That's a pretty uneven comparison. Hardware manufacturers sell hardware not drivers! If they can get someone else to write drivers for them their costs will go down. They currently pay people to write drivers for commercial operating systems. Why wouldn't they want free drivers for free operating systems?
It has puzzled me for a while how hardware companies can be so stupid as to not be giving kernel developers documentation for their hardware. They give a lot of reasons for it but none of them seem to stand up to logic well. You have to be an IP lawyer to get it I guess.
Third party drivers run inside the kernel. If they have security flaws there's nothing the rest of the kernel can do about it. Even a microkernel OS will have a hard time being completely secure without trusting the drivers. At some point it's going to have to touch hardware and it's not easy to abstract that away. After all that's what the device driver is there for in the first place. It's not Apple's fault if someone released a crappy device driver. This is why I like all my Linux drivers to be free instead of that binary crap ATI/Nvidia do. Go Intel!
My computer is running fine networked without inetd/portmap. So are my servers. Inetd is only needed for services that don't do their own daemon and these days that's pretty much none. Portmap is used for RPC so if you're running NFS/NIS you might still need it but it certainly isn't a standard thing either these days. Distros should not enable these by default since they're very much corner cases now.
Having >4GB of memory is easy today and 64 bit pointers makes that transparently usable to a single process without jumping through hoops. In a laptop chip it's going to take longer to get there but the current ones can already take 2GB so it should actually be pretty soon.
An easier way to do this is probably to crop the image, say 10 pixels all around the image and only publish that. Then only you can show the image with the extra pixels all around showing that you have the original and no one else.
>and have their first generation cams fighting gen 3 and gen 4 cameras from Canon and Nikon. No chance.
Funny you should say that because the 7D/5D blow away the offerings from Nikon and go head-to-head with Canon. Reviews consistently put the 5D above Nikon's D70 which is supposedely in another range. The 7D is a good competitor to the Canon 20D only losing on the pixel count and winning in other places. Since Sony now owns the camera division and makes it's own sensors, we're in for a some real competition.
>Nikon continues to make their top of the line F6. It's hard to imagine a better 35mm SLR.
In a lot of ways the Minolta Maxxum/Dynax 7 and 9 film cameras are better than the F6. Ergonomics being the greatest difference. Anyone that has used a recent Minolta body and then uses one from Nikon or Canon will really feel the difference. Minolta cameras are designed by photographers. While the Canon and Nikon cameras will make you press 3 buttons and check a menu the Minolta will have an old-fashioned knob that you can set without taking your eye out from the viewfinder.
Sure but this doesn't mean decrypting a DVD shouldn't be ilegal, it just means making it illegal is very ineffective.
The reason it shouldn't be illegal is that viewing the DVD I just bought is perfectly within my rights and for that I need to decrypt it. If we were talking about military secrets it would be legitimate to expect that decrypting them would constitute a crime[1] because I don't have the right to view that information. I do have the right to view the information in the DVD because I bought it!
[1] Altough I'd expect military secrets to have real encryption.
Yes but designing your own OS from scratch does not give you the right to install it on Apples hardware for instance.
Sure it does. If I bought said apple, it's mine. I can smash it with a hammer, drop it in an acid bath or do any other crazy thing with it including installing my own software in it. I shouldn't expect Apple to help me do any of these things, but I do expect Apple to stay out of the way while I do them since after all the hardware is mine.
With DVD's the same happens. I bought the right to view that movie. If I want to put the DVD under a microscope write down all the bits, do the CSS math in my head, do the MPEG decoding in my head, and then create an image of each frame I can do it. There's nothing wrong about this since it's a fair use of the DVD. The same is true for viewing it in my linux PC. I'm not creating copies or doing a public broadcast I'm just viewing the freaking movie, and that is just plain legal.
JV: I do not believe that you have the right to override an encryption. Because if you have the right to do it, everybody can do it. For whatever benign reason you have, somebody else has got one even more benign. But once you let one person deal in a digital copy -- and I don't have to tell you; you know far better than I that, unlike in analog, the ten thousandth copy is as pure as the original -- it is a big problem. So once you let the barriers down for your perfectly sensible reason, you gotta let it down for everybody.
The problem witht this argument is that it says that what should be illegal is breaking the encryption and not actually doing something with the decrypted data. I agree that taking a DVD, transcoding it to divx and posting it on the web should be illegal, but the illegal part shouldn't be decryption of the data but the actual copying and posting on the web. This actually solves the problems of both sides, it lets me see my DVD's and lets the copyright holders have their rights. Why doesn't the MPAA want this?
