CAFC Kills Means-Plus-Function in Software Patent

Last week the United States Court of Appeals for the Federal Circuit issued a curious decision in Ergo Licensing, LLC v. CareFusion 303, Inc.  This decision should send chills down the spine of everyone who has patents in the software space, and needs to be carefully considered by all those who draft patent applications in the software space.  The panel was made up of Judges Newman, Linn and Moore.  The opinion was delivered by Judge Moore, and a dissent filed by Judge Newman.

Judge Newman explained in dissent that the way in which the Ergo patent was drafted was similar to the way that thousands of similar patents are drafted and that this decision upsets the overwhelming weight of Federal Circuit precedent on the issue.  Judge Newman is correct.  Whether this becomes one of those decisions that is an aberration of is something to be afraid of remains to be seen, but moving forward your software patent applications should become far more detailed in an effort to deal with this case.

The history of this case begins with Ergo Licensing, LLC and Dr. Uvo Ho?lscher (Ergo) accused CareFusion of infringing claims of U.S. Patent No. 5,507,412 relating to an infusion system. In construing the claims, the district court held that the terms “control means” and “programmable control means” are indefinite. The Federal Circuit panel (per Judge Moore with Judge Linn) agreed because the district court correctly held (in their view) that no corresponding structure was disclosed in the specification.

Ergo sued CareFusion for infringement of claims 1-12, 15-16, and 18-20. Prior to the Markman hearing, the parties stipulated that several terms were means-plus-function terms, including the terms “programmable control means” and “control means.” The majority correctly pointed out that “if an applicant does not disclose structure for a means-plus-function term, the claim is indefinite.”  The question hotly disputed was whether there was sufficient disclosure.


Curiously, the despite this being a case about whether the claims were indefinite the majority never set forth the claims.  Judge Newman did, however, in her dissent.  The claims at issue were Claims 1 and 18, which read:

1. Multichannel metering system for metering preselected fluid flows, comprising:

a plurality of individual fluid flow sources;

a plurality of discharge lines, each line of said discharge lines being connected to a corresponding one of said fluid flow sources;

adjusting means associated with said fluid flow sources for acting on said fluid flow sources to influence fluid flow of said fluid flow sources;

programmable control means coupled with said adjusting means for controlling said adjusting means, said programmable control means having data fields describing metering properties of individual fluid flows;

an operating surface connected to said control means;

data input means for input of data into said control means, said data input means being at least partially connected to said operating surface;

data output means for output of data from said control means, said data output means being connected to said operating surface;

selector switch means forming a part of said data input means, said selector switch means including a plurality of selector switches, each selector switch being associated with a set of said fluid flow sources for representing segments of data fields belonging to a corresponding set of fluid flow sources on said operating surface, said each selector switch functionally connecting said data input means with said data fields belonging to said associated set of fluid flow sources.

18. Multichannel metering system for metering preselected fluid flows, comprising: a plurality of individual fluid flow sources divided into a plurality of sets; a plurality of discharge lines, each line of said discharge lines being connected to a corresponding one of said fluid flow sources;

adjusting means associated with said fluid flow sources for acting on said fluid flow sources to influence fluid flow of said fluid flow sources;

control means coupled with said adjusting means for controlling said adjusting means, said control means having data fields describing said fluid flow sources and metering parameters of said individual fluid flows, said control means including a meter management mode for editing and regulating said metering parameters;

a[n] operating surface connected to said control means;

data input means for input of data into said control means, said data input means being at least partially connected to said operating surface;

data output means for output of data from said control means, said data output means being connected to said operating surface;

a plurality of selector switch means, each of said selector switch means being associated with one of said plurality of sets of said fluid flow sources, said each selector switch means placing said control means in said meter management mode for said associated set of fluid flow sources.

Emphasis is added to draw attention to the aspect of the claims at issue in the case.  Perhaps after reading the claims it becomes clear why the majority did not set forth the claims.  Yes these claims use “means language,” but would someone who is familiar with the art think that these terms were indefinite? Absolutely not!  The “programmable control means” have associated “data fields describing metering properties of individual fluid flows,” and the “control means” have associated “data fields describing said fluid flow sources and metering parameters,” as well as “a meter management mode for editing and regulating said metering parameters.”

