Re: [bc-gnso] Finding Common Ground Between Markholders and Legitimate Domain Registrants

[George Kirikos <icann@xxxxxxxx>]

Hi Rick,

On Mon, Jul 6, 2009 at 11:00 PM, Rick Anderson wrote:
> What I see wrong in this notion - at least as described here - is that it 
> probably encourages tasting, squatting, speculation as much as it assists 
> TMholders.  The unintended effect of subsidizing these activities is not a 
> great plan.

Thanks for the feedback. One method to refine the concept further is
to truly limit things to defensive registrations (as opposed to
speculative registrations at lower cost) through a link to an active
"base" domain name (one that does resolve). For example, the domain
typo generator at DomainTools.com spits out a number of matches for
"Verizon":

http://www.domaintools.com/domain-typo/?q=verizon&mode=reg&status=b&rules%5B%5D=qwerty&rules%5B%5D=swap&rules%5B%5D=sticky&rules%5B%5D=look

Let's say that the "base" domain name is declared to be Verizon.com.
Then if Verizon wanted to own verizoln.com or verizom.com, but the
traffic from those domains wasn't worth $7/yr to Verizon (i.e. it
doesn't "pay" for them resolve), they could pay say $3/yr to register
them but have no nameservers, at the same time linking it to
Verizon.com. They could do the same for domains in other TLDs,
declaring them "defensive registrations" that all link to one base
domain that does resolve.

One could develop an algorithm to test whether a domain that is
declared as "defensive" is similar enough to that base domain name to
qualify (e.g. a certain number of common characters, common typos like
wwwdomain.com, etc.). An algorithm probably wouldn't capture 100% of
defensive registrations, but it could probably reduce costs for a
healthy fraction of them.

There could also be a function to list all defensive registrations
(with no nameservers) for a given base domain, to make abusers easier
to bring to justice. For example, let's say someone other than Disney
did own wwwdisney.com and used that as their active "base" domain for
speculative but low traffic domains (which didn't generate $7/yr worth
of traffic) such as wwwdisney.org. A markholder would be able to more
easily capture the entire set of typos that didn't resolve (and thus
were registered under the lower cost system) in one action because of
that linkage.

> As well, what actually makes sense with these secondary TM registrations is 
> to point them at the primary site (rather than to leave them to non-resolve). 
>  That's a better user experience, and if the holder has to go to the effort 
> of registering them (a bigger cost really than the reg cost), whatever 
> traffic they may generate may as well find its destination.

Sometimes yes, sometimes no. If the domain doesn't generate $7/yr
worth of traffic, a markholder might still keep the domain registered
in order to avoid facing the UDRP and legal costs of $5,000+ if the
domain is abused by someone else. If these marginal names could face
lower carrying costs (say $3/yr instead of $7/yr), that cost savings
could be dramatic, thousands or even tens of thousands of dollars per
year.

Registry operators might not be happy by the loss of "fully priced"
defensive registration fees that they're used to currently, but that's
not a suitable business model to begin with. Depending on the
elasticity of demand, ironically registries might even actually
increase the number and total revenues from defensive registrations,
as the lower price for domains deemed "defensive" would actually
increase the total number registered and possibly the total
profitability for the registry.

Sincerely,

George Kirikos
416-588-0269
http://www.leap.com/