I was hoping to see some discussion about hashgraph on this topic, but ill just leave the concept here for people to find and come back to discuss further if others find it interesting.
Hashgraph is an interesting consensus protocol that has been shown to yield high throughput. However this has taken place in a static and private testing environment, it does not mirror real usage. It may be fast, fair and secure within the limited environment within which it operates,
however, if and when used in a public setting, Hashgraph will face the same issues that other public blockchains are facing today and may not be able to maintain its security and performance.Scalability is still an open problem for public blockchains - a much more developed technology. Think of POS for Ethereum, NEO with dBFT, EOS with dPoS, or the concept of sharding. All the different solutions that have been proposed by each ecosystem are interesting, but none has been widely adopted and certainly they are far from perfect.
Also we should consider what we are trying to scale with blockchain? Users, network, or the sheer number of transactions able to take place simultaneously? If we were to design a network, like Hashgraph, that could provide infrastructure for thousands of transactions simultaneously, would that be the end of scalability? Where does the limit lie when we say a solution is scalabe? Does 24k/second like VisaNet count as scalable? If so have we managed to reach it? Hashgraph, as it stands, only scales in the number of transactions, but what about the number of nodes/size of network, or the number of users?
The road to adoption with any new technology is tough. Blockchain is far from obsolete, Hashgraph is just getting started.
