An Irreverent Web3 Mobile Review Article

Author: Jason, Puzzle Ventures

A web3 researcher who doesn't evaluate phones is not a good researcher.

As a web3 researcher who hasn't fully grasped on-chain data, I never expected to evaluate phones one day. The concept of web3 phones is akin to the concept of handheld computers from years ago; while the combination of phone and web3 seems fitting, it also appears somewhat strange. On a larger scale, it could be seen as a revolutionary product marking the shift from the internet to mobile internet, while on a smaller scale, it is just an ordinary phone with some added web3 software and hardware. In today's world, where web3 has not yet been fully embraced, web3 phones add more possibilities to web3, but of course, they also bring uncertainties.

Regardless, any new thing is born in scorn, grows in doubt, and eventually becomes indispensable. So, let's rigorously and seriously study the overall picture of web3 phones; what if it becomes a historical turning point for mass adoption? Alright, as usual, we will start our research from the earliest web3 phones. Wait, the earliest web3 phone is actually the Metavertu from the half-dead luxury brand Vertu? Forget it, web3 is a bizarre world; we just have to get used to it. However, to enhance the rigor of our research, it is necessary to clarify the panorama of the web3 phone market first, and then… we'll talk.

Panorama of the Web3 Phone Market

The web3 phone market is a typical emerging market characterized by non-standardization, chaotic pricing, non-mainstream manufacturers, and uneven user experiences. However, by horizontally comparing various phones, we can still observe some clues.

Figure 1

As shown in Figure 1, there are currently five main web3 phones, and only the first three can truly meet the basic requirements of being web3 native. Therefore, it is difficult to summarize a regular development trend at this stage. Regardless, we should still analyze the underlying development logic from some less obvious trends:

1. Manufacturers generally lean towards non-mainstream brands. This is not hard to understand; it is simply because web3 phones are products with too much uncertainty. Major phone manufacturers have standardized R&D plans and production requirements, making it unlikely for them to quickly enter the market due to the new concept of web3 + phone. At the same time, major manufacturers are still competing in the software ecosystem, and from the perspective of web3 phone software, they would need to cut off their own arms to adopt web3 OS, dapp Store, and DID, which would disrupt their own ecosystems. From an investment return perspective, this is completely illogical. On the other hand, web3 phones themselves are bizarre combinations, and when combined with small manufacturers that are already pursuing differentiation, it does not rule out the possibility of web3 phones forming a bottom-up growth trend in the future.

2. The basic configuration of web3 phones is at an above-average level. This is also easy to understand; the main concept of web3 phones is web3 integration, not competition in specifications. Even if some phone functions are added to cold wallets, I believe it would still be an eye-catching product. Of course, the two leading phones, METAVERTU and Solana Saga, use top-tier processor configurations among current Android phones, along with relatively strong camera and storage capabilities, reaching the level of first-tier phones. The remaining few have uneven configurations, resembling a haphazard assembly of a basic, moderately satisfactory configuration. However, to be honest, any user pursuing specifications is unlikely to buy any web3 phone, and web3 degens should at least have a commonly used main phone. Therefore, the configuration of web3 phones is basically "tasteless to eat, a pity to discard."

3. The software layer of web3 phones mainly consists of dapp store, wallets, and DID. This involves a very important question: what is the technological development path of web3 phones? Is it a fusion model of OS + dapp store on the software layer, or a major overhaul model of cold wallets + Blockchain SIM cards on the hardware layer? We will discuss this question later. Purely from the software perspective, the current popular web3 integration method is to embed a dapp store or curated dapps to enhance dapp accessibility, then use wallets and DiD to build a relatively private and secure operating environment, and finally add some identity verification functions utilizing the phone's camera, fingerprint recognition, and facial recognition, along with some benefits like NFT airdrops. Additionally, Solana Saga has developed a Solana Mobile Stack to facilitate the development of web3 dapps, providing some toolkits for dapp development in the Android environment. The true effectiveness of this web3 combination punch needs to be judged in conjunction with the actual user experience, and we must consider an ultimate question: Is it better for any phone to install a universal web3 OS or to adopt a pre-set single-ecosystem web3 system?

