this is an interesting idea...
as i see it, any miners working on this new system would not use any advanced features, making them still valid bitcoin blocks, just without any special transfers included in the block.
optionally, all bitcoins before block X will be valid paracoins, but after that any block that used advanced signatures is invalid.
Not exactly...at least in my current hazy visualization.
Only 'bitcoins' (actually addresses) that meet a criteria would be 'paracoins'. Generating slightly special ones with vanity-gen would be an easy way to do this. The rational would be to keep the circulation reasonable since the entire goal of the project is to focus on being light-weight and thus conducive to extreme decentralization.
the problem is that even if bitcoin blocks that used advanced features were allowed to exist in a chain, any new transfers that used advanced features would basically be in purgatory until a normal bitcoin miner manages to find a block.
if blocks with multi-sig transfers and whatnot aren't allowed, that would probably cut the blockchain down a whole lot, meaning the new chain wouldn't be accepted at all because it's so short, but that might be made up for by lower memory requirements and lower block sizes, which should increase hash power just from the fact that you don't have to look at as much data to start hashing.
I'm negative about the 'advanced features' being developed within Bitcoin (but warming to some of them.)
Remember there would be two distinct phases. 'parasitic' and (maybe) 'stand-alone'. In 'parasitic', the Bitcoin blockchain is the source of truth. Addresses which were both certified (e.g., '1para1...') and also legally assigned value within the 'paracoin blockchain' would take that value reading from the Bitcoin blockchain.
In 'parasitic' mode, the 'paracoin miners' would mostly be analyzing the Bitcoin blockchain and rolling with whatever punches the Bitcoin system threw it's way. 'paracoin miners' would be mining in a formal sense as well, but only just for practice and development reasons so they would be ready to switch to 'stand-alone' mode if need be.
I think that sha256 mining is conceptually broken for a true distributed system. A big goal of the project would be as a platform to explore other forms of mining which are more conducive to decentralization.
has any progress actually been made? as far as i can see, all it would take is an older version of the client; it wouldn't know what the new transactions meant and would ignore them, and as long as it's a version new enough to have valid blocks they'll be accepted if the chain gets long enough.
No progress. I've not done jack shit on it. From the start I mostly I just wanted to throw the idea out there.
Since I cooked up the idea, I've become increasingly paranoid about the trustworthiness of hardware. I think that a logical first step for a robust solution is to build a foundation on true open-source hardware. All the way down to the silicon level ideally.
My current interests are hardware related, and not even necessarily driven by crypto-currencies. My main interest in Bitcoin is in planning a significant exit to be honest.
