This article explains the Fear, Uncertainty, and Doubt (FUD) factor with patent searching. FUD also applies to aspects of patents other than searching. The Internet has made patent searches easy, but there is always a chance that something will be overlooked. The main reason to search patents is to evaluate the likelihood that an invention is new and unobvious enough to be patentable. However, there is no law or Patent Office regulation that requires you to perform a patent search before filing a patent application. In fact, the Patent Office will always conduct a search once you file a patent application whether you perform a search or not. However, patent search is a cost-benefit decision, because if your searching uncovers evidence demonstrating you probably will not get a patent, you will save the money you would have spent to have a patent application prepared and filed. In the final analysis, business decisions need to be made despite FUD. Today, the data set upon which patent business decisions are based, although still incomplete, is at least a lot easier to build.
When an inventor is thinking about patenting an idea, it's a good thing to know what came before it and what is making its way through the patent process. That knowledge tells the inventor what features of the new idea can be claimed as novel and are therefore patentable.
Never before has the do-it-yourselfer had so much data available to him in the world of patents. Patent searching is now much easier than in the good old days when you had to actually travel to the Patent Office or one of its depository libraries and manually sort through stacks of documents. One thing that hasn’t really changed, though, is the FUD Factor—the fear, uncertainty, and doubt associated with patent searching.
As a patent attorney, I do a lot of searching, usually through the U.S. Patent and Trademark Office's Advanced Search Option at http://patft.uspto.gov, which accesses the database of patents and published patent applications. The information there is mostly current within a few days of publication. For any patent found in a search, you can also review the earlier patents it cites, and by clicking on the “referenced by” icon, you can explore later patents citing the patent you found. This backwards and forwards searching technique often allows you to uncover the earliest patents for a given technology and also the latest and greatest engineering efforts relating to that technology. When I begin to loop back to patents I’ve seen earlier in a given search, I’m usually satisfied my search is fairly complete.
I also regularly search patents using the Patent Office's classification scheme. From the main search page, click on “Searching by Patent Classification” and you can explore the various classifications, and search for patents and published patent applications by classification. For example, I was recently searching for prior patents having to do with the shape of a parachute canopy and, sure enough, parachutes are in class 244, subclass 142, and there is even another subclass called “canopy construction.”
To learn more about a given patent or application, once you’ve found it, you can use the PAIR (Patent Application Information Retrieval) system. This is another often-used site in my “favorites” folder. PAIR is reachable from the USPTO's main search page.
Once in PAIR, “Image Filewrapper” allows you to view the complete history of a given patent. Did the scope of the original patent application change in order for the patent to be granted? Questions like that can be answered here. Is the patent still in force? You can also click on “Fees” to see if the patent owner has paid the necessary maintenance fees in order to keep the patent alive. Are there any related patents or applications? Click on “Continuity Data” for a full report.
I do not, however, print out patents from the Patent Office Web site. It's too much of a hassle. Instead, I use Google Patents where, once the patent number is known, it's a snap to view and print out a PDF version of any patent as issued by the government. On the other hand, I don’t generally search for patents using Google, but that's probably because I’m used to the Patent Office Web site for searching. I do, however, use regular old Google from time to time to search for patents and also to uncover news stories regarding patent infringement lawsuits. For international searching, I often use WIPO (The World Intellectual Property Organization), whose search page is at http://www.wipo.int/pctdb/en/.
The main reason to search patents, of course, is to evaluate the likelihood that an invention is new and unobvious enough to be patentable. Note, however, there is no law or Patent Office regulation that requires you to perform a patent search before (or after) filing a patent application. In fact, the Patent Office will always conduct a search once you file a patent application whether you perform a search or not.
So, why conduct a patent search? It's a costbenefit decision. If your searching uncovers evidence demonstrating you probably will not get a patent, you’ll save the money you would have spent to have a patent application prepared and filed.
In some cases, however, the answer to the question of whether or not to search is “not.” The engineer or scientist who devised the innovation may already know quite a lot about the current state of technology in this particular field of endeavor. For example, she reviews the latest scientific literature and attends all the relevant conferences. The inventor may also hold previous patents in the field and may regularly keep abreast of later patents filed by others.
Such a person may actually know more than the Patent Office because the patents and even the published patent applications that can be searched are behind the times. Once filed, a patent application can take one and a half years to publish and three to five years to issue. Patent filings for the latest and greatest technologies, then, are simply not searchable. So, in these cases, a company that is already aware of what's out there doesn’t need to spend the time on research that may be outdated.
In other cases, a patent search can be worthwhile—for example, when moving into a new technological field or industry, or where company engineers are not aware of the state of the art for a given technology. Searching can also be useful to track competitors in order to see what they are patenting.
