Determining the Value of a Patent Portfolio
BY AMIR H. RAUBVOGEL
Business decisions often depend upon accurately determining an asset's value. Intellectual property is a particularly difficult type of asset to appraise because of its intangible nature and the multitude of complex factors involved in accurate valuation. Yet intellectual property, particularly a patent portfolio, is often among a company's most important assets. Businesses need to accurately assess patent quality and value to answer questions such as:
- Does existing intellectual property protection cover the company's products, services and research areas sufficiently?
- Are the company's patents strong enough to be asserted against a competitor?
- Are the patents a potential source of licensing revenue?
- Are continuation applications, reissue applications or any additional patent applications warranted?
- How much is the patent portfolio worth in the context of the value of the company as a whole?
- How good is the company's patent counsel?
A number of situations exist in which a business may wish to assess the quality of a patent owned by another company. The risk of potentially infringing another company's patent depends upon that patent's applicability, the strength and breadth of its coverage and the viability of a challenge to its validity. It also is important to evaluate the quality of another company's patents to determine:
- How crowded is the technological field?
- Should a competitor's patents affect the company's decision to enter a product or development area?
- Should the company offer to take a license on a patent?
- Should the company attempt to acquire or merge with a competitor?
- Should the company strengthen its own portfolio for defensive purposes or for negotiating leverage given a competitor's patents?
The following is a step-by-step checklist for patent-quality assessment:
First, it is important to consider the overall subject matter addressed by the patent and whether it is related to one's own market or technological area. Of course, the patent may have some general intrinsic value, and certain companies are interested in acquiring patents regardless of the technological area. Generally, however, it is impractical to evaluate every patent that comes along, regardless of relevance to the company's business.
Claims define the limits of the property right conferred by a patent. Therefore, once a company determines that the patent's subject matter is sufficiently relevant to warrant further analysis, the next step is to study the claims. A good set of claims describes the invention at several levels of abstraction, each providing a different breadth of coverage. Broader claims cover the overall concept. Other claims, such as dependent claims, introduce limitations relating to particular implementations of the invention.
While such claims provide narrower coverage, the patent holder may be more successful defending narrower claims against invalidity arguments based on prior art. An alleged infringer usually challenges patent validity whenever a patent holder asserts the patent against it. Thus, it is important for a patent holder to have a robust claim tree, including claims at different levels of abstraction and breadth as fallback positions.
Another issue is whether the claims include all of the invention's parallel and alternative forms. For example, software patents should typically include a combination of method claims (a sequence of steps such as a process, application of an algorithm or business method), system claims (an apparatus, functional architecture or machine) and computer program product claims (software encoded on a medium, such as a disk).
The company also should examine whether the claims specify steps to be performed by more than one party, entity, person or machine. For patents involving multiple entities (such as client/server architectures), it is important to include separate claims for the steps performed by each entity. For example, one set of claims may cover steps performed by the client and another set may cover steps performed by the server. A competitor then could be liable for direct infringement by operating or selling just the server or just the client.
In general, any invention involving communication or interaction between nodes, entities, components or users should include separate claims for as many of these entities as is practicable. This avoids any requirement that the infringer sell or operate all entities.
One also should note whether the claims include "means plus function" language, such as "means for encrypting the received signal." Such claims attempt to define an element of a system according to the function that it performs, rather than its structure. However, the scope of such claims is narrower than it would appear, due to statutory language specifically limiting such claims to the structure linked to the function in the specification, plus equivalents. Although there is nothing wrong with including "means plus function" claims, it is important to also include other claim formats.
A patent's specification can have a significant effect on its scope and validity. Courts will read the claims in light of the specification and use the specification for guidance about the terminology used in the claims and what the applicant intended the claims to protect. If the specification contains limiting language, a court may use that language to narrow the claims, even though the claims do not recite those limitations.
