Why software patent eligibility is not an issue at the EPO — and what the USPTO can learn from this

How the USPTO should deal with patent-eligibility

Talking about software patent eligibility, allow me a quick rant about the flaws of the USPTO‘s infamous “abstract idea” test:

Honestly, I believe the Mayo/Alice two-step test causes much more harm than good:

  1. What kinds of ideas are too “abstract” for a patent? Doesn’t every invention start with an idea? Who can tell the difference between an abstract idea and a “concrete” idea (I wonder what that is)?
  2. When is an invention “significantly more” than the abstract idea? How much “more” does it take?
  3. Some court rulings suggest that the Mayo/Alice test requires applying more than “well-known” concepts. Is this a hidden novelty test? Isn’t a hammer patent-eligible simply because it’s a man-made technical device, irrespective of it being known for decades?

From what I hear from my US friends, I’m not the only one left confused.

But enough ranting, let’s talk about solutions:

A more pragmatic approach to patent-eligibility

What if the USPTO just copied the EPO’s “any hardware” test when it comes to patent-eligibility? (If you skipped the video above, now is a good time to watch it for an explanation of the “any hardware” test)

This way, any invention which uses technical means (e.g., a computer) is automatically patent-eligible — and we can finally stop the endless discussions about what it takes to have “significantly more” than an “abstract idea”.

software patent eligibility

In case you wonder whether this would open the floodgates for all sorts of bogus software patents, the EPO also has a solution for that:

Let’s focus on the real question: obviousness

Although any patent claim can pass the eligibility hurdle when properly drafted, at the EPO the real test is that of inventive step (aka obviousness).

This is the “technical contribution” test (also known as Comvik approach) which my colleague Stefan Steinbrener introduced back when he was board of appeal chairman at the EPO.

Need a brief explanation? Watch this:

So in the end, what it all comes down to is that a (software) invention must solve a technical problem compared to the prior art—and that the invention’s technical contribution must be non-obvious.

So where does that all take us?

Having navigated countless software patent applications through the EPO system I firmly believe that it provides a reasonable, fair and—most importantly—predictable framework for separating the wheat from the chaff.

Put another way, making the “non-obvious technical contribution” the decisive factor helps to acknowledge real technological software innovations while sorting out trivial ideas that are simply done “with a computer” or “on the Internet”.

To keep a long story short, here’s my two cents:

Dear USPTO, please just copy the EPO’s “any hardware” test when it comes to patent-eligibility, and let’s all focus on the real question—obviousness—, e.g., using the “technical contribution” test.

What do you think? Let me know in the comments!

Are you up to date?

P.S.: This was just the tip of the iceberg. If you want more, get on my email list. I usually write an update every two weeks or so – it features inside knowledge and practical tips on European software patents. You can subscribe here.

Hi, I'm a European patent attorney with a computer science background and a partner at one of the largest IP firms in Europe. About | Contact