Hello, Mr. Anderson.

At least, that's the greeting I now expect to see whenever I fire up a page on SourceForge.  And before you ask, no, the Wachowski brothers haven't bought the rights to the Web site.  The open source software world is huge--billions of dollars huge--but trying to figure out its breadth makes me think of The Matrix.  Or, at least, a construct of Matrix-like proportions.

Amazingly enough, a company called Black Duck Software has taken on the task of creating a complete and compelling picture of open source software development.  And I'm not just talking about a simple Linux survey or two. Black Duck has used everything from the largest of the open-source operating systems to the smallest of massively-multiplayer frameworks to develop an epic valuation of open-source software.  It's been running these numbers and scanning for projects since the company's founding in 2002, if that helps you to visualize just how deep the rabbit hole gets.

And what have they found?  Enough code, representing enough cash, to create a little Matrix of your very own. According to Black Duck, there are more than 200,000 open source projects being developed or distributed on the Internet today.  That represents more than 4.9 billion individual lines of code, a metric Black Duck developed by running a counting application on every source file, in every project, the company tracked. More than 4,000 individual Web sites were used to generate this large database of projects and related files.

From there, Black Duck used the Constructive Cost Model developed by University of Southern California professor Barry Boehm to determine just how much this heap of software is worth. The model estimates the number of people-hours a particular piece of software would require based on the amount of code (in thousands) that has to be generated.  Different versions of the model take more complex factors into account, including estimated resource constraints, the desired software reliability, and the experience of the coders, amongst other factors.  In layman's terms, the more code a person has to write multiplied by the complexity of the project determines just how much time one can expect to spend.  Go figure, eh?

Running the model based on the 4.9 billion lines of code approximation from earlier, Black Duck determined that the current open source world would take 2.1 million people-years worth of development to replicate.  You're either in for a long haul behind your computer, or you and your 2.099 million closest friends are racing the Shakespeare-typing monkeys to achieve a great dream. Factoring in an average United States programmer salary of around $75,000, as well as a multiplier that represents additional development costs not directly related to salaries, Black Duck generated a figure of $387 billion for just how much these 200,000 or so programs are worth.

The company also estimates that at least 10 percent of all proprietary software development is redundant, given the propensity of alternative (and less expensive) open-source code and applications. That translates out to roughly $22 billion lost in just the United States alone, or approximately $65 billion globally. 

While it's difficult to find an exact replication of these figures by a third-party (Black Duck is in the business of bridging corporate business models with open-source alternatives), it's not that tough to find a wealth of stories that substantiate the basic claim that open source saves money.  Companies care about saving money, especially as global financial issues plague their bottom lines. Open source is not a panacea--it's no Matrix in form or practice, but you could certainly make an argument for size.