Linux vs Microsoft XP: Optimizations Make Linux the Killer Desktop

Last month, when I tested Linspire 5.0 for my series on Linux desktops for government enterprises, I discovered NVU. At the time, I realized something special had happened to the Linux application inventory. As the NVU Website states:

Finally! A complete Web Authoring System for Linux Desktop users as well as Microsoft Windows and Macintosh users to rival programs like FrontPage and Dreamweaver.

A tad skeptical, I didn't totally believe it. So, I downloaded NVU and tried it on a Fedora Core 3 desktop before I turned it lose on other Linux distributions. Having had to use MS Frontpage more than once on projects, I knew the good, the bad and the ugly of that application. The "good" was tainted with Frontpage extensions on the server side that slowed websites and took massive amounts of disk space. The supposedly WYSIWYG editor didn't always translate to what I had in mind when it went up on the web site.

NVU had all the good qualities of Frontpage without the unnecessary overhead and quirks that take up so much development time. What you see in NVU turns out to be exactly what you get. It's a better application than Frontpage and it's free, open source software.

That's just a prelude to this article. It demonstrates an example of how far Linux has come in the last year as a desktop. For end users Linux provides a superior user experience.

Making Headway

In April 1999, D.H. Brown Associates, Inc. published a report called Linux: How Good Is It? Hardly any archives exist today mentioning that story with the possible exception of this C/Net News article. The study dinged Linux for lacking features needed to make it a serious consideration as an operating system. The report said that Linux was good for file and print servers, Web servers, some scientific computing, and thin client computers. But, the DH Brown report said Linux lacked support for computers with multiple processors; failover and a "journaling" file system needed to reboot a crashed machine without having to reconstruct the system files.

From that point on, the kernel developers began focusing on Linux as an alternative to top-end, trusted UNIX and Windows NT servers. In short order, Linux became much more than a hobbyist system. It became a commercial, industrial strength system capable of competing and surpassing existing business servers. Only market politics, extensive lobbying and sweetheart deals have kept Linux from taking the server business completely away from Microsoft.

Now, that Linux has become the leading server platform in the commercial world, development has shifted to the desktop. Like the post-D.H. Brown period, a flurry of activity has put the Linux desktop in a position to become dominant. People who only looked at Linux six-months ago when O'Reilly & Associates released Exploring the JDS Linux Desktop will not believe the progress made since then.

JDS looked like the leading Linux desktop at the time. Today, it's behind the competition. And while a new release of JDS is forthcoming, it may also trail other Linux desktops in features and capabilities.

Desktops for the Enterprise

Imagine a situation where you don't have to activate Microsoft Client Access Licenses (CALS), manage Windows 2003 License Servers and an Active Directory or worry about support for old world Windows distributions. Also imagine an alternative desktop that provides complementary innovation, works well in existing Microsoft infrastructures and provides a real reduction in costs. Linux provides all of that.

Linux also can logon to an Active Directory like Windows desktops. The Linux logon manager, GDM can handle expired passwords. You can run Windows applications through terminal services using rdesktop without Citrix server extensions. The Evolution workgroup client works well with Exchange. You have interoperability with Microsoft Office file formats using Openoffice.org. You also have a safer browser experience with Firefox. Tweak Linux for pure desktop performance and it's fast, safer and more fun.

A decade ago, PC's were not the dominant corporate infrastructure. Though gaining market share, PC's still had to use terminal emulators to connect to mainframes and run applications off of a variety of heavy metal. I remember programming Oracle Financials where the PC logged on to a HP-UX server 1500 miles away. I also remember having to test forms and reports on another system on the other side of the continent. I also remember supporting a triage HMO application running on a mainframe 3000 miles away on terminal emulators.

The supposed magic of Microsoft and Sun has nothing to do with their great cleverness, marketing prowess or innovation. They happened to have the products people needed to expand during the uptake in the World Wide Web adoption period. IBM mainframes using SNA instead of TCP/IP and Novell with its ill advised bet on IPX/SPX instead of TCP/IP doomed both companies. Apple didn't join the world until the release of OS X. People bought computers in droves because they wanted to be "on-line" and Microsoft offered a low-cost way to do that. Profits were high in Internet Services and among telecom companies and Sun's expensive but stable servers didn't present a problem on the bottom line.

Today, the infrastructure of the Internet is fairly build-out. Companies all have Local Area Networks and high speed Internet access. Sun and Microsoft do not have the same "value proposition". Both companies still have expensive products at a time when they aren't the only game in town. Enterprises now look for value. The shift to value has opened the market for Linux on Mainframes, in clusters and giant grids of PCs. When you look at Google, you see a company that arrived because it went with value and exploited their low cost computing power. I can't imagine Google achieving what they have with expensive Sun or Microsoft servers.

