Approved

JBP-450: Fund Juicebox Protocol Emergency Developers

Cycle

62

Requesting

180,000 USD + 15,000,000 TOKEN

loading

Proposed Transactions

Payout action loading...
Transfer15,000,000TOKENto0xd0a1...2A49

Synopsis

Pay Dr Gorilla, Ba5ed and Filip to look after the protocol, be on-call for bugs and related emergency communications, maintain communication of technical nuances with precision in documentation as doubts arise, and generally attend towards technical risk mitigation. If all goes well, we won't be needed much during these next 20 cycles, at which point the DAO will have even more confidence that the protocol is self sufficient.

We'll support our work by building a few simple security workflows:

  1. Alerting for rapid balance changes in any terminals.
  2. Periodic fork/simulation testing of key functions (maybe pay, redeem, create project) for projects at random.
  3. Alerting for rapid movements in price feeds.
  4. Maybe we can watch new blocks for "suspicious" interactions with JB contracts (repeated internal txs, unusual gas usage, lots of reverts).

The folks being paid through this proposal will be spending most of their time on Applications and R&D efforts funded outside of this payout, like Bananapus.

Motivation

The Juicebox Protocol works. Necessary or suggested changes to it have become scarce, but the risk of unknowns and memories of sleepless nights remind us that insurance is still valuable to recognize as the DAO shift most of its resource allocation to growth.

In my view there are three paths to growth:

  • Don't spend. Let all value accrue to $JBX as it grows on its own.
  • Spend on iterative strategized growth efforts.
  • R&D and artful experimentation that open up new possibilities.

These are all more important to allocate to right now than protocol development, though making sure we have staff in place to keep things from going to 0 in extreme circumstances should remain a priority.

Specification

  • $9k / cycle for 20 cycles to Juicebox Protocol Emergency Developers
  • 15m $JBX upfront to Juicebox Protocol Emergency Developers (0xd0a12b8279614A46b75Dc23DBB3245094a3f2A49)

Rationale

Much smaller comp (3k per dev / FC) compared to active contract dev expenditures throughout the DAO's lifetime, but longer cycle duration that acknowledges the perpetual nature of this work. After 20 cycles, I'll feel confident that we can reassess.

Risks

3 people will formally monitor the protocol and be on-call to address mishaps. This gives us redundancy and presence around various time zones, but there's risk of uncaught emergencies that aren't mitigated in time before worsening.

Timeline

Next FC.

Votes

loading