[Review] MD-38: Provide for Native Bridge and FFS Usage with Fixed Token Supply by l-monninger · Pull Request #38 · movementlabsxyz/MIP

l-monninger · 2024-10-22T22:50:05Z

Summary

MD-38

ganymedio · 2024-10-23T10:22:04Z

So the network has a fixed supply of MOVE? Could you link to documentation about that @l-monninger ? It's news to me. The general tokenomics and validator emissions should be included in this discussion, if there is an established design for those components. Without that information, I'm not convinced that deflationary tokenomics means the supply will converge to zero. If there is such a possibility, the mathematical evidence for it should be spelled out, to inform solution ideas.

0xmovses · 2024-10-23T20:53:37Z

@andygolay 10 billion MOVE ERC20 tokens. When gas is used it is effectively burned, on the L2, this dwindles the supply. We are currently doing 5 - 7 million transactions per day on testnet, it wouldn't take too long for the total supply to be completely burned via execution (gas).

l-monninger · 2024-10-23T21:33:34Z

@andygolay Yes, but his MD requests generally how to supply that.

The statement about deflationary token going to zero is not about deflationary tokens in general. It is about the particular case where you are simply burning gas in a gas algebra that does not work on fractions of the total circulation. You will deflate and then eventually reach 0 simply because you are taking money out of the system in amounts that are not asymptotic and you are not adding any token back. The formalization is trivial and algebraic $T - g \dot n \leq 0 \iff n > \frac{T}{g}$ where $T \in \mathbb{R}$ is total supply $g \in \mathbb{R}$ is average gas per transaction and $n \in \mathbb{N}$ is the number of transactions haha

ganymedio · 2024-10-24T09:58:00Z

According to a recent activity (only a month, more data would be better but this gives a rough idea) it looks like gas consumption is currently somewhere about 2000 MOVE per day:

10 billion / 2 thousand = 5 million days or a little over 13,698 years worth of MOVE left.

Even at 5000 MOVE per day burned, we have 2 million days or 5,479 years remaining.

We shouldn't expect this specific blockchain to last more than a few hundred years, so maybe I'm being totally naive but I don't see the gas depletion issue being urgent even if we are emitting many tens of thousands of MOVE per day.

What if there's a certain amount MOVE set aside for FFS rewards? Has that reward amount been established? From [review] MIP-34 + MD-34: Fast-Finality Settlement #34 it seems like the answer is "not yet" and that amount can be established:

To achieve crypto-economically secured Fast-Finality, we need to solve the following problems:

design a staking mechanism for the validators to stake assets, distribute rewards and manage slashing

Could we engineer the reward amount to have the supply deplete so slowly that we're practically guaranteed a few hundred years, say 300 years worth of MOVE supply? This would mean we can burn about 33 million MOVE per year, or a little over 91,000 MOVE per day.

franck44 · 2024-10-28T04:45:08Z

When gas is used it is effectively burned,

I don't think this is true. The gas fees are transferred to the operator for them to be dispatched to the sequencer, executor and DA and L1 (root hash) transaction fees.

I don't understand what "effectively" means here.

l-monninger · 2024-10-28T07:06:17Z

@franck44

I don't think this is true. The gas fees are transferred to the operator for them to be dispatched to the sequencer, executor and DA and L1 (root hash) transaction fees.

I don't understand what "effectively" means here.

I believe he's referring to the case where nothing is implemented for this transfer, as is the subject of this MD.

apenzk · 2024-10-28T16:05:16Z

can someone link me to the reason why

the L1MOVE =/= L1-bridge-token
the L2MOVE = Move Gas token =/= L2-bridge-token

l-monninger · 2024-10-29T00:09:01Z

can someone link me to the reason why

the L1MOVE =/= L1-bridge-token

the L2MOVE = Move Gas token =/= L2-bridge-token

This MD does not strictly require that. But, numerous points have been made about why this is potentially dangerous here: #20

There are three key points:

Fallible bridge means if compromised in L2 direction, sybil resistance deteriorates.
Fallible bridge means if compromised in L1 direction, staking costs are lessened.
Fallible bridge means any failure breaks fixed supply requirement.

apenzk · 2024-10-29T15:18:42Z

Fallible bridge means if compromised in L2 direction, sybil resistance deteriorates.

Fallible bridge means if compromised in L1 direction, staking costs are lessened.

