close

For more than four decades, Fenwick & West LLP has helped some of the world’s most recognized companies become, and remain, market leaders. From emerging enterprises to large public corporations, our clients are leaders in the technology, life sciences and cleantech sectors and are fundamentally changing the world through rapid innovation.  MORE >

Fenwick & West was founded in 1972 in the heart of Silicon Valley—before “Silicon Valley” existed—by four visionary lawyers who left a top-tier New York law firm to pursue their shared belief that technology would revolutionize the business world and to pioneer the legal work for those technological innovations. In order to be most effective, they decided they needed to move to a location close to primary research and technology development. These four attorneys opened their first office in downtown Palo Alto, and Fenwick became one of the first technology law firms in the world.  MORE >

From our founding in 1972, Fenwick has been committed to promoting diversity and inclusion both within our firm and throughout the legal profession. For almost four decades, the firm has actively promoted an open and inclusive work environment and committed significant resources towards improving our diversity efforts at every level.  MORE >

At Fenwick, we are proud of our commitment to the community and to our culture of making a difference in the lives of individuals and organizations in the communities where we live and work. We recognize that providing legal services is not only an essential part of our professional responsibility, but also an excellent opportunity for our attorneys to gain valuable practical experience, learn new areas of the law and contribute to the community.  MORE >

Year after year, Fenwick & West is honored for excellence in the legal profession. Many of our attorneys are recognized as leaders in their respective fields, and our Corporate, Tax, Litigation and Intellectual Property Practice Groups consistently receive top national and international rankings, including:

  • Named Technology Group of the Year by Law360
  • Ranked #1 in the Americas for number of technology deals in 2015 by Mergermarket
  • Nearly 20 percent of Fenwick partners are ranked by Chambers
  • Consistently ranked among the top 10 law firms in the U.S. for diversity
  • Recognized as having top mentoring and pro bono programs by Euromoney

MORE >

We take sustainability very seriously at Fenwick. Like many of our clients, we are adopting policies that reduce consumption and waste, and improve efficiency. By using technologies developed by a number of our cleantech clients, we are at the forefront of implementing sustainable policies and practices that minimize environmental impact. In fact, Fenwick has earned recognition in several areas as one of the top US law firms for implementing sustainable business practices.  MORE >

At Fenwick, we have a passion for excellence and innovation that mirrors our client base. Our firm is making revolutionary changes to the practice of law through substantial investments in proprietary technology tools and processes—allowing us to deliver best-in-class legal services more effectively.   MORE >

Mountain View Office
Silicon Valley Center
801 California Street
Mountain View, CA 94041
650.988.8500

San Francisco Office
555 California Street
13th Floor
San Francisco, CA 94104
415.875.2300

Seattle Office
1191 Second Avenue
10th Floor
Seattle, WA 98101
206.389.4510

New York Office
1211 Avenue of the Americas
32nd Floor
New York, NY 10036
212.921.2001

Shanghai Office
Unit 908, 9/F, Kerry Parkside Office
No. 1155 Fang Dian Road
Pudong New Area, Shanghai 201204
P.R. China
+86 21 8017 1200


Intellectual Property Bulletin Fall 2004

Federal Circuit Whacks the Doctrine of Equivalents Again

By: Heather N. Mewes

Whack-a-mole is a carnival game that involves quickly and repeatedly hitting the heads of mechanical moles with a mallet as they pop up. At the Federal Circuit, they have their own variation of this game—every time the doctrine of equivalents pops up, a gavel-wielding en banc panel strikes it down.

After the Supreme Court resuscitated the doctrine of equivalents in Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722 (2002), the Federal Circuit struck back. In its remand decision, an en banc panel set forth "general guidance" that effectively limited the patentee's ability to rebut the presumption of surrender. Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 304 F.3d 1289 (Fed. Cir. 2002) (en banc). More recently, in Honeywell Int'l, Inc. v. Hamilton Sundstrand Corp., 370 F.3d 1131 (Fed. Cir. 2004) (en banc), the Federal Circuit expanded the set of claim limitations potentially subject to prosecution history estoppel.

