"Since each new kernel contains new bugs, upgrading isn't very useful because it just replaces bugs you know about and have worked around with ones you don't know about."
How is this different from any other system? So you're expecting a bug-free kernel someday? LOL!
Also, reread what you just wrote - they're fixing one bug and replacing it with another. So you admit they're fixing the bugs.
I think the issues - if significant - have to do more with the PACE of new additions to the kernel than the TYPE of new additions, which appeared to be CA's complaint. They fingered media drivers or whatever, while you have complaints about specific system drivers. While the CD burning bug is obviously serious for desktop users, your SATA problem would likely only affect new hardware users and server users - and server operators tend not to upgrade their hardware (except as performance demands) OR their OS that frequently for exactly the reason you cite - new bugs they don't know about.
2.6 is under heavy development, as you state. While from a server operator's view, this might be unfortunate, the solutions are the same as always: don't upgrade when you've got one that works unless you NEED the performance enhancement of the new one.
CA's complaint is still irrational. They're demanding that a kernel under heavy development not have any bugs. Again, LOL...
And specific bugs also don't necessarily translate into instability which is a different issue. What are they doing that is so affected by X specific bugs that the system is "unstable" to them?
CA is complaining about compiling tons of drivers into the kernel. As has been pointed out my everybody including you, Linux doesn't do this anyway.
My point was that Windows knows about as many or more drivers than Linux (including even at install time) so what is CA's problem? Windows HAS to install as many drivers as the USER wants TOO! Just like Linux! (They may not do so during a server install - I had to download and install Intel drivers for the onboard sound chip on a college lab machine recently on Windows 2003 Server - but they still have to HAVE such drivers if the user wants them installed.)
My point is there is NO DIFFERENCE between Windows doing this and Linux doing this. Therefore CA is wrong to complain about Linux.
Especially when I've yet to hear anyone identify a specific instability produced by any given driver or group of drivers or subsystem. I've heard some unsubstantiated complaints about "bugs", so far.
Where would you look to find out? A text config file and a man page?
That's my point. The whole Group Policy Process is a mass of menus, dialog boxes, etc. No where is there a place to ask, "Why is this not working?" Maybe Event Viewer - I can't recall if we looked there for a clue.
It ACTED like everything was working - except it didn't work.
And that was only one of TWO such failed assignments that night - the other one failed for the entire class and the teacher dropped that assignment from being part of the required work.
If you can't get a SCHOOL LAB assignment to work in Windows 2003 Server (reliably and understandably), I shudder at the results for real corporate deployment.
In other words, if you want the "Aero user experience", you get to do this:
"The default Aero user experience is built on the low-level Longhorn graphics API called Avalon and will require a DirectX 9-compliant 3-D graphics processor with at least 32MB of RAM and an Intel AGP 4x bus. Aero will require a minimum resolution of 1024 x 768 (XGA)...
Aero Glass, the higher-end user experience, will be a true superset of Aero and will come with higher hardware requirements... Aero Glass will require a DirectX 9-compliant 3-D graphics processor with at least 64MB of RAM, although Microsoft will recommend 128MB to 256MB of RAM."
And that's on the VIDEO card.
Granted, most NEW video cards will be at those specs by the time Longhorn is out, but it bodes ill for those people not upgrading.
With Windows, the issue is different - it is "featuritis".
"We need a complicated Group Policy system because we have to lock down the desktop!"
"Why?"
"Because our users will screw up their machines if we allow them to do anything!"
"Why?"
(Because we designed the system wrong in the first place - giving users the ability to screw up their machines by doing anything at all...)
As a result, Windows has built "system management" into a totally unmanageable mess that hardly any sys admin can figure out anymore.
As an example, last night in my Server 2003 class, we did a SIMPLE exercise that created an Organizational Unit, assigned a Group Policy to it, and used it to control the behavior of a user assigned to that OU. Worked for most people in the class.
Didn't work for several people in the class.
Teacher couldn't figure out why it didn't work. It just didn't. The appropriate permissions were assigned to the user account, gpupdate was run, the permissions showed as established in the Security tab - then you reboot as that user: nada. Zip. Zero. Default permissions. Everything done just as the textbook instructed - didn't work. Why? Who knows? Some sys admin who would have to spend an hour or ten to find out - or call Microsoft at $275 to ask them.
If Linux (and the desktops and system management tools layered on it) follow this course, then, yes, Linux will turn into a disaster. Fortunately, there is a treatment (if not a cure) - forking.
