I believe The Wind Done Gone was, in fact, the subject of a lawsuit.
However, it is a parody, protected under fair use.
The first question is whether a work is derivative.
Much of the debate over the "viral" nature of the GNU GPL is actually a debate over what constitutes a derivative work.
Does inclusion of a header constitute derivation?
What about a plugin, which may essentially use a naming convention?
I don't know the ins and outs of NWN modules, but I wouldn't dismiss the possibility that one could be held to be derivative.
Wake up, no matter what some silly EULA says, aftermarket modifications that you spend your time on are yours....
no judge would let a company steal those rights from you because of some contract you did not sign before the sale.
While the enforceability of EULAs is still being tested in the courts and tweaked by the likes of UCITA, it is copyright law that dominates, here.
Copyright law grants to the author the exclusive right to create derivative works.
You cannot mod, so to speak, "Gone With the Wind."
There was (and may still be) a group working on a Matrix FPS mod.
It's a waste of time, because Warner Bros. has granted an exclusive Matrix license to a commercial game company.
If the mod is any good, it will be suppressed, just like the Alien TC for Doom.
The NWN EULA adds to, rather than detracts from, fair use.
I just don't find the additional grants compelling.
now all the big distros are going to scramble to put out another release
Mandrake 8.2 shipped with KDE 2, but now has an update to KDE 3.
It wouldn't be too much of a stretch to do the same for GNOME 2.
I'd give it a couple of months, though, for testing.
I don't know that GNOME 2 will drive up too many distro version numbers.
It's pretty safe to assume that Red Hat is already working on 8.0 with GCC 3.
it does not appear to harm the free software community for the moment
Not to start a GPL-free v. BSD-free flamefest, but the Tux, real-time, and secure Linux patents harm BSD, which is part of the free software community.
The GNU GPL is not an implied patent license.
Its purpose is not "to induce you to infringe any patents or other property right claims or to contest validity of any such claims."
However, if I release software under the GPL, but withhold the patent, then the release is a sham.
If I don't own the copyright to all the code, then it may be illegal.
The point is, we'll have to look carefully at the patent license for boilerplate lawyerese, like a termination clause, that would be incompatible with the GPL.
A GPL-compatible patent license must be written very broadly.
Game companies never buy ideas from people, because EVERYONE has ideas....
I can find no fathomable reason for the INCREDIBLE number of "me too" games
I couldn't agree with you more.
I keep reading that game companies already have more game ideas than they can use.
Then why do they keep reusing the same old ones?
The game industry is still in its infancy, still discovering its medium.
Gamers still ooh and aah at the latest 3-D engine out of Texas.
Simple mechanics are still a differentiator among flight sims and driving games.
From time to time, PC Gamer publishes a list of the best games "of all time."
Is anyone really still playing Duke Nukem, Doom, or Wing Commander?
Does anyone play WarCraft, instead of its sequels?
Strange, because movie sequels are lucky to be considered in the same league as the original.
Maybe we just have to wait out Moore's Law, but I think there will be opportunities for indie developers to show up the big game studios with a new idea, or two.
I don't want to take anything away from the 99% perspiration of games like Half-Life and Max Payne.
I just want to see that 1% inspiration show up a little more prominently.
Perhaps Blizzard shouldn't piss off its customers
on
Warcraft III Gone Gold
·
· Score: 2
Blizzard CAN write a Linux version of all their products. If you don't want it, OK.
But don't do an Al-Queda for the Linux gaming industry... (like the bnetd guys are)
First, comparing bnetd developers to Al-Qaeda terrorists pretty much puts you on the losing side of Godwin's law.
Blizzard's harrassment of the bnetd project has nothing to do with Linux or open source.
It's an abuse of the legal system.
Likewise, Adobe's abuse of the DMCA should offend you, even if you're not interested in cracking eBooks.
The main clusters to benefit would be ones like Beowulf which tend to use "commodity" hardware
TruClusters can use Ethernet as the cluster interconnect, but Memory Channel is recommended, because of its lower latency.
I was also thinking of Oracle 9i RAC.
The Oracle / Dell / Red Hat bundle uses Gigabit Ethernet.
I'm not familiar with the Oracle cluster file system, but I suspect that it would benefit from a low-latency interconnect.
Where it comes in useful is for high-end servers, network backbones and (possibly) clusters
HP's TruCluster is designed to use a proprietary cluster interconnect called Memory Channel, with a bandwidth of about 100 MB/s. Gigabit Ethernet is cheaper, but can't compete when it comes to latency.
