Huh? The G5 is a faster chip than the AMD64, and it uses HyperTransport. If Apple just wanted to stick the fastest chip they could in machines, they would have stuck with the PowerPC. If Apple was just after the fastest chips, they wouldn't have moved to x86 in the first place. But they don't want the fastest chips, they want the most functional, and the AMD fanboys can't get their heads around it. Also, the GMA950 in the Mac Mini, while it doesn't have good 3D acceleration, it has much better multimedia acceleration than most GPU's. Do you think that just maybe Apple was interested in a good multimedia chip instead of a good gaming chip? (Not that that changes I won't be buying one...)
Um. Nothing opens it automatically after downloading. The user has to double click it to open it. Safari will not automatically open things anyway, and will ask you before decompressing anything with an executable in it.
Apple did bundle iPod Minis with their Macs last year for students.
Also dual booting OS X for Intel on a normal PC is a pain. A lot of stuff doesn't work properly. Not to mention no one knows how to install the latest version. And even if someone figures it out all Apple has to do is patch again and add new security to keep it from booting on non-Macs.
Except Cell doesn't support out of order execution making it not suitable for any normal operating system... i.e. Linux. This is the the same reason why Apple rejected the Cell.
"I _want_ people to expect that interfaces change. I _want_ people to
know that binary-only modules cannot be used from release to release.
I want people to be really really REALLY aware of the fact that when
they use a binary-only module, they tie their hands"
Huh? He wants people to code for Linux after saying this? I'm sorry, but if this is his attitude I will never ever write a driver for Linux. He's telling me my driver won't be compatible from release to release BY DESIGN? Does he really want people to recode each release to get their driver to work? Does he think this looks friendly to developers? Does he think commercial companies are really willing to do this every time? Does he think end users will really be willing to wait for recoded drivers every release? C'mon, this is ok for an OS you're writing for a hobby, but this doesn't fly in the real world (where I'm sure a lot more people would like to see Linux).
Yes, adding that abstraction is going to slow you down. Darwin is slower than Linux. Why not fork it? I mean... Sacrificing slowness for usability? An operating system can be faster than anything else out there, but I won't code for it unless it has a stable usable API and my end users can easily install the drivers. Sorry. Apparently Linux is anti that for device drivers.
Huh? We're not talking drivers that come with the operating system.
This is like not being allowed to put a custom engine in your car unless it's approved by your auto maker according to their standards.
If Linux wants to be an operating system about choice they need to let people choose whether or not they want closed source drivers in their system. It's not their job to enforce ideals. That's what Microsoft does.
What sense does this make? You can run closed source software under Linux. Why not closed source drivers? It's this kind of thought that is killing Linux's potential while Mac OS X passes it by.
The Cell is not good for mainstream computers. It's way too hard to write software for, and moving to Cell probably would have meant re-writing all the current Mac applications. The Cell isn't really just an upgraded PowerPC chip. Instead it's mostly built for very specialized applications. I even have doubts it was a good choice for the PS3 seeing as how much work publishers are having to put into making PS3 games.
Actually, no. The PPC version is also 32 bits. It just has the ability to run 64 bit applications and address more than 2 gigs of memory per application. The OS itself is not compiled for 64 bit.
Read your own article. That's retail market. Like you know... It's only counting stores you walk into. It's not counting companies like sayyyyyyyyyyy Dell?
You're missing the point. When the iTunes Music Store came out, it didn't revolutionize music at first. It's a gradual process.
Don't think it's impossible to add video support to the AirPort Express either. Then the user just has to add another box to their stereo setup.
Or I could ignore your long, drawn out, and inaccurate comment and continue watching iTunes shows on my computer with tv out.
Or maybe even hook a video iPod up to a tv via it's tv out jack.
You're missing the point. Having more than one Linux operating system is going to confuse consumers. They don't want that. They want simple. They want easy. How about this... Let's make one Linux distribution. The Debian people can manage the server aspect and applications, the Fedora and SuSE people can manage desktop experience, Mozilla can manage the internet stuff, etc. This is ideal. Everyone is working on maximizing the same OS for different needs. Instead, the server people want to code their own version of Linux, the desktop people want to code their own version (with a different version depending on which desktop experience they think they should be having). The end result? Instead of having "Linux Server" and "Linux Client" you have tons of different versions that people who don't want to have to think about their computers, much less choose an OS. If Linux proponents want Linux to be mass market, they're going to have to realize people don't want to think about what they want their computer to do. They want it to just work. Linux people always look at Linux from the geek perspective. They think the more powerful and complex the OS is over Windows the more people will adopt it. People don't want a complex geek toy OS with options. They want something they buy, install, and use, no thinking involve. Yes, I know geeks will find this a horrible idea, but thats the real world. In the real world people don't care if they are running Windows 95, Windows 98, Windows ME, Windows 2000, or Windows XP. It's just Windows. It just works. They picked it up at the store when they asked for Windows. And they don't want to care which version they are running. As long as they can go to the store and buy their software and it works on their machine it's ok. If Linux people think normal users will be attracted to Linux's "diversity" they're wrong. People see that as complicated, something only geeks can understand, and something to be avoided.
