🍞 Upgrades Baked In β€” Issue No. 52

πŸ“° News

This week, the Tezos blockchain network successfully deployed its first backwards incompatible upgrade via an on-chain vote. Tezos has a novel mechanism that allows upgrades to happen natively via a proposal and voting system that occurs on the network itself. This week's upgrade marks the first time that mechanism has been exercised. Tezos did a high profile ICO during the peak of the last bubble, but the launch of the mainnet was long delayed by (ironically) community infighting. This week's successful on-chain upgrade is a big milestone for the project as it seeks to leave those struggles behind. Link.

Blockchain network upgrades usually take place when users voluntarily switch to new node software that includes incompatible changes. Such an organized upgrade requires massive collaboration between various parties in the ecosystem. The Tezos creators' aim was to enable their network to be upgraded by community consensus, but without the arduous process of coordinating the software upgrade itself. On Tezos, any party can submit source code files representing a change to consensus. Every holder who stakes their coins -- referred to as "bakers" on Tezos -- has the right to cast a stake-weighted vote on the various proposals. After a voting period, if the vote reached  quorum and was approved, all participants node software automatically activates the sandboxed source code to run. For a more in-depth, but still approachable, introduction to the Tezos amendment process, check out this excellent write up by Jacob Arluck. Link.

The two changes that activated in this week's upgrade were themselves relatively minor. First, the patch raised the network "gas limit", i.e. the upper bound on the amount of computation any given transaction can utilize when interacting with a smart contract. Second, the upgrade modified the so-called "roll size", the minimum amount of staked Tezos required become a block producing baker. It was reduced by 20%. If you want to read more, I'll again direct you to a post by Jacob Arluck, who provided a top-notch summary of the proposals in the lead up to their activation. Link.

Let me lay my cards on the table: I'm actually quite skeptical about Tezos. I'll explain why.

For one, I remain skeptical of any Proof-of-Stake based consensus algorithm. I would love for them to work. I'm hoping in the long run they do. But the truth is, compared to Proof-of-Work, they remain relatively unproven and experimental. Can PoS networks stay decentralized and censorship resistant in highly adversarial conditions? We're still figuring that out, and the on-chain upgrade mechanism just adds another potential attack vector to the mix.

Beyond that, I'm not sure I actually see the problem that Tezos is trying to solve. What has prevented networks like Bitcoin from being upgraded isn't getting people to upgrade nodes-- it's getting the community to agree on the upgrades in the first place. By definition, that process is political, and it has to happen off chain. I'm not sure how Tezos changes that. Indeed, the authors of this simple upgrade spent enormous time and resources championing it to the Tezos community.

As a thought experiment, imagine a universe where Bitcoin had stake-weighted on chain voting like Tezos. Would the block size debate have been any less contentious? Would the Bitcoin Cash hardfork have been avoided?  It's not clear at all the answer is yes. And if it is yes, why? Because the people who had the most coins would have voted for an increase? If that's the case, haven't we created a plutocracy? Is that what we want cryptonetworks to be?

Finally, it's unclear that base-layer blockchain networks even need to be continually upgraded. The Bitcoin community has decided they don't. Even the Ethereum community, which is pursuing a dramatic upgrade to that network over the next few years, seems to believe that at some point the base layer will be pretty much finished. If either community turns out to be right, Tezos' raison d'Γͺtre becomes unclear.

With all of that said, I still think this upgrade to Tezos is a milestone worth paying attention to. The developers behind Tezos are talented, intelligent folks. The growing community around the project now seems vibrant. They've staked out a unique set of tradeoffs for a cryptonetwork, and whether it succeeds or fails, it is a valuable experiment to have run. Advertising the network as easily upgradeable may have caused a self-selection bias for folks who are excited about pushing cryptonetworks forward. Even if the on chain governance mechanism turns out to be only modestly valuable, having attracted that kind of community might turn out to be a big win.

On the list of projects I might be very wrong about, Tezos is probably at the top. So I'll be keeping an eye on the project moving forward. I'd suggest you do the same!

πŸ“Š Statistics

462. The number of block producing "bakers" on the Tezos network at the time of this writing. While this seems low for a network aiming to be decentralized, this is is actually relatively high compared to other Proof-of-Stake networks. Cosmos, for example, currently has 140 active validators. Link.