Williamson v. Citrix: En Banc Opinion on § 112, para. 6

By Jason Rantanen

Richard A. Williamson v. Citrix Online, LLC (Fed. Cir. 2015) Download Opinion
Panel: Moore, Linn (author), and Reyna.  Part II.C.1. decided by the court en banc.
Judge Reyna concurred as to the conclusion of indefiniteness, but maintained his dissent as to a claim construction issue decided by the panel.  Judge Newman dissented as to Part II.C.1.

In the original panel opinion in Williamson v. Citrix (discussed here), the majority held that the use of the word “module” does not invoke the means-plus-function language of 35 U.S.C. § 112, para. 6, and thus the claim was not indefinite under the Federal Circuit’s § 112, para. 6 precedent.  In reaching that conclusion, the majority held that because the claim did not use the word “means,” there was a “strong” presumption that § 112, para. 6 does not apply.  Citrix sought en banc review.  (Disclosure: I joined an amicus brief encouraging the court to grant en banc review due to the intra-circuit split on this issue).

This morning the Federal Circuit withdrew the earlier opinion and substituted a new one, with an en banc section addressing the means-plus-function issue.  The en banc court reversed the precedent creating a “strong” presumption,  holding that the standard is “whether the words of the claim are understood by person of ordinary skill in the art to have a sufficiently definite meaning as the name for structure.”  Slip Op. p. 16.  If the words of the claim do not meet that standard, § 112, para. 6 (now § 112(f)) applies.

35 U.S.C. § 112, para. 6 states:

An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding
structure, material, or acts described in the specification and equivalents thereof.

Although there is a presumption based on the presence or absence of the word “means,” that presumption is rebuttable:

In making the assessment of whether the limitation in question is a means-plus-function term subject to the strictures of § 112, para. 6, our cases have emphasized that the essential inquiry is not merely the presence or absence of the word “means” but whether the words of the claim are understood by persons of ordinary skill in the art to have a sufficiently definite meaning as the name for structure.

Id. at 14. Under Federal Circuit precedent from the 1990’s, “the presumption can be overcome and § 112, para. 6 will apply if the challenger demonstrates that the claim term fails to “recite[] sufficiently definite structure” or else recites “function without reciting sufficient structure for performing that function.”  Id. (citing Watts v. XL Sys., Inc., 232 F.3d 877, 880 (Fed. Cir. 2000).

Subsequent cases raised that presumption first to a “strong” one (Lighting World), then to a “strong one that is not readily overcome” (Inventio), and then to an even higher bar: “[w]hen the claim drafter has not signaled his intent to invoke § 112, ¶ 6 by using the term ‘means,’ we are unwilling to apply that provision without a showing that the limitation essentially is devoid of anything that can be construed as structure” (Flo Healthcare Solutions) (emphasis added by court).

The court considered this heightened standard and eliminated it:

Our consideration of this case has led us to conclude that such a heightened burden is unjustified and that we should abandon characterizing as “strong” the presumption that a limitation lacking the word “means” is not subject to § 112, para. 6. That characterization is unwarranted, is uncertain in meaning and application, and has the inappropriate practical effect of placing a thumb on what should otherwise be a balanced analytical scale. It has shifted the balance struck by Congress in passing § 112, para. 6 and has resulted in a proliferation of functional claiming untethered to § 112, para. 6 and free of the strictures set forth in the statute. Henceforth, we will apply the presumption as we have done prior to Lighting World, without requiring any heightened evidentiary showing and expressly overrule the characterization of that presumption as “strong.” We also overrule the strict requirement of “a showing that the limitation essentially is devoid of anything that can be construed as structure.”

The standard is whether the words of the claim are understood by persons of ordinary skill in the art to have a sufficiently definite meaning as the name for structure. Greenberg, 91 F.3d at 1583. When a claim term lacks the word “means,” the presumption can be overcome and § 112, para. 6 will apply if the challenger demonstrates that the claim term fails to “recite sufficiently definite structure” or else recites “function without reciting sufficient structure for performing that function.” Watts, 232 F.3d at 880. The converse presumption remains unaffected: “use of the word ‘means’ creates a presumption that § 112, ¶ 6 applies.” Personalized Media, 161 F.3d at 703.

Applying the pre-Lighting World standard, the court concluded that the term “distributed learning control module” was governed by § 112, para. 6.  That term is part of a larger passage that is “in a format consistent with traditional means-plus-function limitations.  It replaces the term ‘means’ with the term ‘module’ and recites three functions performed by the ‘distributed learning control module.'” Slip Op. at 17.  That term – module – is “a well-known nonce word that can operate as a substitute for ‘means’ in the context of § 112, para 6.”  Id. “Generic terms such as ‘mechanism,’ ‘element,’ ‘device,’ and other nonce words that reflect nothing more than verbal constructs may be used in a claim in a manner that is tantamount to using the word ‘means’ because they ‘typically do not connote sufficiently definite structure’ and therefore may invoke § 112, para. 6.” Id.  And here, “the word ‘module’ does not provide any indication of structure because it sets forth the same black box recitation of structure for providing the same specified function as if the term ‘means’ had been used.”  Id. at 19.  Nor was there any other evidence that indicated that the term recited sufficiently definite structure.

The consequence of concluding that § 112, para. 6 applies was that the court moved to the second step of the means-plus-function construction: looking to the specification for corresponding structure.  Because the specification did not disclose adequate corresponding structure, the claim was indefinite.

Judge Reyna concurred in the conclusion reached by the court en banc, but expressed concern that the court’s approach sidesteps “underlying fundamental issues involving the development of functional claiming law since 1952 when 35 U.S.C. § 112, paragraph 6 was passed.”  Concurrence at 3-4.  Judge Reyna also dissented as to a claim construction issue decided by the panel.

Judge Newman dissented from the en banc ruling in Section II.C.1.  In Judge Newman’ view, the result of the court’s decision is clear: “additional uncertainty of the patent grant, confusion in its interpretation, invitation to litigation, and disincentive to patent based innovation.”

Update: Prof. Risch offers his thoughts on Williamson on the Written Description blog: http://writtendescription.blogspot.com/2015/06/the-past-and-future-of-functional.html

318 thoughts on “Williamson v. Citrix: En Banc Opinion on § 112, para. 6

  1. Stemming from my (unusually detailed and enjoyable) discussion with MaxDrei, a straight up question for Ned Heller and his Belieb system:

    Why did Congress NOT write into the Act of 1952 the actual requirement of structuring claims such that the PON was separate from all else (as in the Jepson claim format)?

    Clearly, this could have been so demanded.

    And yet, the Jepson format is not ONLY (and merely) an optional claim format, but an archaic and little used option?

    I have the distinct feeling that this is (yet another) inconvenient discussion point for our dear Ned Heller, to which there will be met only the screams of silence, as no inte11ectually honest response will be forthcoming.

    1. Good question anon.

      As you know, Westinghouse v. Boyden Power Brake, citing and relying on Corning v. Burden, construed a combination claim, claim 2, to cover the corresponding structure recited in the specification and equivalents thereof in order to preserve its validity. Claim 2 read,

      “In a brake mechanism, the combination of a main air-pipe, an auxiliary reservoir, a brake-cylinder and a triple-valve having a piston, whose preliminary traverse admits air from the auxiliary reservoir to the brake-cylinder, and which by a further traverse admits air directly from the main air-pipe to the brake-cylinder…”

      The Supreme Court held that “[i]n this view, it becomes unnecessary to express an opinion whether the second claim be valid or not, since in the aspect of the case most favorable to the complainants, it is necessary to read into it something which is not found there, or, in the language of complainants’ brief, “to refer back to the specification; not, it is true, for a slavish adoption of the identical instrumentalities therein described, but for the understanding of the essential and substantial features of the means therein illustrated.” In thus reading the specification into the claim, we can adopt no other construction than to consider it as if the auxiliary valve were inserted in the claim in so many words, and then to inquire whether the defendants make use of such valve, or its mechanical equivalent.” Id. at 557-558.

      The practice thereafter developed to use the means-plus-function format for claiming machines, with the understanding that such claims would automatically read into the claims the corresponding structure, and that proof of infringement would require that the defendant use the same structure or an equivalent thereof.

      Then along come Halliburton where the Supreme Court declared the practice illegal when the novelty was in the one element. This threw into question the vast majority of claims to machines which at that time amounted to thousands of issued patents. Kinda of like Williamson, not so?

      As we describe in our article, Hofmann Jr, Rudolph P., and Edward P. Heller III. “Rosetta Stone for the Doctrines of Means-Plus-Function Patent Claims, The.” Rutgers Computer & Tech. LJ 23 (1997): 227, the Milwaukee Patent Law Association proposed a new section to follow section 103 in the new act that read,

      “Where the essence of the novelty of any claim is recited in terms of means or method for performing a specified function without the recital of structure or acts in support thereof, the claim shall not be invalid because of its functionality, but shall be construed to cover the corresponding structure or acts described in the specification and equivalence thereof.”

      This proposal, with slightly altered language, was added to the drafting committee’s proposed amendments to HR 9133 in late 1950. On April 18, 1951, representative Bryson, with the help of house committee counsel L James Harris, first added the new section to HR 3760. At that time it read,

      “An element in a claim for combination may be expressed as a means or step for performing a specified function without the recital of structure or acts in support thereof and such claim shall be construed to cover the corresponding structure, materials or acts described in the specification and abundance thereof.”

      With a minor modification, this is the form that was actually enacted on May 12, 1952, HR 7794.

      But note that the language was intended to enact the rule of Westinghouse, where proof of infringement in court would involve determining whether the accused device was the equivalent of the disclosed corresponding structure. The determining of an equivalent was not part of claim construction. Neither was there any intention that the rule of construction be used to determine patentability in the patent office.

      But the intention was to preserve the validity of those thousands of patents that were using means-plus-function where some of them may have been, like Halliburton, functional-at-the-point-of-novelty. The scope problem was automatically addressed. The indefiniteness problem was put to one side. And it made no difference whether the particular element was at the point of novelty.

      Regarding the problem of validity, this occurs if one does NOT choose to use 112(f) and Halliburton controls. If the claim is effectively a single means claim (old elements behind the colon), then the claim probably is invalid.

      1. Ned,

        Respectfully, that was a far too-long-winded reply that did not answer my question.

        So let me ask again:

        Why did Congress NOT write into the Act of 1952 the actual requirement of structuring claims such that the PON was separate from all else (as in the Jepson claim format)?

        1. (and it is not lost on me that the case you and Stern flubbed so badly also had the court rebuking your rather hallucinogenic alternate-history doozy that 112(f) should somehow not apply in the Office)

          How much did you drop in the 60’s?

    2. I can understand a “point of novelty” in the context of the EPO’s “Problem and Solution Approach” to obviousness, where it is legitimate to identify a “realistic prior art starting point” reference D1 and craft the claim in “characterized by” form to distinguish the invention from the disclosure of D1.

      But in a jurisdiction where identifying a “prior art starting point” reference is seen as impermissible hindsight, and therefore anathema, I don’t see how one can fairly perform a “point of novelty” enquiry. There is nothing new under the sun, and most non-pharma inventions are new combinations of features all of which are in themselves known. So in most cases, there is no “PON”.

      At the EPO, the “characterized by” form is not obligatory because under the Problem and Solution Approach it is often not appropriate. But for those situations where it is “appropriate” it can demonstrate patentable non-obviousness like nothing else can.

      For the 1952 US patent statute to have made it compulsory for the drafter of the claim to pick out and announce the “PON” strikes me as unfair to inventors, unworkable, and liable to bring the US patent system into disrepute.

      anon, below you cast aspersions on the EPO “Guidelines”. These are there to guide the Examiners. Successive editions of the Guidelines draw down the evolving jurisprudence of the EPO Appeal Boards. Only the “established caselaw” of the Boards of Appeal makes it into the Guidelines. For Examiners, the Guidelines is like their 8i6le. If you want to know more about the evolving caselaw of the Boards of Appeal, consult the so-called “White Book”, a resource which retired English Court of Appeal patent judge, Robin Jacob, tells us should be the first thing to consult when it comes to the law of patentability under the EPC.

      1. But in a jurisdiction where identifying a “prior art starting point” reference is seen as impermissible hindsight,

        That’s not the driver.

        Simply put, the law written by Congress is not the law as Ned requires it to be for his PON views.

        It just is not.

        It is that simple.

  2. At 28 I asked a question. I got responses at 28 and 29 in the thread. I’m sorry chaps, but I have no idea how to answer your questions because in my jurisdiction we don’t do “indefinite” and we don’t do 112(f).

    What we have is a system in which you can claim your invention how you like and getting a patent issued on it unless your claim founders during examination at the EPO.

    How might it founder? Well, the EPO is supposed to object, over and over, until satisfied that Applicant has complied with each and every one of the requirements of patentability set out in the EPC.

    What are those requirements? Art 52 eligibility, Art 54 novelty, Art 56 non-obviousness, Art 83 enablement and Art 84 clarity.

    That’s the irreducible minimum. You cannot strike any of those out of any patent law.

    It is also sufficient. What more do you need?

    I suppose Art 84 clarity subsumes 112 indefiniteness.

    The interesting thing about “clarity” of course is that no set of words can ever define an inventive concept with 100% clarity. Every claim ever issued by the EPO is unclear, indefinite, to some extent. The argument under Art 84 is therefore whether the claim is clear enough, whether it is fit for purpose. The purpose is to define the scope of protection. The public has to know whether it infringes or not. The English patent judges have declared that a degree of fuzziness at the boundary of the claim is not enough to invalidate the claim. That works in Europe because the scope of protection is defined by the Protocol on the Interpretation of Art 69 EPC.

    And so, in Europe, whether any given functional language meets the clarity requirement of Art 84 EPC depends decisively on context, and the decision in any one patent application and claim text says nothing about what the decision should be in another claim, another application or another invention.

    Night, Ned, thanks for asking, but I suggest you direct your questions to your colleague US patent attorneys, not me.

    1. I have to say, MaxDrei, that that was one of your best posts ever.

      Amazing then the result when you don’t try to force fit your paradigm into other jurisdictions and show some suitable respect for the differences in law.

      And thank you for the comparative references. As I read Rule 43(1)(a), I wonder if it would be fair to characterize this rule as mandating the Jeppson claim format (which here in the States is but an optional form – and one that has been largely abandoned)….?

      1. Americans often suppose that Jepson = c-i-t. False. The first part of an EPO 2-part claim recites those technical features shared with D1, the art reference that is the starting point for the analysis of the obviousness issue. The extent of the admission as to what is old goes no further than that. There is no admission that anything in this first part of the claim is known otherwise than in D1.

        And, most important of all, only when all the features of the pre-characterizing part of the claim are disclosed in combination in D1, is it proper to recite them in the first part of the claim.

        And so it is, that a c-i-t claim can point up non-obviousness better than most anything else.

        1. Can you explicate?

          I just don’t see a difference, let alone one that makes a pertinent distinction.

          As I read rule 43, there is no mention of putting in the claim any such D1 source or reference to an actual source as you seem to be saying here.

          Perhaps you are not aware of the U.S. jurisprudence concerning the fact that a reference admitted to be prior art by an applicant is deemed prior art even if the applicant was incorrect in her admission.

          1. Two points

            The 2-part form is not obligatory. It is fitting only when it fits to the EPO problem and solution approach to obviousness.

            The Rule includes the words “in combination “.

            It is like learning to ride a bike. Hard to explain. You have to practice. Then you come to wonder, how you could ever have had difficulty, to get the hang of it.

            1. Two counter points:

              I noticed that you have attempted to deftly dodge the point I made above concerning Rule 43 (no actual D1 reference), and that you have not yet explicated a difference or meaningful distinction.

              The rule 43 is not obligatory only in the sense that the claim does NOT contain “those technical features which are necessary for the definition of the claimed subject-matter but which, in combination, form part of the prior art

              To this second counter point I will fully grant that any claims that are fully and completely groundbreaking need no U.S.-Jepson-style two-part treatment. But tell me MaxDrei, what percentage of claims are of this nature versus the percentage that are actually improvement patents?

              You are aware that here in the U.S. the percentage of improvment patents is extremely well North of 50% and the number of claims written in Jepson format is at the polar opposite of well South of 50%.

              I would posit that the U.S. jurisprudence instead simply lacks an equivalent of Rule 43.

              1. Rule 43 prescribes “in combination”. For us in Europe, that means not only just a single source document within the prior art universe but also a single substantiation (embodiment/Examople) within that single doc. Hence “D1”.

                It is instructive to take a typical US-drafted claim and render it into c-i-t form relative to D1. Often, all you have left for the “characterizing part” of the claim is something airy-fairy like “serpentine” or “relatively thin” or “resilient”. the exercise thus provides an early warning system or a wake-up call, for how tough the obviousness issue might turn out to be.

                If you think EPO-PSA is unfair to inventors, cling to the imperative, that basing the analysis on “D1” is out of order unless D1 is truly a “realistic prior art starting point” for the obviousness analysis.

                Yes, of course, the USA lacks anything like Rule 43. As to whether or not that is a “good thing”, opinions differ.

                1. MaxDrei,

                  You are repeating yourself again without explaining the jump to getting where you are at.

                  The phrase “in combination” in rule 43 is used to describe technical features of the claim. Somehow, you have arrived at using this phrase differently – to some item outside of the claim and to a separate prior art document.

                  You have now taken a further “step” and now talk about “rendering” a claim into c-i-t form – but that is quite different than the point that I am focusing on here: that rule 43 requires a certain form from the start for ALL non-breakthrough claims.

                  Also further, the attempted introduction of “if you think… is unfair” is misleading at best, as nowhere have I indicated ANY sense of “fairness” or “unfairness” into the discussion, and any such emotional injection is just NOT helpful in resolving the form of the claim (for all non-breakthrough claims) as dictated by rule 43.

                  Let’s keep a sharp focus uncluttered by emotion, shall we?

                2. Anon, look again at the text of EPC Rule 43(1)(a). It speaks of features in combination in the prior art.

                  It is you who announce that:

                  “The phrase “in combination” in rule 43 is used to describe technical features of the claim.”

                  Evidently, you and I have diffeerent understandings of the meaning of Rule 43 EPC.

                  You write that:

                  “rule 43 requires a certain form from the start for ALL non-breakthrough claims.”

                  I don’t dispute that.

