Insight

The blind spot in
software protection

Why copyright and trade secrets are often not enough—and where patents fill the gap.

Copyright protects how software is written. Trade secrets protect what remains hidden. Neither protects what is often the software’s most valuable asset: what it does. That is precisely the gap that patents are designed to fill. Many software companies assume their intellectual property is fully protected. The source code is covered by copyright, while sensitive know-how is safeguarded through confidentiality obligations and access restrictions. Yet the core of the software’s value—what the program actually does, rather than how its code is written—often falls between these two forms of protection.

Copyright

Copyright protects the expression of software, not its functionality. The European Software Directive is explicit on this point: what is protected is the form in which a computer program is expressed, not the ideas and principles underlying it, including its interfaces. In SAS Institute v. World Programming, the Court of Justice of the European Union drew the logical conclusion: the functionality of a computer program, its programming language, and the format of data files are not, as such, protected by copyright. A competitor is therefore free to study how software behaves and recreate that behaviour from scratch using independently written code, without infringing copyright.


The Directive even provides the necessary legal tools. Lawful users may observe, study and test a program and, under certain conditions, decompile it to achieve interoperability. These are deliberate rules intended to promote competition.

Trade secrets

Trade secrets cover a different area—and leave a different gap. They protect precisely what copyright does not: the model running on a server, the training pipeline, tuned parameters, internal architecture, and other confidential technical know-how—but only for as long as these elements remain secret. A trade secret exists only while it remains secret and while its holder takes reasonable steps to preserve its confidentiality.


Trade secret protection offers no remedy against someone who independently arrives at the same result. The EU Trade Secrets Directive expressly permits both independent development and reverse engineering of a lawfully acquired product. It also safeguards employee mobility: the experience and skills honestly acquired by employees remain their own. Once software is distributed, exposed through APIs, or analysed through benchmarking, secrecy inevitably begins to erode.


This is the blind spot. The value of software lies in its technical behaviour, yet copyright deliberately refuses to protect that behaviour, while trade secret protection cannot preserve it once the product enters the marketplace. A competitor who independently develops new code implementing the same technical solution infringes neither copyright—because no expression has been copied—nor trade secret law, because the solution was independently discovered.


A patent is the only conventional intellectual property right that protects the technical solution itself, irrespective of how it is expressed in writing or whether it remains secret. Whether the infringer has ever seen your source code is irrelevant. Even if a competitor independently invents the same patented method, that is no defence to patent infringement.

Disclosure

The price of patent protection is disclosure. Anyone seeking a patent must describe the invention in sufficient detail for a person skilled in the art to reproduce it, and that description becomes public. This is the essence of the patent bargain: a temporary exclusive right in exchange for permanent public disclosure.


The strategic question is therefore not merely, “Can we patent this?” but rather, “Are we willing to disclose this invention in exchange for exclusivity, or is it more valuable as a trade secret?”


The fact that software can indeed be patented still surprises some people. The relevant question is not whether something is software, but whether it solves a technical problem by technical means.


Conclusion

The conclusion is therefore not about choosing one form of protection over another, but about adopting a layered strategy. You do not choose a single intellectual property right for “the software.” Instead, you analyse the system and determine where the value resides at each layer: copyright as the foundation, protecting against literal copying; trade secrets for transient, hidden details that are difficult to reconstruct; and patents for stable technical solutions that competitors could otherwise reinvent with impunity.


Patent the architecture; keep the implementation details and configurations secret. Companies that rely solely on copyright and trade secrets protect only the form of their software, leaving its functionality—and often its true commercial value—unprotected.

Share