One should review the specification for completeness. A patent must describe the "best mode" of practicing the invention known to the applicant at the time of filing. It must describe the invention at a level of detail sufficient to allow an individual having ordinary skill in the relevant art to re-create the invention without undue experimentation. If doubt exists as to whether the description is sufficient, then the patent's value is questionable; a defective specification can lead to invalidation of the claims.
The prosecution history of an issued patent provides information as to the claims that the patent holder originally filed, the number of office actions, responses and interviews that took place and the rejections, arguments and amendments that were made. This information can have a dramatic impact on the value of a patent and can reveal vulnerabilities that are not always visible from the face of the patent itself.
Narrowing amendments made during patent prosecution can disqualify a claim (or a claim element) from having its scope expanded by the doctrine of equivalents, which otherwise can extend a claim's coverage beyond its literal language.
Additionally, an infringement defendant can use the prosecution record as a basis for construing the claims narrowly, arguing that statements in the record indicate what the applicant intended to patent (and the examiner intended to allow) and that the court should thus read the claims in light of such statements. In general, excessive amounts of argument, characterizations and assertions during prosecution increase the risk that a patent's scope may be thrown into doubt at some point.
Priority Date and Prior Art
Patents with earlier priority dates are more valuable than later-filed patents because the earlier filing date means that less prior art is available to defeat the patent. In considering the priority date, one should investigate what prior art existed at the time, as well as what products may have been launched before the application was filed. In the United States, an inventor must file a patent application within one year of any public disclosure of the invention. Therefore, it is important to consider what disclosures may have been made more than one year before the application's filing date (or priority date).
If relevant prior art does exist, one should investigate whether the applicant properly cited the art to the examiner in the course of the patent's prosecution. Art that was considered by the examiner may make the patent stronger, because there is a presumption that the patent is valid over art that was considered, although the presumption is rebuttable. If the applicant knew of relevant art but failed to cite it, then the patent may be invalidated due to the applicant's failure to fulfill the duty of candor, which requires an applicant to disclose any relevant art of which he or she is aware in the course of prosecution.
Patents that claim priority from other patents or applications are subject to additional avenues of attack. If a patent claims priority from an earlier provisional application, and the provisional application fails to fully support the claims, then the priority claim can be defeated. Similarly, a patent that claims priority as a continuation or continuation-in-part can lose the benefit of the earlier priority date if the specification of the parent application does not adequately support its claims. Therefore, for any patent that claims priority from another patent or application, one should review the specification of any parent application.
Evaluation of a patent is a multifaceted exercise that often requires the skill and expertise of an experienced patent practitioner. However, these basic principles are the cornerstones of such analysis and a good starting point in any attempt to ascertain the overall value of a patent.
The Potential Perils of Provisional Patent Applications
BY LAURA A. MAJERUS
A 1995 change in the patent law allows an applicant to file a provisional patent application for an invention instead of a regular "utility" patent application. During the Internet boom of the late 1990s, more companies began filing provisional applications, thinking that the significant difference in preparation cost and filing fees made a provisional application "the poor man's patent application."
Although provisional applications are useful under some circumstances, they can result in certain problems if not managed properly. Many applicants are only now beginning to understand the downside of having filed a provisional application instead of a regular one. Some problems are due to faulty or hasty legal advice from patent counsel during the boom years. Some problems are due to good, but poorly understood advice of counsel and some are due to the fact that economic decisions caused the company to wait until the last minute to follow up on a provisional application.
Applicants may file a provisional patent application, which lasts for one year, for a nominal filing fee. By the end of the year, the applicant must complete and file the "real" utility patent application and claim the benefit of the filing date of the provisional.
Nothing is wrong with the filing of provisional applications per se. They allow an applicant to file quickly and cheaply and to take advantage of written materials, such as functional specifications, already in the applicant's possession. They also do not count toward the 20-year term of the patent and are, therefore, useful if the invention will have a long market life. Additionally, they allow a company to mark the corresponding product as "patent pending."