                3. Now that you are “not disputing” the clarification I offered when you kicked up dust to my original question, let me ask it again:

                  …Rule 43(1)(a), I wonder if it would be fair to characterize this rule as mandating the Jepson claim format [for ALL non-breakthrough – or mere improvement claims] (which here in the States is but an optional form – and one that has been largely abandoned)….?

                4. Rule 43 recites “wherever appropriate….”. I don’t know whether that fits with your notion of “mandating” but anyway it is the answer to your question.

                  If an Applicant wants to resist imposition of the 2-part form of claim, the way to do it is to plead that it is “not appropriate”. See the EPO’s MPEP (called “The Guidelines”) on how to argue the point.

                5. I applauded you for your discretion at post 30, but now am perplexed by your unwillingness to take a stand here.

                  You want me instead to explore the items in your backyard…

                  Perhaps then a clue from you: which part of the guidelines shall I explore?

                  link to epo.org

                  I tried key word searching the alphabetical list, but Rule 43(1) does not come up.

                6. …(btw, Part F, pages 44-45 is of little help to the pertinent question here – except mayhap the generous “wiggle room” akin to the “per se” and “as such” language elsewhere in your laws of paragraph 2.3.2.

                  One of the take-always here in the comparative law contrast is that U.S. Law lacks the PON aspect that Ned so craves.

                  I am just not seeing the supposed higher level of clarity in your system, but I am no expert in that system.

                7. On top of the already “whimsical” nature of the “wiggly” writing itself, I find this on the wiki to add yet another layer of “squirm:”

                  According to several decisions of the Boards of Appeal of the EPO, the Guidelines are only general instructions intended to cover normal occurrences. An Examining Division can depart from them provided it acted in accordance with the European Patent Convention (EPC). It is normally desirable for Examining Divisions to act in accordance with the Guidelines, but these are not rules of law, so failure to follow a procedure set out in them is not in itself a substantial procedural violation. Nevertheless, the users of the European patent system may reasonably expect, in accordance with the established “principle of the protection of legitimate expectations” (also referred to as “principle of good faith”), that the Guidelines will be followed by the departments of first instance of the EPO.

                  From: link to en.wikipedia.org

                  I find myself less and less impressed with “the superior system.”

  3. “Why is screw any better than means for helically attaching?”

    Tell me Max what would you use if the term screw did not exist? That is the case for many solutions to problems in mechanical engineering and information processing (CS/EE). So, you need to claim all the solutions to a screw in a claim, but say screw does not exist?

    Problem: Giant set of solutions for problems that those skilled in the art know in light of the specification. The only way to capture it is either 1,000 pages or a functional term.

    That is reality! Reality. Reality. Anyone who does not acknowledge that is intellectually dishonest, unethical, and immoral and should not be practicing law or teaching at a university.

    1. I should add too that I don’t think I have ever used the word module in a claim. You know why? Because I’ve worked for the best law firms and corporations in the world and attorneys identified this term as a possible judicial activism problem area about 10 years ago.

      Module is used just like circuit.

      Scope of enablement is the way to police claim scope. This “definite” structure business belays the reality of the vast number of solutions for problems that can only be specified using functional language like “screw” is really a functional term.

      Actually, screw is very interesting. Joachim Martillo: you like word origins. Try to be honest about the word screw.

      Screw probably referred to only a few solutions back 20,000 years ago. There probably were only a few types in the beginning. And then what happened Lemley? Well, little Mark what happened was many more solutions to the “screw” were invented, but the term kept being used to encompass all the solutions. So, now the term is really functional and encompasses 1,000’s if not 10,000’s of solutions.

      It is hard to even write to people like Lemley because you can feel his oily self misrepresenting facts to try to wiggle out of this.

      That is reality.

      1. A screw identifies a huge class of components.

        A screw for attaching the back panel to a middle enclosure could easily and reasonably be viewed as a means+function limitation that would require consulting the specification for a determination of meaning.

        Back in the late 1990s and early 2000s there was a lot of (patented — I think) innovation

        1) that used helical attachment for modular component assemblies in computer server systems and

        2) that really did not fit into the screw class of components.

        BTW, a bolt + nut also performs helical attachment.

        1. “Screw” is not a static term (for example see US Letters Patent 9050141), and the emphasis here is really just dust kicking over a different topic: the ladders of abstraction.

          Sadly, that topic remains misunderstood and even the mention of the word “abstract” sends people away to clutch their pearls.

        2. JM, that is an interesting point about the back panel, but my guess is the way that would be handled in a patent application is that the term screw (or a substitute) would be used and then the novel features would be claimed over the current understanding of what solutions a screw represents.

          And isn’t what I just wrote the whole point? That is to discuss how to claim real solutions? Real inventions.

          Module does convey structure within the context of a patent claim. This was a witch burning result. Shameful. I think it might be better to just abolish all patents and start over. Let the fed. cir. judges be put out to pasture. They are very good at shoveling sh$t and should try some real world application of what flows from their mind (don’t forget that their information processing occurs in the spirit world with their mind doing all the work and where all the equations are already there waiting for their minds to find them–right out of 1910.)

          1. I’ve already told you that “module” does not connote structure in every art, and it has been mentioned as a nonce word in the MPEP since before this decision. The fact that you didn’t respond to that post and are still bullheadly saying it always connotes structure (because it doesn’t!) is unfortunate.

            1. It is far less “bullheaded” and “unfortunate” than you imagine, given the context to which Night Writer’s discussion points go to.

              It is just not a mystery as to which battle that Night Writer is engaged in, and if there be any propensity to over-apply the word out of context, my dear friend Someone, it would be you.

              To think that this is somehow different than the propaganda battle that has rA ged on and on and on across threads of many subjects is to be obtuse.

              1. Is there any one of the four particular cases that you would rather point to than the non-law-and-binding-only-on-examiners MPEP, Joachim?

                After all – and as I am sure that you are aware – the MPEP is chock full of incorrect statements about what is actually law.

                1. I recommend that you do more than just depend on the MPEP.

                  (btw, I am still waiting for you to tell me which case you think applies)

              2. A circular argument. She’s a witch! Why? Because it says so right here in the paper I wrote that says she is a witch!

                Joachim, you should be ashamed of yourself for being part of the witch tribunals. When you come to the twilight of your life and look back, you will be ashamed. Lemley will never be able to look at his life after the things he has done.

                It harms all our culture to be intellectually dishonest.

            2. Someone: I think I did respond to you. I said as in Phillips it is within the context of one skilled in the art in light of the specification.

              I also said that it is a circular argument to say something — in face as in reality—does not convey structure based on what the PTO says. That is a circular argument whereas elance.com is the reality of how the word is actually used for real purposes by real people skilled in the art.

              I would go as far as saying that this opinion in view of elance.com should be enough to start impeaching judges or abolishing the fed. cir. They clearly have no interest in applying the law. They have turned into a court of Google criminals.

  4. At 23.1.3 a nice question is posed:

    “Why is screw any better than means for helically attaching?”

    My first thought is that I know what a screw is but I don’t know what “helically attaching” means. Why “helical”? What is so special about it that it has to be recited in the claim? What property of “helical” is it, that makes it an invention?

    Think about buttons and cuff links and rings on fingers. Or, come to think of it, any plug and socket combination. If the fit is so tight that they won’t engage without some component of rotation superimposed on the relative translational movement, is that “helical” during the “attaching” step?

    I know for sure I have no “screw”. But do I have between my fingers something in the nature of “means for helically attaching”? And that “for”: is it to be construed as “intended for” or merely “suitable for” (because it might make a difference in the case of a plug and socket)? Methinks M+F at the point of novelty is often problematic.

    1. Max, invention is a picture frame with four frames having shaped ends that might be arranged in an overlapping fashion, and which are held together with screws or the like, but angled at 45 degrees from any frame.

      I claim:

      1. 1. A picture frame, comprising:

      four frame members, each having shaped ends such that the shaped ends may be arranged in an overlapping fashion when the four frame members are arranged in a square; wherein the four frame members are attached together at each corner when arranged in a square with an elongated member angled at 45° with respect to any frame, each elongated member having structure disposed upon its external surface in substantial contact with a respective frame member such that only rotational movement of the elongated member is permitted, the structure further adapted to cause the elongated member to insert into or to withdraw from a frame member upon rotation, depending on the direction of rotation.

      Let us presume that picture frames with screws are old, that the overlapping ends are old, and that the point of novelty is the angle of the elongated member with respect to the frames. What, if anything, is indefinite about this claim? Secondly, why should 112(f) be employed?

      1. Well Max, let me say that I described a screw functionally. But that is known structure. So why in the world would I be limited in any way to a particular screw disclosed in the spec when then invention has nothing to do with the particulars?

        This is why the Federal Circuit jurisprudence in this area is and has been so wrongheaded. Functionally claiming old structure has never been a vice. Neither has functionally claiming elements of a combination where the combination defines new functionality between the elements.

        The singular vice that is prohibited is claiming the Invention functionally — in terms of a result.

        But, by not making distinctions, by ignoring the Supreme Court once again, the Federal Circuit continues to make a hash of patent law in the US.

        1. What is so wrong-headed is Ned’s view that ANY court – even and especially the Supreme Court has authority to write patent law (through the mechanism of common law development).

          The days of shared authority (by Congress to the judicial branch) to develop the meaning of the word “invention” – Ned’s ever tenuous tie to “the singular vice” ended in the Act of 1952.

          At IPKAT there is a thread “The Supreme Court in Starbucks (HK); on adjectives and reluctance to embrace change” on which there is an Anonymous comment addressing a bit of common law judicial law writing dealing with passing off and goodwill (topics which may very well legitimately fall under the bonafide legal authority of judges to write common law).

          There is in particular one comment in the decision for which the IPKat author expressed some confusion, and for which the first Anonymous poster deftly answered to dispel that confusion. That comment from the foreign jurisdiction decision was: “ Nonetheless, we should bear in mind that changing the common law sometimes risks undermining legal certainty, both because a change in itself can sometimes generate uncertainty and because change can sometimes lead to other actual or suggested consequential changes

          Perhaps more appropriately here on this blog (and for this jurisdiction), and according to the well-known precept that our patent law is statutory law and not common law, here is my rejoinder to that conversation:
          link to img1.blogblog.com
          Anonymous @12:57,

          Nice job. You explain well how care with common law development should unfold.

          This brings to my (American) mind, two tangential thoughts that might be a divergence from the intended direction of this thread. Please feel free to not take this different direction as anything but a comment on the US state of patent law.

          1) The notion of writing with care because of the consequences of the change in law can not be stressed enough. Our own Supreme Court is quick to denigrate patent attorneys as scriviners gaming the rule of law as written by Congress and use this accusation as justification for its insertion into the law new caveats that are said to be “implicit” in the words of Congress, or of recent vintage are nakedly explicit additions. And yet, our Supreme Court lacks due care with its own scrivining, leaving terms undefined, and sweeping in broad “powers” for the judicial branch below that have no basis in the words of Congress, because, at least in the US, patent law was expressly set up per point 2), below.

          2) In the US, we have a few strict rules of law to abide and limit the application of governmental power. We have a rule of law of limited federal authority, of requiring an actual case or controversy (no “opinion pieces” at the Supreme Court level), and a separation of powers doctrine that was meant to keep a balance and not let ANY one branch of our government (including the judicial branch) garner power in excess of the power explicitly doled out by the rest of our constitution. There are many points that flow from these foundational conditions, but the point explicitly here that your post reminds me of is that here in the States, our patent law is explicitly NOT a common law vehicle and thus, there is NO place** for common law change to patent law.

          **I would place a caveat here in that our system does allow one branch to share limited authority with another branch with certain conditions met. Those conditions include a clear intent and limit of the sharing. For example, in our patent law, Congress has shared common law authority with the courts in setting remedies for patent infringement. Further, prior to 1952, our Congress had shared the power with the judicial branch to use common law evolution to set the definition of the word “invention.” Of course, our Congress reacted to a Court that had turned excessively anti-patent and removed this authority in 1952.

          1. What is really weird anon is a you don’t know a thing about the wall of the court system in our government. You act as if the courts do not interpret the law. You act as if Marbury v Madison did not exist. You act as if you lived in some country other than the United States.

            Regarding the ’52 act, it was Congress expressed intention to codify the law and not to change it. Clearly it was the objective of Judge Rich and certain members of the patent bar, particularly of the Milwaukee Patent Law Association which originated what is now 112(f), to overrule Halliburton. But that requires one to know the holding of Halliburton in order to know what they were trying to do and how to interpret 112(f).

            Your interpretation of 112(f) is that it applies to any functionally expressed element in the claim. Because if that is not what you are arguing, I have no idea why you argue against me.

            But your interpretation is absolutely totally absurd given that there intention was to overturn Halliburton.

            Your complete insistence that we should not look at Halliburton to determine what the intent of congress was, is absurd. It is almost as if you are trying to find buried treasure that is marked on a map 25 paces from a certain landmark, while insisting we cannot first look for the landmark. That would be so ridiculous that it would have to be a joke.

            It is a singular vice of both you and of the Federal Circuit not to look at the holding of Halliburton to determine what it was that Congress was attempting to do.

            1. Interpretation is just not the same as writing Ned.

              Clearly, your view crosses the line and it is not I that lacks a proper viewpoint, but as it is you.

        2. Post with possible double link stuck in the filter, so this might end up being duplicated…

          What is so wrong-headed is Ned’s view that ANY court – even and especially the Supreme Court has authority to write patent law (through the mechanism of common law development).

          The days of shared authority (by Congress to the judicial branch) to develop the meaning of the word “invention” – Ned’s ever tenuous tie to “the singular vice” ended in the Act of 1952.

          At IPKAT there is a thread “The Supreme Court in Starbucks (HK); on adjectives and reluctance to embrace change” on which there is an Anonymous comment addressing a bit of common law judicial law writing dealing with passing off and goodwill (topics which may very well legitimately fall under the bonafide legal authority of judges to write common law).

          There is in particular one comment in the decision for which the IPKat author expressed some confusion, and for which the first Anonymous poster deftly answered to dispel that confusion. That comment from the foreign jurisdiction decision was: “ Nonetheless, we should bear in mind that changing the common law sometimes risks undermining legal certainty, both because a change in itself can sometimes generate uncertainty and because change can sometimes lead to other actual or suggested consequential changes”

          Perhaps more appropriately here on this blog (and for this jurisdiction), and according to the well-known precept that our patent law is statutory law and not common law, here is my rejoinder to that conversation:

        3. link to img1.blogblog.com
          Anonymous @12:57,

          Nice job. You explain well how care with common law development should unfold.

          This brings to my (American) mind, two tangential thoughts that might be a divergence from the intended direction of this thread. Please feel free to not take this different direction as anything but a comment on the US state of patent law.

          1) The notion of writing with care because of the consequences of the change in law can not be stressed enough. Our own Supreme Court is quick to denigrate patent attorneys as scriviners gaming the rule of law as written by Congress and use this accusation as justification for its insertion into the law new caveats that are said to be “implicit” in the words of Congress, or of recent vintage are nAk edly explicit additions. And yet, our Supreme Court lacks due care with its own scrivining, leaving terms undefined, and sweeping in broad “powers” for the judicial branch below that have no basis in the words of Congress, because, at least in the US, patent law was expressly set up per point 2), below.

          2) In the US, we have a few strict rules of law to abide and limit the application of governmental power. We have a rule of law of limited federal authority, of requiring an actual case or controversy (no “opinion pieces” at the Supreme Court level), and a separation of powers doctrine that was meant to keep a balance and not let ANY one branch of our government (including the judicial branch) garner power in excess of the power explicitly doled out by the rest of our constitution. There are many points that flow from these foundational conditions, but the point explicitly here that your post reminds me of is that here in the States, our patent law is explicitly NOT a common law vehicle and thus, there is NO place** for common law change to patent law.

          **I would place a caveat here in that our system does allow one branch to share limited authority with another branch with certain conditions met. Those conditions include a clear intent and limit of the sharing. For example, in our patent law, Congress has shared common law authority with the courts in setting remedies for patent infringement. Further, prior to 1952, our Congress had shared the power with the judicial branch to use common law evolution to set the definition of the word “invention.” Of course, our Congress reacted to a Court that had turned excessively anti-patent and removed this authority in 1952.

  5. At the end of every patent application, we need to add the following to fight the out-of-control Federal Circuit.

    Unless the claims specifically use “means for” or “step for” in a claim or a claim element, applicant specifically intends that the claim or claim element as the case not be construed under 35 USC § 112(f).

    1. This is the same as a comment I just posted on Gene’s site.

      The Federal Circuit and the patent office are both pretending that cannot claim structure functionally. There is no statute and there is no case anywhere that I know of that is actually held this. Halliburton as clarified by Faulkner v. Gibbs held that a claim functional at “the exact point of novelty” is indefinite. That leaves open functionally claiming structure which is not novel, or which is not at the exact point of novelty. Indeed, Faulkner involved functionally claimed elements were the novelty was in the combination, rather than a particular element. The vice that the Supreme Court is seeking to prohibit is not functionally claiming structure but functionally claiming the invention. If a combination claim is equivalent to a single means claim because all the other elements are old and there is no new functional relationship between the one element that is new and the other elements which are old, then the claim is effectively a single means claim.

      We should also note that §112(f) is permissive. It is by its own terms, optional. It is invoked by the applicant. If a claim that is functional at the point of novelty does not use the statute however, the claim may be invalid as functionally claiming the invention if the applicant does not invoke §112(f).

      So what I am going to do henceforth is add a statement to the specification that states that the applicant does not intend, and specifically intends not to have his claims construed under §112(f) unless the claim at least uses “means for” or “step for.” This would end any debate, any legitimate debate that is, whether the applicant is intending to invoke the “optional” benefit of §112(f). A court would then have the find some case support for the proposition that one cannot claim structure functionally in order to invalidate a claim that is not functional at the point of novelty. That case support does not exist.

      This, of course, does not eliminate the requirement that the claim be supported under 112(a). But when one is claiming known structure, or software, that is fully enabled by a description of what it does, then one should have no problem under 112(a).

      1. Ned,

        You are writing phantom law based on phantom history in your attempts to have the pre-1952 Supreme Court cases survive the changes in law as enacted by Congress in 1952.

      2. Help me out Ned, where are the words of Congress that limit the law as you would limit to this phantom ONLY “point of novelty”…?

        My version of 112(f) merely states “an element” and does NOT read as your twisted version attempts to read.

