Leading-Edge Law Group, PLC

Last month, Google won an important battle in the trademark-infringement suit filed against it by auto insurer GEICO over Google’s Adwords advertising program.

Yet, it will be awhile before we know whether Google has saved the Adwords program, which is a crucial component of its revenues.

First, a little background is necessary. In its Adwords program, Google sells advertisements linked to queries run through its search engine. A prospective advertiser selects the search terms that will trigger the ad. Someone who enters a query term still gets the usual Google search results, but the paid advertisement appears alongside the results under the heading “Sponsored Links.”

Last February, I noted that eventually the courts would test whether keying an Adword ad off of someone’s trademark would constitute trademark infringement. Sure enough, last summer GEICO sued Google, claiming that the selling of Adwords ads triggered by searching the GEICO name was trademark infringement. Because this case concerned a well-known trademark, this case was poised to produce a significant decision.

It did just that. In December, U.S. District Judge Leonie M. Brinkema ruled from the bench in Northern Virginia that the Adwords program does not necessarily infringe the trademark rights of entities whose trademarks are used as search terms. For example, it was not trademark infringement for someone to purchase an ad that was triggered by searching in Google for the term “GEICO” absent some special circumstances I discuss later.

The court got it right. A trademark owner does not have a monopoly that prevents any use of the trademarked word or phrase by anyone else, including competitors.

The central question to answer in considering whether trademark infringement has occurred is whether the use of someone’s trademark by a competitor will tend to confuse consumers. A company can use a competitor’s trademark in comparative advertising as long the advertising does not suggest falsely a relationship with the trademark owner.

This case adds to a growing but conflicted body of cases concerning the use of trademarks on the Internet. Broadly speaking, many trademark owners have been arguing that, effectively, no one should be able to mention or use their trademarks on the Web except them or those to whom they give express permission. For example, this issue has cropped up with pop-up advertising keyed to trademarks, and in cybersquatting cases brought against people run Web sites criticizing particular companies.

In some of the cases, courts have been distracted by the opportunistic behavior of the defendants and, consequently, used the club of trademark infringement to attempt to stop conduct even where no likelihood of confusion has existed. If affirmed and followed, the result in GEICO’s case against Google may contribute to unifying the law and making it right.

Yet, the case contains an interesting and not-yet-resolved issue that is crucial to Google. The news reports of Judge Brinkema’s decision generally overlooked it.

Google maintains an absolute policy that, in an Adwords ad, the trademark used as a search term may not be mentioned in the Adwords ad if the owner of the trademark objects.

Thus, for example, Google will sell you an Adwords ad that keys off of the word “GEICO” but purportedly will not permit you to mention GEICO in the body of the Adwords ad because GEICO objects. Google apparently missed some violations of this policy. GEICO contends that that usage of its trademark in the body of the Adwords ad was a trademark infringement for which Google should be liable.

Google’s policy is more generous to trademark owners than it needs to be. One can mention a competitor’s trademark in the body of an advertisement as long as the mention does not suggest a relationship. Perhaps Google does not want to review individual ads because it operates its Adwords program on an automated basis and it would be uneconomical to have trained individuals review each Adwords ad before posting.

Nevertheless, the court orally held that posting Adwords ads that mention the distinctive trademark of a competitor could be a trademark infringement for which Google might be liable. That ruling strikes me as too broad – in some cases the conduct won’t confuse consumers so it isn’t always trademark infringement. Hopefully, this aspect of the court’s ruling will be clarified when the court issues a written opinion to explain the ruling it made from the bench.

We’ll see whether that written ruling leaves the Adwords program in a financially viable condition. If the ruling imposes nearly strict liability on Google for trademarks appearing in Adwords ads, and if Google cannot screen out such trademark uses with a low-cost procedure, then its Adwords profits could be at risk.

Consumers may not hear much in the news about the Digital Millennium Copyright Act. Yet, a couple of recent court cases interpreting the DMCA are having a big impact on what product makers can charge for replacement parts, such as auto parts, and replenishable supplies, such as toner for printers. But will these decisions stand?

The DMCA became law in 1998. Its perceived purpose was to prevent the use of hacking or electronic “burglar’s tools” to break technological protections deployed to guard against piracy of electronic goods, such as music CDs or movie DVDs. The music and movie industries deployed “digital rights management” technology so they could sell electronic copies of their wares without havingsuch legal copies pirated into zillions of illegal copies.

Yet, DRM pits spy versus spy. For each advance in DRM technology, some hacker tries to (and often does) devise a way to disarm the protection and make the protected content freely copyable.

The DMCA made breaking DRM protections illegal in most cases. The relevant provision of the DMCA prohibits circumvention of any technological measure that controls access to a copyrighted work. It also prohibits trafficking in devices designed primarily for the purpose of such circumvention.

Some clever product manufacturers see the DMCA as an opportunity to electronically lock out competition for replacement parts and replenishable supplies, although the DMCA’s authors probably didn’t intend that result.

For example, printer maker Lexmark used a computer lock-out code to attempt to prevent use of cheaper replacement toner cartridges from other suppliers. Also, garage door opener maker Chamberlain Group used an embedded computer code on its garage door openers that prevented use of universal remote controls sold by others. One could ascribe malevolent intent in these cases – preventing price competition in replacement parts.

To be fair, Lexmark and Chamberlain Group both proclaimed benevolent reasons for their actions. Lexmark runs a “prebate” program under which one could purchase toner cartridges at a discount in exchange for agreeing to buy only Lexmark replacement cartridges. Lexmark tried to use technology to prevent those who bought cartridges at the lower “prebate” price from later breaking the deal by buying cheaper replacement cartridges elsewhere.

Chamberlain contended that its “rolling code” technology keeps would-be burglars from stealing the fixed electronic codes ordinarily used for garage door openers.