Some pitfalls, however, are associated with the filing of provisional applications. Some companies are only now coming to grips with these issues. The U.S. Patent and Trademark Office holds a provisional application to the same standard of disclosure as a utility application. That is, the provisional application must show a person of ordinary skill in the art how to make and use the invention with a minimum of experimentation.
Furthermore, a provisional application must contain a written description of the invention and disclose the "best mode" of making the invention. Thus, for a utility application for a particular invention to claim the benefit of a provisional application, the provisional application must meet these tests for the specific invention.
Why does it matter whether the provisional patent application is adequate or inadequate? It matters because the U.S. patent law is set up so that an inventor only has a certain amount of time to file a patent application after his or her invention is used, offered for sale or made public in certain ways.
In general, an inventor can take up to one year from the earliest of these dates to file a patent application. If he or she files an application more than one year after any of these occurrences, then his or her own actions will bar the application. This is called a "statutory bar," because it is created by statute, specifically by 35 U.S.C. Section 102(b).
Problems can arise when an inventor files a provisional application after an activity that creates a statutory bar. If the inventor then waits a full year more to file a utility application based on the provisional application, then he or she has filed his utility application more than one year after the barring activity.
This will not be a problem if the inventor is able to claim the benefit of the provisional filing date. On the other hand, if a court later finds the provisional to be insufficient, then the inventor cannot claim the benefit of the provisional application and the filing date of his or her regular application is past the bar date, creating a bar to patenting the invention.
It comes as a surprise to many people that the Patent Office does not check if the provisional application is good enough to rely on. The office merely keeps a provisional application on file for up to a year until the corresponding utility application is received. The Patent Office then proceeds to examine the utility application.
The Patent Office has made a decision that it will not routinely investigate the sufficiency of benefit claims based on provisional applications. Thus, the first place that the sufficiency of a benefit claim usually is tested is in court.
The decision of the U.S. Circuit Court of Appeals for the Federal Circuit in New Railhead Mfg. L.L.C. v. Vermeer Mfg. Co., 298 F.3d 1290 (Fed. Cir. 2002), provides an illustrative example of what can happen if a company relies on a false sense of security because a provisional application is on file.
In New Railhead, the plaintiff filed a provisional application for a drill-bit invention less than one year after the invention was sold. More than one year after the sale, the company filed a utility application that claimed the benefit of the provisional application.
When the patent was litigated, the accused infringer raised the defense that the claims of the utility application were not supported by the provisional application, and thus that the utility application was not entitled to the benefit of the provisional application.
The claim recited a drill bit with "the unitary bit body being angled with respect to the sonde housing." The court found that the provisional application did not discuss the drill bit being angled with respect to the housing, and that, therefore, the utility application could not rely on the filing date of the provisional. The court found the patent invalid because it had been filed more than one year after the sale.
New Railhead might be just the tip of the iceberg. Many patent applications based on badly written and badly conceived provisional applications are likely wending their way through the Patent Office. Because it takes four or five years for a patent to issue in certain technical fields, such as software, some patents claiming the benefit of provisional applications filed in 1998 or 1999 are just now starting to issue. It may be several more years before parties assert these patents in litigation. Thus, attorneys will probably be seeing more fact patterns like New Railhead in the future.
Many provisional applications filed during the boom years contain several common problems. First, they often lack sufficient disclosure to support the later-claimed invention. During the time pressure of the boom, some provisional applications were filed that are only one or two pages long, often little more than marketing materials touting the invention's advantages and cost savings. Sometimes, these marketing materials are deficient in the details required to teach a person of ordinary skill in the art to make and use the invention.
Second, many of the follow-on utility applications were filed more than one year past a date of public disclosure or offer for sale. The existence of a provisional application gives many companies a false sense of security about their patent rights. Some companies therefore did not file their follow-on utility application within one year of any known bar dates, relying on the fact that they already had a provisional application on file.
