Shannon asked about what’s involved in challenging a patent. Here’s a really, really brief dissertation on how patents get issued, and how they get busted.
How to get a patent
First, there’s some background information here. Note the bit about the requirements that the invention must be new, useful, and non-obvious. Those standards are assessed (depending on what country’s patent laws apply) as of the date the patent application was first filed, or the date of invention. In other words, the test for whether an invention is “inventive” (a circular description, to be sure) is not considered as of today; it’s considered as of the earliest date that the inventor can legally claim.
That background information also mentions that the patent application undergoes an examination process. In particular, the patent claims — the numbered paragraphs consisting of run-on sentences — are examined to ensure that they cover only new and inventive subject matter, and that whatever they cover, it was sufficiently described in the rest of the patent application. The rest of the patent application can describe much more than the invention, but the claims themselves must read only on something new and inventive, because this is the part of the patent that defines the exclusive rights that are going to be granted. The rest of the patent application must give enough information to allow a worker skilled in the relevant field to make the invention described in the claims. (Some information can be assumed, based on the level of knowledge that the reader is expected to have; not everything has to be explained. For example, if the invention involves a circuit, the patent application doesn’t have to explain what a transistor is.)
So, during the examination process, the patent examiner focuses on the scope of the claims. The examiner will search the prior art (usually published patents and patent applications, hopefully from more than one country; sometimes published articles) that seems relevant to the invention claimed in the patent. The examiner will also make sure that the claims are supported by the description in the patent application.
Often, the examiner will raise objections against the patentability of the claims on the basis of prior art. For example, the examiner can conclude that the claimed invention isn’t new, because he or she found a published document that discloses the very same thing. Or, the examiner can conclude that the claimed invention is obvious (i.e. not “inventive”), because the various parts of the claims are found in a number of prior art documents, and an uninventive person who had knowledge in that field of technology or art could have put them together to arrive at the claimed invention. If the examiner comes to this kind of conclusion, he or she issues a rejection of some or all of the claims. It’s up to the applicant (or the applicant’s patent agent) to either amend the claims to avoid this objection (often the claims are originally drafted too broadly at the outset, so they cover prior art and therefore need to be revised during examination), or to argue against the examiner’s conclusion. If the amendment or argument is successful and there are no further objections, then the patent application will be allowed and will be issued as a patent, with the content of the patent application as amended during examination. Of course, this whole process isn’t free or instantaneous. It can take at least one year, usually two, and often much longer, for a patent to issue from an application. And most steps taken during the process cost money — either paid to the patent agent for doing work, or paid to the patent office for official fees.
Now, in many countries, the burden can be on the examiner to find all the relevant prior art that should be considered. Because every examiner is expected to examine a number of applications while maintaining some decent level of productivity, they can’t track down every single reference that might possibly exist. They’re limited by the resources that are available to them.
In the United States, there is an obligation on a patent applicant to disclose any prior art publications that were known to the applicant, the inventor, or the patent agent that are relevant to the invention. In other countries, the examiner may be able to requisition the applicant to disclose any prior art that was cited in related patent applications in other countries. That helps cut down on the burden of searching. But there are still limitations — folk wisdom or common sense that wasn’t actually recorded in a meaningful way isn’t going to be available to the examiner; and an examiner can’t (or shouldn’t) simply rely on “common sense” without any documentary backup to reject a claim.
How to bust a patent
Short, cynical answer: got money?
In order of decreasing cost and risk, here are the ways that patents usually get invalidated (this is a generalized overview of procedures in more than one country, so not everything described here will literally apply in a single country):
- Infringer gets sued for infringement; infringer challenges validity of patent in defence and also seeks impeachment of patent in a counterclaim.
- An interested party brings an action to impeach the patent (there may be some legal requirement to demonstrate status as an interested party); if party is a potential infringer, patentee may counterclaim for patent infringement.
- A party applies to the patent office for a re-examination of the patent in view of published prior art documents.
The first two options involve court actions (quite expensive). The third option doesn’t involve the court — at least, not right away — and depending on the process before the relevant patent office, is often less expensive, although there are specific procedures to be followed.
In a patent office re-exam, the examination board is usually restricted to considering novelty and obviousness in view of printed prior art publications (the scope of the inquiry can vary from country to country, but note again that folk wisdom or common sense that is undocumented may not be accepted), and the patentee may have the opportunity to submit replacement claims that will avoid the prior art and still cover the invention. The challenger is permitted to make written arguments against the patent. So, while it’s lower cost for the challenger, it’s also lower risk for the patentee, because at the end of the re-examination the patent may still stand, with either the same or narrower claims. If the challenger’s goal was simply to get the claims narrowed so that the challenger could avoid infringing them, that could be considered a success.
And when I say “lower cost”, it’s a relative term: re-examination fees payable to the patent office can range from about $1K to almost $9K. As much could be spent again on patent agent fees, if one is retained by the challenger to do the work. And the patentee or challenger may have rights of appeal if they’re not happy with the decision.
In a court action, while the costs and risks are higher for the challenger, the risks are higher for the patentee, too. Other grounds for invalidity besides prior art can be raised, such as procedural and equitable issues, like a failure to pay appropriate fees, abuse of patent rights, or failure to act in good faith before the patent office; but most patents are invalidated on the grounds of novelty and inventiveness in view of prior art, not on technical grounds (however, a court can consider non-documentary evidence of prior use or knowledge). And the court doesn’t have the power to “fix” the patent so that it is no longer invalid, so if it turns out that there is prior art that is covered by some or all of the claims, those claims will be invalid — no chance for the patentee to fix them.
Things can get more complicated if, during an infringement/impeachment action, the patentee or the infringer applies to the patent office for a re-examination.
In some countries, there’s an opportunity for other parties to write to the patent office before the patent is granted to protest against the application. You can submit prior art and arguments why the prior art invalidates the proposed claims. But the examiner may not be required to respond to your submissions; they’ll get considered, for what they’re worth, and your arguments may or may not wind up being used.
Obviously, if avoiding infringement of someoone else’s patent is of concern to you, it goes without saying that it is strongly recommended that you consult a patent attorney. The actual rules of interpretation of patent claims and the rules for determining whether a seemingly different device or method can be captured by those interpreted claims are extremely complicated. Even if it seems that the language of the claims is comprehensible to you, the meaning that you might happen to attribute to them isn’t necessarily the legal meaning. (To make reference to other impeachment-related proceedings, it really does matter what your meaning of the word “is” is.)
As a general comment, it’s a lot easier for laypeople to dismiss the invention claimed in a patent as “not new” or “obvious” than it is for the people who have to scrutinize the patent and evaluate its validity. Years after the patent application was filed, and the patentee starts enforcing its patent against others, it’s easy for commentators to say “yeah, well, that’s old hat — look at any university textbook.” Well, perhaps it’s old news now — but the body of knowledge by which inventiveness and novelty is measured isn’t the standard you would apply now, it’s the knowledge that a person skilled in the art would have had back then. It’s also easy to dismiss the invention as merely being the application of plain common sense, which any uninventive drone could have done, but proving that is a challenge. In other words, hindsight is 20/20: it’s easier to look back from today’s date and say “well of course, this is how you’d do it!” than it is to place yourself in the position of someone at the patent application filing date or invention date and look forward.