4. The hardware layers of web3 phones have their own merits. METAVERTU uses a dual-chip model to enhance the security of the execution environment, adding a Snapdragon A5-level security chip. In simple terms, the working principle is to establish a set of isolated environments based on TEE (Trusted Execution Environment) + SE (Secure Element), while also allowing interaction with the original system via API, similar to bank-grade security hardware like USB keys. This dual-chip security solution clearly involves significant investment, but the actual operational effectiveness remains to be verified. Looking at Solana Seed Vault, its essence is a software that manages encrypted information, storing it within a secure environment in local hardware, which is a compromise between software and hardware. The early Finney Phone simply added a complete hardware wallet and used a physically encrypted safe screen (similar to a sliding lock) for extra protection, but it currently seems to have received little attention. Overall, web3 phones have room for hardware integration, including the recent BSIM card released by China Telecom (which stores private keys within the SIM card) as an innovative hardware security solution. As a physical hardware link to the digital world, phones seem to reflect value more than software.

5. The pricing of web3 phones currently lacks standards. With an exaggerated price range from $3350 (Metavertu's lowest configuration) to $376, the current market positioning of web3 phones is still in a state of confusion. From the horizontal comparison of price gradients, there are at least three types of positioning for web3 phones: high-end cool phones, backup phones for web3 enthusiasts, and low-end non-mainstream phones. Analyzing from a relatively rigorous perspective, we need to determine whether the main application scenario for web3 phones in the future is as a primary device or a backup device. As a primary device, Solana Saga is the closest, but from several public practical evaluations, Solana Saga does not meet the standards of a first-tier primary device in terms of UI smoothness or configuration richness. As a backup device, Solana Saga and several other low-priced web3 phones seem to meet the needs, but is the original intention of web3 phones to become a backup device? At the same time, the single Solana and Polygon ecosystems as backup devices cannot fully meet the basic needs of web3 enthusiasts. Therefore, the positioning of web3 phones in the future has only two paths: gradually evolving to meet the basic universal needs of web3 and approaching the universal needs of primary devices, or gradually becoming a niche, grassroots, or even obsolete product.

Through the above analysis, we find that the current web3 phones need to solve several problems: first, a reasonable combination of software and hardware and optimized adaptation solutions; second, handling the competitive relationship with major phone manufacturers; and third, clarifying the positioning between primary and backup devices.

Technical Path of Web3 Phones

The technical path of web3 phones involves three areas: hardware, OS software, and Dapp software. Web3 hardware wallets and phones can combine into a "hardware + hardware" solution, while Web3 OS and Dapps can interact with the blockchain through light nodes or third-party nodes (including wallet encryption) to achieve the purpose of running web3 on phones.

Hardware Path of Web3 Phones

From the current attempts at web3 phone hardware, they can be roughly divided into the first-generation solution of hardware wallets + phones that emerged around 2018 and the second-generation solution of SIM cards + phones and chips + phones that appeared around 2022 (in fact, hardware wallets are essentially chips), as shown in Figure 2.

Figure 2

First, let's look at the first-generation solution. HTC Exodus 1 and Finney Phone launched web3 phones with built-in hardware wallets and some simple operational software at the end of 2018. If I remember correctly, the end of 2018 was around the time of the last bear market bottom, and from this release time, we can roughly sense the lack of professionalism in both companies' understanding of web3, as their phone sales hardly made any waves.

From HTC Exodus 1's promotional materials at the time, the phone was pre-installed with a set of management hardware wallet (chip) Zion App to achieve similar private key storage, ERC-20 and ERC-721 token payment and remittance functions, while also pre-installing basic software like the Brave browser and market software, thus completing the so-called web3 phone solution. On the other hand, Finney Phone was relatively more innovative, adding a Safe Screen component, as shown in Figure 3. This means that when you need to use the hardware wallet, you need to physically toggle the safe screen, thus disconnecting the phone from the external network and allowing wallet operations to occur in a secure environment.

Figure 3

This retro design reminds me of BlackBerry's physical keyboard; while it effectively and simply solves some problems, it still cannot withstand the trend of technological advancement. Web3 phones are the same; the first-generation solution, while providing a basic solution, faces serious challenges from the trend due to its retro design, incompatibility with existing systems, and UI smoothness issues.

Now let's look at the second-generation solution. METAVERTU uses a TEE + SE secure environment based on another independent chip and switching OS, while the BSIM + phone solution, primarily based on the blockchain SIM card (BSIM) launched in cooperation with China Telecom and Conflux, is also a unique hardware technology path.

