← All calls
ARBORIST#63

Zebra 1.3.0 and SDK 2.0.3

Zebra 1.3.0 fully supports Spend-before-Sync, and the SDK 2.0.3 release is imminent.

Agenda summary 5

  1. Zebra 1.3.0 release

    Zcash Foundation · Marek

    • Zebra 1.3.0 came out this week. The spend-before-sync feature is fully supported, and the progress bar can be used in the default build without special compile flags.
    • The progress bar must be turned on in the config file; when on, log output is redirected to a file and progress is shown in the terminal. A performance bug was fixed in the mining-related RPCs that had been slow when requesting network solutions for many blocks.
    • After the release, a PR enabling all mining-related RPCs in the default build was merged and will be reflected in the next release.
  2. ECC update: mobile SDK 2.0.3 and Edge wallet integration

    Electric Coin Company / Edge · Zooko, Paul Puey, Daira

    • SDK 2.0.2 fixed several bugs, and 2.0.3 is due soon. A new bug reported by Edge remains, so 2.0.4 with the fix may come out around next week.
    • The Zashi private beta is rolling out in stages with the new UI design partly implemented. Nighthawk, Edge and Unstoppable are integrating the SDK to support spend-before-sync.
    • Edge's Paul Puey said the roadmap is to follow all the features the SDK supports, including unified addresses, automatic shielding, memo support and full Sapling pool support.
    • Paul judged that fast sync works, but that wallets over a year old are still slow when synced from scratch. Edge will freeze its code the next day for store submission, and expects a store release by Monday or Tuesday next week at the latest.
    • Paul said the need to change the SDK twice among the release candidates is a concern for schedule delays. Daira explained that the recently fixed bug was one where notes could be wrongly merged in the transaction view, and that it will be handled as a point release with no API changes.
  3. Zcash Shielded Assets (ZSA) update

    QEDIT · Vivek

    • Optimization of the ZSA protocol circuit is nearly done, and PRs are open on the halo2 and orchard crates. The team is waiting for review from ECC and others.
    • The draft ZIPs for Transfer and Issuance (226, 227) are about to be merged into main of the zips repository, awaiting Daira's review.
    • The change of the issuance authorization signature from redpallas to a BIP-340 Schnorr signature was reflected in the ZIP, and the Orchard crate implementation is in progress.
    • The changes are being moved under a feature flag into a separate v6 transaction data structure.
    • With the Zebra team, a wallet that first supports v5 transactions and then ZSA is being built, and the team is preparing to move from Zcashd to Zebra.
    • For asset swaps, a draft document close to the initial ZIP structure is ready and will soon be shared with ECC and others.
  4. Trailing Finality research: Crosslink and the simulator

    Electric Coin Company · Nate, Daira

    • The first draft of the Crosslink design is fairly complete. The liveness security argument is complete, and the safety argument is still in progress.
    • The TFL book is being updated to fit Crosslink, and development discussion is moving to the Proof of Stake channel on the R&D Discord, opening it to the community.
    • The goal is to make the book consistent with Crosslink and finish the safety argument within two weeks. The simtfl simulator, started by Daira, runs the network faster than real time through discrete-event simulation.
    • A message-passing network is implemented so far, and an abstraction of the best-chain protocol is being written. The UTXO, anchor, nullifier and note commitment structures, which add complexity to the TFL design, are simplified for the simulation.
    • Task dependencies can be seen in the published DAG.
  5. FROST audit report and next steps

    Zcash Foundation · Pili

    • The last comments for the final audit report were resolved, and it will be shared once NCC publishes it. It is expected within a few days to a week.
    • The next step is publishing a stable release candidate, and if there is no feedback for about a month, a stable release with a stable API will be tagged.
    • The communication method between participants is not yet decided, so the team is investigating several P2P protocols while first implementing the simplest case using a trusted dealer. It is a key recovery scenario in which one user manages key shares across their own devices.
    • An option to turn off cheater detection in the signature aggregation step was added to the reference implementation. When it is off, you can tell whether someone cheated but not who.

This summary is based on the original meeting notes; check them for exact details.