Overall, some legal commentators have worried that product makers would begin using lock-out codes to prevent after-market competition in many areas and then use the DMCA to legally stymie anyone who tries to bypass the lock-out codes. For example, one can imagine a car manufacturer deploying microchips throughout a car so that a replacement part would work only if the car’s computer system authenticated it.

Suppliers of replacement parts who managed to defeat the lock-out codes challenged Lexmark and the Chamberlain Group in the marketplace. In Lexmark’s case, Static Control Systems produced a cheaper replacement toner cartridge that mimicked Lexmark’s authentication code. In Chamberlain’s case, Skylink Technologies produced a universal garage door remote control that could unlock Chamberlain’s security measure.

Lexmark sued Static Control Systems for violating the DMCA by circumventing its security code system, and Chamberlain did the same to Skylink. Lexmark and Chamberlain both won at the trial court level but lost on appeal.

Essentially, the appellate courts said that the DMCA doesn’t protect a technological access control unless that control prevents access to or copying of a copyrighted work. For example, because the purpose of the Lexmark lock-out code was only to keep a prebate customer from using another manufacturer’s toner cartridge (as opposed to a purpose of preventing illegal copying of something copyrighted, such as a song), the DMCA didn’t forbid breaking the Lexmark lock-out code.

These cases are big wins for consumers and replacement parts suppliers. But the issue isn’t settled yet.

The three appellate judges in the Lexmark case couldn’t agree on important legal points, and the Chamberlain opinion used yet another legal rationale. The differing interpretations have differing consequences for what product makers can do. Also, the plain language of the DMCA may not support some of the reasoning of these appellate courts. The Supreme Court could reverse these decisions.

And even if these decisions stand, they might not prevent all legal hanky-panky. One could envision a clever product maker employing a technological control that prevents some electronic piracy but also just happens to lock up the market for replacement parts. “Sorry, but our car bumpers contain our copyrighted software that ties into our accident avoidance system, and we must protect our intellectual property, so you can buy replacement bumpers only from us.”

On top of this, other laws, such as antitrust law, may forbid trying to lock up the replacement parts market. The relationship between the DMCA and these other laws must be sorted out.

Overall, early attempts to use the DMCA to control the marketplace for replacement parts have failed, but the law has miles to go before it’s settled.

The House passed an anti-spyware bill in 2004 but the Senate never took it up. Rapid growth of the spyware problem and a rising patchwork of state laws has put pressure on Congress to not punt again.

Optimistically, Congress won’t get bogged down in the “what exactly is spyware?” navel-gazing exercise. If Congress adopts as the cornerstone of anti-spyware legislation a requirement of obtaining clear notice and consent before loading software on a user’s computer, the precise answer to this question won’t matter much.

Admittedly, wordsmithing a crisp definition of spyware is tough. Consumer groups and technological interests each pull on the definition to achieve their policy goals.

To me, spyware is software loaded on your computer without your knowing consent that monitors some aspect of your computing activity and provides (or makes available) that information to someone else. For example, software that logs your keystrokes and reports them to someone without your knowledge is spyware.

Consumer interests want to expand the definition of regulated spyware to cover adware, too. Adware usually takes the form of annoying pop-ups that arise from a program embedded in your computer. Adware often comes bundled with a free download of another program, such as P2P file-copying software.

On the other hand, various technological powers want to keep adware unregulated. They also want to exempt certain practices often used by mainstream online advertisers, such as using cookies and Web bugs in Web pages and HTML-based e-mails. (A Web bug can report back on when an HTML page is viewed, what page was viewed and the IP address of the viewer.)

These technology interests have persuaded the sponsor of anti-spyware legislation pending in the House of Representatives to exclude these technologies from coverage.

In addition, some fret that anti-spyware legislation could outlaw or restrict benevolent monitoring technologies, such as parental-control software for Web surfing or the category button in the popular Google toolbar.

These details in coverage matter to the extent the new law will make spyware illegal. But, to paraphrase Voltaire, Congress should not let its striving for definitional perfection prevent it from doing good.

The cornerstone of anti-spywarelegislation should be a requirement that, before software is loaded on a user’s computer, the user must be given in-your-face, understandable notice of what software is being presented for loading and what it does. I’m not talking about weasel words buried in a 10,000-word end-user license agreement that you view through a tiny scroll box.

Instead, I refer to splash screens that say something like “This Web site wishes to load on your computer software that pops up ads on your computer, even when you are not on this Web site, that match your tastes based upon your computer habits – Do you agree to download it?” In order for the download to occur, the Web surfer would have to click “yes.”

Requiring clear disclosure and affirmative consent harms no one. If the proposed software is benevolent (such as parental-control software), the user will consent. Also, in a notice-and-consent regime, a broad definition of “spyware” is fine; there would be no downside to covering adware, too.

Regardless, those who strive to hold back federal anti-spyware legislation until all of their demands are met underestimate the pressure for such legislation.

Microsoft and Dell both report that spyware-caused problems are now the leading cause of calls to their consumer tech-support operations. Many are frustrated.

States are beginning to pass their own anti-spyware laws. Virginia Attorney General, Judith Jagdmann and her predecessor, Jerry Kilgore, have shown strong interest in enacting state anti-spyware legislation. California and Utah already have enacted anti-spyware legislation. At least 19 other states have anti-spyware bills pending.

What probably will happen here will be the same thing that happened with anti-spam legislation.

For a while, electronic-marketing forces kept anti-spam legislation flummoxed on Capitol Hill. Then California passed tough anti-spam legislation, and other states began to pile on. These tough laws caused the electronic-marketing forces to rush to Capitol Hill to obtain federal anti-spam legislation that was less restrictive and that would pre-empt the more onerous state laws.

Similarly, with spyware, the threat of harsh and patchwork state laws probably will persuade the foot-draggers to strike the best federal deal presently available.

