It’s a debate I hear often these days on a variety of enterprise implementation projects. Should the company invest in a commercial application or go the “open-source” route. I’m sure you’ve heard this talk before. A hundred thousand programmers are intent on giving away increasingly impressive software for free and making it available on the open market. The goal of this exercise seems focused around disturbing the future residuals of Bill Gates and people like him.

I took the opportunity recently to stop by sites like SourceForge to look at the various tools there, and I was quite impressed. There are tools in the ERP domain, collaboration domain and, my favorite, the project management domain. There are some major implementers now who are making a big play for this kind of software. “Go with our plan,” they say, “and don’t worry about paying another dime for software licenses. Don’t worry about software at all. Our team of implementers will custom-deliver a solution designed to fit your requirements exactly.”

It’s a compelling argument and seems attractive to companies facing a large license fee to purchase or upgrade to the next version of a commercial package. But, there’s a double-edged sword here that is all too familiar to those who have been in the industry for awhile.

There’s no doubt that free software has tremendous inherent value. Numerous programmers work on contributing to the project and many of these packages do work and do produce results that they intend. Also, in the last couple of years, if you were unfortunate enough to use systems of companies that didn’t survive the IT downturn, you may have thought it would have been better to own the product yourself.

However, when you make a decision to go with one of these systems, you are adopting responsibility for several aspects of the project yourself. First, you must be responsible for testing and quality assurance. Yes, I know, thousands of people test the product. Yet they’re each testing for their own needs. If you’re going to put something as critical as your corporate project data into such a system, you’ve got to test yourself. Next, you’re inherently adopting the entire future development path yourself. A commercial product moves forward based on the huge incentives of the market. It will adapt to new software environments and new business conditions over the course of time. This may happen with open source products – it may not.

The end result may be that you’re trading license cost for an ongoing cost of developers, testers and designers. You should make sure you really like the implementers who are proposing an open source solution as they may be around for quite some time. In the end, some organizations will find themselves spending a significant amount of time working on the design, development and completion of software which has become part of their infrastructure. That really only makes sense when your particular needs aren’t well served by software available in the market today.

It’s not new news. That’s pretty much the state the industry was in at the end of the 70’s. Custom software was the rage then. One of the reasons commercial software took off in the 80’s was that designing and developing custom software isn’t the core business for most companies. It’s still