« Mixed Claim Types | Main | Commissioner Cack »

March 30, 2008

An Algorithm Shy

Australia-based Aristocrat threw away money getting gaming machine patent 6,093,102, and then threw away money trying to enforce it. Means-plus-function with no unction. From the CAFC, a crash course in means-plus-function claims for computers: the claimed function is bounded by the specifically disclosed algorithm. Lacking an algorithmic tether, a means-plus-function processor claim is indefinite.

Aristocrat v. International Game Technology (CAFC 2007-1419)

The game disclosed in the '102 patent purportedly increases player interest in slot machines by providing the player with greater control over the definition of winning opportunities. It allows the player to define the winning opportunities based on symbols displayed on the top and side of a multi-line screen representing slot machine reels. Using the invention on a 3x5 screen, for example, the player can define numerous different arrangements that will allow the player to win for some subset of the 243 possible winning combinations. The player can do so by selecting symbol positions and thereby activating winning opportunities for combinations in which the symbols are not necessarily aligned with one another. The only constraint is that the selected combination must contain at least one symbol from each column. Figure 2 from the '102 patent shows a 3x5 screen with selections, and the figure on the right shows one of the winning combinations for the selection in Figure 2.

'102 had only one independent claim, the crucial element a "game control means," aka "control means."

A gaming machine
    having display means arranged to display a plurality of symbols in a display format having an array of n rows and m columns of symbol positions,
    game control means arranged to control images displayed on the display means,
    the game control means being arranged to pay a prize when a predetermined combination of symbols is displayed in a predetermined arrangement of symbol positions selected by a player, playing a game, including one and only one symbol position in each column of the array,
    the gaming machine being characterised in that selection means are provided to enable the player to control a definition of one or more predetermined arrangements by selecting one or more of the symbol positions and
    the control means defining a set of predetermined arrangements for a current game comprising each possible combination of the symbol positions selected by the player which have one and only one symbol position in each column of the display means,
    wherein the number of said predetermined arrangements for any one game is a value which is the product k1 . . . x . . . ki . . . x . . . km where ki is a number of symbol positions which have been selected by the player in an ith column of the n rows by m columns of symbol positions on the display (0 < i ≤ m and ki ≤ n).

The district court noted that the parties agreed the term "control means" is a means-plus-function term that invokes 35 U.S.C. § 112 ¶ 6. As such, the scope of that claim limitation had to be defined by the structure disclosed in the specification plus any equivalents of that structure; in the absence of structure disclosed in the specification to perform those functions, the claim limitation would lack specificity, rendering the claim as a whole invalid for indefiniteness under 35 U.S.C. § 112 ¶ 2. See In re Donaldson, 16 F.3d 1189, 1195 (Fed. Cir. 1994) (en banc).

The court noted that there were slight linguistic differences in the parties' characterizations of the functions performed by the "control means," but that the differences were unimportant, because there was no adequate disclosure of structure in the specification to perform those functions, regardless of how they were defined.

The district court poohed Aristrocrat's hand waving that there was "appropriate programming."

The court ruled that "[m]erely stating that a standard microprocessor is the structure without more is not sufficient." In particular, the court noted that the specification did not create any specific structure or new machine because "it does not set forth any specific algorithm" for performing the recited function.

On appeal, Aristocrat argued that a general purpose microprocessor was sufficient for satisfying § 112 ¶ 6. Laughter peeled from the appeals court panel. Such a claim would be so broad.

In cases involving a computer-implemented invention in which the inventor has invoked means-plus-function claiming, this court has consistently required that the structure disclosed in the specification be more than simply a general purpose computer or microprocessor. The point of the requirement that the patentee disclose particular structure in the specification and that the scope of the patent claims be limited to that structure and its equivalents is to avoid pure functional claiming. As this court explained in Medical Instrumentation & Diagnostics Corp. v. Elekta AB, 344 F.3d 1205, 1211 (Fed. Cir. 2003), "If the specification is not clear as to the structure that the patentee intends to correspond to the claimed function, then the patentee has not paid the price but is attempting to claim in functional terms unbounded by any reference to structure in the specification." See also Biomedino, LLC v. Waters Techs. Corp., 490 F.3d 946, 948 (Fed. Cir. 2007) ("[I]n return for generic claiming ability, the applicant must indicate in the specification what structure constitutes the means."). For a patentee to claim a means for performing a particular function and then to disclose only a general purpose computer as the structure designed to perform that function amounts to pure functional claiming. Because general purpose computers can be programmed to perform very different tasks in very different ways, simply disclosing a computer as the structure designated to perform a particular function does not limit the scope of the claim to "the corresponding structure, material, or acts" that perform the function, as required by section 112 paragraph 6.

