Free Patents Online: IP Research and Community

Is Software Patentable? Wanna Bet?

First, let me give you a little background so that you can at least understand, and maybe share, my frustration. We had a patent application that had been pending for 6 1/2 years with the USPTO. Various things happened along the way that contributed to this lengthy, if perhaps all too common, pendency period. For example, at one point the examiner to whom the application had been assigned was “on vacation.” It was a really long vacation. So long in fact, that when we started asking when the application was going to be acted upon, the case was reassigned to another examiner. The new examiner, presumably having his own backlog, then didn’t touch the application for years. The USPTO didn’t even reply to written requests to update us on the status of the application.

Finally, when the application was acted upon, the first office action was to deny all claims, with prior art citations that made it apparent that the examiner didn’t really “get it.” Now, lest that sound too critical, I fully sympathize with the impossible job examiners have. With the amount of time they have per case, I’m sure I wouldn’t really “get it” in plenty of cases either. So, we replied, explaining the differences between the prior art and the present invention, pointing out that despite superficial similarity, the methods of the prior art were quite different and literally could not be used in conjunction with the present invention. Of course, you know what the second office action was: Deny all claims.

So, we appealed to the Board. Years later (seemingly in egregious violation of MPEP 1207.02), and after several phone calls during which the examiner really didn’t seem to know what was going on and furnished us with misleading information, we got the Board’s decision. This was just the other day. I don’t need to tell you what their decision was, do I? Deny all claims.

Now, you haven’t read the application, or the office actions. I would completely understand if you are thinking “Maybe they were correct to deny all the claims.” Maybe they were. I don’t think so, but maybe they were. But, that’s not what this is about. All that boring background material was just frustrated complaining on my part. The real point here is not how long this took, or what the Board said, but rather how they justified it.

I quote from the Board’s decision:

"The instant claims [are directed to a] software system, which is per-se non-statutory. For a system claim to be found statutory, it has to have a physical device or processor."

Sure, there was more in their reply than that, but on this topic that was largely it. No case law cited. Bilski wasn’t even mentioned, nor is the “machine-or-transformation” test. So, forget about whether the invention satisfied the requirements put forth in Bilski. I would argue that it did, but that doesn’t seem to be the issue. Rather, the issue here seems to be that we have three examiners, considering something on appeal, who find it legally correct and sufficiently explanatory to simply state that “software… is per-se non-statutory.”

What?? Did the Supreme Court render a new decision that I missed? We all know the gist of what Bilski says, but let me quote directly from the decision to make my point absolutely clear:

“…although invited to do so by several amici, we decline to adopt a broad exclusion over software…”


“…the proper inquiry under § 101 is not whether the process claim recites sufficient "physical steps," but rather whether the claim meets the machine-or-transformation test.”

In other words, software is not non-statutory, and the proper test does not necessarily hinge upon the recitation of physical embodiments. It seems to me that either the Board is wrong or the Supreme Court (and CAFC, whose decision was upheld by the Supreme Court) is wrong. Of course, one could argue that the wording is simply unclear, and that, for instance, by reading an awful lot into the Board's use of "per se" one can reconcile the Board's statements with the decision in Bilski. That's a stretch, especially from a Board who attempts to devine exactly meaning from language on a daily basis. Of course, it seems equally ludicrous that three examiners are all ignorant of Bilski and the machine-or-transformation test.

I don't know how to interpret this. I'd love to hear other thoughts on the matter. I do know that this has been a huge waste of time, highlighting some of the glaring problems with the patent prosecution process. This confusing, seemingly incorrect, response from the Board (after 6 1/2 years no less) should have been impossible. With the right tools available to the examiners, it would have been. But, that's a topic for another blog.


If you case was directed to application level software, you experience is not at all surprising.  I have a case that is over 10 years old, we have been through four appeals – the first three the examiner withdrew the rejection and cited new art.  We won the appeal, then the examiner cited new art.  Then we negotiated some claims that were allowable and then that examiner moved on and the new examiner cited new art.  We negotiated some even narrower claims that were allowable and are waiting for a Notice Of Allowance.  Oh by the way the examiner in one interview told me they did not want our patent to end up on the front page of the Washington Post.  In other words, the PTO is more worried about PR than following the law.
Now to your actual question:
1) Software is not patentable per se – this is absolutely correct since software is just a list of written instructions.  Only when software is put through a compiler (other conversion device) and then used to wire an electronic circuit known as a general computer is it patentable.  (I suspect you know this)  Does this make sense?  I would argue that it is an absurd formality.  The patent rules long ago gave up the requirement to explain features of an invention that are well known.  Thus we do not show how a super-heterodyne receiver works in a patent application.  In engineering we do not show the processor that is going to run an error correction algorithm, because it adds nothing to our understanding.  I suggest the formality of including the general purpose computer in the claims is absurd. 
2) Should software be patentable.  Software is just a way to wire an electronic circuit.  Unless you believe that electronic circuit should not be patentable, then software has to be patentable logically. 

Thanks for the great reply. Yes, I am aware of the "per se" prohibition on software. But, let me dissect that a little, although I realize I'm preaching to the choir on this since we both agree there are absurd formal requirements here.

First, the examiner doesn't say "software", he says a "software system" in the reply. It is unclear to me what a "software system" is, but I take it to include the hardware, otherwise I would just call it "software." Perhaps that's not what he meant.

Second, the invention is an internet-based one, and requires the infrastructure of the internet to be present to make any sense. So, not only is hardware obviously implied, but devices such as routers and servers are recited in the specification and the claims.

For example, there is a claim that talks about communication components and processing components, and the examiner's response says "... such components can be implemented in software alone. Claims directed to software alone per-se is [sic] non-statutory."

Huh? So, I recited hardware in the claim (granted, in what could be construed as an MPF manner in some cases, though not all), and the examiner essentially said "Well the hardware isn't needed, so you have only claimed software." How is that his determination to make? Keep in mind, this wasn't a Jepson claim. There was no question as to whether this material was preamble or actually part and parcel of the claimed material. It was part of the claimed material. It seems to me that the examiner took it upon himself to rewrite my claim in his head and then make up an objection to his mentally-rewritten version of the claim.

In short, I agree with you on the whole "per se" thing: Software, without any hardware, without meaningful transformation of data, without a tie to the "real world," isn't patentable. But, I don't think that was the issue here. Or, if it was perceived to be the issue by the examiner, it shouldn't have been. I'm reviewing the claims (one needs to do that when they were first written 6 years ago) as I write this to make sure I'm not saying anything unfair to the PTO, and I'm finding myself aghast at how many of the claims explicitly require a hardware component, yet were rejected as non-statutory.

I wholeheartedly agree with your take on software patentability and the parallel to circuits: At its most basic level, software provides instructions for logical gate-based operations performed in hardware. If I hardcoded the logic into a chip (a circuit, as you say), it would be patentable. And, our trials and tribulations notwithstanding, if I recite software with the appropriate hardware means, it SHOULD be patentable. Yet, if I just talk about software, which we all know ends up getting run on hardware (and it cannot logically be otherwise - there is no alternative), it's not patentable.

It seems to me that this is actually a manifestation of some level of discomfort with software patents. Some people don't like them. Plenty of people think they shouldn't exist. Yet, under current patent law in the US, they do. So, either accept that, or change the law. Don't make these byzantine requirements and inconsistent decisions.

Well, enough preaching to the choir lol. And, your quote about the Washington Post is priceless. Thanks!