January 2010

Wikipedia Citations in Patents Up 59 Percent

Patent

Patent Librarian Michael White recently posted an interesting graph on the number of U.S. patents referencing Wikipedia articles:

The number of U.S. patents issued last year that contain one or more references to Wikipedia articles totaled 809, a 59 percent jump from 2008. Several years ago the USPTO banned patent examiners from using Wikipedia as a source of information for determining patentability of inventions. However, examiners and applicants continue to cite it.

White's graph is shown below.

Of course, I would also offer my own graph (below) that shows the percentage of patents citing Wikipedia grouped by issue year.

Wyeth v. Kappos: Appellate Panel Confirms that PTO Is Miscalculating (Undercalculating) Delays in Prosecution and the Corresponding Patent Term Adjustment

Patent,

Wyeth and Elan Pharma v. Kappos (as USPTO Director) (Fed. Cir. 2009)

The Federal Circuit has decided an important case involving the length patent term adjustments (PTAs) that are granted due to PTO delays in patent prosecution.  Many if not most newly issued patents will be positively impacted by the decision. Those patents will be enforceable for an additional period of protection tacked to the end of the standard 20–year term that is in addition to the previously calculated PTA.

Prior to 1995, US patents were enforceable for 17–years from the date of issuance. The new rule changed the patent term to 20–years from the filing date.  In order ensure that the change did not adversely impact patentees, Congress created several patent term guarantees that are codified in 35 U.S.C. 154(b)(1)(A)-(C).

  • Part A guarantees “prompt” PTO responses.  The section identifies specific types of PTO delays and the patent term is adjusted one-day for each day of PTO delay.
  • Part B guarantees “no more than 3–year application pendency.” The section indicates that the patent term should be adjusted one-day for each day from the end of the three-year pendency until the patent issues.
  • Part C guarantees that the patent term not be reduced based on interferences, secrecy orders, or appeals.

Several provisions limit these adjustment by reducing the patent term based on patent applicant actions (or delays) that unjustifiably lengthen prosecution. 

Overlap Limitation: Parts (1)(A) and (B) are also subject to an overlap limitation: “To the extent that periods of delay attributable to grounds specified in paragraph (1) overlap, the period of any adjustment granted under this subsection shall not exceed the actual number of days the issuance of the patent was delayed.”

The fight in this case centered around the calculation of when “periods of delay . . . overlap.” USPTO’s interpretation of the statute is that the overlap is property avoided when the term is adjusted only by the greater of guarantee (A) or (B).  Wyeth, on the other hand, argues that the guarantees only overlap when the PTO-delays identified in (A) occur after the first-three-years of prosecution have passed.

Both the DC District Court and the Federal Circuit have agreed with Wyeth’s statutory interpretation:

This court detects no ambiguity in the terms “periods of delay” and “overlap.” Each term has an evident meaning within the context of section 154(b). The limitation in section 154(b) only arises when “periods of delay” resulting from violations of the three guarantees “overlap.” 35 U.S.C. § 154(b)(2)(A). Significantly, the A and B guarantees expressly designate when and for what period they each respectively apply. Thus, this court can easily detect any overlap by examining the delay periods covered by the A and B guarantees.

At oral arguments, Judge Moore was sympathetic to the PTO’s policy position and the opinion reflects that same concern.  However, the panel held that the language of the statute should prevail.

Regardless of the potential of the statute to produce slightly different consequences for applicants in similar situations, this court does not take upon itself the role of correcting all statutory inequities, even if it could. In the end, the law has put a policy in effect that this court must enforce, not criticize or correct.

Deference: The Federal Circuit gave no deference to the PTO’s interpretation of the statute — holding that “[b]ecause the language of the statute itself controls this case and sets an unambiguous rule for overlapping extensions, this court detects no reason to afford special deference to the PTO’s interpretation.”

Affirmed. The PTO must now revise its PTA calculations.  Patentees should consider whether they are eligible for a recalculation of their PTA adjustment.

