September 2007

Most Litigated Patents 2005-2007

PatentLawPic047The following table lists the patents asserted most often in patent infringement litigation since January 2005.

  • Patent No. 5,790,512. Asserted by electronics giant Koninklikje Philips in 29 separate lawsuits.
  • Patent Reissue No. 38,014. Asserted by Mag Instruments in 29 separate lawsuits.
  • Patent No. 5,313,229. Asserted by F&G Research in 24 separate lawsuits.
  • Patent No. 4,792968. Asserted by Ron Katz in 19 separate lawsuits.
  • Patent No. 6,766,304. Asserted by Trading Technologies in 17 separate lawsuits.
  • Patent No. 5,352,605. Asserted by Monsanto in 17 separate lawsuits.
  • Patent No. 6,241,739. Asserted by Altair Instruments in 16 separate lawsuits.

Interestingly, only two of these patents appear to be controlled by non-practicing entities.

(From Westlaw Patent Litigation Data)


In November 2007, 37 CFR 1.78(f) will become effective (along with a host of other rules).

Recently, the PTO made public its draft form SB206 for complying with rule 78(f).

1.78 Form

The instructions for completing the form are as follows:


Under 37 CFR 1.78(f)(1), applicant in a nonprovisional application that has not been allowed (37 CFR 1.311) must identify by application number (i.e., series code and serial number) and patent number (if applicable) each other pending or patented nonprovisional application, in a separate paper, for which the following conditions are met:

  • (A) The nonprovisional application has a filing date that is the same as or within two months of the filing date of the other pending or patented nonprovisional application, taking into account any filing date for which a benefit is sought under title 35, United States Code;
  • (B) The nonprovisional application names at least one inventor in common with the other pending or patented nonprovisional application; and
  • (C) The nonprovisional application is owned by the same person, or subject to an obligation of assignment to the same person, as the other pending or patented nonprovisional application.

Under 37 CFR 1.78(f)(2)(i), a rebuttable presumption shall exist that a nonprovisional application contains at least one claim that is not patentably distinct from at least one of the claims in another pending or patented nonprovisional application if the following conditions are met:

  • (A) The nonprovisional application has a filing date that is the same as the filing date of the other pending or patented nonprovisional application, taking into account any filing date for which a benefit is sought under title 35, United States Code;
  • (B) The nonprovisional application names at least one inventor in common with the other pending or patented nonprovisional application;
  • (C) The nonprovisional application is owned by the same person, or subject to an obligation of assignment to the same person, as the other pending or patented nonprovisional application; and
  • (D) The nonprovisional application and the other pending or patented nonprovisional application contain substantial overlapping disclosure. Substantial overlapping disclosure exists if the other pending or patented nonprovisional application has written description support under the first paragraph of 35 U.S.C. 112 for at least one claim in the nonprovisional application.


  1. In the column on the right, list the numbers of the applications and patents that meet the conditions set forth in 37 CFR 1.78(f)(1)(i). For supplemental lists, please do not list previously submitted applications and patents.
  2. In the first column on the left, place a check mark in the box if the cited application or patent also meets the conditions set forth in CFR 1.78(f)(2)(i) and applicant attached an explanation of how the application contains only claims that are patentably distinct from the claims in the cited application or patent.
  3. Place a check mark in the box in the second column on the left, if the cited application or patent also meets the conditions set forth in CFR 1.78(f)(2)(i) and applicant attached a terminal disclaimer in accordance with 37 CFR 1.321(c).
  4. Place a check mark in the box in the third column on the left, if applicant also attached an explanation why there are two or more pending nonprovisional applications containing patentably indistinct claims. See 37 CFR 1.78(f)(2)(ii)(B).

PTO Claim Construction: “Flexible Foam” Is Not “Rigid Foam That Is Flexible When Crushed”

In re Buszard (Fed. Cir. 2007)

The Buszard application is directed to a flame retardant composition that includes a flexible polyurethane foam base. The PTO Patent Appeal Board rejected Buszard’s patent application as anticipated by a prior patent.

On appeal, the CAFC reversed — Judge Newman penned the opinion that focused on the PTO’s practice of giving claims their ‘broadest reasonable interpretation.’

“Buszard’s specification and claims specifically state the requirement of a flexible polyurethane foam…. No matter how broadly “flexible foam…” is construed, it is not a rigid foam…. The [cited prior art] reference describes only a rigid foam reaction mixture that produces a rigid product. Only by mechanically crushing the rigid product into small particles is it rendered flexible, as a rock can be mechanically crushed to produce particles of sand. This description cannot reasonably be construed to describe, and thus to “anticipate,” the flexible foam product of a flexible foam reaction mixture. We agree with Buszard that it is not a reasonable claim interpretation to equate “flexible” with “rigid,” or to equate a crushed rigid polyurethane foam with a flexible polyurethane foam.”

