Original title: "Starknet Goals and Roadmap for 2023 》

Original title: "

Original Author: StarkWare

Compilation of the original text: Kate

This post will lay out Starknet's roadmap for 2023 and explain the considerations driving it. The roadmap focuses on performance and user experience.

Our goal is to provide the transparency the community needs, so we share the roadmap as early as possible. Of course, sharing the roadmap early also means things could change.We would like to thank the Starknet developer community, especiallybuilders council

, who provided valuable insights and feedback in defining this roadmap. Special thanks to Sylve from briq, Federico from LambdaClass and Jorik from Nethermind for their valuable feedback. We invite the community to continue to participate in this important process.

Journey to Starknet

Starknet is designed to allow developers to leverage the power of STARKs and Cairo to build their applications. So far, Starknet development has focused on providing developers with future-proof features. With the support of Cairo 1.0, except for Regenesis, the functions of Starknet are now basically complete.

Regeneration

Starknet will undergo regeneration this year. This will be a seamless event for the network as it will cause no downtime and will not reset the network state.

The Fresh Cairo 1.0 contract will not require any action. Rebirth is not a big deal for them.

Cairo 0 contracts need to complete the transition to Cairo 1.0. Cairo 0 will be shut down only after we have ensured that the majority of the community is able to make this transition.

Regeneration will be the last planned breaking change on the Starknet roadmap, and we expect any future breaking changes to be scrutinized for governance decisions, as is done in Ethereum.

Or to explain regeneration in Cairo 1.0 itself:

Roadmap to 2023

Starknet's current development focuses on performance and user experience. Here is the estimated timetable:

performance

Our main goal in the coming months is to improve network performance to support the expected growth in user and developer numbers. As we can see, performance is clearly defined by throughput and latency, but also by transaction costs.

Here we list the focus areas by priority and (where possible) state the planned releases for them.

Throughput and Latency

The upcoming Starknet v 0.12.0 will contain significant throughput and latency improvements. This is the culmination of the past six months of Rust-ifying the Starknet stack. This release will include the transition to the Rust-based Sequencer (developed by StarkWare) and the new Rust-Cairo VM (cairo-rs, developed by LambdaClass) - two open source projects. We expect to have performance benchmarks soon.

This shift will significantly reduce block execution time and thus improve throughput. In the absence of congestion, we also want to improve transaction latency, since the main cause of latency is block execution time.

Even after v 0.12.0, higher throughput and lower latency will remain top priorities, and work on improving them will be the backbone of Starknet development.

transaction cost

In v 0.13.0, Starknet transaction costs will be significantly reduced. This will be achieved by targeting a major component of transaction costs: L1 (Ethereum) data costs, which account for 95% of transaction costs today. Volition will allow developers to build Starknet applications using a hybrid Data Availability (DA) model: on-chain and off-chain data. An article about Volition is coming soon, detailing its design and developer interface.EIP-4844 ethereum

(Sharded Blob Transactions) is expected to be released in Q4 2023. Starknet will adapt and benefit from this as soon as possible, in addition to the introduction of Volition.

Due to Ethereum's EIP-4844 and Starknet's Volition (off-chain data availability), we expect data costs to decrease significantly.

faster end

As the release of Starknet v 0.14.0 and v 0.15.0 approaches, we remain committed to providing users of Starknet with a seamless and predictable network experience regardless of congestion levels. To achieve this goal, we are prioritizing development in two key areas: fee markets and block intervals.

fee market

For developers and users, the planned performance improvements will be noticeable as long as the network is not congested. However, when network congestion occurs, it will result in increased wait times for everyone. To address this, Starknet will introduce a fee market in v 0.14.0 to efficiently allocate Starknet's finite resources based on users' willingness to pay for transactions, not just their queued position.

Shorter fixed block interval

In v 0.15.0, Starknet will switch to a constant and shorter block interval. Currently, each block has its own proof, and the interval between blocks is variable - the conserved amount is the block cost. To reach a fixed block cost, the network waits for enough transactions to accumulate, which results in variable block intervals. To address this, we plan to decouple the 1:1 relationship between Starknet blocks and their proofs. As of v 0.15.0, proofs will prove the integrity of one or more Starknet blocks. This will fix the block blocking interval and improve the Starknet user experience.

tradeoffs and considerations

What tradeoffs did we consider when determining this roadmap?

Performance is the most important thing - and this is the loud and clear feedback we get from the ecosystem. As of v 0.12.0, this will primarily be improved by increasing the throughput of the Sequencer.

In v 0.13.0 we had to choose between lower transaction costs and better user experience (shorter/fixed block interval, and more predictable network response during congestion). We decided to focus on lower transaction costs rather than user experience because we expect v 0.12.0 to bring better latency (on the order of seconds). As mentioned above, the main means of reducing transaction costs will be the introduction of Volition, with more details emerging - EIP-4844.

Improving network behavior during congestion (by introducing a fee market) may have to wait until version 0.14.0. Although congestion is very likely after v 0.12.0, we hope that significant improvements in throughput will make this less likely. Therefore, we decided to prioritize reducing transaction costs.

Summarize

Summarize

The release of Cairo 1.0 marks the stabilization of Starknet feature development. The remainder of 2023 will be dedicated to improving Starknet performance and user experience. By the end of this year, we expect the base layer of the network to reach a healthy and stable state in terms of functionality, performance, and user experience. We expect the pace of major changes to drop significantly, naturally under the control of all network participants. By 2024, decentralization will be the focus of network operations and decision-making.