The TEE + SE independent chip trusted execution environment technology used by METAVERTU is, to be honest, relatively reliable. TEE technology has been forming standards and specifications since the 1990s; it is an independent hardware execution environment that can only be accessed through private keys, and it records the hash state before and after execution, ensuring that there are no issues during execution or providing a basis for tracing if problems arise. Additionally, the data within the environment is encrypted by SE, so even if the operating system is exploited, it can still provide a final layer of protection for the data within the environment. Therefore, as long as the information entering the executable environment is correct, the execution process is relatively secure, similar to a hardware wallet. However, this technology still has two issues: first, the cost is too high; the computing power and storage space required for the execution environment determine the chip level requirements, and a chip like METAVERTU's Snapdragon can raise the price above $3000 (although there are also brand premium reasons), which is extremely unfavorable for mass adoption. Second, this TEE method is more suitable for security protection within relatively closed systems like banks; if it needs to connect to the entire web3, especially in scenarios involving MPC and Oracles that require extensive data interaction with the outside, more complete privacy computing and customized smart contract infrastructure are needed to ensure efficiency, privacy, security, and scalability in the data processes.

The basic principle of the BSIM card is to enhance the storage space and computing power of the SIM card by dozens of times, enabling the generation and storage of users' public and private keys within the card. According to official explanations, "The BSIM card can reduce the risk of malware attacks by adopting hardware security protection technology, and it can also bind digital identities to real identities through the real-name system of mobile phone numbers. In addition to basic private key management and digital signatures, the BSIM card can also achieve encrypted storage, key recovery, and even integrate traditional USB key functions." Aren't these functions essentially the basic functions solved by hardware wallets? In other words, the BSIM card essentially provides the capabilities of wallets and DID for web3 to operate on phones. Samsung once collaborated with Ledger to develop the Ledger Live APP to connect Ledger to phones via Bluetooth, thus achieving the capabilities of web2 + web3. When the future BSIM card becomes a more widespread paradigm, a single phone achieving the capabilities of web2 + web3 will clearly be a more attractive and imaginative solution.

It is worth mentioning that when I searched for materials related to BSIM, at least 50% of the information discussed the development potential of BSIM in Africa, which is both unexpected and reasonable. First, the BSIM card effectively addresses the chaos in Africa's banking and fiat currency systems. Compared to Alipay and PayPal thriving in a well-developed financial system, Africa's inherently imperfect financial system has created an excellent soil for web3 payments. According to World Bank data, as of the end of 2021, only 53% of individuals aged 15 and older in sub-Saharan Africa had bank accounts or mobile payment software based on bank accounts. Therefore, when the BSIM card provides web3 infrastructure + cryptocurrency as a payment equivalent + personal on-chain KYC + mobile payment methods based on phones, a complete chain is formed, and we can imagine the enormous potential for mass adoption of web3 among young people in Africa.

OS & Dapp Path of Web3 Phones

The OS or dapp path of web3 phones is actually quite easy to understand; web3 OS is about establishing a separate decentralized (phone) operating system with light nodes to achieve on-chain interaction throughout the system, while phone Dapps interact on-chain through the interaction facilities provided by the dapps within a centralized phone OS. This topic is very broad, ranging from web3 browsers to in-app stores within web3 wallets, to complete web3 mobile OS, all of which can be understood as a solution. Therefore, I will simplify it here and only discuss the pure web3 phone OS.

Currently, relatively pure web3 phone OS includes different standalone products like ethOS, Conflux OS, dappOS, etc., while Near, Polkadot, and others are also developing operating systems based on their respective chain ecosystems. These OS systems are attempting to establish a unified interface for the application layer, improve interaction efficiency through light nodes, and provide more complete developer tools, thus laying the groundwork for potential mobile web3 in the future. But is the reality as beautiful as we think?

