Is it under a 4-clause or 3-clause BSD license? OpenSSL is _still_ under the 4-clause license, with the `obnoxious advertising clause' which says that you have to mention the developers in all advertising materials. Not such a big deal, you might say, but there are two big problems with this: 1) It's incompatible with GNU GPL, so no straight GPL software can use OpenSSL, and 2) it causes huge practical problems.
File something, anything, even if you know it won't stand up in court, even if you know there's well known prior art. Find a law firm willing to work for a portion of the settlement you'll get when someone happens to infringe on your obvious idea.
As for the complaint that NVIDIA is no better than ATI because of a binary driver release: that is not NVIDIA's fault.
Yes, it is. If they say differently they are lying. Where is the register-level documentation for NVidia cards? NVidia owns the rights to that, but I don't see that anywhere.
VIDIA tries to make as much of their driver open source as possible (which is kind of a necessity because of the plethora of kernel configurations out there).
No, they don't. The only thing that they release as `open source' is a stub that lets them a) support more than one kernel version and b) avoid complying with the spirit of the kernel license.
However, the closed-source portions are kept closed because of SGI's patents on OpenGL. Assign blame where blame is due, please
Another lie. Why would SGI stipulate that the code must remain closed? And if they did, why has NVidia not publically encouraged them to change their mind? Anyhow, if it's impossible to open up some parts, why hasn't the rest been released? Even if it's incomplete and doesn't compile, it's something!
In conclusion: maybe NVidia can't Free the source to their drivers for some reason, but they haven't told anyone the real reason. Methinks that it's just some upper level management folks having trouble with the idea of being honest and open with the world.
Seriously, why should they be able to stop people?
If I buy a copy of some mix CD that happens to sample Britney, surely records companies don't actually think they're missing out on a Britney sale? I'm not even a 'potential' customer, so they're not even losing a pretend "if it weren't for napster we would have sold 10 trillion copies of the latest Backstreet boys album, therefore napster has to pay us <pinkypoint to="mouth">one hundred trillion dollars</pinkypoint>"-type sale.
I think everyone can accept that the 2000 US Presidential election was a complete and utter fuckup (surely even the 'winners' will agree). The solution now seems to be to come up with more and more complicated, computersied ways to fix it. Why not just use paper and goddamn pencils? It's works fine here in Australia; a good 10 million people vote in each election, but the election is almost always decided by about 11pm on the night of the vote!
The AEC (the national body in charge of running Federal elections) has reported that there is only one case of enrolment fraud in every two hundred thousand voters.
I think people sometimes need to think 'do we want this complicated electronic system because it will produce a fairer, quicker result or is it just cooler?'
tech:"HP tech saupport, how can I help you?" you:"Well, I've got a problem with my printer. It's an 845c and prints pages upside down." tech:"We've had a few reports about this, and we've come up with a fix. When then pages come out of the printer, rotate them 180 degrees." you:"Thanks!"
Don't tell me a large company bought out a smaller competitor, just so they could kill them!
Next you'll be telling me MS has been buying up small companies to save them doing research or that Disney has been buying senators so they don't have to produce a useful product!
One interesting thing about OS X is that it supports 'fat binaries', which are binary files that contain executables for different architectures. If Apple ever releases OS X for Intel, and it's source compatible with OS X for PPC, developers can ship a CD with their app compiled for both architectures, and it will Just Work(tm) on whichever platform. Sure, it's a waste of space, but disk is cheap and convenience is king.
Uh, where? Wine isn't an emulator; it's a Free re-implementation of the Win32 API, plus some nifty loader software to load Windows binaries and execute them natively.
An emulator is something that emulates a hardware platform, allowing all software written for that platform to run, but at a hugely reduced speed.
Don't be a moron. Australia has in the top ten in the world for a) home computer numbers, b) home internet usage, c) and mobile phones. Oh wait, we're ahead of the US and the UK with mobile phones. About the only thing I think we're behind with is broadband, but that's due to a moronic government going nuts with the largest telco in the country.
Uh...y'know, they may have a pile of money, but they still don't have a product people want.
LA: "Uh, hello, would you like to buy music in an encrypted format that will only play on one machine and can be revoked at any time"
User: "You want me to pay for music in an incompatible format, that won't play on my friends computer and that you can take away whenever you feel like it. Yeah, I'll get right on that, once I finish this download from Kazaa"
Re:Slashdotted, but GNOME2 *is* leagues better
on
KDE Gets The Hat
·
· Score: 1
Yes the issues were fixed, no RMS is way not cool with the QPL. The solution was for Trolltech to license Qt under a dual GPL/QPL license. I can't think of anyone out there who actually uses the QPL side of it, but people are certainly happy with the GPL part.
