Open source and IP: striking a balance between sharing and protecting

Open-source software has transformed the way we develop and use technology, facilitating a collaborative approach that allows developers to access, modify, and distribute code, often without any obligation to pay licence fees to an upstream licensor. This collaborative spirit has fuelled innovation and lowered barriers to entry, allowing even small companies and individual developers to compete on a global scale.

However, this open model can be at odds with traditional intellectual property (IP) practices, which aim to secure a monopoly for innovators to provide competitive advantages. Balancing the benefits of collaboration with the need to protect proprietary technology is a complex undertaking that requires a nuanced approach.

Open-source arrangements can vary widely, but typically they are characterised by:

• Access to source code – source code is made widely available to others (Users);

• Freedom to modify and distribute – Users are authorised to use, modify, and distribute the software;

• Defined licensing terms – the ‘authorisation’ referred to above is defined in an open-source licence agreement; and

• Collaborative development – open-source licences typically require Users to distribute any modified versions of the software, often under the same licence terms, which encourages further development.

The appeal of open-source software lies in the collaborative innovation it fosters. Developers from around the world can contribute to a project, improving its functionality, fixing bugs, and enhancing security.

This collective approach often leads to faster development cycles, lower costs, and highly secure software, due to the large number of contributors who improve the code.

Open-source arrangements contrast sharply with traditional software development because the copyright subsisting in proprietary software entitles a developer to prevent others from copying, adapting, or distributing its source code. While the open-source movement emphasises sharing and collaboration, IP law traditionally focuses on exclusivity and protection. IP rights, such as patents and copyright, are designed to give creators control over their development outcomes, entitling them to prevent unauthorised use and monetise their work. This monopolistic strategy provides companies with a competitive edge but often stands at odds with the open-source philosophy.

Not all open-source licences were created equal, so choosing the right licence is key to finding the balance between leveraging the benefits of open-source software and protecting proprietary technology.

Some licences, such as Apache 2.0 and MIT, are relatively permissive, allowing Users to use and modify open-source code with minimal restrictions. Others, such as the GNU General Public License (GPL), have requirements that compel the User to distribute modified code to anyone who receives the software, along with the same rights to use, modify, and distribute it.

Furthermore, the GPL prevents the software from being integrated into proprietary software made available under terms that restrict User freedoms. This ‘openness’ makes it impossible for a User to control or limit downstream access in exchange for ongoing licence fees. Although a User is not necessarily prevented from contractually obliging a downstream licensee to pay a fee for accessing and using code modified under a GPL (despite the fact that this code must be made available), doing so contradicts the values associated with the GPL and could damage the reputation of the User.

Clearly, the GPL is designed to ensure developer freedom and discourage monetisation through restrictive licensing, making it inherently incompatible with models that rely on the payment of licence fees for use. In contrast, the Apache 2.0 and MIT open-source licences enable Users to modify the software and distribute derivative works without sharing the source code. Users can offer modified versions under these types of licences and a separate proprietary version with additional features, support, or warranty, in respect of which fees are levied. Users could charge fees to access premium versions, updates, or enterprise-level features that are not part of the open-source version, or fees in respect of support, maintenance, consulting, or customisation of the source code.

Patenting provides a way for businesses to protect critical technology while still benefiting from collaborative contributions to the development of enabling source code. As a User, it is worth bearing in mind that while an open-source licence may grant a licence under copyright in the source code, permitting a User to copy and modify the code, it may not grant any patent licences, so that the User is nevertheless prevented from running code that implements a patented process in a particular territory.

By way of example, the Apache 2.0 licence includes the explicit grant of a licence to any patents held by the contributors that apply to the software. This means contributors cannot later sue Users for patent infringement based on their use of the software. Conversely, the MIT licence does not include the grant of such a patent licence – so Users are still exposed to potential patent infringement when they use, modify, and distribute the code.

Understanding the differences between open-source licences is crucial for businesses that want to avoid unintentional legal exposure. Companies that use open-source software should establish a compliance programme to track the use of such software in their developments and to ensure that it complies with the applicable licence terms. Regular audits, proper documentation, and legal reviews of open-source licences can help to mitigate the risk of legal issues.

Many companies, including tech giants, actively participate in open-source development. For example, Google released TensorFlow in 2015 under an Apache 2.0 licence and it quickly became a popular framework enabling developers to easily build, train, and deploy machine learning and AI models, even without deep expertise in the field.

The open-source nature of TensorFlow fostered widespread collaboration, with contributions from researchers, developers, and other tech companies. Google actively maintains TensorFlow, continually updating and improving the software based on community feedback and contributions. Collaboration such as this significantly lowers barriers to entry for the latest developments, spurs innovation across industries, and helps to democratise access to technology.

By strategically contributing to open source, companies can influence the direction of development while gaining access to innovations that they may be entitled to incorporate into their own products.

The interplay between IP and open source is growing more significant. Businesses will need to proactively audit and manage legal risks, and balance the collaborative strength of open-source development, while maximising the competitive edge offered by proprietary technologies.