In an economic slowdown that's getting worse, open source matters, now more than ever. Only it's not in the way you think. Downloading "free" software has never been all that great a path to budget savings for most companies, given all the hidden costs that came with it.

What open source has long promised, and is only starting to deliver consistently to business, is an alternative both cheaper and more effective than proprietary code. Open source code still can't touch the scope of proprietary suites, but it's closing the technology gap. And being lightweight can be its own advantage, rather than throwing ever-more software and servers at a problem. With its transparent and standards-based development, open source code can cut the complexity and risk of custom coding for integration or niche needs.

And those hidden costs? The time it takes to vet a new piece of open source code, to nurture in-house experts who can test and integrate it, to negotiate solid technical support? It's getting easier for companies to get through all that, as open source code is no longer relegated to the fringe of the data center, running a few stray Web applications. Over just the last two years, open source has gone from illegal alien to full-blooded IT citizen. Think of the barriers to open source use that have faded in just the past year or two.

In intellectual property, all the noise about the illegitimacy of open source, how Microsoft found its code in Linux and other open source projects, has withered. Microsoft partnered with Novell, and skilled programmers within Microsoft's own ranks pointed out how much open source code is developed on Windows, and how much open source code interoperates with Windows. And everyone saw how, if Windows Server 2008 is to have a place in the data center, it will have to work with open source code. Proprietary software companies from Citrix to IBM to Sun proved their faith in community-built code by buying open source vendors.

In the courts, open source licenses got a boost this year when the U.S. Court of Appeals overturned a District Court ruling that had cast doubt on how enforceable open source licenses are. In Jacobsen v. Katzer, the court showed an understanding of open source code as a new form of intellectual property, one deserving of protection through the force of copyright.

In security, open source has been dogged by the notion that it didn't pay attention to exposures and vulnerabilities. Through a Homeland Security grant, Coverity, which sells software for spotting security flaws in code, tracked 55 million lines of code in 250 popular open source offerings over three years for security exposures. Apache, Linux, Samba, and many other of the best-known projects showed their defect counts declining rapidly and security standing improving. The Coverity counts finally let open source quantify its security compared with commercial code.

These developments make open source code a safer choice at a time when IT managers are looking for safe harbors.

Perhaps most important of all, it's not just the old standbys of open source driving this change. New open source products are competing directly with commercial code for such deep infrastructure roles as message routing and management (MuleSource's Mule and Apache's ServiceMix enterprise service buses), data integration (Jitterbit), application integration (Talend), and middleware (JBoss, Geronimo, GlassFish, SpringSource dmServer). There are even new models emerging, such as "aggregators" that combine the code from more than one open source project--such as content management, portal, and business intelligence--to create a new product.

As these risks and barriers fade, instead of being on the margin of the data center, open source is assuming tasks close to the heart of production systems. To see how open source is weaving its way into business infrastructure, look at Continental Airlines.