There’s reason to hope that a federal anti-spyware law could make a real dent in spyware. Of course, it will not eliminate rogue operators, especially those overseas.

Yet, to contract spyware, you have to take a voluntary first step – download software, visit a Web site, load a CD. If a Web site or software supplier must provide upfront and clear notice of what it wishes to put on your hard drive, certainly that will reduce the flow of spyware from online folks who seek to gain your trust.

Piece by piece, Virginia is earning the status of having the most cyber-advanced legal system in the United States.

The 2005 General Assembly passed, and the governor signed, an amendment to Virginia’s computer-crimes law that expressly outlaws phishing (more on that below). As far as I can tell, Virginia is the first state to enact a criminal law directly aimed at phishing.

This isn’t the first instance of Virginia being first in the area of cyberlaw. Virginia essentially tied with Maryland in being the first to enact the Uniform Computer Information Transactions Act , which is a comprehensive commercial code for transactions that concern software or computer-based information.

Virginia also is one of the pioneers in anti-spam law. While Virginia wasn’t the first state to enact such a law, it may have been the first to base illegality on falsifying the origin of e-mail (so it will get past spam filters, and to try to trick you into opening it).

A later, federal anti-spam statute put some limits on state anti-spam laws, leaving states room only to regulate falsification in sending unsolicited e-mail. This federal limit made Virginia’s statute the model for a tough anti-spam law that isn’t knocked out by federal law.

Virginia recently demonstrated the bite of its anti-spam law by obtaining a nine-year prison sentence for the felony conviction of Jeremy Jaynes in Loudoun County. To my knowledge, this was the first prison sentence handed down anywhere in the U.S. purely for spamming.

Indeed, two of these three cyberlaw firsts - the anti-spam law and the anti-phishing law - flowed from the attorney generalship of Jerry Kilgore. His office teamed up with two legislators to craft the anti-spam law (and prosecuted the Jaynes case), and his office constructed the anti-phishing statute.

Phishing occurs when a bad actor sends an e-mail seeking personal information that purports to come from a legitimate source, such as a financial institution or retailer. The phisher wants to deceive the e-mail recipient into believing that the request comes from a legitimate source so that the recipient will volunteer information that can be used for identity theft, such as a Social Security number, mother’s maiden name or credit card number.

These scams can be convincing. The e-mail address will make the e-mail look like it comes from a real and famous source, such as AOL or Capital One. Often the e-mail will contain links to fake Web pages that look like they’re part of the real company’s online operations. Sometimes the e-mail will contain trust-building links to parts of the real Web site of a famous company.

For more information about phishing, see www.antiphishing.org and a brochure on the Web site of the Virginia attorney general’s office (www.oag.state.va.us) on “spoofing.”

Under Virginia’s new anti-phishing law, which takes effective July 1, using a computer to obtain certain identifying information by “material artifice, trickery or deception” will be a Class 6 felony (admittedly, the lowest level of felony). If the violator also sells or distributes the identifying information gained or uses the information to commit another crime (such as identity theft), the crime becomes a more serious Class 5 felony.

One reasonably asks, though, whether this anti-phishing law will make a difference. Of course it won’t eradicate phishing by itself, but has any criminal law entirely eliminated the crime it proscribes?

Instead, it’s best to view this statute as one prong of a broader attack on phishing. Certainly making this conduct expressly criminal, coupled with a few high-profile criminal prosecutions and substantial prison terms, will deter some folks.

Also, with each new indictment, trial and sentencing, the public will become a little more educated about what phishing is and how to watch for it. I’m constantly amazed by the size of the computer-literacy gap between heavy but nontechnical computer users and casual computer users. As with spam,identity theft and phony online financial pleas, phishing is best defeated by educating people about the problem. Prosecutions and sentencings make the news and spread the word.

I expect we’ll see a few high-profile prosecutions in Virginia for phishing shortly after July 1. The Computer Crimes Unit of the attorney general’s office has been aggressive in prosecuting new cybercrimes. It has entered into a Virginia Cyber Crimes Strike Force with the U.S. attorneys’ offices for the Eastern and Western districts of Virginia, the Virginia State Police and the FBI.

The spamming conviction and sentencing of Jeremy Jaynes demonstrates that enforcement and prison time can result from these bloodless crimes. Thus, there will be more headlines and, hopefully, more awareness.

Moglen is the attorney who controls the settlement of issues involving the mixing of important items of open-source software, such as Linux, into software sold by software makers. He has been quietly resolving disputes that had the potential to cause negative headlines for software makers.

Some background is necessary to understand what Moglen does. For-profit software companies almost never give out the source code to their software – the code in language a human programmer can read. These proprietary software companies also tightly control the copying and distribution of their software.

In contrast, as the name indicates, the source code to open-source software generally is released to the public. Usually one can copy, distribute and modify the software without having to pay anyone. While Linux is the famous example of open-source software, there are thousands of open-source programs out there.

Yet, open-source software isn’t in the public domain. It’s the copyright property of the programmers who wrote it. Open-source software is licensed, meaning that the folks who wrote it put some restrictions and obligations on users.

Within the world of open-source software, a license agreement called the “General Public License” keeps proprietary software makers awake at night. Linux is one of many popular items of open-source software licensed under a version of the GPL.

The steward of the GPL, the Free Software Foundation, would like all software to be open-source. To push toward this utopia, the GPL requires anyone who incorporates GPL code into a larger piece of software to make the source code of the larger piece of software available to anyone who receives that software. So, theoretically, if you buy a copy of a computer game built upon GPL software, the maker also should give you access to the game’s source code.

Here’s what scares proprietary software companies to death: They fear a programmer will incorporate GPL code into their product without management knowing about it. If that happens, they fear that they will be forced to release the source code to their entire product, which could do to their product what file “sharing” software has done to the music industry.

