Recently, I have had the opportunity to work with a number of clients on their strategy to leverage the advantages that open source can achieve across their IT organizations. I found the major motivator and attention getter for open source adoption is the promise to cut costs; but increasingly, interests now include ways open source can provide control and enable agility.

Many folks in the early stages of open source evaluation think "Linux," which originally created a lot of the open source buzz. In reality, open source has expanded into just about every aspect of IT and software development you can imagine. It is even becoming a major part of commercial software according to Gartner who predicts "...by 2012, 80% of all commercial software will include elements of open-source technology."[1]

But despite its breadth, there is a tendency to lump "open source" into an initiative that does not fully take into account the different groups involved with it. It's a simple idea, but when discussing open source adoption, it's worth considering two perspectives present in any IT organization that develops portions of its own solutions.

The data center refers to "standing up" solutions and ensuring they are stable and meet operational requirements. This audience tends to see open source solutions as an alternative to commercial products such as operating systems, application servers, and databases. It is a fairly common understanding that "packaged solutions," offered by open source communities, are moving up-the-stack from the operating systems to full-fledged applications such as CRM or ERP solutions. Licensing obligations are very visible in the enterprise and the majority of cost associated with open source is frequently connected to purchase support.

The world of the development center is developing and integrating in order to bring solutions online for internal and external clients. These adopters, who are building solutions, see open source as an accelerator; a way to not spend time reinventing the plumbing, a cheap way to experiment, a community solving similar problems, and a rapid application development mindset. Like the commercial vendors referenced by Gartner, these in-house developers are taking advantage of developer-oriented software components that you can find through Google searches and on well-known open source haunts. Ideally, they are paying attention to version and licensing concerns, using permissive licenses such as GNU's Lesser General Public License (LGPL).

What attracts these two groups to open source, and do these two groups, the data center and the development center, value these characteristics in the same way?

"A recent survey of 1,150 enterprise software developers by Marina Del Rey, Calif.-based open source consultancy Simula Labs, revealed that 33% said better control over software code was their main reason for using open source-based development tools. Less than 25% cited cost savings. Another 18% cited the ability to avoid ‘vendor lock-in,' and 19.5% said open source tools offered a ‘lower barrier' to evaluation and use."[2]

It's easy to imagine that the value of "increased control" by virtue of having the source code results in similar perspectives from the two groups. A developer can evaluate the quality of the source code and have a deeper knowledge on how to use it based on his or her purpose. Similarly, a data center can view entire packages, built from source, and has the ability to include what is desired. This eliminates complexity and vulnerability, reduces footprint, and enables other tangible benefits. The control is valued in both perspectives and viewed in a similar way.

While I am not aware of any studies in this area, anecdotally there do seem to be characteristics of open source that are viewed differently. Let's focus on the Low Barrier to Entry.

IT operations want to minimize support, training, and other resources. To achieve this goal, it's a priority to minimize the number of different components that provide the same function, including differing versions of the same packages. In a sense, it is best for the data center operation to limit the number of options - control, limit, and govern.

Good developers, on the other hand, are always looking for opportunities for reuse whether at runtime through SOA, assembly-time through use of components and libraries, or compile time via inclusion of common source code. It is natural to constantly expand the palette, to include more features, better meet schedules, and ultimately help innovate.

While there are certainly commonalities among groups interested in open source, there also can be differences in motivations and perceptions of value as well. Attempting to address two audiences with different concerns can often create artificial challenges to adoption. While they both are leveraging the many benefits open source can bring, they are likely approaching the task from different perspectives.

Viewing open source from the perspective of only one community can create some misperceptions. For example, an organization may conclude that it does not use open source based on the response from the data center, which has not adopted open source in production. This may miss the fact that the development organization, off-shore teams, and even COTS applications have brought open source squarely into delivery. This misunderstanding can be reflected in procurement processes, which are based solely on the commercial model, and do not take into account the developer's ability to "procure" source and components from a vast number of sources not currently incorporated into the process.

When creating a plan for open source adoption, it's important to take into account the perspective of the groups that will participate in its adoption. The explosion of open source communities means most everyone involved in IT has an interest. At least one characteristic of open source is low barrier to entry, which enables a developer to be looking for new solutions and opportunities. However, in the process it poses new challenges to the data center, which is trying to rein in the number of deployed products and versions. Appreciating the breadth of open source and the differing perspectives that might be present in your organization to shape your strategy is based on simple observation and is a corollary to the old adage "know your audience."

References

• Paul, R. Gartner: 80% of commercial apps to use open source by 2012.
• Simula Labs Survey