It's interesting, read some of it. Points I thought interesting [besides the mention of water payloads:].

"Since range costs are a large component of the total flyaway costs of launching into orbit, they could be potentially reduced by launching outside of the United States. This is the rationale behind the Sealaunch concept, which launches in international waters, but use of foreign ranges has a significant regulatory risk: ITAR (International Traffic in Arms Regulation). "
&
Launch Facility Costs 2,000,000 [for launch vehicle with 1000 lb payload to LEO] & 10,000,000 [for launch vehicle with 100,000 lb payload to LEO].

Now, I think this supports the value of using ELV*. And I think it supports what posted below. Which in summary is if were to launch a rocket once per week or less which had 4000 lb payload you could get to point of $500 per lb for water payloads (though I using tonnes and kg) [http://www.habitablezone.com/space/messages/376131.html ].
This is assuming you can launch quickly and week after week, and assuming that even if you were using a govt range, you should able to get a reduction in range costs.
Now, if you using a mothership and runway [and friendly local govt], it would seem you could lower these range costs. Of course, if you were to use my pipelaunch idea, your Launch Facility costs could easily be less than 2 million per launch and you could launch over open ocean. Sea Launch is an international operation whereas using a pipelaucher would necessarily involve international companies [though I don't know the law, so maybe it helps as far as ITAR is concerned [because the laws can be screwy] by involving international companies.

*The paper did not discuss a rocket with 2000, or 4000 lb payload to LEO but it's reasonable that the major costs wouldn't significantly increased for these bigger rockets.