All of this might be confusing to a Judge who is unfamiliar with computers, but that is not the test, is it?  The question is supposed to be whether it would be confusing to a person of relevant skill in the relevant technology area.  Indeed, disclosure sufficient for means-plus-function claiming may be implicit or inherent in the specification if it would have been clear to those skilled in the art what corresponds to the means-plus-function claim limitation.  See MPEP 2181 and Atmel Corp. v. Information Storage Devices, Inc. (Fed. Cir. 1999).  Indeed, the Federal Circuit in Atmel concluded that the title of the article incorporated by reference in the specification may, by itself, be sufficient to indicate to one skilled in the art the precise structure of the means for performing the recited function.  So the focus is supposed to be on one of skill in the art even when interpreting whether the specification provides adequate support for means-plus-function claiming.  Nowhere in the majority opinion is it stated that a person of skill in the art would not have known and would have found the claim vague or ambiguous.  The closest the majority comes is to say: “As Ergo’s expert testimony demonstrates, the ‘control device’ is not a specific structure.”  That, however, is not the right question or at all important.  The critical question is whether one of skill in the art would understand what it means.  Thus, the majority relied on a wholly irrelevant matter when it focused on Ergo’s expert.

Ergo argued that the corresponding structure for “control means” was the recitation of “control device” throughout the specification. Indeed, Judge Newman would point out later that there was voluminous description of the claim terms at issue in the specification, saying: “From the description in the court’s opinion, one would not suspect that the system is described in ten columns of specification, plus four figures.” Notwithstanding Ergo’s argument and Judge Newman identifying where in the specification there was support the majority in a rather cursory way simply concluded that the terms were indefinite.

As already briefly mentioned, to support its decision the majority tried to rely on the testimony of Ergo’s own expert who testified that those skilled in the art would not recognize a “control device” as a particular, known structure.  Rather he testified that there were at least three different types of control devices commonly available and used at the time to control adjusting means; namely microprocessors, discrete circuits connected to step motors and analog circuits.  Of course, proper interpretation of the law would have the term cover each of these appreciated mechanisms by invoking the term, a point seemingly lost on a majority more interested in requiring disclosure of a specific structure rather than an informing disclosure.

Incredibly the majority acted as if this small, closed universe of options would require a hunting expedition, saying: “Although one of skill in the art may have been able to find a structure that would work, that does not satisfy § 112 ¶ 6.”  That is really nonsense.  Had this been an obviousness case the panel would have concluded just the opposite, finding that such a small number of known substitutes would render the trial and error that would be necessary to be well within the knowledge and abilities of the person of skill in the art, thus rendering the invention obvious.  Sorry.  You just can’t have a small number being onerous for disclosure and a small number being trivial for obviousness determinations.

The majority also took issue with the lack of disclosure of an algorithm.  The majority explained that until recently a general purpose computer was not sufficient, but that In re Katz it was determined to be sufficient if there was a corresponding algorithm disclosed that would enable the general purpose computer to operate specifically for the necessary purpose.  The majority explained:

The “control means” at issue in this case cannot be performed by a general-purpose computer without any special programming. The function of “controlling the adjusting means” requires more than merely plugging in a general-purpose computer. Rather, some special programming would be required in order to control the adjusting means.

In dissent Judge Newman would take issue with the majorities characterization of Katz, saying:

My colleagues misread Katz. The court ruled in Katz that the district court interpreted our prior cases too broadly, and reaffirmed that the requirements of §112 ¶6 are met when claimed functions are “coextensive with the structure disclosed.”

She then goes on for three-quarters of a page citing case and holding after case and holding, for example citing Telecordia Technologies v. Cisco Systems for the proposition that the term “controller” provides sufficient structure for “monitoring means” that “an ordinary artisan would know how to interpret the specification and actually build a circuit.”

Of course, we have already established that the majority wasn’t worried about whether the disclosure would inform the person of relevant skill in the art.  Instead the majority seemed to want pinpoint and precise description of everything; namely pinpoint description of “control device,” regardless of whether someone of appropriate skill in the art would appreciate the invention by reading the disclosure.


Pinpoint precision without regard for what would be understood by the relevant actors in the industry has never been required in a patent application prior to this case.  For example, the enablement requirement requires the disclosure to explain how to make and use the invention to the relevant audience so that it could be made and used without undue experimentation.  Without going into a long dissertation on what is undue experimentation, even the casual observer should notice that if the prohibition is for “undue experimentation” that must as a matter of fact, law, logic and reality mean that at least some experimentation is acceptable.

Judge Newman criticized the majority opinion in the dissent because there was detail in the application sufficient support for the claimed terms. She was further frustrated by the fact that the majority did not specifically identify the deficiencies or what should have been done to satisfy them.  Newman summarized:

The panel majority is not explicit as to the specific area of inadequacy of structure on which it relies for its rejection as fatally flawed under §112 ¶6. No error is attributed to the patent examiner. The patent specifica- tion is clear that the invention is the arrangement of the components of the system, not any special or nonobvious structure of a component. The specification is clear that the inventor is not claiming the control device as its invention.

