Evidence-0002: Retention at Rank III

Report Date	2024/12/02
Submitted by	ordian

Member details

• Polkadot address: 13aYUFHB3umoPoxBEAHSv451iR3RpsNi3t5yBZjX2trCtTp6
• Current rank: III
• Date of initial induction: Seeding
• Date of last report: 2024/06/21
• Area(s) of Expertise/Interest: Parachains Consensus

Reporting period

• Start date: 2024/06/22
• End date: 2024/12/02

Salaryless

As you can see here, I haven't claimed any fellowship salary and not going to in the next salary period as well.

Evidence

ParaInherent.enter weights

I've investigated a long-time issue about ParaInherent.enter weight being overestimated. It was one of the remaining blockers allowing Polkadot to scale to 1k validators and 200 cores.

Along with resolving the issue in the following PRs: 1, 2, a long-standing low-impact security issue have been addressed as well.

While digging into the ParaInherent.enter code, which serves as a critical on-chain part of shared security aspect of Polkadot (Parachains Consensus), a few refactoring ideas have been suggested to make the code more resilient and future-proof.

Runtime API calling unmigrated storage

After our team have investigated a finality issue that happened on Polkadot after a runtime upgrade, we have decided to prioritize fixing another long-standing issue.

Looking into the code that is outside of my expertise (substrate executor logic), I've created a fix. The PR has been later adjusted to the newly opened RFC 123 and is looking for feedback and reviews.

Spammening

I have provided help with the (parachains) spammening project spanning across multiple areas

• Getting funds to register 30 parachains. This has proven to be a challenging task involving multiple teams at Parity, such as Finance and Security. But as a result, our team got some experience working with multisigs, pure accounts and proxies. Multix added support for coretime chains thanks to another fellowship member. Hopefully, our setup will serve as a reference for similar future endeavors.
• Working on a browser script, an issue with subxt_signer on no_std platform has been discovered and later fixed.
• Working on a feeless extrinsic, some limitations have been discovered.
• I've worked on several subxt scripts (e.g. coretime orchestration) and gained insights into constructing nested subxt calls dynamically, coretime inner workings and API among other things.

After the relay chain spammening, I've also started an investigation into potential overweight issue:

• https://github.com/ordian/replay-kusama-25876038

Misc

As usual, I'm providing PR reviews and direction around my area of expertise.

Voting record

Ranks	Activity thresholds	Agreement thresholds	Member's voting activities	Comments
III	70%	100%	During this period, I have voted on 75 out of 81 referenda in which I was eligible to vote (i.e 92.5% voting activity) with 100% agreement with higher ranks.	My overall voting activity is at 75.5% according to this script.
I	90%	N/A
II	80%	N/A
III	70%	100%
IV	60%	90%
V	50%	80%
VI	40%	70%