Patently Ridiculous

It is my considered opinion that the time has come for a serious revision of the patents system.

Patents are often confused in the minds of the general population with copyrights and trademarks, so before I begin, let me clarify what is meant by a patent:

From thefreedictionary.com:

A grant made by a government that confers upon the creator of an invention the sole right to make, use, and sell that invention for a set period of time.

Unlike copyright, which is inherently granted for anything you create, a patent must be applied for from the government, and is subject to several conditions which determine whether it will be granted, summarised below:

• A patent is by definition an invention - you have to have invented something to get a patent for it. You can't patent a painting, (though you could patent a method of painting), and you can't patent a discovery, such as a dinosaur skeleton, because you didn't invent it, you just happened to find one. You also can't patent the theory of evolution that explains its existence because evolution already existed, you just discovered it.
• A patent must be original - you can't patent the wheel, even though no patent already exists for it, because you didn't invent it. The legal term for this is prior art - the artistry of it existed prior to your claim.
• A patented concept must also be non-obvious. This is a tricky thing to prove, since most ideas seem obvious once you've heard about them, but essentially this means "if any expert in a given field was given the same problem to solve would they have been likely to solve it the same way?" If the answer is yes, then the idea does not warrant a patent.

The purpose of patents is right there in the dictionary definition; They exist to allow a creator of an invention to profit from his idea for a reasonable period so as to justify the time, effort and expenditure put into the research. Patents have a twofold benefit - they encourage innovation by ensuring that inventors have the opportunity to be rewarded, and they allow inventors to share their ideas early on, rather than being forced to keep them a secret for as long as possible to avoid others from stealing and capitalising on them.

And up until about twenty years ago, this was exactly how patents were used.

But then something changed. The pace of technology cranked up a notch. Computers in particular made new forms of research possible, and the technological innovation increased. Most importantly a new kind of technology emerged that had not existed before, that of software applications.

Software is fundamentally different from hardware in that it is very easy to duplicate. Initially this was mostly a concern for copyright holders, since their work can now be effortlessly duplicated and redistributed. Clinging to the glories of yesteryear, they have fought back against this wave of change by concentrating all their efforts on innovative new ways to prevent the copying of digital data (using technologies such as DRM), rather than seeking innovative new ways to monetise their intellectual property.

Digital data's primary benefit is that it is easy to duplicate and distribute. This is why it has triumphed over the analogue technologies that preceded it. People like DVDs because they can be played again and again without degradation, and because they are more faithful to the original film in terms of quality than VHS. Both of these traits come from the fact that digital data is more robust when copied.

For the media companies to have jumped aboard the bandwagon of digital media, and to now suddenly declare that they want to find a way to make it harder to duplicate, is one of the more ludicrous things to have happened in recent times. They've even expended huge sums on developing DVDs that degrade quickly! Who wants to bet that the 3rd generation DVD will be the VHS cassette?

A few months ago, I suggested that copyright "theft" is a phantom menace, and that rather than trying to stave off the inevitable march of progress, copyright owners need to adjust their perceptions and start focussing on how to make money out of giving people what they want, instead of trying to charge them more for something they don't (when the same product is already freely available with fewer restrictions via the black market).

But where copyright is failing in the world of digital media, patents have flourished into something much more powerful (and sinister) than was ever envisaged at their inception.

Patents were created in a time when the following things were true:

• Innovation was infrequent. New ideas came along at a slow enough rate that it was possible for the patent office to keep track of all claims.
• Most innovation came out of the UK (and later the US, which had a similar patent system, and good communication).
• Most patent holders would develop and market their own inventions, seeking financial backing where necessary, but usually retaining the development rights for themselves, and acting on them. They generally didn't lock them in a box and wait to see who came up with the same idea next so they could sue them (or sell them to a company which intended the same).

