Andreas Schachtner <firstname.lastname@example.org>
Michael Schneider <email@example.com>
One's own name has traditionally been assigned a special role in our society. Nevertheless, a structured system protecting legal rights to a name historically developed relatively late. Earlier legal regimes--also the medieval ones, such as the Prussian General Common Law or the French Civil Code--contain no regulations recognizing a subjective right to the use of a name or granting legal protection of a name under civil law. Legislation that has led to the complex legal regulations in force today began in the nineteenth century and has been continually developing ever since.
Internet name rights are currently at a stage comparable to the general situation during the last century: Subscribers and the "authorities" have just begun to realize that safeguarding identification on the Net is indispensable. However, as there exists no traditional applicable law that contains regulations specific to the Internet, the amendment or extension of existing laws is being attempted. This paper will show to what extent existing law principles are transferable to the constant use of a name on the Internet.
In the Middle Ages, the protection of personal rights--and thus the protection of a name--was irrelevant. However, the necessity of protecting intellectual assets--particularly inventions--had already been realized. This protection was granted in various ways by ruling sovereigns. The nature and duration of the protection, however, was a matter of each ruler's discretion. Nevertheless, the rights of inventors in the economically significant states of those days were effectively protected. The city state of Venice, for instance, issued a pioneering patent protection law as early as 1474 that remained unaltered for more than 100 years. As we know from a procedure dating from 1594 , this order was applied in a manner similar to today's patent protection procedures. England, too, created legal provisions for patent protection with the "statute of monopolies" in 1623, which survived until the 19th century.
The right of one's own name was not considered equivalent to technical rights of protection. In this respect, parallels between this historical development and the debate regarding domains that has been carried out on the Internet can be clearly recognized. Everyone knows that a name is an important instrument to distinguish among individuals, companies, and goods. But in the past, it was by no means concluded that the authorities should regulate the choice of name. On the contrary, the opinion was held that a name was fundamentally insignificant from a legal perspective. Hence, it was considered everyone's right to freely choose a name and later abandon it at his or her discretion. The majority of countries rarely intervened in the freedom to choose a name--usually only if a name was misused to commit fraud or other crimes.
The first attempts at a fundamental change in the assessment of name rights are revealed in the decision of a German court in 1781. The gist of this ruling is that the freedom to choose a name should result in no disadvantages for any third party. In the view of the court, the basic principal can be deduced that every person must have the right to forbid others to use his or her name. However, it was not until almost another century had passed before this view gained worldwide acceptance. Only then did a comprehensive system for the civil law protection of individual and company names develop, which was finally extended to cover the field of trademark protection.
Different legal regulations have been developed for the administration of name spaces and the assignment of names. The more important it is to be able to differentiate between each name bearer, the more detailed and restrictive are these regulations in the respective fields of law.
In private life the capacity of a name to individualize consistently is of secondary importance. Therefore, all legal systems known to us come to terms with the clash of individuals' names. As far as it is necessary in particular cases to differentiate more accurately among individuals, further personal details (first name, date of birth, place of residence, profession, etc.) are added. The relevant regulations in force can be reduced to two principles in many countries: an individual cannot change his or her name without official consent, and someone else's name cannot be used unlawfully.
In business life, greater requirements are demanded for naming for a number of reasons. First, the distinctiveness of a name is of particular economic significance not only for the company but also for the relevant commercial sector. Whoever is not recognizable as the supplier of a product will find it hard to establish lasting relationships with customers. Furthermore, the danger exists in business that a name will be exploited by competitors. Therefore, company names are, as a rule, regarded as especially worth safeguarding. However, all legal systems take particular care not to grant absolute protection, as this would result in a disproportionate restriction of the name space. In order to reach a reasonable compromise, various concepts are applied:
A particularly formalized system for the protection of denominations has been developed for the naming of products, services, and--to a certain extent--creative works. This system is called "trademark law." Trademark protection can be acquired in two ways: by the official registration of a trademark or through the acquisition of general acceptance in business as a result of intensive commercial use of the trademark.
The brand name has a dual purpose: It serves the company by distinguishing its products, and it serves consumers because a brand name is often more easily recognizable than the name of the company. Encouraged by advertising, the targeted consumers tend to choose a product as a result of its social prestige and the illusion created by this advertising rather than applying more objective criteria. The economic significance of this development becomes obvious when marketing concepts such as merchandising are considered. The additional sales achieved by a product in comparison to "no-name" articles is largely based on the strength of the trade name connected with the product. It is obvious that the effects as described can only be developed to the full if different suppliers of the same goods or services do not use the same mark. Therefore, legislators worldwide have decided to grant the owner of a trademark a monopoly of limited range. So as not to restrict the area of available names, regulations have been developed to limit the monopoly of a trademark in a fair way. To do this, two different strategies are used:
The solution for cases of conflict usually follows the so-called "principle of priority," which means nothing more than the fact that the holder of a prior right can assert that right against someone else who wishes to use a name at a later date. The principle of priority is usually also applied to the administration of TLDs.