In Honeywell, the Federal Circuit considered the question of what constitutes a narrowing amendment under Festo. Prosecution history estoppel operates to limit application of the doctrine of equivalents and to restrict patentees to the literal scope of the claim language. Under Festo, a narrowing amendment may trigger application of prosecution history estoppel. Broadening the definition of a "narrowing amendment" means that more claim limitations will potentially be subject to prosecution history estoppel and more patentees can expect to be limited to the literal scope of their claims.

Honeywell involved a situation where the patent applicant, during prosecution, amended a dependent claim to rewrite it in independent form. Consider a simple analogous example. The applicant submits the following claims for examination:

1.   An apparatus comprising:
  (a)  a gizmo; and
  (b)  a widget.
2.   The apparatus of claim 1, further comprising:
  (c)  a thingamajig.

The examiner rejects claim 1 over prior art, but indicates that claim 2 would be allowable if rewritten in independent form. Thus, claim 2 is objected to but never rejected. The applicant accordingly cancels claim 1 and amends claim 2 as follows:

2.   An apparatus as claimed in claim 1, further comprising:
(a)  a gizmo;
(b)  a widget; and
(c)  a thingamajig.

Claim 2 is allowed as amended and the patent issues. The patentee then asserts claim 2 against a device with (a) a gizmo, (b) a widget and (c) a thingamabob. While this device would not fall within the literal scope of the claim as amended, the patentee presents evidence that a thingamabob is equivalent to the thingamajig recited in limitation (c) of claim 2. The accused infringer invokes prosecution history estoppel, arguing that claim 2 was the subject of a "narrowing amendment."

The scope of claim 2 in this example (and in the claims at issue in Honeywell) is exactly the same before and after amendment. This is because a "claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers." 35 U.S.C. § 112 ¶ 4. Thus, unamended claim 2 in the example above always included limitations (a) and (b) regardless of the form in which it was presented. Nonetheless, in Honeywell, the Federal Circuit concludes that such an amendment is a "narrowing amendment" within the meaning of Festo.

The Federal Circuit reaches this result by focusing on "whether the amendment narrows the overall scope of the claimed subject matter" rather than on whether the amendment narrows the scope of the claim. Honeywell at 1141. This paradigm shift finds support in language from the Supreme Court's decision in Festo. In Festo, the Supreme Court found that "[e]stoppel arises when an amendment is made to secure the patent and the amendment narrows the patent's scope." Festo, 535 U.S. at 736 (emphasis added). Accordingly, the majority holds that "the fact that the scope of the rewritten claim has remained unchanged will not preclude the application of prosecution history estoppel if, by canceling the original independent claim and rewriting the dependent claims into independent form, the scope of subject matter claimed in the independent claim has been narrowed to secure the patent." Honeywell at 1142 (emphasis added).

This holding has broader implications for future cases. Consider another example where the applicant has two independent claims of varying scope as set forth above:

1.   An apparatus comprising:
  (a)  a gizmo; and
  (b)  a widget.
2.   An apparatus comprising:
  (a)  a gizmo;
  (b)  a widget; and
  (c)  a thingamajig.

Claim 1 is rejected in light of prior art, while claim 2 is allowed. The patentee elects not to challenge the rejection of claim 1, and the patent issues. Under Honeywell, the original, unamended claim 2 may nonetheless be subject to prosecution history estoppel. By canceling claim 1, a broader claim, the patentee has surrendered the subject matter between claim 1 and claim 2 in order to secure the patent. There is no substantive difference between this example and the rewritten dependent claim. Accordingly, it seems likely that the Federal Circuit would extend its ruling to cover this situation.