Another caveat is that highly relevant patents may not be uncovered in your search efforts. For example, suppose an engineer is looking for patents in the field of liquid crystal displays. Entering the key words “liquid,” “crystal,” and “display” in the abstract search field will uncover over a thousand patents that were issued in just the last three years. The searcher will likely then narrow the search to only those patent abstracts discussing a particular environment for LCD displays—cellular telephones, for example. But a patent can cover liquid crystal displays for use in cellular telephones without the abstract ever mentioning cellular telephones.
In addition, the choice of language by the patent drafter can throw off the search results. Perhaps the abstract of the patent mentions “hand-held communication devices” or “wireless devices” instead of cellular telephones. Relevant patents in the field of LCDs for cellular telephones could exist that don’t even mention “liquid crystal” in the abstract, or for that matter,in any part of the patent. Instead, “pixel electrodes” or other equivalent terms could be used throughout the patent instead of “LCD.”
A patent attorney might, for example, call a common screw a “mechanism for converting rotational motion into linear motion.” You say “fan.” I say “wind generator.” She says “motorized air movement apparatus.” That's one reason I like to search using both key words and the Patent Office's classifications. Again, though, be careful: there is no guarantee the Patent Office always correctly classifies a given patent.
Search options: The Internet has made patent searches easy, but there's always a chance that something will be overlooked.
Finally, even if your data set is fairly complete, at best only an educated guess can be made in attempting to predict patentability. You never know, for example, if the Patent Office will combine two prior patents which, according to an examiner's subjective viewpoint, together render your invention unpatentably obvious. Accordingly, the question, “Can I patent this?” can never be answered with absolute certainty.
A typical question is, “If I make and sell this, will it violate anyone's patent?” My usual answer is “probably.”
An easier-to-answer question is, “Can I patent this, given these three patents I do know about?” But, given the millions of prior U.S. and foreign patents, there is a high likelihood the Patent Office will find additional relevant prior patents rendering the question itself too narrow.
After nearly 20 years in the patent game, however, I can tell you that my attempts at securing patents probably look almost like a bell curve with a small minority of the patent applications on one end never getting through the system, a small minority on the other end sailing through without any objection by the Patent Office, and the vast majority achieving patented status via a give and-take with the Patent Office over the course of a year or two and resulting in patents perhaps not as broad as originally hoped for, but still broad enough in coverage to adequately protect a client's engineering efforts.
Another typical question is, “If I make and sell this, will it violate anyone else's patent?” My usual off-the cuff answer is “probably.” Law firms and several patent search outfits offer so-called “clearance searches,” which attempt to answer the infringement question. The cost of these searches, however, is very high, and they are usually not undertaken unless a given invention is a bet the-company type product. A better question which can lead to less uncertainty is “given our competitors’ X patents (where X is less than about 10), are we clear to make this product?” But, even then, the cost of ascertaining the answer can run as high as $50,000.
Inevitably, at least one such patent is or might be infringed leading to the next logical question, “Is this patent valid?” Since only the Patent Office or a U.S. Federal Court can decide the validity question, however, your attorney's answer matters to a certain extent, but it is not definitive. Trying to predict how the Patent Office, a judge, or worse, a jury, will rule ahead of time is difficult.
Indeed, according to one study, parties attempting to fully eradicate a patent via the Patent Office's reexamination procedure succeeded only about 12 percent of the time. You can assume that most of the 88 percent of the failures thought they had good cases going in.
Another related question I often hear is, “Will we get sued for patent infringement?” Even if you are armed with a clearance search and an attorney's opinion stating there is no infringement, and/or the patents in question are invalid, no one can answer this question. A patent owner whose patent was not found in the clearance search could sue; so too could the owners of the patents you do know about. Right or wrong, they have the right to disagree with you about infringement and validity.
Is there any limit to the uncertainty? As far as I know, only in two scenarios. The first is you will not get a patent if a single prior patent discloses your exact invention. The second is, if that prior patent is expired, you can usually sell the previously patented product without liability—provided you add no bells or whistles covered by other non-expired patents and also provided there are no related unexpired patents covering the same thing or anything else in your product. Generic drug manufacturers, for example, lie in wait for drugs to “come off patent.”
A final common scenario involves the desire to offer a product similar to a new product on the market advertised as “patent pending.” You cannot find the patent application because it has yet to be published. You are in the dark and uncertain. What does the patent application cover? Will it issue? When? This is the FUD principle behind “patent pending.” Note, however, it's a two-way street: your competitors don’t immediately know what your patent pending means either.
FUD also applies to aspects of patents other than searching. I’ve witnessed first hand, for example, the fear expressed by clients who believe no one can enter certain markets (for example, microprocessors) without supposedly infringing numerous patents. There is also a general uncertainty over patent reform including whether or not it will ever really happen, and if it does, whether or not it will work as intended. And, you don’t have to search very hard for expert opinions and op-ed pieces which cast doubt on the U.S. patent system as a whole.
In the final analysis, business decisions need to be made despite FUD. Today, the data set upon which patent business decisions are based, although still incomplete, is at least a lot easier to build.