ZSF excluded from NU6, ZSA encryption dispute
ZSF will not be in NU6, and opinions differed on ZSA verified encryption.
Agenda summary 7
Zebra update: zcash_script and preparing Zebra 1.8.0
Zcash Foundation · Marek
- They released a new minor version of zcash_script. It no longer depends on zcashd's Rust dependencies, so Zebra no longer needs to keep its librustzcash dependency aligned with zcashd.
- Previously zcash_script was tied to this dependency through zcashd. They updated the librustzcash dependency and fixed the Windows tests, putting Windows back as a tier 2 supported platform.
- They merged a PR that drops transactions with unpaid actions and removed the scanner functionality from the zebrad process.
- They will merge the custom testnet and regtest documentation soon, and plan to release Zebra 1.8.0 containing these changes next week.
ECC update: ZIP-320, note commitment tree fix, Arti exchange rate lookup
Electric Coin Company · Daira, Kris, Str4d
- For ZIP-320, they are sorting out edge cases such as handling dust notes and inputs smaller than the ZIP-317 marginal fee, and it is nearly at the merge stage.
- Kris fixed a problem where an assertion added in the note commitment tree shard boundary bug fix fired even in non-error cases. Besides ZIP-320, this fix was the last item blocking the next SDK release.
- Str4d's exchange rate lookup PR fetches from three sources, Binance, Coinbase and Gemini, through Tor via Arti, and takes the median.
- Gemini almost always succeeds and Binance mostly succeeds, but Coinbase appears to block Tor exit nodes and always returns a 400 error. They are looking for other data sources such as aggregation services.
zcashd deprecation progress
Electric Coin Company / Zcash Foundation · Str4d, Arya
- They merged Kris's PR, balancing the representation of shielded and transparent funds in zcash_client_backend and zcash_client_sqlite.
- The backend wallet is still some way from making zcashd-like transparent spends, but this is big progress. Zebra scan is almost split out as a standalone binary, and Alfredo is moving the tests.
- On the ZF side, they started work on an indexer RPC for the replacement wallet.
FROST 2.0.0-rc.0
Zcash Foundation · Conrado
- They released FROST crate 2.0.0-rc.0. It has a breaking change that simplifies the serialization API and supports no_std. Only the ed448 cipher suite does not support no_std.
- After getting user feedback, they plan to release the official 2.0.0 within a month. The refresh shares feature is nearly done. They are adding user registration to the FROST server.
- With registration, it becomes easier to find the other party and receive public keys to authenticate the connection.
Zcash Sustainability Fund update
Shielded Labs / Eiger · Tomek, Str4d, Dodger
- Work on ZSF resumed in the past few weeks.
- Tomek is changing librustzcash so that the transaction ID, sighash and transaction read/write include the ZSF deposit field, and has put it behind the zcash_unstable Rust flag.
- It is also being reflected in the Zcash test vectors. Str4d asked whether the intent is to meet the ZIP deadline for NU6 activation (Tuesday, with the final deadline two weeks later).
- Dodger relayed that Jason McGee said on the forum that ZSF will not be ready for NU6. Tomek also answered that they had not been targeting NU6.
- They decided to discuss on the R&D Discord whether to wait on ZSF until after the zcashd deprecation. They are also taking suggestions for ideas to rename ZSF to something friendlier.
ZSA update and new ZIPs
QEDIT · Vivek, Jon
- They are responding to Str4d's review on the zcash_note_encryption PR.
- They are moving the librustzcash and Orchard work to an NU7 basis and putting it behind the zcash_unstable=nu7 feature flag to prepare for merging the ZSA work.
- The test vectors that separate out the Orchard ZSA part are ready for review, and the vectors that change even the transaction format are being tidied up. A Halo 2 lookup range check generalization PR will also go up soon. ZIP changes continue in the new PR 854.
- Jon introduced two new ZIP efforts: receive control, which makes it possible to reject unwanted transactions, and verified encryption, which encrypts the transaction information of that asset for the asset issuer.
- The latter is a feature that is not privacy-focused, aimed at adoption by stablecoin issuers.
ZSA verified encryption dispute
Daira, Jon, Dodger, Kris, Arya
- Daira opposed verified encryption, saying it makes the issuer key a giant honeypot. If the key leaks through law enforcement or hacking, the entire transaction graph could be exposed, and it must be safe against quantum computers too.
- Daira also worried that regulators could pressure to apply it to other assets or to ZEC. Jon answered that this feature is an opt-in limited to that asset and is a signal of a community that values legitimate use.
- Jon said this feature does not apply to Zcash as a whole. Dodger said it is an opt-in that the issuer chooses to activate and must be made clear to users, and that consensus should be sought through transparent discussion.
- Kris raised governance issues: that ZCG funded this without community consensus, and that naive users could be exposed to surveillance unintentionally.
- Kris instead mentioned the opportunity cost of the resources that could go to shielded programmability, and Jon answered that no team has made such a proposal. Daira asked about designs such as threshold decryption, saying there is no reason the key holder has to be the same as the issuance approver.
- Arya said they are not sure whether there is a good method.
This summary is based on the original meeting notes; check them for exact details.