Any idea how we can expect this new standard to compare?
We're going to need something to replace PCI before we can use 10Gbit ethernet fully though.
If a network has more than two nodes, as most networks do, then each node isn't expected to saturate the network.
Think of adding more lanes to highway, rather than increasing the speed limit.
You'll never hear, "We used off the shelf hardware and stock footage to make
this scene, and saved a few dollars."
I saw a neat demo at a Maya seminar last week.
They're looking to take panoramas to the next level--not only can you rotate the camera, but you can translate it.
Combine this with some effects (e.g., moving water) and low-end compositing, and you can put together a passable demo reel with little or no video footage.
As for production, we're seeing subtle effects in main-stream films.
I was amazed by the behind-the-scenes footage of Cast Away, because I had no idea the extent to which CGI was used.
This will eventually filter down even to the million dollar or less films.
The difference between Linux and proprietary Unix will make a difference, there.
I have a genuine Intel 2GHz processor. The "1.7+" and even "1.6+" AMD machines kick my Intel's butt
Tom's rhetoric is based on the impressive benchmarks of the Northwood (0.13u, 512 KB) Pentium 4 with a 533 MHz FSB and dual-channel Rambus 1066.
I suspect that you have a Willamette (0.18u, 256 KB) Pentium 4 with a 400 MHz FSB and Rambus 800 or DDR.
I agree that the rhetoric is exaggerated, but the Pentium 4's high-MHz design is starting to pay off.
If AMD is to stay competitive, they'll have to look at a larger L2 cache and a faster FSB to match the DDR 166 and 200.
ready to roll out
whatsoever needed to get the new Thoroughbred because it can squeeze out
couple of more 3DMark points.
Clearly, the Thoroughbred would be a more compelling upgrade with a bigger L2 cache and a faster FSB, but the die shrink is worth something: it's down about 10 W.
Still, for a quiet system, I'd consider the 1 GHz C3, which runs at a miserly 12 W.
It's not possible to approve LGPL without opening up a hole that allows J. Random Megacorp to... use the patents with closed source proprietary apps
I really wish Red Hat hadn't restricted the licensing of its patents to free software.
I'd much rather see it form a defensive patent pool, as described in "mixing patent and copyright."
Regardless of its seemingly noble intentions, Red Hat is positioning itself as a patent aggressor.
Licensing only to (a subset of) free software is not defensive; it's offensive.
I would rather pay couple of thousand dollars to have stable, well supported API
Maybe that's where you want to spend your money.
A couple of thousand dollars per developer is not a reasonable price for a company experimenting with the idea of supporting this Linux thing.
I have to fight to get $700 for another Solaris compiler license.
just Linux (and a small matter of testing)
on
United Linux is Here
·
· Score: 3, Informative
the fact remains that glibc is glibc, libm is libm, the kernel is the kernel, etc. Red Hat, Caldera, et al differentiate themselves using little chunks of code like package managers and installers, but when the system is installed it's all just Linux.
Actually, distributions also differentiate themselves by adding patches to things like glibc, GCC, and the kernel.
As I mentioned in "Red Hat's little forks," there are over 100 patches in kernel-2.4.18-4.src.rpm, including a 20 MB whopper from Alan Cox.
As I recall, SuSE incorporated ReiserFS, JFS, and LVM before they were in the Linus kernel.
Wearing your optimistic programmer hat, it should still just work.
Wearing the pessimistic hat of a user or a tester, it has to be retested.
It will be interesting to see the extent to which a "Powered by UnitedLinux" distribution is allowed to add patches.
'rpm' runs natively on AIX5L. So in that sense, IBM has already determined that AIX is going the Redhat way.
SuSE, Caldera, and TurboLinux also use RPM.
Connectiva uses APT over RPM.
I don't see anything in the UnitedLinux pages that specifies a package format, but I'd be surprised if it's not RPM.
RedHat will not join while their list of "Main components" includes KDE 3.0 and excludes GNOME completely.
If you look at the main components, it could very well describe Red Hat 8.0, except for Acrobat Reader.
Red Hat 7.3 already ships with KDE 3.0.
Red Hat 8.0 is supposed to be LSB compliant.
It depends on the conditions of joining.
To advertise "Powered by UnitedLinux," does your installer have to force the installation of KDE?
This could be a deal breaker.
Do you really have to ship Acrobat Reader?
This would definitely be a deal breaker.
they enforce the use of KDE 3, which means I'm not interested, thanks.
I assume you're looking at the main components on the release plans page.
