2.6.x.y are temporary branches with critical and trivial fixes which cause “real” problems to end users. The patches in those branches are fairly small.
That said 2.6.11.y is dead now because 2.6.12 was released today, so the next maintenance release will be 2.6.12.1.
Is actually far superior, like FreeBSD’s development system.
Having to wait for huge release cycles is ridiculous, remember how long a lot of people had to wait for USB support and what not because the only kernels which had it were unstable and/or heavily patched ones available only from some distros.
The new method is a much more incremental and deliberate pace.
But in the new versioning system, aren’t the odd point releases basically relatively stable development releases of the even point releases? As such, we’re rather unlikely to see a maintenance release for 2.6.12, right?
Based on my understanding of it, I’ve just been avoiding 2.6.11 waiting for 2.6.12; but if I’m wrong maybe I’ll just go and compile 2.6.11.12 now that 2.6.12 is released.
If you want to be sure you use a tested kernel, use the one provided with your distribution
Ugh, no thanks. Arch seems to be better than most, but I’d still rather not have my lack of PCIE slow my boot by 15 seconds, or generally the loads and loads of stuff I just don’t need in there.
the question of a 2.7 tree is an interesting one.. I think we’ll begin to see it arise when major kernel changes arise(i.e. if I may, when the 2.5 tree moved to ALSA instead of OSS, NTPL instead of linuxthreads), large patchs that change the key parts of the kernel.
You simply can’t change a key kernel architecture between say 2.6.12 and 2.6.13, I kind of like the new pace of development and improvement, but I hope everyday that they will open a 2.7 tree..
@Michael Sailver
Your incorrect in thinking that. It used to be 2.even(2.2, 2.4)were stable and 2.odd were upstable(2.3, 2.5), but now its only the 2.6 tree. There isn’t any(planned) stability difference between 2.6.11 and 2.6.12.
If you guys don’t have this link already, take a look:
Any of you guys know what the status of Smart Battery support in this kernel version is? I have an acer laptop which has this kind of battery and it does not read its power properly.
Because not all of us have 1,500-2,500 for a midrange iMac or PowerMac.
and although apple is good, I for example only have enough money to buy a new computer every 4-5 years. I got one in 2003, so I guess I’ve gotta make use with what I have, don’t I?
Yeah, I realize the odd tree being development in the past. I had heard that, in the new development model (after community comments about quantity of maintenance releases, I assumed), they were going to aim for even points to be the culmination of heavier development in odd points.
But I guess I should give in and search out comments from Linus and Alan before I decide on how to keep up to date without having to build a new kernel every week or two.
Ivan:
If the truth matters so much, why do you support a company that sues it’s customers when their employee’s leak them information? Or one that limits what they give back to the community as much as possible? (in the case of khtml)
Maybe you should try extrapolating Apple’s behavior to a company the size of Microsoft.
1.) Make your script that installs it on a machine
2.) Name it foo.sh
foreach some-list-of-all-machines
scp foo.sh
ssh host ./foo.sh
Well, it’ll likely be more than 3 lines but you get the idea.
And of course the script to build the kernel will be a bit longer as well.. Course, if you wanted you could probably also just build it once and scp all your install files carefully (bad idea I think).
Can you block IPs and or Ip ranges of illicit users? that would probably help.
as for Apple’s behavoir, its better then you could expect of nearly any other commercial company, they did open source darwin, thats a major step beyond what nearly any other company would do.(Darwin’s core is BSD liceanced, unlike GPL they weren’t required to open source it, they just had to “give credit”)
Can you block IPs and or Ip ranges of illicit users? that would probably help.
Won’t solve the problem. May slow them down, those that are not too knowledgable, but the real aggressive trolls will have already known how to defeat such measures. There are trivial ways around an ip banning.
as for Apple’s behavoir, its better then you could expect of nearly any other commercial company, they did open source darwin, thats a major step beyond what nearly any other company would do.
Yes Apple should be commended for those actions but then they turn around and do stuff like this:
Just to clarify the situation about kernel release numbers, it was decided that there will be no development tree in the near future unless something really big show up.
Incremental development is taking place in the “stable” tree managed by Linus Torvalds making releases every 2 months approximately.
