To say there is general confusion about Linux and open source software within many organizations is an understatement. The confusion seems to stem from misunderstanding the Linux and open source philosophy but is due in no small part to misinformation. Attempts at understanding Linux and open source are hindered by the decentralized nature of Linux and open source software.

Out of the confusion we find popular myths and misconceptions about Linux and open source. These myths and misconceptions result in the oversight of the greatest benefits of Linux and open source and hinder their adoption in businesses big and small. Within this article, I highlight some of the more common myths in hopes of clearing them up.

Myth #1:
All Linux is the same
There is a common perception that all distributions of Linux are the built equally, and when someone refers to Linux they are referring to all open source software. I can't count the number of times someone has asked me "Oh, you use Linux?" when it turns out they really meant to ask if I used a particular distribution. In addition, the perception is that all of the software that comes with a particular distribution of Linux is Linux as well. For example, over the last year a number of studies purporting to be scientific compared Linux and open source security to Windows. The gaping flaw in those studies is that they compare all open source software with just the core Microsoft Windows operating system. Not even Internet Explorer, which many people consider to be part of Windows, is included in these comparisons. The result is that the studies end up comparing the thousands of software packages included with a distribution like Red Hat with a very small subset of the entire Windows environment. An accurate comparison would include all components and all software, including third-party software, that can run on Windows.

Questions of GNU/Linux origins aside, Linux is, quite simply, the kernel - the core of an entire operating system that consists of many parts, including software that runs on top of the kernel. Most Linux distributions include (or have available) a common set of software such as KDE, GNOME, GCC, and other *nix type utilities. However, many distributions also make enhancements to that software and even to the kernel itself. For example, Red Hat utilizes the Bluecurve interface for those that use X Windows. Red Hat also adds a number of customizations to the kernel that are specific to the version of Red Hat being used. Other distributions feature advanced package management such as Debian's apt (dpkg) system, which makes updating a large number of computers incredibly easy. None of these components are Linux, but they are parts of a particular operating system distribution.

Open source does not always mean Linux. Open Office, a powerful office suite, runs quite well on Microsoft Windows, Mac OS, Solaris, and Linux and features compatibility with other office suites, such as Microsoft Office. The Mozilla Project, including the Mozilla Web browser, the Firebird Web browser, and other related software, runs on Microsoft Windows very well and has some advanced features just becoming available in Internet Explorer, such as pop-up blocking that really works.

Myth #2:
If we ignore Linux and open source, they will go away
While this might have been a tenable position in 1995, it is no longer an acceptable answer for technology architects and decision makers. Linux and open source are much too valuable to simply be ignored. More and more vendors are porting their software to or building it specifically for Linux. The proven and excellent track record of Linux and open source in critical projects shows that it is relevant, stable, and powerful.

Additionally, since Linux and open source aren't centralized there is no single-vendor lock-in. With Linux and open source you don't have to put all of your proverbial eggs into one basket. This grass-roots, decentralized development model is the most difficult for many people to grasp, especially those who would compete with Linux and open source.

Speaking of Linux and open source competitors, the anti-Linux and anti-open source campaigns flowing out of Microsoft show how serious Linux and open source have become. If Linux and open source weren't already players, Microsoft would've continued with their policy of ignoring them in hopes that they would go away.

In today's IT environment, you need to accomplish more with less. You need to fulfill more projects, administer more servers, provide more ongoing support. Linux and open source enable you to do more while requiring less. With this statement, I'm not solely talking about hardware; today's processor will always beat yesterday's. Linux and open source are key to being adaptive and delivering the return on investment necessary to gain a competitive edge.

Myth #3:
There are no support resources for Linux and open source
Obtaining support for Linux and open source has never been a problem. In fact, I've had much more difficulty obtaining support from companies with which I have a paid support contract. The key? Rethinking how you go about obtaining support.

Here's my secret for obtaining Linux and open source support. I keep this in mind whenever I find myself needing assistance: "I am not encountering a problem that someone else hasn't already seen and solved."

I call it a secret because it seems as though too many people overlook the great support resources right under their nose. With that secret in mind, a quick search of Google or Google News (formerly Deja News) seemingly always yields results. For those rare problems for which no Google answer is readily available, a post to the appropriate mailing list, Web forum, or Usenet newsgroup usually gets an answer within a few hours.