Interestingly, Judge Newman’s claim construction intertwines analysis of the claim language with analysis of the scope of prior art disclosure.  Although the PTO may give an applicant’s claims broad interpretation — the scope of the prior art cannot extend beyond its disclosure.

Judge Prost dissented. In Prost’s view, any ambiguities in patent claim terms should be construed against the patentee during prosecution of the patent. That approach avoids the usual ‘guessing game’ of Phillips-style claim construction and it is quite easy for an applicant to make clarifying amendments to claim language.  

Peter Detkin on the New Patent Market

DetkinPeter Detkin is Intel’s former director of patents, licensing, and litigation who coined the term patent troll. More recently, Detkin co-founded Intellectual Ventures. IV’s product is intellectual property and IP licenses. The company buys patents from small companies and individual inventors and also files patents on its own inventions.

Detkin has an interesting new article in John Marshall’s RIPL journal titled “Leveling the Patent Playingfield.” In the article, Detkin does a great job of explaining IV’s role in the process.

Normally small players have great difficulty simply avoiding being marginalized by major players.  One part of IV’s business is to aggregate the patent rights held by small players and then use its newfound size and power to license those portfolios.

Although IV’s purpose is clearly to make money, he also sees the results as helping the little guy:

“By matching patent owners with patent users, this market may enable small inventors to have a greater stake in their technological efforts.”

The article is an interesting and quick read (especially for a law review).

Ford Design Patents Block Import of Repair Parts – Infringers Call for Patent Reform

Patent.Law031In re Certain Automotive Parts, 2007 WL 2021234 (ITC 2007)

Ford Motor Company owns dozens of design patents covering various aspects of its vehicle designs. D496,890, for example, covers a vehicle grille; D493,552 covers a vehicle head lamp; D503,135 covers a bumper lower valance; and D496,615 covers a side view mirror. In 2005, Ford initiated a Section 337 action before the International Trade Commission (ITC) asking for an exclusion order against various auto parts importers whose imports violated the Ford design patents. For the most part, the accused parts are used to repair post-crash vehicles.

An administrative law judge (ALJ) found that the majority of the asserted patents were valid and infringed, although some patents were invalid.  The ITC then issued an order to exclude importation of the unlicensed repair parts.

Repair doctrine: Over the years, courts have created a non-statutory doctrine of permissible repair. Under the doctrine, post-sale repairs made to a patented object are not considered actionable. On the other hand, reconstruction of the patented object will be actionable as an unlicensed “making” of the invention.

Although interesting, the repair doctrine does not apply here. The accused infringers are not repairing anything — rather, they are importing replacement parts that on their own are infringing.

Calls for Patent Reform: After losing on the merits, the would-be importers have jumped on the patent reform bandwagon asking for a “repair parts” exception to the design patent laws. [Quality Parts Coalition]. Car manufacturers would like to go the other way — adding vehicle hull design protection to the Copyright Act. [See Boat Hull statute]

Appeal: The case, captioned as Ford v. ITC, 07–1357, is now on appeal at the CAFC — A decision is expected summer 2008.  The critical question for Ford — will the patents pass the CAFC’s dreaded points-of-novelty test?

DC Court Rejects PTO’s Novel Theory of Double Patenting

Takeda v. Dudas (D.D.C. 2007)

Takeda’s priority date stretches back to 1974 when the company filed a Japanese patent application disclosing a set of anti-microbial agents known as cephem compounds. The first US patent did not issue until 1988 (‘606 patent). In 1996, Takeda’s ‘216 patent issued – also reaching back to the 1974 priority date. The ‘216 patent claims a process of compounds claimed in the earlier ‘606 patent — those process claims were not, however, added until 15 years after the priority date.

During a reexamination, the PTO Board of Patent Appeals rejected the ‘216 patent after applying the non-statutory double-patenting doctrine. Under the Board’s decision, the Takeda was guilty of attempted double patenting despite the fact that the two sets of claims were not obvioius variations. Instead, the Board found that the second patent was invalid becaues it “unjustly extends the patent rights of the first patent.”

In particular, the Board held that a process patent with an extended patent term will be invalid for double-patenting “where (1) the claimed method for making the [compound] is described in the [original] patent and (2) there is no credible alternative method for making the [compound] which does not involve an infringement of the method patent?” The Board also faulted Takeda for its 15 year delay in presenting the new claims.

Takeda appealed the BPAI rejection to the DC District court. That court granted summary judgment — ordering the PTO to issue a reexamination certificate.

Only way to make a previously patented product: Takeda submitted an unrebutted declaration that convinced the court that there were multiple ways to create the cephem compounds. Thus, the PTO’s conclusions regarding the lack of alternative methods were factually incorrect.  In its decision, the court rejected the PTO’s argument that the alternative methods must have existed as of 1974. Rather, the court held that those methods may be shown in subsequent technological advances.

