Positioning SeedSigner for Maximum Impact

Wait… You Want to Order 1,000 SeedSigners?!?

A couple of weeks ago I received a message from Fadi Elsalameen, a Palestinian human rights advocate that I was fortunate to meet at last year’s Oslo Freedom Forum. Fadi wanted to discuss acquiring ~1,000 SeedSigner devices, primarily for distribution among Palestinians beginning to learn about saving with bitcoin.

Our project has been unfortunately hit particularly hard by global supply chain issues, with Raspberry Pis of all varieties becoming exceedingly difficult to source. This took the Raspberry Pi Zero v1.3 SBC that is a critical part of our recommended build from being generally the easiest component to source, to becoming arguably the most difficult component to source for those wanting to build a SeedSigner.

While these availability issues unfortunately made it impossible to consider Fadi’s initial proposal, his request served as a catalyst to begin conversations on how to continue to expand SeedSigner’s impact and reach.

The initial value proposition of using a Pi Zero 1.3 was primarily it’s lack of WiFi/Bluetooth hardware, as well as a mature accessory ecosystem and the ease of building with it as essentially a general purpose Linux-powered computer. Basically, it was a very tinker-friendly platform for someone like me with limited development skills.

But over time some drawbacks of the Pi platform have emerged. The first to manifest itself was the ~45 second or more startup time that SeedSigner users contend with. Another issue is the enhanced attack surface that a general purpose operating system presents. And then of course, came the aforementioned supply chain and availability issues.

We’ve Stumbled onto Something Important

Despite these challenges, the Pi Zero platform has given us the immense gift of being able to prove out the model of a stateless, air-gapped, DIY bitcoin signing device. Who would have thought removing private key storage from a hardware wallet’s functionality set could be a compelling user experience? It’s the kind of thing you just have to try to see how it’s going to work and whether it will be useful to people. And while it may not end up being everyone’s preferred solution, our enthusiastic and growing user community consistently reminds me that it feels like we’ve stumbled upon something important.

Removing private key storage from our signing device presents the unique advantage of being able to use a SeedSigner to manage multiple cosigners in a multi-signature wallet; to put this another way, our device not storing your private key means you don’t need to buy a different hardware wallet for every private key that you want to use. It also opens up the possibility of multiple trusted individuals sharing a single signing device.

Another design decision that arose from my unique professional background was SeedSigner’s exclusive use of an animated QR-code exchange process to enforce severe constraints on the signer’s ability to communicate with the outside world while in operation. We’ve come a long way in removing friction from the QR exchange process, and an unintended positive consequence of this design choice was making SeedSigner potentially compatible with any computer or mobile phone with a camera.

These unique properties of being something people can build themselves, statelessness (not storing keys), and usage of QR-exchange for external communication, uniquely position SeedSigner to help solve what I refer to as the bitcoin storage ecosystem’s last-mile problem.

The Saving-with-Bitcoin “Last Mile” Problem

People in developing economies may first begin to interact with bitcoin via remittance payments, or as a medium of local exchange, and many as they become more familiar with the use of bitcoin will begin to carry a balance over time. Perhaps as the bitcoin price appreciates, or perhaps as their own local currency depreciates, many will turn to bitcoin as an inflationary hedge or savings mechanism, building a larger and larger balance as time goes on. They may begin with a lightning wallet, or an on-chain hosted wallet but as their balance grows, they will (or at least should) become uncomfortable with carrying their savings around in their pocket or trusting its safety to a 3rd party. And as they research bitcoin security, they will want to hold their own keys, they will want to hold them offline according to best practices, and they will want to leverage multi-signature wallets where they make sense

This challenge is unarguably coming and I don’t see the hardware wallet market as positioned to solve for it. The common “hardware wallet” model of storing a single private key enforces a 1:1 relationship between private keys and hardware wallets (and you need to store a secondary copy of your seed somewhere besides). This 1:1 correlation serves to benefit the hardware wallet makers & their bottom line, but for users it can get expensive quickly to set up even a smaller multi-signature wallet. Also worth considering is the issue of crossing political borders with one or more hardware wallets; a device that stores private keys is undoubtedly a potentially tragic seizure target.

But again, if the device used to generate signatures does not persistently store a user’s private key, such a “signing device” that can be used with multiple keys becomes more akin to a simple, inexpensive ink pen rather than any kind of security-critical digital vault. This model also encourages users to leverage math and the bitcoin protocol for their security, in the form of BIP39 passphrases (“the math”) & multi-signature wallets (“the protocol”). Imagine: “…can you take this signer to grandma to let her use it for a few minutes and then bring it back to me?”