That was the point made by this court in WMS Gaming, Inc. v. International Game Technology, 184 F.3d 1339 (Fed. Cir. 1999). In that case, the court criticized the district court, which had determined that the structure disclosed in the specification to perform the claimed function was "an algorithm executed by a computer." The district court erred, this court held, "by failing to limit the claim to the algorithm disclosed in the specification." Id. at 1348. The rationale for that decision is equally applicable here: a general purpose computer programmed to carry out a particular algorithm 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." Id., quoting In re Alappat, 33 F.3d 1526, 1545 (Fed. Cir. 1994). The instructions of the software program in effect "create a special purpose machine for carrying out the particular algorithm." WMS Gaming, 184 F.3d at 1348. Thus, in a means-plus-function claim "in which the disclosed structure is a computer, or microprocessor, programmed to carry out an algorithm, the disclosed structure is not the general purpose computer, but rather the special purpose computer programmed to perform the disclosed algorithm." Id. at 1349.

In a later case, this court made the same point, stating that a "computer-implemented means-plus-function term is limited to the corresponding structure disclosed in the specification and equivalents thereof, and the corresponding structure is the algorithm." Harris Corp. v. Ericsson Inc., 417 F.3d 1241, 1253 (Fed. Cir. 2005). The court in that case characterized the rule of WMS Gaming as follows: "[T]he corresponding structure for a § 112 ¶ 6 claim for a computer-implemented function is the algorithm disclosed in the specification." 417 F.3d at 1249.

Aristocrat was not required to produce a listing of source code or a highly detailed description of the algorithm to be used to achieve the claimed functions in order to satisfy 35 U.S.C. § 112 ¶ 6. It was required, however, to at least disclose the algorithm that transforms the general purpose microprocessor to a "special purpose computer programmed to perform the disclosed algorithm." WMS Gaming, 184 F.3d at 1349.

Aristocrat's sorry arguments confused outcome with algorithm, confused enablement with § 112 ¶ 6 requisite. Aristocrat's cited case law was off-point.


Lead counsel for Aristocrat was Brian E. Ferguson of McDermott, Will & Emery.

Posted by Patent Hawk at March 30, 2008 1:02 AM | § 112


Hawk, help me with one thing, will you. The claim: it's not invalid under 101, 102, 103, right? It is clear, and of scope commensurate with the contribution to the art. The supporting disclosure is sufficient for the skilled reader to put the invention to use. In any other country, that would be enough. The inventor has met all reasonable public policy requirements. It is just this extra 112 hurdle that the inventor couldn't clear, right? So, what's the point of the hurdle?

Posted by: MaxDrei at March 30, 2008 2:41 AM

Hi Max:

35 U.S.C. § 112 ¶ 6: “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.”

§ 112 ¶ 6 means-plus-function claims are a short-cut claim form, allowing means that reference the specification for the claimed the means, without requiring that the claim spell the means out.

For a computer processing means, a general-purpose processor means would encompass practically everything, because a processor can be programmed to do any sort of processing. So, for a valid means computer processing means under § 112 ¶ 6, that means is limited to the algorithm(s) disclosed. Aristocrat’s ‘102 patent didn’t disclose an algorithm. It only disclosed what possible outcomes may be had. Hence, ‘102 was invalid for not meeting § 112 ¶ 6, as it was claiming unbounded functionality.

“supporting disclosure is sufficient for the skilled reader to put the invention to use” is enablement.

Hope that helps.

Posted by: Patent Hawk at March 30, 2008 10:29 AM

My understanding is that a simple flowchart probably would have been enough to satisfy the court. For example, display selectable tiles on screen --> input user selection of tiles --> calculate ..., etc. The patent doesn't provide any evidence of any operations, even at a high level, that would be performed by the general purpose computer.

Posted by: anon at March 30, 2008 4:38 PM

anon is right, the bar wasn't very high. Hence my comment on Aristocrat stupidly throwing away money.

It's no guess as to my regard for the prosecutor and any litigator who thought this patent was a grade above junk.

Posted by: Patent Hawk at March 30, 2008 5:32 PM

Means-plus-function is, in general, a crutch for the lazy. In exchange for an easy claim format, you get a tougher rule of interpretation. The only thing surprising about the 112/6 decisions in the last few years (at last!) is the surprise of so many prosecutors that the law really is the law.

Posted by: Steve Sereboff at March 30, 2008 7:08 PM

Sorry to be so dense, but what I'm driving at is "What's the point" of 112.6 and it's case law. You say how stewpid the Applicant and his attorney were, for neglecting to insert the magic fix for the USA into the app, but think of the intended addressee, the PHOSITA. Compare the doc, before and after the magic fix. Makes absolutely no difference to the information content for the intended "technical" reader. Only makes a difference for the lawyers. Form over substance, or what? I see the same thread is now running, on Dennis Crouch's thing.

Posted by: MaxDrei at March 30, 2008 10:38 PM

OR: putting it another way, if the evil is "unbounded functionality" does 112,6 really fix it? OR: putting it another way: now that you have Liebel-Flarsheim, isn't 112,6 superfluous?

Myself, I can't see the evil. Nothing wrong with "unbounded functionality" if that claim defines clearly a contribution to the art that embraces nothing old, nothing obvious and nothing of subject matter excluded by Statute, and what's inside the claim is apparent enough to a PHOSITA that it can be practised, over the full width of the claim. Is this an example of a claim that is valid everywhere else in the world, but fails in the USA and ONLY because of 112,6?

