Why IPWatchdog is Wrong and Computer Programmers Aren’t “Clueless”


By Joseph Kamien

Have you ever wished your computer could respond to emails for you? I’m sure you can picture it now; it would be like having your own virtual secretary. You just sit back and relax on a beach while your little machine takes care of all your important correspondence. “But wait,” you may think. “My computer may be pretty smart but it doesn’t know everything. What if there’s an email it doesn’t know how to respond to? Or what if there are certain emails I always want to answer personally?” Simple enough: we’ll just add a feature to this hypothetical computer program which classifies emails into ones that can be handled automatically and ones that require a personal response. “Amazing,” you probably think next. “If I could create this program, I could make a lot of money.” There’s a problem though: another company has already created a program that does this. “So what?” you probably think. “This is America. I’ll just create a better program, and compete with that other company. There’s nothing illegal about doing that, right?” Well, actually…

The program I’ve just described is patented as U.S. Patent No. 6,411,947 (the ‘947 patent). This patent has had a long and storied history, and its validity is a rather contentious issue in the blogosphere. One blog, aimed at computer programmers, calls the patent ridiculous and obvious. Gene Quinn at IPWatchdog, a venerable blog on intellectual property law, argues that the patent is not obvious and that computer programmers who complain about software patents are “just about completely clueless, at least with respect to patent law.” Mr. Quinn has two arguments for why the patent isn’t obvious, despite the “rants” of computer programmers who think otherwise. First, Mr. Quinn argues, computer programmers are clueless and didn’t read the claims of the ‘947 patent to understand how they limit the scope of the invention. Second, software is often not obvious because it’s so difficult to make software that works (“things don’t go as they are predicted, that there is not cross platform operability, and that there are truly challenged users who deviate from the norm and find new and ever more creative ways to accidentally crash whole systems”). Though I respect Mr. Quinn, he is wrong on this issue. The ‘947 patent should never have been granted because the invention was obvious at the time of invention.

First, computer programmers are not clueless, at least when it comes to understanding computers. They tend to understand what a computer is at its most basic level: a device that takes an input, runs it through a series of rules, and then produces an output. Mr. Quinn’s demonstrates how clueless computer programmers are by quoting the first (and broadest) claim in the ‘947 patent:

1. A method for automatically processing a non-interactive electronic message using a computer, comprising the steps of:

(a) receiving the electronic message from a source;

(b) interpreting the electronic message using a rule base and case base knowledge engine; and

(c) classifying the electronic message as at least one of (i) being able to be responded to automatically; and (ii) requiring assistance from a human operator.

Does this look familiar? It’s essentially a patent on taking an input (“receiving the electronic message from a source”), running it through a series of rules (“interpreting the electronic message using a rule base and case base knowledge engine”) and producing an output (“classifying the electronic message…”). In other words, it’s a patent on using a computer as a computer. To those who understand computers, this is plainly apparent, and that’s why this patent is “ridiculous.” It doesn’t make any more sense than a patent on a process for using a hammer to put a nail through wood or a process for using a typewriter to write a letter. The Supreme Court has made clear that an invention isn’t patentable if it is merely a combination of two previous inventions using each prior invention’s known function. See KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 416 (2007). Logically, using only one existing invention for its known function is certainly obvious. Computer programmers aren’t clueless; it just appears that way sometimes because they see computers for what they are.

Understanding this, it’s clear why Mr. Quinn’s second argument (that the patent isn’t obvious because software is difficult to write) isn’t valid either. Writing a letter is difficult. Most letters don’t work (in the sense that they don’t get their intended messages across). They often contain spelling and grammatical mistakes as well as errors in tone and voice. Still, this doesn’t mean that one could get a patent on a process for using a typewriter to write a letter. The immense effort and creativity required to write a good letter make it a worthwhile subject for copyright law, not patent law.

Ultimately, Mr. Quinn is wrong when it comes to the patentability of this invention and the cluelessness of programmers. The USPTO was right to invalidate claim 1 and many of the other claims of this patent.

Image from Microsoft Office clipart.
Correction An earlier version of this article erroneously summarized the second part of Mr. Quinn’s argument as claiming that all software is patentable.

Leave a comment


  1. I would love for you to point out where I have every said that all software should be patentable. I have numerous times taken the opposite position, so I am just curious why you would make such an erroneous and provably false claim as if it were fact.

    I’d also love for you to explain why software so frequently doesn’t work if it is easy to write.

    By the way, software code is not patentable. Code writters are not inventors.

  2. jtkamien

     /  June 28, 2011

    You make a fair point in the first paragraph of your comment. When I first read your post, it sounded like you were arguing: software is generally difficult to write, therefore all software patents are valid, therefore the ’947 patent is valid. Re-reading it after seeing your comment, it sounds more like the second part of your argument is: software is difficult to write, therefore not all software patents are obvious, despite what critics may say. Is that a more accurate summary?

    As for the second paragraph of your comment, software is definitely not easy to write. In fact, it’s often very difficult to write, difficult to debug, and difficult to maintain. When it comes to patentability, however, the issue is not difficulty, but rather obviousness. Software programmers are not “clueless” for thinking that the ’947 patent is obvious. It’s a patent on using a computer to take an input, run it through a series of rules, and produce an output, which is exactly what computers are supposed to do.

    As for the third paragraph of your comment, I agree. Software code is not patentable.

  • Categories

  • Archives

  • Disclaimer

    The content of this website is not intended to be legal advice, nor should it be interpreted or used as such. The authors and editors make no representations or warranties as to the accuracy of the information posted.