Independent Validators Support

Ecosystem
,
1

Currently, any validator with 7,000+ KSM nominations should be 50% active.

At today’s KSM price ($14.30), 7,000 KSM = $100k stake - which is outside reach of the average validator operator …

The larger validator operators (with 10+ validators) dominate the list of validators, and seem to attract all the nominations) - it’s hard for indie operators to get visibility.

The Decentralised Nodes (DN) program is operated by web3 foundation to “Help Make Polkadot the Most Decentralized and Secure Network”. And DN Cohort 3 supports 99 DOT & 194 KSM validators.

Full disclosure: Interweb was selected for cohorts 1, 2 and 2a. I was not selected for cohort 3, and shortly after losing the DN nominations, most of my validators a now inactive. I plan to apply for the next cohort (each cohort lasts 4 months).

As you can imagine, each cohort is over-subscribed and there are many independent operators that don’t get selected. These operators have to wait months - inactive - to apply for the next cohort.

I propose to start a fund (my own small stash, contributions from other operators, and poss. a loan from treasury) to support nomination of indie validators. If successful, this initiative could morph into a community operated DAO.

This initiative will augment the DN program, with some differences.

  • lead by the community
  • require identity, not require KYC
  • limit applicants who already have multiple active validators outside the program
  • incentivise validators to be independent (i.e. lower commission is better, higher self-stake is better)
  • round-robin selection algorithm / simple, rules-based scoring
  • exclude VOps who are selected for the current DN cohort

How could it work technically?

To start with, starting on Kusama, I propose a combination of

Nomination Pool:

  • INTERWEB POOL (50 on Kusama) will support this scheme
  • Anyone can support by
    • joining the pool, or
    • Granting a Staking Proxy to the control account

Nomination Proxy Control:

  • a nomination proxy account will manage the nominations for all participants
  • according to a calculated rotation, the nominations will move to the next validator

Joining:

  • VOps register a node and corresponding self-stake stash
  • Self-stake has Staking proxy granted to a Nomination Proxy Control account
  • Nomination Proxy Control account manages nominations on a round-robin basis
    • Higher stake = longer in the round-robin…

Leaving:

  • Operators simply remove Staking proxy from stash - e.g. if they get selected for DN
  • Nominators leave the pool or revoke Staking Proxy

What if the fund is not sufficient to support at least 1 validator? In this case I propose to nominate a selection of validators from DN to ensure continued staking rewards.

There are many finer details to work through, and I would love to hear from others about their opinion on this.

Thanks

2

Your Polkadot validators were recently slashed for double signing which is a violation of the DN Program and this means that you should not be accepted anymore. Why should we support you to get in the active set?

3

It’s a fair observation that one INTERWEB validator was slashed. I provided a full explanation why this happened in the DN channel, and the slash was refunded to the nominator.

I don’t believe a slash is a disqualification from all future cohorts, the rules state: “Any slashing events in the previous 4 months.”

I have contributed to the ecosystem in various forms for some years - I built tools to help validators in 1KV and DN. This post is not about me specifically, but about the broader topic of supporting indie operators.

Yes, I could benefit as a participant, but I don’t plan to charge anything for my time. If anyone else would step forward to promote the idea, that would also be great! If not, then I have relevant skills to make it happen.

I have been contacted by other operators who support this idea, yet individually we all lack sufficient stash to make it happen.

To answer your question: Why should you support indie validators? Because it promotes network security, and rewards the people who invest substantial time and hardware to keep the network healthy.

4

I also think it would be good to get feedback on what metrics should be used to evaluate possible candidates for nomination:

  • 1KV had a detailed scoring system that was easy to understand, and technically straightforward to measure
  • DN has rules that require offline response (eg KYC and proof of admin experience) and ‘scoring’ is done on a private backend

At the least, I would suggest:

  • validator performance when active, as measured by the turboflakes app
  • slashes
  • identity - the goal is to support known indie operators. All nodes must be tied to a verified ID
  • hardware - no hosting on prohibited providers
  • updates - 24 hours grace to upgrade to latest release
  • telemetry

Nice to have would be geographic location and operating system diversity.

5

The technical implementation for this algo present quite a challenge. Essentially, we need a combination of on-chain control and off-chain data.

The on-chain parts can come from existing features like proxy for Staking. A proxy account will set the nominations for all the stashes that participate.

The off-chain data needed depends on the rules we adopt

  • performance - turboflakes api / other?
  • identity - People chain
  • rotation of nominations - some storage (on-chain in a smart contract/off chain in the oracle)

I envisage an external oracle service, trusted, open source that pushes data to a smart contract, which in turn makes calls to rotate nominations to the validators.

Quite a ‘fun’ project: smart contract, XCM, multi-chain, oracle/monitoring service

6

Hello all,
first of all i wanna thank Derek for planning this and sharing it.
I appreciate someone outlining clearly the current challenges for independent Kusama validator operators, this environment ain’t easy for smaller indie guys.
I don’t wanna copy and paste everything Derek wrote down, I’m simply saying that i agree with this idea and i it would be nice to have some funds in order to implement it.
Also, i know i he got slashed recently, but i also know he always helped the community in different ways from tools to knowledge.

1 Like