Software Patents and Free/Libre and Open Source Software (FLOSS) (Part 1)

I was invited to speak at the IT360 conference on Software Patentsand Open Source, which I did in the afternoon of April 9. While my slides are online, I believe it is worth discussing the presentation here for the benefit of those who did not attend.

I offered a few bullets as a summary.

  • What is a patent, contrast with other PCT : most people are not clear about the different types of intangible exclusive rights such as Patents, Copyright, Trademarks, and the various related acts, so I summarize them.
  • Should software be patentable? : It is important to ask this question, including having an overview of the history in various countries and the different levels of exclusive rights offered to software.
  • Software patent issues specific to FLOSS : While it is important to differentiate software from other subject matter, we also need to ask just how different FLOSS is from non-FLOSS when it comes to patents.
  • How realistic are different types of threats? : People in the FLOSS sector presume that Microsoft is the greatest threat to FLOSS as they do the most sabre rattling, and I suggest this is not appropriate.
  • Policy proposals : I knew there would be people from government there, so I offer some proposals
  • Discussion

While I had discussion last, the session was quite informal and the entire hour was open for discussion.

I am not a fan of the term “Intellectual Property” for many reasons.One is that it is confusing as it makes an analogy to tangible propertywhich confuses non-lawyers, and lumps together very different areas oflaw. This is why many people in policy circles use the acronym PCT which stands for “Patents, Copyrights, Trademarks and other related rights”.

  • Patent: applies to “art, process, machine, manufacture or composition of matter” which are workable, new, and ingenious (useful, novel, unobvious). You need to apply to get a patent, and patents are then granted by the patent office.
  • Copyright: applies to an expression of an idea. No registration is required for copyright at this time, although there are benefits to registration if you ever need to sue for infringement.
  • Trademark: word, symbol or design, or a combination of these, used to distinguish the goods or services of one person or organization from those of others in the marketplace. This area of law was intended to be a form of consumer protection, different than other areas of PCT which are intended to provide incentives to authors and innovators. You need to apply for and receive registration for trademarks.
  • Also includes: Industrial Designs, Printed Circuit Topographies, Plant Breeder Rights, …

The discussion of whether PCT is similar to tangible property goesback hundreds of years. I discuss Thomas Jefferson’s contribution tothe debate in 1813 in an essay Jefferson Debate: A Godwin’s law for copyright discussions? where I challenge property language used in political debates.

The generally accepted purpose of PCT laws is to create exclusiverights to provide incentives for creativity and innovation. The idea ofcreating a property-like system is that property law providesincentives for investment in what is owned, so it has been presumedthat the same model will work for intangibles as well. Exclusive rightsworking like a functioning property system encouraging investment needsto be understood as a policy goal that needs to be analysed andtweaked, and not an axiom that doesn’t need proof. Many of the failureswe see in practises are examples where this policy is failing to actlike a property system. For these failures to be fixed they needs to berecognized as not only possible, but inherent in the differencesbetween tangibles and intangibles.

A recently released book titled Patent Failure: How Judges, Bureaucrats, and Lawyers Put Innovators at Risk by James Bessen and Michael J. Meurer offers considerable detail on this argument, and is highly recommended.

Historically software was not patentable, and it is only recent that this changed.

The US Patent and Trademark Office (USPTO) fought software patentsprior to 1981. This changed after a 1981 US Supreme Court Case (Diamondv. Diehr) which involved an industrial process for the molding ofrubber products, which included software. While this was not really asoftware patent, but an industrial process that happened to includesoftware, this was seen as opening the doors to pure software patentsand then later to business model patents. Many of these decisions weremade by the US Court of Appeals for Federal Circuit which specializeson patents, and is seen by many to be a biased interest group in thedebate about what should be patentable.

In Europe they had a chance to do things more recently, and specifically excluded “programs for computers” in article 52 of the European Patent Convention (1973.This exclusion was reflected in the 1978 Examination Guidelines for theEuropean Patent Office (EPO), but changed the guidelines in 1985 tointerpret the exclusion list such that software patents were thengranted. Policy discussions are active as part of the proposedCommunity Patent. The European software community does not trust EPO asthey ignore article 52 exclusions, and many practitioners feel thatsoftware should not be patentable.

The Canadian Patent Act does not mention software, and has a veryminimal exclusion list which states that “27(8)No patent shall begranted for any mere scientific principle or abstract theorem.” Thecourt case that is often mentioned is Schlumberger Canada (1981), whichinvolved the use of a computer to analyze data concerning soilcharacteristic measurements for oil and gas exploration. This was yetanother example of a situation where a more traditional patentableprocess happened to include software, but where the software was notitself the subject of the patent.

The Canadian Intellectual Property Office (CIPO) sets guidelines for patentability in their Manual of Patent Office Practices (MOPOP).This manual was updated in 2004 as a result of a closed “consultation”that largely only includes special interest patent agents. You can readabout this consultation in the results from an Access to Information request I have posted online.

The new Chapter 16 on “Computer implemented inventions”has language that would suggest that pure software is not patentable inCanada, but it requires understanding a distinction that doesn’t seemto be generally understood. The chapter includes the key phrase, “Thepresence of a programmed general purpose computer or a program for sucha computer does not lend patentability to, nor subtract patentabilityfrom, an apparatus or process.”

This should suggest that if you have a manufacturing process thathappens to include software, that the presence of the software does notalone render the manufacturing process unpatentable. This would confirmwhat was said in the Schlumberger Canada case, which did not involvepatents on software.

This should equally suggest that if you have a general purposecomputer and add software, that you still simply have a general purposecomputer and no additional patentable subject matter. This should denyinformation and mental process patents, such as methods involved inencoding or deciphering information (file format patents, codecs, orother patents on computing interfaces, etc).

What has been happening is that a broad range of computer relatedsubject matter has been granted by patent offices around the world,including information and mental process patents.

While the above answers the question of whether software ispatentable (it is in an increasing number of countries), it requires adifferent type of analysis to answer whether this is subject matterthat should be patentable. I will continue this subject in the secondpart, although people should feel free to dive into the discussion inthe comments. This article is largely background needed for a much moreinteresting future discussion.

Related Download
CanadianCIO Census 2016 Mapping Out the Innovation Agenda Sponsor: Cogeco Peer 1
CanadianCIO Census 2016 Mapping Out the Innovation Agenda
The CanadianCIO 2016 census will help you answer those questions and more. Based on detailed survey results from more than 100 senior technology leaders, the new report offers insights on issues ranging from stature and spend to challenges and the opportunities ahead.
Register Now