Thread by @hus_qy

@bascule @ercwl 1\ Our approach is indeed somewhat similar to the one taken by Avalanche but this is also not very surprising considering that they are using the same data structure as we do (which was by the way not referenced in their paper) and are trying to solve the same problems.

@bascule @ercwl 2\ That being said it is only similar at first glance. If you look a bit closer, it is quite different as our approach is not "metastable" and always converges to a decision. Especially the Cellular Automata approach with its unique "immune system" is very different.

@bascule @ercwl \3 It is not just much faster (due to the fact that you don't need to "query" other nodes) but also much more secure (due to the additional "immune system").

In addition IOTA does only vote on conflicts, whereas avalanche votes on every single transaction.

@bascule @ercwl \4 So TL;DR: Our voting mechanism is different, our consensus properties are different, the sybil protection mechanism is different, the additional security mechanisms are different.

The only thing that is left is that we both use a dag (which IOTA introduced first).