Double Patenting Based on Delay: The court went on to find that a patentees delay in pursuing claims could never, on its on, be grounds for a double patenting rejection. Rather, the grounds for double patenting always primarily rest on the lack of patentable differences between the two inventions.

WegnerRejection Reversed


  • Hal Wegner along with Ed Polk, Leon Radomsky & Steve Maebious handled the appeal for Takeda
  • BPAI Decision

Continued Patent Trouble for Vonage

Verizon v. Vonage (Fed. Cir. 2007)

Judge Dyk wrote the majority opinion. Each of the five substantive sections of the opinion garnered two votes. Chief Judge Michel joined Parts I and V while Judge Gajarsa joined Parts II-IV. Both Judges Michel and Gajarsa filed additional dissenting opinions.

Earlier in 2007, Vonage was found to infringe three of Verizon’s IP Telephony patents. The verdict awarded $58M in damages along with an ongoing 5.5% royalty rate and an injunction against further infringement.

On appeal, the CAFC affirmed that two of the patents were valid and infringed and that an injunction had been properly granted. As to the third patent (the ‘880 patent), the CAFC vacated and remanded after finding claim construction errors, potential errors in application of the obviousness doctrine under KSR, and errors in determining injunctive relief under eBay. The damage award was also vacated because the jury verdict did not apportion damages according to the various patents.

Thus, Vonage still infringes, but perhaps not as severely. It is unclear at this point whether avoiding infringement of one patent helps Vonage in any substantive way…

Chief Judge Michel would have affirmed across the board. And, interestingly, he would not have disturbed the $58M damage award even though one of the three patents was not infringed.

“When, as here, the evidence shows that each of the accused products infringes all of the patents-in-suit, and the infringer fails to make any showing on appeal that the damages award would not be supported by only those patents for which we affirm liability, we must affirm the damages despite our reversal of part of the infringer’s liability.”

Judge Gajarsa found claim construction errors in one other Verizon patent, but could not convince a colleague.


Sprint v. Vonage (E.D.Kan. 2007) On September 25, a Kansas City jury found that Vonage also infringes a patent held by Sprint Nextel — tacking on another 5% in royalty revenue. [LINK].


Need for Patent Reform?

Won’t it be interesting if the internet telephone industry is decimated through the assertion patent rights held by the largest telephone companies? How does this reflect on patent reform issues? Are there any examples industries being destroyed (or companies crippled) based on patent troll activity/

Supreme Court to Decide Patent Exhaustion Case

Quanta v. LG Electronics (Supreme Court 2007)

The Korean chaebol LG licensed a set of patents to Intel.  The license expressly excludes Intel's customers and similarly does not cover any customer product made by combining an Intel product with a non-Intel product.

When LG sued a host of Intel chip-based computer manufacturers for infringement, they complained of attempted double-dipping.  According to the defendants (including Quanta), their use of licensed Intel chips implicated the first sale doctrine of patent exhaustion.

Now the Supreme Court has granted certiorari to determine whether LG’s narrow license terms avoid exhaustion.

Question presented: “Whether the Federal Circuit erred by holding, in conflict with decisions of this Court and other courts of appeals, that respondent’s patent rights were not exhausted by its license agreement with Intel Corporation, and Intel’s subsequent sale of products under the license to petitioners.”

Exhaustion – also known as the first sale doctrine – serves as a default rule in both patent and copyright laws. Under the principle, a license fee is only be charged one time per object. Thus, a rights holder controls the first sale of a protected object, but does not control subsequent sales. The copyright rule is codified in Section 109. The patent rule, however, is only based on case law.

Lower Court Ruling: The lower court found that Intel’s "unrestricted" license and Intel's sale of its chips under the license exhausted LG's rights to any additional patent recovery.

CAFC Ruling: On appeal, the Court of Appeals for the Federal Circuit (CAFC) reversed, finding that when restrictions are placed on a license, "it is more reasonable to infer that the parties negotiated a price that reflects only the value of the ‘use’ rights conferred by the patentee." The appellate panel found that there was no exhaustion because of express conditions on the use of Intel's licensed products -- specifically that the products could not be combined with non-Intel parts.

Quanta's argument is quite simple:

  • Intel sold chipsets to the petitioners;
  • Those sales were fully authorized under the LG license and were made without restriction;
  • Therefore, LG's patent rights are exhausted.

Quanta goes on to say that under Supreme Court jurisprudence, a sale cannot be conditional. (No improper restraints on alienation).  And, without citing Arizona Cartridge, the petition lists a set of upcoming problems if conditions for sale are allowed for patented items:

[T]he Federal Circuit's new jurisprudence threatens to kick off a new era of "notices" attached to sold goods.  Obvious candidates include "single use only," "no use outside of Massachusetts," "no repair," "no resale," or "no resale for less than the price of purchase."

On the other side, LG points-out that Intel made a business decision not to cover its customers.  Because of that decision, Intel received a cheaper license, but now the customers must make-up the difference.