Note: Dozens of parallel cases are pending in DC & VA district courts. Many of these have been stayed pending outcome of this appeal.

Example: I pulled up recently issued patent number 7,640,000.  The application was filed on November 28, 2006 and was issued December 9, 2009.  It turns out that patent was issued on a first-office-action-allowance. That NOA was mailed October 6, 2009.  Under Part (A), the PTO was required to issue a Rejection or NOA within 14–months.  The 617 days beyond those 14–months formed the Part (A) adjustment.  Under Part (B), we look at how many days beyond three-years it took the patent to issue. Here, that calculation is 11 days.  Under the old calculation, the PTA would be 617 days because that is greater than 11 days. Under the new calculation, the PTA would be 617 + 11 = 628 days (since the Part (A) delay did not overlap with the Part (B) delay).

Dropping: Average Number of Claims Per Patent

Patent

In addition to a reduction in the number of patent applications  being filed, we are also seeing a reduction in the number of claims in each issued patent.  All things being equal, patents with more claims are typically deemed more valuable.  The reduction in claims is likely associated with cost-cutting measures taken by patent applicants. 

The Chart below is based on a random sample of 22,000 issued patents.

PatentLawPic864

Means-Plus-Function Claims

Patent

In December 2009, the USPTO issued 16,000+ utility patents.  I used a software program to tally those patents that include claims with means-plus-function (MPF) claim language by looking for the terms “means for” or “means of” in the claims.

Only about 15% of the issued patent included at least one claim having an MPF limitation. Of patents that include an MPF limitation, half recite fewer than five such limitations. It appears that attorneys are using MPF claim language primarily as a back-up.  Longer claim-sets are more likely to include means-plus function language.  Thus 21% of patents with above-average-sized claimsets (by character count) include a means-plus-function limitation while only 11% of patents with below-average-sized claimsets include an MPF limitation. 

Looking claim-by-claim, I found that only 3% of the patents in my sample included means-plus-function limitations in every independent claim.

Means-plus-function terms are also also associated with non-assigned patents (usually individual inventors).  23% of non-assigned patents included an MPF limitation compared with 14% of assigned patents.

Patents arising from certain countries are much more likely to include MPF limitations. The chart below looks at the most common foreign filing jurisdiction for priority claims and reports the percentage of US patents arising from those foreign filings that include MPF limitations. 

PatentLawPic862

Means-plus-function claiming is primarily focused outside of the chemical and biotech art units. Thus, only 2% of patents examined in TC1600 (Biotech & Organic Chemistry) and 7% of patents examined in TC1700 (Chemical & Materials Engineering) include MPF limitations. The remaining technology centers averaged over 18% of patents including MPF limitations.

The chart below shows means-plus function claim prevalence of the top-20 patent owners (based on the number of patents owned in the sample). These companies are all primarily focused on electronics and are all headquartered in either the US or East Asia.

PatentLawPic863

The final chart shows the change in patent prosecution practice over time. Clearly, the use of means-plus-function claims has dropped over time. This is true even in the face of rising claim counts. The chart below is based on a sample of 22,000 randomly selected patents that issued 1976–2009.  I grouped the patents according to their decade of issuance and calculated the percentage that included at least one means-plus-function limitation.  *The chart also includes the December-2009 data mentioned above.

PatentLawPic865

Notes: My search is likely both over and under inclusive.  Although a claimed element that uses the terms “means for” is usually interpreted as an MPF claim, it is not always so interpreted. Likewise claim terms written in functional language without any structure in the claims can be interpreted as means plus function even if the absence of the “means for” language. Also, “step for” language is typically when a method step is functionally claimed.

No Surprise: Broadly Written Means-Plus-Function Terms Limited to Structure Described in Specification

Patent

Jersey ShoreRestaurant Technologies Inc. v. Jersey Shore Chicken and Klee’s Bar & Grill (Fed. Cir. 2009)

RTI sued two restaurants for infringement of its chicken fryer patent.  The maker of the accused fryer (Oilmatic) then filed a declaratory judgment action. Those actions were consolidated and are all appealed here. The claims at issue here involve a mechanism for filtering and then disposing of waste cooking oil. After construing the claims, the district court issued a summary judgment of noninfringement. RTI appeals.

