Monday, August 27, 2012

BPAI affirms obviousness of substituting parallel connector for serial even if reference did not teach equivalence

Takeaway: A Patentee in an inter partes reexamination appealed an obviousness rejection of a dependent claim. The Examiner used the substitution rationale from KSR, substituting a Firewire connector for a USB connector. The Applicant argued that "[the reference] does not disclose that the pin configuration, the wiring scheme required or operation of the two [connectors] are equivalent." The Board affirmed the rejection since the Examiner made the requisite findings of fact discussed in MPEP § 214(B), and the Patentee did not show error in these findings. The Board explained that a teaching of equivalence is not required support a rejection under the KSR substitution rationale. (Milan v. Apple, BPAI 2012.)

Details:

Milan v. Apple
Appeal 2012008569; Reexamination No. 95/001,083; Tech. Center 3900
Decided  Aug. 7, 2012

The patent under reexamination described a flash memory device designed for use with a set of converter plugs. As shown in the figures below, the pin configuration (416) on the flash memory device (400) was configured to mate with several different converter plugs, including a 4-pin USB plug (FIGs. 12A-C) and a 5-pin mini-USB plug (FIGs. 12A-C).




The Patentee appealed various rejections, including independent claim 1 as obvious over two non-patent references ("Palm V Handbook Travel Kit" and "Handbook for Palm Tungsten Handhelds"), and dependent claim 3 as obvious over Palm, Tungsten, and Sirichai.
     1. A flash memory device comprising:
     a housing having opposed first and second ends;
     a flash memory drive enclosed in said housing; and
     a quick connector mounted in said housing and having a plurality of pins exposed at said second end, said pins being configured for electrical connection to a selected one of at least a first interchangeable connector and a second interchangeable connector ...

     3. The flash memory device according to claim 1 wherein said plurality of pins is six female pins arranged in two parallel rows of three pins each.

After rejecting independent claim 1 as obvious over the combination of Palm and Tungsten, the Examiner added Sirichai and used the substitution rationale to reject dependent claim 3. In rejecting claim 3, the Examiner made several findings of fact, including the ones discussed in MPEP § 2143(B) for the substitution rationale:
  • One type of Firewire connector is known to have a six pin configuration arranged in two parallel rows of three pins each.
  • The prior art (Palm combined with Tungsten) contained a device which differed from the claimed device by the substitution of one component (Sirichai's connector) for another component (Palm's connector).
  • Sirichai disclosed USB connectors and Firewire connectors, and described them as known equivalents.
  • A POSITA could have substituted one known element (Sirichai's Firewire connector) for another (Palm's USB connector), and the results of the substitution would have been predictable.

The Patentee attacked the obviousness rejection of claim 3 by disagreeing with the Examiner's finding that Sirichai taught the equivalence of USB and Firewire connectors. According to the Patentee, Sirichai's discussion of lighted electrical connectors instead taught that:
... the connector tip is a USB connector tip and in another embodiment the connector tip may be a firewire connector tip. The connector tip 150 of Sirichai is simply referring to "an opaque metal shell[.]" Sirichai only provides that both USB and firewire connector tips could utilize the light source disclosed; it does not disclose that the pin configuration, the wiring scheme required or operation of the two are equivalent.
(Emphasis added.)

The Patentee then concluded that "simply substituting the USB connector of Tungsten with the connector body for a firewire connector will not result in functionality of the resulting connector" and as a result, the combination would not result in claim 3.

The Board was not persuaded by the Patentee's "no equivalence" argument. According to the Board, "a teaching of such equivalence is not required" to support the substitution rationale described in KSR. The Board found that the Examiner had made the requisite findings of fact for a substitution rationale, and the Patentee had not demonstrated error in these findings. The Board therefore affirmed the rejection of dependent claim 3.

My two cents: The Board reached the right result -- at the level of detail expressed in the claim, it is obvious to substitute a USB (serial) connector for a Firewire (parallel) connector. Yes, the operation of a serial bus and a parallel bus is different. But none of those differences have anything to do with what's in the claims. Yes, the pin configurations and wiring schemes are different. Nonetheless, substituting one connector for the other is within the knowledge of a POSITA, and the Applicant didn't argue otherwise.

I do have a problem with the Board's statement about findings of equivalence, because I think it's misleading. Of course the claimed feature and the substituted feature from the reference aren't required to be equivalent in every way. But the two must be equivalent at some level, otherwise you couldn't substitute one for the other, right?

And you can fight an obviousness rejection on this point, i.e., by explaining the differences and why the reference feature wouldn't act as a substitute for the claimed feature. Maybe the lesson here is not to characterize such an argument as being about equivalence, but instead to couch it in terms of "results would not have been predictable" (i.e., the third finding mentioned in the MPEP discussion of substitution)?

