Patent information for free software developers

Posted 10 Nov 1999 at 08:00 UTC by advogato Share This

Steven Young recently wrote an article in Slashdot on the basics of patents, with a special focus for developers of free software. He has graciously granted permission for the story to be reprinted here. A fine first article for Advogato's archive of useful information and resources.

By Steven Young, reprinted by permission.

As an intellectual property attorney, and a regular Slashdot reader, I would like to share a few of my thoughts regarding patent issues, especially as they relate to programmers. Although patents (for better or worse) are playing an ever-expanding role in the software field, many programmers either do not know much about them. I am a former patent attorney. During my relatively short career as a patent attorney I secured patents (including software patents, in the U.S. and abroad) for many well-known technlogy companies. My personal views regarding patents evolved to a point where I could no longer prosecute patents in good conscience, so I moved into other areas of intellectual property law. I will try to present some facts about patent law without getting on a soapbox about my own views (too much). My remarks here are intended to be of general use, but (of course) they should not be taken as legal advice -- if you have questions about any particular patent issues, you should see your friendly neighborhood patent attorney.

People sometimes get confused about whether or not independent invention (i.e. inventing something without reference to the work of an earlier inventor) protects them from patent liability. This confusion is likely due to their familiarity with copyright law. Under copyright law, you only have liability to a copyright owner if you actually copy their copyrighted work. If you create a work of your own without reference to their copyrighted work, you are not liable to them for copyright infringement, even if your work closely resembles theirs.

Liability for patent infringement, however, does not depend on your having copied the work of another. You can be held liable for patent infringement if you have made, used, sold or imported, without a license, something that is claimed by a valid patent owned by another. Your lack of knowledge about the work (or patents) of another is irrelevant to the question of whether you are liable for infringing that person's patent.

Because you cannot realistically compare your own creation against the millions of existing patents, or even against the thousands that might be in the same field, there is no way to completely shield yourself from liability for patent infringement. Even if you could compare your work to those existing patents, there would be ever present danger from those patents that might issue next Tuesday. I see this potential "gotcha" as one of the most problematic aspects of current patent law.

The legal fiction that supposedly justifies this result is that a patentee (i.e. patent owner) is given this very strong right in exchange for disclosing an invention to the public. Upon the publication of a patent, everyone is constructively (i.e. by legal fiction) put on notice that they are no longer free to do the particular things claimed by the patent. As I mentioned, however, there are millions of patents currently in effect -- no one is really aware of the scope of coverage of any sizeable portion of them. Although there might be some individuals who are aware of the general state of patent coverage in particular niches, even they would be exposed to potential liability for those patent claims they are not aware of.

To compound the silliness of all this, the courts have held that no one but a patent attorney can really know what the scope of coverage of a patent claim is. Therefore, even the theoretical justification for absolute liability in patent law is not applicable to individuals who cannot afford to keep a small army of patent attorneys on retainer.

Software patents (of one kind or another) are valid in most countries of the world (even those that officially do not grant software patents).

In the U.S. software is clearly patentable. What many people are unaware of, however, is that, for practical purposes, software has been patentable in the U.S., and most other countries of the world, for quite some time. The current debate about software patents in certain non-U.S. jurisdictions is a matter of form over substance.

While you may not be able to patent "a computer program that performs the steps of X, Y and Z" in some countries, you can generally claim something like,

a computer apparatus consisting of a processor and memory, wherein the memory contains instructions which, when executed by the processor, cause the processor to perform the steps of X, Y and Z
Because this type of claim is directed to a computer programmed in a particular way, and not just a program itself, most countries will allow that claim. Strictly speaking, it is not a "software claim", but the effect is nearly the same. Anyone wanting to use the program described by the steps of "X, Y and Z" will necessarily infringe the claim, so this is one effective method of getting coverage for software. There are other forms of such claims, such as a claim to "a computer readable medium storing computer instructions which, when executed by a processor cause the processor to execute the steps of X,Y and Z", or even a simple method claim such as "a method comprising the steps of...". The computer readable medium claim would cover a diskette or CD-ROM containing the "X, Y and Z" program. The simple method claim broadly covers the steps of doing X, Y and Z, regardless of whether a computer is used or not. Some of these claim forms may not allow a programmer to be sued for direct infringement (for example, if the programmer is not distributing a pre-programmed computer, the programmer is not directly infringing the "computer apparatus" claim), but a programmer could be sued for contributory infringement on any of these claims. Legal form aside, the end result to a programmer is the same: using or distributing your programs puts you at legal risk.

The scope of a patent is defined by its "claims" -- patents generally cover much more than one specific product.

A lot of people think of patents as covering particular products -- a new speaker might be marked with one or more patent numbers, suggesting that there is a patent on that exact speaker. Actually, things are more complicated than that. Each of the patents indicated on the speaker includes one or more (usually more) "claims." Each claim specifies one or more characteristics which must be present in any device for that device to infringe that claim.

For example, there might be a claim specifying that the speaker cone angle is between 40 and 45 degrees, and that the speaker magnet is made of an alloy of iron and nickel, in which nickel makes up at least 5% of the alloy by weight (clearly I am just making this up, and I don't know anything about speaker cones, magnets or metallurgy). In order for the claim to be allowed by the Patent Office, a patent examiner must be unable to find an example of a speaker with a cone of the claimed shape and with a magnet of the claimed type. To infringe that claim, a speaker must have both these claim elements. Any speaker with both elements is an infringing device, and any speaker without both elements is (probably) not an infringing device.

Other claims in the patent may claim other combinations of elements, and each claim essentially stands alone. If you infringe one claim of the patent (i.e. you make, use, sell or import something that has all of the elements of the claim), you are liable for infringement, even though you are not infringing the other claims. It is as though each claim is an independent patent. To be free from infringement, you must clear every claim of every patent.

