A casual stamp collector does not have the need for a complex layout. It is not difficult to make sure that a page gets rendered properly (or is usable) in Lynx unless it's a very sophisticated/hackish layout. People will need to learn to use ALT tags and the validator, but it's not such a big deal for a personal/small website.
Also, keep in mind that if the court agrees that the ADA applies to web sites, every HTML writing program will probably include ALT tag capabilities. Legislation like the ADA is a good way to ensure that software makers write programs that generate standards-compliant code. Standards compliance is always a good thing.
Now, we add a whole new method of content rendering. We can't even impliment the main standards properly. How do we plan to ensure that an audio interface can successfully read a website, as well? Keep in mind that this is not what the web was originally designed to handle.
Actually, this IS what the web was developed to handle. Take a look at any HTML/1.0 page and you will notice that it can be read perfectly. Everything became screwed up when the <table> tag was introduced and people started to use HTML as a substitute for PDF, and later Flash. HTML is a markup, not a layout language.
As far as the internet being "hard to convert": no it isn't. Some businesses just need to get hit with a couple of lawsuits to figure that one out. Yes, websites do need to take that into account, so some webmasters will have to change their ways. But, you can make a perfectly functional audio website by using a content management system that supports it, without too much effort. Audio internet is not a hard problem unless you are an amateur trying to do a professional's job.
and tags are what mozilla shows when it doesn't get anything from a server. It looks more like connection problems on your end than anything else. However, it seems to me like these pages are cgi scripts designed to spit out the search terms that people find them with to increase their search ranks. A lot of them simply redirect to other sites when you click.
You are aware that DHCP servers have logs that can map a particular computer to an IP address? If you use cable/dsl, the DSL/cable modem has an internal hardware MAC address that uniquely identifies you as a subscriber - that's how they know which modems to grant access to. The DHCP server logs this stuff. I'm sure they can get usage statistics for each subscriber if they wanted/were forced to.
A gig of flash is about $500 now. This would mean that the price is about 50c/meg. Which would mean that putting Mozilla instead of Opera would cost maybe $2 more (4 megs vs 8 megs), while providing about 3 times more features, standards-compliance and better compatibility. I personally haven't seen a single complex page which Opera displays 100% correctly.
Legacy CPUs? Is that what Macs use? What's the fastest G4 speed - 800 MHz? The x86 architecture is in the double-gigahertz range now. I don't think that's what you call "legacy."
Also, OS X about as close to UNIX as Cygwin running on top of Win98 is. They still have a bunch of NeXT stuff under the hood (Darwin) and do most things the NeXT way (display postscript, etc). The only thing that makes it "unix" is the fact that it runs some unix commands. But you can make DOS run unix commands, so that's not really a good argument.
I would be very skeptical of using something like Darwin/OS X on an industrial-class machine. It's worse than Win2K in terms of overhead (can you even boot without a GUI?), and runs a weird microkernel.
Yes, it makes a good desktop. If you hate computers and love the Apple way of doing things, this is the OS for you. If you switch from Linux to OS X, you probably shouldn't have been using Linux in the first place.
Apparently, you don't know how the FCC works. The frequencies will be auctioned off to the highest bidder, for private use (probably cellphones). Not to mention that gigabit wireless cannot run at those frequencies anyway (band is too narrow).
The point is choice, really. If you like depending on one vendor for everything, go ahead. If you think the price of MS products is fair now, you're not alone. But what keeps MS from charging anything they want once they implant their OS on every desktop machine out there? What makes you think that Windows 2005 will be a few hundred bucks a seat? You do notice that you start to depend on Microsoft for everything once you start using their OS? And is anyone naive enough to think that hardware DRM and "secure platform" initiatives are designed for anything other than making Windows an inseparable part of every PC? If you believe that, simply consider the fact that the entertainment industries have been whining about DRM since 1996, but Microsoft just started to implement it.
If someone doesn't start promoting alternative platforms now, it will be far too late to do anything once MS decides to start tightening the screws.
.deb is no better than RPM, and is much less popular (how many people use debian, anyway?). If you are thinking that APT is somehow tied to.deb, you're wrong. Mandrake has an apt-like tool called URPMI, and it works quite well.
Yes, but who gives a fuck? Yes, you could make a DVD player for $20 if the parts are cheap enough. So what? The players are already around $60. Who cares?
