Software Patents are Stupid

Software patents are stupid.  I know – that’s not original nor is it a particularly new concept. But it is still a topic worthy of discussion.

A Google search for “stupid software patents” yields more than 20,000 results.  Still, software patents are getting renewed attention because of a recent dispute between Oracle and Google.  In mid-August, Oracle filed a lawsuit against Google, claiming infringement of Oracle’s patents and copyrights by Google for unlawfully using Java in its Android mobile platform.  In the lawsuit, Oracle seeks “appropriate remedies” because Google “knowingly, directly, and repeatedly infringed Oracle’s Java-related intellectual property” while developing Android.

The lawsuit has caused some uproar because it stems from Oracle’s recent acquisition of Sun Microsystems, the original developer of Java (and, therefore, the owner of patents related to it).

In contrast to Oracle, Sun was not a litigious company and for the most part, avoided aggressive legal action in defense of its own intellectual property.  But Oracle’s actions shouldn’t come as a surprise.  They’ve always been known as one of the most aggressive and confrontational powers in the technology industry.  After Oracle acquired Sun, Larry Ellison stated that Java was the “single most important software asset” that Oracle had ever acquired.  With that perception, it shouldn’t shock anyone that Oracle (and Ellison) would use it to aggressively pick a fight with Google.

I won’t try to decipher the legal foundation for the lawsuit here.  Did Google violate Java patents?  Probably – it’s nearly impossible these days to develop any substantial piece of software without infringing on somebody’s patents.  A 2004 study of Linux identified at least 283 patents that the open source operating system could potentially violate.  Every month, it seems, there’s a story of some big dispute where Company A claims that Promising New Product from Company B violates its patents.

Major technology companies work to build huge portfolios of patents, partially to provide negotiating options in these disputes.  Microsoft, IBM, Oracle, Apple, HP, and many others actively pursue this approach.  Disputes between these big vendors are rare, perhaps because of the strength of these portfolios – it’s sort of like the threats of “mutually assured destruction” from nuclear arsenals of the US and USSR in the 1970s.

The patent system wasn’t designed for the patenting of software.  The U.S. patent code provides for the protection of “any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof.”  The original intent was for protecting inventors of machines or devices or manufacturing processes.  A series of legal decisions in the 1970s and 1980s opened the door for widespread approval of patents for processes defined in computer systems – i.e., software.

Unfortunately, patent examiners are often ill-equipped to evaluate software patents.  Descriptions are arcane and can be extremely technical.  The claims (which define the scope of the protection that the patent will provide) are often written by the patent applicant to be as broad and general as possible.  With that approach, if the patent is granted, it will provide broad and general protection, allowing the patent holder to reap benefits (by preventing others from developing similar processes or products).

The other primary problem patent examiners have is determining whether “prior art” exists.  If the invention has been described publicly prior to the patent application, then the patent won’t be granted.  Inventors and patent attorneys are legally required to identify any relevant references when they submit an application, but they may not be aware of (or make the effort to discover) whether that prior art exists.   Too often, examiners miss something, the patent is approved, and (too late) it’s discovered that the idea wasn’t really original at all.

Here are some examples of patents I think are stupid:

The company I work for has a number of software patents.  We have some innovative technology that solves technical problems in unique ways.  Given the way the patent process currently works (and the way intellectual property protection is established), getting those patents was a sensible business decision for us.

As a business leader, I find myself in an uncertain position.  To be responsible in managing my company’s assets (the intellectual property of our innovations), I need to protect those assets within the existing system.  Unfortunately, as I described above, that system is deeply flawed and poorly implemented to manage software and technology products.

So, it’s a dilemma.  Do we submit patent applications, protecting our ideas?  Yes, we do.  But that doesn’t mean I like it – I’d much prefer for the system to be restructured to reflect the current needs of today’s inventors.

For now…I’ll leave you with this thought:

Patent unique software products that make sense, not broad software processes.

Share
This entry was posted in General. Bookmark the permalink.