Does ETH really need L2s?

[TastyChillySauce00]

I bet other L2s aren't as "decentralized" as they claim to be. Perhaps, Polygon is the only one on the list that is decentralized. You can see why centralized L2s often offer cheaper fees and faster settlement times. It's because they've sacrificed decentralization and censorship-resistance. Either a chain sacrifices decentralization in favor of greater transaction capacity (scalability) or all the other way around. You can't have the "best of both worlds". At least, ETH's improved scalability will make L2s a thing of the past. It should only be a matter of time before this happens.

You're not wrong, in fact many L2s are controlled by single entity. Linea and base is among the most centralized blockchain.
Linea have centralized sequencer controlled directly by the foundation and also the governance is foundation-led. At least the other L2 only have security council but this one is pretty much centralized.

[d5000]

In general terms I disagree with most of you here

I think layer-2's are always needed for a cryptocurrency to stay decentralized if it reaches a certain adoption, because otherwise you will make full node operation so expensive it can only be afforded by deep pockets. That's more or less the "big blocker" dilemma. As Ethereum has a less decentralized "ethos" than Bitcoin, there are less barriers for the developers to move to a big block model (i.e. "all nodes store everything"), following Solana. But if it really became mass adopted eventually, node operation would become much more expensive. The current low fees are very much a consequence of Ethereum's increase of relevant block limits.

Where I agree is that the current L2 landscape in Ethereum does not look really promising, with many essentially centralized rollup/sidechain projects. And PeerDAS is a technology that I've read in the future could also serve as a base for L1 scaling, even if it currently only handles L2 data blobs. I have however not really looked at that technology in depth to evaluate if it can replace L2s fully in the long term.

[Abiky]

You're not wrong, in fact many L2s are controlled by single entity. Linea and base is among the most centralized blockchain.
Linea have centralized sequencer controlled directly by the foundation and also the governance is foundation-led. At least the other L2 only have security council but this one is pretty much centralized.

It's no wonder why most L2s are able to offer ultra-low fees and blazing-fast speeds. Because they have sacrificed decentralization in favor of convenience. Even Bitcoin's Lightning Network is in the same boat. Guess what? It turned out to be a huge failure. With so many altcoins available on the market, people can just switch to an altcoin of their choosing instead of an L2 network. It's much easier this way.

I guess L2s will become useful when the main Blockchain becomes too crowded. Considering that demand for crypto has stalled a bit, it's going to take quite a long time before we see a resurgence in L2 usage (if any). At least we know, ETH will outlast many L2s and "shitcoins".

[Conqueror777]

Gas fees are so cheap right now, compared to what they were in the ICO era of 2017. Fractions of a penny (USD), according to Etherscan. Vitalik and team continue to work on the main Ethereum blockchain, focusing heavily on the network's scalability. This, combined with lower market prices, resulted in lower gas fees.

That makes me wonder, does ETH need L2 networks? At this point, they seem highly-unecessary. Wouldn't you agree?

Your input would be greatly appreciated. Thank you very much.

Yes in my opinion L2 networks are unnecessary right now. So yes I totally agree with you!

[nelson4lov]

Yes in my opinion L2 networks are unnecessary right now. So yes I totally agree with you!

Gas fees are cheap because no one is actively using Ethereum Mainnet like they used to as most of those volumes and transactions have since moved to L2s. Imo, if we scrap all the L2s right now and all go back to using Mainnet, we'll arrive at the same bottlenecks, congestions and high gas fees that plagued the Ethereum Mainnet for years. AFAIK, the underlying problem—scaling the L1 is yet to be fixed.  Mainnet can't currently scale right now.

[shinratensei_]

Yes in my opinion L2 networks are unnecessary right now. So yes I totally agree with you!

Gas fees are cheap because no one is actively using Ethereum Mainnet like they used to as most of those volumes and transactions have since moved to L2s. Imo, if we scrap all the L2s right now and all go back to using Mainnet, we'll arrive at the same bottlenecks, congestions and high gas fees that plagued the Ethereum Mainnet for years. AFAIK, the underlying problem—scaling the L1 is yet to be fixed.  Mainnet can't currently scale right now.

The cost of decentralization and security is real, Ethereum could become faster but it won't be as decentralized and secure.
I prefer the Ethereum of today. Having 30 tps is fine without sacrificing decentralization, we don't need another solana.

[Abiky]

The cost of decentralization and security is real, Ethereum could become faster but it won't be as decentralized and secure.
I prefer the Ethereum of today. Having 30 tps is fine without sacrificing decentralization, we don't need another solana.

True. But isn't Ethereum already centralized? Ever since it made the switch to PoS (Proof of Stake), ETH went all the way down to "rock bottom". I mean, in terms of decentralization and censorship-resistance. It's validators' (stakers) game now. Institutional investors also have a hold on ETH, via the spot ETFs. Although, they hold more BTC than ETH in their portfolio these days.

Still, ETH is a far better choice when we compare it against coins of the same kind. L2s are even more centralized. They might be unecessary right now, because the main ETH blockchain has low network activity. But once the main chain gets congested again, interest in L2s might resurface. We can't tell what will happen in the future. Better have multiple solutions just in case.