Bitcoin hardware expansion comparable to the Lightning Network can be achieved with Teechan, but is the cost a bit high?

2016 is coming to an end, and the development of Bitcoin is still at a crossroads.

Take the Lightning Network scaling solution, for example. This open source project aims to increase Bitcoin’s capacity to millions of transactions per second (extending the value of this technology). Today, the Lightning Network’s code is nearly ready. However, the Bitcoin community is not quite ready for its arrival.

Although some form of Lightning Network can be deployed today, it requires Segregated Witness activation to work, and Segregated Witness requires near-unanimous support (95%) from miners and users before it can be activated. (Currently, miners’ hashrate signals support for Segregated Witness at only about 25%, far below the required 95%.)

This bottleneck has led some researchers to explore other solutions for obtaining payment channels similar to the Lightning Network.

One of the new proposals is Teechan, which was created by the blockchain academic group "Initiative on Cryptocurrencies and Contracts (IC3)" led by Cornell University. The team explained in a statement that Teechan is an off-chain transaction protocol similar to the Lightning Network, which also allows millions of transactions per second - but Teechan's architecture is different from the Lightning Network.

Specifically, Teechan relies on a special type of hardware - Intel Software Guard Extensions (Intel SGX) , which can mask data information and protect it from external detection.

The benefit of this approach is that the hardware is already in the market, so it can take advantage of Intel's ubiquity in the Internet industry.

Ittay Eyal, a computer scientist on the Teechan team, told CoinDesk:

“Essentially, Teechan already runs on the existing Bitcoin network, so the implementation of Teechan does not require a lot of changes. We have already run a transaction channel on the Bitcoin test network.”

According to the team’s tests, the Teechan prototype achieved a throughput of 2,480 transactions per second per channel with a transaction latency of just 0.4 milliseconds.

However, one thing to keep in mind is that Teechan participants need to purchase a specific type of computer with special hardware in order to open a Teechan transaction channel (currently only a subset of Intel’s products run Intel SGX).

Some bitcoin developers said they were not opposed to the idea of ​​implementing trusted hardware across all blockchain networks.

Decentralization Debate

The main problem at present is that there is a conflict between trusted hardware expansion and Bitcoin's decentralized security model , because the former inherently requires a certain degree of trust in Intel.

The use of trusted hardware in cryptocurrency projects, which are largely software-driven, has been met with skepticism before.

Ari Juels, professor of computer science at Cornell University, said:

“That is indeed the case.”

But he believes it is "difficult" to escape all forms of centralization. He said trusting Intel is difficult to avoid because so many devices use Intel chips, even the nodes and wallets that Bitcoin users rely on.

According to data from 2015, nearly 90% of computers run on Intel chips.

In addition to the Bitcoin community, other open source blockchain communities also seem to be interested in the idea of ​​hardware expansion.

For example, Ethereum founder Vitalik Buterin posted a poll on Twitter last week asking his followers what they thought about using trusted hardware as a component of blockchain projects, and the answers were varied.

There may not be a clear answer to this question, said cryptographer Matthew Green, who believes Intel could be a single point of failure.

“It depends on what’s at stake. A global network and millions of dollars? Um, no way.”

Green may have made this statement in reference to the consensus algorithm developed by Intel, Proof of Elapsed Time (PoET).

Other Use Cases

With hardware scaling in mind, IC3 has been investigating other trusted computing use cases beyond micropayment channels. (Intel even recently offered a grant to the academic blockchain group.)

Additionally, of the 13 blockchain projects listed on the IC3 website, at least three use trusted computing to some extent to protect data integrity or privacy.

One of the reasons is that blockchain smart contracts are increasingly being used as data verification tools.

Juels said:

“Almost any interesting smart contract needs to have some awareness of the state of the real world in order to function. Any financial instrument needs to know what’s going on in the real world, whether it’s a stock price or a commodity price.”

But while it sounds easy, retrieving trusted data for smart contracts is a "considerable stumbling block" because the data could have been manipulated before it could be programmed into a smart contract.

Juels believes that the current Ethereum (fork) cannot provide a high level of trust. On the other hand, the Town Crier smart contract project uses SGX to collect data from a website trusted by participants, and the project guarantees that the data will not be tampered with before reaching the smart contract.

The Town Crier whitepaper has now been published, and IC3 plans to release a public version on Ethereum in early 2017.

Another Ethereum project, Hawk, uses Zcash’s privacy technology to hide the sending and receiving addresses of Ethereum transactions.

Programmers can write smart contracts and then run them through the Hawk compiler to mask the contents of the smart contract.

By acting as a 'manager' in some Hawk applications, Intel SGX can help execute Hawk smart contracts and see the input values ​​of a certain user, but you can trust Intel SGX to keep the user's input private.

Future Use Cases

The above use cases may just be the beginning.

Juels believes that trusted hardware will have many applications in the future, and these applications have yet to be explored, and developers have more experiments to try.

For example, Juels pointed out that the bug bounty model can provide rewards to coders, which benefits open source projects and those coders who are capable of finding bugs. However, Juels pointed out that this model often has problems in terms of fair trade. Coders who find bugs do not always get the rewards they are promised, while others who do not find bugs or only find useless bugs are rewarded.

He proposed an idea for a bug bounty for smart contracts. By using SGX’s strong integrity guarantees, bounty hunters could prove that they found a bug and didn’t disclose it to the company.

In the future, IC3 will work on improving existing concepts and continue to develop protocols that use trusted hardware.

Although the public opinion is not currently focusing on this technology, it is safe to say that this technology may become a big trend in 2017 as the major blockchains will compete to improve their capabilities.