the algorithm support 256-bit operation. Rijndael, the heir apparent to DES, does support 256-bit operation modes.
I'm so glad my school doesn't buy into the silly homecoming court pageants that happen at high schools. Our King and Queen elections have been tongue-in-cheek for the last thirty years or so; this year, the King was the new chef at one of the cafeterias (anyone who makes good dorm food can be King:) and the Queen was the new (male) traffic officer who has been ruling the parking lots with an iron fist for the past few weeks.
Another point is that the average joe may not be authorized to know the information that goes into the politician's decision-making process. Presidential administrations, for example, cannot fully justify their foreign policy decisions to the American people, because those decisions may be motivated by classified information (or the desired result of the decision needs to remain a secret.) A nation must keep its secrets, so unfortunately, the voting populace can't learn exactly what those secrets are.
The end result of this is that you have to elect men and women whose independent judgement you trust, whose principles parallel yours, and whom you trust not to abuse the privleged position they're given. You must, in short, choose someone with good character. Good luck finding that person.
Assuming our goal is more freedom for users, let's analyze the tradeoff involved in writing a closed-source, non-free abstraction layer for the PS2 API:
Cost: One more piece of closed, non-free software is created. A developer who normally writes open, free software must write this. Users are still deprived (as they were before) of knowledge about the PS2 API. Users can freely share, but cannot easily learn from or change, this piece of software. (Whether this is depriving the users of their rights, or simply not granting them a right they didn't have in the first place is largely an issue of perspective.)
Benefit: An entire computing platform is opened so that free, open software can be written for it, whereas it was impossible to write such software in the past. Many existing pieces of free software can be immediately used on the PS2.
I think it's fairly clear that the benefits outweigh the costs (if these costs can be said to exist at all.)
Unfortunately, it's not in Sony's interests to disclose their secret API to someone who will make it possible for free software to work on the PS2. Their whole business model is about taking a significant loss on the hardware, then making that loss up in license fees from the sale of software. If there were lots of free software available for the PS2, sales of console games would dip and cut into Sony's revenue stream.
There *are* ways to enforce the loss-leader model. You can patent the interface between the nominal "razor" and "blade" (or even the actual razor and blade - ever wonder why every Gilette product has a different attachment setup?).
You can require, via clearly stated contractual obligation, that those who purchase the razor must also purchase a sufficient quantity of blades (this is the arrangement Netpliance finally and correctly reached, after weeks of corporate bumbling and stumbling that probably cost them a real shot at their market.)
You can loan the razor to the consumer, and make the purchase of blades a condition of that loan. The fact that there is a loan, with conditions attached, must again be clearly stated.
DigitalConvergence appears to be trying to use the third method of enforcement. According to their business plan, they still own all of the CueCat devices; however, they haven't done a very good job of making this clear to people. Radio Shack hands these things over the counter without ever mentioning any conditions. They send the things out to Forbes subscribers, possibly running into interesting postal law that makes those scanners the property of the recipients. The notion of a loan with conditions is hidden deep in the fine print in the package. Why not be forthright? Because DigitalCovergence is getting and correlating information about everything a CueCat user scans, and it's best if that ugly fact is hidden as deeply as possible.
In this case, the loss-leader model is being poorly implemented, and DigitalConvergence is trying to cover their original mistake by bandying about vague assertions about their "intellectual property." It's a bad business model, and now we see where it breaks down: any business model that assumes that the product will not be reverse-engineered is based on a faulty assumption, and is therefore itself faulty. The fault lies with those who conceived the bad idea, not with the people who, acting legally, point out the flaws in the idea.
No one is entitled to a profit; DigitalConvergence may be about to pay for their mistake in not properly enforcing their razor/blade model. It's not a crime to be part of the market forces which bring a poorly conceived business down, any more than it is not (yet) a crime to not see a movie and thus deprive its makers of royalties.
Once upon a time, I helped set up a new Radio Shack franchise - the barcode reader there simply plugged off a Y-adapter into the AT keyboard port, and when it scanned an item, it simply typed the decimal UPC number out. It's quite likely - especially since it plugs into a keyboard port - that this scanner works the same way. There might be some odd control codes in there, but fundamentally these things represent themselves to the computer as keyboards.
In a
previous Slashdot story,
it was reported that some of the USB specs were being closed to non-members; speculation in the discussion indicated that the closures were made so that CSS-style trusted-client content scrambling could be introduced into USB2 devices (for instance, encrypted signals to digital speakers so that the audio can't be captured over the USB bus.)
Does anyone know what became of this? Does the final USB2 spec contain any such anti-consumer features?