This is likely to result in an increasingly complex "narrowing" analysis. The focus shifts from one claim to the entire set of claims. The cancellation of one claim can resonate throughout all the claims, foreclosing application of the doctrine of equivalents to claims included in the original application and never rejected or amended.

The Federal Circuit's holding draws the ire of Judge Newman, who authored a dissent. Judge Newman objected to the majority's reliance on Festo for its holding. The language relied on by the majority could be explained away as a loose use of the term "patent" in place of the correct term "claim"; a use that is unsurprising given the rarity of patent cases before the Supreme Court.

This issue was not directly presented to the Supreme Court for decision, and reading such a ruling into Festo could be construed as overreaching. On the other hand, Festo supports an equitable approach to prosecution history estoppel, an approach consistent with divining the substantive effect of an amendment rather than its technical form. Consider the following: What if, in the first example above, the applicant, instead of canceling claim 1 and rewriting claim 2, canceled claim 2 and rewrote claim 1:

1.   An apparatus comprising:
(a)  a gizmo; and
(b)  a widget; and
(c)  a thingamajig

The amended independent claim 1 has exactly the same scope as the rewritten dependent claim 2. Moreover, there would be little doubt that this amendment was narrowing; the "thingamajig" limitation has been added to claim 1 and narrowed the scope of this claim. However, there is no reason that the result should be different in these two examples. The patentee ends up with exactly the same claim for exactly the same reason. Equity demands that these situations be treated the same.

Judge Newman prophesizes that patent applicants will have to use more independent claims and will do away with dependent claims. However, under the logic of Honeywell, the use of an independent claim in place of a dependent claim is a distinction without a difference. Nonetheless, the majority's decision is likely to result in increasing limitations on the doctrine of equivalents and potentially, increasing costs of prosecution associated with attempts to avoid these consequences.

Prosecutors, faced with Honeywell, may have few good options. Because the Supreme Court is focused on the substantive effect of an amendment on the patent scope, practices that achieve the same result using different tactics appear to be destined for the Federal Circuit's trusty mallet. One possible prosecution tactic is to invert the usual prosecution practice; instead of starting broadly, and then seeking progressively narrower claims, it may be possible to circumvent the Federal Circuit's strictures by starting narrowly and then seeking progressively broader claims. This would permit the narrow claims to issue without any negative prosecution history relating to potentially broader claims. However, it is also likely to increase prosecution costs and delay issuance of broader and usually more important claims. It is also no guarantee; the Federal Circuit may invoke the prosecution history of a later related application against the earlier application. Using different language in different independent claims (a practice already prevalent) may also avoid some problems by making it difficult to line up unamended claims with amended claims such that an amendment to one will impact another.

However, perhaps the more sound practice is to simply recognize that the doctrine of equivalents is unlikely to be available and act accordingly. For example, instead of rewriting a dependent claim in independent form, rewrite the independent claim and add only what is necessary to overcome the rejection. While equivalents are unlikely to be available, at least there is no surrender of the entire ground between the original independent claim and the dependent claim. With a Federal Circuit hostile to the doctrine of equivalents, the best bet may be to bet on not being able to count on this doctrine.

The Role of Dictionaries in Patent Claim Construction

By: Michael J. Sacksteder

The process for using dictionaries in patent claim construction continues to allow flexibility in the hands of creative advocates and judges. Over the past several years, Federal Circuit judges have engaged in a debate over the extent to which a patent's specification should be used to narrow the constructions of claim terms. While everyone agrees that limitations are not to be read from the specification into the claims, there is a difference of opinion as to the extent to which the specification should be used to narrow the claims. This debate is analogous to the debate between strict constructionists and contextualists in constitutional interpretation, with the patent law strict constructionists preferring to stick with the "ordinary and customary meaning" of the claim terms themselves, and viewing reliance on the specification with suspicion.