Get a clue? No thanks, I have one. The truth is, it is very difficult to develop a complex system as free software. You can't develop a Photoshop replacement, a Quicken replacement, or an Office replacement (not just Word and Excel, but Access and the rest) as free software, simply because much of it is not coding, but rather design and ergonomics and basically the amount of money put into it.
A killer app can only bring popularity to a new system when you can do on it everything you could do on its predecessor. When Linux can run Photoshop, Office, Quicken, and 90% of the other applications, then you can talk about a killer application.
Please note that every version of Microsoft's software is backwards-compatible with almost every other version. Do you seriously think this is not one of the major reasons they are so successful? Do you think people would switch to WinXP if it couldn't run programs for Win95?
In any case, backwards compatibility is job #1, and replacements don't really cut it. People want to use (initially, at least) MS Office and not (star|open)office or abiword. Which is what my point originally was.
What you say is true, but you're overlooking an important point. Ports are very expensive to develop, unless the program has been designed to be portable from the ground up (like Quake3 or NWN). Companies can not really afford to develop ports for a niche audience. Adobe knows that if you use Photoshop, you will buy WIndows. It's not the other way around. Therefore, they have zero incentive to port the program, which is why it would not get ported.
Also, if the company doesn't want to pay for the development of a port out of its own pocket, it would have to sell the port for the full price, which is one of the things you frown upon. If you had Windows versions of all the Adobe tools ($10k+), would you consider switching to Linux and pay for the same software again? Highly unlikely. So how would Adobe pay for the port?
I'm not saying I don't pay for software. I just have serious doubts that there are many people out there who would pay $600 for a Linux version of Photoshop.
Actually, Apple's solution is very, very far from FreeBSD. They don't use very many parts of it in their OS. The core is Darwin, and on top of that is a Display PDF-based GUI and all kinds of stuff. It's somewhat UNIX-compatible, but it doesn't really use UNIX much internally.
In any case, I don't think simply making a good desktop and even porting a few apps to it is enough. Apple did that, and are they gaining users? Yes, but very slowly. They'll never get any major market share, simply because their machines are expensive (price/performance ratio) and the benefits aren't very significant to a typical Windoze user. Yes, the iPrograms are nice, but users don't really care much about convenience or elegance. They mostly care about getting work done and having fun - how elegantly they do it is not very important.
Yes, linux certainly needs to be improved, but the main problem that needs to be tackled is the lack of applications, and not the lack of polish. If you could run as many programs as you could on Windows, you would probably consider switching. If you can't, it doesn't really matter whether it's polished or not. That's one of the reasons why Apple is losing right now.
The ones that are switching were previously using expensive UNIX-based solutions. People are switching from SGI Irix and Sun to Linux - yes, because PCs are cheap. They've already had 90% of the software they need ported, anyway, and they don't need Photoshop. However, Linux is not yet a threat to Windows or Mac, because it simply can not run the right combination of programs for users to even consider switching. It's a chicken and egg problem: the users have no incentive for switching other than minor price differences and ideology, and the developers won't bother writing anything until there is a user base that is ready to buy their programs. There either needs to be a very strong incentive for users to switch or a portability solution that would allow developers to port their stuff over easily (like Wine). Yes, it could be solved without Wine, but it would involve Adobe, Macromedia, and countless others dumping major cash into the platform (which won't happen for sure).
Face the music: there are not enough users on Linux to justify having any developers work on a port of, say, Photoshop. It would take millions of dollars to port, and nobody will buy it. Given that Linux has maybe 0.5% of the desktop, and that maybe 1% of that will ever buy software that costs more than $30, I doubt the expense is justified.
How about promoting more useful projects like Wine/Winelib instead? A company with even marginal resources (Codeweavers) can do wonders with Wine, such as run MS Office and MSIE quite well. If some other company spent some more resources on improving it, it would be able to run 90% of the apps out there, including Photoshop and all the other stuff. It would also have a good chance of increasing that 0.5% market share to something more reasonable.
If you still don't believe me, just consider what would happen if Adobe ported Photoshop to Linux. 10 or 15 people would actually buy it. It would get press coverage. And then, nothing would happen and no other company will bother porting anything. Kind of like what happened to Loki.
How many good OpenGL drivers have you seen from ATI? The only companies that made good GL drivers were 3Dlabs and Nvidia. The rest completely sucked. If the OpenGL driver for a certain card only implements a part of the API, you can't use apps which depend on the missing pieces.