CA is not a "corporate user" - they are a software marketing outfit. They want to market their stuff on Linux, fine. They want to market their own distro, fine. They want to hire a kernel hacker to do that, fine. They want the Linux developers to do it for them, not so fine - particularly when they have no specifics for why.
Secondly, no ordinary corporate user needs a kernel hacker. If they did, they'd sure need it with Windows - assuming of course that was even possible with a totally closed source system.
Third, "TCO" - which in itself is based on air - has nothing to do with this or anything at all to do with the kernel (except to the degree that a kernel improves performance on high-end hardware and might allow purchasing either less expensive hardware or fewer pieces of same.) Anybody running a high-end server on Linux knows how to tweak Linux to get stability and adequate performance and isn't worried about an extra "music driver" that isn't even loaded.
Fourth, while price is indeed important to Linux's success (even if in fact it's not that big a factor in actual deployment and is more a perceived value), Linux has no significant competitors (other than Windows) in terms of pace of development, support, applications, etc., all of which are much more significant than whether the OS costs money or not. Even Solaris isn't in the ball park. None of the bigger iron UNIX variants are significant - not HP/UX, not AIX. The FreeBSD variants are also rans.
Linux is the only game in town against Windows. Period.
The ONLY threat to Linux is if the desktops - and more importantly, the OS system services and their configuration - get SO bloated and insanely complex - like Windows Server 2003 - that nobody can figure out how to use it. And the desktops can always be replaced by something better.
It's manipulating and configuring the system that needs to be kept straightforward and task-oriented - not filled with thousands of menus, dialogs, Management Consoles, Control Panels, ad nauseum, like Windows - which has a fatal case of "featuritis" and absolutely NO usability engineering.
Try figuring out "effective" permissions, end-user lockdown, and Group Policy application in Windows 2003 Server. They have to give COURSES in this stuff, for Baron von Christ's sakes! I know, I'm taking one.
I suspect the ACTUAL cause of instability was not merely "adding drivers", but more along the lines of "adding crap drivers to a crap driver architecture of a crap OS architecture".
Unless you have more specifics, this sounds like oversimplification to me.
It's ridiculous to suggest that the kernel layout should be restricted to the level of a 486.
First of all, you can already do that if you know what you're doing. People in the Third World either know what they're doing or get their machines from people who do - just like in the rest of the world.
Secondly, there are tons of stripped down distros. Pick one.
This is merely asking for your cake and eating it, too - you want the latest kernel and everything it can support to run on the oldest hardware.
Try it with Windows 20003 Server.
Then go back and read the specs for Longhorn: a GB of RAM, a terabyte of hard disk, and a minimum 3GHz CPU.
The Linux kernel is intended to push the boundaries of OS technology - not run on every Third World machine in existence.
Yet, at that, as I pointed out, Linux is incredibly flexible in what it will run on compared to virtually every other OS in existence.
Have you ever installed a late version of Windows?
Watch the installer load device drivers for every known weird form of RAID before it even begins to ask you how you want to install the OS?
And then how long does it take to do "hardware detection" - versus Knoppix that does it all in the three minutes or so it takes to boot from CD?
Yes, Windows is bloated - bloated with (so-called) "features", not drivers. If Linux makes THAT mistake, we can complain. Having a bunch of drivers and support for oddball subsystems loaded into the kernel is not serious and until somebody DEMONSTRATES a stability problem, it's bullshit.
So far I've heard nobody say the 2.6 kernel is in FACT unstable because of x, y, z drivers or subsystems.
You need to try jail sometime, asshole, if you think it's fun. And I use the term asshole advisedly, since if you're as young and stupid as I suspect, yours would get a workout there.
Also if you think prisoners get direct Internet access, you are obviously a total moron.
You have absolutely no comprehension of the concept of personal responsibility.
NONE of the cases you cite has ANYTHING to do with ignoring anybody's so-called "rights" (which are a fiction anyway, but that's another rant.)
Anybody who gets hooked on addictive substances (and that includes my fat ass on pizza and Haagen Daaz) deserves what happens to them. This has absolutely nothing to do with their "rights".
Arbitrarily declaring certain substances as "illegal" - especially with the hidden agenda of criminalizing entire groups of people for the express purpose of putting those people under violent state control - IS a DIRECT violation of "rights".
Any statistics will tell you the drug problem is as prevalent today as it ever was. Massive failure all down the line. The only time it was a minor problem was back in the 1920's - BEFORE they passed the fucking drug laws intended to be used against minorities.