Indeed, the panel majority does not explain the nature of the inadequacy, merely concluding that an inadequacy exists and presumably overlooking not only the invention being a system but also overlooking the detailed description of the pieces and parts of the system present in the application.

Exasperated by the fact that the majority did not describe what should have been present to have a satisfactory disclosure in their opinion, Judge Newman wrote:

This court’s foray into patent draftsmanship, finding standard presentations now to be fatally deficient, adds grievous unreliability to duly granted patents. The invention patented by Ergo displays the established protocol of specification content as characterizes many thousands of computer-assisted procedures. The court now rules that “more” was needed, although I cannot discern what more, except for a five-foot shelf of zeros and ones.

Taking her argument one step further, Judge Newman pointed out that CareFusion’s own patents use the very same language.  Judge Newman wrote:

Although CareFusion sold its theory to my colleagues, it is curious that the description in CareFusion’s own U.S. Patent No. 5,171,301, entitled “Multiple mini-pump infusion system” tracks that herein, for CareFusion recites “programmable electronic means” in some claims and depicts, as the structure that performs the function, a microprocessor as a rectangular box. ’301 patent, col.4 ll.37-44; id. at Figure 2. Indeed, so do thousands, perhaps hundreds of thousands, of issued patents.

Curious indeed!

Judge Newman ended her dissent with a flourish:

The great weight of authority ratifies the presentation in the ’412 specification. Neither CareFusion nor my colleagues on this panel have shown by clear and convincing evidence that the ’412 claims are invalid for indefiniteness. The court’s new position simply taints thousands of heretofore innocent patents, adding a further infusion of unreliability to the patent grant. This is of particular concern because the panel majority is not relying on newly cited references or other possible examination oversights, but is invalidating a patent on purely formalistic grounds—grounds on which the expertise of patent examination is normally superior to that of judges.

This destruction of a granted patent based on a presumably flawed disclosure in the application, at a time when it cannot be remedied, is not only a disservice to inventors who expect a reliable patent upon examination and grant, but an injury to the public that is served by patent-supported innovation. I must, respectfully, dissent.

So what is the take-home lesson?  Regardless of whether things will be well known and understood by those of skill in the computer software arts, the question seems to now be whether they would be well known and understood by a Judge who is not a person of skill in the art.  That being the case, applications should include an abundance of disclosure, diagrams and images providing as much tangible description of everything you mention in the claims, particularly if you are going to rely on means-plus-function claiming as is frequently done in the computer software area.

Furthermore, disclose algorithms for goodness sake!  If you are going to have a computer do something we still have to live in the world where black magic is involved in transforming a general computer into a specific use computer through the use of a special algorithm.  That is what Judges seem to want, so assume no familiarity or knowledge.  Also assume that they will not concern themselves with whether one of skill in the art would understand what is going on.  It is as if they think “it is just a computer” so they can understand things and to them it is hardly clear or informing if you don’t disclose every bloody aspect of every bloody thing regardless of whether it would be easily appreciated by one of relevant skill in the art.

In the computer age it seems that more and more Judges are erroneously assuming they are skilled enough in the computer software arts to make technical decisions.  We don’t have to like that, but we do have to deal with it, which means that software patent applications will need to become longer and more detailed, which means they will cost more to prepare.  In the end clients lose because they will need to pay more, but society doesn’t gain because those of skill in the art will merely see the resulting disclosure as overkill; merely describing what they already appreciated.


Warning & Disclaimer: The pages, articles and comments on do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of

Join the Discussion

11 comments so far.

  • [Avatar for kai ng]
    kai ng
    April 17, 2012 03:41 pm

    In this embedded software application, it all relies on how technically innovative either the mechanical or the software apects of the design are. As the hardware isn’t all that remarkable at all, and noting that the first claim relies on the “plurality” of items (presumabling there’s no chance of claiming invention on any single item), then the software becomes most significant invention of the disclosure. By being lazy and not putting in enough details of the algorithm, the applicant has disclosed nothing new.

  • [Avatar for jon]
    April 9, 2012 11:46 pm


    and this precautionary claim set is what exactly? so a claim set must now disclose the exact substance of the specification, anything that appears to fall outside of the scope of the specification would be considered indefinite?