Nonetheless, in practice, the registration of identical, legally independent marks of the same class happens again and again. Such problems can (and this is another clear parallel to the Internet) be the result of the fact that the registration of the marks has been applied for in different countries. The reason is the so-called "principle of territoriality." In accordance with this, the validity of the respective national trademark law is confined to the sovereign territory of the state in question. In other countries, a mark can be registered anew for comparable products. In the past, various attempts were made to overcome territoriality by supranational treaties, such as the Madrid Treaty; however, this has not been entirely achieved. The result is that in trademark law, too, the multiple use of marks in different regions--here defined by national boundaries--has occurred.
In some legal systems, conflicts over names can also arise at a national level. In many cases, although the holder of a prior right enjoys priority, he or she must consent to the use of this mark by a third party. This can occur, for instance, if the registration of an identical mark is not opposed in good time or if the earlier mark has not been used over a long period of time. There is no legal solution in such cases of conflict; the disputed marks must exist alongside one another, and their owners are compelled to compromise.
On the Internet, the clash of names is tolerated even less than in the legal sphere. To avoid inconsistencies when registering a name, it is necessary that the same IP number or the same set of IP numbers are supplied. A further differentiation is not possible.
During the early days of the Internet, all connected end-systems were entered into a globally managed file "hosts.txt" with reference to the respective IP number. Local changes were centralized and current versions from hosts.txt were downloaded by individual Net administrators. This led to a central bottleneck in the updating of the central file. Moreover, duplicated names, in particular the favorite names at that time (e.g., trillion), could only be eradicated with difficulty and repetitive effort.
Therefore, in the 1980s, the concept of the central administration of the name area was superseded by the Domain Name System (DNS). The DNS is based on a hierarchical, decentralized approach; responsibility is represented in a tree-like structure and delegated to administrators, each of whom is responsible for a part of the name space. The essential design criterion was the provision of a technical method with which, by delegation, mnemonically meaningful names could be decentrally coordinated with (technical) IP numbers . Implications of name rights have not been taken into consideration in the DNS concept. The Internet Assigned Numbers Authority (IANA), which has the ultimate responsibility for the administration of DNS, has drawn attention to this on several occasions . Therefore, the practice of issuing names in the 1980s was largely determined by technical criteria or by procedures introduced through ad hoc processes that could be implemented and carried out by the administrators with little effort.
Only with an increasing use of certain domains--mostly the com-TLD, but also country-TLDs--it turned out that the same name was requested by several organizations. Such conflicts could not be resolved because of the necessity for unambiguous domain names. Therefore, applicants who wished to use a domain already registered were advised to change to a similar name or acronym. However, this has become practically impossible. Numerous second-level domains (SLDs) have already been reserved as a result of the WWW boom in 1994 and 1995, so that in many cases no satisfactory alternatives can be found. In addition, online presence in the commercial sector has become so important that many companies no longer want to refrain from preserving their corporate identity on the Internet.
A concept that seems to suggest itself is to apply the legal rules mentioned above to the assignment of domains. However, in every respective field of law we discussed--even in trademark law--cases of conflict must not be solved in a way that names are assigned in an unambiguous manner. Because of the existing structure and the users' expectations, such distinctions are not possible in the DNS. A name can be used, with few exceptions, only once. Here a division ensues between the sphere of law and the sphere of the Internet. This cannot be resolved by existing rules, as long as the Internet does not provide the technical or organizational preconditions to enable more accurate differentiation.
To solve this dilemma, various approaches are discussed.
The proposal to extend the existing name area by introducing additional TLDs was presented by Postel  in August 1996 and has meanwhile (January 1997) been drawn up by an international ad hoc committee .
In accordance with the above, seven new TLDs are to be created, with competing registrars enabling entries under these TLDs. The objective is to allow an extension of name areas at the root of the DNS to prevent further registration under the already overcrowded domains (.com). The names of the TLDs should in some way be meaningful. TLDs "will consist of letter strings of three to four letters each. IAHC has also decided that each TLD should have general, contextual meaning."
It is vital that the names reflect in a simple fashion the categories of trademark law. Otherwise, the fundamental problem (inconsistencies in the application of trademarks in trademark law and names in the DNS) will not be resolved.
