Some days ago the outstanding Come From Beyond published an article addressing Economic Clustering and (indirectly) Network Bound Proof of Work as his futuristic vision of how IOTA will work in some years.
This is always good brain food and I spent up to 30 hs. doing some research on what this Network Bound PoW was with the help of several IF/IEN friends.
To put it shortly, Come From Beyond piece addresses a solution to the scalability issue on IOTA by presenting an Economic Clustering model in which you have the current Tangle, referred to as Cluster0, and other interconnected Clusters (subtangles) you can connect to while moving geographically. Clusters are interconnected as rings and as you move far away from Cluster0 you start “jumping” to other Clusters that are not connected to the main one.
But what about my iotas on this unconnected clusters? According to CFB you would have to resend yourself the iotas (which I guess will be automated) from the cluster you are in to the next one as you move forward if you want to have them available once you arrive to your destiny.
You don’t want to do this? No problem, you will probably be able to use services that will charge you a small fee to connect the cluster you are in with the Cluster0.
Mmhh. Ok. Now, how will each of this clusters manage users or small IoT devices transactions? If they are not connected to the main cluster, wouldn’t this make them vulnerable to someone that has 51% of that cluster hashpower?
This was the first thing that came to my mind when trying to digest this concept so I just asked CFB. Somehow I knew the answer:
“EC and network-bound PoW will solve that. The details will be revealed later.”
After this I contacted Come From Beyond and other IF/IEN members I know to start building something around this new not-yet-documented type of proof of work. Results are not as good as I would like and they were summarized by me on a Reddit post.
What to invade next?
To be honest that Medium post on EC caught me unexpected. I was still digesttrying to break down Qubic and all its consecuenses and suddenly I found this “now you see it, now you don’t” effect product of a brilliant mind that is just far ahead from us.
The thing is that, as CFB stated on his Twitter status, Qubic will solve smart contracts and that is now “fixed”. Next challenge is scalability and that explains the EC post direction.
Personally, I think that having such a brilliant mind on board is a great plus but I also think that IOTA needs to focus strongly on the development of the basis: we need to improve network performance by working on better Nodes, we need the datamarket, to start working on Qubic and all the things it implies (such as Abra, IOTA’s own programming language).
So, while I love to see how we are always some steps ahead in terms of what limitations the project is going to find on the way, I also think we need to keep our feet in Cluster0land for a while. There is a lot yet to be done here.