It is hardly surprising that the strict constructionists give dictionaries an exalted place in the hierarchy of claim construction evidence. Liberal reliance on dictionaries permits courts to focus on the claim terms themselves. While the specification and prosecution history must be considered, this approach defers consideration thereof until the court has already examined the plain language of the claims and tentatively ascribed meaning to that language. Pro-dictionary judges effectively view dictionaries as an extension of the language of the claims, enabling one to divine the ordinary and customary meaning of such language.

Accordingly, the pro-dictionary, strict constructionist opinions have erected the following analytical framework, built solidly on the foundation of the claim language:

  • First, the court consults relevant dictionaries to determine the "ordinary and customary meaning" of the claim terms. This consultation occurs before the court looks to the patent's specification and prosecution history.
  • Since words often have multiple dictionary definitions, the court then consults the specification and prosecution history to determine which dictionary meaning is "most consistent" with the words used by the inventor. If the inventor's use of the words is consistent with more than one definition, the claim terms may be construed to encompass all such definitions.
  • The court must also consult the specification and prosecution history to determine whether the patent applicant "acted as his own lexicographer," and clearly gave a claim term a special meaning.
  • The court also must determine whether the specification or prosecution history uses "words or expression of manifest exclusion or restriction, representing a clear disavowal of claim scope" that would otherwise result from using dictionaries to determine the meaning of the term.
  • If no clearly inconsistent definitions or clear disavowals of claim scope appear in the intrinsic evidence, the court typically uses the appropriate dictionary definitions of the claim terms as its construction.

This seems like a fairly rigid analytical structure, and in the hands of some judges, it might be. But as demonstrated in the recent Novartis Pharmaceuticals v. Eon Labs case before the Federal Circuit, a creative court—or advocate—can find significant wiggle room. The majority in that decision indulged in remarkable contortions to find an ambiguity in dictionary definitions, which required an excursion into the patent's specification to be resolved.

The patent at issue in Novartis claimed a "hydrosol" for administering the drug cyclosporin. The patent's specification described a formulation in which the cyclosporin is dissolved in a water-miscible solvent, to which water is then added. The resulting mixture can be administered intravenously. Neither party disputed that the defendant's accused product was not such a mixture when sold or administered. Rather, the defendant's cyclosporin was in capsule form. Novartis alleged that the infringing hydrosol was created "when the capsule mixes with the aqueous environment of the user's stomach." The district court disagreed, and granted summary judgment of noninfringement. The question on appeal was whether the term "hydrosol" was limited to products prepared outside the body, or whether it included products formed within the patient's stomach after being ingested.

The majority started, logically enough, with the definition of the claim term—finding that according to Webster's Dictionary, a "hydrosol" is a "sol in which the liquid is water." This led the majority to ask what a "sol" is. The same dictionary provided the answer: "a dispersion of solid particles in a liquid colloidal solution." But what's a "solution"? Webster's provided two definitions: (1) "a liquid containing a dissolved substance" and (2) "a liquid and usually aqueous medicinal preparation with the solid ingredients soluble." While the first definition was broad enough to cover the accused product when combined with the water in the patient's stomach, the second, as the majority went on to explain, was not.

That explanation rested on further definitions from the majority's now well-thumbed dictionaries. The word "medicinal" means "of or relating to medicine," and a "medicine" is "any substance or preparation used in the treatment of a disease." (The majority recruited the Oxford English Dictionary on this definition.) The majority then synthesized definitions from medical dictionaries to extract the conclusion that a "preparation" is a substance that is made prior to being administered. Thus, a "medicinal preparation" (from definition (2) of "solution") is a "preexisting product that is administered to treat disease and therefore must necessarily be prepared outside the body." Hence, assuming that definition (2) for "solution" is the correct one to apply in this context, the construction of the claim term "hydrosol" would be limited to "a medicinal preparation consisting of a dispersion of solid particles in an aqueous colloidal solution formed outside the body."

