Mini Programs in the Web3 World: Understanding MetaMask Snaps

MetaMask Snaps is a project with grand ideals, but setting aside its idealistic facade, we prefer to view it as a mini-program in the Web3 world, or dApplet.

For developers, especially those in China, this concept is not unfamiliar. In recent years, we have often heard terms like "ecosystem building" and "super App" in the internet space, where large companies always try to create a more closed internet environment in order to shift from service providers to standard setters. Nowadays, this trend seems to be spreading into the Web3 domain as well.

As of June 2024, despite Snaps having existed for nearly a year and the concept being proposed at least four years ago, ordinary users still know very little about it. Even some people who are very focused on the cryptocurrency market and open MetaMask every day still do not know what MetaMask Snaps are. However, when explained "it's like mini-programs," they always immediately show an indescribable smile on their faces (most of them are developers, mainly front-end developers).

This situation is reminiscent of when WeChat Mini Programs were first launched. Due to the unclear access to functions, the initial usage was very low. It was only after the access was changed to display the Mini Program list in the dropdown on the homepage that usage began to grow significantly. If MetaMask wants to build an ecosystem, how to attract users to enter the ecosystem may become an urgent issue that needs to be addressed.

Since its inception, mini programs have sparked controversy in the internet industry. Front-end engineers generally hold a critical attitude towards them, believing that they are merely commercial tools for large companies to monopolize traffic and create closed ecosystems, contributing little to technological development. The extent of contributions that Web3's Snaps can bring to the community remains to be seen. Browsing the documentation provided by MetaMask, it increasingly feels that future Snap developers will face numerous restrictions and challenges.

It has been nearly a year since the public beta of Snaps was launched by MetaMask, yet the number of available Snaps listed on its official website is not many; as of June 2024, there are only 68 in total. It is worth noting that the concept of Snaps was proposed by the official around 2020, well before the official public beta, meaning Web3 companies and developers have actually had a considerable amount of time to understand it.

Currently, there are no dedicated developers for Snaps, but there are quite a few mini-program developers from Web2. The general view of mini-programs is that the development experience is poor. Mini-program development still uses the front-end ecosystem, but in a stripped-down version; platform providers restrict what developers can do for various reasons. Moreover, the technical capabilities and documentation quality of each platform vary widely, leading developers to have to go through various strange pitfalls across different platforms before being considered "experienced".

Returning to Web3, for security and other reasons, Snaps is likely to face a similar situation. It must operate in an isolated secure environment, using "secure ECMAScript," which effectively imposes some restrictions on the JavaScript API, such as preventing access to the DOM, Node.js, browser plugin APIs, etc.

Due to security considerations, a basic Snap can actually do very little. Most of its functionalities require obtaining the relevant permissions first, so the Snap needs to request these permissions from the user during installation. The available permissions include lifecycle, transaction, signature, CRON, etc.

Compared to the thriving Web3 ecosystem, the number of available Snaps is relatively small. MetaMask categorizes them into several types.

Snaps Type

According to the classification on the official website, it can be roughly divided into four types:

Account Management

This type of Snap primarily enhances the security of private keys through MPC (Multi-Party Computation) technology.

Traditionally, users typically store their mnemonic phrases on their computers or mobile phones. For users with larger amounts of funds, they may use more secure methods, such as hardware wallets. However, most people simply keep their mnemonic phrases in easily accessible places to prevent forgetting them.

This storage method may cause some issues:

1. If you forget the location of your mnemonic phrase, you may lose access to your wallet assets.
2. If the device storing the mnemonic phrase is attacked by a Trojan virus, it is difficult to prevent hackers from stealing the mnemonic phrase.

MPC technology can automatically split private keys into multiple parts, which are stored in different locations, and only combine them to generate a complete private key when a transaction signature is needed. During this process, a complete private key is never generated, maximizing the protection of the private key from being leaked.

Currently, there are only 3 account management Snaps: Capsule, Silent Shard, and Safeheron.

The appeal of this type of Snaps is not clear yet. According to data from the official website, the number of this type of Snaps is relatively small, and the installation rate is also low.

interoperability

This type of Snaps primarily provides compatibility with non-EVM networks, including multiple chains such as Solana, Cosmos, Near, Sui, etc. Currently, this type is the most prevalent, accounting for more than half of the entire Snaps ecosystem.

Notification and Chat

Although this type of Snaps has its appeal, the number of users willing to send messages via blockchain is still limited.

Safe

The demand for this type of Snaps is quite large, especially considering that MetaMask's own warning features are not adequate when users encounter phishing transactions. Transactions, as the most common operation in the cryptocurrency and blockchain space, are characterized by high risk and high sensitivity. Through the transaction insights feature provided by Snaps, developers can present users with richer transaction information, offering analysis and insights to minimize the potential loss of users' assets.

According to data from the Snaps official website, secure Snaps are the most in-demand type aside from multi-chain support, currently accounting for 20% of the total.

MetaMask Snaps provides the necessary permissions for Snaps to read the raw transaction or signature payload when the user initiates a transaction or signature. The Snap can analyze it and then present a more detailed security explanation to the user.

Future Outlook

MetaMask is attempting to expand its domain to other chain platforms through Snaps, but it is still hard to predict how far it can go. Referring to the experience of mini-programs in the Web2 domain, many apps, although they offer mini-program versions, often have limited functionality, making it difficult to compare with native apps. The follow-up on new features is often lagging, and some mini-programs even guide users to their own apps.

If Snaps develops well, more wallets may follow suit. In this case, Web3 developers may need to develop corresponding versions of "Snaps" for different wallets, which will pose a significant challenge for them. However, if each platform creates its own "Snaps" ecosystem, it seems equivalent to not having a unique ecosystem, which aligns with their concept of "decentralization."

In the future, various "Snaps-like" platforms may emerge, with significant differences between them. Compatibility across platforms may become a routine task for developers, until it reaches an unbearable level. At that time, a Snaps Standard may appear, along with various "SIPs" for discussion, and the community will be thriving.

Despite concerns about the prospects of MetaMask Snaps, it is undeniable that it is a product built by a group of passionate developers aimed at solving real problems, rather than a financial scam.

However, for most of MetaMask's current users, Snaps is not yet a particularly necessary feature. MetaMask may need to do more work on how to promote it more effectively.