The technician who installed my DSL explained that the point to PPPoE is to ensure one (and only one) IP address is given to each customer; you have to authenticate before you can receive an IP address. Contrast with DHCP, where (apparently) anyone with a valid DSL line can grab multiple addresses. That's what the technician told me, anyway; I wouldn't be surprised to learn that that wasn't the real reason.
IP addresses are just numbers, but now they're hoarded like the precious commodities they are becoming. Yet another reason for IPv6 and soon...
Judge Kaplan drew an analogy between DeCSS and
computer viruses in order to justify the censorship
he is engaged in on the basis of public harm.
There's just one little problem with his analogy -
as far as I know, writing a virus is perfectly
legal.
The criminal act embodied in spreading a virus is
not the authorship of the virus, it's the
malicious (or negligent) unleashing of that
virus into the wild. It's not free speech to
deliberately trick someone into running a
destructive program.
It is, however, protected speech when you write
one. Academics and security specialists must have
the right to author virus code in order to
demonstrate potential real-world security threats,
to critique existing security and anti-virus
systems, and generally express their technical
ideas in the most convenient manner possible.
In this context, it's perfectly reasonable to
allow someone to write a dangerous virus and
distribute it to others - the intent is to
convey ideas, not to wreak havoc. It is only
when the act of malicious
distribution
occurs that the distributor becomes a criminal.
In the same way, DeCSS is a Constitutionally
protected embodiment of ideas about
DVD content control, and deserves the full
protection of the First Amendment. Those who
use it for movie piracy can rot in jail for all
I care, but DeCSS is more than just movie piracy.
It is the expression of someone's knowledge and
understanding, and when we forbid that, we lose
a very real piece of our free society.
Does it ban
the specific utility DeCSS, or all software
which decrypts CSS? Does it only apply to
Emmanuel Goldstein and 2600, or to everyone in the
court's jurisdiction? The ruling bans linking -
is the court's definition of linking restricted
to hypertext tags, or is it more expansive than that?
I've got a single-head G400, so the dualhead support doesn't appeal to me as much as increased 3d performance and the ability to run in 32bpp mode. Unfortunately, when I downloaded the binary and put it where Matrox said to, everything fell back to software 3d rendering, and the driver continued to complain about 32bpp not being supported. Am I missing something obvious? Has anyone gotten 32bpp and/or 3d out of this driver?
The motion picture industry is actually pursuing the trade secret argument in a separate case they brought in California state court (DVD-CCA v. Brunner et al). This article is about MPAA v. 2600, a suit brought in federal court in New York against 2600 and Emmanuel Goldstein under the DMCA.
What the industry is doing, of course, is following two different legal tacks, either of which may if successful result in DeCSS and similiar utilities being found illegal. The reason you don't see trade secret arguments being made here is that those aruments are being made on the West Coast.
Slashdot Mirror
What is disallowed is a viral licensing fee, like that used by shareware authors ("You can try this program and redistribute it, but if you use it for more than 30 days, you owe me $20."). You can't require that anyone who receives the software (not necessarily from you!) has to pay you money. The difference is that you might agree only to send the software to someone if they pay you money (CheapBytes would quickly go out of business if they were required to give away CDs for free), but you can't control what they do with the software afterwards. Imagine what would happen if CheapBytes required everyone who redistributed code from the CDs they sell to pay them a fee for the privlege of doing so!
The GPL does not enforce that software be available at no cost, but the market forces inherent in free software do (would you buy a Red Hat CD for $50, without any manuals or support, if you could get the same for about five bucks with shipping or download it off the Net for pennies?)
Of course, this only applies to licenses that are a grant of copyright rights like the GPL. Licenses that take away rights (such as the "no reverse engineering" clause in nearly every piece of commercial software) are another story; this is why the industry is trying so hard to get UCITA passed.
I'm in the process of moving to a new apartment right now, and I've been setting up utilities. Phone and power went like a breeze. The customer service reps (with whom I was able to speak to nearly immediately) took down my name, address, social security number, asked me for a date of activation, and then thanked me and hung up.
That was easy. The local cable modem company (RR) is that easy, too; I called to ask how long an installation took and the rep seemed surprised that I didn't want an install in the next couple of days. This is service, people.
I decided to go with DSL, instead, because there was *one* ISP doing business in town that I could get a static IP from for a reasonable price (Southwestern Bell will give me five static IPs for $80/mo, or one dynamic for $40/mo, but nothing in between - does this make any sense?)