The majority then had to back up to make sure that definition (2) of "solution" was the right one. In order to resolve this definitional ambiguity, the court went where some Before Dictionary Era opinions would have started: to the patent's specification. The specification distinguishes the patented hydrosol from the prior art on the use of cyclosporine particles "in an aqueous hydrosol form for intravenous injection purposes." The majority also found it significant that the specification repeatedly describes the invention as a "pharmaceutical composition." This sent the majority back to Webster's for one more definition: Since a "pharmaceutical" is a "medicinal drug," the majority circled back and reentered the definition (2) branch of the freeway via the "medicinal" on-ramp. Thus, the majority adopted the construction of "hydrosol" springing from definition (2) of "solution" (requiring formation "outside the body"), and affirmed the district court's grant of summary judgment.

The Novartis majority employed what may be the most vulnerable breach in the plain meaning fortress—the determination of the dictionary definition that is "most consistent" with the intrinsic evidence. Granted, in Novartis, the majority had to extend its analysis several degrees of separation from the claim term at issue—from "hydrosol" to "sol" to "solution"—before it found the ambiguity upon which it based its invocation of the specification, and several more degrees of separation—from "medicinal" to "medicine" to  "preparation"—before it could arrive at its construction. But such a journey is not always necessary. Dictionaries often provide more than one possible definition for a term, and the resolution of conflicts between these definitions will often require an examination of the specification. The same is true for definitions appearing in different dictionaries.

Courts also must determine whether to rely on general-usage dictionaries, technical dictionaries or, according to some non-strict constructionist opinions, no dictionary at all. According to recent Federal Circuit precedent, if a claim term would have a special meaning to one skilled in the relevant art, a technical dictionary might be the place to find the correct definition. Conversely, if a skilled artisan would find the term at issue to have the same meaning as the term would have in its common usage, a general-usage dictionary might be appropriate. How does the court determine which is the case? According to the Federal Circuit's recent Vanderlande Industries v. ITC case, the answer is sometimes found in technical dictionaries (presumably when a claim term is not defined), and other times courts should rely on expert testimony to make this determination.

Then, according to the Vanderlande court, there is the situation where a skilled artisan "would attach no meaning at all to that claim term (independent of the specification)." Under those circumstances (according to this contextualist opinion), resort to any dictionary would likely result in error.

Thus, notwithstanding their apparently authoritative nature, dictionaries are not always decisive in construing claim terms. A number of factors—some exceeding those sanctioned by the strict constructionists—may allow a court to circumvent the "plain meaning" of a term as found in a dictionary and explore the contextual meaning found in the patent's specification and prosecution history.

In July of 2004, the Federal Circuit granted an order for an en banc rehearing of the case Phillips v. AWH Corp. in order to clarify the law on various claim constructions issues, including the proper role of dictionaries. Whatever the merits and shortcomings of dictionary use in claim construction, at least the ongoing Federal Circuit debate on this topic should soon be settled.

Quick Updates

A Generic Term Cannot Be Protected Under the ACPA

In Retail Services, Inc. v. Freebies Publishing, 364 F.3d 535 (2004), the Fourth Circuit ruled that a generic term is not a "mark" as defined under the Anticybersquatting Consumer Protection Act (ACPA) and that, accordingly, an owner of such a term cannot bring a claim under the ACPA because it would not be able to satisfy the ACPA prerequisite that an owner establish rights to a "mark."

The case involved the name FREEBIES. Freebies Publishing used the name for its magazine and website, both of which provided information about free mail order offerings. In 1993, it successfully obtained a trademark registration for FREEBIES. In 2001, Retail Services, Inc. began operating a website at www.freebie.com. RSI used this site to manage Blockbuster, Inc.'s promotional freebie program, in which Blockbuster, Inc.'s customers could exchange "Freebie Points" for free video and DVD rentals. Later that year, Freebies Publishing contacted RSI and demanded that RSI cease use of www.freebie.com. When RSI refused, Freebies Publishing brought a Uniform Dispute Resolution Policy action, which Freebies Publishing won. RSI appealed to the district court in the Eastern District of Virginia. The district court overturned the decision, forcing Freebies Publishing to then appeal to the Fourth Circuit.