Back to Moglen. Moglen is the person who deals with software companies who initially resist complying with the GPL.

He is general counsel to the Free Software Foundation and founder and chairman of the Software Freedom Law Center. Among other things, the SFLC assists creators of open-source software in persuading software makers that utilize GPL software to comply with the obligation to release source code.

Listening to Moglen gives you the feeling we will smoothly glide into a world where open-source software dominates but no company’s blood ends up on the floor.

In a recent telephone interview, Moglen explained to me how he has persuaded major companies to comply with the GPL. He says that, when he’s brought into the case, he contacts the company that’s used GPL software in its product but did not make its source code available to its customers.

He tells the contacted company bluntly, “I need compliance with the GPL.” But he also offers a carrot – he doesn’t require money or publicity if he gets compliance.

Moglen claims he has had to intervene about twelve times a year to produce GPL compliance and, due to the terms he offers, he rarely has to call twice to achieve his goal.

Moglen focuses on going-forward compliance. In other words, the software maker either needs to start releasing to its customers the source code to its product or it needs to remove the GPL code from the product.

Crucially, Moglen appears to be flexible in addressing a company’s past noncompliance with the GPL. He usually requests that the source code to the company’s product be made available to past customers that the company can locate. Yet, Moglen spoke of alternative means of addressing past noncompliance when the immediate release of source code for existing products might severely harm a company.

Ultimately, a lot rides on Moglen continuing to control the game. He strikes me as a reasonable man who wants to see the goals of the GPL met without having to crush anybody.

But what if someone intent on bringing down proprietary software companies replaced Moglen? For example, in a recent German case, a programmer of GPL software persuaded a court to halt the distribution of an electronic device because it contained that programmer’s GPL code and the device’s maker hadn’t released the source code.

Beyond that, think about what would happen if a software maker harnessed the power of the GPL to wipe out a competitor? What if, hypothetically, a software maker discovered GPL software in a competitor’s product, bought copyright ownership of some of the code at issue from a programmer, and then tried to use that stake as a tool to force its competitor’s source code entirely into the public? Doing so could destroy the sales of the target company.

I do not know if this scenario will occur. A lot depends on Eben Moglen staying in control of the legal demands of the GPL software community.

A recently enacted federal law called the “CREATE Act” made it easier to conduct collaborative research without having that collaboration potentially undermine the prospective patent rights on the inventions that may be generated.

To understand the usefulness of the act, one first must understand the problem it addresses. This problem, arising from a patent concept known as “obviousness,” can damage patent rights any time inventors work together, even if they work for the same employer.

Some background is necessary. There are three criteria for obtaining a patent on an invention: the invention must be novel, useful and not obvious from the “prior art.”

Asking whether an invention is “novel” essentially asks whether the inventor was the first to invent the device or process.

Obviousness builds on the novelty requirement. Even if you invent something novel, to be patentable, your invention must not be an obvious modification of the technology that exists at the time of your invention. This base of existing technology is called “prior art” in patent-speak.

Here’s an important detail on obviousness that many folks don’t know – the universe of prior art can include earlier inventions made by co-employees or company contractors, even if those earlier inventions were kept secret by the company. Thus, an invention, even if kept secret, that one employee makes can make obvious and, consequently, unpatentable, an improving invention later made by another employee of the same company.

For example, suppose two employees, Andrew and Betty, work together for Acme Co. Andrew invents something patentable. Later, Betty invents an improvement to Andrew’s invention. The company keeps both inventions a secret, and wants to apply for a patent on the invention and a patent on the improvement.

Suppose Betty’s improvement would be obvious to anyone who had access to Andrew’s invention and who was skilled in the technology at issue. If so, Acme Co. could get a patent on Andrew’s invention, but it couldn’t patent Betty’s improvement due to obviousness.

Doesn’t seem right, does it? After all, Andrew and Betty work for the same employer and kept both inventions secret. But the federal courts have held that even such “secret prior art” (e.g. Andrew’s invention) can make a subsequent invention (e.g. Betty’s) obvious and, consequently, unpatentable.

The patent law provides a way to avoid this problem. If, before any improving inventions are made, the employees of a company agree to assign their patent rights to that company, then the initial invention will not be considered prior art in judging the patentability of later, improving inventions. (With rare exceptions, employees generally own patent rights unless they agree to assign such rights to their employers.)

So, for example, if Andrew and Betty had assigned their patent rights to Acme Co. before Betty does any inventing, the obviousness problem goes away. Hypothetically, the company could patent both Andrew’s invention and Betty’s improvement.

In fact, this obviousness-avoidance tool can work even if Andrew and Betty work for different companies, provided that, before any improvements are invented, they assign their patent rights to the same person or company. Thus, this tool can be used when utilizing independent contractors on an Research and Development team and in collaborative research between companies.

Here’s where the CREATE Act adds to the intellectual property protection toolkit. Suppose the two inventors work for different companies collaborating on research – Andrew still works for Acme Co. but Betty works for Better Co. And suppose neither company wishes to assign all of the prospective patent rights that may result from collaborative research to the other company or to a joint venture entity.

Once again, obviousness could stymie the patentability of Betty’s improvement to Andrew’s invention, even if the collaborating companies keep both inventions secret until filing patent applications.

The CREATE Act provides that, if the Acme and Better companies enter into a “joint research agreement” before any improvements are invented, they avoid the obviousness problem. Such an agreement can keep a pre-agreement invention of one of the companies, and any of the inventions created in the joint research, from counting as prior art that might make later, improvement inventions obvious and unpatentable.

While the act doesn’t require doing so, a joint research agreement could carve up patent ownership between the parties or save the issue of patent ownership for later resolution.

Thus, the Acme and Better companies hypothetically can enter into such an agreement and save the patentability of Betty’s eventual improvement from an obviousness problem.

