Case law you can use: improper for Examiner to rely on the same structure to disclose two separate claimed elements

Takeaway: The BPAI has interpreted Federal Circuit case law to say it's improper for the Examiner to rely on the same structure as disclosing two separate claimed elements. In several decisions where this issue arose, the Board used the following cites:

Lantech, Inc. v. Keip Machine Co., 32 F.3d 542 (Fed. Cir. 1994)(in infringement context, a single conveyor held to not meet claim element requiring at least two conveyors).
In re Robertson, 169 F.3d 743 (Fed. Cir. 1999)(claim requiring three separate means not anticipated by structure containing two means where one of the two means was argued to meet two of the three claimed means).

Here are some BPAI decisions where the Board has applied In re Robertson and Lantech in this manner.

Ex parte Weideman: The claims were directed to a vehicle seat. The Board agreed that the reference disclosed the claimed seat back and seat bottom, but found the Examiner's rejection was "flawed."

The Examiner finds that the horizontally extending portions of frame members 3, 4, also identified more specifically by reference numeral 7, not only constitute part of the seat bottom, but also constitute the structure in Lohr that reads on the claimed “center support extending forwardly from [the] central pillar”. Consistent with the principle that all limitations in a claim must be considered to be meaningful, it is improper to rely on the same structure in the Lohr reference as being responsive to two different elements (seat bottom and center support) in claim 1.
(Weideman, internal citations omitted.)

Ex parte Koutsky: The claims were directed to a vehicle seat suspension apparatus. The Board agreed that the flange and rollers in the reference correspond to the claimed "means for allowing reciprocating movement," but found the Examiner's reading of the reference with respect to other claimed elements was in error:

However, the Examiner uses the same [flange] 56 and guide rail 34 as the elongated guide and guide engaging member. Thus, the Examiner has used the same element or structure in the Koutsky reference to satisfy two different limitations from the claimed subject matter. However, when a claim requires two separate elements using one element construed as performing two separate functions will not suffice to meet the terms of the claim. Accordingly, it is our finding that Koutsky does not anticipate the claimed subject matter on appeal.
(Koutsky, internal citations omitted.)

Ex parte Konstant: The claims were directed to a storage rack, which included a horizontal deck beam interconnected between two columns. The beam included an upper flange supporting a locking cross bar, and an outer face in which an aperture was located. The Examiner applied this reference:

The Examiner found the reference's support surface (18) satisfied the "upper flange" (with locking cross bar 30), and that the same support surface 18 satisfied the claimed outer face containing an aperture (20). The Board reversed the rejection:

The Examiner erroneously read both the upper flange/upper support width and the outer face on the same structure in Mulholland. Consistent with the principle that all limitations in a claim must be considered to be meaningful, it is improper to rely on the same structure in the Mulholland reference as being responsive to two different elements (the upper flange/upper support width and the outer face) in claims 1, 10, and 18.
(Konstant, internal citations omitted.)

Note that these are all mechanical cases, by which I mean there was a lot of structure in the claims. I don't know whether this reasoning can be adapted to cover functional claims. That is, if the Examiner uses a single box in a block diagram for teaching the claimed elements "network traffic monitor" and "policy mechanism for applying a policy to the network traffic," I'm not sure arguing "improper to rely on the same structure in the reference as teaching two different elements" will work — a box in a block diagram of a computer system may not be "structure" in the same way that a flange and a roller are "structure."

Case law you can use

One of the benefits of reading BPAI decisions is that occasionally I see the Board citing specific case law supporting an argument that makes sense to me, but which I can't find addressed in the MPEP. For example, I cite In re Arkley for the proposition "the Examiner can't pick and choose from unrelated embodiments in an anticipation rejection." I couldn't find anything in the MPEP about this, but when I saw the Board refer to it in a decision, I started using this argument and citing the case. (See my discussion of In re Arkely here.)

Now that I'm reading lots of BPAI decisions, I'm making a concerted effort to capture these great case law quotes. And when I find a good one, I'll be posting about it as "Case law you can use."