Re:New meaning to "Red" in Red Hat
on
KDE Gets The Hat
·
· Score: 2, Informative
+3 Informative? More like -1, Talking Out My Arse.
A rather important Boolean expression: Sarge == Testing != Unstable.
Firstly, for people who aren't up with the whole gcc 3.2 thing: The newest GCC release, 3.2, changes the C++ ABI. The ABI is basically the set of rules for how binary programs call functions in libraries, and how those functions are named in the libraries. Because C++ supports things like polymorphism and overloading, you have to 'mangle' the function names so that int blah(int k) gets a different name, in the library, to int blah(string k). The rules for how to mangle functions names has changed from release-to-release, but GCC 3.2 is (supposed) to be the last change for a long time.
The problem is that all current C++ libraries and programs (in Debian at least) are compiled with GCC 2.95 and they can't (easily) interoperate with things compiled with GCC 3.2. So, to switch to GCC 3.2, you pretty much have to recompile everything with GCC 3.2, which is a bitch. I say 'pretty much' because there are a few ways to hack around the problem, but none of them are very neat. This is an especially big problem for Debian because it allows upgrades from release to release; I can stick in some install disks from Debian 1.0 and (if my hardware is still supported;) install Debian, configure it and upgrade to the current unstable with out a reboot. Thus, there needs to be some way to move from GCC 2.95 C++ programs and libraries to the GCC 3.2-compiled versions of same with breaking things in the process.
Why Sarge won't break anyway: Packages only hit testing after surviving in unstable for at least two weeks without a Release Critical bug report. Thus, the gcc 3.2 transition will happen in unstable, and pretty much everything (C++ stuff anyhow) will be re-compiled with gcc 3.2. When everything is pretty much working, it will trickle down into Sarge (testing), no breakage required.
Slashdot Mirror
Is it under a 4-clause or 3-clause BSD license? OpenSSL is _still_ under the 4-clause license, with the `obnoxious advertising clause' which says that you have to mention the developers in all advertising materials.
Not such a big deal, you might say, but there are two big problems with this: 1) It's incompatible with GNU GPL, so no straight GPL software can use OpenSSL, and 2) it causes huge practical problems.
Theses issues are a big problems for Debian, in particular.
What if you're using it on a non-Intel platform?
File something, anything, even if you know it won't stand up in court, even if you know there's well known prior art. Find a law firm willing to work for a portion of the settlement you'll get when someone happens to infringe on your obvious idea.
It will definitely make your case look tidier and make later work inside it more easy.
All this technical discussion threw me off the scent for a while, but I was right, it's Slashdot!
[Uh, yeh, you've got to mod me up now. I'll settle for a Funny, but an Insightful would give everyone a laugh.]
As for the complaint that NVIDIA is no better than ATI because of a binary driver release: that is not NVIDIA's fault.
Yes, it is. If they say differently they are lying. Where is the register-level documentation for NVidia cards? NVidia owns the rights to that, but I don't see that anywhere.
VIDIA tries to make as much of their driver open source as possible (which is kind of a necessity because of the plethora of kernel configurations out there).
No, they don't. The only thing that they release as `open source' is a stub that lets them a) support more than one kernel version and b) avoid complying with the spirit of the kernel license.
However, the closed-source portions are kept closed because of SGI's patents on OpenGL. Assign blame where blame is due, please
Another lie. Why would SGI stipulate that the code must remain closed? And if they did, why has NVidia not publically encouraged them to change their mind? Anyhow, if it's impossible to open up some parts, why hasn't the rest been released? Even if it's incomplete and doesn't compile, it's something!
In conclusion: maybe NVidia can't Free the source to their drivers for some reason, but they haven't told anyone the real reason. Methinks that it's just some upper level management folks having trouble with the idea of being honest and open with the world.
If your local CS department won't let you use some of their space, then you must be in Waterloo.
Seriously, why should they be able to stop people?
If I buy a copy of some mix CD that happens to sample Britney, surely records companies don't actually think they're missing out on a Britney sale? I'm not even a 'potential' customer, so they're not even losing a pretend "if it weren't for napster we would have sold 10 trillion copies of the latest Backstreet boys album, therefore napster has to pay us <pinkypoint to="mouth">one hundred trillion dollars</pinkypoint>"-type sale.
I think everyone can accept that the 2000 US Presidential election was a complete and utter fuckup (surely even the 'winners' will agree). The solution now seems to be to come up with more and more complicated, computersied ways to fix it. Why not just use paper and goddamn pencils? It's works fine here in Australia; a good 10 million people vote in each election, but the election is almost always decided by about 11pm on the night of the vote!
.