Note the emphasis above on the word “before.” You have to be aware of this potential problem and implement the solution (an inventions assignment agreement or joint research agreement) before any improving inventions are made. Indeed, the wisest move is to implement the solution before any inventions are made.

Yet, it isn’t obvious to many companies that a failure to get patent assignments before R&D takes place can kill some potential patent rights on obviousness grounds (pun intended). Hopefully the publicity around the CREATE Act will shine a light on this pitfall.

At the end of June, the Supreme Court issued the most important decision concerning copyrights in the past 20 years – the decision concerning the copyright-infringement liability of Grokster, a company that offers peer-to-peer or P2P file-“sharing” software.

The news reports relayed the result – Grokster lost and the recording industry won. But the various opinions issued by the justices contain some important statements and implications that didn’t get much play in the mainstream media.

Before doing so, full disclosure: I have represented the Recording Industry Association of America, although not in litigation against P2P software distributors or users. The opinions I express below are mine, not the RIAA’s.

The holding in a nutshell. In the lower courts, Grokster had defeated the recording industry’s copyright claims by arguing that its P2P software has “significant, non-[copyright] infringing uses.”

The Supreme Court held, unanimously, that if you intend for your software to be used for copyright infringement and it is so used, it doesn’t matter whether there are such noninfringing uses. The evidence was overwhelming that Grokster had encouraged piracy of copyrighted music by use of its software.

Music copiers are copyright infringers. In order for Grokster and its ilk to be liable for the copyright infringement of others, one must prove that use of P2P software to copy copyrighted music (or movies or software) is copyright infringement.

While the court’s decision concerned the liability of P2P software distributors and not individual downloaders, the court’s unanimous opinion made clear that it considers downloading such copyrighted material to be copyright infringement.

In every P2P software case in the lower federal courts, either such user liability has been expressly found or not disputed. The RIAA has sued more than a thousand P2P software users, and no one has yet to put up a successful defense.

Folks still can argue about what the law should be. The argument concerning the present legality of using P2P software to copy copyrighted music for general use is over.

A blueprint for a legal P2P software service? Grokster lost this case because of the legion of evidence that it promoted use of its software for music piracy. Hypothetically, someone could start a new business offering P2P software that has the same illegal and legal uses as Grokster’s and not be liable for copyright infringement under the inducing-infringement liability rule announced by the Supreme Court. Just put the software out there without comment on possible uses. Folks will catch on.

But investing capital in such a venture would be risky. Content owners would sue anyone trying such a business, and not many folks build expensive litigation into their business plans. Also, it would be hard to remain clean on this issue when promoting the product.

Debate on intellectual property versus technology. Although the court’s opinion was unanimous, two justices issued concurring opinions in which they disagreed with one another over the future relationship between copyright and technology development.

The fight is over the continuing meaning of the court’s decision 20 years ago concerning whether the maker of the Betamax, Sony, could be held liable for the copyright infringement that might be committed by folks copying TV programs. The Sony case essentially held that it’s legal to sell a product that can be used for copyright infringement if it has “substantial, non-infringing uses.” The Grokster case adds the gloss that such sales are illegal if the maker promotes use of the product for copyright infringement and such infringement occurs.

Justice Ruth Bader Ginsburg wrote a concurrence joined by Chief Justice William H. Rehnquist and Justice Anthony M. Kennedy. She wrote that Grokster hadn’t fully proven that its software had “substantial, noninfringing uses.” She didn’t say that Grokster’s software couldn’t meet this test, but she called for meticulous proof of such uses.

Justice Stephen G. Breyer wrote a concurrence joined by Justice John Paul Stevens and Justice Sandra Day O’Connor. He criticized Ginsburg’s call for hard proof of “substantial, noninfringing uses.” In fact, he advocated holding that it should not be illegal to sell a product that can be used for copyright infringement unless it “will be used almost exclusively to infringe copyrights.”

This is a fight over the future effect of the Sony case, which remains in effect. The issue is how much copyright infringement can a new device permit its users to engage in before the device maker becomes liable for copyright infringement even if it doesn’t market the device as a copyright infringement tool.

Ginsburg leans toward protecting intellectual property rights (which might chill some technological development), while Breyer leans toward protecting new technologies (which might be used to infringe on intellectual property rights).

This decision won’t end illegal file copying. The P2P software loaded on the computers of millions of folks should continue to work even if every supplier of P2P software is shut down. Also, this decision will encourage companies who offer such software to take refuge in lawless, third-world outposts. Suits against P2P users will continue, and technology and public relations battles will still be fought.

Here’s a list of pointers I’ve accumulated about negotiating and managing contracts.

While these tilt toward purchasers and technology deals, they apply generally to all contracting.

• If it's not in the contract, it's not in the deal! DO NOT rely on vendor promises that it will take care of it for you, that it's your partner or that no one has ever had a problem before. If that is so, the vendor should be willing to put all promises in the contract.

• Strive for completeness and clearness in contracts. The contract should come as close as possible to explaining the entire deal to an intelligent reader unfamiliar with the deal, because that’s what a judge interpreting it would be. If a dispute arises, the other side will forget or fudge oral promises and construe ambiguous terms to its advantage.

• What are your biggest worries and biggest goals for the deal? Make certain the contract covers the worries and specifies the goals with clarity and detail.

• Some types of contractual provisions frequently have way too little detail and clarity – statements of work (e.g. a description of the service to be performed) and terms concerning training, software maintenance and software support. Most statements of work are filled with ambiguous technical jargon and don’t fully describe the services to be performed.

• Be on your company’s side in negotiations. Many customer personnel become advocates for the vendor just to get the deal done, casting aside concern for the risk to the customer. A good lawyer will keep records of decisions you make, so you will bear the responsibility if something blows up that you blew off.

