Linux-distributions are like Dell, HP, IBM, Sony, etc. They provide the same product that is "pretty much" the same, but there is still enough competition between them so that no vendor sits on their laurels.
Today, with different sockets, different RAMs, PCI, API, etc. computers are no longer compatible with everything. Just like you can't put an Athlon into an Intel-socket, you can't use a rpm on debian.
And when you look at the whole x86-market, these differences haven't hurt the market at all.
While I agree that the real incompatibilities between Unices have hurt the whole Unix market, I disagree that different distros do the same to the Linux market because there are no real incompatibilities (any program runs anywhere, it's just a matter of choosing.debs or.rpms which usually isn't necessery anyway because the stuff comes with the distro).
Oh, and Sun/Solaris isn't the "last man standing", Linux is.
Actually despite the public image being the contrary, Microsoft is abandoning it's products quite fast. Examples are Hailstorm, Passport, standard Visual Basic, etc.
Great analysis, I disagree with one thing, though:
we will be stuck with Unix for as long as it took to get rid of Microsoft -- maybe longer.
Unlike Windows, the source for Linux is readily available. So if you really have something better it will be no problem to write a 100% compatible layer that will run Linux-apps on your OS. Actually such a thing already exists in BSD and AFAIK also on AIX-L.
You also miss the biggest advantage of Unix:
It's simplicity. Take the permissions for example. You can argue all day about ACLs being "better" than the classic Unix uga-bits, but at the end of the day, the Unix-way is adequate for almost everybody and more importantly can be understood in theory and praxis by almost everybody. On the other hand, ACLs may be undertood in theory, but they will turn into a terrible unmaintainable mess once you start to use them on a larger scale.
This is the reason why large parts of the Windows-world completely ignores everything about users/permissions and still runs everything as administrator.
So the big plus of Unix is it's simplicity. The "everything is a file" approach is yet another example.
As long as humans have to program computers, the easy and simple way is superior to the more complicated way some professor may have come up with.
You got 1) reversed, on Linux everything works instantly, even a word processor, while you have to install everything on Windows.
2) and 3) are pretty much the same point and you are right here. However, things are improving fast - with governments (like China, Brazil, Venezuela, Germany, France, etc.) going to Linux, more and more software makers will have no other choice than to support Linux.
Without WinFS, there isn't anything interesting left for me. Avalon is just useless eyecandy (so my windows are now 3D-accellerated? So what? Will that make me more productive? Will it empower me? And BTW it will be available for 2K/XP too, so it isn't even a Longhorn-feature.) and DRM is more of a turnoff than a turnon.
Microsoft must deliver an improved version of its browser...
1) The original team is gone. Therefore there is a lot of code that nobody really understands and it will take a lot of time to catch up on that. Just look at how long Mozilla took. Just look how long Microsoft took to create IE. Of course Microsoft has lots of ressources, nevertheless it will take time, no matter how much money they throw at the problem. They can consider themselves lucky when they have an improved version as soon as Longhorn ships.
2) The other, more serious problem is that the better they make the browser, the more attractive web-based development becomes. More (especially in-house) web-development means less Win32-only development, which means (best case) less money for Microsoft because there are much, much less development tools needed and (worst case) a very easy migration path away from Windows. Actually I think that was the reason why IE-development was stopped in the first place. With Netscape gone, Microsoft had nothing to gain from IE-improvements, but lots of MSDN-subscriptions to lose.
after all, [Microsoft a]re the only people who know what all the random gibberish in the registry means, or whether mswin03.dll really belongs in 2003 server's windows directory
Well, actually I wouldn't be sure about that.
When you look at their latest problems - the endless developing of Longhorn (I often ask myself: "What the hell are they doing all the time? They throw out all the interesting features like WinFS, what is taking them so long?"), the complete lack of progress in IE, etc. - it seems a lot that the Windows codebase has become so bloated and complicated that no single person actually understands it.
XP was no problem, but fedora didn't support the SATA chipset contained within the box, so I either had to go with an older IDE based hard drive or just go with XP until the driver was updated by the manufacturer of the chipset.