The primary disputed claim term was the “means for metering oil.” As a means-plus-function term, the element is construed to cover the structures found in the specificatoin that corresponding to the claimed function as well as their equivalents. Here, the specification discloses a squeezable trigger valve for metering out oil.  To be clear, the seemingly broadly written claim term “means for metering oil” is here interpreted to mean “a squeezable trigger valve with a nozzle and its equivalents.”  The accused device uses a dipstick assembly with a rotating positive displacement gear pump.  The appellate court agreed that “no reasonable fact finder could conclude that the accused structure is idenetical or equivalent to the relevant disclosed structure.”

[A]lthough the dipstick assembly, in conjunction with the pump [of the accused device], may perform the identical function as the [claimed] squeezable trigger valve, the differences between the dipstick assembly plus a pump on the one hand, and a squeezable trigger valve with a nozzle, on the other, are not insubstantial. For example, the dipstick assembly and pump are activated with a push button on the dipstick assembly that energizes the pump; fluid pressure in the line then overcomes a spring-loaded check valve to pour into the fryer. The accused system thus uses a pump and fluid pressure to overcome a check valve, whereas the patent claims a valve that is squeezed.

Hal Wegner writes: Restaurant Technologies represents yet another teaching example to remind the patent drafting community to either stop using “means” claiming or to recognize the severe limitations of this claim form and to comply with the requirements set forth in 35 USC § 112, ¶ 6.

Non-Infringement Affirmed

Patently-O Bits and Bytes

Patent
  • Need some help from SPEs: Stanford Law Professor Mark Lemley is working on a project involving an extensive amount of patent prosecution data. He would like to clarify a few issues on patent office practice and could use the help of current or former SPEs.  One particular question that he is asking involves the assignment of examination duties within an art unit. I.e., once an application is classified, what is the process of deciding which examiner will handle the case?  You can contact him at mlemley@law.stanford.edu.

Patently-O Bits and Bytes

Patent
  • Need some help from SPEs: Stanford Law Professor Mark Lemley is working on a project involving an extensive amount of patent prosecution data. He would like to clarify a few issues on patent office practice and could use the help of current or former SPEs.  One particular question that he is asking involves the assignment of examination duties within an art unit. I.e., once an application is classified, what is the process of deciding which examiner will handle the case?  You can contact him at mlemley@law.stanford.edu.

Enforcing Contracts between Joint Owners: WARF v. Xenon

Patent, , , , , ,

Wisconsin Alumni Research Foundation (WARF) v. Xenon Pharma, Appeal No. 08-1351 (7th Cir. 2010).

WARF and Xenon jointly filed for patent protection for a cholesterol lowering enzyme known as Stearoyl CoA Desaturase (SCD). Warf also granted Xenon rights as an exclusive licensee and Xenon agreed to pay royalties for sales or sublicense fees. Xenon then sublicensed its rights to Novartis but did not pay royalties to WARF. Xenon's argument is based on the law of concurrent patent ownership. Generally, a patent co-owner is not required to share licensing revenue with other co-owners. The district court rejected that argument here — finding that the contract between WARF and Xenon is controlling over the patent law default rule. At trial, a jury awarded WARF one-million-dollars in royalties. That award was reduced then by the judge to $300,000.

The Seventh Circuit has now affirmed the lower court ruling that Xenon breached its license agreement by granting a sublicense without paying WARF its share.

35 U.S.C. 262 makes clear that each Joint owner of a patent right may make use of the rights "without the consent of and without accounting to the other owners." However, §262 has a major caveat in that it only applies "in the absence of any agreement to the contrary." Picking up on that caveat, the Seventh Circuit made clear that the "statutory default rule therefore controls unless there is an agreement to the contrary." The appellate panel went on to confirm that the WARF/Xenon agreement is certainly an "agreement to the contrary" despite the fact that it does not include a specific and explicit revocation of § 262 rights.