I will grant that there are exceptions to the rule. Many times this seems to result from postings that begin with, "It doesn't work, please let me know what is wrong with your software." This type of attitude doesn't win friends regardless of whether the support is for open source or closed source applications. There is no replacement for solid troubleshooting skills by system administrators, whether using open or closed source software.

Many companies that produce open source software offer direct support contracts as well. A great example is MySQL AB (www.mysql.com), which offers a number of levels of paid support for the MySQL database server. Their support offerings run small to large, depending on your needs, and include everything from basic support to having a MySQL developer come on site and develop custom enhancements for your organization. MySQL AB is just one of the companies offering paid support for their open source software. Another example is Open System Consultants (www.open.com.au), which has a variety of products and also offers support and training. I could go on, but you get the point.

Many consulting companies now offer support on a wide variety of open source software as well. I can guarantee that if you ask for it, they will consider it. Consulting companies make money by being adaptive to change in the environment. If the consulting firm that you're with doesn't have at least one Linux and open source expert on staff, you might be well served looking for a new consulting company.

Myth #4:
We have no internal expertise for Linux and open source
Having no internal expertise for Linux is a problem. More important, it is also a self-fulfilling prophecy and an excuse that can be used only once or twice. In other words, if you're in a visionary or decision-making role and you haven't encouraged Linux and open source knowledge within your organization, it might never get off the ground.

The seed of Linux and open source curiosity is probably already there in many of your technology experts. The expertise and organizational knowledge will expand if you allow it to grow. The key is to encourage and reward learning and innovation. If your staff is required only to maintain Windows then there isn't any incentive to learn Linux and open source.

Challenge your technology experts to learn Linux. Encourage the use of Linux on your admins' desktops. One of the best ways to learn something is by repetition. Using the excuse of having no internal knowledge of Linux indicates that the decision-making process is much too stagnant and is not adapting to change in the way it should be. Cultivating Linux and open source knowledge enables you to leverage the strength of Linux and open source for a competitive advantage.

There is much anecdotal evidence that many system administrators already have some Linux experience. The admins might be experimenting with it at home. This was the case for a large financial firm who found after migrating to Linux that their admins already had a foundation of Linux skills upon which they could build.

Myth #5:
Linux is difficult to secure
Linux and the open source software that runs on top of Linux are no more difficult to secure than any other software. I believe that open source software, is many times easier to secure than its closed source counterpart. The configuration options for security are usually laid out in the documentation with open source software and configuration changes are almost always simply a matter of updating a text file and restarting the process (as opposed to restarting the entire computer). Anyone who has ever edited the registry to enable a security option can relate to the seemingly hidden nature of some configuration settings. By itself, having to edit to registry is no more difficult than recompiling a Linux kernel, but finding documentation for a closed source software vendor's security settings is too frequently a task that requires an extra support contract on top of the fee for the software itself.

Since the source code is open and freely available, Linux and open source enjoy some of the best peer review in the world. If someone submits code for addition to a project, the code is examined by other developers who may improve upon it, reject it, or add it to the project. If the code is poorly written or will cause problems, it can be caught by any number of developers working on the project.

Some argue that Linux and open source are in fact more susceptible to attack because hackers can access the code to look for vulnerabilities. This argument is quite narrow minded and in many ways ignorant of history. The experiences of secret writing and eventually cryptography show that some of the biggest failures in intelligence throughout history have come about because someone relied on keeping the encryption method a secret.

Along the same lines, keeping source code a secret does nothing to enhance security. Keeping the source code secret actually hinders security when a vulnerability is released or the source code is compromised. Users of closed source software are at the mercy of the vendor. Will the vendor take the time to fix the security issue? When will they release the patch?

Myth #6:
Linux is free
Actually, Linux is free. So why is it a myth that Linux is free? The word "free" has many people confused. Some believe, when referring to Linux, that free means free lunch. On the contrary, when referring to Linux and most open source software, think of free as in free speech. You are free to do with the software as you like. It's the freedom to open the software, look under the hood, twist something here, turn something else there.

However, many distributions of Linux and countless open source software packages are also free as in free lunch. There is a large amount of press coming from those in the proprietary, closed source world that Linux isn't free (as in free lunch) and that open source solutions cost more money than their closed source counterparts. Such claims are nonsense. I've never seen nor heard of a valid independent study done that claims such a thing, nor have I ever seen evidence that Linux and open source are more expensive to administer in the long run.

