When is an invention obvious?

One of the most complex aspects of patent law is the inventive step or obviousness of an invention. To be patentable, an invention must be novel (i.e. not exist) and be the result of inventive activity. This is the case when the measures taken in the invention were not obvious to the skilled person. But when is this the case?

The claims

To correctly judge the patentability of an invention, the invention must be defined clearly. To this end every patent contains at the end of the description a series of so-called "claims". Claims are enumerations of measures, properties and/or components that are essential to the invention.

There are two types of claims: independent and dependent. The latter case can be recognized by the fact that they contain a reference to another claim (for example: "A method as claimed in claim 1"). Such a dependent claim usually only contains one or two measures. These however must be read in combination with all the measrues of the claim to which it refers.

Patent infringement can only occur when doing what is stated in at least one independent claim. This requires that all measures (also called "limitations") of the claim are used by the allegedly infringing device. In other words, the claim defines the scope of protection of the patent. This means that a patent can only be attacked by demonstrating that the scope according to the claim is too broad.

Attacking the claims

Attacks based on novelty

There are basically two ways to attack a claim: novelty and inventive step. For novelty the requirement holds that all measures from the independent claim are described in a single document. When even a single, possibly trivial measure is missing, the claim is novel anyway. It is also necessary that all measures are described in the same combination in the single document. An electrotechnical reference work will almost certainly show all (electronical) components of an electrotechnical invention, since such an invention almost always involves the use of known components, such as relays, resistors and capacitors. However, the reference work will most likely not contain the specific circuit according to the claim.

Attacks based on inventive step

The second way to attack a claim is on the basis of inventive step. This requires a combination of documents which describes all elements from a claim. The next step is to argue why a skilled person would (not merely could) combine those documents so as to arrive at the claimed invention. In this argumentation it is not permitted to apply hindsight. You must base the reasoning on the situation the day before the date of filing of the patent application and the knowledge a skilled person had on that day.

A combination of two documents is the most common. It is also possible to combine more documents, but it is often more difficult to present a reasonable case why someone would combine all those documents. The most common way to attack a claim based on lack of inventive step is to identify a problem in the first document that is solved by the second document. The argumentation then is that the skilled person would identify that problem, and while searching for a solution would encounter the second document and apply the solution described therein. This then results in the invention.

Apart from documents general technical knowledge can also be used. A handbook (such as Knuth's The Art of Computer Programming) is a useful proof of what constitutes general technical knowledge. If the above combination of document misses a measure, which can be considered to be common technical knowledge, then the invention is still obvious.

The specific way in which the European Patent Office assesses the inventive step is explained in Treatment of non-technical features in the problem-solution approach at the EPO.

An example

Suppose there is an invention on a method of safely transmitting e-mail, in which claim 1 enumerates the following steps:

  1. allowing the user to enter a text,
  2. allowing the user to enter an e-mail address,
  3. formatting the text in accordance with RFC 821/822, and
  4. transmitting the formatted text to the e-mail address, characterized by the following steps:
  5. computing a digital signature over the formatted text, and
  6. transmitting the computed digital signature to the e-mail address.

Let's assume that no document can be found which mentions all the steps. If this were possible, the invention would not be novel.

We do, however, have a document describing steps one through four, and another document concerning placing digital signatures on electronic documents in general. This other document explains how the digital signature can be used to determine the integrity of the document.

To attack this invention as being obvious, I begin by observing that the method according to the first document has the disadvantage that the recipient cannot determine whether the message arrived without modification. In other words, it is not possible to determine the integrity of an e-mail message using the method of the first document. This is a generally recognized problem for e-mail.

While searching for a solution to this problem, a person skilled in the arts would realize that the contents of an e-mail message is an electronic document. The skilled person would therefore search for documents concerning the integrity of electronic documents, and in the search would come across the second document, since this concerns that very topic.

The second document recognizes the problem and provides a solution for the problem. This provides an incentive to the skilled person to apply this solution to the method of the first document. This combination results in the method as claimed in claim 1. Ergo, the measures according to this claim are obvious and hence the claim is not the result of inventive activity.

Making the combination the other way would require an argumentation why a skilled person would go from a general text on digital signatures to a document regarding e-mail. This seems a more difficult approach to me. What problem exists in the method of guaranteeing the integrity of an electronic document that would be solved by sending the document by e-mail?