Open source software is systemically relevant and needs greater social support. In 2011, US software developer and entrepreneur Mark Andreessen argued that “software is eating the world”. A good ten years later, Joseph Jacks, founder of the OSS Capital, a venture capitalist exclusively supporting commercial open source software start-ups, replied that “open source is eating up software faster than software is eating up the world”. The trend towards open source software is unmistakable and irreversible worldwide – even though the value of open source software has long been undeterminable.
Public property as an economic concept
The open source concept is one of the most successful modern examples of the medieval economic form of the “commons”, an alternative to private property: at that time, farmers cultivated meadows and forests that belonged to them collectively. At the end of the 1960s, the US evolutionary biologist Garrett Hardin linked the idea of the commons against the backdrop of overpopulation with the demand to consider and tackle problems not only as individuals but also as a community. The “tragedy of the commons”, or the reason for the failure of the concept, is that individuals always try to capitalise on their own advantages.
However, when the political scientist Elinor Ostrom, who researched the commons in the 1970s, turned her attention to the question of the conditions under which commons can be organised, it became clear that the concept can indeed work. Ostrom was awarded the Nobel Prize in 2009 for her research into user organisations’ successful management of common property.
For companies that use open source software, it makes economic sense to contribute to creating and maintaining open source software.
The study “The Value of Open Source” by Harvard Business School, published in early 2024, is a decisive step towards determining the value of the digital commons, where the medieval commons once failed. In economics, the value (v) created by a good or service is traditionally measured by multiplying the price (p) by the quantity sold (q). In the case of open source software, however, not only is the price usually zero, but the quantity sold is also unknown, as the source code is publicly accessible, and its use does not have to be disclosed. In the past, there have been several studies in which researchers have approximated values for price and quantity. For example, labour costs that would have been required to rewrite existing open source software were assumed for the price. This is a very good way of estimating what it would cost to replace all existing open source software if it were to disappear tomorrow. However, the problem of quantity could not be solved.
Other research has focused on web servers that are publicly available on the Internet and, therefore, easily measurable. Using data from the United States, the resulting estimates showed a value of 2 billion dollars for the OSS Apache web server in 2012 and a combined value of 4.5 billion dollars for Apache and the increasingly popular OSS web server nginx in 2018. However, although web servers are an important part of the open source software ecosystem, they only make up a small part of it.
Value measurement in width and depth
The approach developed at Harvard Business School in recent years now looks at the value of open source software in both width and depth:
- The study follows the literature mentioned above and uses the labour replacement value to estimate the price (p).
- For the quantity (q), the researchers use two data sources, which together form the most complete measurement of the use of open source software to date: the ‘Census II of Free and Open Source Software – Application Libraries’ and the BuiltWith dataset. The two datasets complement each other: the first focuses on open source software integrated into the software sold by a company, and the second on open source software integrated into a company’s website.
This means that the approach takes into account not only the price of creating open source software—the supply side—but also, to an unprecedented extent, the usage or demand side. In addition, the value of the open source software used in companies is explicitly quantified.
The study also shows that users’ active participation in maintaining the open source software they use is crucial for the success and growth of the ecosystem. Companies should realise that contributions to the design of open source software only account for a fraction of the costs they would incur if open source software did not exist.
”Today, open source software forms the basis for cutting-edge technology in various areas such as artificial intelligence, quantum computing, big data and analytics.”
Dr. Manuel Hoffmann
About the author
Manuel Hoffmann is a Postdoctoral Fellow at the Laboratory for Innovation Science at Harvard. His research focuses on social and behavioural aspects around open source software and artificial intelligence under the broader theme of innovation and technology management, with the aim of better understanding strategic aspects for large, medium-sized, and entrepreneurial firms. He further collaborates with the Center for Population Health Science at Stanford University and the University of Heidelberg. He obtained his doctorate in economics from Texas A&M University. In early 2024, Dr. Manuel Hoffmann and his colleagues Frank Nagle and Yanuo Zhou published the Harvard Business School working paper “The Value of Open Source”, which is the basis for this article.
The proportion of open source software compared to proprietary software in the codes of all codebases worldwide is 77 per cent. 96 per cent of all codebases worldwide – and therefore practically all of them – contain at least one open source component. The average number of open source components in an application is 526.
Source: 2024 OSSRA Report, Synopsys
The value of OSS on the supply side is 4.15 billion dollars, i.e. this is the amount companies would have to spend on the theoretical new development of the most commonly used open source software products today. The value of open source on the demand side is 8.8 trillion dollars, i.e. this is the amount that companies using open source software would have to spend if they had to buy or develop the systems themselves. The figure is around 8 billion dollars for the logistics sector (transport and storage). Of the 20 sectors analysed, logistics is in 15th place: the potential for using open source software is, therefore, still high.
Source: „The Value of Open Source Software”, Hoffmann, Nagle, Zhou, 2024
Without open source, companies would have to spend 3.5 times more on software than they currently do. 84 per cent of the additional costs would be incurred for the six most important programming languages alone, i.e. C, Java, JavaScript, Python, Typescript and Go.
Source: „The Value of Open Source Software”, Hoffmann, Nagle, Zhou, 2024
The increase in code contributions to open source projects via the GitHub portal in a country by 1 per cent leads to 0.2 to 0.4 per cent more start-ups of IT companies in general and 0.03 to 0.1 per cent more new open source companies. Above all, start-up activity increased in countries with higher per capita incomes and better-educated populations. The new companies generate services in the double-digit billions every year.
Source: „Open source software and global entrepreneurship”, Wright, Nagle, Greenstein, 2021
This article was published in the fourth edition of the Open Logistics Magazine. You can read the full magazine and register for future editions here.
