Proposal: Validator Resource Center and Ranking Website for Kusama (Phase 2)
Phase 2 of the Validator Resource Center and Ranking Website (VRC) further improves upon the ranking method from Phase 1, by making the scoring model more customizable and thereby better representing the user’s preferences. Also, the VRC score is now a major metric in the manual selection process. Additionally, we include several new metrics (e.g. performance based on historic era data) and thereby reduce the load of information on nominators. A new filtering technique helps to reduce the overall set of validators based on their quantitative properties and make it easier for nominators to make an appropriate selection.
Aside from that, Phase 2 introduces a staking dashboard, which gives important visual and numerical information about the network, allowing comparisons of sets to the state of the network. It will be even possible to import your current nominations (via the polkadot.js extension) and check whether some critical events happened to nominated validators (e.g., an increase of commission or a slash).
You can review the full proposal here. Please leave your comments!
Code-of-conduct for the implementation
We are aware that this project bears a large responsibility towards nominators and validators. Therefore, every decision on this implementation follows these principles as much as possible:
-
Act in the interest of the network as much as possible.
-
Act in the interest of the nominator whenever possible.
-
Objectivity on validators selection as much as possible. Don't discriminate against individual validators.
-
We value user preferences and try to give them as much freedom as possible while maintaining the trade-off with usability.
A link to the last proposal can be found here.
Comments (8)
Great! The benefits of this tool are out of any doubt. Congrats everyone for current achievments. It is already a nice tool, and it will be much better in a couple of months.
This are my 2 cents of feedback for the new milestone:
I think this is something we could educate in to our nominators. It is so important for the network security that you are required 10k DOT of self-stake to participate in 1k validators program.
The bigger the cluster, the lesser visibility it deserves. We need to face centralization harder. Spite it implies some hot discussions I see a big value on acting in favour of decentralization. We must not surrender in front of short term clusters profits, and we have here a chance to differenciate this network from many others competitors in the internet of blockchains.
The way it is currently calculated it excludes and punishes validators that have been recently inactive. Era points average should be calculated taking into account only the active eras, not all eras. Something similar happens to autopayout. We should not punish a valitador for not asking rewards in its not active eras.
Keep going with this incredible job!
Hi derfredy
thank you very much for the feedback :) Let me address your points:
"Skin-in-the-game", i.e. self-stake is a major metric in the VRC 2.0 and therefore should receive the attention it deserves. We will include information boxes which further explains the metrics and also stresses their importance. What do you exactly mean with "skin-in-the-game check"? With regard to your comment to the 1k-Validator Program: We don't have an influence on the rules there.
VRC 2.0 offers a way to exclude all but one cluster members which hopefully increase decentralization, while also honoring larger operations who are currently doing a good job. The platform is thought to be as neutral as possible and therefore it only offers the choice to exclude larger operators (all nodes but one). It is important to stay fair towards larger operators as they are doing also a very good and important job. We think this is a good middle-way. Additionally, it does not makes sense to overly punish larger operators as there are easy (but not cost-less) ways to split identities.
The calculation of the average era points takes into account that some validators have not been active in some of the historic eras. In that situation, a validator receives the average era points of that era and therefore the final "average era points" should be comparable, irrespective of whether a validator was not active in a few eras. With respect to the autopayouts that's a viable point and we should have a look into how it is currently done.
Thanks again for your feedback!