Most people also realize that developing parts of the world have bypassed desktop & laptop computers with most internet users potentially classified as “mobile only”. Connecting a hardware wallet to a mobile device is a bit clunky and often requires an adapter to make the connection. Running a sneakernet via MicroSD may be a little less cumbersome with Android devices, but it’s a no-go for Apple devices without again, some kind of adapter. SeedSigner conversely is compatible “out of the box” with just about any mobile phone.

Also a Compelling Experience for 1st World Bitcoin Savers

But to take a step away from developing economies, we’re also seeing significant SeedSigner enthusiasm in North America, Europe and other more developed parts of the world where conventional hardware wallets are significantly easier to find and afford. For many of the people who are drawn to SeedSigner, privacy is a significant concern and they prefer to source and build their own bitcoin storage tools to avoid somehow finding their way onto one or more “lists of bitcoiners”.

And in these same areas of the world we’re also seeing uptake from people who just recognize that over time we’ve developed pretty darn intuitive models and workflows for interacting with the building blocks of the bitcoin protocol (private keys, extended public keys, and partially signed bitcoin transactions). SeedSigner can make working with these bitcoin primatives much easier to understand and manage, without dumbing anything down.

So Where Do We Go From Here?

I’ve digressed a bit from the original topic of “Where do we go from here, to position SeedSigner for maximum impact?” As I’ve pondered maximizing the project’s reach over time, Fadi was a uniquely positioned individual to get in touch — he has a unique potential symbiosis with our project given that he straddles the humanitarian and capital formation worlds simultaneously. Fadi acutely understands the personal challenges faced by people in some of the most challenged parts of the globe, especially so given his personal background as well as his work with Alex Gladstein and the Human Rights Foundation. Fadi conversely fluently speaks the language of business and has the attention of capital allocators in the bitcoin world and beyond.

The ultimate solution as we consider how to move forward is becoming financially self-sufficient. We have received many generous donations to support our development work and help compensate/motivate contributors, especially donations that were received to fund Keith’s full time work on SeedSigner in Q1 2022, pulling a lot of future development goals very quickly into our present users’ hands. But a reliance on donations can make it difficult to plan for the future — most open source projects will wrestle with this challenge as they grow larger and more complex over time. The most sustainable model that has evolved in the FOSS world is to develop some kind of for-profit consumer product or service offering that can “pay the bills” and support ongoing open source development.

And Where Will the Money Come From?

My experiences in pursuing grant funding over time have only served to reinforce my belief that we have to find some kind of “path to a profit”. Though I consider SeedSigner to be a software project, the goal of SeedSigner is to have people building their own signing devices, and most of the grant organizations I’ve interacted with have provided the feedback that “hardware projects” typically have a “path to profitability” and thus aren’t eligible for grant funding. (The one exception to this perspective so far has of course has been the Human Rights Foundation, who generously provided us with not only a development grant in the Fall of 2021, but also much appreciated public recognition of our bitcoin saving model.)

Conversely, in many of our preliminary conversations with capital allocators it has become clear that “hardware projects” are not anticipated as offering the alluring potential investment multiples that other products that approach a near-zero marginal cost of production can offer (software, scalable services, etc.). (I should also note that we have not been entirely rebuffed by venture allocators and have been able to successfully initiate a very small number of early but potentially productive conversations.)

I am very sensitive to how this potential funding direction will concern some members of our community, given that SeedSigner has tried to embody a FOSS mindset from the earliest beginnings of the project’s existence. Having investors, employees, contracts, deadlines, expenses — all of these can impact underlying motivations and practical development, design and operational decisions. From our perspective, I am optimistic about our ability to “walk and chew gum” simultaneously in good faith, and feel as though an earnest and honest approach to these potential tradeoffs can keep our efforts firmly inside of the net-positive side of the spectrum of outcomes. And the nature of open source development will also perhaps provide the ultimate check on us; if you disapprove of our decisions or behavior: fork the code and continue building your vision of what the technology and approach should look like.

DIY Support Isn’t Going Anywhere

In practical terms, I want to be clear that this does not mean any sort of pivot away from either open source or our ongoing and perpetual commitment to DIY. Getting SeedSigner to where it is today has been made entirely possible by the network effects of the larger Open Source movement and it would be utterly foolish of us to change to a “source available” or closed source operational stance; a closed source approach flatly makes no sense from a logical perspective in an endeavor to build a *stateless* signing device. The security assurances of the stateless signer model require that it be auditable and verifiable as such. We additionally remain committed to our Raspberry Pi builders, recognizing that the RPi ecosystem is the most ubiquitous, consumeraccessible single board computer platform in the world. I feel confident that RPi availability will rebound over time and we want to continue adding new features and refining the platform to continue to serve DIY-focused users. The ability to source components and build a bitcoin savings tool in private is critically important for so many reasons.

