Getting a Bang for Your Proprietary BuckBy Edward Cone | Posted 2003-11-03 Email Print
Re-Thinking HR: What Every CIO Needs to Know About Tomorrow's Workforce
Can You Save Money Using Proprietary Software instead of an open-source development language?
That's what Scott Metzger, chief technical officer at TrueLink, decided when comparing the costs of proprietary software to those of using an open-source version of the Java programming language for developing new applications.
Metzger looked beyond the price tag of developing new code in proprietary software vs. an openread, "free"programming language. Particularly, he focused on how much staff time each method would require to put applications in place for his San Luis Obispo, Calif., company, which provides online tools for managing consumer credit.
Working from research and his experience with both options, he calculated that TrueLink could put useful tools in place with a proprietary system 25% faster than an open-source approach.
TrueLink chose application-infrastructure software from BEA Systems, moving quickly to implement a beta-test version of the WebLogic Platform 8.1. "[BEA's] technology stack is richer, so the effort needed per line of code is less," Metzger says. "It paid for itself in the work we didn't have to do."
Metzger estimates his company will save nearly $800,000 in staff costs with the BEA software in the first year alone, and TrueLink will recoup the full cost of the software, including license fees and staff time, in roughly nine months.
Most important, TrueLink was able to get its products to market faster, starting with the launch of a new credit-analysis application this summer. That product is in use at banking customer TransUnion, TrueLink's majority owner.
TrueLink also sells services directly to consumers and in partnership with other financial institutions. The company will implement a mature version of the BEA system this fall.
Still, Metzger isn't doctrinaire about the "open source vs. proprietary" debate. Three years ago he switched from Microsoft NT to the Linux operating system on company servers. Again, that decision was based on project cost, speed and personnel.
Not every aspect of TrueLink's software strategy showed up on the calculator that Metzger developed and continues to refine. There were human factors to consider, too, especially when it came to matching staff skills to the appropriate development tasks. That exercise had a long-term payoff of its own, Metzger discovered.
"As our path toward BEA became clear, the real open-source zealots moved into systems and network administration, where they can still work with open source," he says.