Among those looking for answers at ICANN’s March meeting in Singapore will be new generic top-level domain (gTLD) applicants who haven’t yet found a path to delegation due to potential name collisions. No longer is the issue of name collisions merely technical. Over the course of almost a year, it has evolved into a potential business and financial roadblock.
Name collision refers to the confusion that may occur when a new gTLD string exactly matches an existing string used on an internal network. So, as new gTLDs enter the root zone, they can potentially “collide” with existing names. Name collision is a problem in current gTLDs too, but has taken on greater significance because of the exponential number and types of strings involved in the New gTLD Program.
The issue first took hold in late March 2013 when Verisign produced an incisive report (pdf) laying out a number of possible security concerns. At first, the report seemed quite damning. Upon closer examination, it became clear that ICANN’s Security and Stability Advisory Committee had already addressed many of Verisign’s concerns. Nevertheless a community-wide conversation was launched and the search for solutions began.
ICANN hired an independent agency to audit the root zone through the “Day in the Life of the Internet” (DITL) project to gain insight into the breadth and depth of potential collisions. The DITL data lists every domain name queried at the root zone over the course of 48 hours each year for a number of years. The domain names ending in new gTLDs were pulled from this list to determine which strings got the most hits. But the audit offered scant data aside from the simple list of colliding domains. These lists are the basis of the mitigation plan in place today.
Each path forward, however, has far-reaching and highly variable implications.
The majority of strings will follow the “alternative path to delegation,” which requires applicants to block every domain on their string’s list of DITL data before they launch. Once ICANN has the time and resources to investigate the blocked names, it might unblock those it thinks will cause no trouble. This path is the easiest and quickest for applicants, especially those with fewer than 100 names on their list.
For those strings with hundreds of thousands of names on their list, the solution is more complicated since inevitably the list will contain high value terms. So what has been a technical issue for some applicants becomes a strategic issue for others.
If corporate applicants cannot register names such as help.BRAND or buy.BRAND for an unknown period of time, their new gTLD business plans could be seriously impacted. How long will these names be blacklisted? Could they be blacklisted forever? What do competitors’ lists look like? ICANN’s timeline for digging deeper into the DITL data and creating thorough recommendations for each term on a string’s list could be critical to many gTLD marketing and use strategies.
Finally, applicants for 25 strings, including .BOX, .CASA and .FAMILY, are unable to use the alternative path to delegation not because they have lengthy lists of domains to block but because their lists are dynamic, with a high rate of change from one year of DITL data to the next. This is the group of applicants that must wait until the next ICANN public meeting to find out how ICANN will go about developing a mitigation plan and how long it will take.
While it is unlikely this group’s Name Collision mitigation will stall applications for more than a few months, once again, a technical issue has become something much bigger with serious financial and strategic implications.