• In any contract negotiation, there are three things on the table: money, risk, and the product or service. You’ve usually already decided that the product or service suits you. Thus, the primary purpose of the negotiation is to allocate risk and to determine the dollar amount and payment schedule. The vendor’s goal is to get your money ASAP and to shift as much risk to you as possible. The customer’s goal should be to place as much risk as possible with the vendor and to string out payments to encourage strong vendor performance.

• Those terms printed in boilerplate in tiny type on the backside of the ordering document usually will govern unless a court finds them unconscionable, which is difficult and expensive to prove. Most vendors craft those terms to be one-sided in their favor. Customers increasingly offer their own boilerplate terms as a substitute for vendor paper; of course, these terms tilt toward the customer. Don’t just sign the other side’s paper without negotiation.

• Try to negotiate legal terms before choosing a winning vendor. Vendors can smell commitment like adog smells fear. Once you’ve picked the vendor, you have almost no leverage, even if you’re a Fortune 100 company and it’s a large deal. Consider requiring two or three finalists to review your draft of the contract and to state in advance what amendments they seek. Grade their responses and consider the responses in picking a winner.

• A really short contract may feel nice, but it frequently doesn’t cover many important issues. Important contracts often should be tens of pages long.

• You’ll incur substantial legal fees getting a contract in shape – maybe five figures if there’s a lot of negotiation or if the contract is long. Yet, in my experience, you’ll spend $100 litigating an issue for every dollar you would have spent nailing down the issue with proper lawyering at the outset. Some of my colleagues say this 100-to-1 ratio is too low.

• Don’t wait until the eve of the deal to call the lawyer. It usually takes several weeks to negotiate a contract with a new vendor. Big outsourcing deals often take three months to a year.

• What good is a written contract if you disregard it after signing? Someone at the customer company who will be aware of the vendor’s actions must understand the contract and keep an eye on compliance with it.

Get the lawyer to walk you through the contract when the deal is done. If the contract and reality start to diverge, either the performance must be brought back into line with the contract (so you get what you were promised) or the contract must be amended to reflect the different performance.

If you allow performance to drift away from the contract’s terms, the contract will gradually become incapable of protecting you.

For the past couple of years, Congress has been dickering lackadaisically over the details of possible anti-spyware legislation.

Statutes recently enacted by Alaska and Utah might force Congress to finally address this issue.

Until now, opposing interests have been in stalemate in Congress over what technology any anti-spyware legislation should cover. Pro-consumer (and, in some cases, anti-capitalism) forces have pushed to outlaw most adware, particularly pop-up ads that come bundled with some software. At the least, such interests want it disclosed, before it’s installed, in a clear and unavoidable notice.

On the other hand, marketing companies have pushed to keep certain advertising technologies both legal and in the Web-surfing background, such as using tracking cookies and Web bugs to monitor when and how individual consumers view Web sites and HTML-based marketing e-mails.

While Congress has fiddled with this topic, the states have acted. Ten states have enacted anti-spyware laws and several others have it in the hopper. Alaska and Utah recently enacted bold legislation.

Under Alaska’s law, adware is illegal if it pops up an ad when someone else's trademark is entered into a computer. It targets companies such as WhenU, which offers free software that performs some useful function (e.g. local weather) and that bundles in pop-up adware that displays ads related to a computer user's actions. For example, a Google search for “Expedia” could cause a pop-up for Travelocity to appear.

The law does not allow a consumer to consent to this adware, such as in a long end-user license agreement (those terms in the tiny gray box that almost no one reads).

Alaska has anticipated that its statute will be challenged on the claim that it unconstitutionally attempts to regulate all interstate commerce. Alaska’s law requires pop-up advertisers to inquire as to the home state of all advertising targets, to use geolocation software to attempt to stay away from Alaskans, and to not download banned pop-up software onto the computers of Alaskans. I can’t predict how the inevitable constitutional challenge will fare here.

Utah took a similar approach, enacting a statute that prohibits pop-up ads triggered by the entry of a trademark term. Yet unlike Alaska’s law, Utah’s limits its coverage to instances where the pop-up constitutes a trademark infringement. That’s a hot issue in the courts, so we don’t yet know exactly what Utah’s law has banned.

A court preliminarily ruled that Utah’s law impermissibly regulated interstate commerce by protecting only Utah residents. Utah then amended its law to require adware distributors to ask whether the computer user is a Utah resident, just as Alaska demands.

These statutes don’t ban all pop-ups. They only ban pop-ups driven by the typing of someone’s trademark into a computer. Still, these bills might sufficiently restrict the latitude of online advertisers to force them to push Congress to move faster on compromise legislation on spyware.

That’s how legislation on cyberspace issues gets done nowadays. An issue arises, and Congress plays around while marketing forces and consumer/anti-commerce forces duke it out. Then a state such as California enacts over-the-top regulatory legislation. All of a sudden, Congress has to pass something so that the new federal law can pre-empt the aggressive state law. To get something passed, quick compromises have to be reached on the issues where things bogged down.

This pressure created a federal law regulating spam. What Alaska and Utah have done with spyware isn’t as strong as what California tried to do with spam (essentially outlaw it altogether), but it might gore a big enough ox to get Congress moving.

While the likelihood of Congress passing an anti-spyware law has risen, no one can predict what it will say. Congress almost certainly will outlaw truly nefarious spyware – stuff that tries to sniff out your credit-card number and report it to an identity thief. It’s also nearly certain that Congress will shelter most uses of cookies and Web bugs from liability.

My guess is that Congress won’t entirely ban pop-up advertising, even if triggered by typing a trademark into a computer, but it will require some type of notice and consent before such adware is installed.

First, will Congress bless obtaining end-user consent to things such as pop-ups in long license agreements viewed through tiny gray boxes?