Start-up companies were particularly subject to this delay, because many were strapped for cash and wished to push off preparation of a utility application for as long as possible. Because the Patent Office does not check the validity of benefit claims, problems with support of the patent application by the provisional generally will not come to light during prosecution of the utility patent application.
In addition to the problem of invalid patents in New Railhead, inadequate provisional applications raise the possibility of increased malpractice litigation. In some cases, these problems occurred because counsel did not advise their clients adequately of the downside of filing a provisional and then waiting a full year to file the follow-on utility application. Even if they did so advise clients, many lawyers did not adequately protect themselves by providing the advice in writing. In extreme cases, some lawyers may have even encouraged clients to wait the full year in order to lighten their workloads.
When such patents issue, are litigated and are found to claim priority to inadequate provisional applications, there will be an upturn in clients' suing patent attorneys either for inadequately advising them or for inadequately reviewing the contents of a provisional application at the time of filing.
Whether one thinks that patents based on bad provisional applications are good news or bad news depends on whether one is an owner of such patents or an accused patent infringer. Patent owners may struggle over the issues in New Railhead during the coming years, trying to rebut the case of invalidity presented against their patents in court. Conversely, accused infringers in patent litigation will have another popular defense to assert: The provisional upon which the patent application is based is inadequate.
Copyright Protection Denied for Public Data in Database — Misuse Musings
Can copyright law be used to block access to data that is not copyrightable? This issue was addressed in Assessment Technologies LLC v. WIREdata, Inc.,350 F.3d 640 (7th Cir. 2003).
Assessment Technologies (AT) created a database product called Market Drive, which was used by municipalities in Wisconsin to compile real estate data for property tax purposes. Individual assessors within a municipality would visit the properties therein and input the tax-relevant information into a computer running Market Drive. As new data was collected, Market Drive would compile it into a master database for use by tax officials.
This same data was of particular interest to WIREdata, which catered to real estate brokers in Wisconsin. Under Wisconsin's "open records" law, the municipalities were required to furnish such data upon request to any person, subject only to a nominal fee (i.e., copying the data to a disk). For years, data providers like WIREdata had successfully obtained real estate data in this manner.
With the advent of the Market Drive program, however, three municipalities refused WIREdata's data requests based on an exception in the "open records" law for copyrighted materials. These municipalities were afraid that furnishing the requested data would violate AT's copyright in the Market Drive compilation. AT intervened in federal court to enjoin WIREdata's efforts to obtain the data.
On appeal, the United States Court of Appeals for the Seventh Circuit confirmed that AT did indeed have a protectable interest in the Market Drive compilation because the structure of the database was not so obvious or inevitable as to lack the minimal originality required. However, AT's interest did not extend to the raw data contained therein. As the court put it, AT "created only an empty database, a bin that the tax assessors filled with data," and had no ownership or other legal interest in the extraction of noncopyrighted material. Since the extraction did not involve copying Market Drive, the court dismissed the copyright claim.
The court also took the opportunity to voice its "profound skepticism" at AT's suggestion that the license agreements for Market Drive would bar the sharing of data contained in the tax-assessment databases. Judge Posner raised the specter of copyright misuse in the event that AT was "to try by contract or otherwise to prevent the municipalities from revealing their own data." The misuse doctrine was developed to prevent copyright owners from attempting to control areas outside the "copyright monopoly." Although the Seventh Circuit has not yet extended the misuse doctrine to cases that do not also involve an antitrust violation, Judge Posner signaled a willingness to move in that direction. The United States Courts of Appeal in the Third, Fourth, Fifth and Ninth Circuits have already extended the misuse doctrine in this manner.
