United States: Intellectual Property In Cyberspace

If you think about it, the first form of intellectual property law was real estate law. From groundings in ground, elaborate intellectual constructs were created to describe interests in land which had no corporeal form, but were concepts or ideas. We are so used as lawyers to dealing with terms of years, easements, rights of way, life estates, leaseholds, profits a prendre, land patents and incorporeal hereditaments that we have virtually forgotten that these are all ideas only - the first intellectual property.

As we move into cyberspace, we find the same four (five if you count the publicity aspect of the right to privacy) forms of intellectual property law which apply to non-computerized intellectual property also apply in cyberspace: trade secret, copyright, patent and trademark.

It may be helpful to review in concept what each of these forms of intellectual property is, before discussing each in a computer law context.

Trade secret law is contract law. A trade secret is protected essentially because the discloser and the recipient of the trade secret have agreed in some way (or are statutorily bound) to keep it secret.

Copyright law protects authorship and principally protects the array or peculiar juxtaposition of words and concepts fixed on a page or in another construct. Copyright is solely the creature of a federal statute in the United States and is subject to a wide variety of international treaties and conventions.

Patent law is statutory as well, though, as with copyright law, the statutes are grounded in the United States Constitution. Patent rights exist because of a quid pro quo. Inventors of useful ideas are entitled to make, use and sell useful things, or, more precisely, to prevent others from making using and selling useful things, embodying novel concepts of the inventor. This right of exclusive exploitation is given in exchange for full disclosure during the patent registration process of the technology involved.

Trademark (and service mark) protection exists to protect the public against confusion as to the origin and quality of goods and services purveyed in commerce. Trademarks rights are derived at common law from their commercial use to identify goods, and are also protectable by registration and filing at the state and federal levels as well as pursuant to international conventions and treaties.

With that as background, let us examine how these principles are being applied in cyberspace.

Patent law was not available for computer programs for a long time in this country, and getting a patent is a long, involved, expensive process in the best of circumstances. Copyright protection for software was also circumscribed, because, once a good programmer or systems engineer could read a copyrighted program and understand how it worked, it would be quite simple to design a fundamentally equivalent algorithm which could duplicate all the functions of the copymarked software without infringing the copyright which only protected the "fixed" ideas of the author; not the operational technology that made the program work.

For this reason, the method of choice for protecting computer technology for years has been the trade secret.

Because trade secret is a contractual protection, people in the computer industry have had to be quite creative to protect their trade secrets.

They started by following the Xerox and IBM models by refusing to sell computer programs, but merely licensing them. Because the license preserved ownership in the licensor, the industry reasoned it could preserve rights in the technology forever, even if the license involved a one-time, up-front fee and provided no set term for use and no duty to return the program to the licensor.

The industry also started taking precautions to embed security technology in software to make it harder to duplicate and began issuing periodic updates and upgrades to which users had to subscribe in order to maintain their software at optimal usefulness.

The terms and conditions of sale or licensure, including the trade secrecy obligations, often were embedded in paperwork printed on the paper, cellophane or other wrapping medium in which the physical object (e.g. CD-ROM, diskette, tape) impressed with the software was contained, and these wrapping paper terms and conditions of sale or license became known as a "shrinkwrap" license. Attempts were made to have the purchaser/licensee of the software do some affirmative act to adopt the trade secrecy obligations, such as sending in a warranty or upgrade entitlement card to the software developer, but, by and large, these were one way contracts of adhesion.

Nonetheless, shrinkwrap contracts became generally enforceable contracts of adhesion - at least for the first sale, although some courts found that such contracts were not enforceable after a given piece of software had been resold by the original "licensee."

With the advent of the Internet, software began to be sent, sold and licensed electronically all around the world. The industry then developed another contractual safeguard - the "clickwrap" or "clickable" license.