Heavy development is being done in the -mm tree (and many others trees) which is managed by Andrew Morton who is the “second in charge” and is employed by the OSDL (as well as Linus Torvalds). Once those changes are more or less stabilized, they will be pushed forward to Linus Torvalds tree for wider testing and official acceptance.
So far this development process has been a success. Kernel hackers likes it because changes are smaller and easier to integrate, if an error show up they can find out where that bug come from. This wasn’t easy with the old development model because after 18 or 24 (or 30) months of development no one can tell for sure why certain bug showed up.
Distributors are happy too because new features are introduced faster, hence, they can attract more customers easily. Another advantage for them is that they are able to track more closely the official kernel tree without the need of intrusive patches to implement new features and release them selves from the burden of maintaining large “out of the tree” patches.
Some users likes the new development model because they can get more features faster without the need of waiting 2 years for the development tree to be stabilized.
But, some users are sad because they suffer some stability issues release after release and you can never tell for sure if the next release will work flawlessly on your setup as the previous release did. With the old model this was unlikely to happen (but still possible) because the changes in the stable tree were limited to bug fixes or back porting stable (and simple) features from the development tree.
The most affected users are the ones who likes to compile the kernel from sources. This is not convenient in the new model because the distributor are left with the responsibility of applying “stabilizing” patches if the need show up.
Now that distributors can track the official kernel more closely and get the new features immediately, there is no much need to compile from source.
That’s all for now, I hope this information can help somebody to better understand the new development process and accept it. The only thing it needs is time to mature and when that happens all the disadvantages that comes implicit with the new model will be reduced to a minimum.
I’m using SuSE 9.3 on AMD64,while it’s not *the* fastest it has the greatest amount of usefull applications i tend to use.They have done a good job with 9.3.Can’t wait until 9.4 is released.The only drawback is the bad configurable SuSE firewall2 (by GUI).Would be nice if they would give SF2 the same configurabillity as kmyfirewall.Being able to quickly disabling all ports and opening only those one really needs is in my opinion a plus.
> Give me a three-command walkthrough for deploying this network-wide
> on 185 machines, and then we’ll talk about whether it’s useful or not …
> Windows fails this test, too.
make && cp arch/i386/boot/bzImage /tftpboot
It requires a little infrastructure (netboot enabled workstations) but it’s not too difficult… You might also need to run some sort of symbol-relocation program to make sure the kernel is compressed in such a way that the boot loader can handle it, but that’s still within your three command limit.
To the person who modded me down: You can’t handle the truth!
How do we know what you say is the truth?
Ok, let’s say you ARE saying the truth…did you try updating udev? Have you tried booting the kernel with devfs instead? How do you know the problem is actually in the kernel? How do you explain that others do not have this problem?
I recently came across a surprising bug in Windows 2000. I had a word document with a particular embedded font (jester.ttf). Opening the document on a Win2K PC that didn’t have the jester font (which is why I embedded the font in the first place) caused the PC to spontaneously reboot. 100% reproducible. Did I run to OSNews to post that “Windows is crap and we must fix it?” No, I didn’t. Because I understood that this was probably an isolated incident, and that what happens on a machine won’t necessarily happen on all machines.
2) did the company certify Jester.ttf to run in WIn2K?
Since when are fonts certified to run under an OS? BTW, the problem was NOT with the font, since OpenOffice opened the document without any problems. So the problem was with MS Office and/or Windows, both made by Microsoft.
In our Linux case, SuSE 9.2 was certified to run on Linux 2.6 kernels.
Did you check to see if this also happened to other SuSE 9.2 users who upgraded to 2.6.12? That should have been the first thing you did, to make sure the problem wasn’t with your setup.
You really don’t have a clue do you?
Wow, that didn’t take too long before you resorted to personal attacks. That just made it even likelier to me that the problem you described was made up. You just lost all credibility on the matter. Thanks for playing.
BTW, I run udev and haven’t had any problems with 2.6.12
However, ATI’s crappy closed drivers once again are broken against a new kernel… doesn’t happen often with nvidia drivers. Listening, ATI? I think not.
Mmm, Ignorant anti-linux trolls without a clue in the world, look at them go!
btw, SuSE 9.2 is certified to run with the kernels they distribute, not ones that you download and compile yourself. They may have things that aren’t in the official kernel tree(or aren’t compiled in)(taking Fedora as an example, Xen, GFS, HAL)
You broke your own system, it was something you download and compiled, not something your distrbuter gave you.