The bargained-for exchange between the parties provided that the Foundation would forego its right to separately license the patent in exchange for receiving a share of the profits from Xenon's commercialization of the technology—either directly or via a sublicense to a third party. Xenon received a significant benefit from the agreement—the exclusive right to exploit the technology protected by the joint patent application. Xenon cannot avoid paying royalties or sublicense fees to the Foundation simply by labeling the Novartis transaction a "license" rather than a "sublicense."

Quiet Title: Xenon also filed for its own patents on PPA compounds used to suppress SCD levels. WARF claimed ownership rights in the PPA because one of its scientists had contributed to the project and sued here to quiet title and also sued for conversion. On appeal, the court sided with WARF — finding that the WARF scientist's assignment of rights to Xenon was void because the scientist has already assigned all interest in future inventions to WARF.

Seventh Circuit: WARF originally appealed its case to the Federal Circuit. However, the Federal Circuit transferred the case to the Seventh Circuit based on a lack of subject matter jurisdiction. In a non-precedential opinion, the court held that the dispute did not "arise under the patent laws." [Link] An interesting aspect of that decision was that the court's statement that the Bayh-Dole Act (35 U.S.C. §§ 200-212) is not "patent law" when considering the question of appellate jurisdiction. The Federal Circuit refused to take the jurisdiction issue en banc. Judge Rader penned the following dissent from the en banc denial:

In an extremely short per curiam opinion with far too little explanation, the panel left the impression that it was giving away this court's jurisdiction over a broad swath of claims potentially arising from the Bayh-Dole Act, 35 U.S.C. §§ 200-212. While non-precedential, the panel's opinion nonetheless unnecessarily suggests that this court's jurisdiction to review these cases is limited. I respectfully dissent from this court's denial of rehearing en banc.

The panel concluded, albeit without much analysis, that WARF's complaint did not allege that the Bayh-Dole Act creates a cause of action. Thus it was entirely unnecessary for the panel to opine as to whether the Bayh-Dole Act is a "patent law" or not. And indeed, the Bayh-Dole Act, in my view, is most certainly a patent law. While the panel dismissed the notion that the Bayh-Dole Act's "mere inclusion" in Title 35 does not make it a "patent law," it seems to me that this is actually a pretty good indicator. After all, Title 35 is itself entitled "Patents," and Chapter 18, which encompasses only §§ 200-212, is called "Patent Rights in Inventions Made with Federal Assistance." Also, the language within some of the sections of the Bayh-Dole Act suggests its provenance as a patent law. For example, 35 U.S.C. § 201(d) defines "invention" as "any invention or discovery which is or may be patentable…." Section 201(e) explains that a "subject invention" is an invention conceived of or first reduced to practice by a contractor. "Conception" and "reduction to practice" are familiar patent law terms of art. Also, § 200 states that the Act intends "to use the patent system to promote the utilization of inventions arising from federally supported research or development…." The Bayh-Dole Act is, "at its heart," a patent law, albeit a patent law that employs some government contract rules to facilitate its patent-related policy objectives.

Although this court's panel opinion did not mention it, the reason that the "improvements" clause in this case does not require an infringement analysis is that the WARF/Xenon agreement features no issued patents, just patent applications. Thus, no court need perform a true infringement analysis. See GAF Building Materials Corp. v. Elk Corp., 90 F.3d 479, 483 (Fed.Cir.1996) (no declaratory judgment jurisdiction for infringement/invalidity of a design patent that had not issued when suit was filed; without an issued patent, there can be no infringement and thus no Article III controversy). Unlike GAF, however, this case features an undeniable controversy between the parties. To resolve this dispute, some court will have to compare the claims of the pending patent applications of the WARF/Xenon agreement with the alleged "improvements," and the contract's use of patent infringement parlance to define this term will necessitate some sort of patent analysis, starting with construction of the pending claims. Thus, this is a patent dispute for still another reason.

