[Someone with a clue one said...*grin*] Hrmmmm, you have a major point. Suppose only routers costing 6 figures or more could do CIDR, while routers costing, say, four figures could do RIPv1? Should we give someone a large block to get "up and running" until they could afford the much more expensive routers? (stuff deleted) See, I just don't think "Address Space" should be a barrier to entry. It's not like ARIN *owns* the space, they're just the caretaker for the IPv4 space in America for the "community". As a member of the community, I want them to be fugal to avoid waste. But unusually large (and temporary) allocations to allow a company to enter the market - a company that would otherwise have to perform Herculean tasks, tasks none of the rest of us have ever had to perform, or maybe even tasks impossible with the State of the Art - is not, IMHO, waste. --- What about the case of an NSP looking to obtain a block to begin business with? A case where the parent company (via direct ownership and mergers) holds a couple of /16 allocations across a nationwide LAN infrastructure, and at the same time is not able to deliver to a spinoff arm of the company a 'clean' chunk of space to bring up a very large Internet Backbone. At the same time, the parent company has 0 SWIP documentation of the existing /16s and is still about 18 months away from a total migration to 10.x.x.x space internally. What is ARINs job at this point? To force the spinoff to document the IP allocations within the Sister company before they allow you to do business, or based on the current and future deployment (and CIDR capabilities) allocate sufficient address space to allow the business to launch and grow.