OK. So like a decade ago, enterprises have an investment in an infrastructure that costs too much to maintain. Mainframes became almost extinct because the maintenance contracts cost more than replacing everything with PCs. The same holds true today only instead of mainframes, people look at the cost of software as a percentage of the cost of hardware.

Once software represented a small percent of the cost of hardware but today the tables have turned. Software, dangerous software, costs more than the hardware on which it sits. It also requires people to upgrade their hardware because the software wants more memory, larger and faster hard drives and more powerful CPUs. What didn't make sense ten years ago, doesn't make sense today either: If maintenance costs more than an alternative solution, then the alternative solution should be vigorously pursued. The cost of maintaining my IT infrastructure today is higher than modifying it. And, I don't have to deal with another monopoly and vendor lock-in if I do modify it. Ten years ago, it was a monopolistic IBM oppressing IT departments, today it's Microsoft.

One notable exception exists today that didn't exist a decade ago: I can use my existing hardware. The same Intel based commodity hardware runs Linux. Ten years ago, when you hauled off the mainframe, you had to buy new PCs, wire the building with ethernet, set up routers and switches and buy new software. Today, you can leave the hardware infrastructure in place, continue to use your investment in old apps or provide terminal servers so that your Linux desktop users can still use their Win32 applications.

If you work in an enterprise, you owe it to yourself to test the performance of the Linux desktops. You can find articles on the Internet about performance tuning by doing a Google search or you can take a look at a book by O'Reilly & Associates called Desktop Hacks. You can also read about performance tuning in this week's Linux in Government: Optimizing Desktop Performance, Part II.

For Consumers

The Linux desktop provides an alternative many consumers have embraced. On mailing lists and in forums you can see thousands of posts where people declare themselves "Microsoft free". People do want certain freedoms and few doubt that as a sociological fact. Microsoft costs money and time. They hassle people. Their product costs are high. Consumers can find comparable open source software for free.

The same links above can serve you well if you want to try Linux and optimize it's performance. You will discover that you can use less expensive hardware and your desktop will run faster than Windows XP. The Windows XP you buy today is four years old and the next release of Windows is slated for 2006. It's still not a safe operating system because security has been added after the fact - when people began screaming about spyware, identity theft, viruses and trojan horses. The open source community designed Linux with security in mind from the ground up. Windows XP with security enhancements is annoying. If you secure your system correctly, you'll spend a lifetime answering questions every time you want to visit a new web site on the Internet.

Consumers face the problem that the Microsoft monopoly dominates the retail market space. You can't walk into a store today and find a laptop loaded with Linux on the shelves of your computer store. When you shop on-line every manufacturer states that they recommend Microsoft Windows. Of course they do, they get money from Microsoft to say that. And as a member of the Justice department once explained to me, you can't stop Microsoft from giving marketing rebates to manufacturers.

Franklin D. Roosevelt once sent a message proposing the "Standard Oil" Monopoly Investigation in 1938. He wrote:

"The liberty of a democracy is not safe if the people tolerate the growth of private power to a point where it becomes stronger than their democratic State itself. That in it's essence, is Fascism - ownership of government by an individual, by a group or by any controlling private power."

Many of us feel stuck today in various areas of our lives. We question many of the policies than govern us and wonder why we have to put up with them. I question how our government has allowed a bully to run amok controlling the computer market to the extent that it pervades what we can study in our educational systems. With enrollment in technical courses at a low peak perhaps it's because people don't see a future where Microsoft is the only option.

With the pending action in the European Union against Microsoft, perhaps the time has come once again for our legislators and regulators to question the existence of Microsoft. Old expensive software from Microsoft dominates the retail market. Microsoft's business practices still look like they effect a restraint of trade. In fact, if you go back to the 1999 trial, the issue of bundling software still remains unresolved. Not in the European Union where that's the central theme of their anti-trust action. But in the United States, Microsoft can still kill an unsuspecting business partner and take their market away. See Paula Rooney's article BlackBerry Killer? .

Some Final Thoughts

A friend and business partner recently commented that he felt like he was watching a comedy when he looked at the SNAFUs of Microsoft over the years. As Linux professionals, we both wonder how such a dull company could get into the position it holds. If people will start making real comparisons between Linux and Microsoft, then perhaps the comedy will end. At this point, you have to decide for yourself which is the right product for the times. I already have.

Respectfully submitted