The Fourth Circuit, however, ended up agreeing with the district court's decision. Like the district court, it also found FREEBIES to be generic, based on dictionary definitions, the applicant's own generic use of FREEBIES and the slew of third-party generic uses of the term submitted by RSI. It rejected Freebies Publishing's argument that a trademark registration, especially an incontestable one, for FREEBIES gave Freebies Publishing a per se defense against genericness. The court stated that a trademark registration only provides a presumption that a mark meets the requirement for registration; such a presumption can be overcome by sufficient evidence, which RSI had submitted. Similarly, it stated that incontestability is not a defense against genericness, only descriptiveness. The court also went on to reject Freebies Publishing's argument that Freebies Publishing had shown proof that FREEBIES was a commercially successful mark and that a number of consumers were actually confused by RSI's use of FREEBIE. It held that such evidence of secondary meaning had no relevance unless a mark was found not to be generic. Finally, the court rejected Freebies Publishing's argument that it was not using the mark in a generic sense, as well as Freebies Publishing's claim that genericness could be overcome by use of FREEBIES in stylized form.

After finding the term FREEBIES to be generic, the court went on to consider whether Freebies Publishing could successfully bring a claim under the ACPA. The court ruled that Freebies Publishing could not. Specifically, it held that the ACPA requires a claimant to show ownership in a "mark." The ACPA defines a "mark" as, among other things, a trademark that identifies and distinguishes the owner's goods from those of others. Because a generic term by definition cannot identify and distinguish a source, the court found that a generic term could not be a "mark" as defined by the ACPA. Accordingly, because Freebies Publishing could not satisfy the first requirement for bringing an ACPA claim, the court dismissed Freebies Publishing's claims and affirmed the lower court's decision.

"Written Descriptions" Under the Patent Act

The Federal Circuit (CAFC) has narrowly denied a request to rehear en banc the case of Univ. of Rochester v. G.D. Searle & Co., 375 F.3d 1303 (Fed. Cir. 2004). The case highlights a debate about the "written description" requirement of the Patent Act, which requires that a patent specification "contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art... to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention." 35 U.S.C. § 112.

Opponents argue that beyond enablement and best mode, the only purpose of the written description requirement is to police priority claims. Prior to 1997, the Federal Circuit and its predecessor court, the Court of Customs and Patent Appeals, held that an earlier-filed priority application must describe any invention claimed in a later application, otherwise the later claims were not entitled to the earlier priority date for lack of written description.

In 1997, the Federal Circuit held in Regents of the Univ. of Cal. V. Eli Lilly and Co., 119 F.3d 1559 (Fed. Cir. 1997), that because the University of California's patent only described human insulin cDNA in general terms, claims to the human insulin cDNA were invalid for lack of written description. For the first time, the court in Lilly did not test a later claim amendment against the specification for priority, but used a new disclosure requirement in place of the enablement test.

Subsequently, in Enzo Biochem, Inc. v. Gen-Probe Inc., 285 F.3d 1013 (Fed. Cir. 2002), the Federal Circuit held that a sequence listing deposited in a publicly accessible depository did not satisfy the written description requirement. Following a public outcry, the Federal Circuit vacated its decision and issued a new opinion, Enzo Biochem, Inc. v. Gen-Probe Inc., 323 F.3d 956 (Fed. Cir. 2002) (Enzo II). Enzo II, seemingly inconsistent with Lilly, held that a sequence deposit available to the public can indeed satisfy the written description requirement.