That would be like taking a Dell computer, throwing in a new standard fan or default heatsink, having your computer not work, and whining that it’s all Dell’s fault.
If you want to find a distrbution “certified to run with all 2.6 kernels” then go find one; SuSE is not one of them.
You always need to reinstall the video drivers, they “install” themselves by compiling themselves against kernel code, you change kernels, and it won’t direct, it happens with Nvidia drivers too. Just reinstall the drivers and they should work.
(@the trolls, Microsoft only has a single kernel, made in 2001, so you don’t have the problem with Windows, but yet you are stuck using outdated software(Think XP vs Win98; thats now vs XP) )
I’m aware of how to install the drivers, the problem is the drivers themselves. Compilation errors. ATI has progressed a bit in the last few months, but they’ve still got a pretty poor product in their Linux drivers.
You’d be surprised at how many CS graduates are out there without a fn clue.
University is not the measure of intelligence if it was then why are the majority Uni graduates not very well off? Oh that’s right, we have Uni and then we have the Real World and sometimes they are mutually exclusive.
As for early adopters, why not wait until your distro has the new Kernel in their respective repositories. There might be a reason distros don’t jump on the kernel update bandwagon ASAP, it’s called testing.
Earlier in this thread it was explained how the versioning and tree handling of the Linux development was done. I’d like to emphasize one of the points in that post.
It is the responsibillity of the distributor to patch the kernel they ship for stability.
There are som important things to learn from this.
1. Linux the kernel is not b2c product it is b2b product. No where in the deal is Linus or any other kernel developer responsible for shipping a product usable by end users. It is strictly a component that vendors can use to build operating systems. Nothing more.
2. Linux is not an operating system. Linux is either a kernel or a family of operating systems. So the statment that “linux will never be a desktop os” is true because linux will never be an os. Any comments on the merits of linux as an operating system is just stupid.
Linux is a kernel, just like Darwin is. Darwin is not ready for the desktop since it’s only a kernel. But MacOX X is ready for the desktop, so it’s not Linux fault just like is not Darwin’s fault.
Personally I think that SUSE 9.3, Fedora Core 4 and Ubuntu Hoary are ready for *most* desktops, but if it’s not as good as MacOS X is not Linux’s fault. It’s KDE or GNOME’s fault.
If Quartz and Aqua were ported to Linux (technically that’s possible), would you still think that Linux is not reay for the desktop?
Ant that’s exactly why its improving everyday. the developers are never satisfied. And they don’t need to stop, because of money things. Other words. Development is not dependent of money, but enthousiasme to make things better.
> and you all need to see the truth and grow up
No, i don’t want to see the truth. I want to see a icecream and a lazychair and my wife and child sitting next to me in the sun. Damn, I need to work and it’s very hot in the Netherlands.
Reading stupid trolling comments such as the ones Ivan posted makes me want to stop reading osnews… We’re closing in on /. quality here. I am all for logged in posing only!!
please osnews – stop these childs, who are desperate, cannot handle their feelings. Ivan’s commentary for example, it doesn’t have any need for being posted here….Osnews is a news feed about operatingsystems – not for trolling!
you don’t know nothing about computers, go and play on http://disney.go.com and let the others know the truth, the truth that LINUX SUCKS and osnews as well who post bullshit about linux.
call me when the companies start making it’s products on linux, like games, etc. call me when you have a decent GUI, and not something like full of hacks like x11, call me when you have a decent OS.
x11 is a graphic server..and you are free the use any compatible *nix wdm. Gnu/Linux is about freedom, if you you aren’t choosing a distro that claims you agains an specific wdm(in the beginning, easy to choose another later).
If you don’t like the oppurtunity to maintain and tweak a specific system – like Gnu/Linux – and maintain it , go ahead use another os if you want.
btw, disney.go.com – the games does work at Mozilla Firefox at Gnu/Linux.
The truth is that Linux sucks for you. If you don’t have any need for it nobody thinks less of you.
The reson people are calling you a troll is not because you don’t like Linux it is because you don’t backup your claims with a valid inference(look it up).
As a Windows user, I must admit that I’m quite impressed.
