Originally Published: Thursday, 16 March 2000 Author: John F. Scipione
Published to: featured_articles/Featured Articles Page: 1/1 - [Printable]

Linux will Change the Way Computers Work

Linux will not only change the way you use your computer, but will change the entire way computers are used. Linux provides opportunities to the desktop not offered by Microsoft's family of operating systems. Even the much-beloved Macintosh OS cannot give the same experience as using Linux on the desktop.

   Page 1 of 1  

First look at http://www.linux-hacker.net/iopener/ to see an example of what I am going to rant about. It is not the best example, and it won't change the world, but it is a beginning.

Linux will not only change the way you use your computer, but will change the entire way computers are used. Linux provides opportunities to the desktop not offered by Microsoft's family of operating systems. Even the much-beloved Macintosh OS cannot give the same experience as using Linux on the desktop. Linux will change the way computers work because it has qualities that no other operating system has ever had; it is a free, open OS, that runs equally well as a server OS, embedded OS, and desktop OS. Linux customizability to the needs of the user, makes it an invaluable resource. But why does this mean Linux will change the way we use computers? Linux will change the way we use computers because it cuts out the middleman, by making only the server and the "appliance" necessary.

I am not advocating that we return to a mainframe environment, nor am I insisting that you sell all your computers to jump on the Internet appliance ship. I am simply protesting that it is a massive waste of resources for a business to provide every user with a $2,000+ workstation and then add server machines on top of that just to manage those workstations. It is also a waste of resources to throw away all your computers every 3 years and buy new ones to hold the latest incarnation of Microsoft. With Linux, organizations can have the best of both worlds, they can run inexpensive Internet appliances on a free operating system and connect to a server (or a Beowulf cluster) to run any application they need. The company will save thousands of dollars on software costs alone, not to mention the hardware cost they will save not having to upgrade to a brand new Pentium V 8000s come 2003.

Certainly, Microsoft and Apple will continue to promote the desktop for years to come, and for many home users, a single home desktop is sufficient for all of their needs. However, power users, companies of any size, educational institutes, and governmental agencies all require powerful server machines to store there Web, FTP and file/print servers. These companies also need access to workstations to do their work and remain productive. Linux provides a means to distribute resources such that an organization's workers may stay productive without the need of an expensive Pentium III or G4 under the hood. Linux can import X session, Linux can connect to the Internet remotely, Linux can mount remote partitions as if they were local disks, Linux has a powerful set of CLI tools that make remote administration much easier, Linux is small and will run on most hardware, Linux has both a powerful command line shell and several powerful GUIs, Some versions of Linux can be installed via FTP, or even run off a single floppy!

Perhaps Microsoft OSs could provide the same set of features, but will still lack at least four items that Linux has. First, Microsoft OSs are notably unstable, and constantly reinstalling the OS is simply not an option on and Internet appliance. Second, Microsoft OSs have very high system requirements which make them unusable on machines of small stature. Third, Microsoft OSs cost a lot of money and need to be upgraded every couple of years. Last, Microsoft OSs have many incompatibilities between their product lines -- Win9x, CE, NT, and 2000 -- which make coordinating a client-server environment difficult. Apple OSs also suffer much of the same fate. MacOS X, while a great desktop OS, with a reported 750 MB+ install cost is much too big to fit on the hard disk of an Internet appliance. MacOS X will also have fairly high system requirements, a least when compared to those of Linux, it will require at least a 300 Megahertz G3 processor to run when it comes out in late 2000/early 2001. However, in Apple's defense, the newest G3s are built without floppy drives, showing the high stability of the OS as compared to Microsoft's OSs (you don't need to constantly reinstall from scratch to keep your system running) and promise in the embedded market for Apple.

So what limits this revolution of the computing world, why has it not happened already? First, Ethernet connections are currently much too slow to accommodate the high transfer rate Internet appliances would require to run the bulk of their application off a server. With fiber-optic networks becoming more and more popular, especially in demanding businesses, this problem should be solved in 2-3 years (bold estimate). Secondly, Linux is not yet friendly enough to accommodate mass usage. With projects such as GNOME, Helix, and Eazel promising to create a user-friendly desktop for Linux, some say even easier than MacOS, user-friendliness cannot be far off. Thirdly, I know this may sound stupid, but not many games run on Linux; in order to get people to use Linux, their favorite game needs to be available, and even businesses suffer this from this stigma. Hopefully, with companies like Loki Entertainment Software, games will come to Linux. With Linux becoming a more and more powerful OS everyday, I think that vendors will soon be more than thankful to code games for Linux. However, games are usually meant to run on the local machine, rather than the server, and tend to take up large amounts of disk space, but I am sure that Linux will find a way around this; have you ever heard of a quake 3 server? Last, LCD Monitors are still of low quality and high cost, but are essential to the embedded device; it isn't very embedded if it has a 30 pound CRT screen stuck to it.

As you can see, the issue is mostly hardware and less software right now, but in the future, the hardware will be much less important. The software side is already falling into place, Linux is a great candidate, but others exist. All of the *BSD's, and BeOS could be used. But I guess that is what open source is all about, choice and new opportunities.

John F. Scipione is an engineering major at Monroe Community College in Rochester, NY. He wrote this essay to express his concern that his beloved operating system would repeat the past that its UNIX ancestors have, the forking of code into specialized and proprietary versions, and to provide solutions for said problem. His main background in writing this essay comes from incompatibility barriers he has met on Linux. "These barriers are for now, only annoying, but will become increasingly destructive if they are not attended to."

This article first appeared on osOpinion and is provided under the OpenContent Public License.





   Page 1 of 1