Willful infringement of a patent exposes you to major damages.

Ordinarily, when someone is found liable for patent infringement, they are prohibited from continuing the infringing activity, and they are ordered to pay the patent holder damages equal to a reasonable royalty for the use of the patent, or the patentee's lost profits. The law permits judges to increase the monetary damages by up to three times, however, if there is a finding of willful infringement, meaning that the infringer had knowledge of the patent before engaging in the actions which constitute infringement.

If someone brings a patent to your attention, and you decide that you are safe because it does not cover what you are doing, you are entering into a legally shaky area. The Court of Appeals for the Federal Circuit (effectively the final word on patent law, since the Supreme Court rarely takes patent cases) has ruled that anyone who is not a patent attorney is not qualified to determine the scope of the claims in a patent, and that it would be unreasonable for you to determine that a particular patent is not applicable to what you are doing unless you first get a legal opinion from a patent attorney. Because, as a matter of law, you couldn't really have believed that you understood the patent (yes, our federal courts can be quite condescending), you will likely be found liable for triple damages if it turns out that you were wrong, and that you really are infringing the patent.

Because of this, lawyers routinely advise their clients to avoid reading patents in areas they are working in. The danger posed by the willful infringement doctrine is seen as outweighing any benefit that can be gained from reading patents. This state of affairs, of course, negates even the theoretical benefit of the patent system, that the public at large learns about new technology once it is patented. As it stands, the people who should be learning from patents in any given field are the same people who stand to lose the most if they dare to take a peek.

The standard of invention for patents is much thinner than most people believe.

When a new patent is announced, one of the most common criticisms is that the patent is invalid because the patented invention is merely an obvious extension of something that is already done. Theoretically, this is a valid criticism. Two requirements for a valid patent are that it is novel (the inventor was the first (sort of) to invent that particular thing or method), and non-obviousness (that the invention is not an obvious extension of something that is already known).

In practice, the level of inventiveness required for patentability is vanishingly small. It is relatively easy to show that a patent claim is invalid for a lack of novelty: you simply find something in the prior art (prior art is typically something that was published more than a year before the patent was applied for, although there are many exceptions) that includes all of the elements of the claim. Showing that a patent claim is invalid because of obviousness is considerably more difficult. First, you have to find examples in the prior art that, when taken together, add up to the patented invention. That is not enough, however. You also have to find something in the prior art that suggests putting these prior art pieces together. That is often difficult to find, even where a modification does seem obvious.

During the examination of patent applications by the Patent Office, many claims slip through that are clearly obvious. This can happen for a number of reasons. One is that the patent examiner has not found any prior art that can be combined to give all of the elements of the claimed invention. Another is that the patent examiner has not found any suggestion to combine prior art that has been found. Still another reason is psychological: a rejection on obviousness grounds is rarely clear-cut, and some patent examiners are uncomfortable making such an inherently subjective call. So, when a claim to an obvious invention makes it through the Patent Office, what happens? Generally, the validity of the patent is only challenged by a defendant when the patent owner sues for infringement (assuming the defendant has enough money for a defense). In court, however, much deference is given to the judgment of the patent examiner who originally allowed the claims, and invalidating an issued patent is a very high hurdle. Unless the obviousness is extraordinarily clear (and it rarely is), the patent will not be invalidated on grounds of obviousness, and the patentee will be able to stop others from using the claimed invention.

The consequence of this is that very little inventiveness is necessary in order for a patent to be valid. As a rule of thumb, it is probably safe to assume that every trifling modification, no matter how small, will be patentable by someone -- as long as that someone gets to the patent office in time. In practice, there is a strong possibility that any given patent will be ruled valid, unless you can find something in the prior art that is exactly what is claimed in the patent.

Patents are national in scope, however...

A patent offers exclusive rights to the owner of the patent only within the country that granted the patent. International treaties, however, have made it relatively easy to get almost identical patents issued in many countries. While each country independently examines each application, patents that are allowed in one country generally get allowed in other countries as well. The existence of a U.S. patent suggests the possible existence of corresponding non-U.S. patents (and vice versa).

The U.S. is one of the only major countries that does not publish patent applications until they are issued as patents. In most other nations, patent applications are published 18 months after filing. This publication gives the public some notice of what is coming down the pike. Although the U.S. does not publish, many U.S. companies routinely file foreign counterparts to their U.S. patent applications, typically using the system put in place by the Patent Cooperation Treaty (PCT). The PCT provides for publication 18 months after the patent was first filed, even if that filing was in the U.S. As a result, even though the U.S. does not publish patents that are being examined, searching through international patent publications can tip you off as to what is currently being examined in the U.S.

Determining whether a patent is valid and enforceable is a non-trivial exercise.

Determining whether a patent is valid and can be enforced is very complicated. Some of the reasons a patent might not be valid or enforceable include:

(i) someone else invented the same thing first;

(ii) the invention was described in a publication more than a year before the patent application was filed;

(iii) the invention was offered for sale (by anyone) more than a year before the patent application was filed;

(iv) less than all of the actual inventors were named in the patent application;

(v) the inventor misled the patent examiner during examination of the application;

(vi) the inventor failed to disclose material prior art to the patent examiner during examination; and

(vii) the patent owner has misused the patent (the "misuse" doctrine is similar to antitrust).

This list is not nearly exhaustive, and each point on it is qualified by many exceptions. I provide this merely as a suggestion for ways to go about trying to get a patent declared invalid or unenforceable. If you intend to invalidate a patent, you will need to consult a patent attorney.

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!

Share this page