Of course if Linux wants to stay an operating system by geeks for geeks, diversity is a good thing. But as long as Linux is an OS built for geeks mainstream users will avoid it like the plague. Microsoft ruled the world by making standards with Windows. Linux needs to do the same. Make a standard.
Please note that all the car brands have steering wheels, an accelerator, a key to turn on, and doors with door handles.
Yes, lots of cars can do different things but they all operate the same way. If there was one kind of Linux and you had a server version and a client version, that's easy. The way it is now Linux doesn't even install the same way for each version.
There is a difference between having one kind of Linux tweaked to do lots of different things, and lots of versions of Linux all tweaked to do the same thing.
And this is why Linux will never catch on in the mainstream.
Choice is bad for mainstream consumers. They don't want to have to decide. They want one option they know will work. Why do most consumers choose Windows? There is one version, it's compatible with everything else built for Windows, and they don't have to think when they choose it.
A consumer is not going to want to have to research which version of Linux they want. They want to go to CompUSA, ask for a copy of "Linux", go home, install, and be able to run Linux software. They don't want to have to decide which Linux is right for them. That's not simple. With Windows they just ask for a copy of Windows and tada, they're up and running.
If Linux wants to be mainstream, they're going to have to sacrifice diversity. Mac has reached %6 of the market using one standard OS, maybe if Linux would all hop on the same boat together they'd to the same.
Slashdot Mirror
Huh? The G5 is a faster chip than the AMD64, and it uses HyperTransport. If Apple just wanted to stick the fastest chip they could in machines, they would have stuck with the PowerPC. If Apple was just after the fastest chips, they wouldn't have moved to x86 in the first place. But they don't want the fastest chips, they want the most functional, and the AMD fanboys can't get their heads around it. Also, the GMA950 in the Mac Mini, while it doesn't have good 3D acceleration, it has much better multimedia acceleration than most GPU's. Do you think that just maybe Apple was interested in a good multimedia chip instead of a good gaming chip? (Not that that changes I won't be buying one...)
If Apple wanted to make AMD Macs, wouldn't they just have ignored Intel in the first place?
And why is Apple obligated to make AMD Macs?
Right. This is where you need to go learn what an executable is.
Um. Nothing opens it automatically after downloading. The user has to double click it to open it. Safari will not automatically open things anyway, and will ask you before decompressing anything with an executable in it.
No, you can't say you told us so.
So a license like the GPL can be ignored also? Careful, the sword you're playing with is double edged.
Huh? You can run e17 inside OS X.
And then only a matter of time until Apple releases an update that breaks it again...
Apple doesn't sell Mac OS X x86. And even if you could get a copy off one of the new iMacs it doesn't work on generic PC hardware.
Apple did bundle iPod Minis with their Macs last year for students. Also dual booting OS X for Intel on a normal PC is a pain. A lot of stuff doesn't work properly. Not to mention no one knows how to install the latest version. And even if someone figures it out all Apple has to do is patch again and add new security to keep it from booting on non-Macs.
How is it that it took you a half hour? I just plugged in my iPod and it automatically transferred all my music over, no problems at all.
Dell isn't selling AMD processors.
Apple has already announced support for Blu-Ray.
Except Cell doesn't support out of order execution making it not suitable for any normal operating system... i.e. Linux. This is the the same reason why Apple rejected the Cell.
And Linus just shot himself in the foot:
"I _want_ people to expect that interfaces change. I _want_ people to know that binary-only modules cannot be used from release to release. I want people to be really really REALLY aware of the fact that when they use a binary-only module, they tie their hands"
Huh? He wants people to code for Linux after saying this? I'm sorry, but if this is his attitude I will never ever write a driver for Linux. He's telling me my driver won't be compatible from release to release BY DESIGN? Does he really want people to recode each release to get their driver to work? Does he think this looks friendly to developers? Does he think commercial companies are really willing to do this every time? Does he think end users will really be willing to wait for recoded drivers every release? C'mon, this is ok for an OS you're writing for a hobby, but this doesn't fly in the real world (where I'm sure a lot more people would like to see Linux).
