Aptos and Sui — Keeping up with the Move
Blink and you’ll miss it.
Innovation in the web3 space is ridiculously overwhelming. Even during 2022's bear market, Layer 1 solutions are still popping up.
So here’s a quick summary of the what Aptos and Sui are looking to accomplish and what’s this relatively new Move programming language.
Aptos
Heir of Meta’s dead project Libra, rebranded Diem, Aptos is the brain baby of several ex Meta employees.
As stated in their whitepaper, Aptos is aiming at providing a better web3 foundational layer, capable of mass-adoption throughput and better overall safety.
The so-called Solana killer’s main innovation resides in its use of the Move language — to which we’ll come back later — and its execution layer: Block-STM. It uses parallel execution to manage an extremely high amount of transactions simultaneously. Most importantly, it allows flexible parallel execution. Meaning that channels for transactions aren’t siloed. Apparently, this means that the worse case scenario for Aptos isn’t as bad as Solana’s — which completely shuts down from time to time. Since transactions can change their path to validation, in the event of an issue on the channel.
But launching during the bear-market isn’t the only difficulty Aptos has been facing. Some critics have shed light at the disappointing tokenomics of the project. Underlying the fact that 80% of the tokens are already staked, when very few tokens have been available. Meaning that VCs will have loads of tokens to dump on retail when they get released.
They stated that their whitepaper will be updated regularly, and will eventually contain more detailed tokenomics.
Most importantly, it seems that very few projects have been launched or tested on Aptos. For now, only an Explorer, a wallet and a Name Service have been developed. But they’ve launched their main net only 10 days ago.
Moreover, hunting for market shares is already difficult without bringing new barriers to entry like a new programming language. We’ll have to give time and space to Aptos to develop.
SUI
On the other side, Aptos competitor SUI hasn’t made its main net launch yet.
SUI is the second project to use the new Move language as its base language. Like Aptos, it’s also strongly committed to and focused on extremely high throughput. Its main innovation to achieve this — beside Move — is the “casual ordering” approach.
Casual ordering means that not every transaction has to be submitted to consensus. It might seem counter intuitive in an industry of decentralised protocols revolving around the one main concept, consensus. But if you think about it, if a transaction T is an automated transaction coming out of a trusted smart contract S, and if every transaction leading to T has been checked, then we might not need a consensus on T.
Another major change, compared to other blockchains, is that Sui validates transactions individually instead of batching them into a block. Once a sender makes a transaction, a request to validate it is sent to the validators. Validators can then vote on the legitimacy of said transaction until the sender attains a Byzantin majority (1/3). After-which, the sender receives a certificate of finality, proving that the transaction will be accepted and is considered as done.
These fundamental differences might have motivated them to describe themselves as a “smart contract platform” rather than a Blockchain, in the subtitle of their docs page. Even though they still present themselves as a Layer 1 Blockchain on their landing page.
Concerning the team, SUI was co-founded by Evan Cheng, ex Director of Engineering at Meta, where he most probably learned everything he needed to know about Diem and its underlying Move language.
Move
It seems that beyond Aptos and Sui, the most important news in all that is this new blockchain specific programming language called Move.
Initially, it’s been developed by Meta to build Libra. Since then it has been shared and is now a state of the art, chain agnostic, blockchain programming language.
Mainly, it has been developed to palliate issues faced by Solidity and other languages. What these other languages did was reuse concepts from more traditional programming languages objects, like arrays, strings and hashmaps. But all the hacks and exploits have brought developers to believe that a new tool was needed.
In essence, blockchain programming languages need to do a few simple things, namely manage digital assets movement, or transactions.
That’s why Move’s objects are not arrays and string but the digital assets themselves. The users can then define different types of objects, resulting in different constraints that can protect the user. For example, imagine having defined some of your digital assets as “economy” and defining that these objects cannot be moved, no exploit will be able to syphon those assets from your address.
The Move language seems to change the paradigm and offer a new way to safety in web3. Putting the focus on the ressources you’re protecting rather than the movement — like other blockchains — seems promising.
Concluding
Innovation is merciless in the field, but I wouldn’t give up on OG Layer 1 just yet.
The network effect has devastating strength and shouldn’t be ignored. Every time new technology or logic is presented to the market, technical teams of old project study them and, if relevant, how to implement them.
And since updating — when it’s possible — is easier than creating an ecosystem, first movers might be able to upgrade to new standards quickly and keep their users and devs.
Nonetheless, Aptos surely launched a trend, and we’ll be seeing Move-based projects popping left and right.
Will their implementation of Move help Aptos and Sui earn their part of the cake ? Or will we see Move-based smart contract be integrated in older blockchains, leaving them in the dirt ?
As always, time will tell.
Like everything on the internet, this shouldn’t be considered as financial advice. It’s a good starting point for your DYOR though ;)
