Grants for voting neurons

Hey,

I’d like to inform you about the outcome of our discussion regarding the estimated effort and grant size for the Subnet Management topic.

We acknowledge that in the coming months, the workload required to perform reviews will be higher than initially anticipated. Therefore, we propose increasing the estimated effort and grant size as follows:

  • Effort: 2h/week → ~4h/week
  • Grant: $750/month → $1500/month

These numbers are based on a rough estimate of 50 proposals per month, with 20 minutes allocated per proposal review. Going forward, we can further adjust these estimates if necessary.

Currently, we have two entities participating in this topic who would be affected as follows:

  • CodeGov (Team Grant) – previously $1500, with the new compensation of $3000
  • Alex Lorimer – previously $750, with the new compensation of $1500

These changes are planned to take effect for the rewards of the upcoming review period in January.

Please don’t hesitate to reach out with any further questions or comments.

Best regards,
Moritz

9 Likes

Thanks @cryptoschindler, this will make a big difference!

3 Likes

Thanks again Lara, I’d like take a moment to highlight some concerns I have while you’re reviewing how best to move forward with this.

I’ve been putting hard work, time and passion into Subnet Management reviews for a fairly long time now (prior to the grants kicking in). I’ve sampled my effective VP over time (both voting at the beginning and towards the end of the voting period). Once you discount Synapse (I’m a member of Synapse) there’s been no noticeable change in my follower-based-VP. I think this is important to note given that an important purpose of these grants so far has been:

  • The funded neurons […] establish ways to convince potential followers that they do good work, for example by sharing how they verify and vote on proposals
  • The NNS community […] gets to know new voting neurons that can be followed already now or taken into account when periodic confirmation is introduced

so that (most of) the funded neurons can move to the long-term solution for incentivising voting neurons when it is available

It’s important to discount Synapse’ influence. It holds a lot of VP, and has clear preferences about who to follow (it’s not unlikely that the current configuration will change soon if certain members continue to push their agent and have their way - Proposal: 134722, Proposal: 134654).


My concern is that once the transition to the long-term solution begins (reviewer rewards based on followership), incentivisation will drop, and the barrier to entry for anyone new will be too high and hard to break through. I don’t expect periodic confirmation to do anything but require users to click the easiest buttons for them once every 6 months (it takes hard work for people to change their behaviours and to research whether they should change who they follow - there’s no clear incentive for them to do this).

I have some suggestions about how to address this that I’d like to put to the community, which I’ll plan to post after work later, or in the next day or two. It would be interesting to see if anyone else shares my concerns first.

4 Likes

Hey @lara I would like to echo the primary concern that @Lorimer has raised here. I still believe that there are no real incentives for people to follow any entities other than DFINITY on the technical topics. This concern was first raised by @dfisher earlier in this thread and I followed up his comment expressing the same concerns. You also followed up with your own thoughts on the topic at the time and I provided an example of an easy step that could be considered to address the issue.

In my observation, CodeGov has seen a percent level increase in followers over the last 4 months, but it has not been as significant as you might hope given how engaged we are with technical proposal reviews on all 5 technical topics (IC-OS Version Election, Protocol Canister Management, Subnet Management, Node Admin, and Participant Management). I think this is another example to go along with the concern raised by @Lorimer about his own observations of voting power increase.

That said, I believe that DFINITY has plans to improve the workflow for known neuron registration that will enable us to claim a specialty in certain proposal topics. This change along with an update to the NNS dApp UI to reveal known neurons specific to each proposal topic could have an impact on decentralizing Followee selection. I hope that is still in the plans and look forward to seeing if it makes an impact, especially as neuron owners are asked to routinely confirm following.

Regardless, I still think it is likely that there will need to be some incentive put in place to encourage people to select Followees other than DFINITY. The natural tendency is for people to follow the dev team on any project. This has been true for every SNS and it has proven true for the NNS even after a concerted effort to remove default following on the Governance topic. Two years ago there were measurably only about 2% of total voting power in the NNS choosing to follow DFINITY on the Governance topic. Today, there is close to 20% following DFINITY. This is why DFINITY now triggers over half the total voting power that typically votes on Governance and SNS & Neuron’s Fund proposals when they vote. If decentralization is the goal, then people need a reason to choose someone other than DFINITY. As described by @dfisher, the risk/reward for following someone other than DFINITY is not currently worth it to most NNS participants.

To be clear, I believe it is important for there to be diverse options on who to follow on every NNS proposal topic. I have no interest in accumulating a high percentage of voting power for CodeGov. I would love to see a substantial amount of voting power being triggered by many of the known neurons who put themselves out there as Followee options. CodeGov should be one of many and I am all for changes that will have an impact on how many known neurons are reviewing technical topics and how much voting power they are able to trigger.

To be clear, the certain member that @Lorimer is talking about here is me. I am an original founding member of the Synapse neuron and fully led the effort to organize the formation of that known neuron. I have also taken initiative throughout our 3 year voting history to ensure we have well known and credible voting members and that we stay on top of voting on every proposal topic, especially Governance and SNS & Neuron’s Fund proposals. In that 3 years history, of course we have unintentionally missed some votes. However, when that happens we take steps to reconcile and improve our performance. I expect that I will always be motivated to lead and manage the group the best I can. However, I don’t control the neuron and the decisions we make. I proposed (134722) that we remove @Lorimer as a Followee for the Subnet Management proposal topic because of his high frequency of reject votes on that proposal topic over the last 4 months. He has voted to reject 34.4% out of 163 Subnet Management proposals since the Grants for Voting Neurons program began. In contrast, CodeGov has voted to reject 4.9% and DFINITY has voted to reject 3.7%. Since Synapse follows both CodeGov and Lorimer, it means that Synapse has voted to reject 34.4% of these proposals as well. In my opinion, this performance does not reflect well on the reliability and credibility of the Synapse decision on who to follow on this topic. Hence, I proposed that we make a change. The other voting members for Synapse did not agree, which is why that proposal did not pass. There was a consensus of our Followees on the Neuron Management proposal topic who are ok with our high reject rate on this topic. It was a helpful exercise to better understand the opinions from our diverse and well respected voting members on how the Synapse neuron should be configured.