If you can't comprehend that, you have no fucking ability to comprehend anything.
"In any event, you will always lose in the court of public opinion, which was my point."
Oh, yeah? Wait until I get the necessary nanotech.
The "court of public opinion" will kiss my white ass before they dissolve into a pile of recyclable chemicals.
And your bullshit about people's habits having side effects on the community merely demonstrates your complete inability to comprehend the term personal responsibility.
Linus has no business being pissed at Tridge for doing what any OSS developer might do - reverse-engineer a useful protocol.
If anything, Linus should be pissed at himself for tying the kernel management to a proprietary product and pissed at Larry for pulling it because of some paranoid notion that Tridge was "destroying" Larry's work.
Nobody said proprietary is always evil. What is being said is that Linus had no business tying other people's work to a proprietary product merely because it was technically better in the short term. For his OWN use, fine, no problem.
And what is being said is that Linus has been unnecessarily harsh on Tridge for doing EXACTLY the same thing he did to make Samba work.
Just last week, at the Open Source Business Conference, Schwartz was attacking the GPL.
How much more obvious does it get? Sun is only interested in their own concept of "OSS" - which is, we own it, we LET you use it - provided you never threaten our revenue model.
This is exactly what this guy did.
Would everybody be happy if Linus, who owns the copyright on Linux, suddenly pulled it because OSDL suddenly stopped paying him to work on it?
Slashdot Mirror
"Since each new kernel contains new bugs, upgrading isn't very useful because it just replaces bugs you know about and have worked around with ones you don't know about."
How is this different from any other system?
So you're expecting a bug-free kernel someday?
LOL!
Also, reread what you just wrote - they're fixing one bug and replacing it with another. So you admit they're fixing the bugs.
I think the issues - if significant - have to do more with the PACE of new additions to the kernel than the TYPE of new additions, which appeared to be CA's complaint. They fingered media drivers or whatever, while you have complaints about specific system drivers. While the CD burning bug is obviously serious for desktop users, your SATA problem would likely only affect new hardware users and server users - and server operators tend not to upgrade their hardware (except as performance demands) OR their OS that frequently for exactly the reason you cite - new bugs they don't know about.
2.6 is under heavy development, as you state. While from a server operator's view, this might be unfortunate, the solutions are the same as always: don't upgrade when you've got one that works unless you NEED the performance enhancement of the new one.
CA's complaint is still irrational. They're demanding that a kernel under heavy development not have any bugs. Again, LOL...
And specific bugs also don't necessarily translate into instability which is a different issue. What are they doing that is so affected by X specific bugs that the system is "unstable" to them?
Still missing the point.
CA is complaining about compiling tons of drivers into the kernel. As has been pointed out my everybody including you, Linux doesn't do this anyway.
My point was that Windows knows about as many or more drivers than Linux (including even at install time) so what is CA's problem? Windows HAS to install as many drivers as the USER wants TOO! Just like Linux! (They may not do so during a server install - I had to download and install Intel drivers for the onboard sound chip on a college lab machine recently on Windows 2003 Server - but they still have to HAVE such drivers if the user wants them installed.)
My point is there is NO DIFFERENCE between Windows doing this and Linux doing this. Therefore CA is wrong to complain about Linux.
Especially when I've yet to hear anyone identify a specific instability produced by any given driver or group of drivers or subsystem. I've heard some unsubstantiated complaints about "bugs", so far.
Nope.
Where would you look to find out? A text config file and a man page?
That's my point. The whole Group Policy Process is a mass of menus, dialog boxes, etc. No where is there a place to ask, "Why is this not working?" Maybe Event Viewer - I can't recall if we looked there for a clue.
It ACTED like everything was working - except it didn't work.
And that was only one of TWO such failed assignments that night - the other one failed for the entire class and the teacher dropped that assignment from being part of the required work.
If you can't get a SCHOOL LAB assignment to work in Windows 2003 Server (reliably and understandably), I shudder at the results for real corporate deployment.
Specify.
Then ask why Linus and Andrew - not to mention the distros - are allowing this, IF true.
I SAID it was during the install.
My point was that Microsoft's kernel ALSO has to KNOW about every weird driver out there EVEN to install.
So why is CA complaining about driver support in the kernel?
In other words, if you want the "Aero user experience", you get to do this:
"The default Aero user experience is built on the low-level Longhorn graphics API called Avalon and will require a DirectX 9-compliant 3-D graphics processor with at least 32MB of RAM and an Intel AGP 4x bus. Aero will require a minimum resolution of 1024 x 768 (XGA)...
