6 ms·

Hm. Seems like it would make more sense for 'buyers' to submit btc along with the problem, which the market could then hold in escrow and release as soon as a p

by Bakkot 13y ago

Hm. Seems like it would make more sense for 'buyers' to submit btc along with the problem, which the market could then hold in escrow and release as soon as a proof passing the verifier was submitted. No need to trust anyone but the escrow service, between btc and machine proofs.

gwern 13y ago
Depends on who you want to put the burden onto, the buyers or the sellers. If a buyer has to pay up front, especially with a marginal service like this, they're exposed to opportunity cost (maybe they were going to do something else with that money, although interest rates don't really exist for Bitcoin yet) and counterparty risk (why do you trust this website to hold onto your bitcoins and not be hacked?). If this market never takes off - as is one's default expectation - it'd not be pleasant for buyers to have paid up front.
- pirapira 13y ago
  I can mix both approaches. The buyer and other people can stash up bounty, which the first prover gets. I have not implemented this lest "bitcoin stolen from Coq proof exchange".
  - tsewlliw 13y ago
    Way to avoid arbitrary risks and complications!
    - pirapira 13y ago
      I added risks and complications. A feature called "bounty" is now available. Anyone can add bounty for a problem. The sum goes to the next solver.
bronty 13y ago
It depends on whether or not the buyer allows for proof irrelevance. There are cases where _human_ buyers care about the structure of the proof. In particular, just like with code, you'd prefer a proof that is maintainable, modular, and not-too-hard to update if your assumptions or claim changes.
bevan 13y ago
I created a marketplace for small coding tasks where buyers pay upfront when they post a code bounty (bountify.co). In practice, the pay-upfront model has worked well- there have been no disputes after several hundred bounties, and about 90% of bounties have received working solutions. However, bountify's domain (coding tasks) is probably better suited to upfront payment than proofmarket's. It's easy for code buyers to estimate the difficulty / value to sellers of small coding tasks, so buyers are reasonably assured of getting a solution within the one-week time limit. It seems like proof buyers would have a harder time estimating the difficulty of creating a proof, and that they might therefore be dissuaded from posting btc up-front for fear of nobody furnishing a solution.
- agravier 13y ago
  About bountify.co, Why does it request profile write permissions when OAuthing wit github? It puts me off.

by Bakkot 13y ago

gwern 13y ago

Depends on who you want to put the burden onto, the buyers or the sellers. If a buyer has to pay up front, especially with a marginal service like this, they're exposed to opportunity cost (maybe they were going to do something else with that money, although interest rates don't really exist for Bitcoin yet) and counterparty risk (why do you trust this website to hold onto your bitcoins and not be hacked?). If this market never takes off - as is one's default expectation - it'd not be pleasant for buyers to have paid up front.

pirapira 13y ago
I can mix both approaches. The buyer and other people can stash up bounty, which the first prover gets. I have not implemented this lest "bitcoin stolen from Coq proof exchange".
- tsewlliw 13y ago
  Way to avoid arbitrary risks and complications!
  - pirapira 13y ago
    I added risks and complications. A feature called "bounty" is now available. Anyone can add bounty for a problem. The sum goes to the next solver.

bronty 13y ago

It depends on whether or not the buyer allows for proof irrelevance. There are cases where _human_ buyers care about the structure of the proof. In particular, just like with code, you'd prefer a proof that is maintainable, modular, and not-too-hard to update if your assumptions or claim changes.

bevan 13y ago

I created a marketplace for small coding tasks where buyers pay upfront when they post a code bounty (bountify.co). In practice, the pay-upfront model has worked well- there have been no disputes after several hundred bounties, and about 90% of bounties have received working solutions. However, bountify's domain (coding tasks) is probably better suited to upfront payment than proofmarket's. It's easy for code buyers to estimate the difficulty / value to sellers of small coding tasks, so buyers are reasonably assured of getting a solution within the one-week time limit. It seems like proof buyers would have a harder time estimating the difficulty of creating a proof, and that they might therefore be dissuaded from posting btc up-front for fear of nobody furnishing a solution.

agravier 13y ago
About bountify.co, Why does it request profile write permissions when OAuthing wit github? It puts me off.