Deep Dive : Solana Mobile Stack & Saga
What if, your phone became your wallet...
In this essay, we deep-dive into the Solana’s mobile stack & Saga - the flagship phone they are building with the team @ OSOM. Thanks to Austin Federa, Shek_Dev, Srijan & Anoushk for discussions that contributed to this essay.
TL;DR
Mobile matters. 4G >> Wifi , Mobile >> (Laptops + Desktops + Tablets) when it comes to adoption. For Crypto & web3 to go mainstream, it needs to go mobile.
Web3 is a paradigm shift: money is baked-in & self-custody is the norm. Traditional phones are not optimised to provide for the design & security needs of the use cases that are now possible. Signing transactions directly through your phone is a super power.
Key features of the Solana Mobile Stack: Seed vaults to secure the private key at the device level. Mobile wallet adapter to standardise signing UX across dApps. Solana Pay support for Android - with tap & scan to pay. A dApp store that doesn’t feel like a toll booth.
Any android hardware manufacturer can use the Solana mobile stack SDK to launch a crypto-native phone. Any software developer can build apps on top of it. It is open-source & permissionless. Saga by OSOM is hopefully the first of many.
Get Involved : This is your chance to shape the future. A privilege often reserved to a small group in a small part of the world. Skip ahead to the RFP section for ideas on what makes for a great grant application.
Why launch a phone?
The first question that comes to mind is, why does Solana or any other blockchain for that matter, need to launch a phone at all? Android & Apple smartphones have seen parabolic growth over the last decade. They are cheaper, faster & more performant than ever. So, why would someone need to launch their “own phone” ? What exactly is a “crypto phone” ? And how on earth will it compete with Apple & Google????!
We’ll answer this question in two parts.
First, if crypto & web3 has to go mainstream it has to deliver a world class mobile experience.
It is generally accepted that mobile phones have provided billions of people access to the internet. But I think we under appreciate that for most of these people a mobile phone is the ONLY way they access the internet.
Here are some stats to really drive home just how far ahead mobile adoption is globally compared to broadband connections. India has a total of 17.8 million broadband connections, amounting to 1.33 connections per 100 people. We also have 1.2 billion cellular subscriptions, amounting to 87.3 subscriptions per 100 people. Brazil has 113 cellular subscriptions per 100 people & 13.7 broadband connections.
While the trend of higher mobile subscriptions per 100 people is generally true across all countries, the extent of difference is particularly stark for developing countries.
The reason for this is obvious - having a broadband connection is expensive. Let’s quickly estimate the cost of this in India. First, you need a laptop which will cost you ~$700 on avg. The most basic broadband plan will cost you ~$7/month. Along with this you need a stable electricity connection & a workspace where you can use this setup. Most people have access to neither. Developing countries have larger household sizes & lesser sq. ft per person. Lastly, you need a permanent address with the required documentation to get a connection. On the other hand, the avg cost of smartphones is ~$200, a prepaid cellular data plan of 1GB/day for 9 cents.
This should help establish that the future is overwhelmingly mobile. Existing devices like laptops, tablets have lost the race for global adoption (while remaining obviously useful to many people, like you & me). More immersive devices like headsets & glasses will need to first achieve mass adoption before they can stake a claim on our metaversal future.
This brings us to the next section, it is impossible to deliver a great crypto experience on traditional Android & Apple devices.
Crypto Native Phone Design
We’ll break this down into three sections - security, design & app distribution.
Security
The most important & novel concept in crypto is the ability to self-custody digital assets. This includes your NFTs, tokens and even your data. We do this by signing messages with our private key. A user can & will have many accounts each having their own private key. These accounts are set up and operated through wallets. A recovery or seed phrase provides root access to all accounts in case a user forgets their password. If you or your wallet leaks access to your seed phrase, anyone can access your wallet and steal your assets, NFTs, data etc.
This is what happened in the recent slope wallet hack. The wallet was sending user's “private key material” unencrypted to a logging service from where it got hacked. So, not only did people lose access to the wallets they created with slope, but ANY of their wallets that shared the same seed phrase.
The team at the Solana foundation argues that this shouldn’t have been possible in the first place. All devices have something called a Secure Element. This is a dedicated part of the hardware that can’t be manipulated externally beyond specific interactions. This is how our biometric data is secured by existing Android & Apple phones today.
And while the capability to do this exists, it can only be done at the OEM level. So, the team at Solana could either lobby android, apple & hope that they listen OR build their own phone and distribute it at sufficient scale which forces larger players to listen.
