What additional details/update do you have for the community regarding your WIP approach?
As you can see on the image below we actually didn't chose the WIP approach, we used a recursive-ZK-Snakrs which is basically a linearisation of the ZK-Snakrs making it much faster, however it require more computation power but our simulations and testes revealed that it's actually widely feasible.
https://image.noelshack.com/fichiers/2018/03/6/1516471445-finalresultsazulik.jpgI would like to ask about hash based signature XMSS, if you were considering pros and cons to add XMSS to the Azulik code and to provide your explanation
For the hash used, we will stay with Blake2, after reading JP Aumasson books it would convince everyone that it's currently one of the most efficient hash function. For Azulik the hash algorithm isn't used for proof of work and the hash chosen doesn't really bring upside whereas using hash based signature XMSS would bring complexity and probably instability. (To be honest we aren't very familiar with XMSS).