I'm paying the cost not in money, but in time. A month without broadband will be a long time; if it weren't for the compelling feature set and the fact that I know I'll be stuck with whatever I'm getting now for a year, I'd be jumping ship to cable in a heartbeat.
Why on earth does DSL take so long to set up? Is there an order backlog a mile long? Does the phone company introduce some sort of mandatory delay into the process?
In the middle of the lecture, something like this transpires (paraphrased):
"And so you see that there can be occurences when...oh, here's an occurence. My phone's ringing."
[answers his phone]
"Hello?"
"Actually, I can't talk right now. I'm sort of teaching a class."
[class laughs]
"Yeah, you can hear them laughing in the background?"
"Okay, I'll be in my office around 4."
[hangs up phone]
[to class] "That was John Markoff from the New York Times. He wants to have a chat with me."
[resumes lecture seamlessly]
It's possible that HALlib is required for the G450, but I'm not sure.
- He pioneered the idea of free software, under the moniker of "free software," and very likely feels upset that those ideals have been usurped by this new movement with a new name (this motivation to have his contributions recognized in name can also be found in his urgings that the OS is called "GNU/Linux," not just "Linux");
- He fears that, by removing the word "free" from the name, we're also removing the concept of freedom from our thoughts about the subject. Freedom of software is important to him and he wants the users of free software to be reminded that they are excercising their freedom by using it. "Open Source," in his mind, dilutes that association in users' minds.
There isn't much difference, beyond the symbolic, between free and open source software. Microsoft's latest overture to allow its biggest customers to view the source code to some of its products does not qualify as open source by any stretch of the imagination - they're forbidding modification, redistribution, even compilation of the revealed code. And open source software cannot be bound by "oppressive EULA's" - see Term 7 of the Open Source Definition, which forbids the execution of additional licenses. Keep in mind that the OSD derives, nearly directly, from the Debian Free Software Guidelines.The difference is a difference of name; a kernel by any other name would smell just as sweet.
If I recall correctly, the timespan of a copyrighted work originated by a corporation (a "work for hire" like a Hollywood movie) is now 90 years, recently extended by the Sonny Bono Act from 70. The fate of the company has no bearing on the duration of copyright.
One of the Debian Free Software Guidelines is "License Must Not Be Specific to Debian." So, no, this piece of software will not be allowed into the Debian system unless and until SuSe decides to relicense it in a distribution-agnostic manner. Remember, though, that the Debian mirrors contain a 'non-free' branch which, while not technically part of the Debian system, allows Debian users to easily install such packages. The net result is that Debian users who are willing to accept the license will probably have an easy way to install the software at the end of February, but it won't be on any "official Debian" CDs for a while after that.
And I noticed one sort of odd thing: it hasn't crashed yet on me. When I started trying Mozilla the thing blew up all the time; now I feel completely comfortable with the idea of Mozilla as my primary browser, particularly with the integrated (and free!) crypto.
The only edge Konqueror has over Mozilla now, in my opinion, is being based on QT - which is the only toolkit so far that's been patched to use the excellent Xft library for antialiased fonts under XFree86 4.0.2 and later. I can't wait for Mozilla to pick up support for this thing.
So what is the status of a US patent internationally? Are there any treaties that give patents international scope?
Most patents awarded in the US are recognized by other countries.
Does FreeType have another website that I'm not aware of?
There are still plenty of issues outstanding, though:
The net effect is that a quantum computer in the hands of an eavesdropper halves the effective keylength - a 128-bit key is reduced to 64 bits of effectiveness. 64 bits is, of course, not enough security to defend against government-level surveillance resources, but all that has to be done to solve the problem is to increase the keylength to 256 bits.
One of the requirements for the AES candidates was that the algorithm support 256-bit operation. Rijndael, the heir apparent to DES, does support 256-bit operation modes.
A few years back, when one of our chemistry profs won a fairly well-known prize (for the discovery of "buckyballs"), we elected him Homecoming Queen. (Don't believe me? Check out his C.V.!)
The end result of this is that you have to elect men and women whose independent judgement you trust, whose principles parallel yours, and whom you trust not to abuse the privleged position they're given. You must, in short, choose someone with good character. Good luck finding that person.
Cost: One more piece of closed, non-free software is created. A developer who normally writes open, free software must write this. Users are still deprived (as they were before) of knowledge about the PS2 API. Users can freely share, but cannot easily learn from or change, this piece of software. (Whether this is depriving the users of their rights, or simply not granting them a right they didn't have in the first place is largely an issue of perspective.)
