At the end of the recent Haskell.love conference, there was a lively exchange of ideas on the ways in which the Haskell Foundation might best use its monetary funds to grow the Haskell ecosystem via grants.
I would like to use this post to record the small ideas and thoughts that came to me afterwards. This post is not meant to provide a comprehensive summary of the above discussion in any way, I would merely like to provide additional input, just in case it may be of use in exploring the design space. I’m not a member of the Haskell Foundation. In no particular order:
- Earmarking donations.
- As there are several design options for community grant program (small grants vs. large grants, acceptance criteria), deciding between them seems difficult. One idea to rule them all could be to set up multiple funding pots, each implementing one of the design options. For example, there could be one pot for small grants, pot pool for smaller grants, with slightly different rules; donors can choose to donate to one of their preferred models.
- Cave: Too much earmarking reduces the impact of donations, because it reduces the ability to pool money.
- Non-monetary rewards
- To increase the value that the Haskell community can exchange for contributions, it makes sense to think about value that is hard to measure in terms of money.
- Community recognition. For example, trophies that you can put on your desk would be a nice touch. That said, there was a sense that the community is already very good at recognition and enthusiasm.
- Haskell itself is valuable. For example, programming in Haskell is also a hobby for me, and the language allows me to control my computer ways that no other language allows me to do. For this ability, I’m happy to give back libraries that enable others to do the same. Measuring Haskell in monetary value is difficult — for how much can you buy a compiler extension?
- Mutual exclusion of ethical values
- It occurred to me that different choices for how to run a grant program can be mutually exclusive in a very strong sense: Choosing one way to run a grants program may be counter to the “ethical” values of running it in a different way. For example, it is is not credible for an oil refinery company to run a grant program on solar panels — one value excludes the other.
- Richard Eisenberg expressed some worry that people would reduce their voluntary contributions if there was money on the table. I think this may be an example of something that touches the “mutually exclusive” issue.
- Non-fungible tokens (NFTs) as quasi-monetary rewards
- Even though I presented this as a “terrible idea”, I would like to record it here anyway because it may prompt better ideas. The reason why it may be a terrible idea is because it touches on mutual exclusivity of values.
- An NFT can be described as a unique digital trophy that can be sold by the recipient (repeatedly even). For example, the main implementor of the “FoobarTypes” GHC extension could get a digital trophy for this contribution, which they can sell on the market (and even get profits from future sales). Like a painting by Picasso, this trophy might become very valuable.
- The unique selling point of this idea is that it decouples value from time. At the time of creation, we may simply not know how valuable the “FoobarTypes” contribution will be. Spending money on this contributions carries the inherent risk of the contribution not working out as intended. In contract, the digital trophy makes it possible to appraise the value of the contribution long after its creation.
- Small grants
- There were some ideas floating around about small grants that are not enough to pay the bills, but can still make a difference.
- It occurred to me that there can be phases in life where an individual can only accept small sums of money. For example, unemployment insurance in Germany only allows 165 EUR per month side income before the insurance payout is reduced. University students often find themselves in similar situations. These may be the times where small sums of money do the most good, by leveraging an existing social net.
- Large grants
- Some infrastructure may require sizeable grants that do pay the bills. For example, the executive director of the Haskell Foundation is a paid full-time job.
- I strongly caution against using the existing system of grant funding in science as a blueprint; it is one of the reasons why I left science. The system is very inefficient in that a lot of superfluous effort is spent on writing and optimizing the grant application, and a lot of effort is spent on reviewing the applications. See also: Dr. No Money: The Broken Science Funding System
- Novel ideas that may work better are:
- Grant lotteries. The existing science funding system is not very far from a lottery anyway.
- “Fund people rather than projects”, that is individuals with a track record. See SciAm article above and Wellcome Trust […].
- Cave: Scientific funding differs from software funding in that science has an inherent risk of unknown outcomes. But I would argue that delivering quality software is also more risky than we would like it to be.