Hi @Jonas thanks for updating us.
I’m not sure I understood the weighting mechanism correctly. Could you please let me know if my interpretation of the following hypothetical scenarios is correct?
Thanks a lot 
@santicristobal, it’s not the average that counts, but the median. See below an example where there are 17 votes that allocate between 10 OP and 4M OP. The average in this example is 247.6K OP, because the one outlier vote distorts everything. The median is 5K OP, which is exactly the middle value.
Thanks a lot Daniel that’s very clear!
Gm Optimism! Is there any other evaluation process for the projects considered spam? There’re no assumptions about any other consideration besides the application process. And Is it possible to get more detailed feedback about projects?
I believe this is what you are looking for:
Does this means a project with 16 votes wont get any funding?
There are 145 Badgeholders 17 votes is 11%.
If you have 800 projects to evaluate the chances that 17 BH get to your project feels a bit hard.
Whats the rationale behind 17?
Isn’t it better to have a multiplier depending on how many BH voted for a project instead of limiting to a fixed number?
I think this makes it difficult for projects that are not well known. The most popular projects will not have problems, but the less known or with less marketing will be more difficult to get this amount of votes. Also, considering the amount of applications from badgeholders, it will be difficult to evaluate each project in depth.
I also share the opinion that this hard cutoff might become problematic.
A fair compromise could be to incorporate some further considerations: Why not capping the maximum OP that can be received if the threshold of 17 votes is not reached?
Doing so would also address the legitimate concern of preventing a few badgeholders from conspiring with each other to give outsized token allocations to a few projects.
It’s protecting against collusion, where a small number of badgeholders could decide the outcome of OP allocation for a particular project/application.
The voting applications allow badgeholders to sort by, and view, the number of ballots an application is included in. This gives a signal on how likely the application is to meet the quorum.
This way badgeholders have a way to coordinate and try to ensure that all applications meet the quorum.
When introducing a new voting mechanism we want to air on the side of simplicity and a minimal set of assumptions. Each additional consideration has its tradeoffs and attack vectors. There are definitely ways to improve this mechanism in the future, but for our first experiment with a median-based voting mechanism, I think it’s best to keep it as simple as possible.
Forgive me for continuing to insist, but I think that small or less known projects will be harmed and already known projects will benefit. We must think that many PGs are dedicated to Optimism where they are not known in Gitcoin or Giverth. Surely this will discourage new PGs or those who struggled for a while to get to this date. In difinitiva what worries me the most is that this ends up discouraging new PG, which I think is the facinante that has retroPGF.
Connecting in the dots here on the governance forum…
I agree with this post. It does seem that the shuffle option could be set as default. I noticed in your post that it shows A through Z is the default for the filter. You might wanna give the team at agora the heads up!
Forwarded this feedback to Agora 
Supermodular already has shuffle as the default option
Hi everyone!
From this message, specifically point 1, I understand that projects that don’t reach the threshold will also receive funding, but with a discount applied. Is that still the case? In case it is, is it possible to know how the discount is calculated?
On the other hand, I also agree with some community members that 17 seems a bit too high. Would it be possible to know what’s the rationale for it being 17 instead of, let’s say 10 or 25? I think having a little more information on the thoughts behind that parameter would add a lot of value to the discussion.
Last, I think it’s a great opportunity to think of creative ideas or mechanisms to help less well know projects get their voices heard without adding further complexity to the voting mechanisms. Shuffling the order in which projects are displayed in Agora sounds like a smart move. Having the number of ballots included as a sorting option in the voting application is also very interesting.
Best,
Santi
Yes i also would like to question the threshold of 17. Where does it comes from?
I feel that this barrier will mostly affects the projects that are struggling and are not well known, and often the ones that most need it.
17 from 145 is a high number, as others said, especially considering that we have different categories of projects from all around the world and that badgeholders will evaluate more than 1.000 projects.
As mentioned before, i also feel this will only discourage new members that want to apply for RPGF.
This is an interesting experimentation and improvement on the voting mechanics. It only takes 11% of voters to agree that your project created impact, I don’t think this threshold is unreasonable. This is a great way to reduce collusion and help voters distinguish impact makers from noise makers.
Are attestations being used for voters as well? This could be interesting to use retroactively to identify collusion clusters.
The sorting and lists could also encourage more pluralistic voting from badgeholders. This would help them discover projects outside of their immediate circle and smaller, less known, impact creators.
As mentioned in the community call last night, there is a slight conflict in that if a badgeholder thinks a project should not receive any funding, the way to signal this is by voting for a 0 allocation. (Abstaining would signal indifference and hand over power to those who want to fund the project).
But a “0 vote” counts towards the quorum of 17 votes that would allow the project to receive funding.
Effectively, as I understand it, a project receiving 16 “0 votes” and one ‘friend’ vote for a OP 5M allocation would end up being allocated quite a lot of money.
I think the obvious solution would be to discount clear “0 votes” when it comes to reaching the quorum.
I agree, by the way, that the goal should not be for any application to not meet the quorum.
However, I think it would be better to allow for a smaller quorum - but only reach it through votes that are actually in favour of funding the project.
As you said, the goal should be for all applications to meet the quorum!
This is not the case, as we’re applying the median. In a case of 16 “0 votes” and one “5,000,000” vote, the result would be 0.
See @danftz’s graphic on this
Ah, you are right.
That makes a big difference. I guess that is good enough for now.
Thanks for clearing that up. 
16 non-zero votes doesn’t meet quorum but adding a zero vote then means the project meets quorum, so there is a risk to voting zero.
That’s a great proposal. One thing I’m not clear about is how we can assess the worth of the product. There are multiple criteria available, but only a few people are actually considered “worth it.” Are there any examples to properly assess these submitted projects?