ATI does not have or sponsor open-source drivers, either. The ones they do have are closed source and far inferior to nvidia's. There is a driver that someone is developing for XFree, but the cards will long be outdated before it's done.
At least nvidia supports Linux and makes very good drivers and good hardware. ATI does not support Linux at all for their consumer-level video cards, and provides bad drivers even for their professional cards (the only reason there is a driver for the Radeon 8500 is because it uses the same core as a professional card). Given that ATI cards do not have working drivers, I do not see why the creators of UT have to support them.
Yes, but black-box testing may not be enough to spot rarely occuring anomalies (intentional or erroneous) that would have otherwise been revealed if the code was open.
If the anomalies are so minor that a proper test can not detect them, they are probably not significant statistically, probably something on the order of thousandths of a percent. Also, how do you make a machine that is intentionally biased against a candidate when you don't know who that might be? I'm sure they didn't hardcode the candidates and parties into the machine.
Simple. By having the MD5Sums of the software and compiled code published, and having the compilation process and MD5Sums witnessed by technical representatives of all candidates in the election.
Still, how do you validate the md5sum of every machine without violating its integrity? In 99% of the cases, if you can read the firmware/software, you can replace it, too. It would be difficult to ensure that every machine is running the correct, valid code if you can randomly check the md5sum.
But if there is ever another election as close as Florida's November 2000 fiasco, using closed source software all across the state(s) involved, you can bet there will be lawsuits claiming that the software was biased in the favor of one party or another.
This is the problem with the US voting process, and not the technical side of things. The proper way to handle the problem would be to have a second round of elections, where the candidates with insignificant numbers of votes (Nader, et al) get thrown out and people have to make a choice. That's how it is done in many countries when the numbers get too close. If the margin between two candidates is less than 3 percent, it's too close to call, especially when you have other candidates with single-digit percentages.
The people pay for a lot of things, the voting machines being fairly minor expenses. For example, US taxpayers paid over $15 billion to private, commercial airlines last year. Capitalism at its finest...
You are aware of the fact that there are things such as independent certification procedures that are designed to find flaws in machines? Who the hell would make code/firmware for a voting machine open-source? Why? How the heck do you know that the program in the machine is the same program that you have the source code for? Who the hell is going to review the code, even if it was released as open-source? Some people have a life, you know.
Also, you do realize that the computerized optical scanners or punchcard readers that were used before these machines aren't open-source, either?
The article merely states the obvious fact: when untrained personnel is responsible for some machines, they break. This has nothing to do with the voting machines, and everything to do with the lack of proper procedures. Also, there is always a certain margin of error in voting. That's why most civilized countries (except the US) hold runoff elections when the votes are close.
Huh? The US system has a hot and a neutral line, just like any other AC power system anywhere in the world. Touching one pin will give you 110V, the other is zero. There is also an earth ground line in US outlets, which connects to metal equipment cases so that they don't become electrified. There is no such thing as +60V or -60V - it's AC. Read a book about electrical wiring or something.
Bullshit. 240V is about as safe as 120V in terms of electrocuting yourself, and is much less of a fire hazard (guess why americans have to use expensive solid-copper wire to not have fires). Try touching the hot wire of a 120V plug sometimes. Rest assured, you'll feel much more than a "tingle". Also, I've come into contact with both voltages several times, and they feel about the same (something along the lines of "FUCK!!!")
The attention came from a change to the license; specifically the removal [archive.org] of an exemption for software players/decoders distributed free of charge.
The real reason they removed that exemption is so that players such as Winamp, which are commercial but distributed free of charge, have to pay a license fee. If you formally list an exception for free decoders, they wouldn't have to do that, because they technically are free (even though they are made by a large corporation and are supposed to make money). Also, pretty much anyone would be able to have an MP3 decoder in their application (even if it costs money) by shipping it separately free of charge. It's a gaping loophole. So, Thomson now is saying that the fees still apply to everyone, although Thomson is not enforcing them for what they consider free decoders. If you show me one free project that has gotten a cease-and-desist letter from Thomson, then I'll believe you. As it is, it does look like a Vorbis publicity stunt.
I highly doubt that Thomson will ever start cracking down on free, open-source decoders, because it just doesn't make any business sense. Besides promoting Ogg Vorbis, it would also generate bad publicity for them and the MP3 format, while not earning them a single cent in extra revenues.
