FYI.

This story is over 5 years old.

Tech

Promises Aside, There’s Still No Fix for Bitcoin’s Fatal Flaw

The GHash mining pool has said it will stop mining bitcoin after reaching 40 percent, but a technical solution is needed to fend off the risk of a 51 percent attack.
Image: Shutterstock/Lightboxx

Perhaps the biggest threat to Bitcoin’s future still has no solution. When the GHash.io Bitcoin mining pool gained over 51 percent mining power in June, it caused a stir. Any mining entity with that kind of control can effectively game the system, as miners vote on every transaction before it's added to the blockchain public ledger. A 51 percent stake would therefore give a group the power to block transactions or to spend coins that don’t belong to them. And it would render null the golden promise of Bitcoin: decentralisation.

To assuage fears around its potential hold over the cryptocurrency, GHash and its owner CEX.io held a meeting in London earlier this month with a number of Bitcoin experts. After the gathering, GHash said it would not allow itself to go over 39.99 percent of mining power—but it won’t be doing so through a technical measure. Instead, as chief editor at CEX.IO and GHash Helga Danova told Motherboard, the company will simply ask participants to leave if its power is becoming too great.

Advertisement

“If we start approaching the respective percentage, we will be actively asking miners to move to other pools through all our communication channels, such as Facebook, Twitter, Blog, BitcoinTalk, Reddit, announcements on the website, personal connections with mass media and other Bitcoin market players which play significant roles in mining,” Danova said. It’s a simple, low-tech way of addressing the problem that doesn’t provide any long-term fix.

According to the post-briefing document from GHash, those who convened in London on 9 July could not agree on which technical patch would be best. Danova said no other pools turned up to the meeting, though they were invited. “If a technical solution had been found, it would already have been implemented,” she added.

One of the attendees, Jonathan Levin, co-founder of cryptocurrency specialist and statistics provider Coinometrics, told me the underlying problem—that it’s more profitable for miners to join a larger pool rather than a smaller one—has not been “fully grasped.”

“For a larger pool, they spend less time on average working on redundant mining problems since they are solving more blocks. Information takes time to travel across the Bitcoin network and as a result some miners spend longer than others working on old blocks that will never make it on to the main chain,” he added. “This increases the expected return per hash done in a larger mining pool.”

Advertisement

One of the more attractive solutions to the problem is that proposed by Cornell University professors Emin Gün Sirer and Ittay Eyal, which adds a computational problem to the mining process. In the current protocol, miners have to solve one problem to get hold of fresh bitcoins, but in the professors’ "Two-Phase Proof-of-Work" mechanism, there would be an additional cryptopuzzle.

Crucially, this second puzzle would require the miner to prove they have access to the private key of the mining pool’s Bitcoin address. This means the pool would either have to enforce levels of trust amongst participants, as any of them could access all the coins collected by the group with that private key, or carry out the work on its own infrastructure, which would be hugely expensive.

“The second phase is architected such that if a miner succeeds in finding a solution that's worth Bitcoin, it can cheat the pool manager and steal the Bitcoin. So either the pool manager trusts the miners to be honest (not likely), or he has to do the second phase himself. This would force large mining pools to make large investments in mining equipment, making it much more difficult to form a huge pool,” Sirer told me.

For miners, this would also make those pools less attractive due to the potential for others to abuse the private key. It might be wiser to join smaller pools as there would be less chance of malicious players, or to set up an independent mining project where private key ownership is limited to one user and the quality of processing power is good enough to crack those tricky cryptopuzzles. And the Two-Phase solution wouldn’t require a massive change in the Bitcoin protocol, “just a few hundred lines of code,” according to Sirer and Eyal.

Advertisement

Such a significant change would, as Sirer and Eyal admit, require all Bitcoin participants to agree on and adopt the updated protocol on the same day and at the same time. Given the fractious nature of Bitcoin support, this would be unlikely. “If most nodes do not agree to the change then the network would also be somewhat in disarray. Suffice it to say that any solution needs to find consensus among the core developers, the miners and the bitcoin nodes, some of which will represent the commercial interests on the network,” Levin told Motherboard.

There are some technical issues to be hammered out too. For instance, small miners might be punished by the proposed system as they won’t be able to afford to do the mining if they don’t want to share their private key.

Whatever happens, the biggest mining pools will need most convincing if Bitcoin is to be patched up. GHash hasn’t been won over by the Two-Phase proposal, having invited Sirer to the meeting. It doesn’t like the idea that inspiring trust among its participants might involve introducing a membership fee.

On the other side, Sirer doesn’t think GHash’s promise goes far enough. They said anyone who has control of more than a third of the network’s mining power can pull off a “selfish mining attack” to trick others into believing they are solving cryptopuzzles that have already been cracked. “GHash’s guarantee is insufficient. In fact, even the 1/3 threshold isn't strict and it may be that much smaller pools can attack the system; we simply don't know,” he added.

Though there kerfuffle over the “51 percent threat” to Bitcoin has been getting louder over recent months, the industry is no closer to eliminating it.