Refusing to Stay Preliminary Relief Pending Appeal

Patent

PatentlyO025Bushnell v. Brunton (Fed. Cir. 2009)

Preliminary injunctions are often incredibly important in patent cases. The civil procedure is also fascinating.  The appeal of a preliminary injunction often involves three steps. First, an emergency request for a temporary stay of relief is filed.  If granted, the temporary stay typically lasts only until the parties can brief the question of whether relief should be stayed throughout the appeal.  The third step finally gets to the merits of the appeal. Remember though that the merits of these appeals are whether an injunction should stay in place throughout the rest of the litigation.  Thus, each step asks essentially the same questions about likelihood of success and the harms of action and inaction. 

Bushnell’s patents cover various aspects of a laser range finder. The company sued Brunton and Lanshoo in Kansas for infringement and also filed a motion for a preliminary injunction to stop the ongoing infringement.  On November 25, 2009, Judge Vratil issued the preliminary injunction and Brunton immediately requested a temporary stay from the Federal Circuit. The rotating motions judge (Judge Linn) granted a temporary stay of relief that same day (November 25, 2009). [Link] However that temporary stay was intended only to last until the court could decide whether relief should be stayed for the remainder of the PI appeal.  The decision on that question moved to a merits panel of Chief Judge Michel and Judges Schall and Linn. 

Under the Supreme Court precedent of Hilton v. Braunskill, 481 U.S. 770 (1987), an appellate court may stay relief pending appeal when the moving party establishes a strong likelihood of succeeding on the merits of the appeal or “failing that” when the party has a “substantial case on the merits and the harms factors militate in its favor.”  In this case, the Federal Circuit ruled that “Brunton has not met its burden” and immediately lifted the stay (December 30, 2009 opinion). [Link]  The only problem is that the decision is too late for the Christmas season.

False Marking: Calculating Damages Part I

Patent, , , ,

The Forest Group, Inc. v. Bon Tool Co. (Fed. Cir. 2009).

The false marking statute provides for a fine of “not more than $500 for every … offense.” 35 U.S.C. 292. Past cases have severely limited the false-marking damage award by holding that the sale of thousands of falsely marked items constituted a single “offense” under the statute. The Federal Circuit has rejected those cases – holding here that a qui tam plaintiff may collect up to $500 for each falsely-marked product distributed. This decision is important because it opens the door to potentially large monetary judgments in false-marking cases. Anyone who marks their products as patented or patent pending should take this opportunity to review those markings to ensure that the product being marked falls within the scope of the listed patent and that the patent continues to be valid and enforceable.

At one time Bon Tool bought & sold construction stilts that were manufactured by Forest. However, Bon Tool eventually dropped Forest as a supplier and began importing a duplicate knock-off version from China even though Forest’s stilts were marked with its Patent No. 5,645,515. Forest sued for infringement. During litigation, the district court construed the claims in a way that made clear that neither the original Forest stilts nor the knockoff stilts infringed the patent. At that point (in 2007), Forest’s claims were dismissed on summary judgment. For the next two years, however, the parties argued Bon Tool’s counter claims – including false-marking. In 2009, the district court held that Forest was liable for false-marking because it continued to mark its products as patented even after the 2007 summary judgment decision. However, the court awarded only $500 in damages.

Parties have a reason to mark their products as patented because such marking serves as constructive notice to potential infringers—allowing a patentee to collect damages for past infringement. Under Section 287 of the Patent Act, “[i]n the event of failure so to mark, no damages shall be recovered by the patentee in any action for infringement, except on proof that the infringer was notified of the infringement and continued to infringe thereafter, in which event damages may be recovered only for infringement occurring after such notice.” In theory, marking a product as patented will deter others from attempting to compete by creating a similar version of the product. In addition, manufacturers may garner some reputational benefit by indicated that their product is patented or that a patent is pending.

