What’s Required for Linux to Overtake Apple Mac OS X?

As this site has noted, Canonical and Ubuntu founder Mark Shuttleworth at OSCON called for Linux distributions to leapfrog the quality of Apple’s Mac OS X user experience. So, what will be needed? Here are some thoughts.

A smooth interface, for one. There need to be application design guidelines, just like those on Mac OS X and Windows Vista. Creating a smooth application interface takes more than just a good looking theme. An interface must also include flow, so applications that are included should work together, and for the most part work alike.

Another goal should be standardizing the interface and applications for Linux by creating a default set of applications that become the focus of development.

More Here



Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

re: What's required for Unoobtu to overtake Mac

Money, lots and lots of money.

Unless Canonical bought Apple and Adobe it will NEVER happen (and since they don't have enough money - it's pretty much guaranteed not to happen).

So until you have the complete line of Apple's iWhatever app's and the entire Adobe software catalog ported to run native on Linux - Unoobtu will NEVER come close to dipping into Apple's market share.

Once again, the article assumes eye candy is ALL the user is worried about - it's not, it's all about the applications.

If you doubt that little bit of wisdom - ask a handful of non-geeks what OS means. Few if any will know it means Operating System - fewer still will be able to even come close to describing what an Operating System does.

Some days, I just don't care . . .

I get tired of all the hand-wringing articles attempting to explain why Linux, though it's FOSS and low-cost, hasn't captured more of the global desktop market share.

In my more cynical moments, I just don't care.

Yes, I do like Linux, and use it almost exclusively. I love the community, the freedom, and the variety.

But, looked at selfishly, sometimes I only really care that Linux gains enough global market share to ensure that the next new computer related techie device I buy from the store will have a fully-functioning Linux driver included with it.

Suppose, for example, that I buy an iPod (which I haven't done, and won't do--I have a Meizo). I would like to know that it will operate with my Linux desktop computer without my having to do 3 days worth of research and fooling about to get a result that works with 85% functionality under Linux.

What will it take in terms of global desktop share to get companies to do this? 5%? 10%? 15%? I really have no idea. But, if Linux should get, say, 10% desktop market share, and this is enough for companies to start bundling Linux drivers with their stuff, then the remaining 90% of the computer using world can go hang.

Don't get me wrong--that's my cynical side speaking. Usually, I really do want others to discover and enjoy the fruits of using Linux and FOSS on the desktop. And I do hope that company/vendor made device drivers will be released on an open source basis. And driver issues are certainly better today than they were 6 years ago when I started using Linux.

But for once, when browsing at Fry's or online at Newegg, I just wish I could buy something without having to worry about Linux compatibility.

And, of course, I couldn't care less about Ubuntu and Canonicals' market share.

Cheer up with ultra light laptop, Linux gets easier, instant-on?

The trend is new form factor and new architecture, maybe instant on.

Linux will be perfectly usable with cpu, nands and USB ports. I hope that USB devices will give you free firmware and drivers.

Its time to change too. Price will be right in the next few years.

The future is faster cpu and nands, plus reasonable wait state to USB bus. Can you ask for more when it is even affordable.

Then you have to own one of those smartphones to beat the ultra light by sheer weight?

Footnote:
Intel atom cpu has 512kB of L2 cache. Since most 32 bit cpu needs only 64 kB of L2 cache, it is sufficient to use swap file on nand memory(4 gB), and use part of the L2 cache as L3 cache; thus eliminating the use of expensive drams. Nand uses raif data compression. Linux file system can serve as controller to rotate nand blocks. Use a block(LBA) for temp data, when transistor fails, switch to another block. Nand is best used by splitting operating system into execution area(cpu address, 100 to 16 mB) and apps area(16 mB to 3 gB), then use L3 cache to prefetch instructions and data; and use swap file upto cpu address 4 gB(in blocks) for temp data. Linux file system for swap will alternately use from address 3 gB and from the end address 4 gB backwards. If swap file nand fails, append another new nand to the main nand? So, you need two slots of nand minimum.