Arbitrum co-founder skeptical of move to based and native rollups: Steven Goldfeder

Offchain Labs CEO Steven Goldfeder’s crypto origin story is classic geek gold. As a student at Princeton, he first heard about Bitcoin in 2013 during a visit day at Princeton.

Professor Ed Felten — the future inventor of rollups on Ethereum, co-founder of Arbitrum and a White House science advisor — gave a five-minute lightning talk about the cost of a government destroying Bitcoin. 

Goldfeder’s curiosity was instantly piqued, and that night, other students hosted a poker game with a half-BTC buy-in. As he says:

“That was back in the days when that was just ten bucks. If only I’d held onto it…”

Goldfeder doesn’t really need to worry about that missed opportunity now that Arbitrum has become a major force in crypto. It was designed to fix some of the issues plaguing Ethereum, such as high gas fees, sluggish transaction speeds and scalability bottlenecks. It’s currently the leader among Ethereum L2 scaling solutions.

But rollups have been criticized for relying too heavily on centralized sequencer revenue and for a lack of interoperability and composability in the ecosystem. Ethereum Foundation researchers have proposed making rollups “based” and/or “native” to fix these problems.

Despite reports coming out of January’s sequencing call suggesting Arbitrum would embrace these solutions, Goldfeder appears to have had a rethink.

Why Arbitrum won’t become a based rollup

Based rollups use Ethereum’s validators for sequencing, opening up the possibility of interoperability and composability among L2s. Native rollups, meanwhile, woulduse Ethereum’s base layer network to replace fraud-proof systems and make L2s as secure as Ethereum itself.

But Goldfeder now sounds a lot more skeptical about the benefits for Arbitrum in adopting the tech.

“Just to clarify the terms, you know, a based rollup is basically a rollup without a sequencer that uses Ethereum for sequencing. Actually, if you look at that initial Arbitrum paper that we wrote in 2018, you might call it a based rollup. It doesn’t have a sequencer; that only came later,” he says. 

As for native execution, he echoes Vitalik Buterin’s sentiment: Ethereum’s strength lies in its diversity. 

“And one thing we experiment in is different execution models. So, Arbitrum is fully EVM but also has something called Stylus, which gives you the ability to write smart contracts in other languages such as Rust, C and C++,” he says.

“But I think having put this system in production for four years, in different testnet phases, I think the design that we’ve come on is actually the correct one.”

“So, I don’t think Arbitrum will become a based rollup anytime soon, and I don’t think that it should when it comes to native execution.”

Magazine decided to seek further clarification and context from Arbitrum on its position after the interview was conducted. Regarding the chances of his L2 becoming a based rollup, Goldfeder confirmed in a statement that he believes Arbitrum’s current design is “more efficient, practical, and cost-effective” than based rollup designs.

He was more circumspect about the chances of Arbitrum making use of the technology that underpins native rollups but made no commitment to do so.

“I support Vitalik’s view of native rollups, which envisions a common core that we factor out that will be useful by many rollups, but also provides for rollups to extend that with additional features,” he said. “Some rollups may choose to stop at the common core, but it’s crucial that we also support continued innovation and experimentation at the execution layer.”

From poker to protocols: Origins of Arbitrum

After Goldfeder’s crypto introduction at Princeton in 2013, his academic pursuits led him into cryptography and multiparty computation, ultimately landing him in the thick of blockchain security research. 

By 2018, he had co-founded Offchain Labs with the aforementioned Felten, reviving an old Princeton research project. The result? Arbitrum — the rollup solution whose history actually predates Ethereum going live.

Unlike the many crypto projects with names that sound like bad sci-fi movie titles, Arbitrum has a clear etymology. “It comes from arbitration,” Goldfeder explains. “So, the idea is we take as much computation off Ethereum as possible while having Ethereum to be the sort of referee. If something goes wrong, Ethereum can, you know, fix things.”

This approach aligns perfectly with Ethereum’s philosophy — decentralized, permissionless and resilient. 

How Arbitrum competes in Ethereum’s layer-2 race

The layer-2 space is crowded, with heavyweights such as Optimism, ZKsync and StarkNet all competing for dominance. Arbitrum has been the top dog for some time, but Coinbase’s L2, Base, is snapping at its heels, and faster rollups that sacrifice some decentralization for speed like MegaETH — already running at 20,000 TPS on testnet — seem likely to siphon some users.