The first "Case law you can use" post will be later this week, and I'll be adding this as a tag to enable readers to quickly find these posts.

Does the BPAI have to decide on all rejections in the case?

From what I've seen, it's quite common for the BPAI to issue a new rejection under § 101. Not too surprising, since many claims now on appeal are obviously vulnerable under Bilski. But I just read a BPAI decision from a couple of weeks ago (Ex parte Lancefield) where the Board raised a § 101 rejection without deciding on the prior art rejections. Technically, they "vacated" the outstanding rejections and remanded back to the Examiner with both the old and the new rejections.

Apparently the BPAI is taking a cue from the Federal Circuit in In re Comiskey, which also refused to reach the prior art grounds of rejection "because we conclude that many of the claims are 'barred at the threshold by § 101'."

I'm really hoping this isn't the start of a trend. The main reason I appeal is to get someone from outside the Examiner corp to decide on the prior art rejection, and vacate-plus-remand puts yet another roadblock in front of me. The thought of having to spend another 12+ months in the cycle of Office Action, Brief, Answer, Reply, wait-for-BPAI decision is infuriating.

Now, perhaps there really *is* no point in deciding on the prior art grounds if the claim has fatal problems under § 101. I'm not sure the vacated Ex parte Lancefield case is fixable by a minor amendment. Here's the only independent claim in the case:

1. A computer system implementing an object oriented data processing schema configured to store objects and to store definitions of relations between the objects, the system being configured so as to support definitions of relationships of two types, the two types including:
a first type of relationship whereby one object in a hierarchy related to another object by a relationship of the first type is deemed to belong to a set defined by the other object; and
a second type of relationship distinct from the first type of relationship and whereby a child object related to a parent object by a relationship of the second type is deemed to inherit characteristics of the parent object and, to an extent specified in the relationship of the second type, a proportion of a characteristic of a further object related to the said parent object by a relationship of the first type:
whereby the system uses the schema to process the objects in accordance with the defined relationships.

As long as the BPAI limits its vacating to these types of claims, I won't be too upset.

New blog: Florida Patent Lawyer

I find most of BPAI decisions that I blog about through various random methods. But in the case of Ex parte Denison, credit goes to the Florida Patent Lawyer blog -- he discussed the Denison decision here.

I've added this newcomer blog to my blog reader. Not only does it discusses patent prosecution topics (rare), Florida Patent Lawyer covers a nice mix of other topics too, including TTAB decisions and ITC decisions. In short, stuff that you won't find on every other patent blog.

Ex parte Denison: BPAI reverses when Examiner relied on "an arbitrary delineation of prior art structure, made in order to meet the claim limitations"

Takeaway: The Examiner isn't allowed to just pick random pieces of a drawing and read it on parts of your claim. The Board put it like this: "[T]hat “structure” identified by the Examiner results only from an arbitrary delineation of Berry’s structure, made in order to meet the claim limitations. "

Ex parte Denison
(Appeal 2009-004110; Appl. No. 10/158,362; Tech. Center 3700)
Decided: January 19, 2010

This application is about a medical device: a stent to hold open a blood vessel. As claimed, the stent includes cylindrical elements with a specific shape. Here's a 3D view of the stent reference used by the Examiner, to give you some perspective.

One of the claim elements at issue was: "each cylindrical element has a plurality of alternating valley portions and peak portions, the valley portions including alternating double-curved portions and Y-shaped portions..." The Examiner used overlapping portions of the reference stent to read first on the double-curved and then on the Y.

The claims require "double-curved portions" and "Y-shaped portions", but the claims are silent as to whether or not the two portions partially overlap in a sense (in other words, whether part of the double-curved portion forms part of a Y-shaped portion).

You can see this overlap here, where the claimed "double curve" looks like the letter "W".

The Applicant argued that the Examiner's interpretation is unreasonable because it "requires the viewer to first visualize one shape and then completely disregard portions of that shape in order to visualize a second, distinctive shape.” The Applicant contended that the claims instead require “[c]omplete and distinct structural shapes depicting both the double-curved shape and Y-shape.”

