An Immodest Proposal – how to solve all domain name disputes at no cost to anyone

November 1, 1998

In its white paper ‘Management of Internet Names and Addresses’, the USDepartment of Commerce proposed a major role for the World Intellectual PropertyOrganisation (WIPO) to initiate a ‘balanced and transparent process, whichincludes the participation of trade mark holders and members of the Internetcommunity who are not trade mark holders’, inter alia to develop‘recommendations for a uniform approach to resolving trade mark/domain namedisputes1 involving cyberpiracy2 (as opposed to conflictsbetween trade mark holders with legitimate competing rights)’. WIPO has acceptedthis role, and issued a document ‘Request for Comments on Issues Addressed inthe WIPO Internet domain name process (WIPO RFC.2)’. As part of the process ofgathering input on the issues it identifies, WIPO has commenced a series ofpublic meetings at which interested parties can present their views.

One of the first such meetings took place in Brussels on 29 September. Thelast will be in Sydney on 4 November but it may already be safe to set outcertain general criteria that people want to see met by any process addressingtrade mark/domain name disputes. In the light of these criteria, one proposalwould seem to have considerable merit. There are always objections, and forevery proposal there is a faction or two whose interests are betterserved by some other approach, but for the present proposal, these problems seemfairly surmountable. The remaining sections of the paper set out, respectively,the criteria, the proposal, the benefits (first in a perfect world, and then inthe real one), and finally some anticipated objections (and provisionalresponses to them).


At the Brussels meeting on 29 September, it became clear that the followingwere all desirable aims in any proposal, namely that:

  1. it should not ask domain name registries to become trade mark lawyers, or trade mark courts;
  2. it should keep to a minimum the burdens, on domain name registrants, of providing information (or other burdens, such as delays before use);
  3. it should leave intact the benefits of a growing body of trade mark/domain name conflict expertise, from court judgments down to registry experience in managing disputes; but (obviously)
  4. it should reduce the frequency of recourse to courts or other formal dispute resolution processes; while
  5. leaving parties free to make use of those existing (albeit slow and costly) avenues for redress in important cases;
  6. it should be cheap (ie save money overall, compared to the present situation);
  7. it should be fast;
  8. it should be fair to small entities while giving ‘famous mark’ proprietors redress from cyberpiracy;
  9. it should be capable of phased, and flexible, introduction;
  10. it should be compatible with the (orderly) introduction of new gTLDs.


At its most naive but pithiest, the proposal is that:

  • Domain name holders (DNHs) should be obligated, on demand from other aggrieved right holders (ARHs), to include brief signposts on their Web site home pages, to the effect ‘WARNING: this site has no connection with X – click here to go X’.
  • Whenever an ARH invokes its rights for a signpost on a DNH’s site, the DNH’s registry becomes obligated to monitor behaviour of visitors to DNH’s site, by monitoring click throughs to ARH’s site via the sign post, as a percentage of all visitors (the ‘%CT’).

Benefits (in a perfect world)

For a moment, let us assume all domain name holders could be immediatelybound by such an obligation, and would comply with it – huge, unrealisticassumptions but, as will become clear later, replacing these assumptions withthe real facts leaves the proposal intact.

  1. Immediately, the bulk of any traffic genuinely misled and diverted from ARH is restored to it with no or little gain to DNH, but (importantly) no or little harm to DNH – except to the extent that visitors detect bad faith, if it exists.
  2. Over time, the information generated on %CT will indicate whether: (i) DNH is a pirate; or (ii) ARH’s objections are ill founded; or (iii) something in between. This information would be a valuable feed into any formal dispute resolution process.
  3. Rules could be introduced that if the %CT over a representative period were greater than, say, 90%, the Registry could close down DNH’s site, and if % CT were less than, say, 5%, DNH could apply for removal of the signpost.
  4. The procedure’s existence, coupled with 1 and 2 above and increasing use of bookmarks, etc, would make cyberpiracy unprofitable for famous marks and also (importantly) other less famous marks.3
  5. Just as the presence of a mediation clause in a commercial contract smoothes the way for the parties to a dispute to climb down from their ‘show no weakness’ moral high ground into ‘let’s talk’ valley, so the obligations on a party to signpost the other’s site (and its right, not always worth exercising but worth considering when neither party is a pirate, to have a reciprocal signpost on ARH’s site) will tend to lead to jaw jaw rather than war war, and this may even often lead to mutual benefit (win, win).
  6. No-one has been denied recourse to any other remedy.
  7. Imposing the overhead of including a signpost is much less draconian than (i) suspending the site, or (ii) forcing parties into a ‘gateway site’ arrangement.4 (The parties could always progress to a gateway site voluntarily, if that makes sense to them both.) In effect, the commandment is not ‘thou shalt share’ thy exclusive domain name rights, but the much softer ‘thou shalt inform‘.
  8. If it worked, the procedure would render unnecessary (or less important) various other changes contemplated, all with overheads of their own, eg the need for registrants to prove location, to pay in advance, and to pre-emptively register in numerous TLDs5 to prevent piracy and conflict.
  9. In general, all the criteria set out in the preceding section appear to be met.

