Intellectual Property Law & Consumer Class Actions
There are several types of patent searches that may be conducted on behalf of an inventor or client. Although they may be called by different names, the four basic types of patent searches are:
1) Patentability searches (sometimes called Novelty searches),
2) Validity searches (sometimes called Invalidity, or Enforcement Readiness searches),
3) Infringement searches (sometimes called Clearance, Freedom-to-Operate, or Right-To-Use searches), and
4) State of the Art searches (sometimes called Collection searches).
Each search has a different purpose depending on the needs of the inventor or client, and each entails a different patent search strategy.
The Patentability search (Novelty search) is the most common type of patent search. As the name indicates, a Patentability search is carried out on behalf of an inventor in an effort to determine if an invention is patentable.
Novelty searches generally have no date constraints on the prior art. In other words, if you find prior art that reads on the invention, it does not matter whether the prior art is from yesterday or 100 years ago - it is relevant. The reason for this lack of date constraints has to do with patent law: Anything that has already been disclosed to the public, in any manner, at any time, cannot be patented.
The idea behind a validity search is that the Patent Office may have issued the patent (or allowed specific claims contained in the patent) in error. Examiners are pretty good at what they do, but they are not infallible and they are often operating under severe time constraints. They may have missed a relevant piece of prior art, and this resulted in them granting a patent with claims that never should have been allowed. The situation with a Validity search is often that a company has made a product that infringes upon another company's patent, and they are being sued as a result. One way to win an Infringement case is to invalidate the patent in question. If a patent search can locate prior art that reads on the claims of the patent in question, those claims will be struck from the patent. The patent owner then loses his legal right to sue over products that would infringe on those claims. Whereas a written description of the invention forms the basis of a novelty search, validity searches are done on existing patents, and that patent will serve as the disclosure. Note that with a Validity search, the idea is to find prior art that is relevant to the validity of the CLAIMS of the subject patent. Read that again: the CLAIMS. Not the general idea. Not the entire patent, but each claim, one by one. As a result, this is normally a more time-consuming search than a novelty search. A successful Validity search finds references that the Patent Office missed (you might think this is rare - but it is not), and thereby provides evidence that the claims should have been disallowed on the grounds that someone had already come up with the invention before the patent in question was filed. Pay close attention to the word "filed". The application filing date is very important in Validity searches because it determines the priority date (sometimes called a "get behind" date, for reasons that should become apparent) of the patent. It doesn't matter when a patent issued - it matters when it was filed. You must prove, for each claim, that inventions were known which were so similar to the claim in question, before the filing date of the present patent, that the claims should not have been allowed. Or, in layman's terms, you are attempting to prove that the inventor of the patent in question was actually NOT the first person to come up with this invention. Only the first inventor can hold a valid patent. Let's use a fictional example to illustrate what we mean: U.S. Patent 9,999,999, in Claim 1, claims a computer chip that is faster and generates less heat than any other chip on the market. The patent was filed for on 1/1/2002, and granted on 2/1/2003. In searching the prior art, you find two pieces of prior art that contain information which is very similar to claim 1 of the subject patent. Piece "A" is an article from a computer magazine, dated 11/01/2001. Piece "B" is a patent which was filed for on 4/1/2002 and issued on 12/1/2002. Which of these references is relevant? The answer is "A". Piece "A" came before the filing date of the patent in question. Although Piece "B" issued before the patent in question, it was not filed before the patent in question. Only the filing date matters, because that documents who conceived the invention first, not who got it through the approval process the fastest. Note: In general, when conducting a Validity search you do not cite any prior art that has already been cited on the front of the subject patent. This prior art in listed on the patent because it has already been reviewed by USPTO examiners and, since the patent was granted, the examiners obviously felt that those references were not similar enough to the invention at hand to stop the patent from issuing. In some cases the examiners may have been wrong, but proving that is an uphill battle as compared to finding prior art of which the examiners were unaware. Note that while it is the CLAIMS of the subject patent with which we are concerned (in other words, we are searching on the claims), the art cited may come from anywhere (just like in a Novelty search). Relevant art does not have to be from the claims of other patents. Let's now revisit the idea of dependent and independent claims. Remember our fictional example where Claim 1 read "A lighter, more aerodynamic design for a bicycle rim by using an oval cross-sectional shape", and Claim 2 read "The rim of Claim 1 where the cross-sectional piece is formed from aluminum?". If you were doing a validity search and you had to prioritize finding prior art to invalidate one of these claims, which would you choose? Well, if you find prior art that invalidates Claim 1, then there is no oval rim to be made of aluminum! Claim 2, being dependent on Claim 1, is destroyed by destroying Claim 1. For this, and other reasons, independent claims are normally the targets in a Validity search - it is more efficient to try and invalidate independent claims because you automatically get the dependent ones in the process. You will know a claim is independent if it has no reference to another claim in it.
Also called "Right-to-Use Searches", "Clearance Searches", or "Freedom-to-Operate Searches", the point of this type of search is to prove that a proposed product or invention does not violate any active patents (or conversely, to at least be aware that it does, and make the appropriate business decision). Keep these things in mind for an Infringement search: 1. You are normally only searching patents. Non-patent literature is normally not necessary, since the issue is whether or not the client would infringe an in-force patent. 2. You must carefully read the Claims of each patent you find that might be relevant. Remember, it is the Claims, not the descriptions found elsewhere in a patent, which determine what it would take to infringe that patent. 3. Only in-force patents are of concern, so make sure you limit your date range to the last 20 years. As we have already stated, Infringement searches are generally conducted by trying to prove that no in-force patents exist with claims that would cause the client a problem. But, what if you are confident that such an invention or patent already does exist, but it is over 20 years old? A viable alternate in some very specific cases might be to try and prove that the invention is in the public domain, and therefore safe to use. This would be done by showing that the invention had at one time been disclosed, but that the disclosure occurred over 20 years ago.
The term "State-of-the-Art Search" often means different things to different people. Therefore, if you are a professional patent searcher it is crucial to understand what the client is asking for before beginning this type of search (as it is with any search, but perhaps more so here). Generally, a State-of-the-Art Search is designed to quickly allow someone to see what is currently being developed in a given field. During such a search, the searcher picks patents representative of a specific technology, but does not cite every patent having to do with the technology, as there would be a great deal of redundancy. However, sometimes a client may want each and every patent on a given technology. This type of search is more properly called a Collection Search, but the terms are often used interchangeably.