This is a good system. After having had time to study its implementation, I have found the main flaw to be the lack of a funding mechanism to support network infrastructure and operation costs.
This can be cured by adding a proof of stake layer to reward network supporters.
Miners are too busy trying to compete for profit that they have no incentive or time to maintain the integrity of the network and fund continued research and development.
This won't work because it would incentivize malicious users to spin up node farms in order to artificially bump inflation rate.
A secondary approach might be to back off the diff adjustment since hashrate activity is tied to market price. Reduce the back pressure and allow inflation to heighten while imposing less difficulty.
The problem with the secondary approach is, we don't have an accurate way of predicting ASIC advancement.
Quite interesting project. I am curator of a micro-currency in Germany called Rheingold.
It is based on cash. Therefore the problem of "trust" is not solved. However, we do not have a central bank giving money as credit, but here, every single issuer of his own money gets it printed himself. So Rheingold is rather a group of many de-central-banks. Therefore the seignorage everybody gains himself.
And our money is debt-free.
I would like to promote your project on our blog http://rheingoldblog.wordpress.com. Maybe you could even become Rheingolder yourself with your own cashbill. Since we pretend to be an art-project, the whole project will be shown in a museum in future. We already have more than 1.000 members, mostly in Germany. Our website http://www.rheingoldregio.de
Any idea about a cooperation?