The AEC (the national body in charge of running Federal elections) has reported that there is only one case of enrolment fraud in every two hundred thousand voters
I think people sometimes need to think 'do we want this complicated electronic system because it will produce a fairer, quicker result or is it just cooler?'
And since I'm asking, does anyone know of a good software solution for backing up a database without stopping it?
Get a real database. Seriously. Oracle, DB2, SapDB, PostgreSQL all support online backups.
You're a tech support nightmare!
tech:"HP tech saupport, how can I help you?"
you:"Well, I've got a problem with my printer. It's an 845c and prints pages upside down."
tech:"We've had a few reports about this, and we've come up with a fix. When then pages come out of the printer, rotate them 180 degrees."
you:"Thanks!"
Or for OSS-based systems: vsound.
Someone actually completed this game flawlessly.
Can I have a +5 too, if I complain about the meta-meta-irony of commenting on a repost of a post attached to a repost of an article?
Don't tell me a large company bought out a smaller competitor, just so they could kill them!
Next you'll be telling me MS has been buying up small companies to save them doing research or that Disney has been buying senators so they don't have to produce a useful product!
One interesting thing about OS X is that it supports 'fat binaries', which are binary files that contain executables for different architectures. If Apple ever releases OS X for Intel, and it's source compatible with OS X for PPC, developers can ship a CD with their app compiled for both architectures, and it will Just Work(tm) on whichever platform. Sure, it's a waste of space, but disk is cheap and convenience is king.
Uh, where? Wine isn't an emulator; it's a Free re-implementation of the Win32 API, plus some nifty loader software to load Windows binaries and execute them natively.
An emulator is something that emulates a hardware platform, allowing all software written for that platform to run, but at a hugely reduced speed.
Don't be a moron. Australia has in the top ten in the world for a) home computer numbers, b) home internet usage, c) and mobile phones. Oh wait, we're ahead of the US and the UK with mobile phones.
About the only thing I think we're behind with is broadband, but that's due to a moronic government going nuts with the largest telco in the country.
Also 90% of the workers in my building are female.
Uh, which building is that...I may need to 'investigate' it myself;)
Hurd this year!
Even more impressive than finding a 'layer' to 'server' papers would be finding a correctly speled post in this thread;)
Face it, there's no unbreakable copy protection except for a populace who refuses to copy copyrighted works!
Or one that fears their government .
Uh...y'know, they may have a pile of money, but they still don't have a product people want.
LA: "Uh, hello, would you like to buy music in an encrypted format that will only play on one machine and can be revoked at any time"
User: "You want me to pay for music in an incompatible format, that won't play on my friends computer and that you can take away whenever you feel like it. Yeah, I'll get right on that, once I finish this download from Kazaa"
Fuck Yes!
Yes the issues were fixed, no RMS is way not cool with the QPL. The solution was for Trolltech to license Qt under a dual GPL/QPL license. I can't think of anyone out there who actually uses the QPL side of it, but people are certainly happy with the GPL part.
It ain't GPL-compatible, but it is Open.
+3 Informative? More like -1, Talking Out My Arse.
A rather important Boolean expression:
Sarge == Testing != Unstable.
Firstly, for people who aren't up with the whole gcc 3.2 thing:
The newest GCC release, 3.2, changes the C++ ABI. The ABI is basically the set of rules for how binary programs call functions in libraries, and how those functions are named in the libraries. Because C++ supports things like polymorphism and overloading, you have to 'mangle' the function names so that int blah(int k) gets a different name, in the library, to int blah(string k). The rules for how to mangle functions names has changed from release-to-release, but GCC 3.2 is (supposed) to be the last change for a long time.
The problem is that all current C++ libraries and programs (in Debian at least) are compiled with GCC 2.95 and they can't (easily) interoperate with things compiled with GCC 3.2. So, to switch to GCC 3.2, you pretty much have to recompile everything with GCC 3.2, which is a bitch.
I say 'pretty much' because there are a few ways to hack around the problem, but none of them are very neat. This is an especially big problem for Debian because it allows upgrades from release to release; I can stick in some install disks from Debian 1.0 and (if my hardware is still supported;) install Debian, configure it and upgrade to the current unstable with out a reboot. Thus, there needs to be some way to move from GCC 2.95 C++ programs and libraries to the GCC 3.2-compiled versions of same with breaking things in the process.
Why Sarge won't break anyway:
Packages only hit testing after surviving in unstable for at least two weeks without a Release Critical bug report. Thus, the gcc 3.2 transition will happen in unstable, and pretty much everything (C++ stuff anyhow) will be re-compiled with gcc 3.2. When everything is pretty much working, it will trickle down into Sarge (testing), no breakage required.