So there are two possible ways to build this out and make it work. In one, we have the merchants running point-of-sale terminals acting as nodes to extend the network through pre-defined hardware that we either donate or subsidize for the merchant. This allows us to control what hardware is running the network, get merchants on board by demoing the hardware and setting it up for them, and makes the network scalable in an easy and straightforward way.

There is a secondary way to do it in which individual users continue to be the sole forgers on the network. In this method, we would build a payment processing system like BitPay that we would make available as a point-of-sale application for android and iPad that could be run from cheap android tablets and allow merchants to accept XCR without actually running any hardware on the network. This version requires PoT motivation for individual users to run nodes, is much more difficult to manage and scale, and has a different set of complications. It would however, build a core community of those users running nodes. This was the original system we had decided to build. Our troubles and hurdles in building it are already very public and I think you understand exactly what happened.

In the second method, if you don't require forgers to compete with established merchant nodes, you don't end up cannibalizing your user run nodes. That doesn't really matter in the long run because you will always still hit a saturation point at which new nodes become unprofitable. Depending on network volume, price valuation, and number of nodes, you still end up with the same problem as you mentioned. In this sense, I think you are right, there is no way for you to allow new users to come on board and earn XCR in any worthwhile amount without self saturating in a "forging" type system.

I think our goal in the user run network wasn't to allow everyone to set one up, start forging, and get rich, but rather to ensure that there was enough motivation to build a core team of "networkers" who would be running the network and scaling it as the price and volume went up. I acknowledge that everyone won't be able to play this game, but the number could be high enough to meet the size of the enthusiast marketplace when we get the transaction volume and price up. I don't think we can properly predict or do analysis right now on exactly what those numbers will look like because we have no idea what the valuation could eventually be or how much interest there will be in running a node. If you have ideas on how to do a predictive model and come up with potential numbers I would be interested in seeing what the breaking points would be at certain volume / valuation levels (in regards to how many nodes the network could support profitably).

At the end of the day I agree with you that doing this project right may require thinking outside the box and bucking the trend. It may also require making the hard choice to go against what would probably be the communities wishes (i.e. that they still run the network) and just focusing on merchants and building a better payments system. We often say (or at least, I do) that our goal isn't to dethrone Bitcoin or NXT, but rather that our competition is Paypal, Visa, Mastercard, or Stripe. To compete against these guys we need low transaction fees, but we also need a simple, fast, and secure (plus stable) network. This needs to be our primary motivating factor.

I have to run for now but I would love to continue the discussion. I value this type of exchange.