So is this the first release after migrating to GIT ?
I guess it should clean up the base code a little
The git system seems good. See here
http://www.kernel.org/git/
I don’t like this kernel development system… is confusionary and contributing to the Linux kernel is always becoming more difficult…
Remember 2.6.11.12? The next year we will have 2.6.12.33!
At my opinion we must return to the old linux versioning scheme… or we will be kicked from FreeBSD…
Since I didn’t see it before I think this is the first stable release with TPM support.
Very surprising the maintainer of the TPM driver in the userland is a IBM employee…
ops, my english is bad
not
kicked from FreeBSD
but
kicked by FreeBSD
sorry for that, I am italian
2.6.x.y are temporary branches with critical and trivial fixes which cause “real” problems to end users. The patches in those branches are fairly small.
That said 2.6.11.y is dead now because 2.6.12 was released today, so the next maintenance release will be 2.6.12.1.
I hope this clarifies your doubts.
-William
Is actually far superior, like FreeBSD’s development system.
Having to wait for huge release cycles is ridiculous, remember how long a lot of people had to wait for USB support and what not because the only kernels which had it were unstable and/or heavily patched ones available only from some distros.
The new method is a much more incremental and deliberate pace.
I have text article over banners in in linuxtday page, either with mozilla and konqueror.
Correct me if I’m wrong, I probably am
But in the new versioning system, aren’t the odd point releases basically relatively stable development releases of the even point releases? As such, we’re rather unlikely to see a maintenance release for 2.6.12, right?
Based on my understanding of it, I’ve just been avoiding 2.6.11 waiting for 2.6.12; but if I’m wrong maybe I’ll just go and compile 2.6.11.12 now that 2.6.12 is released.
Not sure, but I believe 2.6.11 was the 2.6 kernel with the longest life so far.
Anyway, I hope the alsa support improves and I don’t have any more issues with my SoundBlaster Live! 24 bits.
I am still using Linux Kernel 2.4.26 without problems…
when there will an interesting switch in 2.6 then I’ll do the switch…
@Micheal Salivar
If you want to be sure you use a tested kernel, use the one provided with your distribution
And we’re likely to see a maintenance release of 2.6.12…
According to William’s post 2.6.x.y are maintenance releases… these should do urgent fixes…
Probably Linux Kernel 2.7 branch will born the Summer of next year, i think
correct me, if I am wrong
I had problems with my home PC ‘s CMEDIA PCI CM8738 sound card with Mandrake 9.2 and FC2 in year 2004.
Try to change distribution and it will probably work.
Also I did this (change distro
)
I am still using Linux Kernel 2.4.26 without problems…
when there will an interesting switch in 2.6 then I’ll do the switch…
A good choice. Also Debian 3.1 sarge comed out last day opted for 2.4.27
What OS do you use on those machines?
“Try to change distribution and it will probably work.”
But it is a known issue that the latest SoundBlaster Live! isn’t fully supported yet. It does work in Debian, but not 100%
It does work fine in Fedora 4, but it has just been released.
If you want to be sure you use a tested kernel, use the one provided with your distribution
Ugh, no thanks. Arch seems to be better than most, but I’d still rather not have my lack of PCIE slow my boot by 15 seconds, or generally the loads and loads of stuff I just don’t need in there.
Besides, I like -cko
the question of a 2.7 tree is an interesting one.. I think we’ll begin to see it arise when major kernel changes arise(i.e. if I may, when the 2.5 tree moved to ALSA instead of OSS, NTPL instead of linuxthreads), large patchs that change the key parts of the kernel.
You simply can’t change a key kernel architecture between say 2.6.12 and 2.6.13, I kind of like the new pace of development and improvement, but I hope everyday that they will open a 2.7 tree..
@Michael Sailver
Your incorrect in thinking that. It used to be 2.even(2.2, 2.4)were stable and 2.odd were upstable(2.3, 2.5), but now its only the 2.6 tree. There isn’t any(planned) stability difference between 2.6.11 and 2.6.12.
If you guys don’t have this link already, take a look:
http://ftp.kernel.org/pub/linux/kernel/people/mbligh/abat/regressio…
http://ftp.kernel.org/pub/linux/kernel/people/mbligh/abat/perf/perf…
Any of you guys know what the status of Smart Battery support in this kernel version is? I have an acer laptop which has this kind of battery and it does not read its power properly.