Slashdot Mirror
A casual stamp collector does not have the need for a complex layout. It is not difficult to make sure that a page gets rendered properly (or is usable) in Lynx unless it's a very sophisticated/hackish layout. People will need to learn to use ALT tags and the validator, but it's not such a big deal for a personal/small website.
Also, keep in mind that if the court agrees that the ADA applies to web sites, every HTML writing program will probably include ALT tag capabilities. Legislation like the ADA is a good way to ensure that software makers write programs that generate standards-compliant code. Standards compliance is always a good thing.
Now, we add a whole new method of content rendering. We can't even impliment the main standards properly. How do we plan to ensure that an audio interface can successfully read a website, as well? Keep in mind that this is not what the web was originally designed to handle.
Actually, this IS what the web was developed to handle. Take a look at any HTML/1.0 page and you will notice that it can be read perfectly. Everything became screwed up when the <table> tag was introduced and people started to use HTML as a substitute for PDF, and later Flash. HTML is a markup, not a layout language.
As far as the internet being "hard to convert": no it isn't. Some businesses just need to get hit with a couple of lawsuits to figure that one out. Yes, websites do need to take that into account, so some webmasters will have to change their ways. But, you can make a perfectly functional audio website by using a content management system that supports it, without too much effort. Audio internet is not a hard problem unless you are an amateur trying to do a professional's job.
and tags are what mozilla shows when it doesn't get anything from a server. It looks more like connection problems on your end than anything else. However, it seems to me like these pages are cgi scripts designed to spit out the search terms that people find them with to increase their search ranks. A lot of them simply redirect to other sites when you click.
You are aware that DHCP servers have logs that can map a particular computer to an IP address? If you use cable/dsl, the DSL/cable modem has an internal hardware MAC address that uniquely identifies you as a subscriber - that's how they know which modems to grant access to. The DHCP server logs this stuff. I'm sure they can get usage statistics for each subscriber if they wanted/were forced to.
A gig of flash is about $500 now. This would mean that the price is about 50c/meg. Which would mean that putting Mozilla instead of Opera would cost maybe $2 more (4 megs vs 8 megs), while providing about 3 times more features, standards-compliance and better compatibility. I personally haven't seen a single complex page which Opera displays 100% correctly.
Legacy CPUs? Is that what Macs use? What's the fastest G4 speed - 800 MHz? The x86 architecture is in the double-gigahertz range now. I don't think that's what you call "legacy."
Also, OS X about as close to UNIX as Cygwin running on top of Win98 is. They still have a bunch of NeXT stuff under the hood (Darwin) and do most things the NeXT way (display postscript, etc). The only thing that makes it "unix" is the fact that it runs some unix commands. But you can make DOS run unix commands, so that's not really a good argument.
I would be very skeptical of using something like Darwin/OS X on an industrial-class machine. It's worse than Win2K in terms of overhead (can you even boot without a GUI?), and runs a weird microkernel.
Yes, it makes a good desktop. If you hate computers and love the Apple way of doing things, this is the OS for you. If you switch from Linux to OS X, you probably shouldn't have been using Linux in the first place.
Apparently, you don't know how the FCC works. The frequencies will be auctioned off to the highest bidder, for private use (probably cellphones). Not to mention that gigabit wireless cannot run at those frequencies anyway (band is too narrow).
The point is choice, really. If you like depending on one vendor for everything, go ahead. If you think the price of MS products is fair now, you're not alone. But what keeps MS from charging anything they want once they implant their OS on every desktop machine out there? What makes you think that Windows 2005 will be a few hundred bucks a seat? You do notice that you start to depend on Microsoft for everything once you start using their OS? And is anyone naive enough to think that hardware DRM and "secure platform" initiatives are designed for anything other than making Windows an inseparable part of every PC? If you believe that, simply consider the fact that the entertainment industries have been whining about DRM since 1996, but Microsoft just started to implement it.
If someone doesn't start promoting alternative platforms now, it will be far too late to do anything once MS decides to start tightening the screws.
.deb is no better than RPM, and is much less popular (how many people use debian, anyway?). If you are thinking that APT is somehow tied to .deb, you're wrong. Mandrake has an apt-like tool called URPMI, and it works quite well.
Samba? Sounds to me like you have bad memory and/or motherboard. Use that anti-static strap next time.
The reason portable players are expensive is because few people want them, not because they're expensive to make.
Yes, but who gives a fuck? Yes, you could make a DVD player for $20 if the parts are cheap enough. So what? The players are already around $60. Who cares?