First, let's discuss the necessity of web3 phone OS. Through comparative analysis of different existing OS, I believe the greatest value of the current web3 OS is: providing a unified identity, wallet entry, and unified scheduling of phone hardware to enhance user experience. As shown in Figure 4, the innermost layer of METAVERTU web3 OS's demonstration diagram is the unified authentication and protection of personal identity, which evolves into different functions in the outer circle. Without using web3 OS, if you want to operate more than two dapps on a phone, you need to repeatedly switch between different dapps and send data across different RPCs, and the experience is certainly not as smooth as operations within web3 OS. However, the web3 OS itself also has certain issues: an independent OS requires a strong technical team for continuous maintenance, and the efficiency of technology and version updates is questionable, potentially leading to security risks; alternatives like web3 browsers and wallets can achieve similar functionalities, and the balance between intermediate experience and security cannot yet be fully validated.

Figure 4

On another note, we must discuss security. Using mobile dapps in an open-source environment like Android OS is certainly the least secure, so is web3 OS necessarily more secure? The reason hardware wallets are secure is that they are completely offline and operate through fully closed-loop built-in software, making them immune to hackers and network attacks. While some web3 OS can achieve soft-hard integration for physical key storage, the web3 OS system itself still has the potential to be attacked; however, I believe the developers of web3 OS will implement targeted security measures. From the perspective of user experience, do users really care whether security is rated 80 or 90? Is the 90% security provided by web3 OS the only factor driving users to abandon their existing 80% security environment? Not really; what users truly care about is usability, fun, and whether they can make money. Therefore, the real homework for web3 OS still lies in the richness of the dapp store, the ease of multi-chain interaction, and the optimization of costs like gas fees. As for security, including data privacy, it can only be seen as an added bonus.

Figure 5

JAMBO understands this well. JAMBO is a web3 superapp aimed at the African market, and its recently launched JamboColor is also a web3 phone (Figure 5). From the current promotional materials, we learn several pieces of information: it sells for as low as $99, comes with wired earphones, has a stylish appearance, and the built-in JAMBO superapp allows users to watch to earn, play to earn, learn to earn, and chat to earn, while security and OS are not mentioned at all. I believe this will be a very effective attempt at mass adoption, because it truly considers what users want, rather than how to hype it up. The iteration of web3 technology is indeed important, but that is a top-down matter for core members of the Ethereum community and A16z to consider. Ordinary users only need a good user experience to appreciate the charm of web3.

User Experience & Community Response

A crucial part of phone evaluation articles is user experience, and the quality of web3 products also partly depends on community response. Therefore, we need a section that combines user experience and community feedback to explore the "naked" state of these web3 phones after shedding their glamorous brochures.

First, let's take a look at METAVERTU. Due to the dual-chip setup, METAVERTU can achieve completely isolated dual-system operations, meaning users can enter the new system homepage through the Meta Space app and perform on-chain interactions including cloud storage (IPFS), Wallet, browser, etc., while also downloading new dapps from the built-in dapp store. Overall, this experience is smooth, but the current dapp store ecosystem is still not well-developed, so many dapp download sources actually just redirect to external links. Additionally, since Vertu's own Life Account requires certain personal information binding to use the system, the so-called privacy and external data cannot be protected.

In terms of community response, I mainly refer to reviews on Amazon and Taobao (Vertu has been acquired by a Hong Kong company and focuses on the Chinese market) and comments on Twitter and Reddit. First, there is a premise: users purchasing such high-priced phones are less inclined to complain about real pain points online, so I can only form a rough perception. From the reviews on e-commerce websites, most complaints are about the Vertu system itself, such as battery life, the smoothness of Vertu OS, heating issues, etc., while the only few feedbacks regarding web3 usage are positive. Overall, due to the novelty of web3 phones, users have a perceptual bonus for new experiences and are unlikely to complain too much. As for the so-called security and privacy, perhaps no one really cares until problems arise. Today's phone market is no longer simply about hardware stacking; the competitive power generated by software capabilities is increasing, and luxury web3 freak phones like METAVERTU face numerous software issues but gain reasonable, even additional explanations from the dual-chip hardware stacking approach. Regardless, borrowing a saying from a Reddit user, those who will truly buy METAVERTU are either ultra-rich or ultra-stupid; the future web3 phone should not use METAVERTU as a milestone for development.