Goldfeder says Arbitrum has full EVM compatibility, enabling developers to port their Ethereum smart contracts to Arbitrum with zero modifications. No new languages, no arcane integrations — just plug and play.

Gas fees are competitive, too. Arbitrum slashes Ethereum’s brutal gas fees by batching transactions, making DeFi, NFTs and gaming apps far more accessible.

But Arbitrum’s true advantages, according to Goldfeder, are its decentralization and security. Some L2 solutions have centralized choke points, while Arbitrum was the first general-purpose rollup to achieve Stage 1 decentralization.

This marks a transition where the chain state is verified with fault proofs but retains an override mechanism requiring consensus from both operators and external stakeholders, reducing dependence on operators alone — a feat that Goldfeder alleges many competitors have yet to reach.

As for Stage 2, which represents the ideal state where the network is fully managed by smart contracts, it eliminates the need for any human operators and achieves true permissionless validation, meaning anyone can submit fraud proofs; that remains a little way off.

Arbitrum’s challenges and roadblocks

Despite its success, Arbitrum isn’t without hurdles, and Goldfeder is well aware that competition is fierce, with rival L2 Optimism’s ecosystem growing, Base aggressively scaling, ZKsync offering cryptographic certainty and StarkNet’s zero-knowledge proofs also making waves.

Goldfeder also wants to make improvements to security.

“So, Arbitrum is like the original layer, the earliest optimistic rollup, the first general purpose layer 2 to launch. And the things that we take very seriously and are paramount to us is that we always put security and decentralization and (values like) censorship resistance over growth.”

Arbitrum DAO really does have control, says Goldfeder

Governance in crypto is often a facade — a handful of insiders pulling the strings while tokenholders merely play along. Not so with Arbitrum, says Goldfeder.

Goldfeder says the Arbitrum DAO is different in that many governance models are advisory — the foundation has the final say. But with Arbitrum, governance is self-executing. Token holders control billions in assets, manage protocol upgrades, and even decide how fees are distributed.

“The Arbitrum DOA is directly controlled by the tokenholders onchain. And by the way, for the fees for the network, those don’t come to me. Those don’t come to some centralized foundation wallet. Those go directly to the onchain treasury,” he says.

That may be so, but last month, a proposal by Arbitrum DAO’s recently appointed growth management committee sparked backlash after recommending that 7,500 ETH be deployed to Lido, Aave and Fluid, three decentralized finance protocols that are not native to Arbitrum.

Take the Gaming Catalyst Program (GCP), a $200-million initiative to boost Web3 gaming. Unlike traditional VC-backed funds where profits go to insiders, GCP’s sole LP is the Arbitrum community itself, which is voted on by the DAO. Every success directly benefits tokenholders, reinforcing a true ownership model.

Beyond L2 scaling, Offchain Labs is also directly contributing to Ethereum’s core development. “We built Prism, Ethereum’s leading consensus client,” Goldfeder notes. “The merge, the shift to proof-of-stake, and even upcoming blob transactions — our team is one of the core developers that are central to that conversation.”

And then there’s Arbitrum Orbit — a framework allowing developers to spin up their own customized chains. With over 100 orbit chains in development, Arbitrum is attempting to become the AWS of Web3, offering tailor-made blockchain solutions for enterprises, DAOs and innovators.

Arbitrum’s future in Ethereum scaling

Arbitrum is bridging the gap between Ethereum’s ideals and real-world usability. By attempting to slash costs, improve speed and decentralize governance, it is trying to pave the way for a future where Ethereum can compete with traditional financial systems without compromising its core principles.

Goldfeder has the final word: “Unlike some other scaling methods, Arbitrum remains closely integrated with Ethereum’s decentralized framework.”

That remains to be seen, but 12 years after an infamous game of poker using Bitcoin in 2013, Goldfeder and his company may yet up the ante.

Subscribe

The most engaging reads in blockchain. Delivered once a
week.

Monty Munford

Monty Munford writes regularly for the BBC, The Economist and City AM and has been a tech columnist for Forbes and The Telegraph. He also runs a growth and visibility consultancy and has appeared at more than 200 events and conferences, interviewing figures such as Tim Draper, the late John McAfee, Sir Tim Berners-Lee, Steve Wozniak, Kim Kardashian, Guns N’ Roses and many others.

Follow the author @montymunford