Requesting community input on the upcoming EA Projects Platformpost by Brendon_Wong · 2018-12-10T17:41:32.212Z · score: 23 (16 votes) · EA · GW · 14 comments
Introduction Benefit 1: Discovery Benefit 2: Evaluation Benefit 3: Coordination Final Points and Related Community Initiatives None 14 comments
Benjamin Pence and I have been working on Altruism.vc, an initiative to make more high EV (expected value) projects possible. We originally focused on increasing funding for high impact ventures, sparked by my initial EA forum post [EA · GW] over the summer. After talking with Good Growth and other community members, we have broadened our scope to project success in general, of which project funding is still a major component.
In October, we announced the EA Angel Group [EA · GW], a group that helps individual EA funders share grant opportunities and grant evaluations with one another. The funders gain the benefits of sharing information and opportunities within a small community. We are interested in bringing these benefits to the EA community at large by creating an online platform that resolves major information and coordination gaps in the high EV project space. Benjamin and I have created an operational prototype of this platform and we are seeking feedback on our design choices.
Benefit 1: Discovery
It is very difficult for individual EA funders to discover projects to fund. With EA Grants, BERI Grants, the newly launched Effective Altruism Foundation Fund, the four EA Funds, the EA Angel Group, Let’s Fund, etc all seeking high impact ventures, we believe a central location for applicants to apply for funding and funders to discover grant opportunities would be very helpful, particularly to EA community members and smaller grantmakers.
The community has been asking for this on an ongoing basis: In the latest Long Term Future Fund grant discussion [EA(p) · GW(p)] in November, the latest EA Grants application round announcement [EA(p) · GW(p)] in September, and in a Facebook discussion on the EA Facebook group earlier this year, community members expressed support for the idea of sharing funding opportunities with the general public or with a subset of EAs.
Currently, it is challenging to request and offer other project-related opportunities such as volunteering, cofounder matching, and advising, as well as other opportunities like office space and web hosting.
We envision that the EA Project Platform will provide “opportunity boards.” For example, in the case of volunteering, a project lead may display volunteer openings or individuals interested in volunteering can specify the kinds of work they are willing to do such as graphic design and translation.
Regarding funding, some EAs have proposed to us that all grant opportunities should be made public, while other EAs have suggested hiding some grant opportunities that could pose risks due to the information they include or their potentially negative outcomes. In order to address these concerns, our current intent is to make our entire platform visible to EA community members only, and not publicly visible or indexable by search engines to encourage greater information sharing and reduce risks.
We will likely also impose a light review process (e.g. other EA funding organizations can delay or reject potentially harmful or concerning projects) or quality check (e.g. requiring at least 1 evaluation) before opportunities are publicly posted. While hiding potentially harmful grant opportunities may reduce risks, posting the rest of the funding opportunities with no way to share information regarding expected impact could be quite ineffective as well. That’s where the evaluation component of our platform comes in.
Benefit 2: Evaluation
Even more problematic than a lack of awareness regarding projects could be a lack of information about a project’s expected impact. Without awareness of expected impact, people could inadvertently execute low EV or even negative EV projects. Funders and volunteers could inadvertently back ineffective or harmful projects, which is especially relevant for an open platform.
We will support project evaluations on the platform. All users or a certain subset of users will be able to create an evaluation and we will strongly emphasize community norms around indicating the rigor that went into the evaluation (e.g. time spent) as well as the depth of relevant knowledge. We believe that publicly sharing all evaluation information makes it easier for potential funders and contributors to get a full sense of all of the relevant considerations around a project. However, a public evaluation system may encounter various pitfalls.
As projects progress and receive feedback, they will likely change in plan or personnel; therefore, the platform will likely strongly couple evaluations of a project with a stage in its development or a certain point in time to encourage project idea iteration, reduce biasing future evaluators, and avoid certain problems (e.g. a self-fulfilling prophecy where a project fails because it did not attract funding because an evaluation predicted it would fail).
Furthermore, it may be valuable to include other measures like anonymous project proposals to be evaluated, anonymous evaluations, and the possibility of limiting evaluation visibility to reduce the negative effects that gaining project feedback at too early of a stage or gaining feedback from the wrong evaluators may have on project success.
Benefit 3: Coordination
Earlier this week, someone pitched a project idea to me without knowing that it was previously an actual EA project with volunteers (a project that I had previously contributed to, in fact). There is currently no easy way for community members to discover active projects, past projects, and project ideas that are in the community. This has undesirable effects such as duplicated efforts, missed collaboration opportunities, and wasted time as people pursue project ideas that have already been tried unsuccessfully in the past.
Our project platform will revolve around projects. It will be easy to find projects that match specific criteria and all their related funding and volunteering opportunities. While we will not remove content from the platform, users will be able to filter by created and last modified metadata to ensure that they are seeing content that is still relevant.
We believe it may be valuable to separate active projects and project ideas. Having a listing of ideas may positively influence the design of projects that are being actively executed as well as help EAs and EA organizations easily judge community sentiment around certain proposals, especially if ideas can be proposed anonymously. Good Growth mentioned to us that supporting project incubation with platform elements like expert-generated ideas, pre-funded project proposals, and pairing project proposals with teams that have the required background and expertise could be highly valuable.
Ideas could improve the evaluation potential of the platform. Project leads looking to get a quick check on community sentiment before starting a project can post an anonymous idea for feedback, and after a project gets off the ground, it can request in-depth evaluations by reaching out to community members and have evaluators post a summary of their evaluation of the project on the platform.
Final Points and Related Community Initiatives
Thanks for reading! Any comments, questions, and evaluations are appreciated. The following paragraphs are my comments on related community initiatives.
It may be the case that an even broader platform that applies to the entire EA community, encompassing all community members and EA organizations, would be even better for increasing the impact of the effective altruism movement. Such a system could make it easy for EAs to share information and collaborate with other EAs in relevant areas, and for the community to see all of the initiatives happening across EA and collectively share evaluations of EA initiatives and provide funding and support. In my mind, such a platform would likely include a distinction between organizations, projects, local groups, ideas, and other entities, as well as provide strong social networking and collaboration features.
Any insights on whether a broader system would be valuable would be appreciated. I believe this is the ultimate vision Rethink Charity’s EA Hub team is working towards. After speaking with the EA Hub, my assessment is that our MVP (minimum viable product) concept and their MVP concept are quite different but may converge in the future; their MVP emphasizes community profiles and geolocation, whereas our MVP emphasizes projects, project evaluations, and project support opportunities. I hope our platform design and launch will provide valuable information for future attempts to improve community coordination with software and we can integrate with or port over all information to future community coordination software initiatives that gain traction.
Regarding more limited applications of our project platform idea, EA Tech Initiatives and the EA Work Club are the only currently active projects I am aware of. The .impact Hackpad is the only past project I am aware of.
EA Tech Initiatives is successful, but it only pertains to technology projects, and all of its project listings and opportunities are in a Google Spreadsheet. Our platform appears to help with the directory functionality of the Google Spreadsheet while enabling the EA Tech Initiatives community to continue collaborating across various mediums like Rethink Charity’s Slack team.
The EA Work Club focuses on work opportunities and differs from our intended platform in many ways. I view our evaluation system as an iteration of its upvote system, and our time-based filtering instead of automatic project listing removal is likely a better implementation that ensures users see relevant content while preserving data that could still be useful.
The .impact Hackpad validated that gathering information about active projects is possible and useful. I believe the ability for our platform to support filtering hundreds or thousands of projects and opportunities based on metadata like cause area, creation date, and stage of project will help it scale and allow users to better discover information.
Comments sorted by top scores.