fix: gov changeable param typo

[caomingpei]

Overview

Commit Summary:

Fix Typo in CIP-13 Regarding consensus.evidence.MaxAgeDuration

Commit Message:

This commit addresses a documentation error in cips/cip-13.md concerning the consensus.evidence.MaxAgeDuration parameter.

The current documentation incorrectly suggests that consensus.evidence.MaxAgeDuration is changeable via governance. This is a discrepancy because consensus.evidence.MaxAgeDuration should be identical with the staking.UnbondingTime, which is not changeable on-chain as indicated by its governance changeability status being set to False.

Additional comment: Allowing consensus.evidence.MaxAgeDuration to be modified on-chain poses a security risk. Such flexibility could enable a node to reduce this duration to under 3 days, which is problematic. In such a scenario, evidence older than this reduced period (less than 3 days) would be deemed invalid, potentially opening a attack window for fraud proofs to be submitted to the rollup. (Especially, the common setting of challenge period of rollup is 3-14 days).

To maintain consistency with staking.UnbondingTime, the parmeter consensus.evidence.MaxAgeDuration should remain unchanged by on-chain governance. This commit fixes the typo.

Checklist

• New and updated code has appropriate documentation
• New and updated code has new and/or updated testing
• Required CI checks are passing
• Visual proof for any user facing features like CLI or documentation updates
• Linked issues closed with keywords

[rootulp]

We shouldn't merge this PR because it isn't a typo. I agree the param likely shouldn't be changeable by governance but it currently is changeable by governance. See this test.

Instead of this PR, we should make a CIP to convert this param from governance modifiable to unmodifiable.

[jcstein]

Agreed on closing this PR. The celestia-app specifications also list it as one that is modifiable by governance.

[caomingpei]

Contributor

Yeah you are right, I agree with you. Simply modifying this CIP would cause the implementation not to match the protocol.
I am interested in these settings (such as consensus.evidence.MaxAgeDuration and consensus.evidence.MaxAgeNumBlocks), because I realized that these settings may introduce security risk (Long Range Attack for PoS). Is there any chance we can come up with a new CIP on this issue?

[jcstein]

Yes, absolutely. Please feel free to do so in the forum: https://forum.celestia.org/c/celestia-improvement-proposal-cip/31

[caomingpei]

Yes, absolutely. Please feel free to do so in the forum: https://forum.celestia.org/c/celestia-improvement-proposal-cip/31

Thanks. I have submitted a placeholder topic, which is under pending. And I will write a draft for discussion as soon as possible.

[jcstein]

what is your ID on the forum @caomingpei ?