On the traditional opening day of the U.S. Supreme Court, the first Monday in October, a legal-related IT posting seems appropriate.
But take this opinion for what it's worth: The sum total of my legal experience consists of passing the Law School Admission Test, four small-claims cases, and religiously watching "Law and Order" on TV.
I believe software, like this blog post, should be subject to copyright, but not patents.
This opinion puts me in rare agreement with Red Hat, whose lawyers filed a friend-of-the-court brief on Oct. 1 with the U.S.Supreme Court against the patenting of software. The case involves a civil suit against the Director of the Patent and Trademark Office (PTO) of the United States. The case has nothing to do with software, but on such twists important points of U.S. law are decided.
However, as usual, I disagree with Red Hat's tactics. Unfortunately, Red Hat likely is fighting the wrong fight in the wrong court in the wrong way.
Taking those points in reverse order, the first half of the Red Hat legal document is full of self-serving marketing claims. Red Hat should have just thrown out the first 11 pages of its brief and stuck with the substantive argument beginning on page 12. A little bombast works when you're a startup, but gets pretty tiring after years of it and especially as a company reaches a half-billion-dollars in sales while still claiming it's the little guy. I hope that Red Hat's poorly substantiated claims about market size and market dynamics are not used by PTO lawyers to harm the case.
For example, the brief cites as important the well-publicized 2008 SAP Research/SAP Labs study about the growing number of open source projects and the fact that the number of lines of code available for license under open source terms and conditions is increasing superlinearly (using the SAP authors' terminology). Using lines of code as a metric for anything is nuts in my opinion. A high number of lines of codes simply means inefficiency to me.
Then Red Hat links number of lines of code to a claim that open source software is "of strategic economic importance." But there is no study cited in the legal filing that makes that claim. In fact, I don't know of any legitimate study that makes that claim although the word "strategic" provides a lot of wiggle room. I've seen a study that says open source might be saving 6 percent of enterprise CAPEX and OPEX, but I can't verify it elsewhere. And I've seen studies (and done a few myself) that show open source terms and conditions have grown to account for a few percent of the quarter-trillion-dollar software market after 20 years. Neither statistic would contribute to a claim of "strategic economic importance."
The importance of open source terms and conditions is the fact that they hold down the research-and-development costs of the major software suppliers such as SAP, which own most of the software patents to which Red Hat and I object. Open source terms and conditions allow SAP, IBM, Oracle, etc. to share R&D efforts without violating various antitrust regulations and without spending a lot of time on administering patent cross-licensing agreements (which they all have with each other). In theory, lower R&D costs decrease eventual IT user costs, but they are just as likely translated to profits for the leading software suppliers.
As for being in the wrong court, the Red Hat brief proceeds to argue that software developed to be licensed using open source terms and conditions is developed differently than software licensed in another fashion, with all the usual canards about "volunteers," faster bug fixes, and user-driven quality assurance. It is untrue, of course, that software licensed with open source terms and conditions has any special claim to those characterisitcs when it comes to development. But opposing lawyers need only to argue that those characteristics are irrelevant when a later case reaches the Supreme Court that is actually about software. Remember that the current case involves the U.S. government's position that business methods can be patented and is not about patenting software.
That's not to mention that Red Hat is arguing to the Supreme Court that the real problem is that patents mean software suppliers have to pay lawyers too much money.
And talk about the wrong fight. With a straight face, Red Hat argues that the now 15-plus-year-old decision to allow the PTO to patent software has stifled innovation and competition in the market. Inc. magazine illustrates how dozens of new software companies start up every year. I guess Red Hat wants us to imagine what might have happened in addition to laser-guided surgery, Garmin, Google and Facebook if it hadn't been for patents. Of course, if only software-enabled mortgage securitization innovations had been stifled.
The brief goes on to talk about the "enormous damage..." patents have done "to the software industry" in the last 15-plus years. This ignores the fact that the software industry didn't effectively exist separate from the computer systems industry before 20 years ago. A separate quarter-trillion-dollar-a-year market has emerged from zero mostly during a period that Red Hat alleges innovation was stifled and enormous damage was done to it. Software as a separate entity did not reach maturity prior to patents as Red Hat contends. Separate protection for software was not needed when it was effectively tied to the minicomputer and mainframe systems on which it ran.
Not that I'm arguing for patents. As with Red Hat, I believe software should be copyrighted as opposed to patented because it is -- unfortunately (but that's the subject of another blog post) -- a stringing of "words" together just like this blog post. You can take my idea if you want and say the same thing by stringing a different group of words together that say the same thing or even by stringing the same words together in a different order. You can take a snippet out of my blog post and repeat it under Fair Use laws. Convention says you give me credit and typically you would want to credit me or otherwise you would not want the snippet. But you can't reprint the blog post in whole without the permission of the copyright owner, IT Business Edge. How much more legal protection do software programmers (or analysts) need? All open source software is copyrighted.
And, of course, it is also licensed under a bewildering array of multiple dozen different licenses. That's another problem in Red Hat's brief; it conveniently blurs the critical distinction between free software and open source software. But that's not a legal issue.
Happy first Monday in October!