How many sTLD's do we need?

["Greg Krajewski" <gkrajews@xxxxxxxxxxxx>]

10?  100? 1000?  How many do we "get into the A root" before we begin to 
realize that the higher the number, the less "significant" the names would be.  
Domain Names by definition whether they are sTLDs, gTLDs or even ccTLD's were 
created so that we didn't have to remember IP addresses.  There is a reason 
phone numbers are only 7 digits in length.  

I think ICANN needs to rethink these rollouts a little bit better before 
approving the next batch.  These "Proof of Concept" trials are a start, but I 
haven't seen the question answered that needs to be: "How many TLDs do we 
need?".  If you don't answer this question before approving new sTLDs or gTLD's 
it would be like starting work on a building you have no idea how many floors 
will be built.  ICANN should treat the DNS no differently then an architect 
would who has the responsibility to provide a complete drawing of how a 
proposed building will look.  

One other troubling thing about the "big question" not being answered is that 
ICANN has to realize that a new board of directors with their own agenda could 
complicate the namespace "down the road" by approving similiar type names.  
Example:  Let's say the current board of directors approve dot travel, which 
isn't necessarily a bad thing right now.  However ten years from now what if a 
new board of directors decides to approve dot vacation because a "lot of the 
good" namespace in dot travel will undoubtedly be taken by "current" 
stakeholders.  What ICANN needs to do is to not blindly approve new domains.  
They need to have more of a rhyme or a reason as to why a sTLD should or should 
not be approved.  Having new registrars submit applications for new domains is 
not a very good way to do a "Proof of Concept" in my humble opinion.  Remember, 
a building is only as good as the "architectual drawings".

One thing I should mention.  My position on domain name expansion has changed 
dramatically after the 2001 rollout of dot biz/info, et al.  I wanted a lot of 
namespace opened up because most names in .com/.net/org were taken.  Now that 
I've had a chance to take a step back I realize that most people still type in 
the .com address.   Adding new names will not get you noticed unless you have 
the most desireable names in a particular namespace.  As we know with the first 
roll-out the average "stakeholder" came away with names that will most likely 
never get type-ins, and probably will never see the 1st page in a Google search.