Correction of Patent Claims — The Opportune Time to Do So
A recent ruling by the United States Court of Appeals for the Federal Circuit emphasizes the importance of proofreading the claims of an issued patent for typographical errors before suing anyone for infringement. In Novo Industries, L.P. v. Micro Molds Corp. (Fed. Cir. 03-1230,-1249)(December 5, 2003) the Federal Circuit held that although a district court can correct certain "obvious errors" in a patent in the process of construing the patent's claims, this ability is sharply limited. The result for plaintiff Novo Industries was invalidation of the patent claim at issue.
Novo had obtained a patent for a carrier assembly for vertical blinds, which was designed to realign misaligned slats. In the process of prosecuting the patent, the patentee amended one of the patent's claims to recite, in part, "stop means formed on a rotatable with said support finger." The amendment added the words "a rotatable with." Novo sued Micro Molds for infringement without seeking a certificate of correction for the claim from the United States Patent and Trademark Office (PTO).
During claim construction, Novo argued that the mystifying phrase was the result of an "obvious typographical error" and asserted two different possible corrections – deleting either the words "a rotatable with," or the words "with said" from the offending phrase. The defendant, Micro Molds, argued that the added word "rotatable" was a noun, not an adjective and since the "rotatable" was not defined in the patent, the claim must be invalid as indefinite. The district court followed a fourth course, determining that the "a" in "a rotatable with" was simply missing a couple of letters. The district court thus substituted the word "and" for "a" in its interpretation of the claim. Under the district court's construction, a jury found that Micro Molds infringed the patent.
On appeal, the Federal Circuit addressed the question of whether a district court can correct an error in a patent by interpretation where the PTO has not issued a certificate of correction. The court held that a district court may do so, but only if "(1) the correction is not subject to reasonable debate based on consideration of the claim language and the specification and (2) the prosecution history does not suggest a different interpretation of the claims." The only alternative is to seek a certificate of correction from the PTO pursuant to 35 U.S.C. Sections 254 and 255. But by statute, the patentee must seek such a correction before suing for infringement.
The Federal Circuit held that Novo's patent did not meet the criteria for correction by the district court and held that the asserted claim was therefore invalid as indefinite in its present form.
First, the Federal Circuit pointed to the multiple proposed corrections suggested by Novo, Micro Molds and the district court in determining that "the nature of the error is not apparent from the face of the patent" and the correction would thus be subject to reasonable debate. In addition, the Federal Circuit determined that the patent's prosecution history weighed against the court's construction, stating that the addition of the words "a rotatable with" clearly suggested that the additional language was "not merely cosmetic, but had substantive significance." "Under these circumstances," the Federal Circuit continued, "in order to make sense out of the patent, the district court was required to guess as to what was intended. That is beyond its authority."
The moral of the story is simple: The process of correcting an error in a patent claim in the district court is risky, presenting the possibility that the claim may be held invalid. Unless the error is clear from the face of the patent and only one correction is possible, seeking a certificate of correction in the PTO is the safer course.
Search Engine Keywords — Trademark Confusion or Not?
In this appeal, the plaintiff, Playboy Enterprises, Inc., argued that the lower court had improperly granted the defendants, Netscape Communications Corporation and Excite, Inc., summary judgment when it ruled that the use of Playboy's trademarks as keywords in Netscape and Excite's search engines did not result in initial interest confusion or trademark dilution. Playboy Enterprises, Inc. v. Netscape Communications Corp., No. 00-56648 (9th Cir. January 14, 2004).Playboy sued Netscape and Excite because the two search engine companies had linked two of Playboy's trademarks, PLAYBOY and PLAYMATE, so that input of these terms would display banner advertisements from competing adult entertainment companies. It argued that use of its trademarks in this fashion allowed the defendants to improperly trade upon Playboy's goodwill in its trademarks. The United States Court of Appeals for the Ninth Circuit agreed and reversed the lower court's decision on both the initial interest confusion and dilution claims.