Last February, a panel of the Federal Circuit held invalid claims of a patent by the University of Rochester for lack of written description for failing to describe a selective COX-2 inhibitor. Univ. of Rochester v. G.D. Searle & Co, 358 F.3d 916 (Fed. Cir. 2004).  Although the patent disclosed sequences for COX-2, it did not identify any compounds that would bind selectively to, and thus inhibit, COX-2.

In voting to deny rehearing, Judge Lourie argues that the law is clear. "There is and always has been a separate written description requirement in the patent law." Judge Newman shares Judge Lourie's view but writes, "this question has percolated enough; it is ripe for en banc resolution." Judges Rader, Gajarsa and Linn analyzed cases decided under the written description provision, maintaining that a separate written description requirement other than for policing priority contravenes controlling precedent. The judges of the Federal Circuit appear both polarized and entrenched with respect to this issue. Ultimately, a case on written description will be heard en banc or on grant of certiorari by the Supreme Court. Until then, the outcome of patent appeals having written description questions may depend largely on the panel that is drawn.

Fourth Circuit Takes Netcom's "Passive Conduit" Standard a Step Further

In a slight expansion of the standard set in Religious Technology Center v. Netcom Online-Communications Services, Inc., 907 F. Supp. 1361 (N.D. Cal. 1995), the U.S. Court of Appeals for the Fourth Circuit in CoStar Group, Inc. v. Loopnet, Inc., 373 F.3d 544, held that an Internet service provider (ISP) is not directly liable for copyright infringement based on the unauthorized posting of photographs of its subscribers—even though the ISP tacitly approved the photographs prior to posting.

LoopNet is an ISP whose website allows subscribers, generally real estate brokers, to post listings of commercial real estate on the Internet. Early on, LoopNet instituted a screening policy to delete photographs showing obvious evidence that they were copyrighted by someone other than the subscriber (e.g., copyright notices). Because CoStar's photographs did not include copyright notices, LoopNet's subscribers ended up posting more than 300 of CoStar's copyrighted photographs. When CoStar notified LoopNet that infringing photographs had been posted, LoopNet removed them. Nonetheless, CoStar commenced an infringement action against LoopNet.

The Court of Appeals affirmed the dismissal of CoStar's infringement claim on summary judgment. Pointing to Netcom, the court upheld the standard that liability will not accrue to an ISP that functions merely as a "passive conduit." That is, there must not only be unauthorized copying occurring on the ISP's website, but there must also be some volitional conduct by the ISP that causes the infringement. "The ISP in this case is an analogue to the owner of a traditional copying machine... the owner of the copy machine is not considered a direct infringer. Similarly, an ISP who owns an electronic facility that responds automatically to users' input is not a direct infringer." According to the court, LoopNet was not engaged in any intervening conduct that actually caused the infringement.

In so concluding, the court rejected CoStar's arguments that LoopNet lost its status as a passive conduit by virtue of its "gatekeeping practices" with respect to the posting of photographs. "Although LoopNet engages in volitional conduct to block photographs... this conduct... does not amount to ‘copying,' nor does it add volition to LoopNet's involvement in storing the copy" because unblocked photographs are, in fact, only posted as an automatic electronic response to a subscriber's request. In fact, the court pointed out that LoopNet's activities tended to lessen the incidence of copyright infringement and "CoStar can hardly request LoopNet to prevent its users from infringing... and then subsequently seek to hold LoopNet liable as a direct infringer."

Consequently, the court affirmed the dismissal of CoStar's direct infringement claims. However, the court's decision would not insulate LoopNet or any other ISP from claims of contributory or vicarious copyright infringement with "a showing of additional elements such as knowledge coupled with inducement or supervision coupled with a financial interest in the illegal copying."

Clicking "Under Protest" Still Constitutes Acceptance of Clickwrap License

A recent district court ruling upheld the noncompete and other provisions of a clickwrap license despite a defendant's assertion that the license was clicked through "under protest." In I-Systems, Inc. v. Softwares, Inc., 2004 U.S. Dist.LEXIS 6001 (D. Minn. Mar. 29, 2004), a Minnesota district court denied summary judgment to a defendant that claimed that it was not bound by the terms of a clickwrap license for several reasons, including the fact that the defendant instructed its users to protest the terms of the license.