          1. I may not have phrased that precisely enough, but the meaning is clear Ned – the words of Congress do not support your view, and it is you that is seeking to twist and re-write history in order to attempt to change law into something that it is not. It is you that seeks to deny the fact that Congress was prompted to act because the Court had become an activist court, and an anti-patent activist court at that.

            Just because you clench tight your eyes, history will not change or go away.

      3. Please, please, please, address the point as explicated by Prof. Crouch’s coined term of Vast Middle Ground (keeping in mind the actual words used by Congress).

      4. To me, Ned, you make good sense, with your “Unless….” caveat to be added to the specification. I do not see how it could fail to be effective.

        And oh so simple!

        Which (sorry to say) is probably a strong indication that it will fail to achieve the purpose you have for it. It is like in an exam, when the answer to the question strikes one as simple. All that means is, you haven’t understood the question fully.

        What seems to me eminently sensible, from my vantage point somewhere else in the world, often fails to come to fruition because of complicating local factors that I don’t understand.

        1. I had a discussion with a professor about this elsewhere. He construes the statute such that any attempt to claim structure functionally is covered. I must ask him or anyone why congress would enact such a statute if the holding of Halliburton/Faulkner limited itself to claims functional at the point of novelty, leaving old structure and new combinations open to functional claiming?

          1. I must ask him if“….

            Ned is like the old man driving the wrong way down a one way street yelling at all the other drivers that they are driving badly.

            If Congress wanted to address this “Point of Novelty” phantom, why did they not say so?

            They did not.

            Even the words they used in the congressional record say something else.

            Sorry Ned, but you just do not have a case for your attempted carryover of pre-1952 Supreme Court common law cases into the post-1952 world.

            Repeat after me: the point of novelty is the claim itself, read as a whole. This then takes care of the concern of single means claims and this then still controls the mere aggregation type of claims that make a mockery out of Malcolm’s pet theory and your over-preoccupation with the use of terms of function at any singular point within a claim.

            Your contrivances and attempted “spin” of the plain words of Congress are just not needed nor are they appropriate.

            1. Anon, the singular problem you have is figuring out what Congress intended to do. I have written the article on that – gone back to original materials that were provided to me and Rudy Hoffman by Judge Rich himself.

              1. And you have clearly let your Belieb system twist your views.

                Even more sad given the wasted opportunity of being close to the sources.

                More the pity for you Ned than anything else, more the shame too that you are so wrong on the law, so wedded to your windmill chases.

    2. An applicant can be his own lexicographer, but if he intends to reject the ordinary or epistemic definition of a term, doesn’t he have to provide his own meaning?

      I have to suspect that in the deciding of Williamson someone just consulted the online definition of module in computer science.

      link to thefreedictionary.com

      Here is definition 4 from the very first cluster of definitions.

      4. Computers A portion of a program that carries out a specific function and may be used alone or combined with other modules of the same program.

      The above seems to describe “module” as a “software means”.

      1. Joachim, there may be another problem here when module is construed as being software itself, as opposed to a machine, process, manufacture… Software is non statutory.

  6. Since this cuts to the chase, it is worth re-posting on top:

    Asked and answered Ned.

    Quite in fact, it was in a conversation with you that I shared the direct words of Congress, the congressional record and the direct implication voiced by none other than Federico. Upon that point in our conversation, you abruptly abandoned the dialogue and refused to integrate the points in your mantra. In other words, and as you have repeatedly done when I have provided counter points that are simply too inconvenient for you, you ran away.

    You now play the Internet style game of pretending the conversation never took place. Stop the silly shouting down games Ned. You may not like the fact that ALL combination claims have been opened up to the words of function in claims and you may not like the fact that old items can be claimed anew if so claimed in combinations, but those are the words that Congress (and not your dust-kicking accusation of Judge Rich) enacted.

    To go one step further, and to highlight the silliness of Malcolm’s pet 101 theory, and the “over” concern here with your own “anti-combination” mindset, do keep in mind that we still have the doctrine that mere aggregations – a listing of elements that have no functional relationship to each other – fail the claim as a whole reading.

    This is true whether or not any element in a claim sounds in a mental step or sounds in terms of being described with language of function.

    This too avoids the goal-post moving and dust kicking attempts to move discussions of those claims in the Vast Middle Ground to discussions of PURE function or claims TOTALLY in the mind.

    If we keep the focus of the discussion clear and if we avoid the purposeful obfuscations, it is abundantly clear that the likes of you and Malcolm want a different law to be enacted from the one that Congress actually enacted in 1952.

    1. How and why do you keep going on, anon? You do realize you’re not going to change Ned’s (or MM’s) thoughts on these issues, right?

      Now, I appreciate what you’re doing, as most of us don’t have time to spar with what is a tremendous barrage of single-minded rhetoric. But I wonder how you continue with the futility of it all.

      1. All tyranny needs to gain a foothold is for men of good conscience to remain silent.

        The fact that it is the same few inconvenient facts and points of law that Ned and Malcolm and the echoes refuse to integrate into the dialogue makes it easy to point out their fallacies.

        They are counting on the act of repetition of their propaganda for it to have its desired effect of “appearing true” – just look at the new poster Genghis, and his “appreciation” of Ned’s twisted history…

      2. PatentBob, anon does not discuss. He simply states that we are wrong without ever discussing the issues. When challenged, he will simply say that he has previously set forth his position.

            1. Ned – you confuse succinctness with “not discuss,” which is just plain wrong – and spawns the typical empty ad hominem line of “med” from 6, who has nothing intelligent to add to the conversation.

    2. Anon it will never end. They are here merely to push points. Notice they never respond to elance which is pretty much irrefutable proof that module connotes structure and is not a nonce word to those skilled in the art.

      We are getting deeper and deeper into witch burning. (And note all those that make fun of me for predicting bad events that I predicted this. I said that Lemley’s shameful paper on functional claiming was a road map to judicial activism and that the central fallacy (among many) is that it does not interpret the claims as one skilled in the art would in light of specification.

      Just bizarre. Alice is contradiction to Graham and this one is in contradiction to Phillips.

      1. If I took a contract via elance to write a new regular expression matching module for node.js, how would that prove that “module” implies meaningful structure in any patent law sense?

        In fact, such a regular expression matching module by Gottschalk v. Benson would certainly not constitute any meaningful structure in patent law.

        1. Your argument is circular–again.

          The issue is: does module convey structure to one skilled in the art in light of the specification. That is the issue. The fed. cir. just said no it does not convey structure.

          Clearly module is not a nonce word or it would not be used to specify work on elance 1000’s of times. Module is a word like circuit that does convey structure to one skilled in the art.

          (Oddly you actually agree with me on that point, but try to wiggle out with a circular argument. The fact that the law held it does not convey structure does not mean that in reality it does not. It means the judiciary is farther and farther off the reality path.)

          1. I could probably come up with at least a half-dozen ways to implement a regex module. At the very least, the structure is indefinite, but I would argue that that the indefiniteness is stronger because the the regex module would be making purely algebraic transforms on string inputs to produce algebraic or boolean outputs. It is practically the the definition of boolean or algebraic indefiniteness in patent law.

            I am not saying that developing the regex module would not be considerable amount of work. I am only saying that listing a regex module in a recitation of elements in a claim would serve as a purely functional limitation.

            Benson showed awareness of the functional aspect in his attempt to patent a BCD to binary converter module. As Ned points out functional claiming at the point of novelty does not work. Hence Benson described the conversion algorithm in gory detail, but ultimately Benson claimed only a mathematical algorithm, and such a claim is 101 ineligible.

          2. >The issue is: does module convey structure to one skilled in the art in light of the specification.

            Structure? Perhaps.

            Definite structure? Definitely not.

          3. Night Writer says: “Clearly module is not a nonce word or it would not be used to specify work on elance 1000’s of times. Module is a word like circuit that does convey structure to one skilled in the art.”

            This is not true. “Module for” does not convey structure in every art. Additionally, the MPEP has indicated “module for” as a non-structural placeholder that may invoke 112(f) since well before this decision came out.

            As with everything patent related, the real answer is “it depends”. But to say that “module for” always denotes structure is incorrect.

      1. Quite frankly Ned, your line of “I just want” is just B$. Your 4:50 pm post below (trailing 23.1.3.1.5.1.1) is mere dust-kicking.

        Note that you DO NOT actually answer my post at EITHER 26 or 8:22 am following 21.1.2.4.1.1…

        You “just want” to continue to think that statutory law is subservient to judicial common law in spite of the fact that the constitution has clearly delineated the proper lines of authority.

        You want to pretend that 1952 did not happen.

        It did.

        As to the Haliburton case itself (see link to supreme.justia.com
        yes I have read it, and yes there is a holding and yes, that holding is quite immaterial given what has happened subsequent to that case.

        Full circle then on your little Merry Go Round. Did you want to throw in more denigration, wrongly aimed, at Judge Rich? The Judge who fully understood the dicta games of the High Court and simply ignored dicta because he knew better? You do realize that dicta is not controlling, right? After all, you were the one in the midst of our Bilski discussion that provided the link showing the danger of confusing dicta and holding, or have you conveniently forgotten that episode as well?

        We have been over what Federico said – including the portion where he acknowledge that Congress opened up the use of terms of function beyond 112(f) – wherever combination claims may be found. Your “response” is merely ad hominem “Anti-Rich” name calling on that point, accusing me of placing form over substance. Excuse me Mr. Heller, but what “super-secret” “Gist” effect-type reading of “combination claim” did you have in mind other than the straight up reading of what you would so eagerly denigrate? At this point, may I yet again remind you that the doctrine against mere aggregation is very much still play and that the claim as a whole reading is fully capable of handling such things (and is precisely why your “over-concern” and Malcolm’s silly pet theory both lose their credibility in the ongoing discussion). So once again, your 4:50 pm post below (trailing 23.1.3.1.5.1.1) is mere dust-kicking.

        1. What is the holding of Halliburton, and how did Faulkner modify it?

          I think the statute is clear that when a combination is claimed, MPF or SPF may be employed at the choice of the applicant. It is entirely optional.

          I suggest now that we all “clarify” our choice in the specification so that there is no ambiguity in trying to determine what OUR is. This serves to clarify the claims.

          1. The onus is on you Ned – your canard of “Point of Novelty” simply did not survive the words of Congress in 1952.

            Trying to quote pre-1952 Supreme Court cases is simply inapposite to this fundamental fact. You need something else to make your case.

            To date, you have made NO case because there is NO post-1952 anything to support your case.

  7. I sat most of this thread out. Did the sky fall or did it not fall? Ned, anon, NWPA, MM, final analysis of the situation or thoughts?

    1. The sky has darkened a bit more for the logic patenting crowd. The CAFC says you can’t use nonce words like “module” to perform an end-around 112P6. If the applicant here had included a corresponding “algorithm” in the specification he/she might have lived another day, or maybe longer.

      Oh, and this is Obama’s fault.

      1. Which means you can write claims for software inventions in “means for” language and say the means is a general purpose processor, and as long as you include an algorithm for each means, you’re OK. (Assuming you meet the Alice rules, assuming you can determine what those are.)

    2. Not fall – in the end this is but about the presumption. The ability to use terms of function already existed outside of 112(f) by design of Congress and this does not change that.

      All this does is create more Patent Profanity.

    3. Not fall. Not even darkened. “Processor” has been held by the CAFC not to be a “nonce” term; use it instead of “module” along with appropriate specification drafting (do your own homework) and this case means zilch in terms of limiting you, against your will, to 112 sixth.

    4. 6, it complicates claiming old structure in claims. One avoided MPF for a good reason.

      Also, from the point of view of “certainty” of the scope of claims, this case is a potential disaster. A claim that was intentionally written without the magic words is now being interpreted in a manner dramatically different. The ordinary reader of the claim, the one of ordinary skill in the art, will have no clue as to the real scope of the claim. How is this at all consistent with the requirements of 112(b)?

      The statute is clear on its face. Use MPF or Step plus function. Otherwise, the statute does not apply, period.

      This needs review by the boss court.

    5. Module is like circuit. It connotes structure. Elance.com proves that.

      This is another opinion where the judicial activist on the Fed. Cir. have given more case law that has nothing to do with reality and specifically attempts to ignore how one skilled in the art would view the claim term in light of the specification. Shameful. I think at this point it is time to start calling for the disbandment of the fed. cir. The streets around their building need cleaning and maybe those bums could be put to use picking up garbage.

  8. link to elance.com

    For those that pretend to have integrity on this blog, please do the following “site:elance.com module.” And, then explain how it is that the word module is used 1000’s of times to describe real programming jobs and yet has no meaning? ==><== (For the Google judges that means a contradiction. You have probably never seen that symbol before.)

    Just another outrage from the Fed. Cir. that completely ignores reality to get their judicial activism result. I think we need to start a discussion about how to abolish the Fed. Cir. which is now stacked with anti-patent science illiterate judges appointed at the behest of Google.

      1. Not sure I understand your question Fish Sticks. The issue is does module connote structure or is it a nonce word.

    1. Not sure I understand. Isn’t a “module” in programming a piece/chunk/portion/file of code / source code?

      1. According to IBM’s definition, see elsewhere in this thread, module is in fact software.

        But, software is non statutory.

        1. Simple set theory explanation Ned (you know, the one you ran away from repeatedly even as I BEGGED you to discuss).

          The set theory that explained that the hands of man for software means that software is indeed a manufacture….

          Other hands of man manufactures that involve writing include magic hat bands and measurement markings on the outside of glass containers…

          “We do not so hold”

          Oops.

        2. …and as you seem to need reminding, software is not the thought of software and a certain aspect of software earns copyright protection, and copyright protection only inures to those things fixed in a tangible medium.

          Shall I quote you from the earliet API thread on the functionality of software? This wondrous manufacture of the hands of man that is not the same thing as the thought of software (or the execution of the manufacture, which itself is a different thing)…?

          Let’s try to be inte11ectually honest here Ned and STOP obfuscating the discussion by pretending that software is not a manufacture and machine component in and of itself.

  9. So, as I understand the opinion, all functional language is now means-plus-function?

    This certainly takes the effort out of claim drafting. The broadest possible scope for computer inventions is now afforded by reciting a combination comprising (a) and (b) means for

    1. Atsby, not all functional language. Functional language can be used to explain what a claimed structure is doing. But that structure must be definite without the function — in other words, the functional explanation should be removable from the claim without affecting patentability.

      I would suggest that the novelty of the claim cannot be in the functional explanation for somewhat obvious reasons.

      1. But that structure must be definite without the function — in other words, the functional explanation should be removable from the claim without affecting patentability.

        The first half of that is right, Ned. The second half is potentially misleading. Some functional language is nothing more than an “explanation.” But functional language can put further structural limitations on structures that are otherwise definite. That kind does not raise definiteness issues, and removing it will broaden the claim, which means that patentability might well be affected.

        1. DanH, prior to the present case, you are right. I think the balance has shifted. The structure itself must be known. If the functional definition has the novel feature, one is selecting or further defining known structure. Thus we are claiming structure functionally.

          1. What if there is no novel “structural feature,” but rather a novel and nonobvious combination of prior art “structural features” that function together in a novel and nonobvious manner to provide a wholly unexpected result? A method claim would perhaps be appropriate, but the assembly itself is novel and nonobvious. Must the specification now list each and every permutation of constituent prior art structural features?

            I agree that trying to the claim “the world” functionally raises over-breadth problems when all that was invented was a particular apparatus, but with the limitations of written language and all that, there should be some functional latitude.

      2. Except: reality. Reality is that the functional language is used to express solutions that are disclosed in the spec and know to those skilled in the art.

        Reality is that this is judicial activism in its worst form further limiting 101 past Alice based on nothing more than the stacked Google judges’ quid pro quo of getting appointed.

      3. Ned –

        If the claim recites a screw is that specific enough?

        What about embodiments that use wood screws verses machine screws. What about 10 x 32 screws verses 8 x 24? Metric thread vs. English? Inch long vs. two inch long.

        What if the specification didn’t specify every (or even any) particular kind, size, thread count, plating, material or screw head configuration?

        Why is screw any better than means for helically attaching?

        1. Les, I think Ned and MM are similar in that they really don’t care what you say or what your arguments are. They simply plow forward with their own commentary, espousing their theories without any regard to counter theories.

          So, if you point out logical inconsistencies in Ned’s (or MM’s) theories, it’s going to go nowhere. This is too bad, as this is a complex situation, and one for which I don’t know what the rules should be. Consider a memory controller, for instance. At one time, you would claim the hardware. But that was when hardware was simple. Now, memory controllers could easily have hundreds or thousands of gates or tens of thousands of gates, and it’s impossible to claim the hardware. You have to claim it functionally.

          I can see the logic behind not allowing claiming functionality (eg, at the point of novelty), but at the same time, many things have to be claimed via their functionality. I can also see where over-broad claim language, such as the slew of business methods being invalidated now under Alice, has lead to problems. Note these are typically claimed functionally.

          It’s a complex issue with no easy solution.

          1. Patentbob: I have written patent applications for the three largest chip manufacturers. I have heard many times by Ph.D.’s: you can push in or out of the circuitry almost any functionality.

            But, I fear that the Google judges on the Fed. Cir. are just like Ned and MM and have been placed on the Fed. Cir. for a purpose and don’t care what the arguments or reality is. Thanks Obama.

          2. MM: memory controllers could easily have hundreds or thousands of gates or tens of thousands of gates, and it’s impossible to claim the hardware.

            Completely false.

            “Biomolecules could easily have hundreds or thousands or tens of thousands of atoms and it’s impossible to describe them all.”

            Been there, done that.

          3. PB They simply plow forward with their own commentary, espousing their theories without any regard to counter theories.

            Keep telling yourself that, PB.

            Meanwhile, I’ll continue to “regard” and respond to Les’ comments (and yours) regularly as I have done so for years, all the way back to when Les would argue that 101 authorizes the claiming of purely mental processes because … “processes.”

            logical inconsistencies

            Again: this entire line of jurisprudence is being driven by the patenting of “innovations” in information processing logic, particularly where that logic is “embodied” in an old configurable machine that was designed to process information (indeed, that is the machine’s only purpose).

            If you are concerned about “logical inconsistency” then you should be concerned about the legal fiction underlying nearly all of the recent jurisprudence, i.e., the fiction that algorithms have patentable structure.

            1. “as I have done so”

              So….

