There's no right way to do it, only ideas that are better than others in certain situations. But if you had the opportunity to head up the design of a new OS, one to Put Things Right, one that could be radical enough to varnish out those UI/X bumps that have clung on for years, but practical enough to be used every day, what would you design? How would you handle application management? What about file types and compatibility? Where would you cherry pick the best bits from other OSes and where would you throw away tradition? I've tackled this challenge for myself and present (an unfinished idea): KrocOS (warning: HTML5 site, will display without CSS in IE/older browsers). OSnews Asks: What would make your perfect OS?
To view parent comment, click here.
To read all comments associated with this story, please click here.
Member since:2009-02-19
I agree completely. It's become kindof a motto for the Linux Hater's Club around here: "X Is Old, X Is Bad, X Is the Problem!" It's basically just a sizable group of people saying the same thing over and over again, and hoping that having it endlessly and oft-repeated makes it meaningful.
X isn't perfect, but it's improving, and it's becoming pretty damned good. It's come a loooong way from several years ago.
That's quite true. There's enough common hardware that's well-supported in Linux that, at least in theory, you could have a vendor start shipping tuned-for-linux systems that'd be about as reliable as OS X is now. Well, almost: the 3D stack is still a little quirky, but, excepting that...
Member since:2009-08-26
So I guess Apple and Google are in the X hater's club too, right? Thom is in there too lol. You must be pretty deluded to believe that all this hatred is baseless or exaggerated.
X was built with poor design descisions that cannot be blamed on the typical excuse of buggy video drivers. There are plenty of cases of X crashing that show this.
Here is my favorite:
X crashes when mouse is unplugged:
http://ubuntuforums.org/showthread.php?t=496059
Member since:2008-03-10
A stable ABI will kill the open source Linux kernel in short order (I'm betting on ~5 years). It will guarantee that in the future your drivers are closed source. Maybe you don't care, but I do and the kernel developers do.
Maybe you should explain how that would happen, there's no reason having to download a driver from a third party or shipping them separately from the kernel will kill it. And how exactly will the demand for an operating system running millions of servers vanish? I can only guess you think that all drivers have to shipped with the kernel, or some other similar logical flaw.
Member since:2005-11-02
It's a simple process. Once commercial companies have a stable target, they target it. They stop releasing hardware specs. People use their drivers because such drivers exist, greatly driving down demand for a Free replacement or for enhancements to a Free version.
Repeat.
After a while you will clearly have *most* drivers for *most* hardware be proprietary. Linux, at this point, becomes fairly useless without such third party drivers... drivers distribution vendors probably cannot ship themselves (or, if they can, joe random hack-my-own-distro developer can't). Think the old nvidia redistribution problem for every single piece of hardware.
You cannot assume that companies will all do what nvidia (eventually) did and permit redistribution by anyone.
This is not certain, but it's certainly likely. Arguments like "It hasn't happened to $FOO, which has a stable driver interface" do not necessarily apply. Most Free operating systems do not have nearly the level of corporate interest that Linux has.
What Linux really needs is an improvement to workflow involved in getting and loading out of kernel drivers. That would solve most of the problems people actually want solved without any of the drawbacks.
Member since:2009-08-26
How will it kill the kernel? A stable abi would bring in better driver support which would bring in more Linux users.
Why are you so worried anyways? The Linux desktop has been in a coma for over 10 years, I'm not sure why you are so concerned with death when desktop Linux just sits there and drools on itself.
I'm really gonna laugh if an alternative like Haiku comes along with a stable abi and trumps Linux on the desktop. Defenders like you will have helped save desktop Linux from its own success.
Member since:2005-11-02
How will it kill the kernel? A stable abi would bring in better driver support which would bring in more Linux users.
Why are you so worried anyways? The Linux desktop has been in a coma for over 10 years, I'm not sure why you are so concerned with death when desktop Linux just sits there and drools on itself.
I'm really gonna laugh if an alternative like Haiku comes along with a stable abi and trumps Linux on the desktop. Defenders like you will have helped save desktop Linux from its own success. "
Who said Desktop Linux? I said Linux, the kernel itself, meaning all Linux.
If Haiku beats Linux on the desktop I will applaud. Free is Free. But, Haiku won't win because of its stable driver interface. It will win because of consistency, integration, etc, etc.. Things users actually see. If there were one desktop and one look and feel on Linux, with a single target for the menu, file locations, etc, etc then Linux would be having an easier time, too.
Desktop Linux has many, many problems that need solutions in order for it to succeed. A stable driver ABI is not essential and won't make a significant difference in user experience.
Member since:2007-11-04
The no stable driver ABI is brought up again and again as the reason that Linux is not succeeding on the desktop. However the people who make this argument fail to give any evidence that this is what is holding linux back. I've yet to see any company saying that the driver ABI is a serious problem for them. To the contrary the NVidia guys said in a recent interview that it's not much of a problem, and they are probably the most relevant provider of closed source drivers for linux. Similarly if the driver ABI is the main thing holding linux back, how come that Opensolaris hasn't overtaken Linux in popularity and why is it that Opensolaris has a lot less hardware support than Linux? (Note I'm not criticising Solaris, I think more open OS are always good).
I've yet to see these two questions being answered.
Member since:
2005-11-02
It seems like I am forever repeating myself. Once more!
A stable ABI will kill the open source Linux kernel in short order (I'm betting on ~5 years). It will guarantee that in the future your drivers are closed source. Maybe you don't care, but I do and the kernel developers do.
X is not the problem. Replacing X doesn't solve any problems and introduces a slew of new ones. There are few good reasons to replace X and a large number of reasons not to replace X. I'd love to hear why you think X is an issue so I can see if you fall in to the 90% of the anti-X camp that just doesn't know what it's talking about.
OS X works as well as it does almost entirely because it is a single vendor system. Linux distributors could do exactly the same thing. It is not a technology issue! The fact is that distribution makers don't want to do that level of integration (or at least they don't do it) and Linux's users would punish them with their feet if they departed so radically from the status quo.
I had high hopes for Ubuntu once upon a time, but they fell in step (in the end) with every other distribution in any number of ways.
I will repeat it once more for the tl;dr crowd: If you want Linux OS X you can build it 99% out of existing software available for Linux right now. Form a company, build boxes with a specific set of hardware, build your distribution, use app bundles. It can work! Today. Without replacing X, or changing the kernel.
If you have a small enough target even the unstable binary ABI is not an issue. If vendors only need to target one new kernel per year (at most) then there is no problem.