Get a clue? No thanks, I have one. The truth is, it is very difficult to develop a complex system as free software. You can't develop a Photoshop replacement, a Quicken replacement, or an Office replacement (not just Word and Excel, but Access and the rest) as free software, simply because much of it is not coding, but rather design and ergonomics and basically the amount of money put into it.
A killer app can only bring popularity to a new system when you can do on it everything you could do on its predecessor. When Linux can run Photoshop, Office, Quicken, and 90% of the other applications, then you can talk about a killer application.
Please note that every version of Microsoft's software is backwards-compatible with almost every other version. Do you seriously think this is not one of the major reasons they are so successful? Do you think people would switch to WinXP if it couldn't run programs for Win95?
In any case, backwards compatibility is job #1, and replacements don't really cut it. People want to use (initially, at least) MS Office and not (star|open)office or abiword. Which is what my point originally was.
What you say is true, but you're overlooking an important point. Ports are very expensive to develop, unless the program has been designed to be portable from the ground up (like Quake3 or NWN). Companies can not really afford to develop ports for a niche audience. Adobe knows that if you use Photoshop, you will buy WIndows. It's not the other way around. Therefore, they have zero incentive to port the program, which is why it would not get ported.
Also, if the company doesn't want to pay for the development of a port out of its own pocket, it would have to sell the port for the full price, which is one of the things you frown upon. If you had Windows versions of all the Adobe tools ($10k+), would you consider switching to Linux and pay for the same software again? Highly unlikely. So how would Adobe pay for the port?
I'm not saying I don't pay for software. I just have serious doubts that there are many people out there who would pay $600 for a Linux version of Photoshop.
Actually, Apple's solution is very, very far from FreeBSD. They don't use very many parts of it in their OS. The core is Darwin, and on top of that is a Display PDF-based GUI and all kinds of stuff. It's somewhat UNIX-compatible, but it doesn't really use UNIX much internally.
In any case, I don't think simply making a good desktop and even porting a few apps to it is enough. Apple did that, and are they gaining users? Yes, but very slowly. They'll never get any major market share, simply because their machines are expensive (price/performance ratio) and the benefits aren't very significant to a typical Windoze user. Yes, the iPrograms are nice, but users don't really care much about convenience or elegance. They mostly care about getting work done and having fun - how elegantly they do it is not very important.
Yes, linux certainly needs to be improved, but the main problem that needs to be tackled is the lack of applications, and not the lack of polish. If you could run as many programs as you could on Windows, you would probably consider switching. If you can't, it doesn't really matter whether it's polished or not. That's one of the reasons why Apple is losing right now.
The ones that are switching were previously using expensive UNIX-based solutions. People are switching from SGI Irix and Sun to Linux - yes, because PCs are cheap. They've already had 90% of the software they need ported, anyway, and they don't need Photoshop. However, Linux is not yet a threat to Windows or Mac, because it simply can not run the right combination of programs for users to even consider switching. It's a chicken and egg problem: the users have no incentive for switching other than minor price differences and ideology, and the developers won't bother writing anything until there is a user base that is ready to buy their programs. There either needs to be a very strong incentive for users to switch or a portability solution that would allow developers to port their stuff over easily (like Wine). Yes, it could be solved without Wine, but it would involve Adobe, Macromedia, and countless others dumping major cash into the platform (which won't happen for sure).
Face the music: there are not enough users on Linux to justify having any developers work on a port of, say, Photoshop. It would take millions of dollars to port, and nobody will buy it. Given that Linux has maybe 0.5% of the desktop, and that maybe 1% of that will ever buy software that costs more than $30, I doubt the expense is justified.
How about promoting more useful projects like Wine/Winelib instead? A company with even marginal resources (Codeweavers) can do wonders with Wine, such as run MS Office and MSIE quite well. If some other company spent some more resources on improving it, it would be able to run 90% of the apps out there, including Photoshop and all the other stuff. It would also have a good chance of increasing that 0.5% market share to something more reasonable.
If you still don't believe me, just consider what would happen if Adobe ported Photoshop to Linux. 10 or 15 people would actually buy it. It would get press coverage. And then, nothing would happen and no other company will bother porting anything. Kind of like what happened to Loki.
How many good OpenGL drivers have you seen from ATI? The only companies that made good GL drivers were 3Dlabs and Nvidia. The rest completely sucked. If the OpenGL driver for a certain card only implements a part of the API, you can't use apps which depend on the missing pieces.
