hmm. what about it makes it more risk than arbitrary domains?
I'm assuming we're talking about things like:
http://search/
http://weather/I'm assuming we're talking about things like:
http://search/
http://weather/I can imagine a lot of scenarios where people could easily be tricked into accessing the wrong URI if these dotless domains become common place.
The same applies to companies whose intranets already use some common words as internal services.