Isn't it funny that you bitch about Linux not supporting that cutting-edge chipset but don't bitch about Windows not supporting AMD64 extensions?
I'm really sick of those double standards. Here we have *exactly* the same problem in Windows and Linux (it doesn't take fully advantage of some cutting edge hardware), yet in Linux it's terrible, terrible, with Windows it's even worth mentioning...
I think open source does create the illusion that it couldn't contain hidden malware because where could it hide in open source, right?
There are numerous examples of malware (like those in Kazaa), easter-eggs (like the flight simulator in Excel and the pinball game in MS Word) and unrequested features (like Windows Product Activation) in proprietary software.
While I agree that something similar could in theory also happen to some one-man or very small open source projects (in theory because I have never heard of any such occurence) there is absolutely no way such code could be smuggled into bigger projects like Linux, Apache, KDE or the like, there are just too many people watching.
So if you compare exactly what the article is talking about, proprietary software has a much worse track record.
If that's true, public institutions must not buy any proporietary software because with public money spent on proprietary software, the public gets nothing.
This weakness made it possible for IBM, Sun, HP, etc. to proprietize Unix and make many incompatible versions that only run on their hardware.
The GPL was only created because of the lessons learned from the whole Unix-fiasco.
If the BSD-license would work so great as the BSD-fans claim, the whole Unix-breakup would have never happened and there would have never been a reason to even create the GPL.
The Xbox is one of the most innovative consoles to ever hit the market. First and foremost, it's the first console to ever include a hard drive. Also, the Xbox was built to be easy to program, utilizing standard libraries (DirectX) and development environments (Visual Studio).
So Microsoft crippled a PC, put some optimizations in and sold it as a console.
Are our standards really that low that this counts as "innovation"?
XBox doesn't contain anything new. Harddrives are hardly new. The x86 architecture isn't new.
The PS2 architecture however was completely different to anything existing. The new "cell" architecture also is completely new.
That's a whole other level than just putting some off-the-shelf hardware parts into a console.
Slashdot Mirror
Right. If anybody wants to provide support, it can be continued forever.
The whole point is that it can be **anybody** and need not be the original author
However, most computers are self-administered or administered by someone who also has another job.
Funny, I haven't had a single application that runs on debian that refused to run on SuSE.
So what was your point?
No, He's not.
Linux-distributions are like Dell, HP, IBM, Sony, etc. They provide the same product that is "pretty much" the same, but there is still enough competition between them so that no vendor sits on their laurels.
Today, with different sockets, different RAMs, PCI, API, etc. computers are no longer compatible with everything. Just like you can't put an Athlon into an Intel-socket, you can't use a rpm on debian.
And when you look at the whole x86-market, these differences haven't hurt the market at all.
While I agree that the real incompatibilities between Unices have hurt the whole Unix market, I disagree that different distros do the same to the Linux market because there are no real incompatibilities (any program runs anywhere, it's just a matter of choosing .debs or .rpms which usually isn't necessery anyway because the stuff comes with the distro).
Oh, and Sun/Solaris isn't the "last man standing", Linux is.
we will be stuck with Unix for as long as it took to get rid of Microsoft -- maybe longer.
Unlike Windows, the source for Linux is readily available. So if you really have something better it will be no problem to write a 100% compatible layer that will run Linux-apps on your OS. Actually such a thing already exists in BSD and AFAIK also on AIX-L.
You also miss the biggest advantage of Unix:
It's simplicity. Take the permissions for example. You can argue all day about ACLs being "better" than the classic Unix uga-bits, but at the end of the day, the Unix-way is adequate for almost everybody and more importantly can be understood in theory and praxis by almost everybody. On the other hand, ACLs may be undertood in theory, but they will turn into a terrible unmaintainable mess once you start to use them on a larger scale.
This is the reason why large parts of the Windows-world completely ignores everything about users/permissions and still runs everything as administrator.
So the big plus of Unix is it's simplicity. The "everything is a file" approach is yet another example.
As long as humans have to program computers, the easy and simple way is superior to the more complicated way some professor may have come up with.