2 Likes

Out of interest Wenzel, how much VP do you think one person (such as yourself) should really be in charge of? I don’t mean to be confrontational with this. I think it’s an important question to discuss out in the open in the context of these grants.

CodeGov has acquired a name for itself. People follow names because it’s easy. This is what breeds centralisation, and stifles decentralisation - in my opinion.

Thanks Moritz. This is a big help indeed. Will this apply to the month just completed?

@cryptoschindler

Hi @Lorimer and @wpb and thanks for your inputs.
Since this is already a huge thread and mainly focused on the voting grants, I suggest that the conversation around incentives is moved elsewhere. In the coming months we plan to discuss how known neurons can better be incentivized and I think it would be best to holistically consider the aspects you brought up together with this.

I think I understand most of your points.
I suggest that as part of this conversation we also think about to what extent we want to try to influence the voting behavior by governance design and to what extent this is more about processes around the voting (making it a habit that new voters see the options who to follow, find good known neurons etc - as someone mentioned perhaps supported by the frontends). I say this because I also think governance is already fairly complex and I fear that by including more incentive systems etc we don’t make it easier to understand. But let’s discuss it when we get there!

2 Likes

Dear Community,

I’d like to share an update regarding Aviate Labs’ participation in the Known Neuron Grant Program. Until now, we have participated as a team, with @quint, @roald-av8 and myself serving as reviewers.

As our priorities have shifted and with a deeper understanding of the scope of work involved in the program, we have decided to transition to an Individual Grant (instead of a team grant - which is how we have been operating thus far). This means that moving forward, I will be the sole reviewer of proposals for Participant Management and Node Admin topics on behalf of Aviate Labs.

We want to assure you that this change will not impact the integrity or quality of the proposal review process. Our commitment to supporting initiatives that strengthen the Internet Computer ecosystem remains as strong as ever.

If you have any questions or concerns about this transition, please feel free to reach out. We are here to ensure clarity and address any inquiries from the community.

Thank you for your continued trust and support.

cc: @cryptoschindler @lara

7 Likes

This change seems reasonable to me @louisevelayo. Thanks for the update. If you don’t mind sharing (which you don’t have to do), does this mean that the Aviate Labs known neuron will solely follow you on these two topics or does it continue following you, @roald-av8, and @quint but they just vote based off of your review? I’m asking out of curiosity simply because I think it’s interesting to learn how different people configure their neurons. It makes no difference to me how the known neuron vote is actually executed as long as it is credible and reliable, which I’m sure is the case.

1 Like

Hey @wpb ,

Thanks for your question. The Aviate Labs Known Neuron will only follow me for voting on the topics of Participant Management and Node Admin, so it will only cast votes for these topics. For the rest of the topics, this Neuron does NOT vote.

1 Like

Do U C Centralisation? :eyes: IC Decentralisation (in Progress…) + ICP Giveaway!

1 Like

Dear Community + Grants Committee,

I’ve been working hard over the last few months, striving to produce the highest quality Subnet Management reviews that I can, while refining tooling to make it easier to do so. The Grants for Voting Neurons initiative has facilitated these efforts, for which I am very grateful :pray:

I’ve recently been in discussions with likeminded, recognisable members of the community who are in the process of refining their own tooling, covering a broader area of Subnet Management concerns than what my reviews cover alone.

In the interest of increasing decentralisation, and broadening and refining IC expertise in the community, I’d like to request that my individual grant for Subnet Management be upgraded to a team grant, to accommodate two team members (a team of 3 in total).

Both are experienced devs. @aligatorr89 is particularly active in the TAGGR community, contributing to the codebase and providing diligent code reviews. @MalithHatananchchige is also a node provider himself, and brings a wealth of knowledge to the team, along with his software development experience.

All aspects of our team are to be decentralised, including access to grant funds, for which we intend to use a threshold-controlled canister to divide up and distribute the monthly allocations.

We would like to announce a decentralised team neuron soon, and over time we would like to build our team and focus on more topics than just Subnet Management alone. In the meantime, please share any concerns you may have and/or ask any questions. I hope it’s considered in the interest of the IC as a whole for my individual grant to be upgraded to accommodate a decentralised team.

Thank you for your support

cc @cryptoschindler, @lara, @marc0olo

6 Likes

This is awesome. I fully support this change. I’m somewhat familiar with the active contributions that @aligatorr89 has made in the Taggr network (both code contributions and code/proposal reviews) and I have recently observed @MalithHatananchchige contributing to technical discussions (example) related to subnet management, node admin, and participant management topics. Both of these folks would make strong contributors to the team that @Lorimer is proposing for his grant. I would very much like to see DFINITY support this change. If it needs further justification from a cost perspective, perhaps the recent reduction in grant for the Aviate Labs team to just @louisevelayo for the node admin/participant management is a sufficient offset of the request to expand Alex’s grant allocation. This would be a net win for the IC and decentralization to allow this change to occur.

5 Likes

Thanks for your kind support!

2 Likes

@wpb appreciate the support

3 Likes