Fallible bridge means any failure breaks fixed supply requirement.

it looks like that all these points are still true for the bridge+DEX design. see
https://github.com/movementlabsxyz/MIP/pull/20/files#r1820933114

apenzk · 2024-10-28T15:52:16Z

+![Fixed Supply Rewarding Problem](./fixed-supply-rewarding-problem.png)
+
+
+Using Fast Finality Settlement with the Atomic Bridge presents a challenge when the token supply is fixed. 


Suggested change

Using Fast Finality Settlement with the Atomic Bridge presents a challenge when the token supply is fixed.

Entertaining different tokens on Layer 1 (L1) and Layer 2 (L2), keeping their added-up supply constant while using the L2-token for fee payment introduces challenges.

The mention of FFS confused the context, i put it further down

apenzk · 2024-10-30T18:32:23Z

+
+Users of the Movement SDK should be **informed of different approaches and trade-offs when using the Atomic Bridge and Fast Finality Settlement with a fixed token supply.** This document provides an initial conceptual framing towards that end. 
+
+This document also seeks a response to the concerns raised in [MD-20](https://github.com/movementlabsxyz/MIP/pull/20).


Suggested change

This document also seeks a response to the concerns raised in [MD-20](https://github.com/movementlabsxyz/MIP/pull/20).

This document also seeks a response to the concerns raised in the withdrawn [MD-20](https://github.com/movementlabsxyz/MIP/pull/20).

apenzk · 2024-10-30T18:46:45Z

+
+### D2: Provide Staged Approaches for Networks Building on the Atomic Bridge and Fast Finality Settlement
+
+**User Journey**: Developers can reference approaches spawned from these desiderata to understand how to build networks using the Atomic Bridge and Fast Finality Settlement with a fixed token supply **in stages.** For example, developers can initially roll out a network using the **"Free Execution" approach and then transition to the "Noble Attester"** approach and so on. Some formats may be more suitable for certain stages of network development and release than others. 


Suggested change

**User Journey**: Developers can reference approaches spawned from these desiderata to understand how to build networks using the Atomic Bridge and Fast Finality Settlement with a fixed token supply **in stages.** For example, developers can initially roll out a network using the **"Free Execution" approach and then transition to the "Noble Attester"** approach and so on. Some formats may be more suitable for certain stages of network development and release than others.

**User Journey**: Developers can reference usage approaches spawned from these desiderata to understand how to build networks using the Atomic Bridge and Fast Finality Settlement with a fixed token supply **in stages.** For example, developers can initially roll out a network using the **"Free Execution" approach** and then transition to the **"Acceptor" approach** (a volunteer attester, see [MIP-34](https://github.com/movementlabsxyz/MIP/pull/34)), and so on. Some formats may be more suitable for certain stages of network development and release than others.

apenzk · 2024-10-30T20:08:31Z

+| Name              | Description                           | Gas Destination       | Reward Form          | Challenges                          |
+|-------------------|---------------------------------------|-----------------------|----------------------|-------------------------------------|
+| **Decoupled**   | • Gas spent on the L2 deposited to pools on the L2. <br> • Rewards on the L1 are issued in a LP token which is demurrage. <br> • Ideally, intermediary tokens are used in the bridge to shield gas and staking token from bridge fallibility. | Pool on the L2         | FFS LP Token; voucher periods can be assigned. | • LP Token's demurrage assigns a lifetime to the viability of the settlement protocol which will require a new or asymptotic token voucher or token generation event at some point. |
+| **Trickle-back** | • Gas spent on the L2 recirculates to pools on the L2. <br> • Rewards are provided on the L2 for bridging back to the L1 at the address of the FFS LP Token. <br> • Ideally, intermediary tokens are used in the bridge to shield gas and staking token from bridge fallibility. | Pool on the L2         | Staking Token               | • Rewards for pool providers to bridge back to the L1 at the address of the FFS LP Token need to be sufficient. <br> • Pool providers likely need to have only custodial access to token, otherwise intended rewarding game theory would likely be intractable. |


