Impact Evaluator Research Retreat Demo Day Iceland 2025

I have seen this over and over, I didn't know it was called "local maximas" but it's true, the hesitance to innovate leads to stagnation, and in web3, that's recipe for self destruction.

I love these points. The opportunity to break out of local maxima reminds me of Planck’s principle in the context of paradigm shifts in the hard sciences: “Science progresses one funeral at a time.” In terms of group dynamics, these structural shifts don’t follow a curve one might expect if given purely rational actors. I think these points are crucial for people to watch for in their own contexts as well. Because, we see this conservativism play out in individual psychology & group dynamics, of which scientific revolutions are a member of a set including families, friend groups, communities, political parties, crypto cultures,… Something can seem wrong not because of its content, but merely its distance from the individual’s/system’s homeostasis. And, sometimes it seems more expedient to take a stance of rational ignorance than invest the time in changing one’s mind.

So i got the chance today to dig deeper with juan on breaking out of local maximas An obvious example of it today is streaming, where despite having many subscriptions you still can't see movies you want (or even impact which types of movies get made or if seasons continue) He analogized it to nature & evolution, where you have the core (local maxima) and different probes that are experimental in nature. If one takes off, then we have more of them seeded across the genetic tree So we should keep our local maxima but invest in ideas that can help us break out to the next level Instead what we often see is the opposite, where once a company captures a local maxima they do all they can to protect their moat & prevent a competitor from breaking out to the next level

IERR Demo Day Iceland 2025

almost missed this exciting stuff ahead

Running STEP has taught us that not all tokenized T bills are created equal. Monthly vs block by block compounding? Not the same. Loosely partnered crypto-native wrapper vs TradFi self issuance? Even if tradfi self issued: digital twin or fully onchain fund? Swiss GmbH debt vs US ETF vs bankruptcy remote SPV? different risks and asset flexibility Transparent fees vs redemptions fees vs hidden fees vs time-based waived fees? Important There is no standard for how RWAs are tokenized and no clear right or wrong way (though some are clearly wrong lol) Every detail has second order consequences when building onchain primitives around yield. With STEP, Arbitrum DAO isn't just rewarding assets for launching. We're pressure testing what it actually means for traditional assets to have better implementations onchain.

Some examples we discussed too in the session with IE framework and challenges they have: Eg 1: Conservation Finance Credit Issuance Process How it works - Credits are derived from measurable metrics - eg: a landowner hires a third-party evaluator to assess weather trees are standing - the evaluator translates this data into credits that can be issued in a market as a reward function Challenges: 1. Estimation Issues: - estimation step is often inaccurate 2. Friction at every step - assigning projects within scope involves significant friction - outside digital domain, processes become even slower and more cumbersome Eg 2: Scientific Publishing How it works - Journals act as IEs for knowledge - Each publication venue carries a reputation score (eg: impact factor) Incentive Loop - Goal for scientists: accumulate reputation through publications Problem/Challenge: - The evaluators (journals and conferences) have been entrenched for over 100 years. - Their reward loop has not been updated which creates rigidity, inefficieny, and misaligned incentives.

Hot springs, waterfalls, mountains that fall into the Atlantic, and incredibly bright people gathering to build new things. The intro weekend to the Iceland hosted by @protocollabs and @juanbenet was fantastic!!

Research Retreat Day 1 🌱 Yesterday after months of organizing, the on impact evaluators (IERR’25) in Iceland finally started and we got together 25 participants from 17 countries and six continents (except Antarctica 🐧), some of the smartest people we’ve ever met. IERR’25 started with every researchers’ dream, a gorgeous research hike to Glymur waterfall followed a visit to the 900 year old Hvammsvik hot springs led by @juanbenet and @momack28 Ideas were discussed and people were nerdsniped on measuring impact, control theory and cooperative game theory. I couldn’t wish for a better start 🌱 Stay tuned for research updates!