Quite an interesting project. It will be necessary to study it in more detail.
Thanks. Please do, any questions or for an invite to the slack then PM me.
UPDATE:
Very exciting few days for me, the team of new dev recruits and the 5 or so alpha testers grappling with a new alpha testnet going live across multiple platforms and hardware devices simultaneously. We have had up to 16 nodes spread across the globe with up to 12 staking simultaneously for a significant number of blocks.
Lots of bug fixes and emergency commits and restarts, some simple and some extremely nuanced.
We had a massive halt crossing the first epoch transition from 0 to 1 at 10000 blocks but that is now sorted. The raspberry pi users reported significant slowdown as we passed 11000 blocks and I tracked the problem down to some inefficient disk access for the chain. Numerous improvements in validation and state checking of the chain as well as node startup performance improvements.
Several key issues remain which need sorting over the coming month:
1) disaster recovery for the POS protocol..
- need more reliable missed block logic
- consensus failure code needs to be improved and built out
- fork detection code needs to be built out and updated to include reorgs capability.
2) complete integration of fees and weighted block reward based upon stake - (requires a chain reset so leaving for now).
3) i have identified a potential MITM attack which could be used to spoof fake messages with our POS protocol. To eliminate this possibility it will be necessary to pass an additional message prior to reveal_one and reveal_two in our protocol which are hash(reveal_one) and hash(reveal_two). This will require a small amount of further time to be added to the existing block time (~40s).
4) addition of a 'key' or bit mask to the iterative hash chains using blockchain seed to reduce security requirements of the cryptographic hash function (requires a chain reset and so will be merged before the next major planned restart).
But the POS protocol is working and following consensus - with our signed iterative hash chains from each stake validator and the PRF sequence from the HMAC_DRBG and entropy from the blockchain system functioning exactly as anticipated.
On a higher level we need to decide our starting network parameters (block time (less puts more pressure on nodes with poor latency, higher allows greater usage of the network which is desirable but reduces performance), initial emission at the genesis block, consensus rules, flexible blocksize parameters etc.
We are also gearing up to announce some form of genesis emission sale. I am a big fan of ethereum so we may issue ERC20 tokens on ethereum although this isn't crystallised yet.
As usual we have a very busy slack, and if you wish to discuss the project, the protocol, the signatures or join as a developer or tester then PM me for invite.
And the website with details of white paper, github and a link to our email list is
http://theqrl.orgInca