Over the past few days I’ve read a bunch of misinformed blog posts regarding software patents and how they apply to the Freedom Scientific v. GW Micro case. I hope, with this post to talk about the law and why it applies to this situation.
Note to readers: I am not an attorney nor have I attended law school. I have attended a number of seminars on software patents, user interface copyright and other IP law that surrounds the high technology industry. Thus, I’m speaking from the view point of a well educated dilettante. Also, way back in the eighties, I co-founded the League for Programming Freedom with my friend Richard Stallman. Our purpose was to oppose user interface copyright (a case we won in a Supreme Court ruling in Lotus v. Borland) and software patents which a pair of Supreme Court decisions ended up with us losing that battle.
Judicial History on Software Patents
A long time ago, IBM filed a suit against the US Patent and Trademark Office. They asserted that the work by Benoit Mandelbrot on fractal geometry and chaos theory should be granted a patent. The court took the position offered by the PTO and ruled that all mathematics exist in nature and, therefore, can only be discovered and not invented. Discoveries cannot be covered by patents.
Years later, AT&T filed a similar suit but, as evidence, brought the design for a bit of hardware that, while probably too complex to actually build, did describe a machine. They claimed that software was merely an implementation of the machine and, therefore, covered by patent. Our side (we filed an amicus brief signed by most of the most important contributors to computer science from Marvin Minsky to Rodney Brooks) used the IBM decision and added that all algorithms, all of software for that matter, could be expressed as a function of the lambda calculus, hence, it was an expression of mathematics and discovered rather than invented. We didn’t believe this was an especially great argument but it was the best we had.
Thus, since the At&T case, software patents have been granted in abundance.
Freedom Scientific v. GW Micro
FS had a patent on placemarkers as a screen reading feature while using an Internet browser. This patent was granted years before GW Micro decided to add the feature to Window-Eyes.
The GW Micro placemarker functionality contains a number of new and innovative features but, at its base, it starts by providing the same action as does the JAWS placemarker feature which, as one can find in a search of the USPTO web site is pretty clearly described in the patent granted to Freedom Scientific that makes placemarkers in screen readers on the web their soul property for 17 years since the patent was issued.
Improvements to an invention covered by a patent held by FS or anyone else for that matter are called “derivative works” as they derive their basis from another invention. So, in this case, it appears as though GW Micro (I say “appears” because I have not heard all of the evidence nor am I member of the jury) based its invention on the one in JAWS which probably means it is a derivative work and, therefore, is in violation of the FS patent.
In most industries, GW would file for a patent on its improvements and then try to negotiate a cross licensing agreement with FS so GW could use the base feature and FS could use the GW enhancements. Unfortunately for GW, they published their beta before filing for a patent of any kind and, thus, created the published prior art for their own invention.
There are a number of arguments that GW can make in a court to defend themselves against FS but I cannot think of one that does not require removal of their placemarker feature from Window-Eyes.
Innovate, Don’t Litigate
By stating the above as clearly as I can, I must also say that I do not endorse Freedom Scientific’s behavior in this matter. I agree with many of the other bloggers that the new Window-Eyes with its very cool scripting facility is a major step forward in screen reading technology.
FS doesn’t advertise the patents it holds in JAWS anywhere in the product or package so GW could have unknowingly added the feature entirely without notice that FS had a patent on it. In my opinion, FS should have sent a warning before filing a suit but I have not been on their executive staff for nearly four years so my opinions are meaningless regarding their strategy on all matters, including prosecution of patents.
But, as far as I can read it, FS has a very solid case and are well within their rights to protect their intellectual property. Do I think this is ethical? Well, as I helped organize the fight in the Supreme Court against AT&T, you can probably guess where my emotional side stands but emotions are worth caca in a patent dispute.
I hope this piece helps the community understand how this tricky bit of law works and why FS has a very legitimate claim – whether you and I like it or not.