The August MimbleWimble update from David Burkett has just been published, covering the latest work done on the project in the last month, along with discussing future plans and the testnet release at the end of September. So without further ado, over to David:
I apologize in advance for the much-briefer-than-usual update. I’ve been busy with tying up loose ends, which are important & necessary, but not very exciting to write about.
As planned, I was able to wrap up the Initial State Download, in particular the validation of the MWSTATE messages. There are still a number of ways where malicious peers can use these new messages to DoS nodes. I’m not yet guarding against all of those attacks, but there will be plenty of time to focus on that after the testnet is released.
I also took advantage of the opportunity to go back and clean up some code & strengthen some of the more fragile logic, particularly around management of the UTXO set (the collection of unspent coins).
Everything is on track for a testnet release at the end of this month. I’m still figuring out a few final details, but I’ll share more information later this month about what the testnet will look like, and how everyone can get involved. I’m hoping to have a basic CLI or JSON-RPC option for advanced users to experiment with while I transition to focusing on adding non-interactive transaction support.
For those interested, I recently gave a talk about future directions in Mimblewimble which highlighted the weaknesses of vanilla Mimblewimble, and what we can do/are doing to try to solve those. Here’s a recording of the talk: https://www.youtube.com/watch?v=x5vJzp1ehaI&feature=youtu.be 16