[When there are no patent misuse or notice concerns,] parceling out of the exclusivity rights is simply a matter for negotiation in the marketplace, not for a legal override of commercially fashioned terms.

An amicus on behalf of Dell, HP and others identifies the origin of the current CAFC precedent: Mallinckrodt.

Mallinckrodt held that a restriction on the post-sale use and enjoyment of a patented article is enforceable under the patent laws as long as it does not “venture[] beyond the patent grant and into behavior having an anticompetitive effect not justifiable under the rule of reason.”

According to the Amicus (and Quanta), the Mallinckrodt decision and its progeny do not follow Supreme Court precedent as stated in Univis Lens.

Univis Lens ... held that the authorized sale of an article manufactured “under the patent” exhausts all patent claims in the article regardless of any purported limitation on the subsequent use and enjoyment of the article. . . Univis Lens held that a condition on the post-sale use and enjoyment of a patented device violated the antitrust laws because it was not enforce-able under the patent laws, and thus was not protected against antitrust scrutiny. See 316 U.S. at 252 (“[t]he price fixing features of appellees’ licensing system, which are not within the protection of the patent law, violate the Sherman Act”).United States v. Univis Lens Co., 316 U.S. 241 (1942)


  • It turns out that the patents were originally issued to Wang Labs and are indeed quite broad. U.S. Patent Nos. 4,918,645; 5,077,733; 4,939,641; 5,379,379; and 5,892,509
  • Mallinckrodt, Inc. v. Medipart, Inc., 976 F.2d 700 (Fed. Cir. 1992)
  • Jazz Photo v. ITC, 264 F.3d 1094 (Fed. Cir. 2001), cert. denied, 536 U.S. 950 (2002)("United States patent rights are not exhausted by products of foreign provenance. To invoke the protection of the first sale doctrine, the authorized first sale must have occurred under the United States patent.").
  • Arizona Cartridge Remanufacturers Association Inc. v Lexmark International Inc, 421 F.3d 981 (9th Cir. 2005)
  • United States v. Univis Lens Co., 316 U.S. 241 (1942)
  • Birdwell, ''Exhaustion of Rights and Patent Licensing Market Restrictions,'' 60 J. Pat. Off. Soc'y 203 (1978)

Let me Repeat: Means-Plus-Function Element Found Indefinite Without Corresponding Structure

Maurice Mitchell Innovations v. Intel (Fed. Cir. 2007) (non-precedential)

Mitchell’s patent is invalid as indefinite. In its opinion, the CAFC used a simple three step analysis:

  1. The claimed “means for causing” is a means plus function limitation.
  2. The specification contains no structure linked to the claimed means.
  3. Therefore, the claim is invalid as indefinite under 35 U.S.C. 112 p2.

The patentee pointed to a structure in the specification that could serve as the claimed “causing means.” That approach failed because the structure must be “linked” within the specification to the claim language.

Quid pro quo: The CAFC noted that a patentee creates additional burdens by using means plus function language:

“The quid pro quo for using a means-plus-function limitation requires specificity in reciting structure and linking that structure to the limitation.”

In recent history, a more accurate version of this quid pro quo may be written as follows: The quid pro quo for using a means-plus-function limitation is that the patent claims will either be found invalid or narrowly construed.

The CAFC had no reason to issue this as a precedential opinion because it almost identically follows the recent Biomedino case: Biomedino, LLC v. Waters Technologies Corp., 490 F.3d 946, 950 (Fed. Cir. 2007).


Invalidation: Corroborating Evidence Required, Invalidity Judged by Totality of the Evidence

PatentLawPic041Adenta GMBH v. OrthoArm and American Orthodontics (Fed. Cir. 2007).


  • OrthoArm patented a new type of bracket for holding orthodontic braces.
  • In 2001, OrthoArm assigned patent rights to American Orthodontics (in a settlement agreement). American agreed to pay an ongoing royalty of 4%. 
  • American distributes Adenta’s products — in a contract between American and Adenta, the German company began to share royalty payments.  
  • Adenta threatened to stop paying its share of the royalties and American responded with threats.
  • Adenta then sued for declaratory judgment of invalidity.

Declaratory Judgment Jurisdiction: In MedImmune, the Supreme Court made clear that declaratory judgment jurisdiction is much broader than allowed by CAFC precedent.  In particular, the court rejected the CAFC’s requirement that a DJ plaintiff show a “reasonable apprehension of suit.”  Under the court’s newly developed precedent, Article III jurisdiction will arise when a patentee asserts its patent rights against another party’s activity or planned activity.  No actual infringement is required nor is any cessation of royalty payments.

Here, jurisdiction was simple because the patent holder had threatened to assert its rights if Adenta failed to pay royalties — “thereby creating a substantial controversy.”