I don't know that UnitedLinux enforces the use of KDE, so much as ensures its availability.
Nonetheless, I think it's premature to standardize on a desktop environment for Linux.
While the GPL release of Qt was terrific, GTK+ is released under the more business-friendly LGPL.
Developers of non-GPL Qt products (e.g., Opera and Kylix) have to pay a hefty fee to TrollTech.
More to the point, why in the world is Acrobat Reader listed as a major component of UnitedLinux?
It's a proprietary product of a DMCA-wielding company.
Perhaps more egregiously, it's ugly.
The recently released 5.05 for Linux (with no search support, mind you) has no place on a KDE desktop.
Someone else mentioned Redhat feeds into an 'illusion' that businesses want - 'shiny support', etc. It's no illusion....
having a *real person* to call who's been trained on that particular distro is invaluable.
The other aspect of Red Hat (and, perhaps, all significant distributions) is the work that goes into developing a stable combination of packages.
kernel-2.4.18-4.src.rpm is a long way from the generic 2.4.18 kernel: it has over 100 patches, including a 20 MB whopper from Alan Cox.
GCC 2.96 is the most visible fork, but hardly the only one.
It's all free software, the majority of which makes it back to the original project, but Red Hat is the first to take advantage of its own hard work.
That's an advantage.
they're using the "viral" nature of the GPL to create a market for the non-GPL'd version...
A stand-alone package couldn't easily do this, could it?
Same situation for Qt.
They got the KDE developers to standardize on Qt, which creates a market for companies like Borland and Opera to buy commercial licenses.
This continues to be an objection to KDE: it establishes TrollTech as the toll collector on the Linux desktop.
As for more or less stand-alone examples, you have CUPS and GhostScript.
CUPS gives away the print manager and some popular drivers, but sells other drivers.
GhostScript gives away old versions after it has earned enough in the first round.
This strategy of dual licensing requires owning the copyright to all the code.
Unless your user base feels very charitable, you end up doing all the development yourself.
So yes, you can easily release a crippled GPL version as a loss leader, while selling a more full featured, proprietary version.
However, you risk someone adding compelling features to the free version that you can't use.
This is kind of the situation that TransGaming has put itself into.
It counted on a free stream of development from the WINE developers.
Its compelling added value is the copy protection code, which won't soon find its way into WINE proper.
A lone programmer has a written a new program...
How does this programmer buy food to eat?
This notion of the "lone programmer" writing the Great American Program in his basement would be absurd if it didn't describe the beginnings of some of the most successful free software projects.
We have to stop thinking of free software as a business model for producers and look at it from the point of view of the consumers.
California formed a department specifically to advise on IT issues.
The state accepted that department's recommendation to buy millions of dollars of Oracle software, apparently far in excess of its need.
California needs software.
Should it enter an absurd contract with a proprietary software company, or should it spend the money on free software that is guaranteed to be there tomorrow, and that won't cost a penny more if the state hires more employees?
Sun, on the other hand, has contracted with Wipro, an SEI CMM level 5 outfit in India to enhance Metacity as a GNOME window manager.
They can do this because it's free software.
There's no reason why they can't continue proprietary development against the LGPL tree.
Actually, someone asked in a TransGaming forum for more frequent binaries.
The official reply was that the copy-protection code is not remotely modular, so it takes a fair amount of release engineering and testing to go from CVS to a full binary release.
I assume that this is why they don't find an LGPL WINE useful.
Slashdot Mirror
The MiniDisc is not so mini, and it requires lossy compression to store a full album.
I've never seen a non-Sony MD player. Is the technology licensed to any other companies?
The first question is whether a work is derivative. Much of the debate over the "viral" nature of the GNU GPL is actually a debate over what constitutes a derivative work. Does inclusion of a header constitute derivation? What about a plugin, which may essentially use a naming convention?
I don't know the ins and outs of NWN modules, but I wouldn't dismiss the possibility that one could be held to be derivative.
While the enforceability of EULAs is still being tested in the courts and tweaked by the likes of UCITA, it is copyright law that dominates, here. Copyright law grants to the author the exclusive right to create derivative works. You cannot mod, so to speak, "Gone With the Wind."
There was (and may still be) a group working on a Matrix FPS mod. It's a waste of time, because Warner Bros. has granted an exclusive Matrix license to a commercial game company. If the mod is any good, it will be suppressed, just like the Alien TC for Doom.
The NWN EULA adds to, rather than detracts from, fair use. I just don't find the additional grants compelling.
