Turning Feature Phones into Bitcoin Hardware Wallets

Kgothatso Ngako’s Machankura startup is proving that Bitcoin can be useful to people even when they don’t have a fancy phone, or even an internet connection.

Machankura, which is slang for “money” in Ngako’s home country South Africa, is a software tool that allows users to transact with bitcoin over the Lightning Network, a Bitcoin Layer 2 payment network, on pre-smartphone mobile devices (feature phones).

It’s the first consumer technology that allows people to use bitcoin without an internet connection — an ingenious way of bringing the asset to the financially disenfranchised.

This feature is part of CoinDesk’s “Future of Bitcoin” package published to coincide with the 4th Bitcoin “halving” in April 2024.

Ngako, a 30-year-old former Amazon Web Services (AWS) developer, released Machankura into the world in the middle of 2022. And it has taken off since Ngako last spoke with CoinDesk in April 2023. Its user base has jumped from 3,500 to 13,600 in just under a year, according to Ngako, and the company is currently providing services to residents of seven African countries. Things are on the up and up, and still Ngako wants to continue pushing forward.

The next stage for Machankura, which is currently only a custodial Lighting solution, is enabling Africans to hold their own private keys, bringing them true financial self-sovereignty.

In January, at Adopting Bitcoin Cape Town this past January, an event Ngako helped organize, he announced Machankura’s Bitcoin Java Card project, which will let feature phones double as non-custodial Bitcoin wallets. Java Card is software that allows applets, basic computer programs that perform specific tasks, to run on smart cards, which communicate with the SIM cards in the phones. In this case, the Java Card software will enable African users to hold their bitcoin private keys, sign transactions and see their UTXO balance.

I spoke with Ngako virtually to discuss this ambitious undertaking and to get some more details about what this next iteration of Machankura will look like in practical terms. We also talked about why self-custody is essential for African users.

Interview has been condensed and lightly edited for clarity.

You recently announced the Bitcoin Java Card project, which will allow Machankura users to self custody their bitcoin using their feature phones. Please explain how this works.

The Java Card project is a chip that runs on your SIM card. You stick it on a normal SIM card [and] it adds the functionality you programmed on this chip on top of the feature phone. With this chip, you can then program an applet, which has Bitcoin-related functionalities like signing. We have programmed the ability for it to produce a SegWit address [and] are now just waiting for the next part to complete where you are able to communicate with a relay, something like an Electrum server, where the server feeds you all the blockchain data and then you can [see] your balance.

Now, we can send a message to a relay from the Java card applet, and what we need to do is successfully receive a message via the Java Card project. The SIM card can receive SMS [with] the chip, but the chip cannot trigger events while it’s on the SIM card. That is the part we are [still working on].

Machakura currently utilizes Lightning, but this would give people access to Bitcoin on the base layer, right?

Right. If you look at Muun wallet, technically it’s an on-chain wallet [with which] you can do swaps. [We thought] ‘Okay, cool. Let’s do that.’ The idea is if we just solve the first part of the problem — How do you self custody Bitcoin on any device? — we can then take all the lessons from the other wallets and [apply] them [to] this, as well.

A wallet like Muun is a bit expensive to use, whereas a wallet like Phoenix has proven to be cheaper to use. Will you be looking at all Lightning wallets to find the most cost-effective ways to have the bitcoin transfer from the base layer to Lightning?

I think both Muun and Phoenix [have] good architecture to look at. With the splicing update, Phoenix is bringing in some of the things that made Muun wallet good. It gives you a good multisig on-chain wallet. Since Phoenix does [this] with Taproot, it gives you similar functionalities to what Muun wallet gave you but for less. We want to [create] a hybrid between Muun and Phoenix.

Are you concerned that people might not want to pay that initial base chain fee to open a Lightning channel, especially since base chain fees continue to rise?

I’m paying attention to the conversation around self-custodial Lightning and the costs associated with it. At the end of the day, we shouldn’t chase it being free. Treat the first cost you pay like opening a bank account. Yes, today, opening a bank account is free, but it hasn’t always been free. [It’s] free in a manner that is detrimental to the user. If you don’t put those costs upfront to the customer to say, ‘As a bank or as a LSP or as whatever, these are the upfront costs that we have to pay to give you an account,’ you end up coming up with weird ways to then offset the cost to have a customer. The beauty of Lightning is that once you’ve paid those upfront costs and you’re transacting, you are in this blissful world (referring to using bitcoin on a second layer at lower costs).

You’re very open to feedback on Machakura. Did users ask for a non-custodial version of the service?

The feedback I’ve gotten has really pushed me against doing self custody. With Machankura being a custodial service, people feel like they can call me to say: ‘This didn’t work. Can you fix it?’ A lot of people are comfortable with that. You could still give great customer support with a self-custodial service, but you are on your own with regards to if you lose your keys, there’s nothing we can do to help you. But there’s no reason for me not to give my users [a] self-custodial [option].

If we look at what’s happening across African countries, Nigeria being the biggest example with what they’re doing to Binance (referring to the Nigerian government blaming Binance for causing economic disruption in the country), it’s not in my best interest to be a custodial service. They’ve detained two directors from Binance and requested the CEO of Binance come in to have in-person meetings. That is a ransom situation if I’ve ever seen one.

So, if bitcoin is going in the same trajectory that we are assuming it goes, [and if] fiat currencies are also going in the same direction that we are assuming they go, which is in a downward spiral, somebody has to be the scapegoat. In Nigeria’s case, it’s Binance, and Binance didn’t do anything to initiate that.

I love living in Africa. I don’t want to move to a different continent just to say, ‘Oh, I’m not affected.’ And I love working with the people I’m working with. I don’t want them [to have] to move from the countries they live in. So, I’m like, ‘Okay, it has to be self-custodial.’

That reminds me of something you said on a panel at the African Bitcoin Conference last year. Someone asked you what African governments could do to support greater Bitcoin adoption, and you responded with “ban it.” Why was that your take and is it still your take?

That is an opinion I’ve held for years, even as I was rolling out the service in new countries. When bitcoin is banned, it forces you to deploy [your] service in a censorship-resistant fashion. When it’s not banned, you do what you can and hope for the best. In Nigeria, before they blamed Binance for the economic fallout, they first unbanned bitcoin and then blamed Binance for why the Nigerian Naira is falling. If bitcoin is banned, that is the most ideal [situation] you could have as a builder in Africa. Yes, it forces you to do things in a cumbersome, decentralized, censorship-resistant fashion, but I think that is the way to go.

We have the technology to actually ensure that some of these services are accessible regardless of approvals. [In regulated environments,] we are not building censorship resistant tech because we are too comfortable. A bitcoin ban makes everyone use their brains again.