Evidence of 102(b) Public Use: Under the CAFC’s “rules of evidence,” a single person’s testimony is never sufficient to invalidate a patent. Rather, some corroborating evidence must always be presented. Once that threshold is passed, a jury may evaluate the validity of a patent considering all of the evidence as a whole (using a “rule of reason.”)

Here, Adenta brought forth five witnesses to testify that the invention had been shown at the trade show. In addition, Adenta produced a letter from a German attorney directing that the US application be filed quickly because of the trade show exhibit. Despite the patentee’s attempts to discredit the evidence, the CAFC found that the corroboration threshold was met and that the fact finder had not committed clear error in evaluating the totality of the evidence.

Sham Litigation: There is some indication that this was a sham litigation between Adenta and American. Although American owns the patent, it was required to continue to pay royalties to OrthoArm under their original settlement agreement. The CAFC, however, refused to hear that argument.


  • John Fredrickson of Boyle Fredrickson represented Adenta
  • Thomas Reynolds of Michael Best (Milwaukee) took the other side

Warner Lambert’s Neurontin Patent May Be Infringed

PatentLawPic040In re Gabapentin Patent Litigation (Fed. Cir. 2007)

Warner Lambert sells its drug gabapentin under the trade name Neurontin to treat epilepsy and as well as post-shingle nerve pain. The related patent covers a method of making gabapentin that is stable enough to work well for pharmaceutical use.  Using tools of the Hatch-Waxman Act, Warner Lambert filed this patent litigation to keep generic versions of the drug off the market.  After years of litigation, the district court granted the generic manufacturers summary judgment of noninfringement.

Summary Judgment Basics: On appeal, the CAFC reversed — finding that the NJ District Court had been too quick to discount Warner Lambert’s expert witness testimony. In particular, the patentee’s expert had performed a test on Teva’s samples and determined that the acidic content of those samples fell within the scope of the claims. On summary judgment, “all reasonable inferences” must be weighed in favor of the non-moving party — here, those inferences lead to the conclusion that Warner Lambert’s expert might be correct and that the summary judgment was improperly issued.

Claim Construction: The CAFC also affirmed that a claimed “anion of a mineral acid” is properly defined as an “anion derived from a mineral acid” based on the plain language of the claim. 

A jury will likely now decide whether the generic manufacturers have done enough to avoid infringement.

Typically, generic manufacturers stay off the market throughout patent litigation in order to avoid the potential risk associated with infringement damages. Here, however, Teva began selling gabapentin in 2004 and several other generic manufacturers quickly followed.


  • Morris Nichols of Delaware represented the Patentee
  • Mr. Haug of Frommer Lawrence Haug; Steven Lee of Kenyon & Kenyon; and Richard Basile of St. Onge represented the various generic manufacturers.



Threatening Letters to Infringer’s Clients Are Not Actionable Unless “Objectively Baseless” (Although They Will Create DJ Jurisdiction)

PatentLawPic039GP Industries v. Eran (Fed. Cir. 2007)

Eran makes “leakproof” rain gutter covers. Several Eran employees were fired during the summer of 2005.  Those individuals reorganized as GPI — a potential competitor to Eran. As a preemptive strike, Eran mailed letters to its distributors and contractors noting that GPI’s planned to sell a competitive product that infringed Eran’s patent.

After GPI filed for declaratory judgment, the district court (Nebraska) granted GPI’s motion for a preliminary injunction — stopping Eran from “(1) seeking to prevent GPI from making and selling its gutter products during the pendency of this litigation and (2) making vague and unspecified allegations of infringement against GPI in the marketplace.”

On appeal, the CAFC reversed — finding that the preliminary injunction had been improperly granted.

The party seeking a preliminary injunction has the burden of showing: “(1) a reasonable likelihood of success on the merits; (2) irreparable harm; (3) a balance of hardships in its favor; and (4) public interest in favor of the injunction.” Once granted, a preliminary injunction will only be vacated for clear error.

Injunction against Communications: Based on First Amendment rights and 35 U.S.C. § 287 (notice), injunctions relating to communications must pass a higher hurdle. In particular, “bad faith” is required in order to bar communications.

“One has a right to inform others of his or her patent rights. See 35 U.S.C. § 287. Thus, an injunction against communication is strong medicine that must be used with care and only in exceptional circumstances.”

In Globetrotter Software, the CAFC held that “bad faith” communications require “objectively baseless claims.”  Notably, these determinations are based on federal law that preempt any state laws that may be less strict.

Here, the CAFC found that Eran’s statements could not be “objectively baseless” since the lower court had noted the real potential that Eran would eventually succeed on the merits of the case.

Preliminary Injunction Reversed


  • Hovey Williams of Kansas City, Missouri represented the patentee, Eran.
  • Kirton & McConkie of Salt Lake City, Utah represented the accused infringer.