              That means more of the same “ecosystem” that we** have witnessed for nine years and running.

              ** “we” being a collective term, seeing as Malcolm has been his toxic self long before I ever posted here.

              Prof. Crouch must be sooo proud.

              /off sardonic bemusement

          4. PatentBob,

            Contrast the word “monologue” with the word “dialogue.”

            The relentless propaganda that you see from Ned and Malcolm and the Echoes wants nothing to do with actual dialogue.

          5. See, this is a actually a good reason for (limited) functional claiming, because the elements cited aren’t the point of novelty, and have corresponding physical structure (e.g. an address conversion module has a definite physical structure made of muxes/demuxes; an adder has a definite structure made up of and & or gates).

            However, in the same memory patents you talk about, I often see the point of novelty claimed functionally, which is problematic.

            Also, there’s a vast difference between this and “module configured to”.

            1. What is the “point of novelty?” No joke, how do you know what the point of novelty is, a priori, unless you’re the PHOSITA — a mythical omniscient superhero?

              1. Peanut, well, the SC is concerned about single means claims. These should be easy to spot — an improved element in an old combination claimed functionally. All the old elements could be placed in Jepson formation with the only new or improved element a single means after.

                Only, MPF format is not used.

                1. The canard comes home to roost – clearly it is the single means claims and ONLY the single means claims for which PON might come into play as it should be understood (by anyone willing to be inte11ectually honest) that the “point” of novelty in combination claims – no matter if any single element of the combination is itself novel or not – is the claim itself, as a whole.

                  Ned’s predilection here with the optional claim format of Jepson claim mirrors the fallacy of Malcolm who also attempts to paint a single optional claim format as a “one true legal-only claim format.

                  But this type of “mandatory” treatment is NOT what the law actually provides, and is a clear indication that the Beleib systems of Ned and Malcolm would require Congress to have written different law.

                2. Stop it anon. You sound just like Judge Rich, form over substance.

                  Put two elements in a claim and presto, you have a combination claim. No looking at the substances.

                  I still doubt that you have ever read Halliburton or Faulkner.

      4. All this yapping of yours Ned sounds so good in theory, but it does not fit the reality of modern information processing inventions.

        The fact is that module is used like circuit. That is reality and all the judicial activism and anti-patent judges in the world can’t change reality. They can only change the law. If the Google judges want to burn information processing inventions as witches, there doesn’t appear to be much we can do about it. But, know there are plenty of us that know exactly what they are doing and history will not be kind to the likes of Taranto.

        1. If the Google judges want to burn information processing inventions as witches, there doesn’t appear to be much we can do about it.

          Sure there is.

          Rewrite the patent statute to expressly allow the patenting of “information processing logic” and maybe think about putting some really specific guidance in there about (1) exactly what kind of logic is eligible; (2) what kind of disclosure is required; (3) what needs to appear in the claims; (4) how claims to “new” logic are to be examined in view of prior art disclosures of “old” logic.

          Good luck. This should be straightforward given the overwhelming support out there for such claims, particularly among software writers.

            1. He knows it too, Night, as he has volunteered such admissions against interests (and his overall view of the Act of 1952 simply confirms his own admissions)

  10. If you have no novel physical structure in your computer-implemented claim (recited directly or via 112P6) or if you have no novel physical structure (e.g., a new machine described in objective physical terms) in your information processing method, you are guaranteed to have either a serious 101 or 112 problem, or both (not to mention the 102 and 103 issues that rear up when the breadth of claims is opened by functional language including non-limiting abstract elements).

    One more time: that’s guaranteed.

    The squeeze play was recognized quite a while back so none of this should come as any surprise to applicants and patentees relying heavily on functional language to protect “innovations” in information and logic. It’s because of these applicants and patentees that are seeing these reactions from the courts and it’s going to continue. The courts will do whatever they can to limit the damage to innovators in the arts which the system was designed to protect but the endgame is already clear.

    Algorithsm aren’t structure. Configurable black boxes that carry out logical operations aren’t eligible for protection with patents, regardless of whether they are claimed as black boxes or methods of “using” the black boxes.

    1. Algorithms are not structure. They define a sequence of steps. At best from a patent law perspective, they potentially form part of a larger process or apparatus.

      Claiming the operation of a machine without more used to be per se non statutory or indefinite until Rich overturned the long accepted CCPA doctrine to the contrary in Tarczy-Hornock.

      1. Ned –

        Algorithms are structure. Please see the bolded definitions. In particular see: 5 : the aggregate of elements of an entity in their relationships to each other . For “elements” read steps.

        Full Definition of STRUCTURE
        1
        : the action of building : construction
        2
        a : something (as a building) that is constructed
        b : something arranged in a definite pattern of organization
        3
        : manner of construction : makeup
        4
        a : the arrangement of particles or parts in a substance or body
        b : organization of parts as dominated by the general character of the whole
        c : coherent form or organization
        5
        : the aggregate of elements of an entity in their relationships to each other

        link to merriam-webster.com

        1. In particular see: 5 : the aggregate of elements of an entity in their relationships to each other

          LOL. Yes, let’s put extraordinary weight on the fifth (out of five) definitions … the same definition that covers the “structure” of fairy tales.

          1. Not sure how a choice of a defintion makes that selection “extraordinary.”

            It’s not like he is trying to dissemble upon a point of law like making one optional claim format be (somehow) the only valid and “legal” claim format like a certain someone does ad nauseum.

        2. Les, that processes have order does not mean that computers have structure. The structure we are talking about here is the structure of a machine.

          That is why the question of the weight of 1 becomes highly relevant to this conversation. How many 1’s and 0’s does it take to fill a breadbox?

          If I load a computer with all 1’s, is it heavier than when I load it with all ‘0s?

          1. The weight of “1” has ZERO relevance Ned.

            Stop kicking up dust.

            What is the difference in weight between three resistors in series and three resistors in parallel?

            What is the structure of Morse’s allowed “space?”

            Do you really believe that software is exactly a “1?”

            Nothing but rhetorical games and dissembling from you Ned.

            Repeat after me: software is a manufacture by the hand of man and a machine component. Software is equivalent to hardware and to firmware.

            1. Morse claim 5 was directed to physical marks and physical spaces.

              Algorithms are not directed to anything physical, but only to steps in a “process.”

              When a computer executes an algorithm, it processes through machine states. At no time is the computer in a particular state comprising the entire algorithm. In particular, no structure in the computer is created by an algorithm.

              1. Ned – you did not answer my question.

                You blandly repeat yourself and say “physical space” but I asked what is the structure of that space.

                The point that you seem intent on avoiding is that your own “spin” of focusing on “1” alone is just a logical fallacy as evidenced by attempting to focus on “space” alone.

                Try again – this time with just a little inte11ectual honesty.

    2. If you have no novel physical structure in your computer-implemented claim (recited directly or via 112P6) or if you have no novel physical structure (e.g., a new machine described in objective physical terms) in your information processing method, you are guaranteed to have either a serious 101 or 112 problem, or both (not to mention the 102 and 103 issues that rear up when the breadth of claims is opened by functional language including non-limiting abstract elements).

      This is in Williamson v. Citrix? I missed that part.

      1. This is in Williamson v. Citrix?

        I don’t think it’s in there expressly otherwise I would have quoted it. Courts typically don’t spend time articulating all (or even most) subsequent straighforward applications of reason and logic to their decisions.

        Here’s the logic: if you can’t obtain a claim to a machine described in functional terms X because the machine lacks sufficient structure description, then you can’t possibly do an end-around that result and obtain a claim that covers the use of this same unpatentable machine to perform the stated function.

        Consider the machine claim: A black box configured to perform function X.

        The claim is ineligible for lack of sufficient structure.

        How can there simultaneously be sufficient disclosure to claim: A method of performing function X, using a black box configured to perform function X?

        Now let’s layer on top of all this reasonableness the fact that information processing logic is ineligible for protection with a patent.

        1. Where is the reason or logic of your continued fallacy of “objective physical structure” being ANYTHING more than a mere option of claim form, and certainly not the only legally valid claim form?

          Or are you too going to try to continue to ignore what Prof. Crouch coined as the Vast Middle Ground?

          (or just keep on your inte11ectually dishonest and dissembling ways)

          1. You and Dennis should write a book together on this “Vast Middle Ground” concept that you apparently share.

            Of course, there’s always a chance that there is a vast middle ground of disagreement between Dennis and you, in spite of your endless efforts to prop your incoherent views up with something he once wrote (in what context exactly? evidently that doesn’t matter to you).

            1. Nice dissembling by attempting to implicate that I have taken Prof. Crouch’s term out of context.

              I have not taken Prof. Crouch’s term out of context.

              Try to say something on point – and be inte11ectually honest about it or kindly stfu.

    3. “Configurable black boxes that carry out logical operations aren’t eligible for protection with patents, regardless of whether they are claimed as black boxes or methods of “using” the black boxes.”

      This is actually true. The problem is that the above does not actually follow from the below:

      “Algorithsm aren’t structure.”

      This is actually where MM goes off the rails. A sufficiently defined algorithm actually is sufficient structure. This is why functional claiming is allowed when applied to old elements. The problem arises in the context of the absence of sufficient specificity for a novel algorithm.

      For example, a claim:

      A system comprising:

      Clause 1: a processor executing instructions for [standard old well-understood computer function],

      Clause 2: means for [function with well-defined and well-described novel algorithm in the specification]

      This is patent eligible and will remain patent eligible. To be sure, there are a lot patents “in force” where what should have been claimed and described as clause 2, is actually claimed in the form of clause 1, and they’re going down by the bunch. But the patent eligibility of the general approach outlined is legitimate.

      1. While I disagree that an algorithm is physical structure, it does define what something does — which is the proper in a method claim or in describing how a system operates. (A system claim is really a form of method claim.)

        “Means for” should be construed as apparatus, such as a computer, that is executing a program so that it performs the recited functions.

        Most programming is self-evident and enabled. We are not talking about physical circuits here.

        There remains only the question of whether a claim is effectively a single means claim. But beyond that, I find no object to most method and system claims claiming otherwise statutory processes of systems.

        1. Where is it written that a system claim is a form of method claim?

          You keep on making the same mistake of trying to equate software and the execution of software – they are just two different things Ned.

          Software is a manufacture and machine component. That is how it is defined.

          You also continuously want to ignore the fact of what it means to be equivalent, and that a physical circuit very much is equivalent to software. You have NO inte11ectually honest response or continuation of dialogue to the FACT that software is equivalent to hardware and is equivalent to firmware.

          Stop avoiding this point.
          Learn this point.
          Integrate this point into the dialogue.

          1. anon, both the Federal Circuit and Supreme Court have said that software is not patentable per se. The sale of software is not a direct infringement. Etc.

            It is you, anon, who are on the outside barking at moons.

  11. Is this not a storm in a teacup? The sky falling, except it’s not?

    Standard drafting advice for some time (as far as I know) has been to write a plurality of independent claims, with at least one in M+F style and at least one that avoids this style. That way, one has one’s cake and eats it too.

    Looking in from Europe, 112(6) was always an absurdity. But still nobody suggests simply striking it from the statute.

    But something has to be done about it. You can either (pace Newman) narrow it down so much that nobody stumbles into its grip unwittingly, or, with the majority, you can pump up its scope (I suspect Federico would approve) so that it captures all functional claiming

    Either way, from now on, one would know how to draft claims.

    And as I say, if you can always have multiple bites at the claiming apple, where’s the harm?

    1. Max, except that all functional claiming is now covered under 112(f) regardless of whether one uses the magic words. In other words, the broader you write, the narrower your claim.

      1. Yes Ned. I think there is no difference between how you see the majority view and how I do. But your comment intrigues me, that, from now on, the wider you claim, the narrower the scope of protection you end up with. That’s fun.

        In Europe, you can’t have more than one independent claim to the same subject matter. I think that much of the difficulty in construing an independent claim in the USA lies in permitting multiple independent claims to the same subject matter.

        The EPO thinks that practice renders fuzzy the precise scope of protection you are asking for. In that, the EPO is correct, isn’t it.

        1. Max, the problem we have in the US is 35 USC 251. If we have a reissue or reexamination, and we have to amend a broad claim, we lose past damages. Thus it is good to have claims of varying scope.

          Regardless, from the legal point of view, all dependent claims are effectively independent after the patent issues. That’s another statute.

          Second, we cannot have two claims of precisely the same scope. If we do, in court, one can assert only one of them.

          1. All good points Ned. Intervening rights? Not in Europe but one can see the point of having them.

            Dependent claims? Yes, of course. The importance of having a well-shaped taper on the dependent claim set will increase, I wager.

            Two independents of the “same” scope? There is much fun in that. US attorneys defending their drafting tell us they are subtly different. You have revealed why they do that. The EPO looks at the substance and sees it is the same in each of the independent claims proffered. One can see why they might do that too.

            1. A driver here MaxDrei, at least in part, is the way out sovereign has written out patent law and segregated individual categories.

              Invention itself typically can be – and must be – “scrivined” to fit into a category, and almost always can fit into more than one category. Thus, to deny multiple independent claims here in this sovereign is to deny to the inventor the full reach of the protection that the system was designed to afford in the first place. Chakrabarty offers some small insight into the topic.

              1. Perhaps a misunderstanding here, anon? In Europe it is routine to go to issue with a slew of different independent claims in the same patent. But each one is directed to (!) a different and distinct “aspect” of a common inventive concept. Each such aspect gets its own claim “category”.

                Thus, you will routinely find in one and the same EPO B publication independent claims to i) a molecule ii) a way of making it iii) intermediates in that process iv) pharmaceutical compositions that include the molecule v) uses for that pharmaceutical. In mechanics you might find independent apparatus (means for X-ing, Y-ing etc), method (steps of X-ing, Y-ing etc) and product claims for an inventive concept XYZ.

                It is just that it is unusual to get more than one independent claim in the same “category” for that implies a different “invention” (WYZ) and each patent is confined by our statute to a single invention.

                Whatever “scrivening” implies, a lot goes on in Europe too, to meet the requirements of our statute. It is the same verywhere, I suspect.

                1. I think the EPO’s “common inventive concept” is different from that applied in the US. Consider a patent application with a semiconductor device and a method for making the same. In the US, that is going to be considered two different inventions and restricted into two applications. In the EPO, it’s the same inventive concept.

                  How does the EPO apply its rules to a case where the method for making a semiconductor device is patentable, but the end result is not (as the structure of the device is already known from the prior art)?

                2. PB: How does the EPO apply its rules to a case where the method for making a semiconductor device is patentable, but the end result is not (as the structure of the device is already known from the prior art)?

                  They reject the claims covering the old device.

      2. Ned, doesn’t this opinion on its face just modify the presumption as to whether an Applicant intended a claim to be covered under 112 p 6? Previously there was a strong presumption that an Applicant did not intend to do so unless means for language was used, but now that strong presumption is essentially gone. However, couldn’t an Applicant still explicitly state either during prosecution or in the spec itself that claims are not intended to be covered under 112 p 6? In such a case, wouldn’t the presence or absence of a presumption be irrelevant? On the other hand, if you were to explicitly state a claim is not to be interpreted under 112 p 6, would you risk running afoul of Haliburton? It seems like, if we continue down this road, eventually Haliburton will have to be reckoned with, rather than just ignored and never cited.

        1. JCD, do you really believe the courts care about what you say in prosecution on this topic. But, go ahead and try. See if it works.

          As to Halliburton, what you are claiming functionally must be the invention so that the invention itself is not in the claim but in the spec. Now this fits quite well when one is claiming structure. But when one is claiming a method in the first place, this doctrine has no place.

          Software describes a sequence of steps, not structure. They are method claims, and have been treated as such by the Federal Circuit for quite some time.

          1. Yes, machines that perform methods have no structure. Your brain has no structure. You have a mind where all information processing is performed in the spirit world. That is how it works according to J. Stevens and his ilk.

            1. You either didn’t understand what I said or don’t understand what presumption is being referenced. The prior strong presumption was a presumption as to what an applicant intended, i.e. whether an applicant intended a claim element to be covered under 112 p 6. If you explicitly state you do not want a claim element to be covered under 112 p 6, then whether there is a presumption is irrelevant. It’s possible that existing precedent is entirely revisited such that Applicant intent is irrelevant, but traditionally the 112 p 6 inquiry is designed to divine applicant intent, not to arbitrarily construe a term free from applicant intent.

              1. JCD, Perhaps we need boilerplate in every application to the effect that the applicant chooses NOT to employ the “benefit” of 112(f) unless applicant specifically uses “means for” or “step for” in the claims.

        2. JCD: couldn’t an Applicant still explicitly state either during prosecution or in the spec itself that claims are not intended to be covered under 112 p 6? In such a case, wouldn’t the presence or absence of a presumption be irrelevant?

          Yes and yes.

          On the other hand, if you were to explicitly state a claim is not to be interpreted under 112 p 6, would you risk running afoul of Haliburton?

          Certainly there is a risk if the functional language in question represents your innovation over the prior art.

            1. Yes it was.

              I would (and I have here – this is not a new conversation point) point to the Act itself, the legislative record and Federico’s commentary.

                1. Asked and answered Ned.

                  Quite in fact, it was in a conversation with you that I shared the direct words of Congress, the congressional record and the direct implication voiced by none other than Federico. Upon that point in our conversation, you abruptly abandoned the dialogue and refused to integrate the points in your mantra. In other words, and as you have repeatedly done when I have provided counter points that are simply too inconvenient for you, you ran away.

                  You now play the Internet style game of pretending the conversation never took place. Stop the silly shouting down games Ned. You may not like the fact that ALL combination claims have been opened up to the words of function in claims and you may not like the fact that old items can be claimed anew if so claimed in combinations, but those are the words that Congress (and not your dust-kicking accusation of Judge Rich) enacted.

                  To go one step further, and to highlight the silliness of Malcolm’s pet 101 theory, and the “over” concern here with your own “anti-combination” mindset, do keep in mind that we still have the doctrine that mere aggregations – a listing of elements that have no functional relationship to each other – fail the claim as a whole reading.

                  This is true whether or not any element in a claim sounds in a mental step or sounds in terms of being described with language of function.

                  This too avoids the goal-post moving and dust kicking attempts to move discussions of those claims in the Vast Middle Ground to discussions of PURE function or claims TOTALLY in the mind.