Upon downloading "clickable" licensed software from the Internet, in addition to arranging for payment, the user would be required to authenticate the user’s identity and "click" (i.e. indicate by "mouse click") the user’s assent to the license terms, including trade secret terms. Because it requires specific affirmative action by the downloading user, in theory at least a "clickable" license should provide even more trade secret protection than a "shrinkwrap" license.

But systems developers have statutory protections for their trade secrets, as well. Indiana, for example, is one of 29 states which has enacted a version of the Uniform Trade Secrets Act, Indiana Code Section 24-2-3-1 et seq., which allows for private actions for damages and injunctive relief whenever a trade secret is misappropriated where there is a contractual relationship between the discloser and the recipient of the trade secret or where the parties otherwise enjoy a confidential relationship or the recipient learnt the trade secret by piracy.

In addition, in 1996, the federal government enacted the Economic Espionage Act of 1996, 18 United States Code Section 1831 et seq., which makes domestic or international theft or misappropriation of trade secrets, including computer data and transactions, a federal crime. Unfortunately, there is no private right of action under this statute.

Finally, software developers have maintained the secrecy of their software by means inherent in the technology itself. Computer programs and data are rendered in machine readable form, that is, as primary data (ones and zeros). Very few people are capable of looking at an array of machine readable data comprised of ones and zeros and stating with certainty how the program works and what it does exactly. Reverse engineering from such data, even if it were possible, would be time consuming, expensive and difficult. For most programs, few would have the patience, particularly if the program were also protected by encryption, which is sometimes the case.

The result is that software developers often can publish their programs in ways which make the machine readable aspect of their programs - the so-called "object codes" -discoverable by their licensees, with little actual fear that such trade secrets will be lost by piracy.

What such companies guard dearly, however, the so-called "source codes" for their programs. A "source code" is a kind of schematic, similar to a critical path or PERT chart, which lays out in diagram and human readable form how the program works and each step it takes to do what it does. Experienced programers upon reviewing source codes can duplicate them or create functional equivalents relatively simply. For this reason, in financings or large project developments, it is not unusual for a "source code escrow" to be set up with a bank or other escrow agent intermediary to turn over the source code in the event its developer defaults or goes out of business, while protecting the secrecy of the source code in the interim.

For many years, it was believed that software was not patentable because it was wrong to patent what was essentially a mathematical algorithm. This changed in 1981 when the United States Supreme Court ruled in Diamond v. Diehr, 450 U.S. 175 (1981), that a computer program to control the temperature of the rubber mold in a vulcanized rubber system was patentable.

Since that time, despite the expense and time consuming nature of the patent process (it can take three or four years to get a software patent), more and more software related patents are being applied for and granted. In fact, it is estimated that as many as 10,000 United States software patents are currently pending.

After 1981, when more and more software patents were being applied for, the patent office struggled to sort out which software was a mere unpatentable mathematical algorithm and which software was a patentable invention meeting the traditional patent law requirements of usefulness, novelty and non-obviousness.

Since the 1990's the Federal Circuit in Washington, which concentrates more on patent matters than the other federal circuits, has clarified matters. On the one hand, if the computer program merely computed, such as a program to convert binary numbers to decimals, then it is an algorithm and cannot be patented. On the other hand, if the computer program evaluates or analyzes data to come to some conclusion about something, such as whether or not a heartbeat is irregular, then the invention is a patentable process relating to real world values despite the fact that it may contain a computing algorithm. In short, the Federal Circuit reasoned, you need to evaluate the function of the programming invention as a whole, rather merely fret over whether or not the program did calculations, in order to decide patentability.

As a result, the Patent and Trademark Office of the United States Department of Commerce in 1996 adopted Examination Guidelines for Computer Related Inventions, which clarifies the application process and which is available through the Patent and Trademark Office.

There are three main issues relating to copyright law in cyberspace that I would like to address here. The first is the extent to which a creator of a copyrightable computer application can rely upon copyright protection. The second concerns how safe it might be to use on your own computer copyrighted materials available over the Internet. The third concerns the extent to which one can disseminate copyrighted materials downloaded from the Internet.

