I'm talking about the protocol. You know, that thing that's actually kind of difficult to change once a blockchain gets established? What have you done that can't just be released as a modified client for the Bitcoin blockchain?
All of the changes can't be released as a modified client for the Bitcoin blockchain; doing so would cause a fork in the blockchain which would be a lot more disruptive than simply starting a new one. In particular, the problem with forking the blockchain is that transactions can be carried from one side to the other and this can be used to double-spend bitcoins - once on each side of the fork - and obtain bitcoins that are valid on both sides, guaranteeing that exchanges and e-wallets will be screwed over.
- Difficulty algorithm aims for 3 minute blocks instead of 10 minute blocks. Achieved by changing one constant. Of questionable merit, due to scalability concerns.
Sounds like a good reason to implement it on a new, smaller network rather than Bitcoin then.
- Difficulty retargets more often. Achieved by changing one constant. Essential for an underdog blockchain to survive, but not really relevant for Bitcoin.
- Difficulty increases limited to +10% per retarget. Achieved by changing one constant. Essentially an extra subsidy to miners, not relevant for Bitcoin.
Not relevant to Bitcoin
now; it's an open question as to whether it will be in the future. Also, the limited difficulty increases aren't just a subsidy to miners, they're also necessary to stop the difficulty from overshooting wildly. (For some reason they're not just implemented as a simple change to one constant either; the way difficulty increases are calculated is substantially modified.)
- Block reward changed from 50 to 32. Achieved by changing one constant. Irrelevant except for the psychological factor, see also: Ixcoin.
It was an arbitrary choice when Satoshi made it for Bitcoin and it's still an arbitrary choice now.
(Also, am I the only person that's more worried about the security implications of their changes to difficulty targetting than anything else?)