Under these conditions the patents system worked very well. Innovations were put into practice quickly, and if they failed then the patent would expire and someone else could have a shot at capitalising on the idea. It was also relatively easy for an inventor to check that his idea was unique, or for the patent office to do so on his behalf. Once a patent was granted, its ownership was indisputable.

But modern patents work very differently indeed. The idea of a lone genius producing patentable ideas is an archaic fantasy. It seems probable that most technologies that could reasonably be conceived from scratch by an individual have already been discovered and implemented. Almost all modern innovations, particularly in the field of software are evolutionary (rather than revolutionary), incremental improvements of existing ideas, many of which already have patents protecting all or part of the concept. Truely original, non-derivative ideas are much less common, and are usually the product of large research groups or organisations.

Even if the inventor is an individual, any patents for ideas they produce belong to the company he or she works for, or the organisation funding their research. Depending on the contracts involved, this may even apply to works they produce in their spare time - which can lead to situations where the inventor is unaware that a patent exists for their invention, and cannot be certain who owns the rights to it.

The next problem is that the pace of these innovations has completely swamped the various patent offices' ability to cope with them. There is no reliable, global, public database of patented inventions, nor any search engine capable of reliably determining if an idea has already been patented. This in itself is a disaster as it leads to patent conflicts or overlaps, but it is made far worse by the fact that many innovations are not patented at all by their original creator, and those are even harder to track down. Proving prior art is often essentially impossible without weeks of careful research by large teams of (expensive) patent lawyers.

Sadly, on more than one occasion free-thinking software innovators have designed new algorithms and published them as free tools for the masses, only for it to be discovered at a later date that the developer was not contractually entitled to release the algorithm at all, and that a patent had been filed for it by their employers prior to publication.

The most famous (and awful) example of this was the LZW compression algorithm. Originally published in 1984 as an article in IEEE Computer magasine, it was quickly adopted by hundreds of independent software developers, and became part of the GIF image standard, used extensively on the Web. Years later it emerged that the patent for this algorithm had been bought by Unisys, a little known company who appeared from obscurity and declared that they were entitled to seek damages from every web site and software application that used (or could output) GIFs! Though the Unisys patent is now expired and the controversy around it all but forgotten, the more recent dispute between IBM and SCO shows that frivolous lawsuits over patent disputes are not going out of fashion.

So this is the sad reality of modern patents. Far from being used to protect the interests of the "small businessman with a big idea", as originally intended, patents are now used as stealth weapons, secretly hoarded by anonymous companies who skulk secretively in the shadows waiting for victims to innocently stumble on one of the ideas in their patent portfolio. They wait until some poor sap achieves moderate market success and then pounce on them with a lawsuit seeking outrageous damages and giving the victim little choice between paying them, or facing bankruptcy from paying legal fees for the duration of an inevitably long-drawn-out court battle.

Many modern patents never result in an actual product, instead they are traded in secret between giant companies so that they may be used as weapons in the legal cold war that erupts whenever anyone is foolish enough to try to bring a new product to market.

Patents are now so numerous and far-reaching that it is essentially impossible to introduce a new software application or electronic device that does not infringe on a patent that somebody, somewhere has filed. Instead, companies must gather their own patent arsenal so that when they are inevitably sued, they will hopefully be able to find something buried deep in their archive that will allow them to launch a countersuit to nullify their attacker.

Of course this just further encourages the growing trend of companies that do nothing except buy patents and file lawsuits. By actually producing any products they are putting themselves at risk, but by assimilating intellectual property rights and preying on competitors, they can earn a healthy living with far less expenditure.

It gets worse. The rate of innovation means the patent office is no longer able to keep track of the state of the art. Many patents are filed that grossly infringe on prior art, and many others push the boundaries of the obvious. More recently, genetics researchers have been patenting minor modifications to naturally occurring gene sequences and molecules for use in drugs. There is already great concern that patents have been granted for parts of the human genome. How is this possible? Surely this violates the principle that patents must be inventions not discoveries? The patents circumvent this by claiming that they cover the use of these sequences only as part of an original process or application (a definition which can easily be extended later by the use of ambiguous language, or further incremental patent applications).

