Let's face it, the new TLD "testbed" is dead, if it ever existed at all. Three years after the ICANN Board used an astonishingly ad hoc process to go "shopping" for .aero, .biz, .coop, .info, .museum, .name and .pro, we're no closer to selecting new ones than we were then. Although the community of Internet users was promised that these initial seven selections were simply selections for a "testbed," we now have little data to make any evaluation about them and no clear ideas about how to test their success or failure. Had this been a real "test," the experiment would have been designed before the selections were made, and the Board's decisions would have been guided by the variables necessary to conduct the test.

Trying to get community agreement about how to evaluate the testbed now, however, is an impossible task. Everyone involved has their own agenda and wants to spin the experiences of the last three years their own way. Trademark owners, insistent on owning "their names" in every conceivable TLD, have been ruthlessly bled by the registries through sunrise periods. Even TLDs like .pro that plainly pose little threat to create consumer confusion have sought to make easy and early profits off the trademark bar. Is it any wonder that the intellectual property and business constituencies aren't eager for new TLDs?

The new registries aren't eager to move the process along either. When asked in 2002 whether ICANN ought to create additional TLDs, Afilias' Roland LaPlante cautioned for restraint. The reason? LaPlante said: "[W]e must allow adequate time... [so] that their introduction does not negatively crowd or disrupt the current registration marketplace by offsetting the existing supply versus demand structure." If you need a translation, Mr. LaPlante was worried about losing market share to new competitors.

With all of the prospective TLD registries kept firmly outside the formal ICANN processes -- remember prospective registries were denied membership in the registry constituency -- who inside of ICANN has any incentive to move this process forward? That's why we're where we are three full years after MdR 2000. Perhaps only the shame of having done virtually nothing on this front for three years will finally move ICANN forward.

Let's put the testbed to bed and start anew. Rather than trying to evaluate the effects of new TLDs in a vacuum, let's just do it. Announce a formal call for applications. See what comes in. Pick the best 10 or 20 and queue them up for launch 2 per quarter for the next two years. If an application is selected, then access a fee. Just do it.