I am in full support of this, a pruning of inactive members is sorely needed plus this has the additional benefit of potentially encouraging people to keep an eye on democracy a bit more even without being prompted.

Feel a little over complicated. How about this:

1. There are 10 motions in previous 3 election rounds
2. Councilor A missed 7 motions voting which >= 2/3 * 10
3. Councilor A can not be candidate for next 2 election rounds

Of course, numbers above are just examples. A abstain call can still be added.

Sharing some interesting points about this initiative during a Council discussion session on Kusama Direction Channel:

In general terms, all Councillors agree that there is not enough pressure on Council seats for participation, but this might be solved long-term by increasing voter participation: The current system allows voters to "set it and forget it". For this reason, some councillors have suggested possible solutions to take action:

A Council Participation Website

The general opinion is that a Council participation website on one of the explorers or on one of the ongoing projects funded by the treasury, like Dotreasury, could help in determining participation levels from different councillors. This website should record past voting and council members centric data (checking that proposeMotion calls or the use of proxies/batches are taken into account). This page should also include personalised information and a list of councillors the user's address is voting for, with all info on participation records - this would help voters decide on who to nominate.

Alternatives to the use of negative incentives

Many councillors believe that the initial approach described on this post incentivises fast participation vs. non-discussion, and is detrimental to the overall process. Many believe that some alternatives, in combination or not with this initial approach, can be:

A. expiry of votes after a certain period;

B. candidacies expiry after a certain period with some protections to avoid fast-cycling;

C. set up a system that puts steady pressure on stakeholders to re-evaluate preferences in vote, by discounting votes over time using some kind of curve (this info will be extended, as it seems to be the preferred option for many councillors).

D. Ponderate votes by voting activity, playing with the idea of dynamic seats.

The goal of these alternatives is to make less-active stakeholders less opinionated than active stakeholders and as a candidate you would see your support slipping away progressively and would need to gather more support behind your seat. An issue with the first two options (A and B) seems to be the risk of low-backed councillors stepping in, opening the door for malicious behaviour.

Regarding negative incentives (something this initial approach relies on), Council members mentioned they could be detrimental to the design: if we strike Council members punishing them for inaction, we would need a steady flow of runners-up to avoid low-backed seats taking over, as well as ways to avoid fast participation with no discussion. A particular comment on negative incentives suggested to focus them solely on quantitative virtues of councillors for which there is broad consensus: a valid point for the design and reflected on the Council participation website idea above.

A short-term solution until expanding on this initial approach needs to focus NOT on Council participation, but voter participation: In this direction, Shawn Tabrizi's Lotteries pallet is a good possibility to experiment with on Kusama.

Happy to hear comments and questions about this summary, please add anything I might have forgotten to include.

This is a good read on the topic of governance, @rrtti-5220 .

Personally, I'm not a fan of negative incentives because there is always room for error when slashing/banning someone who might be perceived as "bad actor" by the code while making genuine human mistakes. A moment of inattention or confusion shouldn't be too harshly punished from the get-go, especially in a space where things move very fast and not everybody is always able to keep up for x and y reasons.

The main issue that I see with the Lottery pallets is that it once again shifts the onus onto stakeholders/users to participate. Meanwhile, I think that Councillors should be the ones making the effort to pay attention to the stakeholders that they supposedly represent, in order to further secure their council seat...at least, in first instance.

From my perspective, there are already too many obstacles to voter's participation in council elections:

1. Understanding the phragmen election algorithm
2. Understanding the delegation process
3. Researching individual councillor's whereabouts
4. Securing enough KSM/DOT to pay transaction fees
5. Pay voting bond that could otherwise be staked to get rewards

I'm not even mentioning the fact that any mistake (i.e rejected calls) in this process costs some chunk of money already at current prices.

I think rewarding councillors who participate regularly and "pruning" (possibly openly downgrading or flagging their reputation without slashing/banning) those who don't within a set number of eras is the way to go. But this has to happen within the existing setup, without requiring any more effort from users/stakeholders.

All in all, the core of the issue is "How do we optimise current processes by cementing councillors' participation?". Dealing with "How can we get more users to participate to council elections?" once the council is already running as democratically competitive as intended will then become easier, because people will SEE that their vote has an impact on individual Councillors' seat and reputation.

For now, I'm going to dive into the code to see what is happening there before I post any other suggestions on this matter.

Thanks again for looking into this problem on behalf of the community. :)

Edit: typos and addendum.

Thanks for the comment, @anaelleltd!

I think the plan outlined here has a different goal than lotteries. As I see it, we have two main issues on the governance side of things: 1. councillors who are lazy and 2. part of the community that does not vote.

For the first problem, the post above with a first initial approach to restructure council is suitable. I have heard many opinions on negative incentives - i think for stakeholders who have 1. decided to submit their candidacy to council, and 2. taken the steps to ask people to vote for them, negative incentives are suitable if theyve become lazy. I get the argument of "inattention or confusion shouldn't be too harshly punished from the get-go" but the expectation for council members is to be available for discussion when needed, and if not, to simply renounce the seat.

For the community vote issue, we need to go a bit beyond: i thin negative incentives is obvs not a reasonable solution here, thats why lotteries might be able to help there and appears as a potential way to experiment. Lotteries adds a positive incentive to participation focusing on specific calls (let's say we open a lottery for referenda voting, when/if a member votes for any proposal in referenda queue they will be able to buy a lottery ticket and participate in the open lottery).

I am studying some alternative options council members gave input on to try and restructure council vote based on participation of councillors, I will keep you updated here. You can always open a chat on Element to discuss this! My ID is @raul.rtti:matrix.parity.io