Posted by: MaxDrei at March 31, 2008 12:49 AM

In contrast to Steve's bleak view of MpF as being a lazy man's tool, I would argue that properly used it is an incredibly effective technique. S/ware is one example. Properly done you lay out your algorithm in pages and pages of spec and a few flow diagrams, and then pop it all into the claim with a MpF, like Aristocrat didn't do.

Or where you have disclosed 10 different structures for carrying out a common fxn, you cover them all in 1 MpF claim. If the new rules on limitations of claim numbers ever pass Judge Cacheris' scrutiny, we'll all be using lots of MpF's.

From Max's point of view, I believe in writing claims for EPO etc. one has to include the reference numerals or fig numbers in the claim to point to the structure in the spec. What ref number would Aristocrat have used for control means -- the one for a CPU? Sounds indefinite to me. You may have structure, but there is no limitation [algorithm] claimed distinguishing that structure from the CPU in my laptop. The algorithm is the limitation of the CPU. It wasn't claimed.

I've not given it a lot of thought, but what is coming across from the CAFC is that enablement is determined with respect to PHOSITA, but definiteness is an objective criterion (independent of skill level) that must adequately link the claim to the description.

Interestingly, Max uses "112.6" or "112,6" to designate paragraph 6. This section is so annoying. Americans are in the habit of writing "paragraph 6." Why in the name of Zeus didn't Congress number those paragraphs? I mean we have 102(a), 102(b), etc, and similar designations throughout the whole flippin' US Code. Why did they screw 35 USC 112?

Posted by: Babel Boy at March 31, 2008 8:49 AM

Babel, there is no "requirement" for ref numbers in the EPO. Where the nature of the case permits, ref numbers are inserted but the EPo Exr can insist on them only because the Statute lays down that they cannot be used to interpret the claim. As you say though, linking claim features with drawing features is a nice discipline that can expose drafting weaknesses.

Posted by: MaxDrei at March 31, 2008 9:03 AM

"Means-plus-function is, in general, a crutch for the lazy. In exchange for an easy claim format, you get a tougher rule of interpretation."

I'm not sure what "tougher rule of interpretation" you get when using mpf.

This is a very simple decision by the court: if you want to use mpf, you have to disclose structure (acts) corresponding to the recited function (steps), and the disclosure of, "well, you'd just program a computer to do it" doesn't cut it. That was the law before this decision, and will remain the law.

The amount of blather this decision has generated is really quite startling.

Posted by: JD at March 31, 2008 1:15 PM

I agree with anon that a simple flowchart would have completed the disclosure. Over the last 35 years, I have prepared countless logic block diagrams and/or flowcharts with inventors to substantiate their methods implemented in algorithms and filed the application with both method and "means" claims. The flow chart blocks illustrate each claimed means and step, and the exercise identifies logical problems in the broadly contemplated "invention" that is typically not yet reduced to practice. This prompts the inventors to actually illustrate steps that work and can be claimed, increasing the value of the patent.

Following this process, I have often been criticized by clients for burdening the inventors and putting too much disclosure into an application as clients believed (due to more recently popularized theories) that the patent is less vulnerable to attack and claim scope is broader if less is disclosed (plus they want to do it on the cheap). Decisions like this vindicate old school techniques of preparing patent applications with sufficient disclosure that can be claimed. (Remember the ancient adage that patent specifications are wells from which claims may be drawn?) Think of how valuable the claims might have been in this patent if the "control means" were illustrated in a few steps and the steps were broadly recited in the claims -- no lessening of scope, immunity from this attack, and the opportunity to draft possibly more valuable alternative sets of claims.

In the past, EPO examiners often insisted on
recasting claims in EPO "characterized by" structure and putting numerals by the claim elements. This exercise often causes the drafter to realize that the specification is insufficient to support the claims. In this case, no "control means" is illustrated in the drawings, and if it had been put in as a block of a block diagram including the keyboard, display, etc., the drafter might have been prompted to expand the control means block into its component steps and prepare the claims properly.

Unfortunately, clients are not willing to pay for the EPO claim drafting effort at the time that a U.S. or PCT application is prepared. Gaping holes in disclosure are often pointed out by astute European counsel reviewing the application. When I started practice pre-EPO, I was fortunate enough to be trained extensively both in U.S. and foreign (particularly German) patent application drafting. So my U.S. applications were subsequently written with an eye as to how they needed to be revised for non-US filing. Apparently, that training is absent today in most U.S. firms and corporate patent groups.

As an aside, where are the method claims in this patent? Again, once method claims are prepared for an application, then one must go back and insert the steps in the specification and illustrate them in the drawings.

Probably, in this situation, the Australian client would have resisted doing any of the above, given Australian patent norms and cost issues, even if advised to do so. Hopefully, these decisions will jog U.S. practitioners out of being lulled into the false sense of security that the less disclosed the better and so advise their clients.

Posted by: jfbpatent at April 1, 2008 7:17 AM