Benefits (in the real world)

One could not impose the proposal on all domain name holders with immediateeffect. But it could be:

  • included as a term in registry contracts for new applicants
  • observed voluntarily as a code of conduct by one or more entities or business associations (this would not help on cyberpiracy, but would smooth the way on ‘genuine’ disputes)
  • introduced, by legislation, as a non-excludable implied term in all domain name registration contracts
  • imposed by domain name registries at renewal time.

These measures would operate cumulatively to create a standard approach overa fairly short time.

Once imposed, some DNHs perhaps would not comply, but this failure to do sowould (i) be a breach of the relevant registry agreement (eventually entitlingthe registry to cancel the registration), and (ii) look very bad in any latercourt or other hearing. A provision in domain name contracts empowering theregistry to take direct action to enforce the obligation to display a signpostwould also help, at least with small operators who rent Web space from theirISP.


Would famous mark proprietors find this proposal acceptable? Well, it isquite compatible with additional special measures in their favour and is, ingeneral, of use to them in combating piracy. Perhaps they would dislike theobligation to signpost others? But it would always be open to a Kodak to refuseto include a signpost (to ‘Kodak-rip-off’ etc.) – and Kodak would probablybe safe if it just waited to be sued by Kodak-rip-off (who of course neverwould).6 On the other hand, if genuine seekers after ‘Colt’air-conditioning find themselves at a gun-maker’s site, why shouldn’t they betold and helped on their way?

How will the proposal work for domain names which do not point at a Web site,but are used only for (say) e-mail? Not at all, is the answer – but domainnames lacking Web sites generate far less confusion and loss of business. By andlarge, e-mail users do not speculatively aim mail at ‘’– they usually have an exact address, and once one has an exact address itdoes not matter any more whether it reads ‘’ or ‘’.(As a gloss on the proposal, one could empower ARHs to require auto-replies toe-mail incoming to the DNH site, stating ‘nb. no connection with X’).

Wouldn’t some sites groan under the weight of signposts? Well, maybe – butone could provide that any site obliged to show more than, say, six signpostscould instead devote the requisite part of its home page to another page,listing all signposts.

Wouldn’t people cheat by fiddling the %CT figures? Well, the %CT etc. couldbe remotely monitored fairly easily, using standard HTML commands within thesource code for that Web site. (An automated process could also monitor changesto the home page HTML to ensure it was not tampered with. Bona fide updates tohome page HTML would need to be notified to the monitoring registry – butlow-overhead procedures for all this are perfectly feasible).

It may be that there is a problem for people who do not read English andcannot recognise the signposts. A standard logo – dare I suggest ‘WIPO(circular logo) Confusion Buster’ – would perhaps suffice.


If this proposal were thought by a large enough constituency to be, inprinciple, a good idea, then a considerable quantity of flesh would need to beput on the bones, but at this point brevity is preferable to detail. The authorcannot speak for WIPO, but would welcome any comments on this scheme, itsworkability and (if workable) how it could be most effectively introduced andoperated.


  1. Covering domain names in the generic Top Level Domains (‘gTLDs’), ie not extending to country code TLDs (‘ccTLDs’)each of which is administered by or with the sanction of the government of the relevant country.
  2. The practice of registering (often, dozens of) domain names containing famous marks with a view to extracting payment from proprietors of those marks.
  3. It is famous marks proprietors who feel most aggrieved by cyberpiracy – they bear the brunt. Another part of the US Government’s brief to WIPO was to ‘recommend a process for protecting famous marks’. This author’s view is that special rules for famous marks are like ‘gated communities’, or ‘neighbourhood watch’ schemes – they simply change the victims of the crimes, to the next easiest targets. It is therefore in the interests of all ‘non famous’ marks proprietors to find solutions to cyberpiracy which are of benefit to all users. To allow special rules for famous marks is to remove the most vigorous opponents of cyberpirates from the field.
  4. An extra Web site shared by agreement between two or more parties sharing the same name, listing each of their individual sites and inviting visitors to select the one they wanted to visit.
  5. Multiple applications clutter ccTLDs, and just soak up extra capacity from new gTLDs – making their introduction pointless.
  6. Perhaps, to assuage famous mark proprietors and others who think they would be subjected to too many signpost requests, it should be permissible, where there are good grounds for believing the requesting entity to be acting frivolously or in bad faith, to post a (substantial) bond to cover any damages in any suit brought by those denied a signpost, instead of actually including a signpost. However, such a right would significantly reduce the effectiveness of the signpost scheme, to the detriment of all (including famous mark proprietors). It is to be hoped that even large entities would instead choose to put the issue to a %CT test and work from there.