Devil’s advocate on open source
I just finished reading Mark Taylor’s recently published blog post Open Source: It’s not just cheap!–one of thousands written about the benefits and pitfalls of open source software.
In his post, Mark referenced a Frank Scavo quotation from Chris Kanaracus’s July 2010 article about Open Source ERP. Apparently Frank, an IT consultant, wrote the following comment in an email to Chris:
“A lot of companies are getting fed up with the restrictive licensing, forced march to new versions, and exorbitant fees for ongoing maintenance. For organisations that are willing to take responsibility, open source provides the ultimate in flexibility, low cost, and control of one’s own destiny.”
Mark remarked that Frank’s opinion is not the usual “Open Source is better because it’s cheaper,” but “Open Source is better because you retain control.”
Nearly all justifications for the use of open source are based on notions about cost or control. However, as I pointed out back in 2008, perceived cost-savings and control are often illusions. We’ve thought about the subject quite a lot since then and we are thoroughly convinced that the attraction of open source is rooted in an illusion of control. More about that in a moment.
Perception vs. Reality
Adopters of open source, like rebellious teenagers, decry parental control. They fantasize about being able to do what they want, when they want, however they want to–many without a thorough understanding of the responsibilities and consequences that accompany such freedoms. The desire to feel that one is “in control” can be a powerful motivator.
Free of their parents’ tyranny, young adults experience the costs, complexity and inconvenience of caring for themselves: paying for rent, utilities, food, gas, insurance, taxes, furniture, doing laundry, preparing meals, cleaning the apartment, replenishing toiletries, and earning an income. They discover that life is not nearly as simple or inexpensive as it was while living at home where their parents managed and supported the household, but they enjoy the perceived freedom and control of living on their own.
And so it is with open source adoption. Free of the perceived tyranny of software vendors, and as Frank put it, “the restrictive licensing, forced march to new versions, and exorbitant fees for ongoing maintenance”, companies must shoulder the cost, complexity, and inconvenience of software development and support.
They are free to adopt whatever open source software application/platform they want, and modify it as needed, on whatever timetable they envision. Of course they are then responsible for the programmers, database experts, user interface designers, information architects, technical writers, QA engineers, project managers, and subject matter experts as well as the development schedule, bug fixes, integrations, porting customizations to future major product releases, and indefinite support of their branch in the code base.
In a nutshell, software development and support dependencies, responsibilities and risk do not disappear following the adoption of open source. They simply shift (in whole or in part) from software vendors to internal development teams, the open source community, or third party providers such as Taylor’s IT consultancy. The perceived illusion of greater freedom and control collides with the cold hard reality of never-ending software development and “lock-in” - there is no free lunch.
Play Devil’s Advocate
There’s a huge difference between using open source software applications as-is without extensive modifications and using them to reinvent commercially available options.
If you’re planning to use open source applications as-is for web hosting, coding, writing, 3D modeling, image editing, etc with only minor configuration changes then there’s virtually no special investment in development or support. This is the ideal use of open source software.
If, however, you’re running a large company with fairly complicated workflows and infrastructure, and you’re in the market for a relatively complex application, then consider your options carefully. Open source may not be the wisest choice. Before moving ahead, ask yourself:
- Why am I interested in open source?
- Perceived cost savings?
- Have I thought about the resource impact of programmers, database experts, user interface designers, information architects, technical writers, QA engineers, project managers, and subject matter experts?
- How about the expense and challenges of product development, bug fixes, integrations, porting customizations to future major product releases, and indefinite support of the application?
- Do I imagine resource sharing across multiple business units, agencies, companies or communities? It can be incredibly effective on the rare occasion that it works well. However, politics, costs and inefficiencies rise alongside project size and complexity. Ask the U.S. Departments of Homeland Security, Defense, and Justice how well large-scale resource sharing works. The answer is “very inefficiently”.
- Perceived freedom and control over development and support?
- Whose freedom? From whom? Locking my company in to my development team, a third party consultancy or development community can be every bit as risky and costly as locking it in to an application vendor.
- Perceived cost savings?
- Would the open source application require extensive modifications? Is software development core to my business? If it isn’t, why am I entertaining the idea of adding software development into the mix?
- Was I approached by members of my IT/development team or a third party service provider who sold me on the use of open source? What are their justifications and motivations?
- Do they believe they can build a “better” product, within my budget, than what is commercially available?
- Do they believe they have the resources and expertise to develop and release modifications more quickly?
- Are the resources in-house and up-to-speed today or will they have to hire and train them?
- Do they have the support staff to document changes, deploy them, test them, and train users?
- What is their plan for managing development team change and succession?
- Have they ever undertaken a project of this scope?
- Is the open source project they wish to use free of intellectual property that may expose my company to future legal issues?
- Are my requirements so specialized that I feel open source is my only choice?
- Have I spoken with the vendor? Is there no interest to include my requirements in a future release? Can I wait?
- Am I trying to make an application work with poor business processes/practices or obsolete systems rather than fixing the root of the problem?
- Are the processes or systems that I am reluctant to change a competitive advantage or an albatross?
- Can I not refine, simplify or re-engineer our processes/infrastructure more affordably than committing to the ongoing expense of custom application development and support? “Simple” feature requests are often solved with inexpensive widgets and process workarounds that allow companies to configure commercial software without custom development. Extensive features requests should raise a red flag that my processes and/or infrastructure may be overcomplicated. Custom development should be viewed as a last resort once all other sources of non-value-added friction have been addressed.
- Was I approached by members of my IT/development team or a third party service provider who sold me on the use of open source? What are their justifications and motivations?
This is by no means intended to be a comprehensive list of considerations for open source software adoption, but it is a list of often overlooked questions every business owner should ask to determine if open source software makes sense.
For more information:
If you have questions about this article I invite you to contact us and arrange a FREE teleconference to discuss your concerns. We encourage you to invite members of your development team and IT staff, and representatives from your application vendors to join the conversation. Follow us on Twitter.
