The biggest improvement is transaction cost. Using an Ethereum Layer 2 network can be dramatically cheaper than interacting directly with mainnet, which makes DeFi, payments, gaming and other applications much more practical for smaller users.

Speed has improved as well. Many rollups confirm transactions quickly enough that the experience feels much closer to a conventional application than older Ethereum usage did during periods of heavy congestion.

But scalability is not only about transactions per second or gas fees. User experience matters too, and this is where I think the rollup model still has weaknesses.

A new user may need to understand which Layer 2 an application uses, how to move assets there, whether a wallet supports the network, which bridge to trust and where liquidity is located. Even experienced users can end up with funds spread across multiple networks.

Liquidity fragmentation is another issue. Two applications may both be part of the Ethereum ecosystem while operating on different Layer 2 networks with separate liquidity. If moving between them requires additional transactions or bridges, some of the simplicity that scalability is supposed to create disappears.

Security can also be difficult for users to evaluate. Different rollups may have different upgrade mechanisms, bridge designs, sequencer models and stages of decentralization. From the user’s perspective, they may all look like “Ethereum,” even though their actual risk profiles are not identical.

I think rollups have solved a large part of Ethereum’s capacity problem, but they have not completely solved the usability problem. The next major improvement may come from wallets and applications hiding the underlying network complexity so users no longer need to manually think about bridges or individual Layer 2s.

Do rollups feel like a real scalability solution in your everyday Ethereum usage?

Have lower fees and faster transactions outweighed the additional complexity of multiple networks?

And what needs to improve most before Layer 2 becomes seamless for regular users — bridging, liquidity, wallet integration, interoperability or security transparency?

Some networks try to scale directly at the Layer 1 level by increasing throughput, improving execution efficiency and optimizing validator performance. The advantage is simplicity. Users remain on one chain, liquidity is easier to concentrate and applications do not need to depend on multiple external scaling layers.

The downside is that higher performance can create more demanding hardware requirements. If running validators becomes too expensive or technically difficult, decentralization may gradually weaken. That is one of the main trade-offs behind high-throughput Layer 1 architectures.

Ethereum has largely moved toward a modular scaling model where Layer 2 networks handle more execution while Ethereum provides settlement and security. This can theoretically support much larger transaction volumes without forcing the base layer to process everything directly.

The problem is complexity. Multiple rollups can fragment liquidity, create additional bridge dependencies and make the user experience harder to understand. From a technical perspective, the architecture may scale well, but that does not automatically mean it creates the simplest environment for users.

Another interesting direction is application-specific infrastructure. Instead of forcing every application onto the same execution environment, some projects are building dedicated chains, rollups or app-specific networks optimized for a particular use case. That can improve performance, but it can also increase fragmentation even further.

I think long-term sustainability will probably depend on more than raw transaction throughput. A successful scaling model needs to balance performance, security, decentralization, liquidity and user experience. Optimizing only one of those variables may create weaknesses elsewhere.

Interoperability could eventually make many of these distinctions less important. If users can move assets and interact with applications across networks without thinking about bridges or individual chains, different scaling architectures may be able to coexist much more effectively.

Which blockchain scaling model do you think has the strongest long-term future?

Do you prefer high-throughput Layer 1 networks, Ethereum-style Layer 2 scaling, application-specific chains or a modular multi-layer architecture?

And which trade-off are you most willing to accept: higher hardware requirements, more network complexity, liquidity fragmentation or slower base-layer performance?

In theory, competition between Layer 2 networks should be healthy. Different teams can experiment with transaction costs, performance, privacy, application-specific environments and developer tooling. That creates room for innovation and allows users to choose infrastructure that fits their needs.

The problem is fragmentation. Liquidity gets split between multiple networks, users need to bridge assets, applications deploy on several chains, and developers have to decide where to focus resources. For someone new to crypto, moving between different Layer 2 environments can make the Ethereum ecosystem feel much more complicated than it needs to be.

I think this becomes especially noticeable when several networks offer nearly identical functionality. If two Layer 2s have similar fees, similar applications and similar user experience, it is not always clear why both need to exist as separate ecosystems.

Liquidity is probably the biggest practical issue. DeFi works best when capital is deep and easy to access. When users and assets are spread across too many networks, market depth can decline and bridging becomes more important. That introduces extra steps, costs and security risks.

There is also a sustainability question. Some Layer 2 networks attract activity through token incentives, airdrop expectations or temporary reward programs. That can generate impressive short-term usage, but it does not necessarily mean the network has durable demand.

At the same time, consolidation may happen naturally. Over time, the strongest Layer 2 networks could accumulate most of the liquidity, developers and users, while smaller networks either specialize or gradually lose relevance.

Do you think the Ethereum ecosystem currently has too many Layer 2 networks?

Would you prefer a smaller number of dominant rollups with deeper liquidity, or is a highly competitive Layer 2 market better for innovation?

And what do you think will ultimately determine which Layer 2 networks survive — fees, security, liquidity, developer adoption, interoperability or ecosystem size?

Layer 1 networks such as Solana aim to process large amounts of activity directly on one chain. The advantage is a simpler user experience: users do not need to think about bridging between multiple networks, moving liquidity across different environments or understanding which Layer 2 they are currently using.

Ethereum has taken a different direction. A growing share of activity now happens on Layer 2 networks, while Ethereum increasingly acts as the settlement and security layer underneath them. This approach can improve scalability without forcing the base layer itself to process every transaction.

The trade-off is fragmentation. When activity spreads across many Layer 2 networks, liquidity, applications and users can become separated. Bridges and interoperability tools help, but the experience is still more complicated than using a single network with unified liquidity.

At the same time, scaling everything directly on Layer 1 has its own challenges. Higher throughput can increase hardware requirements, make validator participation more demanding and create difficult engineering trade-offs between performance, decentralization and reliability.

I think the most likely outcome may be a combination of both models. Some applications could remain on high-performance Layer 1 networks, while other ecosystems use Layer 2 networks for specialized execution and rely on a base chain for settlement.

The bigger question is what users will actually care about. Most people probably will not choose an application because it is technically Layer 1 or Layer 2. They will care about speed, fees, security and whether the product works. If infrastructure becomes sufficiently abstracted, users may eventually stop knowing which layer their transactions are happening on.

Where do you think most blockchain activity will ultimately live?

Do you prefer the single high-performance Layer 1 model or the Layer 2 scaling model built around a settlement chain?

And do you think users will still care which blockchain or layer they are using five years from now?