Sadly, it doesn't matter to most companies that their patents may not stand up in court because the mere threat of court action is sufficient to blackmail businesses into coughing up licence fees. If the choice is between an extortionate licence fee or even more extortionate legal fees (with no guarantee of success) most companies just aren't prepared to take the risk.

Laws designed to prevent frivolous lawsuits are often ineffective in these cases due to the level of technical knowledge required to prove that an idea is obvious, or that it is conceptually similar to an existing invention. Also, sitting on patents until violaters achieve market success is not frivolous according to the legal definition, and the company in question can easily claim that they were "unaware" of their patent until recently, or that they had only just acquired it as a result of a merger.

So how can we change this?

Numerous suggestions have been made to combat this new patent plague. Ideas range from reducing the maximum software patent expiration period (currently between 15 and 20 years), to simply rejecting the concept of patents for software altogether. Patents were invented in a time before software existed, even as a concept, so one could attempt to argue that they don't apply to it.

Despite some benefits, I think this solution is disingenuous. While it is true that patents pre-date software, the principles of their application are a logical extension of the same idea. And banning software patents doesn't really help. It would reduce the incentive for large companies to fund software innovation, and it wouldn't even solve the problem - modern digital electronics devices suffer the exact same issues, with companies attempting to patent the specifics or generalities of particular user interface paradigms, or even vague concepts like "look-and-feel".

And if we were to go down this road, I believe the coming revolution is going to blur the hardware/software distinction so much as to make this unenforceable. In the end, we would would either have to ban all patents, or none at all.

So what can we do?

I think the better solution is to take a cue from the way that trademarks are handled. Trademark licences come with a "use it or lose it" clause. You cannot secretly trademark a name and then wait quietly for someone to come along and take it up before suing them for infringement. If you don't actually trade under your trademark then it expires and may be employed by someone else who actually intends to use it.

If patents were to come with an additional clause stating that they would be considered null and void if the company made no efforts to develop a product with them within a fixed (short) period, they would be much less of a menace to innocent parties, whilst still serving the useful purpose of protecting investments in innovative research.

The patents system also needs to be brought out into the open, with rules requiring companies to publish lists of all their patents on a single, publicly (and freely) accessible database. Any company that made use of an invention prior to it appearing on this list would not be liable for damages.

The patent ratification process also need to be tightened up. Patent applications should be required to undergo a public review, during which time anyone would be given the opportunity to provide evidence of prior art, or submit challenges to the innovative content of the proposal. Patents, if granted, would be deemed retrospectively applicable from the beginning of this process, so there would be no danger of copycats abusing this waiting period to steal the idea and capitalise on it.

Finally, these rules need to be agreed on an international basis, or at the very least the interactions between patents laws in different regions need to be clarified.

At the moment it is not clear if any infringement is taking place when a developer posts a US patent-violating freeware program on a server that is accessible from the US, but not located on US soil. Would he be required to paste a disclaimer stating that the material was not permitted in the US? Or even take stronger action to block such downloads? And what if the programmer himself was a US citizen, or based in the US?

The global nature of the Internet means that international borders are a very poor method of enforcing boundaries between contradictory intellectual property restrictions.

The key point here is that patent laws, like copyright laws are currently fighting against the tide of progress instead of guiding it. While their purposes may seem noble, the reality is that they cannot possibly function in this environment. Laws can only work if they reflect will of the majority. When we have thousands or even millions of people pouring their creativity into circumventing these regulations, we cannot expect the legal profession to be able to put the same magnitude of effort and creativity into frustrating them. And we have to ask ourselves whose interests are they serving by doing so?

In the end, whatever we decide to do, it must be done soon. The patent laws as they stand are failing, and with the power for defeating these barriers that the Internet community has demonstrated, we face a simple choice between reforming our concepts of intellectual property once and for all, or watching as they are trampled to dust under the steady march of progress.