Re:So You Prefer Fragmentation over Cooperation
on
XFree86 Alters License
·
· Score: 2, Insightful
Now I'm sure I've made a mistake there, but I've never been able to find out what it was:)
Your mistake is not reading the license. The GPL says nothing about the code "being re-licensable under later versions of the GPL" it's the COPYING file in most software that says this. The linux kernel is a notable exception since it's licensed under GPL v2 and nothing else:
From the kernel 2.6.0's COPYING file:
Also note that the only valid version of the GPL as far as the kernel
is concerned is _this_ particular version of the license (ie v2, not
v2.2 or v3.x or whatever), unless explicitly otherwise stated.
I should have said that in many implementations of C and C++, time() still returns a 32-bit value.
And this is not a problem at all. Just declare a wider variable and assign the return value of time() to it, an you'll have a wider time to play with. The fact that it returns a 32 bit value will only matter 30+ years from now.
If Acme Software modifies a GPL program and gives it to the UK government only, they are only obligated to provide the source to the UK government. If nobody else has the binaries, nobody else gets the source.
Nope. If you take a GPL program, make changes, and distribute your changes in any way, you have to provide those changes to everyone, you can only keep them secret if you never distribute them.
Quoting the GPL's section 2
b) You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.
so we don't have to hope and pray that a new installation doesn't overwrite a system file with a weird buggy version, or that our OS won't decide to go tits-up in the middle of an important process. Since all us good Slashdotters KNOW there will still be crufty, evil OS's around in 10 years, even if WE aren't using them:-)
Then maybe the solution isn't using aditional bug-prone software to try to recover fast from failures but to actually replace the crufty, evil OS's
Ironic, isn't it, how quickly we forget about the First Amendment when it's somebody else's speech being protected instead of our own?
My country's constitution isn't the same as your's but it doesn't protect the right for other people to force their speech on me. If the US constitution is diferent you need a few more amendments.
Re:Journaling File System: for those who don't kno
on
Looking at Longhorn
·
· Score: 1
When ext3 was created it was a fork form the ext2 code base... so the above is mainly just misinformed/. crap. It's like saying OpenBSD doesn't share code with NetBSD.
I wasn't able to google up a link but I remember an interview with Stephen Tweedie were he stressed the fact that ext3 was a block level implementation of a journaling filesystem that used the ext2 on-disk structure for convenience.
I checked the first revisions of ext3 and it indeed forked from ext2. I didn't know that and apologise for the mis-information.
My point still stands though. Ext3 is not just ext2 with a journal hacked on top, it's a major change to ext2 and the stability of one doesn't reflect on the other.
Re:Journaling File System: for those who don't kno
on
Looking at Longhorn
·
· Score: 2, Informative
My statement about ext3 being more mature than ReiserFS is based on the fact that ext3 is a journal add-on to the now very mature ext2; that is it's an evolution of an older filesystem, not the revolution that is ReiserFS.
This is not true. ext3 and ext2 have the same disk representation but they don't share code, at all. The fact that ext2 is mature doesn't really help ext3. People think ext3 is just ext2 with a few hacks to add journalling but it's actually a block level implementation of a journaling filesystem that just happens to use the same disk layout as ext2 for convenience. Your statement is sort of like saying that the NTFS code in Linux is mature because Windows has had NTFS for a few years now.
I disagree. You can't legally pull my GPL'd code off the net and start selling commercial versions.
Sure I can. If I couldn't do that commercial Linux distributions would be in trouble.
you are absolutely NOT allowed to sell my changes. Get it?
Sure I am. The GPL only makes sure that you get changes back GPL licensed, it doesn't have any restrictions on the buying and selling of software. Get it?
As I said towards the end, though, if you haven't accepted any GPL'd changes, then you have your original, non-released, non-GPL'd version that you retained sole copyright to. That's the version you can sell.
I can sell any version. What I can't do with GPL that I can do with BSD is take the code and change it, not giving my changes back, and sell *that*.
Another thing you're overlooking in the BSD vs. GPL discussion is that while in GPL'd code the changes have to be given back as GPL'd code with BSD code there's no such guarantee. Taking that into account, if you release some piece of code BSD licensed you might not be able to use some of the changes you get back in a version you sell commercially because it's not guaranteed that those changes are covered by the BSD license, with the GPL that doesn't happen since all changes have to be given back as GPL and you can sell them at will.
