[wg-c] nine principles for domain names

[Jonathan Weinberg <weinberg@mail.msen.com>]

Kathy --

	Philip Sheppard mentioned in a recent post to WG-C that you and he had
collaborated on a set of nine priciples / criteria for domain names.  I
want to share with you part of a longish note that I sent him a few months
ago, somewhat skeptically commenting on an earlier -- and shorter --
version of the principles; I'd be interested in your thoughts.

Jon

---------------

>Philip --
>
>	I think most of the principles are controversial to some degree, and I
surely agree that they are worth discussing on the list.  I'll try to
explain some of my own thoughts about the principles, taking them out of
sequence in order to convey my thoughts more clearly.
>
>"Findability — the gTLD should assist a net user to find a particular
domain name."
>
>"Semantics — the gTLD should be meaningful in English or a language with a
significant number of net users."
>
>"Differentiation — a gTLD should not confuse net users and so should be
differentiated from all other gTLDs."
>
>	I think that everyone on the list subscribes to a weak version of the
findability principle, in that nobody would support a rule under which
domain names were required to be meaningless alphanumeric sequences.  For
this reason, I don't think you'll get much argument on the semantics
principle.  The controversy here, I think, comes with regard to the
question of how important findability should be in structuring the name
space, and how it stacks up against other desiderata.  Some list members
believe that findability should be the primary concern; others take the
view that the DNS is by its nature ill-suited to be used as a directory
system, and that we should not try to structure the DNS to replace search
engines and yet-to-be-developed directory tools.
>
>	Specifically: If one believes in the pre-eminence of the findability
principle, then presumably one would not want ICANN to create additional
general-purpose TLDs.  Further, one would want differentiation, so that one
would not want to see more than one TLD serving any particular market.
(That is, one would not want the name space to include both .sports and
.athletics, b/c the user wouldn't automatically know in which one he should
look for a particular sports-related domain.)  Yet I don't think I agree
with either of those results.  It seems to me that it would be useful to
have new general-purpose TLDs, to provide competition to .com.  Right now,
.com stands astride the name space as the dominant commercial TLD.  It is
the 500- pound gorilla of TLDs, and domain names in .com have tremendous
(artificial) market value; companies that currently have a domain name in
the form of <www.companyname.com> have an extremely important marketing and
name-recognition tool.  They have an advantage over all other companies
that do not have addresses in that form, because they are the ones that
consumers, surfing the Net, will be able to find most easily.  Adding a set
of limited-purpose TLDs, it seems to me, would not change that: there would
remain .com (and to a lesser extent .net and .org) atop the TLD pyramid,
and a mass of special-purpose TLDs below.
>
>	Alternative *general-purpose* top-level domains, by contrast, could
provide effective competition to .com.  With the addition of new
general-purpose TLDs, shopping.com might face competition from shopping.biz
and shopping.store.  Those businesses will have to compete based on price,
quality and service, rather than on the happenstance of which company
locked up the most desirable domain name first.  This would more nearly
level the playing field for individuals and businesses seeking attractive
domain names, and would diminish the ability of a minority of e- businesses
to collect rents based simply on their registration of good names in the
"best" TLD.
>
>	With regard to limited-purpose TLDs as well, I think competition concerns
cut in the opposite direction from differentiation concerns.  The existence
of many competing TLD registries, without regard to differentiation, will
diminish the market power that any particular TLD will exercise.  Users who
are unimpressed with the performance of one registry can instead acquire a
new domain name in a different top-level domain, run by a different
registry.  Imagine, thus, that the name space contains a single registry
called .sports.  Without more, this registry has market power corresponding
to the degree that it is a better TLD for certain registrants than any
other, and it can use that market power to extract inefficient rents.  If,
on the other hand, there are many TLDs, and relatively free entry into the
TLD namespace, then the market power of the .sports registry can be
constrained by the creation of .athletics.  That's inconsistent with
differentiation, but it think it redounds to the benefit of domain-name
registrants.
>
>	(As an aside, I don't think that the existence of TLDs whose target
registrant population is substantially similar to that of other TLDs is
"confusing" to Net users.)
>
>"Trust — the gTLD should give the net user confidence that what the name
purports to stand for is actually the case."
>
>	This is an interesting one; it applies, I assume, to limited-purpose
TLDs, since general- purpose TLDs don't purport to stand for much.  There
are two ways, it seems to me, to set up a limited-purpose TLD.  One way to
have some policing body checking credentials, to ensure that each SLD
applicant in fact belongs in the TLD.  This has the advantage of creating
"trust," in the sense that you define it above.  On the other hand, it can
also have disadvantages: for example, to give a trade association of
industry members the authority to say whether an applicant is really a bona
fide industry member and entitled to register in the TLD, creates the
possibility of undesirable cartel behavior.  Further, this approach may not
be necessary.  In the US, for example, there is almost no policing of
business' choices regarding where in the Yellow Pages their telephone
listing will run, and yet consumers nonetheless find the Yellow Pages quite
useful — an auto repair shop has no incentive to buy a listing under
"flowers."  I expect that in an expanded name space, the policing model
will be appropriate for some TLDs, and the unpoliced model for others.
>
> [rest snipped]