1/ All you folks out there talking about how valuable "governance" is have lost the plot.

TL;DR:
1. Generalized governance over DeFi building blocks is BAD.

2. Trust-minimization is GOOD

3. DeFiers should strive for trust-minimization, not community governance
2/ Bitcoin benefits enormously from *trust-minimization*, and Szabo has argued effectively that trust-minimization is the most scalable form of governance.

Why?

Because ecosystem participants don't want rules to change on them.
3/ Changing the rules of a cryptographic primitive, against the wishes of even a relatively small segment of the actors of that system, will erode trust in that system.

Erosion of trust leads to defection, reducing the value and scalability of your system.
4/ IMHO, BTC is currently more "money-like" than ETH because Bitcoin's governance is more trust-minimized than Ethereum's.

(IMO this is due to the Ethereum's upcoming transition to ETH2.0, and Ethereum can become as trust-minimized as BTC after that transition.)
5/ Similarly, ETH is more money-like than any of the ETH-killers because all of the purported killers have extensive (aka non-scalable) governance systems

What they tout as a benefit really just hamstrings them from being able to achieve scalable moneyness.
6/ For a recent DeFi example, Sushi has governance, while Uniswap does not.

But this governance has only been detrimental to LPs, who faced the acceleration of the migration to Sushi due to arbitrary actions from an untrustworthy governor https://twitter.com/NomiChef/status/1302508803309498369?s=20
7/ While Uniswap LPs can "set it and forget it" without fear of arbitrary changes to the Uniswap protocol, Sushi LPs can't do this.

If they stop paying attention even for a few days, they might be force-migrated to a new system far earlier than expected: https://twitter.com/NomiChef/status/1302139173864792066?s=20
8/ So at the very least, it's clear that governance is COSTLY. It costs time, effort, and attention of the participants.

And maybe sometimes that's important. But it certainly is not in keeping with the ethos of trust-minimization, and reduces the scalability of money legos
9/ "But if we don't build governance into the system, how do we ever make progress", I hear you asking.

Well, that's the beauty of new systems. Uniswap already did this once, deploying V2 after a successful V1.

Importantly, migration was voluntary, and V1 is still used!
10/ Similarly, I'm sure ETH1.0 will remain after the migration to ETH2.0.

IMO, ETH1.0 will likely fade, just as Uniswap V1 faded. Because most people in the ecosystem will migrate to ETH2.0.

This is GOOD! This is trust-minimized governance in action!
11/ This is how you build trust-minimized DeFi, which is what we all want.

If we don't restrict what governance systems can do, they will trend towards CeFi in the long run, just as all systems that don't have strict constraints on governance trend towards centralization.
12/ Generalizing, Governance of money legos should ADD NEW things. It should very rarely, if ever, change the way the current thing works.

Addition > Migration
Optionality > Force
Trust Minimization > Majority Rule

/fin
You can follow @brendan_dharma.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled:

By continuing to use the site, you are consenting to the use of cookies as explained in our Cookie Policy to improve your experience.