A structured extension can be seen as a variant on the simple extension. The former transfers the distinctions of trademark law and matches the extension with the internationally unified categories of trademark law. This enables an analogous registration of domains and trademarks.
Nevertheless, this proposal does not take the regional distinctions into adequate consideration. Furthermore, legally defined categories are not intuitively comprehensible for laymen. Thus, the effect of an Internet presence based on corporate identity is not or only insufficiently realized.
Instead of extending the sphere of TLDs, it is possible to extend the second-level domains. With this approach, the above-mentioned extensions of the name area (categorization by trade name and region, respectively) would be carried out under the country TLDs.
A consistent implementation of this plan would, however, result in structuring according to regional aspects and categorization according to trademark law or vice versa. This would make the area of names extremely confusing and would probably be accepted neither by users nor providers.
What stands in the way of all these approaches for the extension of the area of names is that the entries hitherto made in the TLDs (whether .com or country TLDs) would have a head start on all the succeeding domains issued according to a complicated scheme. This could trigger a conflict over entries in these privileged domains.
An approach circumventing this dispute is a strategy to depart from the significance of domain names altogether. Domain names are frequently used as a simple way to acquire information about a company via the name of the enterprise . If a mechanism were available that enabled such access even though a company's Internet presence could not be inferred from its name, the domain name would recede in importance and could be superseded by other constructs (see above) or the use of names entirely unrelated to those of the companies .
This approach has great potential for a long-term transition from a name-based, referential search to a content-related, associative one. At the moment, however, there is reasonable doubt as to whether users will accept this approach. The name of a company is frequently the only reference the user has at hand. Searching for SUN Microsystems GmbH in Germany  is an example of how a search with Alta Vista (one of the most comprehensive and powerful search engines on the Internet and therefore very popular among users) supplies a number of references without making the desired URL easily to find . Another disadvantage of making the domain names unrelated to companies represented is that an accurate entry can no longer be recognized by the URL.
The solution we propose is based on a domain that can be jointly used, in case several applicants raise a rightful claim to the related name.
The joint administration of a domain aims to make the services under a domain equally accessible to a group of users. Here service dependent solutions must be found. We shall describe this with examples of a joint home page, generic e-mail addresses, a whois directory entry and, of course, maintenance of the DNS.
In the case of joint use, for each domain under a TLD <domain>.<tld> the name www.<domain>.<tld> is reserved for the jointly used home page. The previous sole use by the first holder is terminated.
On the home page, its joint use as well as a list of the participating organizations with reference to the URL of their respective home pages are to be prominently displayed. The joint home page is to list the users in a neutral form. Graphic elements are only to be employed in the form of small icons referring to the relevant URLs.
A domain, shared.<domain>.<tld> is to be kept free for the provision of a name space for joint users. Here, succeeding joint users of the domain will be able to establish their own name infrastructure under an unambiguous subdomain (given as D-<no> along with the serial number of the claim registration).
The joint Web server is to be operated by the administrator of the domain.
A common e-mail address info@<domain>.<tld> is to be established as the generic address of the users of the domain. Every joint user of the domain can be entered here using an alias. The responsibility of the further maintenance of this alias rests with the user.
In addition, mail distributors info-<descr>@<domain>.<tld> are to be set up under which the user alone can be specifically reached.
Directory entries shall be portrayed using the example of the RIPE and InterNIC whois directories.
Essential components of these directory entries are fields for the description of objects and the naming of addressees (admin-c, tech-c). Directory entries of a shared domain should indicate joint usage:
*descr: Domain in shared use by
*descr: <list of orgs>
*descr: additional information may be obtained from http://www.<domain>.<tld>/
As far as the objects are concerned, the addressee is the respective administrator of the domain. He or she is responsible for the maintenance of the objects.
The maintenance of the primary master name servers lies within the responsibility of the administrator of the domain. He or she must provide CNAME-RRs for both a Web server and a mail server.
The delegation of the jointly used part of the area of names shared.<domain>.<tld> is effected by the administrator as soon as a new member registers and names name servers that satisfy criteria for operating a zone.
To ensure effective joint use of a domain, an administrator must be appointed who is responsible for maintaining the name servers, Web servers, directory entries, etc. With regard to this, there are various conceivable alternatives:
Each of the three variants above possesses advantages and disadvantages. With regard to the neutral status of the administration, (2) and (3) are better. (3) does not rank particularly high in the case of a great proportion of jointly used domains. (2) has a good chance of creating a market and thus cost-effective tenders for the administration of such domains.