Originally Published: Sunday, 18 June 2000 Author: Tommy
Published to: featured_articles/Featured Articles Page: 1/1 - [Printable]

Community Forum: Somebody, Quick! Give Linux a Fork!

Many Linux industry watchers claim that a fork is unavoidable and will destroy Linux's chance at being a mainstream OS. I say that it is unavoidable, and it will help Linux if it's done right.

   Page 1 of 1  

Many Linux industry watchers claim that a fork is unavoidable and will destroy Linux's chance at being a mainstream OS. I say that it is unavoidable, and it will help Linux if it's done right.

For Linux to be "mainstream," it must supplant Microsoft Windows 9x. To do this, Linux must become easier to use than consumer MS-Windows. The Linux community also needs to keep its collective eyes firmly on the future. If Linux can stay a step ahead of consumer Windows on upcoming technologies, it will usurp users as Microsoft slowly loses them. This is already starting to happen with NT/2000 users, but for 9x users it's a much different process.

What about forks? Where a fork is desperately needed in Linux is the same place Windows has a "fork" -- between its Server/Workstation "distributions" (NT/2000) and its consumer "distributions" (95/98). A well-placed fork could do lots for Linux.

Let's say that Corel Linux and Caldera OpenLinux (since they are aimed at the desktop) decided to remove all the traditional *nix tools that home users don't need or want. On top of that they replace X with a new GUI, one designed for graphics, not networking. This would absolutely be considered a fork. However, it would allow them to turn Linux into an OS that's faster and easier for home users at a much faster pace than trying to get the "Linux community" to adopt something new.

On the same side of the coin, binary compatibility could still exist. To port a well-written application from Corel/Caldera to Red Hat or TurboLinux would not be any more difficult than porting an application from Windows 98 to Windows 2000.

One thing Linux needs to compete on the desktop is to learn from others. BeOS, for instance, is faster and easier to use than Linux's best commercial "desktop" offerings. Sure, Corel/Caldera has about the same out-of-the-box functionality as BeOS. But, in some respects, BeOS has better (out-of-the-box) hardware support. You do not have to recompile your kernel just to use a standard ATAPI CD-RW. You can use your softmodem. Not to mention the BeOS filesystem is ready for the future -- it can throw around an 18 petabyte size file. Now that's some serious media! While Microsoft cannot possibly bandage FAT enough to do this, Linux is nimble enough to compete with BFS. Unfortunately, EXT2 will probably not be replaced or upgraded until another mainstream OS's filesystem is smoking it.

I believe Windows will slowly disappear from home PCs regardless of what happens in court, its just too old and all the patches in world can't save it. Anyone who has ever used Linux or BeOS on a regular basis has seen what an OS can be like. Linux users tend to dismiss BeOS just as quickly as Microsoft dismissed Linux just a short time ago. If Windows 9x should start losing its market, BeOS is already ahead as far as technology or ease of use goes. But applications sell the OS. Well, BeOS may be a winner there too.

While I wouldn't dare to argue the overall quality of BeOS vs. Linux applications let me throw some numbers at you:

LINUX: http://www.linuxlinks.com/Companies/Software_Development/

BEOS: http://www.belinks.com/be_softcos.html

BeOS already has more commercial desktop application companies springing up than Linux. Most of the commercial software developers listed for Linux do not create desktop applications, where most of the BeOS ones do. There are more Windows software developers supporting Linux than BeOS, but this could change in a second considering the "Linux business model" is still suffering industry growing pains. BeOS is commercial, and has a comfortable business model.

Firm eyes on the future. What Linux needs is a fork, not just any fork, but a standard fork -- one path for the server/workstation and another for the home user/media. In my opinion, if Linux does not get a fork soon when Linux gets to the supper table, Be Inc. may have already eaten Windows.

Tommy is a PC technician and salesman in a small PC shop in Indianapolis IN. Flames can be sent to tom6789@sprynet.com, but they won't be answered.

This article is provided under the OpenContent License, and was first published in osOpinion.

Please send your editorials for consideration in the Community Forum to editor@linux.com.





   Page 1 of 1