So, that’s reason #1 : Only by working directly with hardware manufacturers can we ensure adequate security of the seed phrase. This ensures that no wallet or dApp can accidentally or intentionally leak it & compromise all associated user assets and wallets.
This is a UX unlock disguised as a security feature. Today, apps require you to open up your wallet to sign a message. On the Saga, they can interact directly with the secure element via an API. This will reduce the friction in carrying out on-chain transactions from within rich applications significantly.
Apps like StepN have achieved this UX today by implementing their own in-app & on-chain custodial wallets. But it is far from ideal. You need to set this wallet up separately & maintain another seed phrase. You can then transfer assets to & from this wallet. Alternatively, you can rely on deep-links provided by wallets. This doesn’t scale because not all wallets may provide them. They will be non-standard across different wallets. And may not encompass all the functionality required by your app. Or import a wallet by manually typing in the seed phrase.
Using the Solana mobile stack, you can basically use your device to sign messages. So each app won’t need to build and maintain a custom wallet, nor will we need to rely on access via deep-links or QR codes from a few wallets. The mobile wallet adapter standard will provide seamless & secure authorisation.
Design
There are two fundamental design assumptions baked into our mobile experience today. The first one is the process of authentication & authorisation via login. The second is the movement of financial assets across products & services.
Let’s start with authentication & authorisation via login. The fundamental assumption in today’s software is that data & code is hosted by the corporation owning the application. Users “request” for access to their posts, tweets, and social connections by logging in. Developers access this data via well defined APIs & publish through an approval process set by the corporation.
I’m not making an ideological comment here, but a technical one. Today’s software is optimised for a single authority reading & updating all code and data within an application. And don’t get me wrong, we’re living in an era of apps far superior to anything we’ve seen so far. But to paraphrase the famous saying - what got us here won’t us there. This design does not scale for apps & protocols that are built around the principle of self-custody.
Apps built using crypto & web3 principles don’t claim ownership over access to their user’s data. The most critical elements will be hosted on a public blockchain (like Solana) & secured through a user’s wallet. Each time you want to access an app, you will connect and provide it access to your data. Another app developer can just as easily request a user to connect their data to a new application they’ve built and also update it. All of this does not require any permission from the application where this data was originally generated.
This is why the ability to authorise transactions using your phone is a superpower. Today “login with google/apple” is a ubiquitous pop-up on our phones and the dominant way of authorising access to our apps. In the future it’ll be via our phone as a universal wallet (see the minting experience in the video below the image).
Flow of money
The great thing about today’s consumer apps is that they’re mostly free to use. This has been critical to achieving mass scale & connecting billions of people. Advertising was the dominant business model of the internet. With platforms like Google, Meta, Amazon & now Apple, Netflix are the ad exchanges. Through their product innovations and network effects they became custodians of our attention & data. Thus effectively dictating terms to advertisers on the various ad formats, targeting methods, pricing models. This allows no room for direct interactions between brands and customers.
Web3 powered apps will most likely still retain advertising led business models but reduce the power of the ad exchanges by transferring ownership of the most valuable asset (data) back to customers. Brave is already pioneering this approach by enabling direct financial interactions between advertisers & customers. Multicoin Capital wrote about the idea of DataDAOs, which have the potential to collect large scale consumer data while allowing people to retain control over how it is used & a share in the financial value derived from it.
The second business model of the internet is subscriptions. A way for consumers to directly pay for the services they want to use. This is accessible to a much smaller and relatively affluent set of people. Today, mobile app subscription transactions are allowed to happen only via the respective app stores. The payment methods used are restricted to the ones supported by these platforms & a 30-50% commission is charged on each transaction. This is why many services like Netflix don’t support purchases through the app store because they feel like they don’t really depend on the app stores for discovery. Subscriptions are typically rigid, billed monthly or annually. Transactions are always structured between the corporation owning the app & the users, never peer to peer - leaving massive value on the table. Reward programs attempt to fill this gap by (trying to) gamify app interactions and assigning an internal token value. Even these aren’t p2p, have limited spending sinks and are unusable outside the originating app beyond some vanilla “partnership” deals. We spoke about the limitations of today’s approach and possibilities of a crypto-native loyalty program in this essay.
