When Ethereum launched it was marketed as a platform that would one day become a worldwide decentralized super computer that could do anything. People don't try to push this narrative much anymore, and for good reason... unless of course they are talking shit about ETH.
The scaling issues are real.
It was perhaps foolish to think that an extremely redundant decentralized database would be able to scale up and perform generic programming tasks without any kind of drawback. Even centralized solutions like Facebook had to work extremely hard to scale up and take over. Years later, we find out the hard way that even long before mainstream crypto adoption, a single bull run can kick transaction fees up to $200 a pop.
ETH maximalists would make the claim that this is just a temporary setback and that scaling solutions will come into play that fix the situation. This is both true and a lie. It's the paradox of decentralization. Almost all scaling solutions reduce the security of the chain and introduce centralized attack vectors into the system.
This is one of the biggest advantages to Bitcoin. We can buy BTC today knowing that ten years down the road the core code is going to be exactly the same as it is now. Even with the second coin by market cap (ETH) we can see that this is far from the case.
There's a lot of drama going on with ETH's transition from POW to proof-of-stake. Many see this as a silver bullet scaling solution. Others are saying it's not going to help much. Either way, it's obvious given the ETH token distribution that the switch to proof-of-stake is an extremely centralizing event in terms of governance and regulation of the network. I'm not even going to pretend I know the in's and out's of everything going on, but I'm sure once this thing goes live we'll have a much better idea of the situation instead of having to game-theory it out and simply guess.
I'm of the opinion that general purpose blockchains don't really make sense. Just from a fundamental perspective of someone who knows how this tech works, why would anyone try to be a jack-of-all-trades when it comes to constantly needing to balance efficiency with security? It makes a lot more sense for there to be a bunch of networks that are all good at their niche. Then together the greater network that is created from interoperability has a lot more power than any one single network trying to do it all.
I think at the end of the day people like the idea of a general purpose blockchain because they are greedy. Plain and simple, they project capitalism onto an open source collaborative environment to their own detriment. They want their general-purpose blockchain to "eat the world" and become the dominance force that takes over everything. They want number to go up, and they think framing it as a competition and "winning" is going to accomplish this goal. That's not how rebuilding the economy from the ground up is going to work.
This is all rooted in greed and imperialism and control, which run contrary to everything crypto actually stands for. Maximalists are perpetually theory-crafting a paradox that makes zero sense.
We have to decentralize the world!
Oh, but also everything else but my thing is shit and eventually everything else will die except for my network. You should just give up and dump all your money into my bags while you can.
The hypocrisy is real.
It's like watching evangelical religious people fight with each other. And it makes perfect sense exactly how this happened. We were born and brainwashed into this system from birth. It would actually be weird if we weren't projecting that reality onto the current paradigm.
In any case
For a second there I was thinking that general-purpose blockchains like ETH and other EVM solutions were destined to die out because of all the bloat they create. However, there is this one thing that they do quite well: incubate other projects during their infancy.
Advantages of EVM
We've already seen a lot of good ideas come out from all the development happening on EVM chains. While AMM wasn't invented by crypto (it's existed for a long time) it was never a viable business model or scalable until EVM devs created yield farms. Now this tech is not only viable and scalable, it is also decentralized, which is pretty crazy when you think about it.
Even though a lot of the early iterations of this tech are absolute garbage and totally centralized, the concept itself is immortal. Ideas can't be killed. Yield farming will now live forever in some form or another. We have to assume it will get exponentially better with more development. I've even said that this tech will allow us to create stable assets with great elasticity that slowly go up over time (whereas fiat slowly goes down). We have EVM devs to thank for that.
There's also something to be said for the dog-eat-dog world of Ethereum and EVM devs. If there is a flaw in the protocol, the code ninjas on EVM figure it out very quickly and exploit it. If you have a project up and running on EVM for over a year and it hasn't been hacked yet, you know the code you're running is pretty damn solid considering such a hostile environment.
Another huge advantage of general purpose blockchains is that all the tokens on the platform can be atomic-swapped for one another. Once again we see that the advent of AMM tech and ultra-deep liquidity pools made these on-chain atomic swaps exponentially more powerful and valuable. Access to liquidity is a huge deal.
Startups in crypto that would be considered penny stocks (or even worse: no stock) end up with gigantic liquidity pools and allow whales to buy and sell x100 more than they would within the legacy system for the same amount of slip. Pretty wild when you actually think about it.
Second layer
While ETH might be a great starting place for a new project, it's a terrible place to get stuck on. I still follow a couple ICOs back from 2018. Imagine working on a project for 5 years on Ethereum and going from $1 transaction fees to $200. That could completely destroy someone's business model and make it totally irrelevant.
I think it's obvious that any kind of legitimate project build on a general computing blockchain will eventually have to leave the chain it was built on or it will never be able to scale up. And that's fine. That's exactly what the second layer is for.
We can think of the second layer and a completely different blockchain. The nodes that validate the second layer are not the nodes that validate the first layer. This makes the solution scalable, because then the first layer doesn't have to worry about the second layer, let the second layer regulate itself. While this technically makes the second layer "more centralized" in theory, it doesn't have to be in reality. It all depends on how many nodes are in consensus and how that consensus is reached.
There are many ways to implement a second layer. In some situations the data exists on the first layer but isn't processed on the first layer (HiveEngine). In other situations the data only exists on the second layer and the first layer is only called to settle disputes. There are also ways to combine both of these strategies together so that data is split efficiently between the two layers (Splinterlands).
There are even ways to create second layers in which the first layer is hardly needed at all. On Hive, it wouldn't be that difficult to create a second layer that only uses Hive for its accounts and private key security. You can log in with your Hive credentials, and that's it. Everything else would be completely separated from the main chain. This could be a great thing or a bad thing just depending on context and what was built. Either way it's a scalable solution and an easy way to prevent bloat on otherwise hyper-inefficient databases.
Conclusion
General purpose blockchains are never going to be the best at anything by definition. Their only true niche is incubating other tokens and platforms that actually have a real niche. Perhaps that's the entire point.
Once we understand this we can see that platforms like ETH (or god forbid, Cardano) are never going to "take over" and make other networks irrelevant. Quite the contrary their purpose is to support and protect other projects during their early stages. Yes, many of those chicks will not grow up to be a Golden Goose. Many will die out before ever hitting their prime. Everything is as it should be.
The big difference between this kind of incubation and what we see in the legacy economy is ownership. Legacy incubation is akin to venture capitalism. The entity fronting the startup money expects to own a big piece of the final product, and for good reason. Why else would they do it? It has to be a symbiotic relationship somehow. Both sides need to gain something out of the deal. Crypto can accomplish this without needing to own a piece of the final product, and that is revolutionary.