Staking token is not defined - at least there should be a reference to the relevant document

apenzk · 2024-10-30T20:15:38Z

+| **Decoupled**   | • Gas spent on the L2 deposited to pools on the L2. <br> • Rewards on the L1 are issued in a LP token which is demurrage. <br> • Ideally, intermediary tokens are used in the bridge to shield gas and staking token from bridge fallibility. | Pool on the L2         | FFS LP Token; voucher periods can be assigned. | • LP Token's demurrage assigns a lifetime to the viability of the settlement protocol which will require a new or asymptotic token voucher or token generation event at some point. |
+| **Trickle-back** | • Gas spent on the L2 recirculates to pools on the L2. <br> • Rewards are provided on the L2 for bridging back to the L1 at the address of the FFS LP Token. <br> • Ideally, intermediary tokens are used in the bridge to shield gas and staking token from bridge fallibility. | Pool on the L2         | Staking Token               | • Rewards for pool providers to bridge back to the L1 at the address of the FFS LP Token need to be sufficient. <br> • Pool providers likely need to have only custodial access to token, otherwise intended rewarding game theory would likely be intractable. |
+| **Settle with Gas**            | • The settlement messages include the amount of gas spent. The LP for the FFS claims token locked in the L1 bridge. <br> • Bridge may still benefit from intermediary tokens.                              | FFS LP Token        | Staking Token               | • Unclear how to handle forks. <br> • Centralization remains as FFS contract needs to have capability to claim token locked in L1 bridge. However, this can be governed on-chain in this case. <br> • Intermediary tokens can still help reduce attack surface via the bridge. Now attacks would likely focus on exploiting capabilities of the settlement contract to unlock token. |
+| **Noble Gasmaster**      | • A trusted signer is responsible for successfully relaying gas spent. The use of intermediary tokens in the bridge is moot.              | FFS LP Token         | Staking Token             | • Centralized signer.   |


similar to the "acceptor" is there a way how any validator can be permitted to transfer L2 pool tokens to the L1 reward contract? (at the cost of L2fees and L1fees for that validator)?

Similar to the Acceptor model, a L1-selected acceptor could be responsible for transferring rewards from L2 to L1, and get rewarded on L1 for this, however a volunteer could transfer also the L2RewardPool balance to the L1RewardPool balance (at their own cost)

apenzk · 2024-10-30T20:17:03Z

+
+
+## Overview
+![Fixed Supply Rewarding Problem](./fixed-supply-rewarding-problem.png)


this figure is displayed without any description or context. it is not referenced in the text anywhere.

at no point burning is mentioned in this file

There is no reference to the model that describes the DEX approach. also for the sake of this MD and the description of the burn problem the DEX is not necessary. I suggest to add a figure that has a simple direct bridge from L1MOVE <> L2MOVE, and where L2MOVE is the gas. Such a model seems to be minimal yet envelop the problem sufficiently.

apenzk · 2024-10-30T20:21:06Z

+|-------------------|---------------------------------------|-----------------------|----------------------|-------------------------------------|
+| **Decoupled**   | • Gas spent on the L2 deposited to pools on the L2. <br> • Rewards on the L1 are issued in a LP token which is demurrage. <br> • Ideally, intermediary tokens are used in the bridge to shield gas and staking token from bridge fallibility. | Pool on the L2         | FFS LP Token; voucher periods can be assigned. | • LP Token's demurrage assigns a lifetime to the viability of the settlement protocol which will require a new or asymptotic token voucher or token generation event at some point. |
+| **Trickle-back** | • Gas spent on the L2 recirculates to pools on the L2. <br> • Rewards are provided on the L2 for bridging back to the L1 at the address of the FFS LP Token. <br> • Ideally, intermediary tokens are used in the bridge to shield gas and staking token from bridge fallibility. | Pool on the L2         | Staking Token               | • Rewards for pool providers to bridge back to the L1 at the address of the FFS LP Token need to be sufficient. <br> • Pool providers likely need to have only custodial access to token, otherwise intended rewarding game theory would likely be intractable. |
+| **Settle with Gas**            | • The settlement messages include the amount of gas spent. The LP for the FFS claims token locked in the L1 bridge. <br> • Bridge may still benefit from intermediary tokens.                              | FFS LP Token        | Staking Token               | • Unclear how to handle forks. <br> • Centralization remains as FFS contract needs to have capability to claim token locked in L1 bridge. However, this can be governed on-chain in this case. <br> • Intermediary tokens can still help reduce attack surface via the bridge. Now attacks would likely focus on exploiting capabilities of the settlement contract to unlock token. |


what means "the LP for the FFS claims token locked in the the L1 bridge"