I’ll assume no:
look for(just ctrl+f) it in the changelog
http://kernel.org/pub/linux/kernel/v2.6/ChangeLog-2.6.12
however:
http://sourceforge.net/projects/sbs-linux/
(or if that doesn’t work)
http://www.softtools.net/oem/sbs3.html
What I found strange when looking for it was this: http://www.hypercore.co.jp/opensource/linux-sbs/ seems someone got it working with the 2.2 kernel, heh.
Yeh, okay, I’ll bite.
Because linux performs well compared to OSX, especially on lower end hardware:
http://anandtech.com/mac/showdoc.aspx?i=2436
Because not all of us have 1,500-2,500 for a midrange iMac or PowerMac.
and although apple is good, I for example only have enough money to buy a new computer every 4-5 years. I got one in 2003, so I guess I’ve gotta make use with what I have, don’t I?
Sepht:
Yeah, I realize the odd tree being development in the past. I had heard that, in the new development model (after community comments about quantity of maintenance releases, I assumed), they were going to aim for even points to be the culmination of heavier development in odd points.
But I guess I should give in and search out comments from Linus and Alan before I decide on how to keep up to date without having to build a new kernel every week or two.
Ivan:
If the truth matters so much, why do you support a company that sues it’s customers when their employee’s leak them information? Or one that limits what they give back to the community as much as possible? (in the case of khtml)
Maybe you should try extrapolating Apple’s behavior to a company the size of Microsoft.
I think its time to eliminate anonymous posting myself to try and weed out the Microsoft employees.
Actually, I think you can do this:
1.) Make your script that installs it on a machine
2.) Name it foo.sh
foreach some-list-of-all-machines
scp foo.sh
ssh host ./foo.sh
Well, it’ll likely be more than 3 lines but you get the idea.
And of course the script to build the kernel will be a bit longer as well.. Course, if you wanted you could probably also just build it once and scp all your install files carefully (bad idea I think).
Can you block IPs and or Ip ranges of illicit users? that would probably help.
as for Apple’s behavoir, its better then you could expect of nearly any other commercial company, they did open source darwin, thats a major step beyond what nearly any other company would do.(Darwin’s core is BSD liceanced, unlike GPL they weren’t required to open source it, they just had to “give credit”)
Can you block IPs and or Ip ranges of illicit users? that would probably help.
Won’t solve the problem. May slow them down, those that are not too knowledgable, but the real aggressive trolls will have already known how to defeat such measures. There are trivial ways around an ip banning.
…which is why an account-based system would make so much sense.
as for Apple’s behavoir, its better then you could expect of nearly any other commercial company, they did open source darwin, thats a major step beyond what nearly any other company would do.
Yes Apple should be commended for those actions but then they turn around and do stuff like this:
http://www.itweb.co.za/sections/business/2005/0505241003.asp?S=Reut…
I’d be weary of Apple just as if they were Microsoft.
Hi there,
Just to clarify the situation about kernel release numbers, it was decided that there will be no development tree in the near future unless something really big show up.
Incremental development is taking place in the “stable” tree managed by Linus Torvalds making releases every 2 months approximately.
Heavy development is being done in the -mm tree (and many others trees) which is managed by Andrew Morton who is the “second in charge” and is employed by the OSDL (as well as Linus Torvalds). Once those changes are more or less stabilized, they will be pushed forward to Linus Torvalds tree for wider testing and official acceptance.
So far this development process has been a success. Kernel hackers likes it because changes are smaller and easier to integrate, if an error show up they can find out where that bug come from. This wasn’t easy with the old development model because after 18 or 24 (or 30) months of development no one can tell for sure why certain bug showed up.
Distributors are happy too because new features are introduced faster, hence, they can attract more customers easily. Another advantage for them is that they are able to track more closely the official kernel tree without the need of intrusive patches to implement new features and release them selves from the burden of maintaining large “out of the tree” patches.
Some users likes the new development model because they can get more features faster without the need of waiting 2 years for the development tree to be stabilized.
But, some users are sad because they suffer some stability issues release after release and you can never tell for sure if the next release will work flawlessly on your setup as the previous release did. With the old model this was unlikely to happen (but still possible) because the changes in the stable tree were limited to bug fixes or back porting stable (and simple) features from the development tree.