  • [Avatar for Gene Quinn]
    Gene Quinn
    April 9, 2012 11:36 pm


    No, software patents live on. Means-plus-function claiming of software has suffered a terrible, perhaps fatal blow. There is still a place and time, but it needs to be a precautionary claim set that is intended to cover whatever the extent of the disclosure is. It cannot be your only claim set, that is for sure.


  • [Avatar for jon]
    April 9, 2012 11:24 pm

    ok so no more software patents?

  • [Avatar for Gene Quinn]
    Gene Quinn
    April 9, 2012 05:57 pm


    I couldn’t agree with you more. This seems like judicial activism pure and simple.


  • [Avatar for TINLA IANYLTINLA]
    April 9, 2012 05:48 pm

    Asked, “What do these Judges want?”

    They want you to stop drafting means plus function claims. They’ve hated such claims since Congress forced them on the Court, and they have addressed the Examiner corps directly to demand that Examiners force applicants to state on the record what each and every means plus function limitation means because “it’s too much work” for the Judges to do it. In short, this is their way if penalizing Applicants who exercise the option granted by Congress to use means plus function claims.

  • [Avatar for step back]
    step back
    April 9, 2012 02:30 pm


    Have you had a chance to look at today’s software killing decision:

    As such, the specification of the ’435 patent must contain an algorithm to perform the function associated with the “access means” limitation, or the limitation is indefinite. This appeal turns on whether the specification discloses an algorithm to perform the function or functions associated with the “access means” limitation.

  • [Avatar for Gene Quinn]
    Gene Quinn
    April 6, 2012 11:44 am


    “What exactly do these judges want?”

    I don’t know. They are obviously not particularly well informed on patent law issues and disclosure requirements. They simply don’t understand the law.

    So what should we do? I think we may have to create disclosures for each patent application that describe a computer as if it never previously existed. I think we will have to describe the processes for every simple, well known and easily understandable computation.

    What we might want to start doing is actually filing software code. Even those not well versed in patent law on the CAFC would be hard pressed to say disclosure is insufficient if the patent application included functioning code.

    On a side note… let’s not forget that this decision was authored by the same person who initially said that the point of novelty for design patents has to be non-obvious. Either she wants to thoroughly and radically re-write patent law or she doesn’t have a clue.


  • [Avatar for EG]
    April 6, 2012 09:00 am


    Yet another case illustrating why I no longer use “means-plus-function.” While I agree with Newman’s dissent, the Federal Circuit case law generally has so “poisoned” the interpretation of 35 USC 112, 6th paragraph that any further use of MPF in claims is (unfortunately) very risky.

  • [Avatar for step back]
    step back
    April 6, 2012 07:39 am


    This is yet another example of judges gone wild.

    Just as they did in Mayo v. Prome, in Bilski, in Gottchalk v. Benson, they start foaming at the mouth about things they clearly fail to comprehend.

    At first it was kind of cute and laugh-worthy. Oh look. Those yesterday’s-generation judges don’t understand what a “computer” is. Ha ha. But now it’s getting beyond embarrassing and moving into the deadly serious and clear and present harm stage. They are killing America’s Golden Goose, the thing that got us to an age of prosperity; modern technology.

    There is no such thing as a “general purpose” computer and there never was. When the judges spout such total nonsense it just drives me up the wall and out onto the swinging chandelier. I mean, no one has ever built a “computer” (analog or digital and with N number of W bits wide digital words) that can do “generally” anything anyone wants them to do. Not even IBM’s Watson rises to the level of the mythical, never been built, “general” purpose computer. How long are we going to continue to pay undeserved respect to judges who clearly don’t know what the f__ (WTF) they are talking about? I can’t imagine how Newman J. keeps her sanity while day in day out listening to all that anti-scientific jabberwhacky.

  • [Avatar for jon]
    April 6, 2012 06:37 am


    i think its not possible to disclose algorithms in full for large and more complex software. what exactly do these judges want?

Varsity Sponsors

IPWatchdog Events

Patent Portfolio Management Masters™ 2024
June 24 @ 1:00 pm - June 26 @ 2:00 pm EDT
Webinar – Sponsored by LexisNexis
August 22 @ 12:00 pm - 1:00 pm EDT
Women’s IP Forum
August 26 @ 11:00 am - August 27 @ 5:00 pm EDT
IP Solutions, Services & Platforms Expo
September 9 @ 1:00 pm - September 10 @ 2:00 pm EDT

Industry Events

Bayh-Dole Act March In Webinar
May 31 @ 1:00 pm - 2:00 pm EDT

From IPWatchdog