April 27, 2022: Innumerable, and Often Idiotic 🐘🐍
Updates on Stake DAO and Vyper vs. Solidity
Of the two newsletters released so far this week, both topics have have tons of updates and would benefit from revisiting already within just 48 hours. The breakneck speed of crypto is such a blessing.
StakeDAO
Monday’s piece about the sdCRV gauge vote was unabashedly incomplete at the time. Throughout I stated that I felt unwilling to issue an endorsement of any kind because of my shallow understanding. Here’s some updates.
To start with, discussion has exploded since the article, hitting the memetically important count of 69 comments. Read them all for yourselves!
Curve itself weighed in against @crypto_condom’s premise.
The Stake DAO pools also received the highest possible blessing, in the form of an explicit endorsement by Curve chief @newmichwill
Convex chads also joined the chat
One piece of the story I’d also missed is that Alchemix is acquiring $SDT 👀
Additionally, in Monday’s piece I’d cited a number of parts of the Stake DAO tokenomics I didn’t fully grok. The team reached out to fill in my gaps in understanding. For those of you curious about how the math problem I’d posited works out, here’s the solution.
The difference here due to a small fee from their gas-optimized locking contracts. They mentioned they’ll work on preparing some more materials to make it easier for smol-brains like me to understand, and if you have lingering questions please hit me up and I’ll be happy to relay them to the team.
I’d also had some questions about the sdCRV-CRV peg mechanic. They confirmed the primary driver would be incentives to the Curve pool, which explains the importance of this vote for the Stake DAO tokenomics. In the course of our chat, the team also confirmed they’re working on a new mechanism for their sdTokens that would address the concerns of a theoretical governance attack.
Users should still remain aware the project is intentionally fairly centralized at this point. The project is explicitly at the phase where they are spurring faster growth, with plans to decentralize as they grow. This is by no means an indictment — I’d prefer a project of their size focus on rapid development, since decentralization introduces a lot of bureaucracy. I don’t doubt the positive intentions of the team. Once the team passes about $1B TVL it will become more urgent to start the decentralization process, to combat the looming risk of dishonest and hostile regulators.
It’s very encouraging to see the team so responsive to the community questions. A big thanks to the team who patiently walked me through a million questions when I’m sure they had more important work to do like making price go up.
Vyper 0.3.3
Just yesterday we touted the awesome efficiency of Vyper, and the story continued more from there.
For starters, Etherscan already leapt to add Vyper 0.3.3 verification.
You can therefore now peruse the historic first Vyper NFT directly on Etherscan, and if you want to be a part of history you can now mint it directly on Etherscan using the minter by simply plugging the value from user_price_for_quantity() into mint() — about a hundred people or so are whitelisted to get a free mint so this function should return 0 for such lucky duckies.
Where the story continues is that some Solidity fans pushed back against team Vyper gloating about gas efficiency.
Team Solidity complained, justifiably, that comparing a custom Vyper contract to using an ERC-721 based on Solidity’s OpenZeppelin is an unfair comparison, since it was the OpenZeppelin templates that are gas inefficient. It’s a little bit like saying Chevies are more fuel efficient than Fords by comparing a Bolt to an F-150.
They pointed in the direction of a more gas-optimized ERC721A introduced by the Azuki NFT project.
The article is indeed a good read, and a bit off my radar as I’m not much of an NFT person. Comparing this ERC721A implementation against a Vyper NFT was indeed a much closer comparison, with Vyper only getting a 6% edge in gas minting as opposed to the 43% versus the Open Zeppelin contract.
I created a public repository showing my work since there was such intense interest in the methodology for the first go-round.
That said, I don’t particularly intend to keep issuing updates on this subject. I have no doubt that devs on both sides of the debate can gas golf their way to squeeze out incrementally better results, with the lead flip-flopping and the contract’s design becoming increasingly tortured. A truly apples-to-apples comparison would require a neutral party specifying exact functionality, and both sides writing a custom contract to fulfill these specs, and that sounds like too much work.
At this point, I’m content to concede that both Solidity and Vyper can generate minting functions that are optimized to be roughly on the same order of efficiency. The rug comes if an NFT project is using OpenZeppelin’s NFT templates, which requires the end user pay higher gas fees to give the developers more flexibility.
Of course, it’s these costly templates that make Solidity so popular in the first place.
For those interested in getting started with Vyper NFTs nonetheless, check out the inaugural Vyper NFT as well as template on which it was based, from which it’s pretty easy to piece together all the mint, burn, and enumerable functionality one might need.
We also suspect Vyper is only going to get better in its upcoming releases, so now’s a good time to jump on the bandwagon.
The Blind Pig and the Truffles
"I am often wrong. My prejudices are innumerable, and often idiotic."
H.L. Mencken, 1920
Longtime readers know this newsletter relies on reader pushback to get us somewhat in the vicinity of truth. We prefer to crank out fast content and try to be upfront about what we don’t understand and mistakes. This may differ a bit from publications like 🤡desk, which prefer to push their bias with a stodgy, authoritative tone and underemphasized corrections. We hope our way is a lot more fun.
Our early pieces were pretty much nothing but mistakes, but lately we’d been a bit sharper. We had a recent streak of dropping pieces and watching them age perfectly. Or, at least, people just ignored the errors rather than say anything.
The steadily-growing follower count makes life tougher though. Yet ultimately the value in this newsletter is the steady give and take between my hasty screeds and the community response, so keep it up everybody!
Heavy social media users claim Twitter becomes more noise than signal above ~10K followers, a number we’re sadly approaching. So, uh, could some of you take one for the team and unfollow already? I imagine a majority of readers hate-follow anyway, hoping to see me embarrassed yet eternally frustrated by my inhuman lack of shame 😁
Thanks for the write up sir. May I ask what’s the command line tool you use in the screenshot of sdcrv calculation?