In crypto, everything can be a financial asset that flows seamlessly through products. I’ll reserve comments on whether this is good or bad, but it is happening. This makes today’s advertising and subscription models look limiting. Real-time streaming of tokens as payments, NFT-gated content access powered by infinite asset types freely tradable between people and apps will unlock a new paradigm of interactions that our existing infrastructure can’t support.
That’s reason #2, self-custody of data & assets requires a signing entity (wallet) to be built into the fabric of the phone.
App Distribution
After reading the previous 2 sections, this is the most obvious next limitation that comes up. Publishing an app is entirely at the discretion of Google & Apple. There have been sufficient instances of apps being delayed, denied approval or taken down. The business model & incentives are misaligned too, as Chris Dixon outlined in his essay on Why Decentralisation Matters. Platforms want to maximise the value they can extract, while publishers want to pay the least possible. Often platforms can entirely vertically integrate into categories where they see potential (like Google or Apple Pay).
I feel for the teams facing the brunt of these tactics but I don’t think these matters are going to be settled in the court of law. These platforms are corporations designed to maximise value to their shareholders. They have built the largest and most thriving ecosystems of our lifetimes, albeit with the help of apps. But they have God mode & they’re going to use it.
Reason #3, The best protest is to build a better alternative. A world where the app store is minimally extractive & more transparent.
So far, we’ve understood why mobile matters & what a crypto-native phone looks like. In the next section we’ll answer the question - what exactly does it mean to launch a phone ?
OSOM, Saga & the Solana Mobile Stack
There are two key components - the device (Saga) & the software stack (Solana Mobile Stack). While they come together in an attempt to build a crypto phone of the future, it is just as important to understand them separately.
The Saga is a phone being manufactured by a company called OSOM, led by Jason Keats. It is designed to be a standalone flagship android phone powered by Google mobile services & the Solana mobile stack.
Let’s take a bit of a detour & get to know the CEO of OSOM (or chief hooligan as he likes to call himself), Jason Keats. He has easily one of the coolest and unconventional backgrounds I’ve come across in a while. Started off with a degree in Astrophysics from Berkeley to call himself a rocket-scientist. After graduating he started a motorcycle company, worked on Solar Panels and then ended up working at Apple on the macbook air, ipad, an infrared camera and many other (to this day) top secret projects. An interesting anecdote from a podcast he did with Anatoly - the camera he worked on produced one handmade unit & sold for a couple of million dollars. If the owner ever uses it and needs to get it repaired - there’s a note in there saying “Call Jason”. He then went on to work at Andy Rubin’s Playground where he worked on Essential. And is now leading OSOM, building the hardware of tomorrow. In fact, he had never worked in crypto until their partnership with the Solana foundation.
The Solana mobile stack (SMS) is an open-source SDK that can be integrated into any Android phone by the OEM (manufacturer). Yep, that’s right - the team is building this open-source from day 1. Any Android OEM, today, can integrate the Solana mobile stack into their devices & compete with the Saga. You’ll probably get a grant from the foundation for that as well.
Quoting from their github page, there are currently 4 key features of the Solana mobile stack.
A seed vault: The Seed Vault is a system service providing secure key custody to Wallet apps. Your keys, seeds, and secrets never leave the secure execution environment, while UI components built into Android handle interaction with the user to provide a secure transaction signing experience to users.
Solana dApp store: It will provide a distribution channel for apps that want to establish direct relationships with their customers, without other app stores’ rules restricting the relationship or seeking a large revenue share. The goal of the Solana dApp Store is to empower the Solana community to eventually play a key role in managing the contents of this app store. One of the hardest problems to solve here will be finding the right moderation policy. While it will take time to nail this down, there will be significantly more community (developers & device owners) participation than there is today. Including governance structures & the ability for anyone to float a proposal to initiate a change they wish to see.
Mobile Wallet Adapter: Mobile Wallet Adapter is a protocol specification for connecting apps to Wallets on mobile devices. It allows Wallet apps to provide transaction signing services for different types of apps on mobile devices. The protocol is not limited to Android devices either; it envisions similar support for iOS devices, as well as the capability of Wallet apps to provide signing services to applications running remotely, such as on other mobile devices, and on desktop or laptop computers.
Solana Pay for Android: A reference implementation demonstrating how Wallet apps can use the system features of Android devices to capture Solana Pay URLs via QR codes, NFC taps, messages, and web browser interactions to launch Solana Pay requests.
The primary reason why the partnership with OSOM is so important (apart from the fact that Jason seems like a legend) is that the SMS needs hardware support & access to the secure element we spoke of earlier, to unlock its full potential.
