DeFi Vibrancy amid Exploding Gas Price — What’s a Dapp Developer to Do?

A Call to Coordinate

DeFi is Winning Hard

Something very exciting is happening with DeFi; in a way it feels similar to the run-up of the 2017 ICO craze. But there is a difference: where the ICO phenomenon was built up of many separate projects — most of which failed, some of which succeeded — DeFi is a conglomerate, an interlocking mass of projects, with strong dynamics of organic growth, evolution, and antifragility.

Of course it’s still made up of separate projects, but these projects seem to deliver even more than the “financial legos” idea thrown around a year or two ago: there is actually a kind of magnetism to these lego pieces, so that when one DeFi system is built, it’s actually pulled toward the others.

As an example, when we started our FRY sale, an independent agent provided FRY liquidity on Uniswap, increasing our exposure and FRY liquidity, all without any permission granted or effort on our side. We built the system and it snapped into DeFi, automatically.

The incentives of arbitrage acted as a kind of fundamental force of DeFi, pulling our system into the larger conglomerate. If the individual dapps and interface like Uniswap and 1inch are the organelles of DeFi, DeFi agents following financial incentives act as the chemical signals between them, regulating the larger system in an emergent fashion and rendering the body as a whole aggressively antifragile, even cannibalizing rotten pieces when they’re discovered.

We are entering a world where a clever user can make $40k profit on a single transaction, and happily pay $2k in fees.

Which brings us to the problem, inseparable from the DeFi excitement.

Gas prices are insane

Gas prices peaked on Thursday at 360 Gwei. It cost $5 to approve any amount of DAI, and an additional $6 to transfer it. Closer to home, our FRY sale now cost around $40 in gas to successfully buy any amount of FRY, and DAIHard, our product for fiat/crypto trades on Ethereum, cost over $200 in gas just to open, let alone push through to completion.

Dapps will have to begin migrating. Where will they go? Anyone who thinks the answer is obviously Ethereum 2.0 is not being realistic about the time scales involved. Maybe the answer is Ethereum 2.0, but we simply can’t know that until it’s built.

A dapp development team with insight into this question will be better prepared for the coming months, because they can start preparing for a migration and be relatively sure that all of their DeFi friends will move with them.

So, let’s do some collaborative sensemaking.

A Call to Think Together

I’ll be hosting a series of open calls over the coming week, to explore this question with dapp developers and other web3 project professionals. We will focus on the options available for dapp developers at a technical level, considering network effects.

The central question will be: what can a dapp developer do today to scale — ideally without losing interoperability with the DeFi system?

Do we wait for Ethereum 2.0? Can we expect dapps to react by becoming more efficient? What are the interopability options like today, between different chains? What are the most promising smart contract languages? What types of applications can we expect to move first, and which ones will remain operational on Ethereum even with high gas prices?

Participants can expect two benefits: first, they’ll get a live view of where “the others” are looking to jump, and how urgently, better preparing them to follow the herd (or some relevant section of it). Second, the collective intelligence of any given chat will highlight differences, challenges, and similarities of the options on the table, generating new insights any one participant couldn’t have reached alone.

Most or all of these chats will be recorded. If there are more than 8 or 10 participants in any given chat, we’ll utilize breakout rooms to promote coherence, then filter the insights back up into the larger plenary.

If you’d like to be involved, fill out this Doodle to coordinate times. I’ll announce the schedule Monday morning UTC as another Medium post, and email the respondents to the Doodle.