PASS Data Community Summit logo

PASS Data Community Summit 2023

Get all the latest announcements direct to your inbox

Software patents – stupid or insane?

Arthur Fuller considers software patents to be both stupid and insane, is it time for a review of patent law?

Lest the headline mislead you as to my biases, I consider software patents to be both stupid and insane. I raise this issue because it is currently rearing its ugly mug in the world of open source software, but it has affected much development in the proprietary worlds of Windows as well.

First of all, patent laws were created long ago, which is not to say the thinking was correct then either, but we have to recognize the intellectual and technological climate back then. It may surprise you that patents go back about 500 years. Britain claims the longest lineage of patent history, beginning with the patent granted to John of Utynam in 1449 for his method of making stained glass. His unique method, for which he was granted a 20-year monopoly, was used for the windows of Eton College.

After that, the history of patents went downhill. Even as early as 1600, patents were more political than intellectual. Elizabeth I granted a patent for the making of knives with bone handles, which was subsequently declared without merit since knives of that type were already widespread before the patent was issued.

Patents for the absurd

In recent years, patents have been granted for the most absurd “inventions,” and nowhere more perniciously than in the realm of software. Ironically, the company that could have legitimately made a case for patenting its ideas (Xerox) chose not to. The employees of the Palo Alto Research Center created almost everything in the current computer world-the concept of a GUI, the mouse, the window, and a whole lot more.

Had Xerox chosen to patent these concepts, the computing world might look radically different. Imagine a computer without a mouse and windows (not Microsoft’s version, but the generic concept of a window).

Apple purloined the Palo Alto concepts and delivered them first in the Lisa, if memory serves. Microsoft copied them and Apple sued. Back then, I suggested forming a SEX (Sue Everyone, Xerox) group-an idea before its time, I guess.

And now here we are in the 21st century, with Microsoft, Novell, SCO and Apple on one side of the fence and the open source community on the other. Actually, the location of the fences seems to change from month to month. What remains constant is that all these firms want to plant their flag on a certain number of concepts, either to prevent anyone from using them or to extract fees for their use. I mention the big players simply because the actual list is so long that it might comprise a volume of Encyclopedia Britannica.

The true meaning of discovery

I don’t want to discuss the particulars of each lawsuit or patent application. That would take millions of pages. Instead, I want to propose a radical alternative. Imagine a distant time when no one knew that the sum of the angles in a triangle totaled 180 degrees-no matter what shape the triangle. Would you grant a patent for that insight?

I would not. I would instead emphasize the meaning of the word “discovery”-it means to dis-cover: The idea was already there, waiting for someone to remove the blanket hiding it from public view.

Should I happen to discover a quicker sort of method than the standard sorts, I do not deserve royalties for this discovery! I deserve to be paid by whatever firm is currently employing me, but neither I nor said firm deserve the right to lock this discovery down and charge fees for its use-not for 20 years, not for one year, not for one hour.

In case you’re wondering, I have already put my money where my mouth is. As principal of a firm, I mandated that we ship the source code to our software products. We charged a fee to license our software-we had to make a living-but we held no secrets. We did not obfuscate our code. We laid it out for all to see: warts, elegance and all.

I don’t think that anything in any of our products was truly innovative, in the sense that Archimedes’ theorems were, but we did what we could and we exposed our ideas to public scrutiny. As a result, several of our most enthusiastic customers submitted enhancements that we incorporated into the products, with grateful accreditations. In one case, a customer was so good we hired him.

Sue me, sue you blues

I write all this because the corporate forces are gathering around the open source community and the concept underlying it, and lambasting it with nonsensical arguments. “Open source is commie.” No one to my knowledge has actually gone to that extreme, but reading between the lines you can feel that sort of intent.

In another case, Microsoft is suing Apple for a patent violation concerning technologies in the iPod-even though Microsoft filed its application after the iPod was released! Apple introduced the iPod in November 2001, but did not file a provisional patent application for various bits of the technology until July 2002, followed by a full application in November of that year. Meanwhile, Microsoft filed an application in May 2002 to patent various components of music players, including the song menu! What a concept! Think “restaurant” but it applies to songs! Is this brilliant innovation or what?

A potential law called The Patent Reform Act of 2005 is currently jumping through the hoops of the U.S. Congress. Among other changes, this law would base the validity of a patent on first-to-file rather than first-to-invent.

Wanting it both ways

So where should we go with software innovation? As you may have gathered, I think patents are the wrong way. Even copyrights pose serious problems, in my opinion. Suppose that I write a knockoff of Microsoft Word, without one shred of inside information as to its architecture. My menus are identical, the overall functionality is equivalent, my code runs more quickly, and my price is 1/10 of Microsoft’s-plus, I include the source code so you can enhance it if you wish. In this case, all I have done is clone Word’s menu-none of the underlying code is copied from Microsoft’s code. (Old-timers may recall an actual case like this, when Borland released its Quattro spreadsheet.)

Meanwhile, the Linux world is knee-deep in legal doggie-do, thanks to SCO (the Santa Cruz Operation), which is claiming that significant portions of its code have been used in the various Linux distributions.

Let’s now look at it from the perspective of the companies so adamant about their patents, copyrights and so on. Grab any license agreement that you have handy: I’ll bet that one of its first sentences resembles “This software is not guaranteed to do anything, and if you try to do something with it and it fails, tough luck, darling.”

Software companies want it both ways: to patent their discoveries and to walk away from any disaster that might befall your enterprise due to a bug. Automobile manufacturers have a little more integrity; when they discover a steering or brake defect, they issue a recall. They may get sued in the meantime, and although they kick and scream throughout the court process, occasionally they end up paying.

Politics and law are arts of negotiation, so I’m willing to split the difference. I’ll recognize software patents the day the software companies step up and stand behind their products. Until then, open source is a much better universe, commie or not.