Next is Solana Saga. There have been quite a few English and Chinese review articles on Solana Saga, and I will simply summarize and discuss trend viewpoints here. The first viewpoint is that Solana Saga has an overall tone of "weak hardware, mixed web3 experience," with complaints about hardware mainly focusing on the camera, lack of facial recognition, rear fingerprint input, screen display quality, battery life, chip quality, etc., with some users even evaluating the actual value of the phone to be around 3000 yuan. In terms of web3 experience, the overall UI and software receive mixed reviews, but still, due to the bonus of new experiences, it can be eye-catching. The most praised feature is the Seed Vault product, which essentially stores key information on a chip processor outside the Android system, so it can be understood as a trusted UI combining software and hardware, allowing quick fingerprint verification for transactions after storage.

The most criticized aspect is the adaptability of web3 software and the lack of a robust dapp store, similar to METAVERTU. Currently, Solana Saga only supports Phantom, Solflare, and Ultimate, limiting the use of some other ecosystem dapps, including Bitcoin (of course, users can still use web versions or other dapps within the Android system). Additionally, various review articles mention issues with wallet connections failing or crashing in different scenarios, as well as missing functionalities in some dapps, and confusion between Google accounts and web3 accounts. I believe this is not Solana Saga's fault, as Solana Saga's SMS is promoting more developers to improve the web3 phone ecosystem. In terms of the dapp store ecosystem, as of June 2023, the dapp store only offered 30 dapps, most of which are non-mainstream dapps (as shown in Figure 6), and even included Ledger Live in the wallet download directory. Imagine a future web3 mobile scenario: users switch between a Ledger wallet and Solana Saga, changing currencies and passwords across different devices to complete an NFT purchase. Is this what Web3 is supposed to be?

Figure 6

Finally, let's take a look at Nothing Phone and HTC, which are relatively marginal products. Nothing Phone focuses on design and UX, already being an oddity in the phone market, but it does look cool enough. In terms of web3, Nothing Phone chose to collaborate with Polygon and developed some small components like PolygonID, NFT widgets, and wallets, even airdropping a Black Dots NFT to community users. These actions seem more like a marketing campaign rather than a complete web3 phone experience. However, this approach is actually a good attempt for phone manufacturers, adding extra charm to the phone through some web3 small components and promoting marketing through NFT airdrops, while not affecting the phone's web2 experience; it can be seen as a nice addition.

HTC Desire 22 Pro, on the other hand, is a relatively less successful attempt. First, the web3 integration solution used by HTC Desire 22 Pro is to build a set of metaverse components within an Android system, including VIVE wallet, VIVERSE, etc. In reality, HTC not only built a wallet but also created a metaverse platform, which goes against the original intention of web3 decentralization. From the experience of VIVE wallet, it is merely a low-end version of a general crypto wallet, supporting about 20 ERC-20 tokens and not providing any redevelopment pathways, representing a regressive web3 experience. Furthermore, the positioning of HTC Desire 22 Pro as a budget phone does not even mention the so-called chip storage, secure environment, and other features that web3 phones should have; it can almost only be seen as a product aimed at the ultra-stupid only.

Table 1

Conclusion

The topic of web3 phones is too vast and new; while we cannot cover all information in detail, we still aim to make a small contribution to a better web3 future through constructive summaries. The core question we focus on is: what kind of web3 phone aligns with future development trends? In summary, there are the following points:

1. Web3 phones need to move towards a combination of software and hardware. A hardware-based secure environment is an inherent advantage of phones, while the integration of software and hardware is key to enhancing user experience. Some existing web3 phones have strong hardware, while others have good UI, and the next generation of web3 phones will inevitably be products with strong hardware and good UI. I believe that through meticulous refinement and an enhanced understanding of web3, creating a well-rounded web3 phone is only a matter of time.

2. Web3 phones need to move towards universality. Currently, web3 phones somewhat suffer from a self-enclosed ecosystem issue, and universal multi-chain interaction is necessary to truly enrich the dapp store and developer ecosystem of web3 phones. At the same time, we need to consider the relationship between the web3 ecosystem and the phone's native ecosystem; the two can be parallel, supplementary, or overlapping. Regardless of which direction we take, ensuring the universality of the web3 ecosystem on phones is essential for generating user demand for the necessity of web3 phones.

3. Web3 phones need to move towards mass adoption. High prices and niche brands are not choices for the masses, and the inherent mass appeal of phones should not be constrained by objective factors like price and brand. Web3 phones should not become bizarre niche products (at least not in the future). JamboPhone serves as a good example, setting aside concepts like dual chips and secure environments, starting from the two most essential needs of low price and good dapps, and iterating gradually from the bottom up.