Benefit: An entire computing platform is opened so that free, open software can be written for it, whereas it was impossible to write such software in the past. Many existing pieces of free software can be immediately used on the PS2.
I think it's fairly clear that the benefits outweigh the costs (if these costs can be said to exist at all.)
Unfortunately, it's not in Sony's interests to disclose their secret API to someone who will make it possible for free software to work on the PS2. Their whole business model is about taking a significant loss on the hardware, then making that loss up in license fees from the sale of software. If there were lots of free software available for the PS2, sales of console games would dip and cut into Sony's revenue stream.
You can require, via clearly stated contractual obligation, that those who purchase the razor must also purchase a sufficient quantity of blades (this is the arrangement Netpliance finally and correctly reached, after weeks of corporate bumbling and stumbling that probably cost them a real shot at their market.)
You can loan the razor to the consumer, and make the purchase of blades a condition of that loan. The fact that there is a loan, with conditions attached, must again be clearly stated.
DigitalConvergence appears to be trying to use the third method of enforcement. According to their business plan, they still own all of the CueCat devices; however, they haven't done a very good job of making this clear to people. Radio Shack hands these things over the counter without ever mentioning any conditions. They send the things out to Forbes subscribers, possibly running into interesting postal law that makes those scanners the property of the recipients. The notion of a loan with conditions is hidden deep in the fine print in the package. Why not be forthright? Because DigitalCovergence is getting and correlating information about everything a CueCat user scans, and it's best if that ugly fact is hidden as deeply as possible.
In this case, the loss-leader model is being poorly implemented, and DigitalConvergence is trying to cover their original mistake by bandying about vague assertions about their "intellectual property." It's a bad business model, and now we see where it breaks down: any business model that assumes that the product will not be reverse-engineered is based on a faulty assumption, and is therefore itself faulty. The fault lies with those who conceived the bad idea, not with the people who, acting legally, point out the flaws in the idea.
No one is entitled to a profit; DigitalConvergence may be about to pay for their mistake in not properly enforcing their razor/blade model. It's not a crime to be part of the market forces which bring a poorly conceived business down, any more than it is not (yet) a crime to not see a movie and thus deprive its makers of royalties.
Once upon a time, I helped set up a new Radio Shack franchise - the barcode reader there simply plugged off a Y-adapter into the AT keyboard port, and when it scanned an item, it simply typed the decimal UPC number out. It's quite likely - especially since it plugs into a keyboard port - that this scanner works the same way. There might be some odd control codes in there, but fundamentally these things represent themselves to the computer as keyboards.
Does anyone know what became of this? Does the final USB2 spec contain any such anti-consumer features?
IP addresses are just numbers, but now they're hoarded like the precious commodities they are becoming. Yet another reason for IPv6 and soon...
There's just one little problem with his analogy - as far as I know, writing a virus is perfectly legal.
The criminal act embodied in spreading a virus is not the authorship of the virus, it's the malicious (or negligent) unleashing of that virus into the wild. It's not free speech to deliberately trick someone into running a destructive program.
It is, however, protected speech when you write one. Academics and security specialists must have the right to author virus code in order to demonstrate potential real-world security threats, to critique existing security and anti-virus systems, and generally express their technical ideas in the most convenient manner possible. In this context, it's perfectly reasonable to allow someone to write a dangerous virus and distribute it to others - the intent is to convey ideas, not to wreak havoc. It is only when the act of malicious distribution occurs that the distributor becomes a criminal.
In the same way, DeCSS is a Constitutionally protected embodiment of ideas about DVD content control, and deserves the full protection of the First Amendment. Those who use it for movie piracy can rot in jail for all I care, but DeCSS is more than just movie piracy. It is the expression of someone's knowledge and understanding, and when we forbid that, we lose a very real piece of our free society.
Does it ban the specific utility DeCSS, or all software which decrypts CSS? Does it only apply to Emmanuel Goldstein and 2600, or to everyone in the court's jurisdiction? The ruling bans linking - is the court's definition of linking restricted to hypertext tags, or is it more expansive than that?
I've got a single-head G400, so the dualhead support doesn't appeal to me as much as increased 3d performance and the ability to run in 32bpp mode. Unfortunately, when I downloaded the binary and put it where Matrox said to, everything fell back to software 3d rendering, and the driver continued to complain about 32bpp not being supported. Am I missing something obvious? Has anyone gotten 32bpp and/or 3d out of this driver?
What the industry is doing, of course, is following two different legal tacks, either of which may if successful result in DeCSS and similiar utilities being found illegal. The reason you don't see trade secret arguments being made here is that those aruments are being made on the West Coast.