UDAs/ZSAs and burning funds for fees

[dontbeevil]

btw can we put up a bounty for coming up with best (UDA) proposal with fee mechanics? @ZcashGrants wdyt?

[aristarchus]

@dontbeevil

what happens to user’s txn when they bid low tx fee?

If a majority of txs have only the low -5% bid, then the miners will actually make the most ZEC by filling the block with -5% txs.

@dontbeevil

Does it mean users will broadcast upto 3 transactions then rest of them will be rejected?

Yes, if a user is willing to pay up to +5%, then they broadcast 3 txs. I’m sure this could be simplified to just one packet of that data in the future, but I suggested this assuming it would require the least amount of engineering to get up and going asap, and not require any major changes to consensus code.

@dontbeevil

then we could pick median fee as the base fee for next block (if Zcash doesn’t enforce all txns in same block to have same fees

It’s important that all shielded txs have the same fee in a given block in order to preserve privacy. Picking the median fee as a minimum for the next block (I think that’s what you mean by base fee?) would not allow for fees to go down when there is lower demand.

[Shawn]

Anyone familiar with the Tezos “Multi-Asset Shielded Pool”? Based on Zcash work, curious how they solve the fee problem?

https://masp.dev/
masp/multi-asset-shielded-pool.pdf at main · anoma/masp · GitHub

[holmesworcester]

My question whenever this comes up is, have any chains dealt with the “top heavy” and “value accrual” problem with UDAs, at all (like, including outside the context of a privacy-focused currency like Zcash)?

Do we have any models for this that are known to add value and security to the “host” chain?

Agree with @secparam that there’s no reason to put solving this problem ahead of implementing UDAs, Unless there’s some known strategy for addressing the “top heavy” problem that you can’t implement once you’ve already started, we should assume that it’s possible to find a solution we can graft on later, when there are bajillions of Tether, DAI, and Bitcoin on Zcash

Startups often have to deploy a feature in its simplest form first and figure out value capture later. It seems like that’s what the situation calls for here.

[holmesworcester]

@Shawn

Would it be possible to set a default fee for UDA related transactions independent of the typical transaction fee?

It would, but then I think you get the worst of both worlds, relative to some approach where the fee is hidden:

1. smaller anonymity set for UDA users
2. smaller anonymity set for ZEC users
3. no solution to the value capture problem, since the fee has no relation to value.

Also, are we sure that proportional-to-value fees address the value capture problem in a way that’s meaningful to ZEC investors, the ZEC dev fund, and Zcash security?

Intuitively this seems like it depends on real-world factors, such as whether money sits or moves, and whether the market can bear fees high enough to add security.

Maybe it needs to actually be based on the value that’s sitting there, no?

(Again, I agree that we should do it regardless, but I’m curious what the answer here is!)

[zorg5]

This project doesn’t use the networks native token for fees and lets the fees be paid in the UDA tokens. Network Incentive: Privacy (PRV) Mining & Distribution - 👾 Tech - Incognito

[Jgx7]

Do UDAs enable devs to create an ecosystem similar to polkadot?

[Shawn]

ECC exploring Zcash-Shielded Assets - Electric Coin Company

[aiyadt]

@secparam

burn the UDA specific portion of the fee

Not a fan of burning hard money. Maybe requiring a higher fee for ZSAs(previously UDA) and paying it to the miner of that transaction would incentivize mining nodes to invest in top quality hardware & processes to optimize Shielded Asset transactions on the Zcash network whilst supporting the standard Zcash transactions?

[secparam]

Yeah, i’m not either. What about burning asset specific gas?

As I’ve said, fees for ZSAs are a massive red haring for version 1. We don’t even really have fee’s for shielded TXs. We can ship and iterate ZSA fees just like zec fees.

Eventually we will end up with ZSA specific fees. And any version of ZSAs that support fees will need to:

1. Respect privacy. This eliminates miners getting paid fess that relate to the amount or type of asset transacted.

2. Preserve value. We need to support things like Tether/USDC. We can’t have ZEC seigniorage apply to assets.

I think this leaves us with burning asset specific gas and a pricing mechanism.
E.g. issue asset specific gas via an onchain auction. Post 1 unit of the asset, a fee, and a reserve price denominated in zec. If anyone pays the reserve price, they get the asset. Otherwise, in x blocks, you get 1 unit of gas for that asset you can use or sell. (this can probably be redone as some kind of yield farming where it merely sets the conversation ratio.). It’s got a lot of downsides. Hope it spurs someone to better ideas.

[amiller]

Agree that ship and iterate is a good strategy to quickly build on the enthusiasm for assets.
Basically it would be a good idea to emphasize the experimental nature of ZSAs so we have the flexibility to iterate on the fee policy without creating some expectation that they’re fixed and final

Disagree with this subpoint tho, which isn’t necessary for the bigger picture we agree on:

@secparam

any version of ZSAs that support fees will need to:

1. Respect privacy. This eliminates miners getting paid fess that relate to the amount or type of asset

This doesn’t have to be the case. For example you could charge a value-dependent fee at the time of shield-minting (minting a ZSA after depositing some existing non-shielded backing asset), or at the time of withdrawing (redeeming a ZSA note to withdraw the non-shielded backing asset). It wouldn’t affect privacy because it’s only computed at the time of deposit/withdraw, which already publicly disclose the value of the backing asset

[secparam]

I meant fees for transacting.

You certainly can do fees for minting and maybe withdrawing (though not sure how you prevent someone doing a bridge which has you withdraw by paying to the 0 address). That would have the nice side effect of encouraging people never to withdraw and only mint once when the asset was cheap.

[aiyadt]

As for asset specific fees, how will the consensus rules differentiate between assets? Would it require reliance on a third-party pricing oracle? This might not be optimal.

Respecting privacy can be achieved by an calculating the fees based off the ZSA traffic compared to last block and making that info available via block headers to reduce complexity of fee calculation by clients/wallets.

Preserving value for stable coins need to be handled by the corp. issuing the assets, ZEC fees will anyways be lower than ETH smart contract transaction fees, and enabling direct L1 ZSA-BTC swaps via Thorchain.

My point is that any fees that are being planned to be burned better be sent to the miners as they are responsible to power the mining nodes and should have incentives to mine and broadcast ZSA transactions.

Also, solid points made by you @secparam on the UDAs panel at Zcon2 lite