Second, for things that Congress makes illegal, will it empower every harmed consumer to sue for a violation and permit class-action suits, or will it empower only the Federal Trade Commission and corresponding state consumer agencies to sue? These governmental actors can’t sue every violator and don’t try – they usually pursue only a few high-profile suits in hope of scaring away other violators.

I don’t know how these issues will turn out, but, thanks to Alaska and Utah, it’s now more likely that we’ll find out soon.

Is it legal to purchase a Google AdWords ad triggered by someone else’s trademark? If so, is it legal to use that trademark in your ad?

The courts haven’t provided us with the final answers, but a recent decision by a federal judge in Alexandria, Virginia, provides comfort for the practice of keying an AdWords ad off someone else’s trademark. Unfortunately, the courts haven’t addressed in detail whether you can use that trademark in your own ad.

First, some background: With AdWords, you can purchase an advertisement that appears next to Google search results for a particular search term or terms. It appears under a heading called “Sponsored Links” (Google’s euphemism for “advertising”).

For example, I could purchase an AdWords ad triggered by the search terms “Virginia intellectual property lawyer.” Google determines whether and where my ad will appear in the “Sponsored Links” section based upon how much I pay per click on my ad and how popular my ad turns out to be.

Not surprisingly, some people want to trigger their ads off the trademarks of others. For example, you might want to offer a competing or complementary product.

The selling of AdWords ads keyed off others’ trademarks received a major test when GEICO sued Google. GEICO contended that its “GEICO” trademark was infringed when Google sold AdWords ads keyed off the mark.

A federal judge in Alexandria orally ruled in Google’s favor last December and, this summer, finally issued a written opinion explaining the ruling.

The key issue in determining whether trademark infringement has occurred is whether ads appearing under the heading “Sponsored Links” confuse search-engine users into believing those ads are part of the organic search results and, thus, might be associated with the trademark owner.

At trial, GEICO offered a survey that purported to show substantial consumer confusion. The court found flaws in the survey and ruled that it didn’t prove consumers were being confused by ads triggered off the GEICO trademark when the word “GEICO” didn’t appear in the ad itself. The court dismissed this part of GEICO’s suit.

To me, this result is correct because the “Sponsored Links” section obviously consists of ads. Yet a survey conducted nearly a year ago by the Pew Internet & American Life Project supposedly found that 38 percent of Web surfers did not grasp the distinction between “Sponsored Links” and the organic search results.

GEICO’s loss does not bind other trademark owners who might sue Google and AdWords advertisers, although the court’s opinion may persuade other courts to rule the same way. Another company, American Blind Co., is litigating a similar lawsuit against Google.

Within the realm of trademark-triggered AdWords ads, the biggest question is whether you can trigger your ad off of someone else’s trademark and mention the entity’s trademark in your ad.

Currently, Google permits you to mention someone else’s trademark in your ad unless and until the trademark owner objects to Google. GEICO objected, but some uses of the GEICO trademark appeared in ads anyway. GEICO claimed this also was trademark infringement.

The judge held that, in this particular case, using GEICO’s trademark in AdWords ads triggered by searching the word “GEICO” was trademark infringement, due to survey results showing that this use of the GEICO trademark confused many Web surfers. Yet the court did not decide whether Google was liable for this infringement, and the advertisers were not parties to the lawsuit. GEICO and Google settled this issue after the court’s ruling.

Logically, it should not be trademark infringement to mention someone else’s trademark in your AdWords ad as long as you do not create confusion as to the source of your goods or services. If your clearly-worded AdWords ad invites consumers to compare the price or quality of a trademarked product to your competing product, that should be considered legitimate comparative advertising.

On the other hand, it could be trademark infringement to onlinedeceive shoppers into believing you carry someone else’s branded goods when you do not, hoping that shoppers will settle for your off-brand goods once they are in your e‑store.

Interestingly, Google recently stated in its blog that it might change its policy and permit the use of someone else’s trademark in an AdWords ad even if the trademark owner objects. Google mentioned legitimate comparative advertising.

The challenge for Google will be to find a cost-efficient way of separating legitimate from illegitimate ads. It’s hard for me to see how this can be done properly without having a person knowledgeable about trademark law manually review each AdWords ad that contains the trademark of an objecting trademark owner.

Yet Google has demonstrated brilliance in devising computer algorithms to filter through chaff and find wheat. Perhaps some smarty-pants at Google can invent an algorithm to separate legitimate comparative ads from trademark-infringing ads.

Imagine this: You’re watching the U.Va.-Virginia Tech football game on TV. Virginia quarterback Marques Hagans rolls out to his right, does a feint to the left and pretends to pitch the ball left, all to get Tech defensive end Darryl Tapp to bite, then zips down the right sideline for a long gain. (O.K., after the recent game, you’ll have to use your imagination really well!)

Amazed, the TV announcer says “Hagans just used his patented feint move to ditch the defensive end and to run for a long one.”

Well, what if Hagans really could get a patent on this avoidance maneuver? Thanks to a recent court decision, such a seemingly far-fetched possibility became less laughable.

In recent years, the U.S. Patent and Trademark Office had followed an internal rule that, in order to get a patent on a process, that process had to be within the “technological arts.” At the beginning of October, a special Patent Office court ruled that no such limitation exists.

This decision may open the door wider, perhaps only for a while, to getting patents on abstract methods of doing business.

To understand why, some background is necessary. Under U.S. patent law, you can get a patent on “any new and useful process, machine, manufacture or composition of matter, or any new and useful improvement thereof.”

On the other hand, the Supreme Court has held that some sorts of discoveries cannot be patented – laws of nature, physical phenomena and abstract ideas.

But what kind of “process” can be patented? If Hagans were the first to conceive of it, could he get a patent on a way for a quarterback to evade a defensive end?

