Why Hashgraph will never replace Blockchain

Many moons ago (last week), I was a HUGE Hashgraph fanboy. I thought it was surely the future, knocking out blockchain and it’s weak, unscalable and insecure infrastructure. Then I learned something really daunting… and now I’m convinced that Hashgraph is absolutely not a viable Bitcoin replacement — at least not if you see a truly decentralized future.

To understand why this is, I think it’s helpful to get a good grasp of what sort of problems Bitcoin and the blockchain solve first, and why Hashgraph is seen by many as a better replacement. Then we can talk about why Hashgraph is actually worse than blockchain.

The Byzantine Generals’ Problem

The Byzantine Generals’ Problem is an agreement problem that was coined after the Byzantine army. This problem happens when a group of generals that each command a portion of the Byzantine army surround a city of 4 walls, with one group of soldiers at each wall. The groups need to be able to coordinate either a full-on attack or a full-off retreat. The important thing is that every general agrees on a common decision. Any confusion will mess the whole thing up.

What could cause a problem is when you have a bad general that tells two groups to retreat and tells the other two groups to attack, knowing that if all four groups aren’t doing the same thing at once then they will fail and lose the battle.

The same sort of problem can be caused if you have a bad messenger, as opposed to a bad general. How do you know the messenger isn’t being stopped half way and bribed by a traitor to give a group the wrong information?

Here’s another way to describe it: imagine we’re in 1951 before wireless cellphones, and there are 3 football stadiums right next to each other with games playing simultaneously. If I’m in stadium 1 and I want to know what’s going on in stadium 3, I’d have to ask someone to run over there and check the score for me. I could pay him $5 to promise he’ll give me an accurate score, but if someone else stops him on his way there and bribes him with $10 to give me the wrong score, or if the runner himself has ulterior motives, then I’m going to get the wrong information.

Now let’s take that example and put it in the context of banks and your grandma’s money. Then it becomes obvious really quick why these are extremely bad situations to be in. It’s a haven for corruption, and the only solution to the problem is technology.

The solution

Bitcoin is appealing to people because it solves the Byzantine Generals’ Problem using blockchain technology. It’s wide open and no one has control over it. Plus, everyone can see exactly how it works because it gives full transparency.

Hashgraph on the other hand, is completely centralized. Many people think it’s decentralized because it solves the Byzantine fault tolerant just like Bitcoin, and in fact it does this much better than Bitcoin does — this is all very true, and it is the reason I fell in love with Hashgraph at first. But everyone seems to be forgetting one important detail: Hashgraph has a patent, and is NOT open source.

Here are the related patents I could find:

I find it interesting that it’s disguised as a “distributed database”. I mean, that’s what it is and that’s how you hear the Hashgraph people describe it if you listen to their lectures and talks.

This is why I consider Hashgraph to be decentralized centralization. What it means is that the algorithm Hashgraph uses is decentralized in theory and in practice, but there certainly will be bad actors trying to control the technology, simply because they own it. It’s their software by law, and no one else has access to it or can see into it. This opens a whole slew of undesirable corruption.

Let’s face it — money talks, and whenever it’s in the conversation, it usually wins.

I’m sticking with blockchain for now. Who knows, maybe I’ll change my mind if I discover that everything I just said is wrong.