"The settlement messages include the amount of gas spent." i assume you mean L2gas? i think we should clearly distinguish the gas kinds.. postconfirmers have L1gas cost but earn parts of the L2gas

apenzk · 2024-10-30T20:21:44Z

+| **Trickle-back** | • Gas spent on the L2 recirculates to pools on the L2. <br> • Rewards are provided on the L2 for bridging back to the L1 at the address of the FFS LP Token. <br> • Ideally, intermediary tokens are used in the bridge to shield gas and staking token from bridge fallibility. | Pool on the L2         | Staking Token               | • Rewards for pool providers to bridge back to the L1 at the address of the FFS LP Token need to be sufficient. <br> • Pool providers likely need to have only custodial access to token, otherwise intended rewarding game theory would likely be intractable. |
+| **Settle with Gas**            | • The settlement messages include the amount of gas spent. The LP for the FFS claims token locked in the L1 bridge. <br> • Bridge may still benefit from intermediary tokens.                              | FFS LP Token        | Staking Token               | • Unclear how to handle forks. <br> • Centralization remains as FFS contract needs to have capability to claim token locked in L1 bridge. However, this can be governed on-chain in this case. <br> • Intermediary tokens can still help reduce attack surface via the bridge. Now attacks would likely focus on exploiting capabilities of the settlement contract to unlock token. |
+| **Noble Gasmaster**      | • A trusted signer is responsible for successfully relaying gas spent. The use of intermediary tokens in the bridge is moot.              | FFS LP Token         | Staking Token             | • Centralized signer.   |
+| **Free Execution**      | • Execution does not charge gas fees, meaning gas token stays in circulation and supply does not change.             | Nowhere         | Staking Token or FFS LP Token             | • No sybil resistance; not suitable for anything but promotional or initial stages of a network. <br> Gas fees do not contribute to rewards, so rewards rely solely on initial grant and perhaps LP. Execution does not have cost in token and so cannot drive work-based benefits in token. |


need a reference to a MD / MIP that exlpains Staking Token and FFS LP Token

Co-authored-by: Andreas Penzkofer <36140574+apenzk@users.noreply.github.com>

apenzk · 2024-11-05T15:02:21Z

+| Name              | Description                           | Gas Destination       | Reward Form          | Challenges                          |
+|-------------------|---------------------------------------|-----------------------|----------------------|-------------------------------------|
+| **Decoupled**   | • Gas spent on the L2 deposited to pools on the L2. <br> • Rewards on the L1 are issued in a LP token which is demurrage. <br> • Ideally, intermediary tokens are used in the bridge to shield gas and staking token from bridge fallibility. | Pool on the L2         | FFS LP Token; voucher periods can be assigned. | • LP Token's demurrage assigns a lifetime to the viability of the settlement protocol which will require a new or asymptotic token voucher or token generation event at some point. |
+| **Trickle-back** | • Gas spent on the L2 recirculates to pools on the L2. <br> • Rewards are provided on the L2 for bridging back to the L1 at the address of the FFS LP Token. <br> • Ideally, intermediary tokens are used in the bridge to shield gas and staking token from bridge fallibility. | Pool on the L2         | Staking Token               | • Rewards for pool providers to bridge back to the L1 at the address of the FFS LP Token need to be sufficient. <br> • Pool providers likely need to have only custodial access to token, otherwise intended rewarding game theory would likely be intractable. |


Suggested change

| **Trickle-back** | • Gas spent on the L2 recirculates to pools on the L2. • Rewards are provided on the L2 for bridging back to the L1 at the address of the FFS LP Token. • Ideally, intermediary tokens are used in the bridge to shield gas and staking token from bridge fallibility. | Pool on the L2 | Staking Token | • Rewards for pool providers to bridge back to the L1 at the address of the FFS LP Token need to be sufficient. • Pool providers likely need to have only custodial access to token, otherwise intended rewarding game theory would likely be intractable. |

| **Trickle-back** | • Gas spent on the L2 is collected in pools on the L2. • Rewards are provided on the L2 for bridging the collected pool tokens back to the L1 at the address of the `$RewardToken`. • Ideally, intermediary tokens are used in the bridge to shield gas and staking token from bridge fallibility. | Pool on the L2 | Staking Token | • Rewards for pool providers to bridge back to the L1 at the address of the FFS LP Token need to be sufficient. • Pool providers likely need to have only custodial access to token, otherwise intended rewarding game theory would likely be intractable. |

feat: suggested solutions for fixed token supply.

