On Friday, November 8, 2019, cryptocurrency exchange, Coinbase revealed its decision to make changes to its existing confirmation requirements for four digital assets, namely, Bitcoin [BTC], Ethereum Classic [ETC], Zcash [ZEC] and Litecoin [LTC].
As per their blogpost, Bitcoin’s current confirmations of 6 will be dropped to 3, followed by ETC’s 5676 to 3527, and ZEC’s 18 to 12 confirmations. Interestingly, the confirmation requirements for LTC has increased from 6 to 12.
Given said changes, one might wonder why they might be necessary to begin with. As per the Coinbase team, the confirmations set in place were done a year ago with no particular approach in mind. The lack of some consistent approach is what led to some assets having an “overly conservative confirmation requirement,” over others.
The reason why the aforementioned three assets saw a decrease in their confirmation requirements rests in Coinbase’s confidence that it was a safe move, adding that the reduction will speed up the acceptance rate for customer deposits. In the case of Litecoin’s increase, it was supposedly done to protect the crypto exchange customers from LTC’s possible 51% attack.
What Really Drove Coinbase to Make Such Changes?
In a separate blog post, the Security Engineer at Coinbase, Mark Nesbitt, shared the underlying reason for making such changes and it rested in their views regarding proof of work security.
Here are the two claims that appear to have supported their decision:
“It is a security feature for a particular coin’s mining operations to be the dominant application of the hardware used to mine that coin.”
Simply put, Nesbitt argued that the proof-of-work coin can reduce the risk of 51% attacks if and only if it becomes the “dominant application of the hardware used to mine the asset.” He compares ASICs to that of CPUs and GPUs. In the case of the former, those who mine can choose to attack the coin, whereas in the case of the latter, just about anyone can attack the coin as long as they own CPU or GPU.
The second claim is:
“Manufacturing and ownership diversity will be improved with ASIC-friendly algorithms.”
In his viewpoint, proof-of-work algorithms can never be ASIC-proof, but only ASIC-resistant. That said, ASIC-resistant algorithms are deemed effective because of its barrier to entry in the mining hardware, which makes it difficult to build an effective ASIC. He went on to state that the only way to build an effective ASIC is if the investment and experience is there – leading to the problem of centralization.
All this being said, Nesbitt notes that the most reasonable approach would be to select an algorithm that is cheap and easy to manufacture an ASIC – resulting in a commodity that requires little expertise and eventually allowing for diversity of manufacturers. The latter is crucial for a network that supports decentralized mining.