This is another spur of the moment post. Only putting on Mirror so it lives on a bit better than a Twitter thread. If it is dumb please pretend someone else wrote it.

I recently wrote about the case for fees that make a protocol stronger.

Lately I’ve been thinking about something else: extractive fees should be accrued to parties at the periphery and not to the protocol/DAO.

“Extractive fee” is kind of a cynical framing. I am referring to fees that are being taken just because they can. The protocol could work without these fees, perhaps even better, but some party wants a reward for their work.

One example would be the fees that some borrowing/lending protocols take, between what the borrower pays and what the lender receives. Another example would be trading fees, such as Uniswap could charge.

Uniswap is a very interesting example, I think. Much has been made of the “flipping the fee switch.” The DAO could set a fee, a percent of the swap fee for a given pool, that goes to the DAO.

But there is another fee option in Uniswap v3. At the periphery, a swap order can send a fee to an arbitrary recipient. Presumably, this exists to allow clients to give themselves a fee: an incentive to build a Uniswap interface. The problem is that Uniswap Labs makes the best interface and they don’t take a fee: hard to compete with 0!

I wish Uniswap Labs did take a fee. Clearly, they are a competent team. They deserve funding that is unencumbered by DAO politics. Also, if they took a fee, it would make more sense for others to build their own interfaces and match or undercut this fee: we might actually see many Uniswap interfaces!

In many ways, I am more interested to see this fee switch get flipped than the protocol fee switch. It is not clear to me that the Uniswap DAO has proven itself to be an effective steward of funds. I would like to see it have funds other than UNI to spend, but I am not sure that giving it more funding is especially warranted right now.

So much is still being figured out with DAOs: their politics and how to best run them. Sending money to DAO is, in many cases today, guaranteeing it will be subject to some bureaucratic mess. I think that enabling and encouraging those adding value to an ecosystem to directly take a cut for themselves would be a great thing: no need for a grant or some DAO benevolence! If done well, this could create a periphery of many organizations profiting from the protocol and highly interested in seeing it succeed. This is in many ways how Ethereum works and how I think the web standards have worked, too.

Clearly DAOs are exciting and probably needed for many things, but I think that many parties taking fees at the periphery, and the core team working to make this highly viable, is under discussed. It strikes me that the dynamic of many productive and profitable teams wanting to see the core keep working might be more generative than many people trying to vote their way to a share of one giant pot.