The software in question was developed by plaintiff I-Systems, Inc. (ISI) according to development agreements it signed with defendant Softwares, Inc. Certain breaches by ISI would result in Softwares receiving a perpetual license to use and sublicense the software without paying royalties to ISI. The development agreements specified that they represented the "entire agreement between the parties with respect to the subject matter" and could only be modified in writing signed by both parties.

The software and later upgrades were covered with ISI's license provisions that provided that the user could not, among other things, "reverse engineer, modify or create derivative works based on the software," disclose confidential information related to the software or conduct any competitive business activity. Two of these licenses were "shrinkwrap" licenses that were installed automatically when the user installed the software, and a third, a "clickwrap" license, required the user to click acceptance of its terms before using the software. As users (as well as distributors) of the product and its upgrades, Softwares installed the software pursuant to these license agreements. However, Softwares's dissatisfaction with the product caused it to instruct its end-users of the product to accept the license in order to install the software but to memorialize written objections to the license. When ISI discovered that Softwares had developed a competing product, it sued, based on breach of the noncompete and non-reverse engineering provisions of the software licenses and copyright infringement, among other theories.

At issue before the court was whether or not Softwares was bound by the terms of the license agreements. In its summary judgment motion to defeat the breach of contract claim, Softwares offered several reasons why it was not in breach. First, it argued that the terms of the licenses did not apply because the development agreements represented the "entire agreement" between the parties and had to be modified through a writing signed by both parties, a threshold not met by the license agreements. Second, Softwares said that it was not bound by the license" agreements because it was not aware of the shrinkwrap agreements and had only "clicked through”the clickwrap agreement under protest.

The court disagreed with the defendant on all counts. It said that even if the license agreements did not apply, the plaintiff's alleged breach of the development agreements did not give the defendant free rein to do "anything" it wanted with the plaintiff's software. Instead, the remedy was circumscribed by the original development agreement, which provided for a perpetual license and nothing more. The court also held that a jury might find that the license agreements did apply on a variety of theories, including that they were new and separate agreements to which the parties agreed, or because the licenses were contemplated to bridge the distinct propriety and distributorship rights contemplated in the distribution agreement.

In this vein, the court held that a jury could find that Softwares accepted and was bound by the terms of the clickwrap license when it clicked acceptance to install the software upgrade, regardless of its protestations. Furthermore, because the clickwrap and shrinkwrap licenses were referenced in invoices provided to Softwares that directed it to "read the software license agreement carefully," the court was not persuaded by Software's argument that it was not aware of the shrinkwrap license provisions.

Several points emerge from the court's decision. First, merely protesting the terms of a license, while still clicking through, will not absolve a user from its terms. Second, "complete agreement" provisions included in licenses will not always be dispositive, particularly when different rights and subject matter are being contemplated between the parties. And finally, alerting users to electronic licenses through other medium, including invoices, can help to overcome any assertion of that the user is unaware of them.


Fenwick & West's Intellectual Property Group offers comprehensive, integrated advice regarding all aspects of the protection and exploitation of intellectual property. From providing legal defense in precedent-setting user interface copyright lawsuits and prosecuting software patents to crafting user distribution arrangements on behalf of high-technology companies and implementing penetrating intellectual property audits, our attorneys' technical skills enable the Firm to render sophisticated legal advice.

The contents of this publication are not intended and cannot be considered as legal advice or opinion.

© 2011 Fenwick & West LLP. All Rights Reserved.

We appreciate your feedback!

If you have questions, comments, or suggestions for the editors of the IPB, you can e-mail them to IPB@fenwick.com.

For subscription requests and address changes, please e-mail IPB@fenwick.com.