When you write software, you retain the copyright. When you release it as OS under a BSD-style license, I would agree that you retain the copyright. You can exercise your copy rights and pull the source code off the net to start selling commercial versions for example.
And how's that any diferent in GPL'd code? You can also "pull your source code off the net to start selling commercial versions" with it.
However, when you release it under the GPL, you essentially branch a version of your copyrighted code and release it to the community. You retain the copyright to your original work, but if I submit a change to your GPL version, my change is also GPL'd.
The only diference the BSD license has in this respect is that it doesn't enforce the changes to come back in any way or license.
The confusion is between OS in general and GPL'd OS, which most people (especially the/. crowd) generally equate with OS in general. If you release BSD, your copyright is still enforcable. If you release GPL, I agree with vosbert in saying it's now the community's code.
Both the GPL and the BSD license effectively give the code to the community, what is different between then is what you expect to get back from that code. With GPL changes have to be released back as GPL. With BSD you can pick up the code and run with it. So, the BSD license is more about *giving* to the community where as the GPL is more about the community *sharing* code and making sure that sharing continues.
You can't exercise any of the normal copy rights on GPL'd code; you gave them up when you released it and accepted GPL'd updates.
You can exercise copyright on your original code and not on others code. That doesn't change if we're talking about the GPL, the BSD license or any other license for that matter. What you can do with the BSD license that the GPL doesn't allow is to take BSD code and use it in a close-sourced app. That has nothing to do with exercising "normal copy rights on [...] code", since you don't have copyright on code you didn't write.
Slashdot Mirror
That's a pretty uneven comparison. Hardware manufacturers sell hardware not drivers! If they can get someone else to write drivers for them their costs will go down. They currently pay people to write drivers for commercial operating systems. Why wouldn't they want free drivers for free operating systems?
It has puzzled me for a while how hardware companies can be so stupid as to not be giving kernel developers documentation for their hardware. They give a lot of reasons for it but none of them seem to stand up to logic well. You have to be an IP lawyer to get it I guess.
Third party drivers run inside the kernel. If they have security flaws there's nothing the rest of the kernel can do about it. Even a microkernel OS will have a hard time being completely secure without trusting the drivers. At some point it's going to have to touch hardware and it's not easy to abstract that away. After all that's what the device driver is there for in the first place. It's not Apple's fault if someone released a crappy device driver. This is why I like all my Linux drivers to be free instead of that binary crap ATI/Nvidia do. Go Intel!
My computer is running fine networked without inetd/portmap. So are my servers. Inetd is only needed for services that don't do their own daemon and these days that's pretty much none. Portmap is used for RPC so if you're running NFS/NIS you might still need it but it certainly isn't a standard thing either these days. Distros should not enable these by default since they're very much corner cases now.
Having >4GB of memory is easy today and 64 bit pointers makes that transparently usable to a single process without jumping through hoops. In a laptop chip it's going to take longer to get there but the current ones can already take 2GB so it should actually be pretty soon.
An easier way to do this is probably to crop the image, say 10 pixels all around the image and only publish that. Then only you can show the image with the extra pixels all around showing that you have the original and no one else.
>The thought of being able to handhold sharp shots with Minolta's 28/2 at ISO800, f2 and 1/8th or 1/15th makes me very envious.
:-P
Yep, I do that all the time with a 24/2.8. 1/8 and 1/5 are easy.
>and have their first generation cams fighting gen 3 and gen 4 cameras from Canon and Nikon. No chance.
Funny you should say that because the 7D/5D blow away the offerings from Nikon and go head-to-head with Canon. Reviews consistently put the 5D above Nikon's D70 which is supposedely in another range. The 7D is a good competitor to the Canon 20D only losing on the pixel count and winning in other places. Since Sony now owns the camera division and makes it's own sensors, we're in for a some real competition.
>Nikon continues to make their top of the line F6. It's hard to imagine a better 35mm SLR.
In a lot of ways the Minolta Maxxum/Dynax 7 and 9 film cameras are better than the F6. Ergonomics being the greatest difference. Anyone that has used a recent Minolta body and then uses one from Nikon or Canon will really feel the difference. Minolta cameras are designed by photographers. While the Canon and Nikon cameras will make you press 3 buttons and check a menu the Minolta will have an old-fashioned knob that you can set without taking your eye out from the viewfinder.
For the $400 it costs buy yourself some more memory or disk. Whatever is the bottleneck.
Sure but this doesn't mean decrypting a DVD shouldn't be ilegal, it just means making it illegal is very ineffective.