Mandrake 8.2 shipped with KDE 2, but now has an update to KDE 3. It wouldn't be too much of a stretch to do the same for GNOME 2. I'd give it a couple of months, though, for testing.
I don't know that GNOME 2 will drive up too many distro version numbers. It's pretty safe to assume that Red Hat is already working on 8.0 with GCC 3.
Not to start a GPL-free v. BSD-free flamefest, but the Tux, real-time, and secure Linux patents harm BSD, which is part of the free software community.
A patent is least harmful as part of a patent pool, as described in "Mutual Defense Against Software Patents."
The GNU GPL is not an implied patent license. Its purpose is not "to induce you to infringe any patents or other property right claims or to contest validity of any such claims." However, if I release software under the GPL, but withhold the patent, then the release is a sham. If I don't own the copyright to all the code, then it may be illegal.
The point is, we'll have to look carefully at the patent license for boilerplate lawyerese, like a termination clause, that would be incompatible with the GPL. A GPL-compatible patent license must be written very broadly.
I couldn't agree with you more. I keep reading that game companies already have more game ideas than they can use. Then why do they keep reusing the same old ones?
The game industry is still in its infancy, still discovering its medium. Gamers still ooh and aah at the latest 3-D engine out of Texas. Simple mechanics are still a differentiator among flight sims and driving games.
From time to time, PC Gamer publishes a list of the best games "of all time." Is anyone really still playing Duke Nukem, Doom, or Wing Commander? Does anyone play WarCraft, instead of its sequels? Strange, because movie sequels are lucky to be considered in the same league as the original.
Maybe we just have to wait out Moore's Law, but I think there will be opportunities for indie developers to show up the big game studios with a new idea, or two. I don't want to take anything away from the 99% perspiration of games like Half-Life and Max Payne. I just want to see that 1% inspiration show up a little more prominently.
First, comparing bnetd developers to Al-Qaeda terrorists pretty much puts you on the losing side of Godwin's law.
Blizzard's harrassment of the bnetd project has nothing to do with Linux or open source. It's an abuse of the legal system. Likewise, Adobe's abuse of the DMCA should offend you, even if you're not interested in cracking eBooks.
TruClusters can use Ethernet as the cluster interconnect, but Memory Channel is recommended, because of its lower latency.
I was also thinking of Oracle 9i RAC. The Oracle / Dell / Red Hat bundle uses Gigabit Ethernet. I'm not familiar with the Oracle cluster file system, but I suspect that it would benefit from a low-latency interconnect.
HP's TruCluster is designed to use a proprietary cluster interconnect called Memory Channel, with a bandwidth of about 100 MB/s. Gigabit Ethernet is cheaper, but can't compete when it comes to latency. Any idea how we can expect this new standard to compare?
If a network has more than two nodes, as most networks do, then each node isn't expected to saturate the network. Think of adding more lanes to highway, rather than increasing the speed limit.
I saw a neat demo at a Maya seminar last week. They're looking to take panoramas to the next level--not only can you rotate the camera, but you can translate it. Combine this with some effects (e.g., moving water) and low-end compositing, and you can put together a passable demo reel with little or no video footage.
As for production, we're seeing subtle effects in main-stream films. I was amazed by the behind-the-scenes footage of Cast Away, because I had no idea the extent to which CGI was used. This will eventually filter down even to the million dollar or less films. The difference between Linux and proprietary Unix will make a difference, there.
Tom's rhetoric is based on the impressive benchmarks of the Northwood (0.13u, 512 KB) Pentium 4 with a 533 MHz FSB and dual-channel Rambus 1066. I suspect that you have a Willamette (0.18u, 256 KB) Pentium 4 with a 400 MHz FSB and Rambus 800 or DDR.
I agree that the rhetoric is exaggerated, but the Pentium 4's high-MHz design is starting to pay off. If AMD is to stay competitive, they'll have to look at a larger L2 cache and a faster FSB to match the DDR 166 and 200.
Clearly, the Thoroughbred would be a more compelling upgrade with a bigger L2 cache and a faster FSB, but the die shrink is worth something: it's down about 10 W.
Still, for a quiet system, I'd consider the 1 GHz C3, which runs at a miserly 12 W.
I really wish Red Hat hadn't restricted the licensing of its patents to free software. I'd much rather see it form a defensive patent pool, as described in "mixing patent and copyright."
Regardless of its seemingly noble intentions, Red Hat is positioning itself as a patent aggressor. Licensing only to (a subset of) free software is not defensive; it's offensive.