                  If we keep the focus of the discussion clear and if we avoid the purposeful obfuscations, it is abundantly clear that the likes of you and Malcolm want a different law to be enacted from the one that Congress actually enacted in 1952.

              1. I’m not trying to be difficult or score points, I’m legitimately trying to understand what it is you are referencing. Can you point me to it?

                Thanks!

                1. Ideally with the same cite you would use in a Federal Circuit brief. That is what I was looking for. Something that can let me easily access the information you are saying is so easy to access.

                2. I don’t have cite to Federico’s commentary. I do have a copy.

                  Here is what he said in full:

                  The last paragraph of section 112 relating to so-called functional claims is new. It provides that an element of a claim for a combination (and a combination may be not only a combination of mechanical elements, but also a combination of substances in a composition claim, or steps in a process claim) may be expressed as a means or step for performing a specified function, without the recital of structure, material or acts in support thereof. It is unquestionable that some measure of greater liberality in the use of functional expressions in combination claims is authorized than had been permitted by some court decisions and that decisions such as that in Halliburton Oil Well Cementing Co. v. Walker, 67 S.Ct. 6, 329 U.S. 1, 91 L. Ed. 3 (1946), are modified or rendered obsolete, but the exact limits of the enlargement remain to be determined. The language specifies “an” element, which means “any” element, and by this language, as well as by application of the general rule that the singular includes the plural, it follows that more than one of the elements of a combination claim may be expressed as different “means” plus statements of function. The language does not go so far as to permit a so-called single means claim, that is a claim which recites merely one means plus a statement of function and nothing else. Attempts to evade this by adding purely nominal elements to such a claim will undoubtedly be condemned. The paragraph ends by stating that such a claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof. This relates primarily to the construction of such claims for the purpose of determining when the claim is infringed (note the use of the word “cover”), and would not appear to have much, if any, applicability in determining the patentability of such claims over the prior art, that is, the Patent Office is not authorized to allow a claim which “reads on” the prior art.

                  You will note that the drafter of the patent act said that 112(f) did not apply in the PTO and for good reasons. Donaldson did not quote Federico’s statement, but pointedly ignored it with this,

                  “In addition, P.J. Federico’s post-ACT “Commentary on the New Patent Act,” 35 U.S.C.A. § 1 (1954 ed., West), reprinted in 75 JPOS 162 (1993), is not legislative history per se that may be relied upon to indicate Congressional intent. Even if it were, the comments contained therein do not suggest that Federico knew of any particular intent by Congress regarding the manner in which the sixth paragraph, then the third paragraph, should be applied. In this particular, he was merely stating his personal views.”

                  In re Donaldson Co., Inc., 16 F.3d 1189, 29 U.S.P.Q.2d 1845 (Fed. Cir. 1994) at fn. 4. link to scholar.google.com

                3. JCD,

                  Respectfully, I am not going to “Bluebook” this for you. I have given you all that you need to have. The Statute itself, the congressional record to the statute, and the commentary post statute (Federico, which is not congressional record, but augments the record).

                  I will add this though: you do need to be careful to take the entire proceedingsand historical context into mind, as witness Ned’s attempted parsing here. The context of the Act of 1952 and its place in history – as a reaction by Congress against a Court that had turned TOO anti-patent must be kept in mind. Beware the false placebo of “just codification.”
                  The single paragraph that Congress took from pre-1952 law and turned into multiple sections cannot – and must not – be attempted to be taken as some sort of word for word transcription pre to post. Much like the addition of section 103 must be kept in mind for looking at the words of section 101 pre and post (since Congress acted to remove the Court’s common law ability to set the meaning of the word “invention” by refocusing instead on a different concept, that of obviousness), so too ALL of the sections of 112 must be kept in mind in looking at “the addition of the last section” and the words that make up the portion of the pre-1952 paragraph that were formulated into the post 1952 multiple paragraphs.

                  Ned’s obsession (wrongly placed) in the ascendancy of the Royal Nine and the Supreme Court simply is not appropriate given the fact that statutory law is ascendant – and explicitly so with patent law expressly because the authority for writing patent law is plainly given in the Constitution – and given NOT to the judicial branch, but to the legislative branch. Judge Rich was keenly aware of this – and this alone is the reason for Ned’s venomous and outrageous attacks on the personhood of that man. Do NOT so quickly gloss over the word “PURELY” that was the focus of adding that new last paragraph (as is so often attempted in the blog comments here). It is beyond clear that Congress allowed the use to terms of function throughout 112 – and not just in 112(f). Prof. Crouch has correctly coined the term: Vast Middle Ground. And instead of being inte11ectually honest about this, you can witness the typical anti-patent attacks that REFUSE to accept this, refuse to learn this, and refuse to integrate this point in the dialogue.

                  Unless and until this understanding is treated with a basic respect, ANY point attempted to be made in denial of this plain historical fact is merely obfuscatory and dissembling, and deserves nothing but scorn and abrogation.

                4. JCD,

                  Further note how Ned attempts (again) to spin (incorrectly) the notion that 112(f) is not to apply to the Office during prosecution. He tried that very same arguement in a court of law and was rebuked there. Instead of learning from his mistake, he is still chasing his windmills.

    2. How bizarre MaxDrie is – “What me, Worry?”

      Never mind such “petty” things like separation of powers and rule of law, one can always (later) react….

      /off sardonic bemusement

    3. Max,

      You say: “Either way, from now on, one would know how to draft claims.”

      I think Newan’s dissent was based in part on not creating, what she called in another case, a “trap for the unwary.” This is inadvertently using what the CAFC calls a “nonce word.” Engineers quite often describe systems as comprised of various modules. Clearly, that word is now verbotten, but some other descriptor may be considered a nonce word when a means plus claim element was not intended.

      1. OK Ned. Do you mind it if I suggest that drafters need to know when they are straying into claiming “by function” and Newman wants to help them.

        I’m not sure whether there needs to be a distinction between claiming “functionally” and not. I think any attempt to make the distinction is doomed to fail As far as I know, under the EPC nobody needs to make any such distinction. Isn’t 112(6) the fount of all your difficulties? Would it not be better just to scrap it? There are other tools to discipline those who draft covetously, aren’t there. remove 112(f) and the courts will be quick to seize one of these other sticks, to beat the claims down.

        1. You (too) seem to miss the distinction between claiming PURELY by function and claiming outside of the realm of PURE functional claiming and using words that sound in function – the Prof. Crouch coined term of Vast Middle Ground.

          It is ONLY when a claim is PURELY functional that the “OhNoes” and pearl clutching should be started, and yet, we have a philosophically driven contingent to broaden that “OhNoes” into the aforementioned Vast Middle Ground by hook or by crook (or by “Gist” or by parse). It is this “philosophically driven” game that is the root of all evi1 here.

        2. Max, you probably are aware of this, but 112 p 6 was enacted in response to court precedent striking down functional claiming at the point of novelty. Getting rid of 112 p 6 would (seemingly) render claims found to utilize functional claiming at the point of novelty invalid, regardless of whether there is corresponding structure in the specification. 112 p 6 simply gives you an out in case there is not structure in the claims, but there is structure in the specification. The strangeness, I believe, comes in in that 112 p 6 can be applied to elements that are not at the point of novelty to unduly narrow a claim to what is disclosed in the specification.

          1. The strangeness, I believe, comes in in that 112 p 6 can be applied to elements that are not at the point of novelty to unduly narrow a claim to what is disclosed in the specification.

            That “undue” narrowing will occur only where the drafter was exceptionally lazy.

            Best practice has always been to provide a substantial list of examples and definitions for every essential element recited in the claims. If you want to take shortcuts, be aware of the risks.

          2. [retry with proper formatting]

            JCD: The strangeness, I believe, comes in in that 112 p 6 can be applied to elements that are not at the point of novelty to unduly narrow a claim to what is disclosed in the specification.

            That “undue” narrowing will occur only where the drafter was exceptionally lazy.

            Best practice has always been to provide a substantial list of examples and definitions for every essential element recited in the claims. If you want to take shortcuts, be aware of the risks.

            1. I don’t think this is as clear cut as you make it. I don’t think it’s productive to have 20 pages of boilerplate outlining definitions for “computer”, “monitor”, “processor”, etc. This type of bloat is the reason that examiners often don’t even bother reading the specification, which results in lowered examination quality.

              Amusingly, I would offer that best practices would be to only consider the point of novelty as “essential”, and thus under your own best practices statement it is only at the point of novelty that you would want to offer examples and definitions.

              1. JCD: I don’t think it’s productive to have 20 pages of boilerplate outlining definitions for “computer”, “monitor”, “processor”, etc.

                I didn’t suggest that “20 pages of boilerplate” was necessary.

                Keep your claims focused on the innovation and you won’t need so much boilerplate.

                That said, diverting the discussion to “examination quality” is a loser’s game. At the end of the day, the application drafter remains in the driver’s seat. You’ve either built the necessary contingencies into the application because you are an experienced prosecutor who knows that patent law is continually evolving or you are a newbie who spends too much time reading finger-pointing blame-the-Examiner patent blogs written by rich entitled attorneys who are wrong about pretty much everything.

                It’s your choice, JCD.

                1. Yep, you’ve definitely accurately described the world and the only two possible choices while fully addressing my comment that your own best practices statement includes reference to “essential” elements that would once again only entail offering detailed structure and examples at the point of novelty. Or the exact opposite. One or the other.

  12. Genghis, I did not give you a “version” of history. I find your remark offensive.

    No offense was intended. I think all histories are versions. Some versions are based on facts and try to be as fair and accurate as is reasonably possible. But they are still versions. I have no reason to believe that your version is not fact-based or that it is not fair and accurate. But that doesn’t mean I’m going to blindly accept it.

    1. …and on the opposite side of Alfred E. Newman…

      Pay no heed to Ned’s rhetorica1 trick of false umbr@ge. He likes to pull that out every once in a while, especially with new readers. He has no substance in his faux m@dness

  13. In the article and in the discussions I noticed the long chain of Fed. Cir. decisions that created the roller coaster ride for the means-plus-function presumption.

    But what about the Supreme Court? Has it weighed in on 112(f) (or 112, para 6 as it was before)? Is this case now headed to the Supreme Court?

    1. The Supreme Court has not addressed 112(f) or Halliburton since 1952.

      The case that should’ve gone up was en banc case of Donaldson where the Federal Circuit unanimously held that the statute applied in the patent office and effectively overrode the requirements of 112(b). The patent office position was that statute was intended only as a rule of construction in court because it was the duty of the patent office to make sure that the claims as issued clearly defined within their four corners what the invention was and that this was impossible if an applicant could distinguish the prior art by reference to structure in the specification that is not in the claims.

      Of course, the patent office was entirely correct and the Federal Circuit was unanimously wrong.

      1. The patent office position was that statute was intended only as a rule of construction in court because it was the duty of the patent office to make sure that the claims as issued clearly defined within their four corners what the invention was and that this was impossible if an applicant could distinguish the prior art by reference to structure in the specification that is not in the claims.

        I understand that there was a time when a patent did not have to include a separate “claims” section, and we were supposed to figure out the bounds of the invention by reading the specification. Then, at some point we switched to the current system where we have a separate claims section. Maybe there was a concern that for some inventions it would not be possible to write a “claim” using the “new claiming system.” Was 112(f) supposed to be a way to allow inventors to claim inventions by using the old approach, i.e., by describing the invention in the specification and essentially writing a claim that says “look at the specification”? If so, then that may have been the reason for the Fed. Cir.’s decision in Donaldson.

        1. Genghis, Generally, one cannot claim new structure in terms of what it does rather than what it is because the claim would cover not only the structure disclosed but all structure that achieved the claimed result. Such claims are deemed too broad and indefinite – claiming the invention in the abstract. See e.g., O’Reilly v. Morse.

          However there was a case called Westinghouse v. Boyden Power Brake where the novel structure (of claim 2) was claimed in terms of what it did. The Supreme Court held that in order to preserve the validity of the claim, the claim would be construed to cover the corresponding structure recited in the specification and equivalents thereof.

          Thereafter, patent attorneys adopted the practice of claiming novel structure using means-plus-function with the understanding that the claim will be construed in court to cover the corresponding structure described the specification and equivalents thereof.

          Then along comes the Halliburton case that says those claims are inherently indefinite because of their apparent scope when claiming novel structure regardless of the rule of construction in court. According to Halliburton, it was the threat of such a claim that was the problem – people would naturally tend to avoid doing the function claimed because of the uncertainty of whether what they were doing was an equivalent. Essentially Halliburton overruled Westinghouse.

          But the bar reacted in horror as so many claims had for so long been claiming novel structure functionally with the understanding that they would be construed in court to cover the corresponding structure.

          So the primary purpose of enacting 112(f) was to restore the ability to claim novel structure functionally as before and to affirm the validity of thousands, if not hundreds of thousands of patents that had been using this practice since 1898, the date of the Westinghouse case.

          But the statute was specific in terms of what one had to do to invoke its protection against invalidity. One had the use means for or step for in the claim. If one did not do that, theoretically, a claim claiming novel structure functionally would be invalid under Halliburton.

          But one has to note that Halliburton was only concerned about claiming novel structure functionally. It was not so much concerned about claiming old structure functionally. Faulkner v. Gibbs clarified this a few years later.

          Thus there is no reason to construe functionally claimed old elements under 112(f). None at all.

          1. Thanks for the history lesson.

            (Post like Ned’s history lesson are the reason why I keep coming back to this blog. Than you Ned and everyone else who post informative material in the comments section.)

            1. Just be advised that Ned views history through his own lens — he does not agree with “functional” claiming, so everything is tainted by that.

              So, you’re getting a “history” lesson, but it’s one-sided.

              1. Backwards, PatentBob. You and your ilk live in the land of make believe and the made up.

                Have you ever heard of Judge Rich? Now there was a real piece of work. He set about to overturn just about every Supreme Court patent case he did not like, and he did not like most of them. So, he simply refused to follow them and/or ignored them. It was on his watch that the CCPA created the fiction of the programmed computer. What fascicle nonsense. What next, the “authored book,” “the composed score,” the arrowed bow. Is the book, the score or the bow a new manuafacture because one changes the meaning of the letters on the page, the music one hears when one plays the score, or is the bow really new when one improves the arrow.

                Rich was among the worst judges in all history, IMHO, unless that is, you agreed that he was right in what he was attempting to do, protect software with patents.

                1. Ned, I have ripped apart your “version” of history many times and you insist on playing your B$ drive-by monologuimg that ignores your massively tainted errors – beware indeed new readers, this is exactly why the propaganda never NEVER ceases in this “ecosystem.”

              2. Understood. I am not blindly accepting Ned’s version of history. I’m just thanking him – and others – for taking the time to clearly articulate their thoughts.

          2. Ned,

            So, you are saying “means for electrically connecting said contacts with the corresponding indicators, means for supplying electric current to said indicators” as in the Faulkner case do not need construing because they are not at the “precise point of novelty” as in the Halliburton case.

            I was under the impression that now all claims are considered as a combination whether the elements are new or not. Nothing in 112(f) or this case as I read it distinguishes between old and new. Please advise.

            1. Troubled, the issue is not whether means-plus-function claim element should be construed but whether they should be limited to the corresponding structure recited the specification and equivalents or whether they should be construed to cover any means that carries out the side of function which is the way they are construed in rest of the world.

              However given the statute, there is a command that means-plus-function claims be construed in a specific way. Every body knows that it advance so when they write such claims and employ the magic words “means for” or “step for” they know what they are getting.

              So the question becomes whether claims not using the means-plus-function format should be construed as means-plus-function claims. The answer clearly is no – they should be given their full scope; and if it turns out that they are functionally claiming the invention, then the claim is invalid.

              1. Ned,

                Based on this case, I think 112(f) will be even less popular than today, so my question was somewhat academic. However, from your answer, I take it that all means for or step for claim elements are treated the same way whether they are conventional or not.

                I also understand your position to be that any claim element reciting a function not preceded by means for or step for is invalid on its face. Is that too simplistic? If not, how does one distinguish OK functions from those that are not. Does one use an O’Rielly v. Morse preemption analysis, for example?

                1. Ask Ned to come to grips with Prof. Crouch’s own coined term of “Vast Middle Ground – or Federico’s pointing out that terms using function was greatly expanded by Congress outside of 112(f), also noting that 112 and 101 prior to 1952 – were one paragraph, and Congress intended – as stated in the historical record – much more than mere codification (including – per Federico – the abrogation of Ned’s wrongly cited Haliburton.

                  Ned utterly lacks inte11ectual honesty on this topic.

                2. No, Troubled. You do not get it. The functional claim element should be given its full scope UNLESS it is claiming the invention. Functional at the point of novelty is the vice.

                3. And, since this is a defense, it should be incumbent on a person asserting the defense to prove that the invention lies in the one element claimed functionally.

                4. The invention is the claim as a whole – your “PON” as you are attempting to use that term is a canard.

                5. Ned,

                  I have to reply to myself because the indent string ran out at your reply to me below.

                  MPEP 2173.05(g) discusses Functional Limitations. Summarizing four columns which discuss inter alia the Halliburton case, “a claim term is functional when it recites a feature by what it does rather than what it is.” There is nothing wrong with this but “A functional limitation must be evaluated and considered, just like any other limitation of the claim.” Citing a case (Innova/Pure Water v. Safari Water), “subject to any clear and unmistakable disavowal of claim scope, the term ‘operatively connected’ takes the full breath of its ordinary meaning,”

                  In seems like the PTO already employs your “full scope” approach. True? But maybe your concern is with issued patents. If so, are you suggesting BRI should be applied to those that contain functional limitations?

                6. The problem with the PTO is that it does not understand Halliburton. Perhaps they have a reading problem. Perhaps they haven’t read Faulkner v. Gibbs.

                  Functionally claiming OLD and WELL KNOWN elements does not have a enablement problem.

                  Neither did Halliburton have any problem with a new combination of functionally claimed elements.

                  The vice that Halliburton was trying to stop was claiming the invention functionally — where the sole novelty was in the one element. This effectively was a single means claim.

                7. “Effective” and “parse” and “gist” just do no comply with the actual law – as written by Congress.

                  Ned, I “get” that you want a different law, one in accord with your Belieb system, but you really do need to take the law as it really is – as it was written by Congress with the changes made in 1952 (and just stop clenching tight your eyes and pretending that “only codification” was taking place, as that is clearly false with the single pre-1952 paragraph being massively changed.

