First, there is the 2000 round applicant for "web", and the unanswered (in the American courts, a venue that may matter) question of whether ICANN's 2000 offer may be unilaterally recinded, by ICANN, after having received a timely and complete acceptence by the applicant. For those who hold that ICANN may receive timely and complete applications, and subsequently unilaterally recind past offer(s), what guarantee do the 2012 applicants have that for reaons as unrelated to their application as the 7-10 limit (which I had a part in developing) was to each of the 40+ 2000 applicants that their applications will not be similarly rejected? If it is proper to credit Chris Ambler $45k against any future application for any string and sollicit a second or subsequent applications for "web" and attempt to monitize a plurality of applications by auction, then it may be proper for ICANN (see "American courts", supra) to recind each such offer until its auction reserve value for "web" is met. Of course, Chris Ambler is not the only 2000 round applicant who's timely and complete application has been treated by ICANN as a loan repayed with out interest, and only in credit against future applications. CORE too is similarly situated, having applied for "web", as is the World Health Organization, having applied for "health", etc. So, where a string did not appear in a timely and complete application in a prior round, and appears in two or more such applications in the current round, then it is possible (at ICANN's sole discression, see "American cours", supra) that ICANN will be the beneficiary of an auction outcome which meets ICANN's unilaterally pre-determined auction reserve price. Second, assuming there is "auction revenue", is novel infrastructure the best, or even better, of the choices, in any region? My experience is no. I worked to get fuel to the NAP at Boutillier Hill in the days after the Hatian Earthquake. Others worked to get pallets of VSAT and WIFI kit to isolated NGOs. The NAP at Boutillier Hill was, after the destruction of the submarine cable landing, via a microwave link to the Dominican Republic, the sole point of connectivity between Haiti and the rest of the net. It was also, after repairs on the land line to the airfield where the Government relocated to an undamaged police building, on day 2, the sole source of Government data network communications. The real issue was diesel. Several times the generator was down to "last hours" and (mercifully new) batteries ran the net. With Becky's help, State was prompted to direct the military rescue mission to resupply the NAP hours before fuel and batteries failed, and the first week of government and data-supported rescue (Vint got a fresh set of Google Earth datasets sent in a disk to the NAP) was survived. The second week was simply a slightly less hectic repeat of the first, with food for the NAP staff, and visas for the families in place of barrels of diesel as the critical issue. In sum, each mile of wireline infrastructure repaired "turned on" many hundreds of residences and businesses, whereas the sprinkling of newtech pallets of VSAT+WIFI kit serviced only a fixed point and nothing else. The recovery effort goal was not "turn on" the NGOs in their scattered sites, but "turn on" as much of the entire country as possible, starting with the government function of recovery coordination. Sending stuff feels good, and it "solves" an ICANN process problem of making disposal by auction (no policy but market policy) appear to be a good thing (TM). However, wireline infrastructure is generally deployed along rights of way of rail, road, and sealane infrastructure, and surface transport cost (measured in time and dollars) in Africa is higher than in Europe and North America. The wisdom of separating the economic development associated with data delivery from the economic development associated with the delivery of tangible goods and services is open to question. Eric