As you are aware, the Internet is a series of protocols linking computers world wide and the World Wide Web is an array of databases stored on computers around the world.

To the extent what you want to protect is a computer program, computer programs clearly are copyrightable in the United States as literary works.

But so what? It is so simple in practice to reconfigure a copyrighted program so as not to constitute a protected "derivative work" that many program creators feel that copyright protection is virtually useless - particularly because operant or command aspects of copyrighted materials have specifically been held non-copyrightable. Lotus Development Corp. v. Borland International, Inc. 49 F.3d 807 (1st Cir. 1995) affirmed 116 S. Ct. 804 (1996).

With the advent of software patents which do protect the command aspects of computer programs, copyright protection for anything but the most cosmetic aspects of computer programs fixed in tangible form has come to a virtual standstill.

What about databases? A database is copyrightable as a "compilation" or a "collective work," 17 United States Code Section 101. The problem is, a mere alphabetical grouping of information in the public domain is not necessarily copyrightable, as some telephone book publishers have found to their dismay. See Feist Publications, inc. v. Rural Telephone Service Co., Inc., 499 U.S. 340 (1991).Consequently, while a database can be protected by copyright, copyright is probably not a very strong protection, particularly when you consider that accessing the database via the Internet may well constitute a permissible "fair use" pursuant to the federal Copyright Act. 17 United States Code Section 107. Furthermore the maintainers of the database and the Internet Service providers can argue that they are not violating copyrights either because they are merely passive service providers and are not themselves publishing or transmitting the materials from the database but merely passively allowing Internet browsers to locate and look at such materials. They are libraries, not publishers, such persons can argue.

Such arguments are buttressed by the new "Digital Millennium Copyright Act of 1998," as enacted by Congress in October 1998, and signed into law by the President. Essentially, in addition to including technical amendments (including new anti-hacking provisions) to enable the United States to join in certain World Intellectual Property Organization Treaties, this new Copyright Act amendment adds a new Section 512 to federal copyright law. Essentially, new Copyright Section 512 will exempt providers of on-line services and Internet access from copyright liability when they transmit or provide access to materials on-line. Providers which do not actively control, participate in or know about infringement will generally be immune from copyright infringement suits under the new Digital Millennium Copyright Act of 1998.

The line of argument leads rather naturally to my second point, which is that users of the Internet who find databases on the World Wide Web and download information from them for their own use probably, or at least arguably, are not infringing copyrights, or, if they might be, such individual perusal constitutes "fair use" of copyrighted materials pursuant to 17 United States Code Section 107.

It is the third category of usage of computerized databases which is more troublesome, and constitutes a more unsettled aspect of copyright law and one which likely will be addressed by Congress in the near future. This third area concerns Internet users who download and then retransmit computerized materials by storing or framing them on their Web pages, using them in their hyperlinks, transmitting them to a wide circle of recipients, uploading them to bulletin boards, or printing them out and using them as text materials in their classrooms. Some of these uses may be deemed "fair use." Some of these uses may be deemed legitimate because of the "first sale" doctrine. (I.e. The copyright protects the book, etc. only through its "first sale." After that, the work is in the stream of commerce and is not protected as the creator’s exclusive property. 17 United States Code Section 109).

But some of these uses may not be immune from copyright infringement lawsuits. With all of the money changing hands over the Internet these days, it is likely that commercial use of the copyrighted materials via the Internet and World Wide Web will be a legal battleground for years to come.

For this reason, some lawyers advise their clients to print Website notices right on their Websites in an attempt to shift liability to the user of the Website. Such notices may include copyright notices, terms of use of the Website, and disclaimers of warranty and liability. It is doubtful that such notices will be completely successful in shifting all liability from the Website operator, and lawyers should advise their clients to be cautious about including, framing or hyperlinking to materials copyrighted or trademarked by others.