55393ec

l-monninger changed the title ~~Provide for Atomic Bridge and FFS Usage with Fixed Token Supply~~ MD-38: Provide for Atomic Bridge and FFS Usage with Fixed Token Supply Oct 22, 2024

l-monninger and others added 5 commits October 22, 2024 16:08

fix: clarify.

37862ae

feat: numbering.

c86c75f

feat: embed fixed supply rewarding problem.

52ac7ab

docs: bolding

1b7d247

fix: typo

1ddd68b

apenzk changed the title ~~MD-38: Provide for Atomic Bridge and FFS Usage with Fixed Token Supply~~ [Review] MD-38: Provide for Atomic Bridge and FFS Usage with Fixed Token Supply Oct 28, 2024

apenzk mentioned this pull request Oct 28, 2024

[draft] MIP-40: NB-FFS Decoupled #40

Open

fix: double d3

0d8ca21

l-monninger requested a review from apenzk as a code owner October 29, 2024 07:31

apenzk requested changes Oct 30, 2024

View reviewed changes

l-monninger and others added 5 commits November 5, 2024 10:07

Update MD/md-38/README.md

83fdc18

Co-authored-by: Andreas Penzkofer <36140574+apenzk@users.noreply.github.com>

Update MD/md-38/README.md

16fb6d4

Co-authored-by: Andreas Penzkofer <36140574+apenzk@users.noreply.github.com>

Update MD/md-38/README.md

b29ace0

Co-authored-by: Andreas Penzkofer <36140574+apenzk@users.noreply.github.com>

Update MD/md-38/README.md

2a4f9df

Co-authored-by: Andreas Penzkofer <36140574+apenzk@users.noreply.github.com>

feat: argument of no-action

b220b2f

apenzk reviewed Nov 5, 2024

View reviewed changes

apenzk added 2 commits November 22, 2024 16:41

Merge branch 'main' into l-monninger/provisions-for-fixed-token-supply

0d2d300

native bridge renaming

0cfadad

apenzk changed the title ~~[Review] MD-38: Provide for Atomic Bridge and FFS Usage with Fixed Token Supply~~ [Review] MD-38: Provide for Native Bridge and FFS Usage with Fixed Token Supply Nov 22, 2024

apenzk added the bridge label Nov 25, 2024

		![Fixed Supply Rewarding Problem](./fixed-supply-rewarding-problem.png)


		Using Fast Finality Settlement with the Atomic Bridge presents a challenge when the token supply is fixed.

	Using Fast Finality Settlement with the Atomic Bridge presents a challenge when the token supply is fixed.
	Entertaining different tokens on Layer 1 (L1) and Layer 2 (L2), keeping their added-up supply constant while using the L2-token for fee payment introduces challenges.


		Users of the Movement SDK should be informed of different approaches and trade-offs when using the Atomic Bridge and Fast Finality Settlement with a fixed token supply. This document provides an initial conceptual framing towards that end.

		This document also seeks a response to the concerns raised in [MD-20](https://github.com/movementlabsxyz/MIP/pull/20).


		### D2: Provide Staged Approaches for Networks Building on the Atomic Bridge and Fast Finality Settlement

		User Journey: Developers can reference approaches spawned from these desiderata to understand how to build networks using the Atomic Bridge and Fast Finality Settlement with a fixed token supply in stages. For example, developers can initially roll out a network using the "Free Execution" approach and then transition to the "Noble Attester" approach and so on. Some formats may be more suitable for certain stages of network development and release than others.



		## Overview
		![Fixed Supply Rewarding Problem](./fixed-supply-rewarding-problem.png)

Conversation

l-monninger commented Oct 22, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Uh oh!

ganymedio commented Oct 23, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

0xmovses commented Oct 23, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

l-monninger commented Oct 23, 2024

Uh oh!

ganymedio commented Oct 24, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

franck44 commented Oct 28, 2024

Uh oh!

l-monninger commented Oct 28, 2024

Uh oh!

apenzk commented Oct 28, 2024

Uh oh!

l-monninger commented Oct 29, 2024

Uh oh!

apenzk commented Oct 29, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

apenzk Nov 5, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

l-monninger commented Oct 22, 2024 •

edited

Loading

ganymedio commented Oct 23, 2024 •

edited

Loading

0xmovses commented Oct 23, 2024 •

edited

Loading

ganymedio commented Oct 24, 2024 •

edited

Loading

apenzk commented Oct 29, 2024 •

edited

Loading

apenzk Nov 5, 2024 •

edited

Loading