In later posts, I will discuss how the different countries deal with software. In this post, I will start having a look at some basic rules for preparing claims for software patent specifications for Australia and the United States. This is a blog, so I'm going to choose random topics re drafting software patent specifications as I please. Eventually, there should be enough tagged together to achieve coherence.
Also, please remember that this is just a distillation of my experience. If you decide to follow this advice, I accept no responsibility if your application is rejected by the USPTO, or your multi-million dollar licence deal falls through.
There are a number of aspects of a software invention that you should look at protecting in the claims.
Some of these are:
- A method that is carried out by a computational device that executes the software.
- A system that uses the software product.
- A software product per se.
- A medium on which the software product is stored.
- A computational device when programmed with the software product.
Why? Particularly in the US, it's useful to have as many independent claims as possible. That's because they have an illogical way of interpreting claims. It's called the "prosecution history estoppel" doctrine. In short, a US Court can take into account the actions of a patent applicant during prosecution to limit the scope of the claims. So, if a patentee decides to amend a claim during the application procedure, the court can estop the patentee from asserting any rights based on that claim that are broader than a literal interpretation of that claim.
If there are a number of independent claims, it may be possible to avoid having to amend one of more of them. Since the independent claims are the broadest, the patentee may be saved from having to give up valuable patent real estate.
Of course, another reason is that it allows the patentee more grounds to sue for infringement.
I will continue on this in a further post.