C. Registering And Protecting Trademarks.

As indicated above, trademarks and service marks are afforded protection at common law, under state law and pursuant to the Federal Lanham Act, 15 United States Code Section 1051 et seq., because they protect the public against confusion as to the origin and quality of goods and services. Note that this differs from the common law tort right of publicity which is an offshoot of the right of privacy. In the case of trade or service marks it is the public which is the intended beneficiary, whereas in the case of the right of publicity it is the owner of the right which is the intended beneficiary.

The first thing to know about trade or service marks is that rights therein derive from commercial use thereof. The mantra "use it or lose it" becomes important, for there are marks out there - "aspirin" springs to mind - which have become unprotectable because not consistently enough used in connection with one manufacturer’s product only so as to protect the public from confusion. The result - a mark that can be used by anyone with anything - a "generic" mark.

Owners of trademarks or service marks can protect themselves at common law, state law and under the federal Lanham Act by bringing lawsuits against those who would use the same or similar marks with similar goods or services. The essence of these causes of action is that the use of the same or similar marks by others confuses the public as to the origin and quality of goods previously solely associated with the plaintiff’s mark. Such actions are for "infringement" and/or "dilution" of the owner’s rights in the mark said to be infringed or diluted. There are also actions available under the Lanham Act for "unfair competition" which may involve "passing off" the infringer’s goods as those of the owner of the mark, and which may not even require proof of a trademark infringement. See 15 United States Code Section 1125(a).

What does this have to do with the Internet? A lot, it turns out, particularly as regards domain names. The practice developed of handing out domain names on a first-come, first-served basis without ever checking to see whether the assigned domain names infringed existing registered trademarks.

Not surprisingly, quite a bit of litigation resulted and Network Solutions, Inc. ("NSI"), which assigned all domain names in the United States, began to be named as a co-defendant. NSI promptly began promulgating new domain name assignment policies, which require, among other things, the applicant to swear that the assigned name is non-infringing and will not be used unlawfully. Also, NSI now will place use of a domain name on hold pending resolution of any challenge thereto by the holder of a registered mark, although this policy also has drawn NSI into trademark/domain name disputes. Complicating matters, NSI now is a "lame duck", having been supplemented in October 1998, by a non-profit corporation called "The Internet Corporation for Assigned Names and Numbers", or ICANN. Consequently, NSI is in the process of partial replacement as the assignor of domain names since ICANN recently named five more companies as registrars of the domain names ending in ".com," ".net" and ".org."

Further complicating matters is the fact that a trade or service mark is an adjective whereas a domain name is a noun used to identify an address. Domain name owners have argued - generally unsuccessfully - that a domain name which is a functional address is not being used in a trademark sense and cannot infringe. This argument has generally lost, based in part on the new federal Trademark Dilution Act amendments to the Lanham Act, 15 United States Code Sections 1125(c) and 1127 (1996), which, in effect, does away with the requirement that the public be confused by an "infringing" use, but only requires that the use be dilutive of the uniqueness and the distinctiveness of the mark claimed to be diluted. (Note: Although a majority of the states have state trademark anti-dilution statutes, Indiana does not.)

How can you help your clients avoid trademark problems in cyberspace? First, advise your clients to do a domestic and perhaps even an international trademark search of any domain or other names which are planned to be used on a Website.

Second, suggest that the client register the domain name not just with NSI, but with at least one state trademark registry. This will ensure that your client’s use of that mark will be picked up as a prior use by anyone conducting a comprehensive national trademark search after the date of state registration.

Third, tell your clients to be careful about what they frame or hyperlink to on their Websites. Litigation has resulted from framing or linking news stories from one medium on another’s Web page on the grounds that it confused the public into thinking that there were actual business links between the original publisher of the news items and the owner of the Web page and that such linkages thus constituted trademark infringement. See, Washington Post Co. v. Total News, Inc., 97 Civ. 1190 (S.D. N.Y. filed Feb. 20, 1997 and later settled).