                  You need to accept the reality of history, and not retreat to your fantasy “spin” land.

            1. The requirement to particularly point out the invention came from this case: Evans v. Eaton, 20 U.S. 356, 5 L. Ed. 472 (1822).

              The requirement is ignored by the Federal Circuit.

              1. Ned likes to continue to ignore the fact that Congress ended the Court’s power of developing patent common law in 1952.

                Come people – this is basic: the difference between statutory law and common law and which branch of the government has authority to write law.

                Ned – watch The Paper Chase and note the comment at the hour and six minute mark.

                Stop
                Being
                Purposefully
                Ignorant

                1. >…watch The Paper Chase and note the comment at the hour and six minute mark.

                  Nobody will listen to your fictions, so you use another work of fiction to do it for you? :3

                2. You are not nearly as clever as you think yourself to be my amorphous friend, as works of fiction can and often do present facts and laws that are in accord with reality.

                  Here, specifically, the quote to be grasped is applicable to this reality of U.S. jurisprudence.

                  You would do well to understand the point that you seek to mock, as your act of mockery paints you to be the f001.

  14. A method claim recites a sequence of steps. An apparatus claim simply adds the phrase “module for” to each step. The apparatus claim is rejected as indefinite under 112(f). Now is the method claim automatically indefinite?

    It seems this decision says that the method claim may still be definite even though the apparatus claim is indefinite. That doesn’t seem logical, but that’s what the decision seems to be saying.

    What is the real impact here? If I’m reading this decision correctly, method claims would survive this Fed. Cir. decision because they don’t trigger the 112(f) analysis. If that is the case, then can’t a patent owner simply assert the method claims against exactly the same infringing product?

    1. Genghis, why of course you hit upon the complete hypocrisy of the argument that an apparatus claim is indefinite while a method claim is definite where they are claim the same thing.

      Halliburton was about claiming structure functionally. Halliburton was not concerned about method claims where the structure itself was unimportant but the actions were important.

      When the invention is in the method, recasting the claim as an apparatus claim should provide no different result. The problem we have is that the Federal Circuit once upon a time concluded that when one is doing this, recasting a method claim as an apparatus claim, that one must disclose an algorithm as if an algorithm were structure. The problem is that they should never allowed one to claim methods as apparatus.

      Perversely all this came out of in re Bernhart that had held that a programmed computer was a new machine for 101 purposes. Thus for a very long time the patent office would allow claims to programmed computers that they would find invalid the claimed as method claims.

      Now we have a mass of confusion where apparatus and method claims claim exact same invention existed in the patent and one is held to be indefinite and one is held not. This is not funny.

      1. I can’t agree with this: “Perversely all this came out of in re Bernhart that had held that a programmed computer was a new machine for 101 purposes.”

        A programmed computer is a new machine for 101 purposes. If you don’t believe me (and you have a windows computer with windows 7), hit the start button, type in “cmd” in the “search programs and files box”, click on the “cmd.exe” link found. In the window that pops up, type “format c:”. Follow the prompts to format your c drive.

        After, all you don’t need any of the operating system, word processor, web browser, etc., because the unprogrammed computer has nothing to offer over the programmed one, correct? I mean, the programmed computer is in no way “new”, thus all you need is the “old” computer without all that pesky software.

        1. Patent Bob, there is no such thing as a programmed computer. A program exists external of the computer and is loaded by the computer and there executed. The program tells the machine when executed what to do in terms of a sequence of computer instructions. Thus a program describes the sequential operations of a computer. It does not describe a new machine.

          1. To me this debate just boils down to where you draw the boundary around the so-called “machine.”

            For the sake of argument, let’s assume that we have a CPU and a non-volatile memory that permanently stores a program.

            If you draw a boundary around the CPU then I think everyone agrees that there is nothing new – the CPU outputs the same output signals for the same set of input signals. But if you draw the boundary so that it includes the CPU and the non-volatile memory, then that arguably creates new “machines” depending on what is stored in the non-volatile memory because it outputs new output signals for the same set of input signals depending on the program that is stored in the non-volatile memory.

            Am I the only one who sees both points of view and is still not sure which one is correct?

            When we have biological computers (see link to extremetech.com), the “program” could be stored in DNA (the article that I linked to doesn’t exactly say that, but I can see how one could extend the article so that we store the “program” in DNA). This DNA is clearly is going to be man made. Patentable?

            1. Genghis, you see the problem — how do we define “computer.”

              Well I suggest we define it the way it existed in 1946 when stored program computers were first invented. Things have changed since in terms of architecture, but not the principle of the design.

              Programs existed in media apart from the computer. They were loaded by some means and executed. The entire program did not have to be in memory at one time to be executed, because program memory was extremely limited and expensive. Think of the early program memories as today’s cache, and main storage as the original paper or mylar tape that stored programs back in the day.

              Now some may want to confuse and conflate microcode with programs. But no one skilled in the art would be confused.

              Some might also want to confuse programs with ROM-based programs like a bios. But, these are distinct.

              Programs by definition are not part of the machine but are objects of the machine. They are external to the machine. That is what they are from the very beginning. The program and the computer are distinct things.

              1. Well I suggest we define it the way it existed in 1946 when stored program computers were first invented. Things have changed since in terms of architecture, but not the principle of the design.

                According to this definition, are artificial neural networks computers?

                1. Genghis, I don’t think so. Computers are generally understood to have a Central Processing Unit. This implies a central unit that processes instructions. Neural Networks do not have a central processing unit that I know of. I don’t even know if they have instructions. They simply have logic and memory.

                2. Poor Ned is stuck in time.

                  He is definitely NOT approaching this like a Person Having Ordinary Skill In The Art.

                  He is stuck trying to dissemble because of his Belieb system (and his clients, right Ned? – oopsie)

                3. Artificial neural networks (ANNs) are usually considered to constitute a family of statistical learning models and not a class of computer.

            2. Genghis: To me this debate just boils down to where you draw the boundary around the so-called “machine.”

              What “debate” are you referring to exactly?

          2. One should distiniguish program executing devices (like general purpose computers) from programmable logic devices. In the past some general purpose computers had writable (or in the case of Prime Series 50 program selectable) instruction sets.

            When they were first introduced aspects of WISC and SISC machines might have been patent-eligible and patent-able.

            WISC processors were programmable /reprogrammable logic devices and could perhaps be considered early examples of Field Programmable Gate Arrays (PFGAs).

            1. But, Joachim, even FPGA’s are programmable. The programming doesn’t change the machine, the FPGA, unless the configured state is made fixed in some fashion, like microcode in a computer. The programming and the machine remain distinct.

              Now, put the FPGA in a larger process or a machine, as in Diehr, then it makes little difference whether one uses an FPGA or a GPDC to perform the calculations. The process or the larger machine is the claimed invention.

              1. But until recently FPGAs were rarely programmed to be program-executing devices.

                Nowadays (at least for the last 10 years or so) one can purchase microprocessor cores like MIPS or ARM to load into FPGAs. Some intelligent data moving or networking chips like the IOP480 come with microprocessor cores.

                Typically, because only one “user” program is ever downloaded into such a device, this type of device should be considered a sort of a hybrid between programmable and program-executing logic device.

                Obviously, the boundaries between programmable logic and program-executing logic has become a good deal less sharp especially because the systems for programming programmable logic devices incorporate languages that look a lot like high-level programming languages for program-executing devices.

                Usually, the programming of a programmable logic device involves a lot more parallelism than we see in programs written in high-level programming languages for program-executing devices.

                Ned is completely correct that the GPDC of Diehr was simply a system component, whose functionality today could easily be implemented with a programmable logic device. I recently made that argument in a response to an office action.

                1. Joachim, Intel has just acquired an FPGA company. I would think that we will soon see FPGA’s incorporated into advanced processors to off-load complex functions involving vector processing and the like, to amp up graphic processors, or perhaps to do something we never thought of before.

                2. Repeat after me:

                  Software is equivalent to hardware and is equivalent to firmware.

                  Let’s have a little inte11ectual honesty on that point, shall we ladies?

          3. Patent Bob may be attempting to describe a modern turnkey system based on a general purpose computer.

            In the early 90s a turnkey system was typically a custom piece of hardware and ran custom software. Nowadays, a turnkey system is often a Linux based system with a virtual file system constructed from data stored in some sort of non-volatile memory (nowadays typically flash, but there are many ways to create the RAM file system).

            In the early 90s or modern turnkey system typically only one special purpose program can be executed. Cisco routers of the 90s corresponded to the first sort of turnkey system while the current generation of Cisco routers correspond to the second type of turnkey system.

            In all cases the Cisco devices do something that cannot be done mentally — switch packets from one network to another, Aspects of the Cisco devices were and are patent-eligible/patentable as would be a Cisco device implemented by porting Cisco software to a GPDC. The GPDC with the Cisco program would still be switching packets although probably not as quickly.

            I wonder whether some of the turnkey word-processor claims of the 80s were so obviously allowable under 101. If they are allowable under 101, should they really have been allowed under 112 paragraph 6?

            Here is a case in point.

            link to google.com

            The independent claims are formulated solely via means+function.

  15. In the big picture, this case is the result of Lemley’s functional claiming paper, which is a total sham. This case is judicial activism that ignores the reality of information processing inventions and creates a whole new world of hurt for innovation.

    That is reality and the big picture. (And, reality: the cost of patent prosecution and litigation is nothing compared with the cost of research and innovation. 300,000 patents is approximately 9 billion with a ($30,000 budget which is very generous.) $9 billion is chump change to price of innovation and research. So, what do patents make up 5% of the cost at most? Again, reality.

  16. MICROSOFT CORPORATION v. PROXYCONN, INC
    link to cafc.uscourts.gov

    While the Federal Circuit affirmed that the claim construction to be used by the patent office in IPRs is broadest reasonable interpretation (“BRI”), the Federal Circuit essentially used its conventional claim construction approach for district courts to determine that the construction used by the PTAB in two instances was unreasonable.

    In particular the court found that that the construction used by the PTAB was not consistent with the usage of the terms in the specification. The court held the PTAB claim construction to be unreasonably broad, reversed that holding, vacated the PTAB decision of unpatentability, and remanded.

    The Federal Circuit also addressed the issue of whether, under the rules, the patent owner had to distinguish the prior art of record or only the prior art used against the claims the substitute claims were intended to replace. The PTAB allegedly did not follow the rules, but imposed additional requirements of an enlarged board case called Idle Free that required the patent owner to distinguish all art of which it was aware. The Federal Circuit held that the patent owner had to distinguish all the prior art of record when proving that a substitute claims were patentable.

    Scope of review: claim construction: intrinsic, de novo; extrinsic, substantial evidence; interpretation of rules: APA, 5 USC 706, arbitrary and capricious.

  17. There is another interesting aspect of this case. Assuming the following characterizations of the prior Fed. Cir. decisions for this legal test are correct, this is another amazing demonstration of how Fed. Cir. panels rampantly ignore the Courts own clear rules of stare decisis, which require later panels to follow legal determinations of prior panels unless and until overruled en banc:
    “Under Federal Circuit precedent from the 1990’s, “the presumption can be overcome and § 112, para. 6 will apply if the challenger demonstrates that the claim term fails to “recite[] sufficiently definite structure” or else recites “function without reciting sufficient structure for performing that function.” Id. (citing Watts v. XL Sys., Inc., 232 F.3d 877, 880 (Fed. Cir. 2000). Subsequent cases raised that presumption, first to a “strong” one (Lighting World), then to a “strong one that is not readily overcome” (Inventio), and then to an even higher bar..”

    1. BTW, several years ago I did a test of how much respect Fed. Cir. panels give to their [en banc adopted] rules of stare decisis by pulling the then recent Fed. Cir. decisions on the test for a 102 anticipation. I found that only one such panel had even cited the first (supposedly controlling) Fed. Cir. decision on that legal test.
      Subsequently I had collected a number of Fed. Cir. “best mode” decisions and noted that most of them also simply ignored [and substantially differed from] original Fed. Cir. panel decisions on what that statutory requirement really meant.

        1. ^^^ oh, the irony of Ned Heller asking this (yet still clenching tight his eyes when it comes to the Royal Nine’s reckless disregard of the separation of powers).

    2. Sometimes I believe they don’t know what their current rules are. It’s as if there are too many cases and too many judges. Certain judges seem to latch onto single cases or lines of cases and ignore (or don’t know about) cases with other holdings.

      1. PatentBob, perhaps it has to do with the way they cite cases. If they required one to cite the original holding first, with subsequent authority cited as following or or distinguishing the first case, not only would they know, but we would know, what the controlling authority really was.

        1. Ned, is what you are suggesting is that a major source of the stare decisis problem is the frequency with which attorneys for both sides fail to cite controlling cases to the Court, sometimes miss-characterize the cases they do cherry pick to cite, and never get sanctioned for it?

          1. Well, the court itself is guilty of this. I am not surprised that attorneys adopt the same style as the court, recent cases first, even where the recent cases only follow earlier cases. That permits one to cite to panels that ignore prior panels even where the facts and issues are substantially the same.

            When I went to law school, I was taught to cite the case that actually made the holding. One did not cite as authority cases that simply followed a prior case. Thus if one found a case on a particular point, one traced the case back to the case that made the original holding and cited it. (I do that here, often referring to Hotel Security as the origin of the printed matter doctrine; or Application of Bernhart for the source of the rule that a programmed computer is new machine for 101 purposes.)

            Now, if a later case did not follow a prior case, but distinguished it based upon the facts, one should site both and state that the one distinguished the other.

            I suppose one could do what you did and cite both lines of authority so that a judge could know that there are two lines of authority on a particular point. This would be similar to citing a split among he circuits, with a majority view and minor view. We could do that by counting judges on the court who held one view or the other.

          2. sometimes miss-characterize the cases they do cherry pick to cite, and never get sanctioned for it?

            Paul, you are talking to one of the chief offenders of that, none other than Ned Heller.

            1. anon, I would appreciate it if you would stop it. We disagree about the interpretation of cases. The problem is, you simply say I am wrong and never explain why or what you think is a proper interpretation. Thus, we never can really discuss things properly.

              All readers have to know is that you disagree with me. That does not make you right.

              1. I have explained – many times.

                I even BEG you to join the discussion of my explanations and explore the logic well known to those who study history, know law, and know facts.

                You just run away.

    3. I was under the impression that the CAFC never treats issues of ordinary English grammar, meaning, or usage as matters of fact. For the CAFC such issues are always matters of law, and it addresses them de novo. Because of the nonsense that I have seen in claim construction (e.g., an attempt to claim that the definite subject of a nominative absolute is the object to which a claim is directed), the refusal of the CAFC to defer to the opinions of linguistic ignoramuses seems like a good thing.

  18. The problem with this decision is the Fed. Circuit’s failure to provide a well-reasoned and factually supported rational for overturning prior precedent. There’s no citation in the opinion to even remotely support the allegation that the prior precedent and the plain words of the statute “resulted in a proliferation of functional claiming untethered to § 112, para. 6 and free of the strictures set forth in the statute.” The Fed. Circuit was supposed to lead to more predictability but has failed time and time again to achieve this goal.

  19. In Halliburton, the Supreme Court
    made the following observations in holding certain claims
    that recite “means” language invalid:

    The language of the claim . . . describes this . . .
    element in the ‘new’ combination in terms of what
    it will do rather than in terms of its own physical
    characteristics or its arrangement in the new
    combination apparatus. We have held that a claim
    with such a description of a product is invalid . . . .
    Id. at 8.
    Arguably, this rationale applies to functional claiming
    generally, not just to claims that recite “means.” Indeed,
    the Halliburton Court relied on precedent invalidating
    functional claims that did not recite the term “means.”
    Id. at 9 (citing Holland Furniture Co. v. Perkins Glue Co.,
    277 U.S. 245, 256–57 (1928).)
    The continued viability of
    this rationale, and its impact on how this Court applies
    § 112, para. 6 merits attention.

    Reyna walks RIGHT UP to the line of intelligence and then fails to take the last step.

    There is no distinction between the requirements of 112f and non-112f claiming. The application of 112f is simply a shift of form – an imputation of spec into the claims where the claims would otherwise have to write out the structure itself.

    One cannot obtain a larger scope than a means-plus scope, regardless of whether one uses means-plus language. There are concrete embodiments to which one is entitled, and there are what the art would consider equivalents to those embodiments, to which one is entitled.

    There are also non-equivalents which also perform the same function. People who use functional limitations to attempt to gain entitlement to those embodiments are simply creating overbroad reaches and invalid scopes.

    1. Newman finds the intelligence that Reyna lacked:

      Today’s ruling is an example. The court holds that
      the clause “distributed learning control module” is subject
      to paragraph 6 because “module” is a “nonce word.” The
      court then finds no “algorithm”
      for “module” in the specification,
      and invalidates the claim for failing to comply with
      paragraph 6.
      However, contrary to the apparent belief of
      the majority, the presence or absence of the paragraph 6
      signal does not affect the requirements of patentability

      While I am somewhat agnostic under what rules are given to invoke 112f (and I can certainly see Newman’s point), the major issue Newman gets totally right: The algorithm disclosure would have been necessary anyway. It just would have been necessary under 112a instead of 112b, and it would have had to have been a limitation within the claim.

      1. Random, if the invention is in “structure” (I mean real structure) actually disclosed in the specification (which is not typical of software), a functional claim to that structure is invalid under Halliburton – invalid under 112(a) and (b).

        As to software, anybody can write code to carry out a functional statement. Software is not structure, but a series of steps. I would suspect that 112(f) should be invoked only if the sequence of steps is not claimed, but merely the result: code to achieve a particular result without more.

    2. RG: There is no distinction between the requirements of 112f and non-112f claiming. The application of 112f is simply a shift of form – an imputation of spec into the claims where the claims would otherwise have to write out the structure itself.

      I agree with this.

      If you are claiming functionality, then the corresponding structure for that functionality needs to be in the specification (i.e., the corresponding means) or recited in the claim if you choose the more verbose route.

      The underlying absurdity, however, and the one that will continue to warp the entire system around it until the absurdity becomes evident to all but the most extreme patent worshippers is the continued reliance on the fiction that algorithms and logical methods are “structures” amenable to analysis under the system of statutes, examination rules and case law designed for an entirely different purpose.

      1. The purpose is innovation and the protection thereof.

        The offer to you remains open – feel free to NOT use any innovation that you are unwilling to provide protection for.

