Enhance grant system with allocating money to dependencies and promouters, to increase components financing and visibility

[vporton]

User Story:

1. Components and fundamental science discoveries authors receive little money.
2. Their projects are not enough advertised, among other reasons because software writing and promouting skills are often disjoint and authors may have not enough money to invest.
3. More SEO for GitCoin.co would be useful.

Acceptance Criteria
GIVEN we agree on the amounts shared to promouters and to dependencies.
WHEN we agree on the amounts shared to promouters and to dependencies.
THEN write code.

Possibly, this feature request should be split into two: about dependencies and about promouters.

Tech Details:
To everybody who donated to a project's grant allocate a (fixed) share of future donations from projects that refer to that project (proportional to the amount of donation). This solves the problem of underfinanced components.

Determining referring projects can be automated using APIs of services like pypi.org or crates.io (I realize that it requires substantial computing power.)

Reward these who lead to a GitCoin grant (who led is to be determined by a referral code in URL parameter) by a (fixed) share of future donations from projects that refer to that project. This solves the problem of advertising grants. It also gives more promoting to GitCoin.co site as a whole.

On GitCoin sites and our advertising network(s) display paid banners advertising grants (with referral codes to give back to advertisers). Banners should be selected by auctions.

This way we create (yet small) venture capital (SEO) business for advertising grants. It offloads advertising their grants from authors to professional marketers ruled by the market.

Allow anyone to create a copy of a (his or somebody other's) grant page with different text to try different SEO or different landing page conversion rate. It is especially useful for our "venture capitalists".

Open Questions:
How much to allocate to promouters? (Allocation to promouters should be high, for them to be able to reinvest, and also to solve the main open source problem - underpromouted projects.)

How much to allocate to dependencies?

Notes/Assumptions:
Google repeatedly claims that duplicated text is not a per-se problem for SEO, so allowing our "venture capitalists" to create different versions of the same grant landing page should not be a problem.

[vporton]

If a project (like a component) is unfinished but in perspective highly useful, the above described system creates the incentive for "venture capitalist" to invest into the future of this project.

[vporton]

The donors should be able to donate even to a project (e.g. on crates.io or pypi.org) that is not presented at GitCoin grants. In this case, the entire donation goes to dependencies.

There should be also an easy way to join GitCoin grants by creating a file at .gitcoin/ right on GitHub.

[vporton]

Note that speaking about dependencies I mean both direct and indirect dependencies.

[vporton]

The grant owner should be able to sell a part of his/her grant, to receive money quickly. So the grant owner should receive a fungible token representing "shares" of his grant.

An unprepared grant owner could spend all of his/her grant quickly and lose money. Against this situation I propose the following:

Pay to the grant owner a personal token every second (dynamically calculate the amount of the token on withdraw). From the donated funds make an exchange of such personal tokens to established tokens. If our system gets to know that the grant owner is defunct (dead, if he is a human), stop financing these personal tokens.

[vporton]

Donations address to be specified at GitHub in a file like http://github.com/vporton/vporton/.gitcoin.

[vporton]

It is important that the user can't opt-out from the affiliate program (or set a prohibitively low affiliate share) for his grant, because doing SEO for free projects is even more important than paying the author.

I am author of ordered semigroup actions (a scientific text). Suppose, I put it at GitCoin grant, disable the affiliate program (or set a prohibitively low affiliate share), then I would block development of science, because there is no SEO for the project, but highly specialized scientists will find ordered semigroup actions by highly selective keywords, what will make them to assume that the project is already done (and it is not because of no SEO to make the project useful).

[vporton]

Create software that will automatically find donations recipient address (e.g. for dependencies) by checking an URL like https://github.com/vporton/vporton/donations.json related to browsed software package or scientific article (as reported by Semantic Scholar).

Also create browser sidebar that will leverage the aforementioned software to automatically find donations recipient address.