During Dappcon Berlin we interviewed Matt Luongo (Thesis/Keep Network) about tBTC, a project that aims to bring Bitcoin to the Ethereum ecosystem. This post is a transcript of the AMA we had as a follow-up with Matt in the Chorus One Telegram channel. It covers Matt's background, how Keep and tBTC came about, how tBTC compares to other interoperability solutions, as well as how the initiative was received in the wider ecosystem.

Welcome Matt, thanks for joining us for this. Why don't you start with giving us a bit of background about yourself and tBTC?

I've been working in the crypto space for a while, and started a Bitcoin 1.0 company in 2014 that's still around - foldapp.com. In our downtime, we were working on how we could decentralize the core market behind our business. In early 2017 that brought us to Ethereum.

Quite quickly, we realized that the issue we were trying to solve required confidential data custody- what I like to call "autonomous privacy". The work we did there became the Keep network.

About 10 months ago, we decided to launch the first app on the platform we'd built. tBTC is that app, a trustless bridge between Bitcoin and Ethereum. Quite a journey from there to here.

Definitely sounds like it! So tBTC is the first (and best?) use case you realized the Keep Network could be used for? Can you talk a bit more about the Keep Network itself and about some other applications you or others are working on that are using it?

Sure! tBTC is maybe the most obvious use case- the market will tell us what's best heh.

The basic idea behind Keep is the idea of, well, "keeps"- multiparty computation groups that allow for secret computation and storage off-chain, while remaining provable on-chain.

Can you please expand on what you mean by "autonomous privacy"?

So instead of per-user privacy, we're talking about a construct that can be used for multiple users - think a private auction where the best price wins trustlessly, without exposing bids.

In my opinion, this is the stuff we need to replace entire institutions. One of my favorite use cases is for trustless credit scoring. That's why we started talking about "autonomous privacy". How can a smart contract read an encrypted file or sign a message off-chain? You can't put key material in the contract, so there's an issue with where you can root trust. MPC gives us strong tools to solve those problems. And it turns out, "decentralized signing" can be extended to build robust cross-chain systems.

So is this a decentralized Anchorage then? The full power of automated asset management without giving up custody?

I think that's a good way to look at it! Though at the end of the day, you still have to protect any keys involved on Ethereum as well. You could certainly build something like BitGo on top of this network.

It seems like extending "decentralized signing" to build robust cross-chain systems is the core idea of tBTC. Trust-minimized interoperability between chains. Now, there appear to be many other solutions working in a similar vein. One of these is the wBTC approach, or other similar federated pegs. How does tBTC compare to these?

The basic idea behind a simple federated peg is that you deposit into a multisig on one chain, then those validators mint a token on the new chain. The problem here is that those validators only have their reputation on the line- if they steal funds, you have no recourse outside perhaps the law. If they get hacked, that's it. In this system, we've added two important components.

  • First, use *many* federations so the entire economy won't collapse from one hack. For us, that's a federation per single BTC.
  • Second, bond the validators on the new chain. If a validator is hacker or absconds with funds in this system, you can take their ETH or other on-chain assets

I've been calling this idea a "bonded multi-federated peg" to distinguish from the more centralized approaches of Liquid or wBTC.

So a bond is used to collateralize the BTC it is securing. Seems like this approach is more expensive than using "trusted" parties. How do the economics work here?

Initially, for sure - it's always more expensive to build a trust-minimized system than it is to trust someone :)

As the network bootstraps, you need a pretty significant signing bond per deposit - think 1.5X the BTC deposit in ETH to guarantee against mischief. As the network grows, you can actually lower that significantly- even below 1X ETH as you allow tBTC to act as a signer bond. On the staking / signing side, we're proposing a low fee for deposits initially- 50 BPS every 6 months - but we want to make that number market-discoverable after launch.

What do you think about other approaches like Cosmos IBC or the Interledger protocol (if you are familiar)? Do you believe there will be one dominant way to move assets across chains or will multiple solutions co-exist?

Big fan of both projects. I'm typically a pluralist- there's room for a ton of different chains and approaches. We're loosely planning to launch on Cosmos, and are considering Polkadot as well. The great thing about interop is that it shouldn't be winner-take-all market - the more we build, the more we all win.

That is a great point. In the interview with Brendan you mention tBTC being not just a technological but also a social bridge from Bitcoin to Ethereum. How did you experience the sentiment of Bitcoiners and Ethereans so far? Was there any significant difference between how these communities view tBTC?

Ooh, interesting question. My favorite part about this project is getting an earful from both sides. In general, the ETH folks are usually considered more "open", but there have been some Ethereans who are worried this might hurt ETH. On the Bitcoin side, we've gotten great feedback on what they need to feel this is a trustless bridge - in particular, our initial price feed was trusted. We thought that was a nice way to reduce scope for v1, but they're not having any of it :D

In general, as a long-time Bitcoiner myself, I'm very sensitive to "shitcoin" and "useless token" concerns. Turns out if you build a thing that has a reason to exist, those don't come up so frequently.

If there are no more questions from our community maybe we can already start to wrap up with a question someone added to our Google doc: what are some other interesting projects in the crypto ecosystem right now that you are excited about?

I'm following everything in DeFi closely- Synthetix, Maker's MCD, the latest from Compound- and obviously I have an interest in interop.

It's about time for many of these fancy (over?)funded L1s to ship, so also watching closely there :)

Most definitely, more reasons for good interoperability solutions :D

Exactly!

OK, so I think we can call this a day. Thanks so much for shining some more light onto tBTC and your view of interoperability in the space!

Thanks so much guys! Feel free to learn more about tBTC at https://tbtc.network and follow along :)