The false-marking statute is intended to promote competition as a counterbalance against scams and potentially overreaching claims. A successful false-marking claimant must prove two elements: first, that an unpatented article has been marked as patented; and second that the marking was done with intent to deceive the public. See Clontech Labs. Inc. v. Invitrogen Corp., 406 F.3d 1347, 1352 (Fed. Cir. 2005). Here, the Federal Circuit reviewed the district court bench-trial finding of false-marking for clear error. It found no clear error. The appellate panel did, however, hold that the district court had improperly limited the damage award to $500.

Under Section 292 of the Patent Act anyone who marks an “unpatented article with the word ‘patent’ … for the purpose of deceiving the public … shall be fined not more than $500 for every such offense.” Although the false-marking statute has been part of the patent law for more than 150 years, it was amended in 1952. That amendment changed the damage calculation from “not less than one hundred dollars” to “not more than $500.” The leading case interpreting the pre-1952 statute is the one hundred year old decision of London v. Everett H. Dunbar Corp., 179 F. 506 (1st Cir. 1910). In London, the court interpreted the statute to impose “a single fine for continuous false marking”—reasoning that a minimum penalty of $100 per falsely marked article would be out of proportion and inequitable.

[I]f we construe the statute to make each distinct article the unit for imposing the penalty, the result may follow that the false marking of small or cheap articles in great quantities will result in the accumulation of an enormous sum of penalties, entirely out of proportion to the value of the articles. . .

Despite the statutory change, recent courts have followed the London precedent – including the Bon Tool district court. On appeal, the Federal Circuit rejected that precedent – holding instead that the statute requires that each falsely marked article can serve as the basis of a separate offense. The appellate court made clear that the reasoning of London no longer applies because the statute now sets a maximum per-offense award rather than a minimum.

This does not mean that a court must fine those guilty of false marking $500 per article marked. The statute provides a fine of “not more than $500 for every such offense.” By allowing a range of penalties, the statute provides district courts the discretion to strike a balance between encouraging enforcement of an important public policy and imposing disproportionately large penalties for small, inexpensive items produced in large quantities. In the case of inexpensive mass-produced articles, a court has the discretion to determine that a fraction of a penny per article is a proper penalty.

Vacated. On remand, the district court must “determine the number of articles falsely marked by Forest after November 15, 2007 [and] the amount of penalty to be assessed per article.”

Links

Supervisory Patent Examiners: Rethinking their Role

Patent, ,

In his most recent blog post, USPTO Director David Kappos addresses problems and potentials of the USPTO's SPE system. A SPE or Supervisory Patent Examiner serves as the direct manager and trainer for around 13-20 patent examiners. SPE's obviously play an important role in the examination process. Kappos writes:pic-84.jpg

Our first-line managers play a critical role in the achievement of USPTO goals. In fact, my management experience has taught me that first-line managers have *far* more impact on the employees they manage than anyone else in the enterprise – more ability to develop employee skills and careers, more ability to produce outcomes that are successful for both employee and the enterprise. Their day-to-day work as coaches, guides, trainers, and mentors for employees is essential to creating a supportive, effective workplace.

On Patently-O and elsewhere, patent examiners have discussed qualities of good SPEs and bad SPEs. Partially in consideration of those remarks, Director Kappos is focusing on revamping the performance review system for SPEs (AKA the performance appraisal plan or PAP). The idea is to focus SPE attention and energy in the right areas. The performance rating will be based on (1) a measure of the quality of examiner work (25%); (2) a measure of workflow productivity toward the goal of pendency reduction (25%); (3) a measure of how responsive the SPE is to Art Unit examiners as well as to inquiries from applicants (20%); (4) a measure of the SPEs service as a coach and mentor to Art Unit examiners (15%); and (5) a measure of leadership activities of the SPE (15%).

Comments on the new SPE PAP can be sent to spepapawardtaskforce@uspto.gov. (Can someone send me further details on the plan?)

Read Kappos Comments