Yes, adding that abstraction is going to slow you down. Darwin is slower than Linux. Why not fork it? I mean... Sacrificing slowness for usability? An operating system can be faster than anything else out there, but I won't code for it unless it has a stable usable API and my end users can easily install the drivers. Sorry. Apparently Linux is anti that for device drivers.
Huh? We're not talking drivers that come with the operating system. This is like not being allowed to put a custom engine in your car unless it's approved by your auto maker according to their standards. If Linux wants to be an operating system about choice they need to let people choose whether or not they want closed source drivers in their system. It's not their job to enforce ideals. That's what Microsoft does.
What sense does this make? You can run closed source software under Linux. Why not closed source drivers? It's this kind of thought that is killing Linux's potential while Mac OS X passes it by.
The Cell is not good for mainstream computers. It's way too hard to write software for, and moving to Cell probably would have meant re-writing all the current Mac applications. The Cell isn't really just an upgraded PowerPC chip. Instead it's mostly built for very specialized applications. I even have doubts it was a good choice for the PS3 seeing as how much work publishers are having to put into making PS3 games.
Actually, no. The PPC version is also 32 bits. It just has the ability to run 64 bit applications and address more than 2 gigs of memory per application. The OS itself is not compiled for 64 bit.
Read your own article. That's retail market. Like you know... It's only counting stores you walk into. It's not counting companies like sayyyyyyyyyyy Dell?
You're missing the point. When the iTunes Music Store came out, it didn't revolutionize music at first. It's a gradual process. Don't think it's impossible to add video support to the AirPort Express either. Then the user just has to add another box to their stereo setup.
Or I could ignore your long, drawn out, and inaccurate comment and continue watching iTunes shows on my computer with tv out. Or maybe even hook a video iPod up to a tv via it's tv out jack.
You're missing the point. Having more than one Linux operating system is going to confuse consumers. They don't want that. They want simple. They want easy. How about this... Let's make one Linux distribution. The Debian people can manage the server aspect and applications, the Fedora and SuSE people can manage desktop experience, Mozilla can manage the internet stuff, etc. This is ideal. Everyone is working on maximizing the same OS for different needs. Instead, the server people want to code their own version of Linux, the desktop people want to code their own version (with a different version depending on which desktop experience they think they should be having). The end result? Instead of having "Linux Server" and "Linux Client" you have tons of different versions that people who don't want to have to think about their computers, much less choose an OS. If Linux proponents want Linux to be mass market, they're going to have to realize people don't want to think about what they want their computer to do. They want it to just work. Linux people always look at Linux from the geek perspective. They think the more powerful and complex the OS is over Windows the more people will adopt it. People don't want a complex geek toy OS with options. They want something they buy, install, and use, no thinking involve. Yes, I know geeks will find this a horrible idea, but thats the real world. In the real world people don't care if they are running Windows 95, Windows 98, Windows ME, Windows 2000, or Windows XP. It's just Windows. It just works. They picked it up at the store when they asked for Windows. And they don't want to care which version they are running. As long as they can go to the store and buy their software and it works on their machine it's ok. If Linux people think normal users will be attracted to Linux's "diversity" they're wrong. People see that as complicated, something only geeks can understand, and something to be avoided.
Of course if Linux wants to stay an operating system by geeks for geeks, diversity is a good thing. But as long as Linux is an OS built for geeks mainstream users will avoid it like the plague. Microsoft ruled the world by making standards with Windows. Linux needs to do the same. Make a standard.
Please note that all the car brands have steering wheels, an accelerator, a key to turn on, and doors with door handles. Yes, lots of cars can do different things but they all operate the same way. If there was one kind of Linux and you had a server version and a client version, that's easy. The way it is now Linux doesn't even install the same way for each version. There is a difference between having one kind of Linux tweaked to do lots of different things, and lots of versions of Linux all tweaked to do the same thing.
And this is why Linux will never catch on in the mainstream. Choice is bad for mainstream consumers. They don't want to have to decide. They want one option they know will work. Why do most consumers choose Windows? There is one version, it's compatible with everything else built for Windows, and they don't have to think when they choose it. A consumer is not going to want to have to research which version of Linux they want. They want to go to CompUSA, ask for a copy of "Linux", go home, install, and be able to run Linux software. They don't want to have to decide which Linux is right for them. That's not simple. With Windows they just ask for a copy of Windows and tada, they're up and running. If Linux wants to be mainstream, they're going to have to sacrifice diversity. Mac has reached %6 of the market using one standard OS, maybe if Linux would all hop on the same boat together they'd to the same.