What We’re Focusing on Right Now

Whew. So now the question in practical terms is where we go from here? I would like to offer up some bulletpoints to summarize what we’ll be working on over the coming months…

Extending Keith’s full-time work on the SeedSigner RPi platform via community donations: Keith has indisputably proved to be highly productive and accountable with the resources the community has generously provided him. He has brought our new code structure and user interface into reality, thoughtfully and with many additional features that were not even planned as a part of the upgrade. We have ample development ideas to keep him engaged and building on the RPi platform for at least Q2 2022, and likely for much longer beyond that. In the very near future we will be publishing proposed RPi development goals for Q2 and will be soliciting the community’s input on and support for his continued work.

Incorporating the important work of several other contributors that has continued in-parallel to v0.5.0 development: @Desobediente_Technologico has done amazing work on a custom BuildRoot OS that we are very excited about; @mutatrum has done important work on optimizing our display performance and expanding our support for other display/control hardware; @samkorn has produced a proof-of-concept standalone SeedSigner emulator that is TailsOS compatible; and there are other outstanding innovations and improvements pending as well. Plainly said, we want to explore and integrate as many of these exciting innovations into the project as we’re able to.

Refining and disseminating our pitch for external grants and investment: Our unique mission and approach are going to entail finding the right match among grantors and capital allocators. I firmly believe that the right parties will rise to support our mission, but we are going to have to do the hard work of widely and effectively communicating both the SeedSigner value proposition, and our ability to execute on it, to bring them into our corner. Getting behind SeedSigner is both good for bitcoin, and good for the world; we just need help the right people understand this.

Exploring alternative microcontroller architectures: The decision of which microcontroller platform to ultimately pursue building a larger offering on is probably the most important decision we will be faced with over the near term. Making the right choice will require leveraging the knowledge and experience of any and all experts and hobbyists that are willing to share their knowledge in this area. We will be actively seeking out people with this expertise.

Continuing to leverage opportunities to spread the word about the project and get SeedSigner devices into people’s hands: A “wallet” that doesn’t store your private key is still a crazy, crazy idea to most people in the beginning. It’s something you either have to do the research and think a lot about (kind of like bitcoin itself!), or alternatively see first-hand, in action and explained by someone who already gets it. I will be actively seeking podcasts appearances, opportunities to present at bitcoin meetups, and speaking opportunities at conferences and other events to help more people understand the value of our approach to bitcoin saving.

Lastly, What You Can Do to Help

We’re going to be busy, but we’re going to need your help too. Here’s how you can help keep the ball moving forward with SeedSigner:

Continue to share the project with other bitcoiners: This is the most important. We are a grassroots effort without any kind of marketing budget. If what we are building adds value to bitcoin (which I emphatically believe it does) please share it with others. Let your favorite podcasters know you’d love to hear an episode featuring SeedSigner. Network effects are powerful and growing the project’s audience not only expands our reach, but helps get us on the radar of new contributors.

Get in touch if you have specific knowledge of and/or experience with microcontroller platforms: We will be weighing the advantages and disadvantages of various options, including: availability, cost, programming language support, DIY-feasibility, security concerns/features, and other factors not yet on our radar. Those with knowledge of the specific benefits and tradeoffs of different architectures will be a huge cheat code for us as we continue to learn.

Consider a donation to our general development fund and/or make a donation to Keith directly to extend his work focusing on SeedSigner. Our unique opportunity to monopolize his attention and development skills can continue to pull features that could be months or more off into the nearer future.

Keep an eye out for a call for translators: Incorporating multi-language support into our user interface (and later into seed word options) is a high priority update for us. We will be making a call for help with translations soon and appreciate everyone who is willing to pitch in with their time and expertise so that many more people can interact with a SeedSigner in their native language.

In Closing

I want to thank Fadi for his very public support of our project on the main stage at Bitcoin 2022, and I want to thank Keith, Nick, and all of our other contributors for their vital work to improve and grow SeedSigner. Just about exactly a year ago, in April of 2021, I tweeted that:

“The intention behind SeedSigner is to provide a high-quality, dead-simple signing experience, with as low of a hardware cost as possible. It can give people across the ENTIRE #Bitcoin net-worth spectrum access to airgapped multisig hodling…”

This pretty succinctly sums up what the project was about then, and what we would like SeedSigner to continue to be about in the future. We appreciate everyone’s interest and enthusiasm and look forward to seeing where this mission takes us.