Untangling code

The Comingled Code: Open Source and Economic Development. By Josh Lerner and Mark Schankerman. MIT Press; 264 pages; $35 and £25.95. Buy from Amazon.com, Amazon.co.uk

IN INFORMATION technology there seems to be a revolution every other week. At least that is the impression one gets when following media coverage of the sector. Yet sometimes the hype is justified, in particular in the case of open- source software, free programs developed by loosely knit groups of developers. Within just 15 years they have completely changed the landscape of the software industry, turning it from a mostly capitalist economy into a mixed one.

The shift should be of interest—and not just to techies. Software is important stuff; it keeps the world moving. No car, no television set, not even a modern toaster works without some code. Take corporate computer programs away, and the economy comes to a grinding halt. In some cases software has changed how humans behave; spreadsheet programs, for instance, have redefined more than one job.

At least theoretically, open source could also resolve the main dilemma that bedevils innovation policy. On the one hand, most inventors need incentives to keep inventing. On the other, the social value of an invention is maximised if anyone—not just those willing to pay for it—can use it. Open source seems to satisfy both conditions. Developers contribute voluntarily, and share code freely.

This makes it all the more surprising that the writing about open source is rather patchy. To be sure, there are plenty of books about it. But many are banging a drum on one side or the other of the heated debate within the software industry. Others lionise open-source stars, such as Linus Torvalds, who wrote the first version of Linux, a popular operating system. None offers a robust survey and analysis of the phenomenon.

With “The Comingled Code”, Josh Lerner and Mark Schankerman, professors at the Harvard Business School and the London School of Economics respectively, are aiming to fill this gap. They have done a good job—although its academic tone makes it unlikely that the book will fly off the shelves, even in areas with a lot of hackers (who are sure to take offence at the fact that the authors took money for research from Microsoft, long the arch- enemy of the open-source movement— although they assure readers that the funds came with no strings attached).

The authors offer a brief history of their subject. Open source is actually nothing new. Sharing of source code was widespread in the early days of computing, when software was not perceived as having market value and most developers were academics. But companies soon discovered that selling proprietary software can be a very profitable business. To recreate a software commons, a group of politically motivated programmers came up with a special usage licence in the early 1980s to ring-fence “free software”, as it was then called.

This approach, known as “copyleft”, as opposed to copyright, languished in obscurity until the internet made it easier for far-flung groups of developers to collaborate—which led to an explosion of open-source activity. Today SourceForge, an online home for such projects, hosts thousands of them. More importantly, open-source software has become an integral part of information technology. In some markets, for instance, Linux is more successful than Windows. And Android, the fastest growing operating system for smart-phones, is largely open source.

Greatest contributions

Although the book is a good primer on all this, its main contribution consists of two surveys—one of users of software, the other of developers—that are unprecedented in both scale and scope. More than 2,300 companies and nearly 2,000 programmers, spread across 15 countries, both rich and poor, filled out questionnaires. And Messrs Lerner and Schankerman asked a lot of questions, from how much open-source software a firm has implemented to whether governments should mandate the use of such programs.

The findings contradict much conventional wisdom. Open-source developers, for instance, are widely believed mostly to be volunteers who just love writing code. While this may have been true in the early days of computing, the motivation and background of programmers is now much more mixed. Many work for firms that develop both open-source and proprietary programs and combine them in all kinds of business models. Nearly 40% of companies surveyed fall into this category.

The survey also indicates that the two software worlds are much more “comingled” than their respective champions would have it. More than a quarter of companies happily mix and match both sorts, in particular in poorer countries. Messrs Lerner and Schankerman view the environment of software developers and users as a complex ecosystem akin to a rainforest. It would be wrong, they say, to see the two types of software as substitutes for another or as interchangeable.

Yet the finding that open-source advocates will like least is that free programs are not always cheaper. To be sure, the upfront cost of proprietary software is higher (although open-source programs are not always free). But companies that use such programs spend more on such things as learning to use them and making them work with other software. Whether one or the other type of software is a better deal depends on a customer's circumstances. Does a firm have cash to spend? Do its employees have the necessary skills? There “is no such thing as ‘the cost of software',” the authors argue.

Given the complex picture, they dismiss the argument that open source can solve the conundrum of innovation policy as being “too optimistic”. They do not believe that governments should intervene in favour of open-source software, as many have done through subsidies or public procurement. Instead governments should make sure that the two forms of software compete on a level playing field and can comingle efficiently. One way of doing this would be to promote open standards to ensure that proprietary incumbents do not abuse a dominant position.

All this sounds reasonable, but these recommendations also point to the book's weakness. Having dissected open source in detail and told governments at length what not to do, the authors' prescriptions remain rather vague. “There is no right answer,” they say in the final chapter, amusingly called “The Takeaways”.

It would also have been helpful to examine the implications of the findings for technology sharing in other industries. Open source has moved way beyond software—into biology, all forms of digital content (Wikipedia, now ten years old, is the most prominent example) and even hardware. “The Comingled Code” is full of insights, but the literature about this important development in recent economic history is still far from complete.