ATI does not have or sponsor open-source drivers, either. The ones they do have are closed source and far inferior to nvidia's. There is a driver that someone is developing for XFree, but the cards will long be outdated before it's done.
At least nvidia supports Linux and makes very good drivers and good hardware. ATI does not support Linux at all for their consumer-level video cards, and provides bad drivers even for their professional cards (the only reason there is a driver for the Radeon 8500 is because it uses the same core as a professional card). Given that ATI cards do not have working drivers, I do not see why the creators of UT have to support them.
Yes, but black-box testing may not be enough to spot rarely occuring anomalies (intentional or erroneous) that would have otherwise been revealed if the code was open.
If the anomalies are so minor that a proper test can not detect them, they are probably not significant statistically, probably something on the order of thousandths of a percent. Also, how do you make a machine that is intentionally biased against a candidate when you don't know who that might be? I'm sure they didn't hardcode the candidates and parties into the machine.
Simple. By having the MD5Sums of the software and compiled code published, and having the compilation process and MD5Sums witnessed by technical representatives of all candidates in the election.
Still, how do you validate the md5sum of every machine without violating its integrity? In 99% of the cases, if you can read the firmware/software, you can replace it, too. It would be difficult to ensure that every machine is running the correct, valid code if you can randomly check the md5sum. But if there is ever another election as close as Florida's November 2000 fiasco, using closed source software all across the state(s) involved, you can bet there will be lawsuits claiming that the software was biased in the favor of one party or another.
This is the problem with the US voting process, and not the technical side of things. The proper way to handle the problem would be to have a second round of elections, where the candidates with insignificant numbers of votes (Nader, et al) get thrown out and people have to make a choice. That's how it is done in many countries when the numbers get too close. If the margin between two candidates is less than 3 percent, it's too close to call, especially when you have other candidates with single-digit percentages.
The people pay for a lot of things, the voting machines being fairly minor expenses. For example, US taxpayers paid over $15 billion to private, commercial airlines last year. Capitalism at its finest...
You are aware of the fact that there are things such as independent certification procedures that are designed to find flaws in machines? Who the hell would make code/firmware for a voting machine open-source? Why? How the heck do you know that the program in the machine is the same program that you have the source code for? Who the hell is going to review the code, even if it was released as open-source? Some people have a life, you know.
Also, you do realize that the computerized optical scanners or punchcard readers that were used before these machines aren't open-source, either?
The article merely states the obvious fact: when untrained personnel is responsible for some machines, they break. This has nothing to do with the voting machines, and everything to do with the lack of proper procedures. Also, there is always a certain margin of error in voting. That's why most civilized countries (except the US) hold runoff elections when the votes are close.
Huh? The US system has a hot and a neutral line, just like any other AC power system anywhere in the world. Touching one pin will give you 110V, the other is zero. There is also an earth ground line in US outlets, which connects to metal equipment cases so that they don't become electrified. There is no such thing as +60V or -60V - it's AC. Read a book about electrical wiring or something.
Bullshit. 240V is about as safe as 120V in terms of electrocuting yourself, and is much less of a fire hazard (guess why americans have to use expensive solid-copper wire to not have fires). Try touching the hot wire of a 120V plug sometimes. Rest assured, you'll feel much more than a "tingle". Also, I've come into contact with both voltages several times, and they feel about the same (something along the lines of "FUCK!!!")
The attention came from a change to the license; specifically the removal [archive.org] of an exemption for software players/decoders distributed free of charge.
The real reason they removed that exemption is so that players such as Winamp, which are commercial but distributed free of charge, have to pay a license fee. If you formally list an exception for free decoders, they wouldn't have to do that, because they technically are free (even though they are made by a large corporation and are supposed to make money). Also, pretty much anyone would be able to have an MP3 decoder in their application (even if it costs money) by shipping it separately free of charge. It's a gaping loophole. So, Thomson now is saying that the fees still apply to everyone, although Thomson is not enforcing them for what they consider free decoders. If you show me one free project that has gotten a cease-and-desist letter from Thomson, then I'll believe you. As it is, it does look like a Vorbis publicity stunt.
I highly doubt that Thomson will ever start cracking down on free, open-source decoders, because it just doesn't make any business sense. Besides promoting Ogg Vorbis, it would also generate bad publicity for them and the MP3 format, while not earning them a single cent in extra revenues.