The most affected users are the ones who likes to compile the kernel from sources. This is not convenient in the new model because the distributor are left with the responsibility of applying “stabilizing” patches if the need show up.
Now that distributors can track the official kernel more closely and get the new features immediately, there is no much need to compile from source.
That’s all for now, I hope this information can help somebody to better understand the new development process and accept it. The only thing it needs is time to mature and when that happens all the disadvantages that comes implicit with the new model will be reduced to a minimum.
Cheers,
-William
User accounts for reporting abuse, and maybe some light ip banning might really clear things up, while keeping the site pretty open and clean.
Few of these trolls are serious ones. Simple countermeasures get rid of the most things, most annoyances aren’t dedicated.
“linux will never be a desktop os, never”
Ignorance? Stupidity? Just trollin’?
I’m using Linux on my desktop, so Linux is a desktop OS.
I’m using SuSE 9.3 on AMD64,while it’s not *the* fastest it has the greatest amount of usefull applications i tend to use.They have done a good job with 9.3.Can’t wait until 9.4 is released.The only drawback is the bad configurable SuSE firewall2 (by GUI).Would be nice if they would give SF2 the same configurabillity as kmyfirewall.Being able to quickly disabling all ports and opening only those one really needs is in my opinion a plus.
> Give me a three-command walkthrough for deploying this network-wide
> on 185 machines, and then we’ll talk about whether it’s useful or not …
> Windows fails this test, too.
make && cp arch/i386/boot/bzImage /tftpboot
It requires a little infrastructure (netboot enabled workstations) but it’s not too difficult… You might also need to run some sort of symbol-relocation program to make sure the kernel is compressed in such a way that the boot loader can handle it, but that’s still within your three command limit.
>Bloody hell. I got the Linux 2.6.12 and now my system (SuSE 9.2) locks after starting udev.
Update your udev.
I can’t make the initrd, fedora lacks the command! oy-oy-oy…. time to
http://rpmfind.net/linux/rpm2html/search.php?query=mkinitrd&submit=…
To the person who modded me down: You can’t handle the truth!
How do we know what you say is the truth?
Ok, let’s say you ARE saying the truth…did you try updating udev? Have you tried booting the kernel with devfs instead? How do you know the problem is actually in the kernel? How do you explain that others do not have this problem?
I recently came across a surprising bug in Windows 2000. I had a word document with a particular embedded font (jester.ttf). Opening the document on a Win2K PC that didn’t have the jester font (which is why I embedded the font in the first place) caused the PC to spontaneously reboot. 100% reproducible. Did I run to OSNews to post that “Windows is crap and we must fix it?” No, I didn’t. Because I understood that this was probably an isolated incident, and that what happens on a machine won’t necessarily happen on all machines.
I’m sorry, but your post reeked of FUD.
2) did the company certify Jester.ttf to run in WIn2K?
Since when are fonts certified to run under an OS? BTW, the problem was NOT with the font, since OpenOffice opened the document without any problems. So the problem was with MS Office and/or Windows, both made by Microsoft.
In our Linux case, SuSE 9.2 was certified to run on Linux 2.6 kernels.
Did you check to see if this also happened to other SuSE 9.2 users who upgraded to 2.6.12? That should have been the first thing you did, to make sure the problem wasn’t with your setup.
You really don’t have a clue do you?
Wow, that didn’t take too long before you resorted to personal attacks. That just made it even likelier to me that the problem you described was made up. You just lost all credibility on the matter. Thanks for playing.
BTW, I run udev and haven’t had any problems with 2.6.12
However, ATI’s crappy closed drivers once again are broken against a new kernel… doesn’t happen often with nvidia drivers. Listening, ATI? I think not.
Mmm, Ignorant anti-linux trolls without a clue in the world, look at them go!
btw, SuSE 9.2 is certified to run with the kernels they distribute, not ones that you download and compile yourself. They may have things that aren’t in the official kernel tree(or aren’t compiled in)(taking Fedora as an example, Xen, GFS, HAL)
You broke your own system, it was something you download and compiled, not something your distrbuter gave you.
That would be like taking a Dell computer, throwing in a new standard fan or default heatsink, having your computer not work, and whining that it’s all Dell’s fault.
