Linux DevCenter    
 Published on Linux DevCenter (
 See this if you're having trouble printing code examples

Open Source Advocacy for the Enterprise

by Jono Bacon

Within the reaches of open source advocacy are a wide range of different sectors and industries. Inside each of these, small sub-communities of people have appeared with the intention of educating and equipping their sector of choice. And while many are spreading the word in more humanistic and ethically satisfying areas such as charities and education, they have left the so-called "enterprise" sector largely in the hands of big-business vendors.

The word enterprise is actually quite misleading. It doesn't really mean anything. In reality, enterprise tags a piece of software with marketing that managers feel comfortable with. Often targeted at high-end large enterprise organizations, the word enterprise also has the side benefit of making their little cousin the Small to Medium Enterprise (SME) feel bigger, and more like an "enterprise." The wacky reality is that if you put the word enterprise in the product description, it will sell better.

In the enterprise sector the guiding light of a community of open source aficionados is unlikely to have an impact. If the technology does not have the language and presentation expected in the sector, it will not sail. Part of the reason why the sayings "No one got fired for buying IBM" and "No one got fired for buying Microsoft" exist is because those organizations have spent years developing, refining, and actually creating much of the language.

There is of course more to doing business than language alone. Another strong tradition is vendors and their competitors batting back and forth case studies, reports, Total Cost of Ownership (TCO) figures, and other facts. This act, something I call "corporate Pong," embodies a culture of demonstrating whose technology performs best under the statistical microscope. As I explained in Unwrapping the numbers game, this is almost entirely redundant. As an open source consultant, every single organization I have ever dealt with has been drastically different in their requirements, abilities, resources, and desires. So-called "case studies" demonstrating a lower TCO or benchmark do not mean anything due to the sheer variations in how technology is deployed in different organizations.

Related Reading

Open Source for the Enterprise
Managing Risks, Reaping Rewards
By Dan Woods, Gautam Guliani

How do you get open source into this challenging sector? The best method is to identify not only the cold, hard requirements but also the soft requirements that transform the solution from simply software into something that actually makes better sense and better business: a solution offering a better way of managing IT, wrapped in "enterprise" language.

Real-World Problems

One of the most fundamental issues in the enterprise world is that a solution hinges on the perception of how well it solves the problem. Decision-makers don't care about the philosophy of software, its ethics, or even its technical prowess in those situations other than the task in hand. For most of these businesses, the prime requirement is that IT does the job and it does it well. The reason for such a cut-and-shut solution is that the decision-makers are not hackers and simply have no interest in a clever, witty, and flexible solution. They are interested in a predictable, measurable solution--a comfortable car as opposed to KITT from Knight Rider.

In addition to a comfortable, predictable solution, enterprises also look for systems that are easy to staff. This is one of the reasons why archaic Unix systems have fallen out of fashion in preference to systems such as Windows. As Windows has provided a conceptually easier path into IT over traditional Unix systems, more people have hit the job market and the cost of finding and employing a system administrator has become more viable. As Linux has developed as a commercially practical operating system, the skills market is improving steadily, but this has traditionally been quite a problem in open source adoption. Some open source advocates tend to forget about this issue or fail to see the importance. In reality, a skills shortage can be problematic enough for a company to actually shift from one platform to another, as the Royal Institution of Chartered Surveyors case demonstrated--a case where an entire organization shifted to a Microsoft platform due to a skills problem with their solution!

Finding Solutions

Open source is actually quite an applicable technology for many of these organizations. The technical stability and security benefits commonly cited with open source software are largely due to its community-driven development process. In addition to this, the combined contribution of paid developers from companies such as IBM, Novell, Red Hat, HP, Oracle, Mandriva, Canonical, and others keeps the open source offering moving forward in a variety of different areas. When you consider advocating open source to an enterprise, identify the technical requirements and then explore which of these major companies contribute to the open source stack. Also find conclusive results about the performance and uptime capabilities as a whole--the established names behind the solution will certainly add confidence in the technology. Gather your results and put them together into a short document. Remember that your manager will not want to read reams about this. Short, sharp bullet points are the order of the day.

One of the most fundamental benefits of open source is its potential integration. In many organizations, multiple separate systems work in largely separate ways, with little integration or communication between them. With open source, there is a huge potential for integration to occur at the operating system, application, and even the code level. To demonstrate this, consider a use case:

A thin client system from the Linux Terminal Server Project provides the desktop for all of the machines on the network with the single sign-on system handled by LDAP and Kerberos. When a user logs into the system, he/she accesses the web-based CRM Sugar and the company website (powered by Mambo) in Firefox, and authentication happens automatically. Sugar can import content and contacts from other tools and output data in XML RSS feeds. These feeds can then be connected into Mambo, read by web-based aggregators, or used with other applications and systems for such uses as generating reports, updating other systems with statistical data (such as stock tracking), and more.

This use case provides a good example of how a variety of open source tools can fit together to form one large and consistent system. The value here is not just in the individual components, but also in the bigger integration picture. This integration goes further when you can hook the open source components up to existing systems using technologies such as XML, RSS, and Web Services. The trick in helping to push open source as an aid to integration is in proving that the integration is not only possible, but also cost-effective. To achieve this you will need to check pricing with vendors and consultants to determine how much it will cost.

Within this sector, cost is important but it is not the only driving factor. In the previous installment of this column, Maria Winslow, the author of The Practical Manager's Guide to Open Source and contributing editor to LinuxWorld magazine, advises that "solid evidence of how you can save money with open source will speak louder than most evangelizing." Maria is correct that saving money will mean more than most evangelism techniques, but it is also key to remember that the importance of costs varies tremendously between different organizations. Within the large enterprise sector, it is not uncommon to spend inordinate amounts of money on IT. In smaller SMEs, cost is more important; the price of a dollar means more the smaller the organization gets.

Experience Paves Way for Successes

There is no silver bullet in advocating open source, but the best approach is to understand fully the sector to which you are an advocate. This article has explored some of the issues involved in the "enterprise" space, but there are still a great many issues to explore. As I explained earlier, corporate Pong means nothing due to the variation between different organizations. The same rule applies to advocacy--there is no single method of advocating that will work everywhere. What can work is to explore real and tangible advocacy methods that improve the quality of your message and better direct it to the right people. This process certainly needs your input, so please share your thoughts.

The next article will explore the challenge of the skills gap and look into methods of advocating open source to train people in the open source skills that are becoming more and more relevant.

Jono Bacon is an award-winning leading community manager, author and consultant, who has authored four books and acted as a consultant to a range of technology companies. Bacon's weblog ( is one of the widest read Open Source weblogs.

Return to the Linux DevCenter.

Copyright © 2009 O'Reilly Media, Inc.