Patenting Everything Is Not A Good Idea

Some software developers think all intellectual property (IP) should be patented.  Others feel – just as strongly – that no IP should be patented at all.  Both have good arguments.

How, the pro-patent people argue, can we make a living from the software development investments we make, if anyone can copy our intellectual property?  (Actually, they usually employ stronger language like “steal” or “rip off” instead of copy.)

How can the art of software development progress, the anti-patent developers argue, if everything worthwhile is protected and must either be licensed, perhaps at high cost, or can’t be used at all?  Anti-patent folks, often allied with the open software movement, like to trot out the “scientists stand on the shoulders of those who went before them” argument.  If they gave us all that help, that head start, they argue, how can we do less for the developers who will follow us? 

I’ve made that argument myself, from time to time.

There’s a third leg to this stool.  Sometimes software developers aren’t pressing for patent protection, but the venture capitalists who are providing their funding (or are promising to) require patent protection to make their investment safer.  Since you can’t write software and bring it to market without money, you can guess who wins this debate.

But what’s going on now is driving me crazy and you need to know about it.  It isn’t new; it’s just getting worse.  Basically, the Patent Office is granting patents on Software and Processes that are not new at all, based on the fact that the Patent Office Process is itself no longer up to the task of understanding what it’s looking at.

In the beginning, the Patent Office used to look at mechanical models of inventions.  Knowing whether something was new and worthy of being awarded a patent was a simple process of looking to see whether the new model was sufficiently different than the models already on the shelves. 

Enter software. 

	It exists in a vast and bewildering array.
	There is no physical embodiment of it usually presented
	In an industry where expertise is varied and scarce, the government finds it impossible to have a sufficient and varied supply to actually know, from an expert point of view, what it’s being asked to grant a patent on
	Many patents today are simply existing processes being moved from the physical world, where they existed as paper (fill-in-the-form-and- pass-it-around) processes, being moved to a computer

You think we’re being too harsh.  In many instances, the Patent Office has granted patents to something that was

1. Already patented by someone else; or
   
   

2. Already existed in commerce (this is called prior art and usually      precludes the possibility of someone being granted a patent); or
   
   

3. Already existed as a commonly used process; or
   
   

4. Seems absurd on the face of it

You may choose your favorite examples.  Here are a few we’ve found.

In the field of encryption software, a company named MAZ Software has gotten a patent called “overly broad” by a number of security software firms and experts.  You can read about it at http://www.staronline.com/vcs/business/article/0,1375,VCS_128_1094576,00.htm, where Phil Zimmerman, the inventor of PGP (Pretty Good Privacy), a well-known security product, says they could go after nearly anyone, including Microsoft, based on the patent that was granted. 

The problem is that companies like MAZ usually try to go after only companies that are big enough to afford their licensing fees but not big enough to actually fight them successfully through the courts to get their patent set aside.  In the meantime, good, small software firms, their business models damaged beyond repair, disappear.

We were once expert witnesses for a medium-sized company who was attacked by a small company created (their attorneys believed) for the sole purpose getting a patent on an aging but still used technology and suing companies they claimed infringed on it.  Unfortunately for their scheme, my client was a very smart and knowledgeable inventor and he had kept very good prior art records.  They lost their patent and their law suit.

If you want to see how silly this gets, on April 17, The Mercury News duly reported that a 7-year old Minnesota boy (whose father is a lawyer) got the Patent Office to grant a patent for a method of swinging a playground swing from side to side by pulling first on one chain and then the other.  If you’ve ever done this (I’ve watched three generations of kids enjoy this trick) you know it’s not original, but the Patent Office apparently couldn’t figure that out and granted him a patent on April 9.

Or you might want to look into the widely reported case of the orange growers in Florida who have been growing Valencia-like oranges from Brazil, imported to extend the growing season for oranges and to permit never-frozen orange juice to reach your supermarket shelves.  Tropicana insists that they are the only ones who should have the right to buy the fruit – they have a process patent to prove that.  The growers, of course, want to sell to the highest bidder.  We thought indentured servitude of agricultural help went out a hundred years ago.  Guess we were wrong.

Does this mean we shouldn’t have patents?  Of course not.  Having intellectual property protection is an important part of the development and commercialization process.  It helps justify investments and long development cycles by providing a measure of protection for the possibility of profit.

But it does mean we need a better method of determining

	What should be patentable
	Who should provide the expertise to determine patentability and what their credentials should be
	What demands we should make for a search of the prior art, especially when that prior art is not embodied in physical objects, but rather in business processes or software code.

(back to top)