I have a problem with this too, however, as the results of a badly-conceived substitution are probably quite predictable: the combination doesn't work as intended, or work as well, or too many modifications are needed to consider it obvious.

So to me, it is about equivalence, not predictability. Because it makes sense to argue that a POSITA would not substitute a blodget for a widget because they are not equivalent in these specific ways which relate to what is claimed. The Patentee here didn't win because he didn't (or couldn't convincingly) explain why the differences matter for what is claimed, and/or why the differences were too great for a POSITA to handle.


Tuesday, August 21, 2012

BPAI decides whether limitation refers to final state of disconnection or action of disconnecting

Takeaway: An Applicant appealed a written description rejection of a claim for a device used in rock drilling. The rejection hinged on the interpretation of  "device is loaded .... only when said rock bolt is disconnected from said coupling sleeve by rotation of the rock drilling machine." The Examiner interpreted this to exclude loading after disconnection, while the Applicant asserted that the phrase excluded loading during disconnection. The Board found that the proper interpretation was "during disconnection," since "disconnected" must be read in conjunction with the modifier "by rotation." (Ex parte Kanflod, BPAI 2012)
Details:

Ex parte Kanflod
Appeal 2012005389; Appl. No. 10/539,148; Tech. Center 3600
Decided   August 14, 2012

The application on appeal related to drilling machines. The independent claim on appeal was directed to a sleeve that couples a rock bolt (1) to a rock drilling machine (2). (See figure below.)

The limitation at issue read:
a locking device (7) ... said locking device is loaded to retain the rock drilling machine connected to the coupling sleeve only when said rock bolt is disconnected from said coupling sleeve by rotation of said rock drilling machine in a direction for disconnecting said first part of said coupling sleeve from said rock bolt for reinforcing a rock with said rock bolt.
(Emphasis added.)

During prosecution, the Applicant added the qualifier "only" in response to a prior art rejection. The Examiner then rejected the amended claim under § 112 First ¶, as lacking written description. The Office Action explained the problem with the phrase "only when ... disconnected" as follows:
[T]here is no structure to prevent the loading of locking device (7) to retain the rock drilling machine (2) connected to the coupling sleeve (3, 5) regardless of whether or not rock bolt (I) is connected or disconnected, since rock bolt (I) has no contact with either of locking device (7) or drilling machine (2) (even when the bolt is installed) and thus cannot prevent relative loading between device (7) and machine (2).
(Emphasis added.)

The Applicant submitted an expert declaration from the inventor to address the written description rejection. The declaration also explained the meaning of the limitation at issue. Referring to this declaration, the Applicant argued that the "only when ... disconnected" phrase should be read in conjunction with the "rotation" phrase:
[T]he locking device is loaded (locked) to lock the drilling machine to the coupling sleeve only when the drilling device is rotated in a direction to separate the rock bolt from the coupling sleeve, and not during a percussion operation.

The declaration, and the argument, then went on to reference portions of the specification that described this loading behavior in conjunction with the rotating.

The Applicant went to appeal on the written description, indefiniteness, and prior art rejection, with both sides essentially repeating the above written description arguments.

The Board found that the written description rejection was based on an improper claim interpretation, one which read the first portion of the wherein clause in isolation from the remainder of the claim.
A person skilled in the art would understand the above limitation to mean that the locking device is loaded just prior to disconnection of the rock bolt from the sleeve, i.e., when the rock bolt is still connected to the coupling sleeve. The Examiner improperly interprets the limitation to mean that the locking device is loaded only after disconnection of the rock bolt from the sleeve.

The Board reversed the written description rejection, finding that the Applicant's specification did convey possession of the properly interpreted claim.
The specification as originally filed explains that (1) during drilling the locking device 7 is unloaded (Spec. at 2:13-14) and (2) after drilling the locking device is loaded to disconnect the rock bolt from the coupling sleeve (Spec. at 2:14-19).

My two cents: Once again, the real issue is claim construction. It seemed to me that both sides recognized this, but could have done a better of job of explaining exactly how they were interpreting the claim. I had to keep reading the claim, and the arguments, over and over again.

The Examiner apparently read "disconnected" as referring to the bolt's final state, so that the phrase "only when" was interpreted to exclude loading with the bolt disconnected. On the other hand, the Applicant apparently read "only when ... disconnected" to refer to the action of disconnecting the bolt, or perhaps the bolt's intermediate state of being disconnected.

The choice of tense in "when disconnected" definitely sounds like a state rather than an action. It's only when the "by rotation" portion of the claim is considered that it even becomes possible to understand "disconnected" as an action. Still, I think the phrasing "only when said rock bolt is being disconnected" does a better job of conveying the Applicant's apparent intent.

