Agnostic of the specific projects mentioned here, as the grantees in Gitcoin ecosystem mature in scale and have a verifiable impact, it would be great to see the outcomes codified via impact certificates open for evaluation and linked to subsequent grant applications, like proposed here. This will help establish traceability between results and doers, especially in areas of collaboration, and donors can make informed choices accordingly.
This is still true. Notwithstanding a granteeās tenure with Gitcoin, the application process requires projects to resubmit how funds received previously have been utilized. If you have specific information that indicates otherwise or have feedback to improve the due diligence process, please contribute here for the GG19 Outline and Strategy post.
This gray area between cooperation and collusion is an extremely important topic for all distribution mechanisms that utilize quadratic funding. I am going to draw heavily from this paper, including the concept of cluster-matching that was implemented in GG18, to counter the spectrum between unintentional over-coordination and deliberate collusion.
The paper argues that āthe issues surrounding ācollusionā cannot be addressed simply by outlawing some subjectively- chosen set of ill-intentioned actions. Instead, we should design mechanisms that seek consilience by accounting for a broad spectrum of social connections (not just the ābadā ones)ā.
Accordingly, GG18 matching pool was allocated by utilizing cluster-matching QF. The signals evident in the data for biases and excessive coordination led to reductions in the match of the most suspicious projects by up to 70%, redirecting those funds to other projects. The results without and with cluster-matching are here.
A quick comparison between rounds shows that the Web3 Community and Climate Round had the most āredistributionā of funds due to cluster-match i.e., these two rounds had the most imbalances arising from pre-existing participant relationships. This highlights a greater need for tying impact with funds received making it open for public evaluation.
On the other hand, a cursory look at some of the above projects tells a different story based on the signals in the data, i.e., these projects gained a larger share of the matching pool with cluster-matching QF as a result of having a more distinct set of donors supporting these initiatives in comparison to biases observed with other grantees in Climate Round.