Aero Glass, the higher-end user experience, will be a true superset of Aero and will come with higher hardware requirements... Aero Glass will require a DirectX 9-compliant 3-D graphics processor with at least 64MB of RAM, although Microsoft will recommend 128MB to 256MB of RAM."
And that's on the VIDEO card.
Granted, most NEW video cards will be at those specs by the time Longhorn is out, but it bodes ill for those people not upgrading.
Okay, the terabyte of HD is an exaggeration. But if the minimum is 3GHz and 512MB, then the ACTUAL needs will be at LEAST 3GHz and 1GB.
And thousands of businesses and millions of home users are not going to upgrade.
Which will not be thousands of businesses trying to extend the lifespan of their current hardware.
To some degree, laziness is probably involved.
With Windows, the issue is different - it is "featuritis".
"We need a complicated Group Policy system because we have to lock down the desktop!"
"Why?"
"Because our users will screw up their machines if we allow them to do anything!"
"Why?"
(Because we designed the system wrong in the first place - giving users the ability to screw up their machines by doing anything at all...)
As a result, Windows has built "system management" into a totally unmanageable mess that hardly any sys admin can figure out anymore.
As an example, last night in my Server 2003 class, we did a SIMPLE exercise that created an Organizational Unit, assigned a Group Policy to it, and used it to control the behavior of a user assigned to that OU. Worked for most people in the class.
Didn't work for several people in the class.
Teacher couldn't figure out why it didn't work.
It just didn't. The appropriate permissions were assigned to the user account, gpupdate was run, the permissions showed as established in the Security tab - then you reboot as that user: nada. Zip. Zero. Default permissions. Everything done just as the textbook instructed - didn't work. Why? Who knows? Some sys admin who would have to spend an hour or ten to find out - or call Microsoft at $275 to ask them.
If Linux (and the desktops and system management tools layered on it) follow this course, then, yes, Linux will turn into a disaster. Fortunately, there is a treatment (if not a cure) - forking.
Try and fork Windows - or even Solaris.
And if it doesn't crash, it's a damn sight better than Windows!
Once again, building a case on nonsense.
CA is not a "corporate user" - they are a software marketing outfit. They want to market their stuff on Linux, fine. They want to market their own distro, fine. They want to hire a kernel hacker to do that, fine. They want the Linux developers to do it for them, not so fine - particularly when they have no specifics for why.
Secondly, no ordinary corporate user needs a kernel hacker. If they did, they'd sure need it with Windows - assuming of course that was even possible with a totally closed source system.
Third, "TCO" - which in itself is based on air - has nothing to do with this or anything at all to do with the kernel (except to the degree that a kernel improves performance on high-end hardware and might allow purchasing either less expensive hardware or fewer pieces of same.) Anybody running a high-end server on Linux knows how to tweak Linux to get stability and adequate performance and isn't worried about an extra "music driver" that isn't even loaded.
Fourth, while price is indeed important to Linux's success (even if in fact it's not that big a factor in actual deployment and is more a perceived value), Linux has no significant competitors (other than Windows) in terms of pace of development, support, applications, etc., all of which are much more significant than whether the OS costs money or not. Even Solaris isn't in the ball park. None of the bigger iron UNIX variants are significant - not HP/UX, not AIX. The FreeBSD variants are also rans.
Linux is the only game in town against Windows. Period.
The ONLY threat to Linux is if the desktops - and more importantly, the OS system services and their configuration - get SO bloated and insanely complex - like Windows Server 2003 - that nobody can figure out how to use it. And the desktops can always be replaced by something better.
It's manipulating and configuring the system that needs to be kept straightforward and task-oriented - not filled with thousands of menus, dialogs, Management Consoles, Control Panels, ad nauseum, like Windows - which has a fatal case of "featuritis" and absolutely NO usability engineering.
Try figuring out "effective" permissions, end-user lockdown, and Group Policy application in Windows 2003 Server. They have to give COURSES in this stuff, for Baron von Christ's sakes! I know, I'm taking one.
I suspect the ACTUAL cause of instability was not merely "adding drivers", but more along the lines of "adding crap drivers to a crap driver architecture of a crap OS architecture".
Unless you have more specifics, this sounds like oversimplification to me.
Where in his post did he say that?
Somebody quoted Morton complaining about insecurities in the kernel several releases back. The responder pointed this out.
You have a list of currently exploitable kernel vulnerabilities and their status on the kernel maintainer's schedule?