2) and 3) are pretty much the same point and you are right here. However, things are improving fast - with governments (like China, Brazil, Venezuela, Germany, France, etc.) going to Linux, more and more software makers will have no other choice than to support Linux.
True.
However unlike in the 90's with Linux there exists an OS that runs on (almost) all hardware and runs a lot of software.
Actually Linux would be perfect for a lot of non-gaming Windows users, the only thing deeply missing is good marketing. And Google could solve that.
If they build an OS, which I really doubt....
Because you can't compile the code, you have no way to verify that it is even the right source code.
The only thing you will get is [i]some[/i] source code. It might be from a 5-year old version of the product, it might even be from another product.
So what's the point?
Why should anybody upgrade?
1) The original team is gone. Therefore there is a lot of code that nobody really understands and it will take a lot of time to catch up on that. Just look at how long Mozilla took. Just look how long Microsoft took to create IE. Of course Microsoft has lots of ressources, nevertheless it will take time, no matter how much money they throw at the problem. They can consider themselves lucky when they have an improved version as soon as Longhorn ships.
2) The other, more serious problem is that the better they make the browser, the more attractive web-based development becomes. More (especially in-house) web-development means less Win32-only development, which means (best case) less money for Microsoft because there are much, much less development tools needed and (worst case) a very easy migration path away from Windows. Actually I think that was the reason why IE-development was stopped in the first place. With Netscape gone, Microsoft had nothing to gain from IE-improvements, but lots of MSDN-subscriptions to lose.
Does anybody know?
More interesting than that would be when those patents expire, actually.
Well, actually I wouldn't be sure about that.
When you look at their latest problems - the endless developing of Longhorn (I often ask myself: "What the hell are they doing all the time? They throw out all the interesting features like WinFS, what is taking them so long?"), the complete lack of progress in IE, etc. - it seems a lot that the Windows codebase has become so bloated and complicated that no single person actually understands it.
So this story is just a piece of evil propaganda.
songs, movies and software programs
They just look at Maya, etc. and "assume" those are downloaded my every P2P-user and multiply the result with their earlier assumptions.
Exactly, I think for the US, the war has become an end and is no longer a means.
Isn't it funny that you bitch about Linux not supporting that cutting-edge chipset but don't bitch about Windows not supporting AMD64 extensions?
I'm really sick of those double standards. Here we have *exactly* the same problem in Windows and Linux (it doesn't take fully advantage of some cutting edge hardware), yet in Linux it's terrible, terrible, with Windows it's even worth mentioning...
When you submitted that story to slashdot, did you use handwriting or voice recognition?
Didn't think so...
There are numerous examples of malware (like those in Kazaa), easter-eggs (like the flight simulator in Excel and the pinball game in MS Word) and unrequested features (like Windows Product Activation) in proprietary software.
While I agree that something similar could in theory also happen to some one-man or very small open source projects (in theory because I have never heard of any such occurence) there is absolutely no way such code could be smuggled into bigger projects like Linux, Apache, KDE or the like, there are just too many people watching.
So if you compare exactly what the article is talking about, proprietary software has a much worse track record.
If that's true, public institutions must not buy any proporietary software because with public money spent on proprietary software, the public gets nothing.
Wrong it is a very bad weakness.
This weakness made it possible for IBM, Sun, HP, etc. to proprietize Unix and make many incompatible versions that only run on their hardware.
The GPL was only created because of the lessons learned from the whole Unix-fiasco.
If the BSD-license would work so great as the BSD-fans claim, the whole Unix-breakup would have never happened and there would have never been a reason to even create the GPL.
XBox was released 2 years after PS2.
When the PS2 was released, the graphics were great by that time. When the XBox came out, everybody has already seen it on PCs.
So Microsoft crippled a PC, put some optimizations in and sold it as a console.
Are our standards really that low that this counts as "innovation"?
XBox doesn't contain anything new. Harddrives are hardly new. The x86 architecture isn't new.
The PS2 architecture however was completely different to anything existing. The new "cell" architecture also is completely new.
That's a whole other level than just putting some off-the-shelf hardware parts into a console.