If only we had an algorithm for efficiently allocating capital… :trollface:
In all seriousness though, this is something we’ve talked about a lot at clr.fund. The TL;DR is nested matching rounds. Recipients at the the top level are the categories listed above, which are each distinct matching rounds with recipient projects within their respective categories.
This way, rather than coming up with some arbitrary (and ultimately biased) capital allocation for the categories, we can just let quadratic finding work its magic.
It also provides a nice opportunity for folks who want to contribute broadly to a category, but don’t know the best way to allocate to projects within then category.
We haven’t implemented this for clr.fund yet because we’re not yet at a scale where it makes sense, but I think Gitcoin Grants is well and truly beyond the scale where this would be the most efficient way to decide the matching pool allocation between categories.