On the initial interest confusion claim, the court found that genuine issues of material fact still existed as to the strength of Playboy's marks and as to Playboy's survey showing actual consumer confusion, and so it determined that the lower court's finding of summary judgment was inappropriate. On the rest of the factors for likelihood of confusion, it found that the factors either favored Playboy or were neutral to Playboy. Based on this analysis, it concluded that there was a strong likelihood that users would be at least initially confused, if not continuously confused, by the appearance of advertisements from competing companies when the users input Playboy's trademarks. The court also went on to reject the defendants' arguments that the use was fair use, nominative use or functional use. The court stated that the use could not be fair use because the use was confusing and confusing use was not fair use. Similarly, it concluded that the use was not nominative, because defendants still had 400 other nontrademarked terms to link to adult entertainment advertisements. Finally, it rejected defendants' claim of functional use as irrelevant.
In making these conclusions, the court specifically limited itself to the case where keywords were linked to confusingly labeled or unlabeled advertisements. It stated that it was not opining on a case where the keywords were used to bring up Playboy materials or materials sponsored by Playboy or on a case where the linked advertisements clearly identified the source of the advertisements. In the concurrence following the opinion, Judge Berzon more clearly repeated this distinction. She indicated that while she agreed with the result in this particular case, she was concerned that the initial interest doctrine outlined in Brookfield Communications v. West Coast Entertainment Corp., 174 F. 3d 1036 (9th Cir. 1999) not be applied too broadly, because it could be construed, if taken broadly, to cover even those uses where no user was ever confused, even initially.
The Ninth Circuit remanded on the issue of dilution as well. Specifically, it found that there were still issues of material fact as to the fame of Playboy's marks and as to the issue of likelihood of dilution, given the new clarification on the standard of dilution recently handed down by the Supreme Court in Moseley v. V Secret Catalogue, Inc., 537 U.S. 418 (2003).
Wyeth v. Natural Biologics: A Cautionary Tale in Trade Secret Misappropriation
Last fall, a federal court in Minnesota issued an unusually harsh ruling against a pharmaceutical preparations company that it found guilty of trade secret misappropriation, permanently preventing the company from researching, developing, producing, making or selling natural conjugated estrogens. Wyeth v. Natural Biologics, Inc., No. 98-2469 (D.Minn. October 2, 2003).
Plaintiff Wyeth Corporation makes and sells conjugated estrogens as the blockbuster drug Premarin for the treatment of severe symptoms associated with menopause and osteoporosis prevention. The trade secret at issue in the case was the process by which Wyeth makes the estrogens from pregnant mare urine. Although Premarin has been sold in the US for more than 60 years, no generic versions of it have been approved because the active ingredients are unknown, in part because they derive from a natural source. Defendant Natural Biologics sought to change this by applying for FDA approval of its version of Premarin, also made from pregnant mare urine. Skeptical that Natural Biologics had figured out how the estrogens were derived on their own, Wyeth brought a suit for trade secret misappropriation.
Tipped off by an unidentified phone number found by a paralegal on a document, plaintiff Wyeth uncovered that the defendant had had extensive contact with a scientist formerly employed by the plaintiff. Worse yet, defendants had attempted to conceal this relationship by redacting records, destroying documents and even lying about it in deposition and in court.
These transgressions led the judge to issue a ruling broader in scope and time than typical trade secret rulings. Stating that the company "cannot be trusted to obey an Order that permits [the defendant] to exercise any discretion," the court barred the defendant from ever engaging in any process or research relating not just to the trade secret but also to any removal of estrogens from urine. Because the injunction is permanent, rather than temporary, the ruling is predicted to effectively shut down defendant Natural Biologics.
The ruling contains lessons for companies and lawyers alike. First, do not steal trade secrets, especially when a $2 billion per year product is involved. Second, do not lie in an attempt to cover up your mistakes — in this case, it was the actions of the defendant after the trial started that arguably caused the greatest problems. And finally, on a less dire note, careful scrutiny of the often-voluminous production of documents in discovery can and often does yield the key to the entire case.
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.