And can compare them to the same in Windows and the Mac and FreeBSD?
Let's see it.
It's ridiculous to suggest that the kernel layout should be restricted to the level of a 486.
First of all, you can already do that if you know what you're doing. People in the Third World either know what they're doing or get their machines from people who do - just like in the rest of the world.
Secondly, there are tons of stripped down distros. Pick one.
This is merely asking for your cake and eating it, too - you want the latest kernel and everything it can support to run on the oldest hardware.
Try it with Windows 20003 Server.
Then go back and read the specs for Longhorn: a GB of RAM, a terabyte of hard disk, and a minimum 3GHz CPU.
The Linux kernel is intended to push the boundaries of OS technology - not run on every Third World machine in existence.
Yet, at that, as I pointed out, Linux is incredibly flexible in what it will run on compared to virtually every other OS in existence.
All of this is just utterly pointless criticism.
Have you ever installed a late version of Windows?
Watch the installer load device drivers for every known weird form of RAID before it even begins to ask you how you want to install the OS?
And then how long does it take to do "hardware detection" - versus Knoppix that does it all in the three minutes or so it takes to boot from CD?
Yes, Windows is bloated - bloated with (so-called) "features", not drivers. If Linux makes THAT mistake, we can complain. Having a bunch of drivers and support for oddball subsystems loaded into the kernel is not serious and until somebody DEMONSTRATES a stability problem, it's bullshit.
So far I've heard nobody say the 2.6 kernel is in FACT unstable because of x, y, z drivers or subsystems.
Thanks for this exposition of conventional wisdom.
When you have something specific to pin this on, I'm sure we'd all like to hear from you again.
Nuke Texas. If Saddam's Iraq couldn't be allowed to exist, neither should Texas.
because they always have at least one thumb up their ass at any given time.
You need to try jail sometime, asshole, if you think it's fun. And I use the term asshole advisedly, since if you're as young and stupid as I suspect, yours would get a workout there.
Also if you think prisoners get direct Internet access, you are obviously a total moron.
You have absolutely no comprehension of the concept of personal responsibility.
NONE of the cases you cite has ANYTHING to do with ignoring anybody's so-called "rights" (which are a fiction anyway, but that's another rant.)
Anybody who gets hooked on addictive substances (and that includes my fat ass on pizza and Haagen Daaz) deserves what happens to them. This has absolutely nothing to do with their "rights".
Arbitrarily declaring certain substances as "illegal" - especially with the hidden agenda of criminalizing entire groups of people for the express purpose of putting those people under violent state control - IS a DIRECT violation of "rights".
Horseshit.
Any statistics will tell you the drug problem is as prevalent today as it ever was. Massive failure all down the line. The only time it was a minor problem was back in the 1920's - BEFORE they passed the fucking drug laws intended to be used against minorities.
If you can't comprehend that, you have no fucking ability to comprehend anything.
"In any event, you will always lose in the court of public opinion, which was my point."
Oh, yeah? Wait until I get the necessary nanotech.
The "court of public opinion" will kiss my white ass before they dissolve into a pile of recyclable chemicals.
And your bullshit about people's habits having side effects on the community merely demonstrates your complete inability to comprehend the term personal responsibility.
San Francisco is not big. The Bay Area IS big - which is irrelevant since the City Council is not talking about the Bay Area, only San Francisco.
Also, San Francisco has FUCKING HILLS! Which happen to be great places to put antennas, numbnuts...Ever see a picture of Sutro Tower?
You just don't get it.
Linus has no business being pissed at Tridge for doing what any OSS developer might do - reverse-engineer a useful protocol.
If anything, Linus should be pissed at himself for tying the kernel management to a proprietary product and pissed at Larry for pulling it because of some paranoid notion that Tridge was "destroying" Larry's work.
Nobody said proprietary is always evil. What is being said is that Linus had no business tying other people's work to a proprietary product merely because it was technically better in the short term. For his OWN use, fine, no problem.
And what is being said is that Linus has been unnecessarily harsh on Tridge for doing EXACTLY the same thing he did to make Samba work.
Absolutely.
Just last week, at the Open Source Business Conference, Schwartz was attacking the GPL.
How much more obvious does it get? Sun is only interested in their own concept of "OSS" - which is, we own it, we LET you use it - provided you never threaten our revenue model.
This is exactly what this guy did.
Would everybody be happy if Linus, who owns the copyright on Linux, suddenly pulled it because OSDL suddenly stopped paying him to work on it?
What's the difference here?