Imagine for a moment that someone offered you the car of your choice for free. I'd probably take a nice new McLaren F1. The car is free, no one can argue that. But the car does me absolutely no good without fuel, oil, various fluids, an operator's license, and probably some insurance.

Now consider the case where no one offers me the car for free. First I'd have to buy the car. But I'd still have to purchase the other vital bits such as fuel and insurance. Now my cost to own that car has skyrocketed because I had to pay for the car.

It can be argued that the actual software cost is a small portion of the total cost of the project. I agree in general with that premise but saving 5-10 percent on a project can give you the advantage you need. The bottom line is that regardless of which operating system or software you use, you'll still end up buying the same components - such as personnel and hardware - to run it. However, with Linux and open source, you save a fair amount of money on the software and operating system. Further, you're not locked into a vendor, at their mercy for licensing costs.

I could also argue that Linux and open source run more efficiently and thus don't need as much hardware as some closed source operating systems and software. In addition, I could argue that administration of Linux is much easier since you don't have to reboot the entire server just to apply a security patch for the Web browser. Alas, those are topics for another column. All of these factors add up to short- and long-term savings on projects that use Linux and open source.

Myth #7:
There is no training available for Linux and open source software
Though becoming less of a myth daily, the perception that there isn't any training on Linux and open source still remains. Typing "Linux Training" into Google yields 1,650,000 results. I'm quite sure that not all of those hits are training offerings, and then even some of those may not be from entirely reputable firms. But with names like IBM offering training on Linux and open source development, one cannot successfully argue that training in Linux doesn't exist.

Myth #8:
Linux and open source software are difficult to deploy
Another self-fulfilling prophecy related to both training and cultivating internal expertise is the myth that Linux and open source software are difficult to deploy and maintain. I will concede that an administrator coming from a Windows background will be slower to deploy Linux than someone who has used Linux for years. By the same token, it would take additional time for that same Linux expert to deploy a Windows-based solution. In the end, you have to look toward the future and start encouraging Linux and open source expertise within your organization.

Some of the most difficult and ultimately failed projects that I've been involved in have been with closed source software and proprietary support. From poor, contradictory, or nonexistent documentation, to poor, contradictory, or nonexistent technical support (even when a support contract has been paid for), the deployment of closed source, proprietary software is no panacea. On the contrary, the deployment headaches tolerated with closed source software are the result of the lowering of expectations that occurs after being disappointed time and time again.

Myth #9:
Linux and open source aren't ready for prime time
Looking to the stability of Linux and open source as a means to delay implementation is, at the very least, questionable. For years, I worked for an Internet provider where all of our services, from Web, to e-mail, to authentication, to ftp, and others, were based solely on Linux and open source applications. The computers frequently had uptimes of a year or more, which is unheard of for computers running other operating systems responsible for thousands of users.

One of the primary lessons learned at LinuxWorld Expo over the past year is that Linux is in business. Linux is being used successfully in countless businesses today and is gaining market share and mind share every day.

Myth #10:
All open source software is created equal
I can't, in good conscience, write about the high level of stability with Linux and open source without writing a little about some of the open source offerings that are available. I will concede that not all open source software is created equal. Some open source software, just like its proprietary cousin, is not that good. Some software is unstable or may not work at all on your platform. Open source software is not immune to the same problems that plague proprietary software.

A key difference is that open source software doesn't require huge investments just to find out that the software won't work. In addition, you ultimately have access to the source code and can adjust the code as necessary for your implementation. How many times can a business sign a large contract with a vendor only to find out that the software doesn't do everything that the salesperson said it could?

Most open source software that you'll encounter for enterprise projects is quite stable, has a good user community, and solid development. However, I did feel it necessary to point out that not all software, whether proprietary or open source, is equal. If you have a bad experience with a piece of open source software, you shouldn't consider it an example of all open source software or of the development model itself.

Conclusion
Linux and open source software are surrounded by myths and misinformation. Decision makers need to cut through both the myths and misinformation to come to a clear understanding of what Linux and open source software can do for their organization. Today's IT is about connecting vertical business silos to an environment that can provide the best value while enabling rapid response to changing business processes. Utilizing Linux and open source successfully within an organization requires expertise, but more important, it requires vision. Armed with the knowledge to challenge the myths, you're one step closer to gaining a competitive advantage.