Ineligible Subject Matter: 35 USC 101 Finds its Teeth (Biting into Nonobviousness)

In re Comiskey (Fed. Cir. 2007); and
In re Nuijten (Fed. Cir. 2007)

  • Comiskey’s patent application claims a method of arbitrating that does not necessarily require the use of any ‘technology.’
  • Nuijten’s patent application claims a signal containing a ‘watermark’ without being tied to a specific type of signal or any technology.

Both applications have been denied by the Federal Circuit under 35 U.S.C. Section 101 for their failure to claim patentable subject matter.

The scope of patentable subject matter is defined by Section 101 as any “new and useful process, machine, manufacture, or composition of matter.”  Any patent claim must fall within at least one of these categories.  Further limiting these categories is our precedential history that not every new and useful process constitutes patentable subject matter. See, for example, Flook & Benson.  In particular, the Supreme Court has noted that a process claim reciting an algorithm must either be (1) tied to a machine or (2) transformative of some composition of matter or device. Flook & Diehr.

Under these precedents, a claim that includes a mental process may be patentable — so long as it is otherwise tied to ‘reality.’

Comiskey: The Comiskey method of claim 1 admittedly does not require a machine, but it does require interaction with a ‘document.’  The reference to a document, however, was not enough to allow the “mental process” claim to be treated as legitimate.

“Thus, like the claims that the Supreme Court found unpatentable in Benson and Flook and the claims found unpatentable in our own cases, Comiskey’s independent claims 1 and 32 seek to patent the use of human intelligence in and of itself. Like the efforts to patent “a novel way of conducting auctions” which Schrader found to be directed to an abstract idea itself rather than a statutory category, Comiskey’s independent claims 1 and 32 describe an allegedly novel way of requiring and conducting arbitration and are unpatentable.”

Some of Comiskey’s claims arguably require a computer to operate through “modules.”  That claim construction transforms those claims to be patentable subject matter.  This means that software remains patentable subject matter if a computer limitation is a proper part of the claim.

Section 101 is Safe: Comiskey appear to reinstate something akin to the USPTO’s ‘technological arts’ requirement that had been eliminated by the Lundgren decision.  For Comiskey, recitation of a microprocessor as a limitation of the claims was sufficient to overcome the patentable subject matter issue. Notably, this is a rule of formality that, in most instances, can be followed without substantively reducing claim scope.  Under Comiskey, there is no requirement that the technical link be something novel or non-obvious. 

Section 103 is Lost: Although the question of patentable subject matter does not require a determination of nonobviousness, the reverse is not true.  Nonobviousness may now require an input from the patentable subject matter stage. The most interesting lines of Comiskey are as follows:

“The routine addition of modern electronics to an otherwise unpatentable invention typically creates a prima facie case of obviousness. Moreover, there is no pertinent evidence of secondary considerations because the only evidence offered is of long-felt need for the unpatentable mental process itself, not long-felt need for the combination of the mental process and a modern communication device or computer.”

The implicit holdings here: (1) evidence of secondary considerations do not apply to portions of an invention that are not considered “patentable subject matter;” and (2) during nonobviousness analysis, any portion of an invention that constitutes nonstatutory subject matter will be considered de facto obvious.

Nuijten: The Nuijten court bit off a bit more — finding that a signal is not patentable even if tied to a transitory form (such as a radio broadcast or light pulses in a fiber optic cable).  The problem – the court could not fit a ‘signal’ into any of the four categories:

  • Process: Process is defined in Section 100 as a “Process, Art, or Method.” The ‘art’ term appears on its face different than a typically process — especially based on the constitutional statement of “useful arts.” However, the CAFC refused to expand the meaning of process to include items that do not require an action. Thus, a signal is not a process.
  • Machine: The Supreme Court has defined a machine as “a concrete thing, consisting of parts, or of certain devices and combination of devices.” Burr (1863). Under this definition, a signal is not a machine.
  • Manufacture: The court limited a manufacture to an “article” produced by man. According to the court, an ‘article’ is not transient and cannot exist in a vacuum — both qualities of a signal. Thus, a signal is not a manufacture.
  • Composition of Matter: A transient electric signal is not a “chemical union, nor a gas, fluid, powder, or solid.” Therefore, a signal is not a composition of matter.

Failing to fit within a predefined category, the only conclusion is that a signal is not a patentable subject matter.

Rules of Practice:

  • Gangrene Treatment: In the political world of the courts, this case might be seen as the CAFC’s gangrene treatment — where the Supreme Court is the disease and patent law jurisprudence is the patient.  Several toes have already been lost, and the CAFC is cutting off a foot to prevent the disease from spreading any further.
  • Morse Code: One important statutory subject matter case is that of O’Reilly v. Morse (1853) – invoving the patenting of Morse code. In that case, the Supreme Court invalidated some claims as overbroad, but found that Morse’s claim to a “system of signals” to be patentable as an ‘art.’  The CAFC, of course, does not overrule Morse, nor does it adequately explain how Morse’s patent would fit within our modern conception of the four categories.