After a 1998 court decision clarifying that methods of doing business can be patented in some circumstances, the Patent Office added a requirement for patents on processes – that the invention have a technological component.

Historically, such a requirement could be satisfied by, for example, having the process performed by a computer or other device. Patent law also requires that to be patented, a process must produce a “useful, concrete and tangible result,” but producing useful information with a computer program usually satisfies this criteria.

The October decision by the Patent Office court wiped away the “technological arts” test in the patent-examination process. In that case, the patent applicant sought a patent on a method for compensating a manager of a privately held business in an oligopolistic industry, in a way that would reduce the incentives for collusion between that manager’s firm and industry rivals.

That concept is a mind-full, but you don’t need to understand the invention to get the point here. What is noteworthy is that the patent described a process that had no technological component. It was just a financial calculation of how to pay someone, coupled with the commonplace act of paying that person.

The patent examiner applied Patent Office policy and held that the “invention” wasn’t patentable due to a lack of a technological component. Ultimately, the Patent Office court held that no such requirement exists under the U.S. Constitution or the federal patent law.

This decision raises two questions: Will the decision stand? Also, what does it mean for business and our society?

On the former question, I wonder whether the framers of the Constitution intended to authorize giving out patents on processes where the purported new development consists of a mental process or financial calculation where the implementation is non-technological.

The “Patent Clause” of the Constitution gives Congress “power . . . to promote the progress of science and useful arts, by securing for limited times to . . . inventors the exclusive right to their respective . . . discoveries."

Also, Congress might wish to examine whether granting these sorts of patents promotes “science and useful arts.”

The Supreme Court recently accepted a case in which it might examine the technological arts requirement. The case concerns whether one can patent a method for diagnosing vitamin deficiencies by testing for an elevated level of an amino acid in the blood. That case doesn’t force the Supreme Court to address the technological-arts requirement but provides an opportunity for it to do so.

Next, assuming the decision stands over time, patent applicants will no longer need to put a technological step or facet into a patent claim. Thus, one could obtain patents on ways to do things, such as performing a useful financial calculation, without having a computer process or other device in the picture.

We’ll see if Marques Hagans takes his cue and scrambles for the Patent Office. Of course, he may have a problem with not being the first inventor of the shake-and-bake move, and to get a patent, you have to be the first person to invent the invention. Michael Vick and, before him, Shawn Moore, might want to remind Hagans of a little college football history.

I love Google’s plan to scan parts of the collections of five prominent libraries and to make those books searchable online.

But will Google’s effort survive the copyright infringement claims made against it?

Google has two projects under way – Google Print for Publishers and Google Print for Libraries – to make the book contents searchable.

You can use the Google “book search” function to search the books Google has already loaded into its database via the Publisher Program and the Library Program. On each book search results page, Google displays options for buying the book from various vendors.

The Publishers Program isn’t legally controversial. Under it, book publishers agree to have their books made searchable on Google. In response to a search query, only some relevant book pages plus two pages on either side are displayed in search results.

But not every publisher will sign up for the Publishers Program. To cast a wider net, Google launched its Libraries Program.

Under it, Google made agreements with five libraries to scan all or part of their collections: the University of Michigan, Stanford University, Harvard University, Oxford University and the New York Public Library.

With the Libraries Program, Google scans the full book and makes it searchable. Yet, the search result for books still under copyright displays only the snippet containing your search terms – the search term in about three lines of text – plus bibliographic information about the book.

A PUBLIC SERVICE – The Libraries Program could be a tremendous educational service. For many, if information is not on the Web, it might as well not exist.

This program also could be a boon for authors. The vast majority of books are practically unknown and undiscoverable to the world. With this program, nearly all authors can hope for more books sales and greater notice.

So why would anyone object? Several book publishers and the Authors Guild sued Google alleging copyright infringement. Google launched the Libraries Program over their objections. The publishers essentially claim that copyright law gives them control over whether a database of their authors’ works may be constructed or utilized online.

In my view, this dispute isn’t really about whether the Libraries Program should exist. What the publishers really want is control over the online-advertising revenue that operating such a database can attract.

No rulings have yet been made in these suits. Below is my take on whether the Libraries Program will withstand the copyright infringement claims.

FAIR USE IS KEY – Fair use is a defense to infringement liability for someone who exercises one of the exclusive rights of the copyright owner without permission, such as the right to make copies of a work, such as a book.

Unfortunately, the law does not define fair use clearly. It’s judged on a case-by-case basis. The most important factor is the effect of the unauthorized use of a copyrighted work on the potential market for it.

If the unauthorized use tends to substitute for sales of the copyrighted work, then the use often is not fair. On the other hand, if the use tends to complement the sales of the copyrighted work, the use often is fair.

GOOGLE’S CHANCES – Consider the legality of Google’s Libraries Program in two parts: (1) making a full copy of copyrighted books for Google’s database and (2) posting snippets from these books in search results.

As for book copying, fair use justifies making an entire copy of a work only rarely. Yet, Google’s complete copying of books is just an intermediate step to generating search results. Google users won’t have access to the full copies.

The posting of snippets easily is fair use in my view. Free access to a few snippets shouldn’t make buying the book unnecessary. Indeed, book sales should increase from exposure.

Beyond the legal analysis, some courts will silently consider whether what Google is doing could be done if consent must be gained from every book publisher. Given the fragmented and old-school nature of publishing, it’s hard for me to see that industry pulling such a project together.

Indeed, except for books published in the past several years, publishers may not have the contractual power to authorize book copying for database purposes. Instead, any such right may still belong to the individual authors.

In addition, I believe courts will find the Library Program’s potential societal benefit to be too attractive to rule against.

Your use of this Web site constitutes acceptance of our terms of use and privacy policy even if you choose not to view them. The information you obtain at this site is not, nor is it intended to be, legal advice. You should consult an attorney for individual advice regarding your own situation.