And this brings us to the next section, what’s the endgame here? Does the Solana foundation want to out-manufacture Apple & Google? If yes, then why open source your SDK? How will they ever compete with the manufacturing might of the existing players? What about all the other crypto phones?
The Endgame
The goal with the Saga is to demonstrate the power of the Solana mobile stack, by building an “unconstrained playground for crypto’s first 50k users”. Instead of continuing to operate within the limitations of today’s phones. The team believes this is a worthy tradeoff which will push the space forward overall.
OSOM is a private company & they do want to maximise the impact they can have with the Saga. At the same time they hope that as they achieve scale, competitors start taking note and integrate the freely available SMS into their devices. Positive sum thinking can be so strange.
Request for Proposals
If you’ve kept up so far, are excited about the future of a crypto native phone AND want to contribute to it - The Solana foundation & Superteam would love to support you.
The Saga is currently shipping to the US, EU, CA & UK. Importing it into other regions makes it prohibitively expensive. If you are an Android hardware manufacturer who wants to run SMS on their devices & ship them in your country - reach out to the Solana foundation or Superteam.
If you want to build applications, tools, infrastructure, etc. on SMS, you can apply for a Solana Mobile grant or a Solana foundation India grant.
What makes a good grant application?
Building core utilities that are necessary but hard to monetise (e.g. a free to use, open-source, mobile-native wallet)
Building infrastructure that helps others build better & faster (e.g. a gasless NFT minting layer)
Building apps with potentially high consumer pull (e.g. games like StepN)
You’re building as open-source as possible.
Here are 9 ideas to get you started:
A “default” mobile wallet interface: So far, wallets were our gateway to using web3 apps. Everytime you carried out an action on-chain you needed to sign a message through your wallet. Now, your phone is your wallet. This makes the core functionality of authorising transactions, viewing & transferring assets a feature not an app. This will push wallet apps to find other ways to monetise, just like UPI is forcing payment apps to monetise using lending, investing etc. An open-source wallet built on top of the mobile wallet adapter that users and apps can use is a valuable public good.
Gasless NFT minting : NFTs will play a critical role in onboarding new users to crypto. Especially as brands look beyond PFP sales & build user-specific NFTs with real utility. We illustrated a few examples in a recent essay about the FTX Crypto Cup. Simplifying the UX & allowing communities/brands to pay gas is key to a magical first onboarding experience (Even though gas is negligible, the experience of buying 0.0000000001 SOL is high friction). Candy Pay has built a no-code mint builder & gasless APIs that work with today’s wallets. Building this experience natively for SMS will be immensely useful to all the other apps.
Pokemon Go: Straight from Austin, a game where you can mint Pokemons based on your IRL location, verified by Helium.
An NFT Gallery: Today, our phones have a stock gallery app (mostly google photos). PFPs remain an important use case of NFTs. Some of the best and most valuable communities will be built around NFTs. There are many reasons why we haven’t seen a gallery take off yet - users typically split NFTs across multiple wallets, stake them, etc. With the phone becoming your universal wallet, a default gallery that displays all of your NFTs is a utility application.
A Web3 profile: An app that consolidates your identity across domain names, NFTs and other assets. Our web3 identity is being formed across the many different communities, protocols & dApps we use. Service that issue domain names is a first step in building a digital home for us. A mobile app that allows us to freely connect NFTs that represent proof of work, attendance, membership etc. has the potential to become the “identity” primitive on Solana.
A Camera app & creator studio: This is an idea inspired from Cini, a project that recently won 2nd place in the mobile track of the Solana Summer hackathon. A camera app that allows people to list their filters, designs, photos, reels as NFTs for others to use & compose on has the potential to drive the next generation of photo sharing apps.
A fitness app (primitive): Build an app that mints steps taken (& other movement) into an NFT. This can become a fitness primitive used by many other apps & services.
Social publishing: An editor that lets you publish to protocols like wordcel, spling, etc.
Location based rewards service: An app that allows people to complete location based tasks for rewards, all verified on-chain. data entry for maps, reviews. E.g. tag information on hivemapper, aggregate reviews, unlock rewards, etc.
Applying with these does not guarantee a successful application. Neither should you constrain yourself to these ideas. This is just meant to help you get started. The possibilities are endless…
Disclaimer: This post is not financial or investment advice. It is meant for informational & educational purposes only. Please do your own research about risks and compliance before buying, investing/ or trading.
Cover image source: Mid Journey