  20. Good thing software companies don’t need patent protection–because things just got even tougher for them.

    Watch for software companies to start filing a lot more method claims in their apps vs apparatus claims. Lots of method claims.

  21. Under Federal Circuit precedent from the 1990’s, “the presumption can be overcome and § 112, para. 6 will apply if the challenger demonstrates that the claim term fails to “recite[] sufficiently definite structure” or else recites “function without reciting sufficient structure for performing that function.”

    Therefore a claim to computer code will always invoke 112f because code in and of itself is insufficient structure for achieving the function. It must be code with a particular (and that particularity is left out of the claim) structure that can achieve the functionality.

    “Generic terms such as ‘mechanism,’ ‘element,’ ‘device,’ and other nonce words that reflect nothing more than verbal constructs may be used in a claim in a manner that is tantamount to using the word ‘means’ because they ‘typically do not connote sufficiently definite structure’ and therefore may invoke § 112, para. 6.” Id.

    Again, computer code, or program, or application, absent a claim to the algorithm, always meets this standard.

    The end result is, as I have always said, that you cannot simply claim “all programs” that achieve a result, but rather “this algorithm achieving a result.” This is probably being done to protect the applicant, because trying to claim the function outside of 112f is invalid anyway, for reasons I have often stated.

    1. One of the problems with this is that the rest of the world does not interpret language like this. So, you’ll get Chinese and European patents written using “modules” and other similar words. They may have no disclosure for the modules. That’s because they don’t need to. Here, what’s a valid patent elsewhere becomes invalid.

      1. Well in fairness that’s because they learned from us and we’ve been doing it wrong for 30 years.

        You can’t on one hand say something is so enabled that I don’t need to teach you how to achieve it and on the other say that the implementation is both non-obvious and more than just an abstract idea to achieve something.

        An invention has to have a teaching at some point. The teaching can’t simply be an idea to achieve a result.

        1. Your last statement to me starts to get at a major point of disagreement between various camps. If you have an idea to do something which no one has thought to do before, even though everyone can immediately figure out how to do it, is that patentable? Some would say yes, others no.

      2. And on the flip side, if you avoid means plus function disclosure and claims in your US original patent, you will not have support for powerful means plus function claims in Japan and Europe.

    2. Random, focus on the bios limitation. Is 112(f) invoked in either claim 1 or 2. Second, is either claim indefinite with respect to this element?

      I claim:

      1. Apparatus, comprising:

      a bios configured to boot an OS; and

      a novel OS.

      2. Method, comprising:

      booting a computer into OS; and

      a novel OS method.

      The spec. states that any boot program may be used. No structure is described.

      1. a bios configured to boot an OS;

        Well there’s a question in my mind whether this invokes 112f at all, since the structure may inherently be sufficient for performing the function. i.e. It would be like saying a cup configured to hold a liquid – the office reads that as a cup and not the cup of the spec.

        Let’s assume it does invoke 112f – that functionality isn’t in the claim for no reason. If the reason is simply to get to the OS, what does it matter nominating a particular bios in the spec and then saying “the important feature of this bios is that it launches the OS, and all other bios which launch the OS would be sufficient” and then doing an equivalency argument since, according to you, all bios ever made will be the same.

        If the reason is that it launches the OS in some specific manner, then you aren’t entitled to protection for another manner not contemplated. There is no harm in holding it to means-plus if there truly is no difference and there IS harm in not holding it to means plus if there is in fact a difference.

  22. GTNX, INC. v. INTTRA, INC.
    link to cafc.uscourts.gov

    Feds again hold it may not hear an appeal from a denial of institution decision under “315/325”. Mandamus denied — no clear right to relief.

    What is new here is that the Board determined that 315/325 were “jurisdictional” in nature. Thus, the patentee had not waived his right to bring a late motion to dismiss. The court did not disagree.

    “We likewise see no clear bar on the Board’s treatment
    of the § 325(a)(1) proscription as a “jurisdiction[al]” limit,
    to be applied without invoking waiver based on the timing
    of the patentee’s raising of the issue. GTNX, 2014 WL
    7723800, at *2 (relying on La. Pub. Serv. Comm’n v. FCC,
    476 U.S. 355, 374–75 (1986) (agency “may not confer
    power upon itself” because “to expand its power in the
    face of a congressional limitation on its jurisdiction would
    be to grant to the agency power to override Congress”)).
    In any event, GTNX cannot find a clear, indisputable
    right to have the Board maintain the proceeding in the
    circumstances present here. We conclude that mandamus
    relief is unavailable.”

  23. This decisions is so bad it is a joke. Halliburton forbad functionally claiming the most crucial element in a combination claim — basically the point of novelty. It did not forbid functional claiming old elements. 112(f) was intended to address the holding of Halliburton – to permit the use of a functionally claimed at the point of novelty. It was not intended to overrule Halliburton to any further extent.

    One knows this because 112(f) was specifically addressed to the Halliburton case. Thus one knows the scope of the statute by reference to the holding of that case.

    Thus if an old element of a combination claim is not using “means for” or “step for,” it should not be construed as being governed by 112(f) thereby requiring some detailed description of old structure in the specification. Such a requirement is complete nonsense, because old structure does not need to be described in any detail, nor does one need to describe how to make and use old and well-known structure.

    This case is a total travesty. One more example of the Federal Circuit that simply does not get it.

    1. Ned: old structure does not need to be described in any detail, nor does one need to describe how to make and use old and well-known structure.

      Can you be specific about the text in this opinion which suggests to you (apparently) that this rule has been changed?

      I’m having difficulty getting from your analysis of the statute, case law and facts in this case to your conclusion about the majority’s opinion.

      1. OK, MM,

        By not knowing whether 112(f) applies with any certainty, the claims are made inherently indefinite.

        Consider the combination of an old bow and new arrow claimed like this:

        1. An apparatus comprising:

        a structure configured to propel an arrow with force; and

        an arrow having — there follows details of novel structure.

        The specification describes no structure at all corresponding to the functional statement. It simply says, use any apparatus that will propel an arrow with force. The invention is in the arrow.

        If 112(f) were to apply to 1, it is not clear that the claim is not indefinite because no corresponding structure is “described.” But everyone knows that such a structure would include a bow, crossbow, a ballista, or the like, and of many different configurations. Etc.

        Now if the spec says “bow,” does the claim cover a crossbow? Why do we have to get into that?

        Another example, many claims include a computer. Do we really have to describe what a computer is, and how to make an use it? What is the difference between claiming a computer and structure configured to automatically calculate an algorithm which is broader than a computer where the invention is not in the means for calculating, but in the algorithm (together with something else as in Diehr).

        A requirement to invoke 112(f) should not take place in the PTO unless the applicant is claiming something new about the functional element in defining the invention. Ditto, in court. If the invention resides at all in the functionally defined element itself, then the claim is invalid. That is a better rule. The alternative only builds a lot of ambiguity into the scope of claims.

        We need clarity and certainty to know in advance that a claim invokes 112(f). Newman is spot on that this case opens the door to a lot of ambiguity regarding the true scope of claims. It is simply one more thing to litigate.

        1. If 112(f) were to apply to 1, it is not clear that the claim is not indefinite because no corresponding structure is “described.” But everyone knows that such a structure would include a bow, crossbow, a ballista, or the like, and of many different configurations. Etc.

          The problem arises when Claim 1 issues before the ballista is invented, or if the ballista is in fact invented but unknown to this arrow inventor. Claim 1 then has the meaning at time of examination and of issuance as being directed to a bow and a crossbow. Yet you would assert that your claim covers the arrow+ballista despite the fact that nobody has done an assessment as to whether that is 1) enabled and 2) substantially similar to the bow/crossbow that you posited. Suppose the arrow has a particular structure such that it can’t be used with the ballista? It makes no sense for the claim to be invalid because of a later development, nor does it make sense for the claim to inherently cover the later development without any kind of analysis that it should.

          Instead, what has to happen is that the claim at issuance be limited to the bow/crossbow, and then you have an argument once the ballista is invented as to whether this is really an equivalent situation. I’m not saying that it isn’t also covered. I’m saying the manner of coverage has to be through equivalency and not literalness. The only manner in which you can have that is to deny the applicant the functional genus to the non-novel part at the time of filing.

          You can’t get around the fact that the scope of a disclosure is, at most, what is known to the inventor and the scope of functional language is anything which achieves the function. From a scientific standpoint one can never prove the two coextensive.

          1. There is no harm, btw, in saying that element X is selected because it achieves a function and that other known things can achieve the function, and perhaps future things will also achieve the function in similar ways. That IS, in fact, what the inventor is thinking in his mind, and is an accurate representation of what his scope should be.

            What doesn’t make sense is to simply skip the above and go directly to the functional result as if all ways of doing something for all time will be considered equal when experience dictates they wont. Nominate a means, claim it in means plus and let the art determine if equal things are equal. Assuming is what gets applicants into trouble.

            1. There is no harm, btw, in saying that element X is selected because it achieves a function and that other known things can achieve the function, and perhaps future things will also achieve the function in similar ways. That IS, in fact, what the inventor is thinking in his mind, and is an accurate representation of what his scope should be.

              What doesn’t make sense is to simply skip the above and go directly to the functional result as if all ways of doing something for all time will be considered equal when experience dictates they wont. Nominate a means, claim it in means plus and let the art determine if equal things are equal. Assuming is what gets applicants into trouble.”

              Why doesn’t that make sense. You just observed, I think, that every patent application could include the boiler plate that you proposed. If you understood that, why burden the system with all that additional text. Why isn’t one of ordinary skill in the art as clever as you and why can’t we take the position that one of ordinary skill in the art would understand that element X is selected because it achieves a function and that other known things can achieve the function, and perhaps future things will also achieve the function in similar ways and skip the above and go directly to the functional result as if all ways of doing something for all time will be considered equal?

              If the claim recites a screw is that specific enough? What about embodiments that use wood screws verses machine screws. What about 10 x 32 screws verses 8 x 24? Metric thread vs. English? Inch long vs. two inch long.

              Why is screw any better than means for helically attaching?

          2. Random, what we seem to be talking about here when claiming old elements functionally is that we are claiming known structure. (The “equivalents thereof” part is not a matter of claim construction, but of infringement. I know, the Federal Circuit is really confused on this issue as well because of cases like Alappat.) So long as we understand it this way, there should be no problem.

            However, the Federal Circuit seems to treat all means-plus-function elements in the same way. They require a that the structure be described (what level of detail has not yet been determined), and limit equivalents to structural equivalents.

            In other words, given our bow example, a crossbow may not be covered if I give a bow as an example. Perversely, giving any example or some examples actually narrows the claim.

            Take for example software. I claim an algorithm and I disclose a flowchart. That is pretty broad. But what if I described an actual program written in a particular language? Does it cover a program written in a different language? You see the problem here? The more disclosure one gives, the worse it gets, when the whole point of the broad functional statement of old elements is to not be limited to particular structure but only to particular functions in a combination.

            I must insist again that these must be known. Whether or not a ballista infringes when a ballista is in fact invented later is not a matter of claim construction but rather is a matter of infringement. The claim does not literally cover the ballista. The claim literally covers only known bows including crossbows.

            1. Ned is making an important point.

              Below is the first official IEEE 802.1d MAC Bridges Specification. It certainly provides enablement. Thousands of POSITAs implemented MAC bridges on the basis of this specification.

              The specification provides pseudo-code, packet formats, data structure definitions, and timer definitions.

              Never once does the specification describe in detail MAC packet reception or transmission, which were old well understood functions.

              DEC never patented the MAC bridge spanning tree method or device, but at some point in the past DEC probably could have. The written specification for such a patent would have referred to MAC packet reception or transmission functionally.

              It appears that DEC senior management believed creating a large open market for MAC bridge devices would facilitate DEC sales in other markets that the strategists considered more significant to DEC’s bottom line.

              link to drive.google.com

              1. Joachim, Seagate management did not enforce its patents where it wanted its competitors to standardize on its component designs. It allowed this in order that component vendors would lower its price to Seagate for components on its initial runs of new designs.

                Take a servo improvement that Seagate hypothetically invents. Seagate did not make the chips for servo systems. Thus if it wanted a vendor to puts its new servo improvement into a chip, that chip was going to cost X or X+ depending on whether Seagate allowed the vendor to sell to it competitors.

                The only way a company like Seagate to use its patents to exclude competition where the invention was in a component would be to make the necessary components in house. But that was unwise for other reasons in that Seagate’s own customers required Seagate to have second sources on its own supply chain.

                So, it gets complicated.

                And then we have some management that did not fully understand all this and wanted to have their cake and eat it too — allowing vendors to use the Seagate designs even while insisting that we preserve Seagate’s right to sue its competitors for using the components purchased from Seagate licensees. Huh?

                Advising top management on patent policy and strategy was always fun.

        2. Ned: The invention is in the arrow.

          I humbly suggest applicants learn to claim their inventions rather than hide their inventions amidst unnecessary and unduly broad recitations of the prior art.

          Once again: the pressure put on the patent system for the past couple decades comes primarily from patentees attempting to claim information and information processing logic. The system was never designed for this and, in fact, for most of its lifetime the system was understood and acknowledged to exclude these things. Then along came the Internet and suddenly (surprise!) up popped up some folks intent on controlling access to (i.e., “monetizing”) information and information processing by any means necessary, including using the patent system.

          To get around the established rules, applicants began playing all sorts of games to hide the ball. The claim you described is an example of one of those games. If this case makes that claim harder to obtain and enforce, then so be it.

          That said, I’m still waiting for you to show me the offending passage in the opinion which establishes the new rule you seem to believe was established.

          1. MM, regarding old elements in claims, a lot of the times the claim drafter does not know what is truly novel. This emerges during prosecution. At the end, it usually is the case that some of the original elements in the claim are old and serve no purpose in the claim — but they still have to be proved for infringement purposes. But, if one tries to remove these “unnecessary” elements from the claims after a NOA, most times the examiner will balk. They simply will not allow it. My luck is about 25-75. I just had an examiner balk when I took out a host of unnecessary limitations in the allowed claims. I filed an RCE and got an immediate NOA.

            As to the passage in the opinion that I am concerned about, consider this:

            “It has shifted the balance struck by Congress in passing
            § 112, para. 6 and has resulted in a proliferation of functional
            claiming untethered to § 112, para. 6 and free of
            the strictures set forth in the statute.”

            What is this balance? The Federal Circuit never comes out and actually discusses the holding of Halliburton leaving the implication that 112(f) was intended by Congress to apply to situations where the Supreme Court had no concern. See, e.g., Faulkner v. Gibbs, 338 U.S. 267, 70 S. Ct. 25, 94 L. Ed. 62 (1949) that expressly limited the holding of Halliburton to the precise point of novelty.

            When the “invention,” as opposed to a particular element in a claim, is claimed functionally we have a problem of indefiniteness and undue breadth. But when an old element is claimed functionally, most everyone knows what structure or set of structures is being claimed. But it must be understood that the element in question is old.

            For example, in the Microsoft case, one of the structures being claims was a cache. Everyone knows what that is in the computer arts. It can be claimed as a cache, or as a memory device configured to contain recently used data such that that data is used rather data located in a remoter location upon a request for that data. What fricken difference should there be if the invention has nothing to do with the particular structure of a cache. I presume that the spec. only disclosed a black box with the word cache on it. What more does one really need?

            The vice of functional claim emerges when one considers what Halliburton invented: a mechanical tuner. But he claimed “means … for tuning … to clearly distinguish the echoes from said couplings from each other.” Everything else in the claim was old, and old in combination: the oil well tubing having collars, means (dynamite) for inducing impulses in the tubes, and the means for receiving the impulses and determining the time lapse between echoes. Had he placed the old elements before the colon, we would instantly see a single means claim.

            Single means claim is the problem of Halliburton claims. The Supreme Court said no to such a claim where the most “crucial element” is functionally claimed. Congress said, OK — but use means plus function to “signal” that one is really claiming the novel structure described in the specification. The signal is important. It is that signal that tells us to look to the specification for the details.

            Of course, any applicant MAY use 112(f) anywhere. But it is not required to be used anywhere. If it not used, and the claim amounts to a single means claim, it is invalid.

            1. The grain of truth in the Lie: single means claims are forbidden.

              Beyond that though, Halliburton was abrogated by Congress (see Federico), and the notion of PON is a canard, as the novelty is the claim as a whole.

              At least that is the law according to the words of the branch of government that has the actual authority to write patent law (hint: it’s not the judiciary).

      2. I am not sure where to jump in here. Apologies for not giving credit to the good comments above and below. I agree with Newman and Ned among others. The old law provided a nice straightforward way to decide. This decision will create all kinds of mischief. It seems like the lazy way to invalidate a claim that seems obvious. (As does the invention to me in hindsight.)

        To MM’s point on old structure, in response to patentee’s expert: “But the fact that one of skill in the art could program a computer to perform the recited functions cannot create structure where none otherwise is disclosed,” at page 19. The court did not dissect the claims into old and new.

        Suppose the spec concludes with “None of the claims avail themselves of 35 USC 112(f).” Is that enough? Suppose there are nonce words. (Would an etymologist on the panel explain why the judges are using the word “nonce,” meaning “for the once” correctly?) 20,000 US patents contain “module” in a claim but do not contain “means.” Are these now all in jeopardy?

        An electrical engineer might claim a system having an element described as a “power supply module capable of delivering 12 volts DC at 1 amp.” If “module is left out, is this now OK? Clearly, leaving out “power supply” is a problem.”

        Although there are cases which use “means” carelessly, e.g., “fastener means,” at least it is clear. What about “fastener?” Is this now a nonce word? If so, one must illustrate, or at least mention, bolts, nails, screws, glue, etc., etc. to avoid being indefinite or limited to the one choice and equivalents.

        1. “A nonce word is a lexeme created for a single occasion to solve an immediate problem of communication, or “an invented or accidental linguistic form, used only once”. All nonce words are also neologisms. Some nonce words have a meaning and may become an established part of the language, while others are essentially meaningless and disposable and are useful for exactly that reason, for instance in child language testing. The term nonce word was apparently the creation of James Murray, the influential editor of the Oxford English Dictionary.”

          The courts use “nonce word” to characterize an effort to obviate the rules of functional claiming by substituting some less obvious word or phrase for “means”.

          1. >The courts use “nonce word” to characterize an effort to obviate the rules of functional claiming by substituting some less obvious word or phrase for “means”.