CAFC Denies Joint Infringement Claim and Maps Out Infringement Avoidance Schemes

BMC v. Paymentech (Fed. Cir. 2007)

BMC’s patents cover a method for processing a debit transaction without using a PIN. The payment industry is divided so that there is no single party that infringes an entire claim. However, when combined, the effort of a payee’s agent, an ATM network, and a financial institution

The usual rule is that direct infringement occurs only when a single actor performs each and every element” of an asserted claim. Likewise, inducement requires proof of underlying single-actor direct infringement.  Under a doctrine akin to agency, all actions ‘controlled’ by the single actor are considered to be performed by the single actor.  Thus, courts have long held that you cannot avoid liability by simply “having someone else carry out one or more of the steps” on your behalf. 

Here, BMC acts for an extension of the law in cases where parties are cooperating in a manner that infringes the patent – even though no single party controls the others. The CAFC, however, refused to extend the law — even while admitting that its decision paves a clear pathway for would-be infringers to avoid liability:

“This court acknowledges that the standard requiring control or direction for a finding of joint infringement may in some circumstances allow parties to enter into arms-length agreements to avoid infringement.”

The CAFC’s concern, of course is that allowing this type of joint infringement would greatly increase potential liability of many many companies. For instance, here, BMC would like to hold ATM networks liable even though their actions do not involve any novel part of BMC’s invention. Without any controlling party, BMC’s claim of infringement fails.

As part of its justification, the court cited Mark Lemley’s divided infringement article. In that article, Lemley argues that patent drafters can avoid most potential joint infringement problems by drafting claims that focus on one party at a time.


  • The doctrine of control is certainly not as strict as typical agency doctrine.  And, in this opinion the court gives some positive language to capture bad behavior. For instance, the court noted approvingly that district courts have held parties liable for who ‘mastermind’ a scheme of contracting-out steps of a patented process.  As a strict liability claim, the CAFC is correct to avoid undue expansion of the class of potential defendants. However, the hole created by the CAFC is potentially quite large — I have proposed a new cause of action for conspiratorial infringement that would not require a single controlling entity, but would require a mens rea requirement similar to that of inducement.  
  • Related Material:
    • Fromson v. Advance Offset Plate, Inc., 720 F.2d 1565, 1568 (Fed. Cir. 1985) (finding no direct infringement by manufacturer who performed the first step of a process claim even where its customer performed the other step of the claim)
    • Cross Medical Products, 424 F.3d at 1311 (rejecting patentees’ efforts to combine the acts of surgeons with those of a medical device manufacturer to find direct infringement of an apparatus claim).
    • An expanded joint infringement notion is supported by the CAFC’s recent On Demand decision where the court noted that “[i]t is not necessary for the acts that constitute infringement to be performed by one person or entity.”
    • In PharmaStem, the CAFC did not decide the issue, but noted that the “viability and scope of [the theory of joint infringement] liability is a subject of considerable debate.”
    • Applied Interact, LLC v. Vermont Teddy Bear Co., Inc.  2005 WL 2133416 (S.D.N.Y.,2005), the district court noted that infringement by separate entities requires “some connection” between the entities.
    • Free Standing Stuffer, Inc. v. Holly Dev. Co., 187 U.S.P.Q. 323 (N.D. Ill. 1974), requires a “sufficient connection to, or control over” third party entities performing some of the elements.
    • Cordis Corp. v. Medtronic AVE, Inc., 194 F. Supp. 2d 323 (D. Del. 2002) (requiring close relationship between defendant and the doctors performing part of the patented process).
    • Shields v. Halliburton Co., 493 F.Supp. 1376 (W.D. La. 1980) (finding defendants liable for infringement based on the combined actions of two entities).
    • Wegner well thought viewpoint from 2001
    • Sriranga Veeraraghavan, Joint Infringement of Patent Claims: Advice for Patentees, 23 Santa Clara Computer & High Tech L.J. 211 (2006).


How not to divide patent rights during bankruptcy

Morrow v. Microsoft (Fed. Cir. 2007)

In bankruptcy, AHC’s title to its patent transferred to a liquidating trust, AHLT. A second trust, GUCLT, was given the right to several causes of action, including patent infringement litigation (generally called Estate Litigation).  GUCLT was not granted a license to make or use the patent.

Through its trustee, GUCLT sued Microsoft for infringing Patent No. 6,122,647 – a patent relating to dynamic generation of hypertext links. Microsoft was awarded summary judgment on the merits. On appeal, GUCLT had trouble even getting a hearing.

Standing: To have standing, a party must show injury:

  • The patent title holder has standing to sue because infringement injures the patent holder’s right to exclude.
  • An exclusive licensee usually has standing to sue because the infringement injures the licensee’s ‘exclusivity.’ When the exclusive licensee sues, the patent title holder must be joined as a co-plaintiff.
  • A non-exclusive licensee usually does not have standing because they have no right to stop others from making or using the invention.
  • A non-licensee also has no standing to sue.