My sense is that the issue of what tense to use comes up more in mechanical cases where you focus on describing structure first, and then on associating actions and states with the structure. Whereas action and state are more central to the claims I deal with every day in computer and electronics applications. Even so, I can think of instances where I've had to think about how to properly convey state, action, or a combination of the two – and it can be a challenge.

Friday, August 10, 2012

BPAI provides concise summary of how to rebut an Examiner's claim interpretation

Takeaway:  In an appeal involving a cartridge for caulking compound, the main issue was the meaning of of the term "punch" in claim limitations such as “punching an opening in the tubular container.” The BPAI explained that the Applicant's mere assertion that claim language did not encompass the teaching of the reference was not sufficient to rebut the Examiner's interpretation. To rebut, an Applicant should point to "precise meaning" in the specification, to a dictionary definition, or to a statement by a POSITA. (Ex parte Longo, BPAI 2012.)

Details:

Ex parte Longo
Appeal 2009014183; Appl. No. 11/328,537; Tech. Center 3700
Decided:  May 23, 2012

The subject matter on appeal was a cartridge for caulking compound. The main issue on appeal was the meaning of the term "punch" in claim limitations such as “punching an opening in the tubular container.”

For the "punching" limitations, the Examiner relied on a reference teaching of a "rapidly actuatable tool for making cuts" which "deforms the cylinder wall to form catches 17 as it cuts." The Applicant argued that the cutting described in the reference was not the same as punching.

The Board presented a dictionary definition ("to cut, stamp, pierce, perforate, form, or drive with a tool or machine that punches") to show that the Examiner's interpretation (punching" includes "cutting") was reasonable. The Board found that the Examiner's interpretation was "essentially unrebutted" by the Applicant.

Appellant, however, points to no specific wording in the specification that shows a precise meaning given to the term “punching” or to any related terms such as “punch” that would amount to interpretive guidance. Appellant also fails to provide support for this argument by pointing to any type of general purpose or technical dictionary as interpretive guidance that punching would not encompass the cutting performed in van Manen, nor does Appellant state that this is a term of art known to those of ordinary skill to exclude van Manen’s cutting. Appellant merely provides attorney argument against the Examiner’s interpretation with nothing further. 
Based on this claim interpretation, the Board affirmed the prior art rejections of the independent claims.


My two cents: I picked this decision not because it was controversial in any way, but because it included a clear and concise statement of what the Board is looking for in an argument about claim construction. A mere assertion in the Appeal Brief that a widget is not a blodget probably won't be convincing. Instead, you should refer to a "precise meaning" in the spec, to a dictionary definition, or to a statement by a expert about what the term means to a hypothetical POSITA.

On this last point, the Board's phrasing suggests that an Applicant statement in the Brief about what the term means to a hypothetical POSITA would be sufficient. However, other decisions I've read suggest instead that you need evidence. This evidence can take the form of another reference that uses the term, or a § 1.132 declaration from a person of ordinary skill in the art about the meaning of the term.

Friday, August 3, 2012

BPAI affirms enablement rejection of "detecting at least one gamma ray" aftering finding POSITA knows that detection requires multiple rays

Takeaway: In an application involving the use of radiation to test materials, the Applicant appealed an enablement rejection for claims that included the step of  "detecting at least one emitted prompt gamma ray." The Examiner asserted that "[i]t is a notorious scientific fact that statistically meaningful data requires collection of information from not one but from a plurality of events and nuclear interactions.The Applicant argued that the specification need not enable every embodiment, and that the burden was on the Examiner to show that all multiple ray embodiments were inoperative or not enabled. The Board affirmed the rejection, based at least in part on the Applicant's admission that detection required more than one ray.  "As such, Appellant has conceded that the disclosure of the present application does not provide an enabling description for the full scope of Appellant’s claims. Appellant, having chosen the broad claim language, must make sure that the broad claims are fully enabled." (Ex parte Akers, BPAI 2010.)

Details:
Ex parte Akers
Appeal 200900591; Appl. No. 10/383,096; Tech. Center 3600
Decided:  March 15, 2010

The technology on appeal was materials testing using radiation. The only independent claim on appeal read:

1. A method for evaluating a material specimen, comprising:
     bombarding the material specimen with neutrons to create prompt gamma rays within the material specimen, some of the prompt gamma rays being emitted from the material specimen, some of the prompt gamma rays resulting in the formation of positrons within the material specimen by pair production;
     detecting at least one emitted prompt gamma ray;
     detecting at least one emitted annihilation gamma ray resulting from the annihilation of a positron; and
     calculating positron lifetime data based on the detected emitted prompt.