Maybe that's where you want to spend your money. A couple of thousand dollars per developer is not a reasonable price for a company experimenting with the idea of supporting this Linux thing. I have to fight to get $700 for another Solaris compiler license.
Actually, distributions also differentiate themselves by adding patches to things like glibc, GCC, and the kernel.
As I mentioned in "Red Hat's little forks," there are over 100 patches in kernel-2.4.18-4.src.rpm, including a 20 MB whopper from Alan Cox. As I recall, SuSE incorporated ReiserFS, JFS, and LVM before they were in the Linus kernel.
Wearing your optimistic programmer hat, it should still just work. Wearing the pessimistic hat of a user or a tester, it has to be retested. It will be interesting to see the extent to which a "Powered by UnitedLinux" distribution is allowed to add patches.
SuSE, Caldera, and TurboLinux also use RPM. Connectiva uses APT over RPM. I don't see anything in the UnitedLinux pages that specifies a package format, but I'd be surprised if it's not RPM.
If you look at the main components, it could very well describe Red Hat 8.0, except for Acrobat Reader. Red Hat 7.3 already ships with KDE 3.0. Red Hat 8.0 is supposed to be LSB compliant.
It depends on the conditions of joining. To advertise "Powered by UnitedLinux," does your installer have to force the installation of KDE? This could be a deal breaker. Do you really have to ship Acrobat Reader? This would definitely be a deal breaker.
I assume you're looking at the main components on the release plans page. I don't know that UnitedLinux enforces the use of KDE, so much as ensures its availability. Nonetheless, I think it's premature to standardize on a desktop environment for Linux. While the GPL release of Qt was terrific, GTK+ is released under the more business-friendly LGPL. Developers of non-GPL Qt products (e.g., Opera and Kylix) have to pay a hefty fee to TrollTech.
More to the point, why in the world is Acrobat Reader listed as a major component of UnitedLinux? It's a proprietary product of a DMCA-wielding company. Perhaps more egregiously, it's ugly. The recently released 5.05 for Linux (with no search support, mind you) has no place on a KDE desktop.
The other aspect of Red Hat (and, perhaps, all significant distributions) is the work that goes into developing a stable combination of packages. kernel-2.4.18-4.src.rpm is a long way from the generic 2.4.18 kernel: it has over 100 patches, including a 20 MB whopper from Alan Cox. GCC 2.96 is the most visible fork, but hardly the only one.
It's all free software, the majority of which makes it back to the original project, but Red Hat is the first to take advantage of its own hard work. That's an advantage.
Well, that's only if I get my six weeks vacation.
Same situation for Qt. They got the KDE developers to standardize on Qt, which creates a market for companies like Borland and Opera to buy commercial licenses. This continues to be an objection to KDE: it establishes TrollTech as the toll collector on the Linux desktop.
As for more or less stand-alone examples, you have CUPS and GhostScript. CUPS gives away the print manager and some popular drivers, but sells other drivers. GhostScript gives away old versions after it has earned enough in the first round.
This strategy of dual licensing requires owning the copyright to all the code. Unless your user base feels very charitable, you end up doing all the development yourself.
So yes, you can easily release a crippled GPL version as a loss leader, while selling a more full featured, proprietary version. However, you risk someone adding compelling features to the free version that you can't use. This is kind of the situation that TransGaming has put itself into. It counted on a free stream of development from the WINE developers. Its compelling added value is the copy protection code, which won't soon find its way into WINE proper.
This notion of the "lone programmer" writing the Great American Program in his basement would be absurd if it didn't describe the beginnings of some of the most successful free software projects. We have to stop thinking of free software as a business model for producers and look at it from the point of view of the consumers.
California formed a department specifically to advise on IT issues. The state accepted that department's recommendation to buy millions of dollars of Oracle software, apparently far in excess of its need. California needs software. Should it enter an absurd contract with a proprietary software company, or should it spend the money on free software that is guaranteed to be there tomorrow, and that won't cost a penny more if the state hires more employees?
The UK spent half a billion dollars on air traffic control software that doesn't work very well.
Sun, on the other hand, has contracted with Wipro, an SEI CMM level 5 outfit in India to enhance Metacity as a GNOME window manager. They can do this because it's free software.
Actually, someone asked in a TransGaming forum for more frequent binaries. The official reply was that the copy-protection code is not remotely modular, so it takes a fair amount of release engineering and testing to go from CVS to a full binary release. I assume that this is why they don't find an LGPL WINE useful.