This article by Hunter Jensen, CEO of Barefoot Solutions, was originally published in the July 2008 edition of PHP Architect, a print magazine for PHP Professionals. It is the fourth article in a five part series on some of the different legal issues surrounding the web development industry.

Modern day patents originated from a specific form of Letters Patent. These were decrees handed down by a monarchy granting exclusive monopoly rights to the recipient for a specific and often broad range of business activities. Patent laws have come a long way since then, but the basic premise is the same. A government provides monopolistic rights to the creator of an invention for a limited time. The justification here is to provide an incentive structure that is conducive to promoting progress and innovation.

But why, as professionals working with PHP, do we need to care about patents? Most small software development shops and startups will run their entire course without ever securing a patent. We should care because most small software development shops will infringe upon a software patent during the ordinary course of business. Patent infringement lawsuits, regardless if you win or lose, are expensive propositions. Further, we should care because they are threatening the software development industry as a whole. They are stifling innovation and creating massive barriers to entry. Major changes need to be made to fix the current system in the U.S. and across the world.

First, a little background on intellectual property laws and patents. The U.S. system will be discussed here because it has one of the broadest covering patent systems in the world, and is the basis for many of the software patent issues that are affecting the industry globally. The United States Constitution states that: “Congress Shall Have Power To […] promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries”. This one sentence contains the entire justification for the U.S. copyright and patent system. Patents are not the right to use your invention, but rather the right to exclude others from making, using, selling, or importing patented goods and services. Inventors must apply for patents which are then reviewed by government officials and awarded or denied based on their merit. In the United States, this is handled by the U.S. Patent and Trademark Office (U.S.P.T.O). They base their judgments on three main criteria. In order to be awarded a patent, the idea presented must be novel (new), useful, and nonobvious.

A patent will not be considered new if:

• The invention was publicly known before it was supposedly “invented” by the patent applicant
• The invention was explained or described in any publication more than one year prior to the date that the application is filed
• The invention was used or offered for sale in a public manner more than one year prior to the filing date.

The actual rules are more detailed than this, but are fairly clear in their application. The Useful requirement for patents is generally used only to throw out frivolous applications and is most often very easily met for software patent applications. The Nonobvious requirement is by far the most difficult to predict and judge properly. The rules state that not only must the invention be different from any prior art (novel), but it must also be a nonobvious extension or improvement on any prior art. This is judged by determining whether a person “of ordinary skill in the art” would consider something obvious. For example, as a person of excruciatingly ordinary skill in the art of computer programming, it seems obvious to me that, in an online shopping system, the less clicks required to checkout the better. Therefore, given prior art of a two-click checkout process, a one-click process seems to be an obvious extension. Unfortunately, as we will see, the judgments of the USPTO do not always follow my intuition.

A common misconception is that you have to actually copy another person’s invention in order to infringe on a patent. While this is true for copyright, it is possible to infringe on a patent while being completely unaware of its existence. Lack of knowledge is considered irrelevant in a patent infringement case. Another area of confusion is where exactly patents are enforceable. While it is true that patents are only enforceable in the country in which they were issued, international trade agreements make it very reasonable to obtain similar patents in many other countries, thereby making them internationally enforceable.

Software patents are only one specific form of patents. Surprisingly, they are available in most countries with a mature set of intellectual property laws; Even those countries that specifically do not allow them. This is achieved by working around the letter of the law and ignoring the spirit entirely. For example, in the United States software patents were not valid in the 1980s because they were regarded as “printed matter”. The federal government ruled that they were simply written instructions on how to perform a task and therefore outside the scope of the patent office. Further, the USPTO has always held that scientific truths or algorithms expressing those truths cannot be patented. No corporation can own e=mc2. In the mid-90s the “Beauregard Claim” was developed and shown to be lawful, and it changed the landscape of software development for the worse. While it was still unlawful to patent a set of written instructions, the Beauregard claim sought patent protection for a computer readable storage device (hard drive, floppy disk, CD-ROM) which contains the specific written instructions. This patent was awarded, creating a precedent that was further litigated, and now nearly all software patents contain some form of the Beauregard claim.

Armed with some general knowledge on software patents, we can now examine where the problems lie. As I said above, the purpose of patents in the United States is to promote progress. Most generally, the problem is that software patents do not achieve their purpose. In fact, the argument can be made that they actually stifle innovation, having the opposite of the intended effect. They are stifling innovation because the USPTO awards patents for very standard software development methodologies. Techniques that any PHP programmer uses in their day-to-day coding can quite often be patented. This makes creating web applications an intellectual property minefield where one misstep could lead to a patent infringement lawsuit.

The most highly publicized case that illustrates this issue is the notorious Amazon one-click patent. The invention they claimed was the concept of storing a user’s personal and payment data so that, in one click, they could complete a checkout. They considered this idea so innovative that they applied for a patent. The scary part is that it was awarded. No prior art could be found, and it was considered nonobvious by a patent reviewer. Later, Amazon sued Barnes and Noble for patent infringement for providing similar functionality on their application. The lawsuit ultimately ended in a settlement requiring Barnes and Noble to remove this feature from their website. It now takes at minimum two clicks of the mouse to checkout on barnesandnoble.com. Anyone can imagine the millions of online stores which could benefit from this feature, but the Amazon patent prevents this from happening.