If you want to find a distrbution “certified to run with all 2.6 kernels” then go find one; SuSE is not one of them.
You always need to reinstall the video drivers, they “install” themselves by compiling themselves against kernel code, you change kernels, and it won’t direct, it happens with Nvidia drivers too. Just reinstall the drivers and they should work.
(@the trolls, Microsoft only has a single kernel, made in 2001, so you don’t have the problem with Windows, but yet you are stuck using outdated software(Think XP vs Win98; thats now vs XP) )
I’m aware of how to install the drivers, the problem is the drivers themselves. Compilation errors. ATI has progressed a bit in the last few months, but they’ve still got a pretty poor product in their Linux drivers.
You’d be surprised at how many CS graduates are out there without a fn clue.
University is not the measure of intelligence if it was then why are the majority Uni graduates not very well off? Oh that’s right, we have Uni and then we have the Real World and sometimes they are mutually exclusive.
As for early adopters, why not wait until your distro has the new Kernel in their respective repositories. There might be a reason distros don’t jump on the kernel update bandwagon ASAP, it’s called testing.
Earlier in this thread it was explained how the versioning and tree handling of the Linux development was done. I’d like to emphasize one of the points in that post.
It is the responsibillity of the distributor to patch the kernel they ship for stability.
There are som important things to learn from this.
1. Linux the kernel is not b2c product it is b2b product. No where in the deal is Linus or any other kernel developer responsible for shipping a product usable by end users. It is strictly a component that vendors can use to build operating systems. Nothing more.
2. Linux is not an operating system. Linux is either a kernel or a family of operating systems. So the statment that “linux will never be a desktop os” is true because linux will never be an os. Any comments on the merits of linux as an operating system is just stupid.
What’s your problem with Linux?
Linux is a kernel, just like Darwin is. Darwin is not ready for the desktop since it’s only a kernel. But MacOX X is ready for the desktop, so it’s not Linux fault just like is not Darwin’s fault.
Personally I think that SUSE 9.3, Fedora Core 4 and Ubuntu Hoary are ready for *most* desktops, but if it’s not as good as MacOS X is not Linux’s fault. It’s KDE or GNOME’s fault.
If Quartz and Aqua were ported to Linux (technically that’s possible), would you still think that Linux is not reay for the desktop?
Ivan said:
> GNU = poor quality
And everything non-gnu is good?
> Debian = stupid guys believing that RMS is god
I like that one. Its funny.
> Linux = poor kernel ever
Ant that’s exactly why its improving everyday. the developers are never satisfied. And they don’t need to stop, because of money things. Other words. Development is not dependent of money, but enthousiasme to make things better.
> and you all need to see the truth and grow up
No, i don’t want to see the truth. I want to see a icecream and a lazychair and my wife and child sitting next to me in the sun. Damn, I need to work and it’s very hot in the Netherlands.
Lighans
Reading stupid trolling comments such as the ones Ivan posted makes me want to stop reading osnews… We’re closing in on /. quality here. I am all for logged in posing only!!
These aren’t any good arguments…
please osnews – stop these childs, who are desperate, cannot handle their feelings. Ivan’s commentary for example, it doesn’t have any need for being posted here….Osnews is a news feed about operatingsystems – not for trolling!
SHUT UP
you don’t know nothing about computers, go and play on http://disney.go.com and let the others know the truth, the truth that LINUX SUCKS and osnews as well who post bullshit about linux.
call me when the companies start making it’s products on linux, like games, etc. call me when you have a decent GUI, and not something like full of hacks like x11, call me when you have a decent OS.
Linux sucks as well as the GNU movement.
x11 is a graphic server..and you are free the use any compatible *nix wdm. Gnu/Linux is about freedom, if you you aren’t choosing a distro that claims you agains an specific wdm(in the beginning, easy to choose another later).
If you don’t like the oppurtunity to maintain and tweak a specific system – like Gnu/Linux – and maintain it , go ahead use another os if you want.
btw, disney.go.com – the games does work at Mozilla Firefox at Gnu/Linux.
The truth is that Linux sucks for you. If you don’t have any need for it nobody thinks less of you.
The reson people are calling you a troll is not because you don’t like Linux it is because you don’t backup your claims with a valid inference(look it up).