The reason it shouldn't be illegal is that viewing the DVD I just bought is perfectly within my rights and for that I need to decrypt it. If we were talking about military secrets it would be legitimate to expect that decrypting them would constitute a crime[1] because I don't have the right to view that information. I do have the right to view the information in the DVD because I bought it!
[1] Altough I'd expect military secrets to have real encryption.
Yes but designing your own OS from scratch does not give you the right to install it on Apples hardware for instance.
Sure it does. If I bought said apple, it's mine. I can smash it with a hammer, drop it in an acid bath or do any other crazy thing with it including installing my own software in it. I shouldn't expect Apple to help me do any of these things, but I do expect Apple to stay out of the way while I do them since after all the hardware is mine.
With DVD's the same happens. I bought the right to view that movie. If I want to put the DVD under a microscope write down all the bits, do the CSS math in my head, do the MPEG decoding in my head, and then create an image of each frame I can do it. There's nothing wrong about this since it's a fair use of the DVD. The same is true for viewing it in my linux PC. I'm not creating copies or doing a public broadcast I'm just viewing the freaking movie, and that is just plain legal.
Your mistake is not reading the license. The GPL says nothing about the code "being re-licensable under later versions of the GPL" it's the COPYING file in most software that says this. The linux kernel is a notable exception since it's licensed under GPL v2 and nothing else:
From the kernel 2.6.0's COPYING file:
ATAPI Floppy support used to work great for 100MB zips.
And this is not a problem at all. Just declare a wider variable and assign the return value of time() to it, an you'll have a wider time to play with. The fact that it returns a 32 bit value will only matter 30+ years from now.
Nope. If you take a GPL program, make changes, and distribute your changes in any way, you have to provide those changes to everyone, you can only keep them secret if you never distribute them.
Quoting the GPL's section 2
Emphasis mine.You take being a grammar nazi to a hole new level.
Then maybe the solution isn't using aditional bug-prone software to try to recover fast from failures but to actually replace the crufty, evil OS's
I wasn't able to google up a link but I remember an interview with Stephen Tweedie were he stressed the fact that ext3 was a block level implementation of a journaling filesystem that used the ext2 on-disk structure for convenience.
I checked the first revisions of ext3 and it indeed forked from ext2. I didn't know that and apologise for the mis-information.
My point still stands though. Ext3 is not just ext2 with a journal hacked on top, it's a major change to ext2 and the stability of one doesn't reflect on the other.
This is not true. ext3 and ext2 have the same disk representation but they don't share code, at all. The fact that ext2 is mature doesn't really help ext3. People think ext3 is just ext2 with a few hacks to add journalling but it's actually a block level implementation of a journaling filesystem that just happens to use the same disk layout as ext2 for convenience. Your statement is sort of like saying that the NTFS code in Linux is mature because Windows has had NTFS for a few years now.
Sure I can. If I couldn't do that commercial Linux distributions would be in trouble.
Sure I am. The GPL only makes sure that you get changes back GPL licensed, it doesn't have any restrictions on the buying and selling of software. Get it?
I can sell any version. What I can't do with GPL that I can do with BSD is take the code and change it, not giving my changes back, and sell *that*.Another thing you're overlooking in the BSD vs. GPL discussion is that while in GPL'd code the changes have to be given back as GPL'd code with BSD code there's no such guarantee. Taking that into account, if you release some piece of code BSD licensed you might not be able to use some of the changes you get back in a version you sell commercially because it's not guaranteed that those changes are covered by the BSD license, with the GPL that doesn't happen since all changes have to be given back as GPL and you can sell them at will.
And how's that any diferent in GPL'd code? You can also "pull your source code off the net to start selling commercial versions" with it.
The only diference the BSD license has in this respect is that it doesn't enforce the changes to come back in any way or license.
Both the GPL and the BSD license effectively give the code to the community, what is different between then is what you expect to get back from that code. With GPL changes have to be released back as GPL. With BSD you can pick up the code and run with it. So, the BSD license is more about *giving* to the community where as the GPL is more about the community *sharing* code and making sure that sharing continues.
You can exercise copyright on your original code and not on others code. That doesn't change if we're talking about the GPL, the BSD license or any other license for that matter. What you can do with the BSD license that the GPL doesn't allow is to take BSD code and use it in a close-sourced app. That has nothing to do with exercising "normal copy rights on [...] code", since you don't have copyright on code you didn't write.How did you arrive at this figure? By my calculations 0.05% applied ten times gives 0.501126501% of growth.