Create a platform for assurance contracts that looks nice.

Longer description of your proposed project

This project will be to make a website that has assurance contracts, of the basic form if X people sign by time Y then the contract activates, else it is void. Assurance contracts that work solve a lot of multiagent coordination problems by indicating to people when it's not personally costly for them to move to the better equilibrium.

Manifold has got surprisingly far importing prediction markets to Earth; I want to try a similar thing with assurance contracts. I am going to first try doing the simple thing, with no stakes other than a simultaneous reminder ping, which seems sufficient for some acts of coordination; if it seems like adding further mechanism design like involving money or smart contracts might be helpful given evidence from this, I might go on to do that.

I think that one reason why simple cheap assurance contracts haven’t took off is that there hasn’t yet been a service for creating them that has a UI that’s good enough; to get more than a handful of users, I think a platform

must parse as a legitimate website and not a phishing scam, by the standards of 2023 webdev

must let users unilaterally create new assurance contracts

must be built on something that lets the developers rapidly iterate if their first try at UX doesn't seem to quite be achieving this

Spartacus (https://www.lesswrong.com/posts/98HihPxpoRFja2JNF/is-there-an-assurance-contract-website-in-work?commentId=FEDDgmpFed7h3XApn) has tried to do this, and has been under construction for two years with an Airtable as UI.

CollAction (https://www.collaction.org/) has a nice UI but the existing projects are all routed through the CollAction team which creates a barrier to use and I think prevents the creation of silly fun small projects that would help the project grow with a small userbase.

Actuator is now defunct, but when it existed had a minimal UI (unstyled HTML tables) and did not get the ten users to coordinate on advertising Actuator.

Bad UI is also probably why my prediction market platform (announced here https://www.lesswrong.com/posts/GHc2cqR6CfKumg6Be/tetraspace-grouping-s-shortform?commentId=JxgbgT4fvhnYSGkcc,though the website is no longer up) didn’t take off, despite solving the same problem as Manifold, which did take off; UI is very important and I had built functionality but no UI.

Describe why you think you're qualified to work on this

I have three years of experience as a software engineer at Sainsbury's. Since May 2022 I have been in a frontend role working on the Argos website team; prior to that I was in a variety of different roles in a mixture of frontend and backend. This means that I do have experience in developing the full stack for web applications, and have seen software project management from the position of being a developer.

Other ways I can learn about you

@TetraspaceWest

How much money do you need?

£9,000 - £18,000; this is to replace full-time salary, for 3 to 6 months of work on this which is my rough range of "can jump to prioritising this without being distracted by money" to "most I can confidently say I can probably use, unless I start seeing actual evidence of success in what I have so far"

Links to any supporting documents or information

No response.

Estimate your probability of succeeding if you get the amount of money you asked for

These markets I made for it are unfortunately unconditional:

https://manifold.markets/Tetraspace/if-tetraspace-makes-an-assurance-co- currently 40% that the website gets >100 users
https://manifold.markets/Tetraspace/if-tetraspace-makes-an-assurance-co-0e23dd873406- market on absolute number of users with

40% is roughly my assessment of getting 100 users given that I try this too, on the outside view grounds that previous projects of this nature have failed and that getting arbitrary websites off the ground is hard.

My comments as an ACX evaluator:

I like Tetra a lot, based on their writing and Manifold usage; I strongly considered offering them a Manifund regrantor budget (and would have done this if our overall budget pool was like 20% larger). That said, I'm a bit skeptical that 1) assurance contracts are a huge unmet need, or 2) they'll be able to create a sufficiently-nice-looking platform. I think "platform that looks nice" is actually very tricky but also necessary for wide adoption.

(I'd feel much better about point 2 if they would pair up with someone whose specialty is web design)

Since then it looks like Tetra is working with Jordan of Spartacus, which seems like a great fit (I would have suggested this if Scott didn't)! I'm a little unsure if "jam two people with no prior experience of collaborating" will actually work well, but tbh this kind of describes me and James/Stephen prior to ACX Grants, so there's at least some precedent. Best of luck!