GUCLT holds the contractual right to sue, but is not a licensee. Ergo, GUCLT does not have standing.  (Under the bankruptcy plan, AHLT likewise has no right to bring litigation…)

Rule of Practice: A non-title-holder must be granted an exclusive license as well as full litigation rights in order to have standing to sue for patent infringement.

IP Hall of Fame 2007

WinningCupThe conservative British publication, Intellectual Asset Management (IAM) Magazine, sponsors an annual induction ceremony into the IP Hall of Fame. This years induction will be part of the Ocean Tomo Fall Auction in Chicago next month. However, IAM has released the names of winners already. Some highlights:

  • Judge Pauline Newman (Federal Circuit Judge)
  • Kevin Rivette (of Rembrandts in the Attic)
  • Gerald Mossinghoff (Former USPTO Commissioner)
  • Jerome Gilson (of Gilson on Trademarks)
  • Karl Jorda (Franklin Pierce Law Center)
  • Joseph Straus (Max Planck Institute for IP)
  • Hugh Laddie (All around UK IP Bigwig)
  • Hisamitsu Arai (Former Japanese Patent Office Commissioner)

I am a voting ‘academy member’ of the Hall of Fame along with a host of better qualified intellectual property leaders.



Patently-O Housekeeping

  • Patently-O continues to grow with over 280,000 visits in August 2007. (Double last-year’s numbers). Please let me know when you run across some interesting patent law news for the site!
  • Patently-O has now been consumed by a Missouri Limited Liability Corporation, Patently-O, LLC.
  • Part of the ‘business’ transition includes site sponsorship. The Chicago Intellectual Property Firm MBHB LLP (my former law firm) has become the first major sponsor.  Current MBHB attorneys author two other well respected IP related blogs: The OrangeBookBlog and PatentDocs. Thanks MBHB!

$156 Million Verdict for Patentee

TGIP v. AT&T (E.D.Tex. 2007)

TGIP has won a jury verdict against AT&T for $156 million. The asserted patents – U.S. Patent Nos. 5,511,114 and 5,721,768 — both involve calling card processing. The jury also reportedly decided that the infringement was willful – District Court Judge Ron Clark will now decide whether to enhance damages based on the finding. (Judge Clark will also rule on JNOV motions).

TGIP appears to be a company that run by Charles Stimson, one of the named inventors. Interestingly, neither patent show-up in the PTO assignment database as assigned to TGIP. Stimson was formerly the president of “Call Processing, Inc.,” which is listed as an assignee (but not a plaintiff). In the late 1990’s CPI sued several companies for infringing the same patent. That case was dismissed without prejudice.

AT&T will appeal.


  • Verizon has reportedly settled with TGIP for an undisclosed sum. Other companies that (likely) settled include: IDT, DFTS Interactive, US South, Pre Solutions, Bell Atlantic, and reportedly MCI.

How long does a BPAI appeal take?

PatentLawPic037In the first two weeks of September 2007, the Board of Patent Appeals and Interferences released over one hundred opinions. I looked at the file wrapper histories of thirty of these recent BPAI decisions to get some picture of the timing of BPAI appeals.

On average it took just under 18 months (542 days) from the filing of the appeal brief until a decision was reached by the BPAI.  The mean hides a wide range of delay: From a minimum of 9 1/2 months to a maximum of 44 months. (Standard Deviation 255). 

Most of the time delay is not attributable directly to the BPAI. Rather, most of the delay occurs between the date that the appeal is filed and the date that the case is submitted to the BPAI.  On average, it took about 11 months (328 days) to ‘complete briefing.’ Typically, that time period involves the applicant’s brief; followed by an examiner’s answer; and finally a reply brief. Once the case is submitted to the BPAI, the average decision time was seven months (214 days).

If you win an appeal, the Examiner may reopen prosecution with an additional rejection, although the more common approach is to issue a notice of allowance.

Based on this information, you can advise clients that an appeal – pushed through to the end – takes an average of 18 months, but that there is a wide variance.  As I noted earlier in 2007, only fewer than of appeals are pushed-through to the end. [Link].  In many cases, Examiners withdraw rejections or applicants file RCE’s with new claims. These calculations also do not include time delays associated with BPAI rejections for improper appeal brief form. Instead, I only began counting once a ‘proper’ brief was filed. In an earlier study, I showed that approximately 25% of appeal briefs are rejected on procedural grounds as either defective or incomplete. [Link].

The prosecution history of a patent application typically includes at least two substantive rejections prior to the appeal brief. In this sample, the average application had been in process for over three years (39 months) before the appeal brief was filed. 

As with essentially every other area of patent law, we can expect that BPAI timing and results will vary by technology area.