As developers, an example that might hit closer to home would be the lawsuit filed by Eolas against Microsoft Corporation. The lawsuit claimed that Michael Doyle, the president of Eolas, successfully developed a manner to seamlessly embed content into a Web page. We know these as the <embed>, <object>, and <applet> tags. Because of the victorious lawsuit and Microsoft’s failure on appeal, Microsoft was forced to change the functionality in Internet Explorer 6 surrounding how they activated ActiveX controls. Microsoft released the patch in 2006, and many websites stopped functioning correctly. Microsoft’s solution was to require the user to click any ActiveX control first to activate it, and then they could interact with it. My company, Barefoot Solutions, was managing several projects at the time that were affected. Most notably, embedded video players now had to be clicked twice in order to be played – once to activate, and again to play. This was infuriating. Fortunately for developers and users, someone developed a workaround in which Javascript was used to output the embed tags, and Microsoft never plugged that hole. But as Tim Berners-Lee pointed out in a letter to the Director of the United States Patent and Trademark Office in 2003, the problem was severe considering the millions of websites with historical significance that are no longer being managed. These were affected and never updated. Regardless, we see how much damage one patent can have on the Web.

These are only two examples of software patents with major implications in which ultimately the common internet user is the loser. Netcraft estimates that as of January 2007, 155,583,825 websites are violating at least one software patent. When these magnitudes of people are violating the law, it is natural to suspect that the law itself might be the problem.

Another major issue with software patents is that avoiding liability is impossible. Not just difficult, but literally impossible. First, there are millions of patents. The idea of reviewing all patents to ensure that your new invention does not infringe on one is just not feasible. The tools for patent research are still very ineffective. More importantly, in the United States, patent applications are not publicized. Even if it were feasible to research every patent to ensure that your invention did not infringe upon one, it would still be possible for a patent to finally be awarded – the application review can take several years – in which case your invention would then be infringing on a valid patent. This is a major flaw in the logic of the U.S. system and is unfair to any inventors, including software developers.

Finally, software patents are causing a huge drain on resources that otherwise could be used on innovation. The End Software Patents project estimates that 15% of all new patents are software patents. Their report shows that approximately $11.2 billion is spent annually on software patent litigation. This number refers only to those lawsuits which actually make it to litigation. Legal fees in the many lawsuits or licensing agreements that are reached before ever getting to litigation are not included, so the actual expense is much higher.

One significant cause for this drain on resources is the existence of so-called patent trolls. These firms buy up patents and seek out infringement cases as their primary means of revenue. Bruce Perens, a leader in the Free Software and Open Source community, dubs these firms patent parasites. Yet we cannot place the blame with these firms, but rather the system that incentivizes this behavior and makes it lawful. These parasites force patent holders to spend their resources on legal fees, as opposed to research and development, thus stifling innovation on a large scale.

Many would say however that we need software patents in order to protect inventors and provide incentives for major investments in innovation. There is evidence however that innovation will occur without the existence of software patents. Through the 1980s, software patents did not exist in the United States. Some of the most important concepts that we rely on today were created during this time, showing that we do not need a software patent system to allow for progress. These concepts include: the word processor, spreadsheet, database, email, the World Wide Web, audio-video compression/decompression, the basic elements of the graphical user interface and the fundamentals of all modern operating systems. Why then do we continue to support a system that stifles innovation and is unnecessary? As Bill Gates wrote, “If people had understood how patents would be granted when most of today’s ideas were invented, and had taken out patents, the industry would be at a complete standstill today.”

This paints a dismal picture for the future of the web development industry. So as developers, what can we do to protect our business? Unfortunately, the answer is not very promising. First, many lawyers will advise you to not read up on software patents. This seems counterintuitive, but is often the safest play. The reason is that willful infringement of a patent can involve much larger damages awarded in an infringement lawsuit. Further, U.S. law states that only patent attorneys can judge whether or not your concept infringes on a certain patent. Laymen are unable to properly determine the scope of a claim in any patent. So, if you were to read a patent and decide that your concept does not infringe, the federal court would not agree that you were able to make this judgment, and you could then be sued for willful infringement. Thus, in nearly all cases, the risks greatly outweigh any benefits, and it is best to not educate yourself on the current patents in software development.

Another recommendation is to be sure to place the liability for patent infringement in the hands of the owners of the intellectual property. In an earlier article I suggested that for a freelance PHP programmer, your development agreement should include a ‘Work for Hire’ clause that provides all intellectual property rights to the client. It is also important to include an Indemnity clause stating that the client will pay any legal fees or damages awarded due to a lawsuit for copyright or patent infringement. This will help shield you from some of the liability in the case that you unknowingly infringed upon a patent.

Another method of shielding yourself or your company is to purchase patent infringement insurance. These policies will pay for legal fees and/or any damages that are incurred as a result of an infringement lawsuit. Unfortunately, the premiums for this kind of policy are often very high, and as such this is not very feasible for freelance developers and small businesses.

In reality though, there is not much you can do to fully protect yourself from liability. In medicine, malpractice lawsuits are just part of the risk of being a doctor. In software development, patent infringement lawsuits are currently part of the risk as well. Ultimately what we must do is get involved to fix our broken system. Supporting organizations like End Software Patents (ESP) and the Free Software Foundation that fight software patents is a good start. Our very livelihood is in danger, and it is our responsibility to protect ourselves, our business, and the Web.

Share this article