The Board found that a POSITA wouldn't recognize the portions, as claimed, in the reference:

However, that “structure” identified by the Examiner results only from an arbitrary delineation of Berry’s structure, made in order to meet the claim limitations. Even giving the claims their broadest reasonable interpretation, one of ordinary skill in the art would not have understood the claim terms requiring “double-curved portions” and “Y-shaped portions” to include the arbitrary portions of Berry defined only by the Examiner.
(Decision, p. 5.)

Seems like "arbitrary delineation of structure" is a great argument to use in mechanical applications. But in the art units I deal with, almost everything is described in functional, rather than structural terms. So I don't know how useful this argument will be for such cases.

Ex parte Nishioka: indefinite when claim includes variables but no meaning is attached to the variables

Ex parte Nishioka
(Appeal 2009-001507; Appl. No. 10/046,224; Tech. Center 2400)
Decided January 12, 2010

This patent application is related to encryption, so I would expect math to be involved. But look at the claim — it appears to me to be nothing but math: 

I'm not afraid of math, but this is too much for me to think about (at least in my free time). So I'm not going to discuss the substance of these claims. But I thought the indefiniteness rejection might be a good lesson for those who deal with math-intensive applications from time to time.

The Examiner viewed the claims as a laundry list of elements, with no relationship. Worse, the elements themselves were "undefined."

[The independent claims] recite the limitations "a key generation step of generating a secret-key" and "a public-key" with a list of undefined elements in between without a relation of how are they being used to generate a key or how are they obtained/selected. The relationship between the elements listed and how they are used to produce a key is not clear. Examiner contrasts the language found in the instant application with the language found in the prior art's claims, where the variables are clearly defined. A clear definition of the variables would overcome this 112 rejection.
(Examiner's Answer, p. 25.)

In response, the Applicant simply asserted that the elements in the claim are well known to one of ordinary skill in the art, then "invite[d] the Board to provide a statement of how to overcome this rejection."

The Board affirmed the indefiniteness rejection, noting that variables are not definite unless they have meaning:

This is an art of very high complexity (FF #3), and the practitioner in such an art is assumed to have advanced math skills. But the attachment of meaning to the variables expressed in the equations of the claims, either through definitions in the specification or in the claims themselves, adds necessary clarity to the claims to be able to properly appreciate the metes and bounds of the protected invention. Appellants have not done this. The Examiner’s contrast of the instant claims to those in the prior art was insightful.
(Decision, p. 9.)

Although the Board didn't come right out and say it, I'd say the reference to the claims in the prior art was a strong hint to the Applicant as to the type of claims that would comply with §112 ¶2nd. I've reproduced the claims from the prior art reference below.

You can see that while variables are used, so are descriptions of the variables. Also, even mathematical functions are described in words (i.e., product of first base-group-number raised to the power of a first monotone function of said first exponent-number).

I think the lesson here is that when dealing with math-intensive subject matter, you should think carefully about simply pasting equations into the claims.

PTO memo on acceptable computer-readable medium claims

A new memo on acceptable language in computer-readable medium claims is available on the PTO website at  http://www.uspto.gov/patents/law/notices/101_crm_20100127.pdf

The memo, signed by Director Kappos himself, is "an effort to assist the patent community in overcoming a rejection or a potential rejection under §101." The memo suggests amending the claims to state that the computer-readable medium is “non-transitory.” The PTO is taking the position that such an amendment is not new matter even if the word "non-transitory" is not in the spec — as long as the spec did describe the medium as *something* other than a signal.

It doesn't appear to me that this memo has yet been officially published to the outside world. Although it's on the PTO website, it's not on the page with all the other examination guidelines (here).

I suspect this is an internal memo that has been leaked, and that the memo will be made official in the near future. I found out about the memo from the IP Spotlight blog.

Effect of BPAI claim construction in litigation

In an earlier post (here), I discussed the BPAI Ex Parte Jenkins decision, in which the Board construed the claimed "storage space" to be bigger than a rivet, given the teachings of Applicant's spec.