            That is a circular definition.

            Not only that module is like circuit. It conveys plenty of structure.

            The problem is that we now have a court that is largely ignorant of science and patent law. The Google judges have never practiced patent law and never took an interest in science. They appear to not grasp the fundamentals of patent law.

            Abolish the Fed. Cir. It is time. Obama has loaded it with anti-patent anti-science witch burners.

    2. This decisions is so bad it is a joke. Halliburton forbad functionally claiming the most crucial element in a combination claim — basically the point of novelty.

      The Point of Novelty rule is only a specific instance of the broader rule Ned.

      The question is always always always “Does the claim define a particular structure of achievement.” When the PON is claimed functionally it never can because it’s attempting to claim not only the invention but other inventions as well.

      But the same threat MAY exist outside the PON as well. When the threat occurs, the rule still applies. This should be obvious for at least the reason that 102 is not the only concern related to patentability. A structure that is non-obvious or definite when applied to a car may not be when applied to a plane. It is no less an attempt to be over-inclusive of your solution by talking about the object in vague terms as it is to talk about the subject in vague terms. Now it may be that threat is less when the feature is not a novelty, but certainly not so much that is becomes a per se rule that you only apply it to the PON.

    3. This is certainly an interesting description of “structure”:

      While portions of the claim do describe certain inputs and outputs at a very high level (e.g., communications between the presenter and audience member computer systems), the claim does not describe how the “distributed learning control module” interacts with other components in the distributed learning control server in a way that might inform the structural character of the limitation-in-question.

      So “structure” can be satisfied by describing how the “module” interacts with other “modules.” I suppose that something like this would suffice:

      A system for presenting messages to a user, comprising:

      a decryption module that, upon invocation with a message encoded with a bidirectional encryption technique, decodes the message with the bidirectional encryption technique to produce a plaintext message;

      a messaging user interface that, responsive to receiving a plaintext message, presents the plaintext message to the user; and

      a decoder module that:
      receives an encrypted message from a message source;
      invokes the decryption module with the encrypted message; and
      responsive to receiving from the decryption module a plaintext message produced from the encrypted message, delivers the plaintext message to the user interface.

      …accompanied by a structural diagram showing the interactions of the modules.

      Apparently, logical architecture = details of the interaction of the modules = structure.

      1. Apparently, logical architecture = details of the interaction of the modules = structure.

        Program = method

        Structure = definite algorithmic steps taken, i.e. take input X, do Y to it to get Z, add Z to A…

        It wouldn’t be enough to say how they interact, but the inputs and outputs are 2/3rds of the way to describing the “black box” that is the program.

        1. What is shocking about all this is all the people that want to tell people skilled in the art how to describe information processing machines. This from science illiterate judges appointed due to Google’s massive bucks being poured into DC.

          You want to know how to describe it? http://www.elance.com. Read.

        2. Random (again) confuses himself with “program = method

          Absolutely and unequivocally false.

          A program being executed is just NOT the same thing as the program itself.

          Software is a manufacture in its own right.

          Let’s try to have a little inte11ectual honesty on this most basic point ladies.

  24. OT: CAFC’s non-precedential opinion in Alexsam v. The Gap came out today. This was the second opinion in a week, I believe, where the Federal Circuit chucked an East Texas jury’s finding that a patent was valid and replaced it with a finding that the patent was obvious or anticipated.

    1. The Alexsam case is interesting because the patent was held invalid based upon a prior invention defense, a rarity.

      What was overturned was the jury verdict that the inventor had conceived of the invention prior to the data reduction to practice by the third-party system. The Federal Circuit held that the inventor had not corroborated his claim of prior invention of at least two of the [novel] elements of the claim.

  25. “That characterization is unwarranted, is uncertain in meaning and application, and has the inappropriate practical effect of placing a thumb on what should otherwise be a balanced analytical scale. It has shifted the balance struck by Congress in passing § 112, para. 6 and has resulted in a proliferation of functional claiming untethered to § 112, para. 6 and free of the strictures set forth in the statute. ”

    Unwarranted? Of course its warranted. Its warranted by the statute! If you want to invoke 6th paragraph, say means. If you don’t, don’t.

    The problem is once again with the courts. Apparently, the courts chose to ignore the presence or absence of the word “means” a time or two and construed the claims the other way because they wanted to “justify” a particular result.

    Solution: Make the presence or absence of the word means even less of an indicator. Ignore the statute. Why not. It works for 101.

    1. If you want to invoke 6th paragraph, say means. If you don’t, don’t.

      You’d think if Congress had intended such a technical reading of the statute they could have made that perfectly clear. Or the great Frederico could have made it perfectly clear in his awesome commentary.

    2. The problem is once again with the courts.

      Don’t forget: this is the same court that created all kinds of convenient fictions so that the software patent lovers could roam wild and pretend they were the most important people in the history of the world. That happened even while people like me were explaining — in real time — that this was going to turn out badly because the exceptions created to benefit a tiny, tiny, tiny number of people would ruin the system to the detriment of pretty much everybody else.

      Reap the whirlwhind. Newman’s complaints about “uncertainty” and dashed expectations fall pretty flat at this late date.

  26. Without intending to, Newman’s dissent presents what may be the best evidence showing that 112P6 did not represent a “green light” for bona fide functional claiming at the point of novelty but rather the exact opposite.

    Frederico’s Commentary:

    It is unquestionable that some measure of greater liberality in the use of functional expressions in combination claims is authorized than had been permitted by some court decisions and that decisions such as that in Halliburton Oil Well Cementing
    Co. v. Walker, 67 S.Ct. 6, 329 U.S. 1, 91 L. Ed. 3 (1946), are modified or rendered obsolete, but the exact limits of the enlargement remain to be determined.

    Even Newman’s own spin acknoweldges the qualified aspect of the changes:

    Federico explained that paragraph 6 enlarges the opportunity to claim a function, but limits how that function is supported and construed

    The fundamental principle underlying the core holding in Halliburton — the proscription against functional claiming at the point of novelty — are unchanged because without them the system turns into a casino for unskilled speculators and free-riders (hence the rush of patent attorney “innovators” into the system when the CAFC foolishly opened the gates). What the statute terminated — and rightfully so — was the ability of a court to identify a single functional element anywhere in a claim (i.e., a phrase such as “means for”) and declare the claim invalid without further analysis.

    Newman ends with this interesting point:

    The court is not powerless to require software claims to comply with the statutory requirements of description, enablement, definiteness, unobviousness, etc. If there have been abuses, as the majority states, the remedy is not to eliminate the statute, but to apply the statute.

    Take the lead, Judge Newman. An algorithm isn’t a structure, as everyone in the world knows. Now go ahead and apply the statute.

    1. MM, but Newman is right. Claiming old elements functionally does not implicate any undue breadth or indefiniteness issues at all. The structure of old elements and how they function in a combination are by definition well known, and the details do not even have to be described the specification so that one of ordinary skill in the art would know how to make and use the old element. Describing how the old element actually functions in the combination at times improve the understanding of the claim.

      As you said in your remarks, what the Supreme Court was concerned about in Halliburton and its predecessor cases was a claim that was functional at the point of novelty. The Supreme Court was not really concerned with new-combination claims or with old elements of either a new or old combination.

      The fact that the Federal Circuit has never looked at this issue seriously was noted by Judge Reyna. The new rule will result in what I believe to be a miscarriage of justice.

      1. Claiming old elements functionally does not implicate any undue breadth or indefiniteness issues at all.

        That’s not what this case was about.

        1. MM, and that is precisely WHY the Federal Circuit so often gets it wrong. It Legislates — it does not decide cases.

          1. I don’t see where this case says that you can’t claim old elements “functionally”, i.e., in a manner that sufficiently relates the known structure of that element to the skilled artisan.

            Where do you find this holding in the opinion?

            1. MM, come on now. The vast majority of old elements in most claims are functionally claimed. Now they will have to have corresponding structure described or else. Further the scope of MPF equivalents is limited to structural equivalents of the described structure. Now that is preposterous.

              The problem with the opinion is that it does not limit it holding to new elements of a combination.

              Further, consider a method claim per below. Does such a claim invoke 112(f)?

              1. Method comprising;

              accelerating a spindle to a rpm between X and Y;

              introducing liquid Q;

              continuing to rotate the spindle for R minutes;

              reducing spindle rpm to Z for S minutes; and

              stopping the spindle, and retrieving from the spindle purified plutonium.

    2. >An algorithm isn’t a structure, as everyone in the world knows.

      Except those that live in the real world. We all know that your brain has no structure MM. Your information processing occurs in the spirit world with no structure associated with your information processing.

      (Get the sticks and matches for a witch burning!)

      Just shameful lack of understanding of basic science. Stomach turning.

  27. Crazy stuff below. So, if one skilled in the art knows how to make say door hinges “cannot create structure where none otherwise is disclosed.”

    What? So, solutions that are enabled by the spec for one skilled in the art have no structure according to this opinion.

    At this point I am VERY strongly in favor of abolishing the Fed. Cir. Thanks Obama.

    Dr. Souri also testified that, “as
    one of ordinary skill in the art, reading the specification, I
    would know exactly how to program” a computer to perform
    the recited functions and further testified that
    structure “could be in software or it could be in hardware.”
    J.A. 1391 (256:12–258:16). But the fact that one
    of skill in the art could program a computer to perform
    the recited functions cannot create structure where none
    otherwise is disclosed. See Function Media, L.L.C. v.
    Google, Inc., 708 F.3d 1310, 1319 (Fed. Cir. 2013).

    1. structure “could be in software or it could be in hardware.”

      “The structure could be some instructions or it could be some doohickey.”

      You gotta love that kind of clear, detailed, and definite description. Not hand-wavey at all.

    2. door hinges

      Pretty sure people skilled in the art of making doors have a reasonable, sufficient and shared understanding of the objective physical structures falling into that category.

      Of course, if your innovation was an “iron door hinge with increased rust-resistance under strong oxidizing conditions” then you bet your sweet patooty you’re going to need more physical description of your hinge and some supporting data as well. Now go ahead and explain to everyone how such a requirement is antithetical to “patent based innovation.”

      1. >>Pretty sure

        I am positive that

        “people skilled in the art of information processing [making doors] have a reasonable, sufficient and shared understanding of the objective physical structures falling into that category. ” for modules and for the structure recited in the claims of this patent.

        1. Please describe for everyone — in objective physical terms — the structure of one example of the hardware such that anyone looking at the hardware would know that the hardware has the functionality ascribed to it here.

          Do the same for an example of software.

          I’m no carpenter but I can draw you a decent picture of a door hinge such that 99/100 other people will know what it is.

          Maybe — just maybe — information processing logic doesn’t fit very well into a patent system that was created and designed for something else entirely. Truly a radical thought, I know …

          1. This has been covered like a million times and you have lost every time.

            Please tell us if there is a structural difference between two resistors in parallel and two resistors in series.

            Moreover, your assertions are just ridiculous on their face. The machine can’t perform different functions without a change in structure. The machines are performing information processing that is beginning to rival human abilities–but somehow this is done without structure. I suppose your brain has no structure.

            This is like going back to medieval Europe and arguing with a witch burner. Only the science illiterate Google judges—thanks Obama–could come up with this stuff.

            1. Please tell us if there is a structural difference between two resistors in parallel and two resistors in series.

              “Please tell us if there is a structural difference between our drug that cures cancer and the drugs that don’t.”

              Yawn.

            2. The machine can’t perform different functions without a change in structure. The machines are performing information processing that is beginning to rival human abilities–but somehow this is done without structure.

              And what is that structure? The question isn’t “is there a difference?” The question is “what does the difference look like?” Define the structure.

              You’re way off the rails as usual. You forget that you can’t claim ends, you can only claim means. A module is saying “any programming that achieves this end” which gives no definition to the means at all.

              1. Random, you and I have been through this before. Tell me should the claim be interpreted by one skilled in the art?

                Patent applications are supposed to be for what is new. What is known by one skilled in the art is not supposed to be put in there.

                Endless garbage flows from you.

  28. Awesome news and congrats on a successful amicus brief.

    This was a much needed correction. Now all the CAFC has to do is stay the course and avoid creating further loopholes whereby applicants freely substitute functionality for structure at the point of novelty in their claims. If she doesn’t like this, I expect Newman is going to feel greater disdain for the other corrections that are surely coming down the pike.

  29. A good knowledge of English philology is helpful here. English does not have a complete linguistic mechanism for constructing paronyms and often interposes a layer of Latin derivation.

    We must look at Latin to understand what module really means.

    Module << Modulus << Modus + ulus (diminutive).

    Modus definition 9 from Lewis & Short: A manner of performing an action, method, mode, way.

    The meaning of module has expanded with time. We can talk about hardware modules (means/mechanisms/components) for constructing a stereo receiver, software modules (mechanisms/libraries/means/components) for doing pattern matching, or mechanical modules (means/components) for recreating vintage cars.

    Practically every invention uses standard components in some way, but "module" in Claim 8 certainly looks like a nonce word equivalent to "means".

    1. >>We must look at Latin to understand what module really means.

      Wow, so do you have an education in computer science? I do. Module is not a nonce word. And try to address my comment about it being used consistently and often for real software projects such as on elance.

      1. My understanding of the caselaw is that you can still disclose a general purpose processor as the “means” (or “module” in this instance) for performing the function. You just need to disclose an algorithm for performing the function.

        1. You just need to disclose an algorithm for performing the function.

          Question: Any algorithm? In any format?

          This is the area in which Judge Newman’s expectations are going to be smashed against the rocks in the very near future.

        2. The problem is they are trying to exclude what a person of ordinary skill in the art knows and exclude it as structure. That is their game from the Lemley functional paper.

          Lemley wrote a paper as a guide for judicial activism and they are following it.

          1. The problem is they are trying to exclude what a person of ordinary skill in the art knows and exclude it as structure

            The problem is that there is no actual definite physical structure that is known by “people of ordinary skill.”

            It wasn’t necessary for you to prove that to anybody, but you did so when you were asked repeatedly over the years to tell everyone what these “known” corresponding physical structures are and you failed every time. This is at the same time you are crowing about your skill in the art and berating everyone else for lacking it.

      2. MIT hired me as faculty back in the 80s because of my knowledge of computer science, telephony, and computer networking.

        Module means a lot of different things in software or hardware engineering.

        Because the patent specifically mentions Java, module might relate to the Java Module System, which specifies a distribution format for collections of Java code and associated resources.

  30. >>It replaces the term “means” with the term “module” and recites three
    functions performed by the “distributed learning control
    module.”

    Except, try to look at elance.com. Programmers use the word to specify programming jobs. The word is not a nonce word and does connote a structural meaning or it wouldn’t be used to specify programming jobs—as in reality.

    Reality—functional terms are used to specify the programming tasks just like in the application. The meaning is that the function can be implemented by one skilled in the art.

    So, we have the fed. Cir. refusing to understand information processing and using their nonsense to limit patents—again.

    1. The word … does connote a structural meaning or it wouldn’t be used to specify programming jobs

      The argument doesn’t make sense given that “programming” is writing instructions for computers.

      You could also state: “module does connote a structural meaning or else it wouldn’t be use to specify a grammar lesson for third graders” and your logic would fail for the same reasons.

      Learning modules don’t have “structures” in the sense that matters for patents, just as sentences, stories and logical arguments don’t have “structures” in the sense that matters for patents.

      1. >The argument doesn’t make sense given that “programming” is writing instructions for computers.

        This again is a psychotic post from you MM. There is simply no meaning in your words. It is like talking to a judge at a witch burning.

        Simply no connection with the real world.

      2. Try to remember too MM that just because Google judges on the Fed. Cir. make some ridiculous holding doesn’t make it reality any more than holding that the sun revolves around the earth.

        (Ned, notice that Newman discusses Haliburton and says it was overturned by the 1952 Patent Act.)

        1. Night,

          I am surprised to find myself agreeing with one of your posts. The reality-defying “ridiculous holding” that comes to mind for me:

          “[Software] programming creates a new machine, because a general purpose computer in effect becomes a special purpose computer once it is programmed to perform particular functions pursuant to instructions from program software.” In re Alappat

      3. >>Learning modules don’t have “structures” in the sense that matters for patents, just as sentences, stories and logical arguments don’t have “structures” in the sense that matters for patents.

        Sure so a machine that processes information using time, space, and energy, (and that must have structure) is the same as a story. OK. Sure.

        Is your brain the same as a story? Again, we have the mystical spiritual world appearing.

        1. NW: a machine that processes information using time, space, and energy, (and that must have structure) is the same as a story,

          More accurately, a one-sentence story about a machine that processes information but which fails to describe any novel structure for that machine is the same as “a story”, at least for the purposes of a sane patent system.

          1. Wow, that was worth a belly laugh MM! A story is the same as an information processing machine.

            You know MM, judicial activist that distort reality like this are perhaps the lowest form of attorneys. If you and you ilk went to Congress and changed the law, well–not good–but OK I can respect that. But, making up nonsense and disregarding previous laws and the Constitution are not actions I can respect. No one should.

            1. the lowest form of attorneys

              I think everyone can see that you just attacked something that you wrote rather than something that I wrote.

    1. Note: have not read the case yet, so this may be moot, but given the vintage here of accepting the “interpretation” for so long, and Congress, having the opportunity to change the actual written words – and choosing not to – is this new “reinterpretation” in fact re-writing of the law?

      Someone needs to take a look at how much power over statutory law the judicial branch is wielding (O’Malley warning anyone?)

      1. Well, anon, you are right here about ignoring the statute and the intent of congress. Clearly the statute says that 112(f) is an optional choice of the applicant. If one does not use the magic words, one does not get the benefit of the statute.

        As correctly noted by Newman, other statutes govern whether functionally defined claim terms are invalid. Obviously, a functionally defined claim term describing an old element in a combination is not only not indefinite, but it is fully supported.

    2. If the spec discloses some structure, the claims should be fairly safe, unless 112 equivalence is somehow narrower than DOE equivalence.
      Computing means claims are more at risk, until the court acknowledges that everyone who rights and reads those claims knows exactly what structure is used to accomplish the computing function. Many means-plus-function terms immediately call forth structure blatantly obvious to the artisan, and this post hoc insistence that reciting things like “means for driving a screw” are indefinite unless the drafter describes a “screw driver” is just childish. The silliness is hidden in fancy language when the court talks about modules and algorithms.

Comments are closed.