The Applicant then went to appeal on several non-prior art rejections, including a rejection under § 112 First, Enablement. According to the Examiner, the specification did not enable detection of a single gamma ray as recited in the independent claim:
The limitation "at least one" admits having only one. There is neither an adequate description nor enabling disclosure as to how and in what manner detecting only one prompt gamma ray and only one annihilation gamma ray can collect adequate experimental data. It is a notorious scientific fact that statistically meaningful data requires collection of information from not one but from a plurality of events and nuclear interactions.

In the Appeal Brief, the Applicant argued that § 112, First did not require enabling every embodiment of the claim (citing Ex parte Saito, BPAI 2008). Since "at least one ... gamma ray" encompassed many embodiments involving multiple gamma rays, and the Examiner failed to demonstrate that these embodiments were inoperative or not enabled, the Examiner failed to carry the burden in showing lack of enablement.

The Applicant did not rely solely on the lack of a prima facie case of enablement, but went further to specifically argue that the claims were enabled to a POSITA. The Applicant cited to In re Skrivan, 427 F.2d 801 (Fed. Cir. 1970), for the proposition that an enabling disclosure was not required to include information presumed to be within the level of ordinary skill in the art. After quoting the Examiner's "notorious scientific fact" characterization, the Applicant asserted that "persons having ordinary skill in the art will recognize that a plurality of gamma rays will need to be detected in order to collect meaningful data." As evidence, the Applicant pointed to two prior art references of record that showed "these kinds of calculations and thus also the underlying data collections have been performed by skilled artisans for upwards of 60 or even 70 years."

The Examiner's Answer challenged the Applicant's use of Ex parte Saito, noting that the Applicant had selectively quoted from the opinion. The Examiner quoted additional material from Saito which said: "if the number of inoperative combinations becomes significant, and in effect forces one of ordinary skill in the art to experiment unduly to practice the claimed invention, the claims might indeed be invalid." The Examiner then applied Saito to the application on appeal, and concluded that:
Neither one, two, three, or even fifty or more prompt gamma rays in combination with each of one, two, three, or fifty or more annihilation gamma rays would not yield an operative embodiment. The number of inoperative combinations is significant and an artisan would have to perform undue experimentation to practice the claimed invention.
(Emphasis added.)

The Board affirmed the rejection. The Board agreed with the Applicant that "an application for patent need not necessarily provide an enabling disclosure of every embodiment of the claimed invention." However, the Board found that the Applicant had not provided any enabling disclosure for detecting one gamma ray, which is explicitly included in the scope of the claim. In fact, the Applicant had specifically conceded that a POSITA would understand that detection of multiple gamma rays would be required to collect meaningful data. The Board concluded its analysis as follows:

As such, Appellant has conceded that the disclosure of the present application does not provide an enabling description for the full scope of Appellant’s claims. Appellant, having chosen the broad claim language, must make sure that the broad claims are fully enabled. In this case, as conceded by Appellant, the scope of the claims, by encompassing a method wherein one emitted prompt gamma ray and one emitted annihilation gamma ray are detected, is not less than or equal to the scope of the enablement, as required by the first paragraph of 35 U.S.C. § 112.
(Emphasis Added.)

Backstory: The above summary of the prosecution history leaves out a lot of details. The actual prosecution history is quite convoluted: multiple Answers and Reply Briefs, a reopening of prosecution, and a return to appeal. I simplified greatly so I could focus on the specific issue of enablement.

Postscript: The Applicant let the case go abandoned after the appeal decision – despite the fact that the Examiner withdrew the only prior art rejection before appeal, the Board reversed the other § 112 rejections, and the affirmed enablement rejection could probably have been overcome by appropriate claim amendment.

On the other hand, this application was in the middle of a continuity chain of three applications, and the Applicant was successful in getting two issued patents out of the chain. So maybe the Applicant got what it needed elsewhere. Also, it may be that the claims in this particular application had no commercial value by the time of the BPAI decision. The decision came down a full six years after filling of the first Appeal Brief, and a full seven years after the application was filed.

My two cents: I don't know much about this area of technology, but I feel that's unnecessary to understand this decision: both sides agreed on the facts, but not the law. I say this was the right result, and it looks like the Applicant misunderstood the law of enablement.

Once the Examiner explained the rejection in the Answer, it was clear [to me] that the rejection could be overcome with a simple claim amendment to exclude the inoperative embodiments.

The most interesting aspect may be: what's the appropriate amendment? Seems to me "plurality of gamma rays" doesn't help, because that still includes small numbers like 2, 3, and 50, which apparently can't be detected.

My first stab at an amendment would be "a stream of gamma rays". Avoids a specific number of rays (important) and doesn't seem indefinite -- to me, but I'm admittedly not a POSITA. Perhaps there is a better term of art.