Would the Applicant have a hard time arguing in litigation that an accused product with a tiny storage space infringes? What about a storage space big enough to hold a couple of fishing flies? That's bigger than a rivet, but not by much.

Or is the Board's construction irrelevant in litigation? If so, why? Can't you say that the size of the storage space was what distinguished over the reference, so a sort of prosecution history estoppel applies, even though it was the Board that made the distinction? Or is the estoppel doctrine not relevant unless the Applicant argued the size of the storage space was a patentable distinction?

I don't know the answer to these questions. I'm just thinking out loud.

Can an Examiner use "res judicata" after a decision by the BPAI?

If you get an adverse decision from the BPAI, you can continue prosecution by filing an RCE with claim amendments and/or new evidence. I came across an application (Appl. No. 10/023,117) in which the Applicant did this, and the Examiner invoked res judicata to reject a claim, even though amendments were made.

Although I've never come across res judicata in patent prosecution, the MPEP specifically provides for it in section 706.03(w). It does make sense to me that once the Board has affirmed a claim rejection, that rejection can be applied again with no further explanation because the issue has already been decided.

What struck me as strange is that the Applicant had amended the claims. But the Examiner took the position that the amended claim was not patentably distinct from the claim presented at appeal:

The sole change to claim 6 is changing the final clause from "outputting ... to a destination identified in the ... instruction" to "outputting ... to a destination register identified in the ... instruction". As it was shown in the examiner's answer to appeal mailed November 14, 2007 that Abbott output to a destination register, the claim is not patentably distinct from that affirmed by the Board of Patent Appeals and Interferences.

I think it's clear that "destination register" has different scope than "destination," such that the Examiner must make a new finding in the post-RCE Office Action that the reference discloses the added feature. Certainly it's not enough that the Examiner's Answer "shows" that the reference teaches the amended feature, because a statement in an Examiner's Answer is an allegation, not a judicially established fact.

In fact, I think the Board's decision implied that the claim on appeal ("outputting to a destination) was patentably distinguishable from the claim in the RCE ("outputting to a destination register"). This happened when the Board pointed out that the Applicant was arguing about a feature which was not present in the claim on appeal:

Initially, we note that although Appellants argue that Abbott does not teach that the output of the transposition circuit 410 is provided to a destination register that is specified in an instruction ... the Appellants have chosen to draft the claims, claim 1 in particular, far more broadly.

By pointing out that the claim on appeal did not specify "destination register", I think the Board in effect said that a claim without this feature is patentably distinct from a claim with this feature. Thus, res judicata does not apply to the amended claim.

I should note that the Examiner did not solely rely on res judicata. As recommended by the MPEP, a prior art rejection which was applied also, in this case, the same rejection which was appealed was "maintained and incorporated by reference."

Ex parte Fabritius: ridiculous definition of "portable"

Ex parte Fabritius
(Appeal 2008-0068; Appl. No. 10/449,707; Tech. Center 2100)
Decided: May 29, 2008

The claim at issue was to a "method for handling data in a fixed memory associated with a portable electronic communication device ... and an operation of storing data in said fixed memory comprises ..."

The reference disclosed an automobile radar system. The Applicant argued that the system was not a "portable communication electronic device." The Board first found that this phrase in the preamble was not limiting. The Board should have stopped there, because the only other Applicant argument was "non-analogous art," which the Board noted was irrelevant for anticipation.

But the Board continued on, and shot itself in the foot. Here is the Board's strained reasoning as to why the automobile radar system was "portable":

At least a portion of the communication device described by Woll is removable, and thus “portable” within some meaning urged by Appellants. RAM card 20 (Woll Fig. 2) may be removed from the system so that the data may be read by a personal computer 60 (Fig. 6).

What's going on here? Did the Board confused "movable" with "removable"? I can see that "movable" is a synonym for "portable", but not "removable". Also, even if there is some overlap between removable and portable, why is the entire thing portable simply because a portion of it is removable? Is my house portable because the door is removable?