It requires lots of processing power. It consumes huge amounts of energy. It is becoming less profitable.

Those are key concerns about mining, the process by which bitcoin (BTC) and other cryptocurrencies have been generated since bitcoin’s earliest days. But mining was never the only model envisioned by blockchain developers, and the movement toward alternate approaches such as proof of stake (PoS) and proof of authority (PoA) continues to accelerate.

“I am seriously looking forward to when the cryptocurrency community basically passes away with proof of work,” Ethereum cofounder Vitalik Buterin said at a private, mid-August event.

Each mining alterntive has its own nuances and approaches to reaching consensus.

Even as long-planned efforts to introduce a hybrid model to Ethereum continue, other platforms, including Tron (TRX), are adopting alternatives to mining. Most important for application developers, enterprise providers say consortia models involving known entities don’t need elaborate models to pick the network participants who will validate new blocks of data.

Mining “works great in anonymous, open networks where competition for cryptocurrency promotes security on the network,” Microsoft software engineer Cody Born writes in a post introducing Azure’s proof-of-authority (PoA) model for Ethereum, which was released in August. However, in private/consortium networks, the underlying ether (ETH) has no value.”

The Proof Is in the Pudding

The proof of work (PoW), proof of stake and proof of authority models each offers a different algorithm-based approach to reach consensus on blockchains.

Proof of work rewards participants who devote computational time and power toward solving complicated mathematical problems. The network participant who solves a problem gets to validate the latest block of transactions and add it to the blockchain. As a reward, this participant receives freshly minted cryptocurrency.

With the growth of massive mining operations and the resulting demand for energy, PoW has generated headlines and headaches worldwide.

Conversely, proof-of-stake models base consensus decisions on users who “stake,” or temporarily freeze or assign their tokens in order to vote on validation issues, with their votes proportionate to the number of tokens they’ve staked. The largest test of the model to date will come when Ethereum adopts a hybrid version of the protocol known as Casper, although the timeline for the shift remains unclear.

Known either as proof of authority or proof of assignment, PoA represents a third approach that essentially assigns validation and other responsibilities to specific users or nodes. That’s the approach taken by Microsoft in its implementation of its enterprise Azure framework on Ethereum, which Born says “is more suitable for permission networks where all consensus participants are known and reputable.”

Along with working well in private blockchains in which all users are known, PoA’s limited processing overhead may be particularly well suited for networks including thousands of connected devices driving internet of things (IoT) applications, proponents argue.

Each model has its own nuances and approaches to reaching consensus. Less than a month after launching its mainnet in June, Tron announced it was adopting a variation of the proof of stake model. Called delegated proof of stake, or DPoS, the model eschews selecting validators based on the number of tokens they’re willing to stake in favor of giving all users an opportunity to “vote” for validators by freezing their own coins. Tron’s governing foundation calls it “a more efficient and democratic form of PoS” in a tweet describing the model.

Resistance to Change

The impact of these shifts, when combined with cryptocurrency prices falling from their 2017 highs, has been felt by chipmakers developing graphics processing units (GPUs) used to mine many cryptocurrencies. Nvidia officials, for example, reported a dramatic drop in hardware sales attributed to mining in the second quarter of 2018, adding that they anticipate future revenue from cryptocurrency sales to be “immaterial.” Some of that decline, though, may be due to serious miners moving toward more powerful application-specific integrated circuits (ASICs).

Even with the moves toward other models, it’s not at all clear that mining will disappear completely. As Princeton’s Arvind Narayanan testified before a U.S. Senate committee last week, blockchains like bitcoin, in which the cryptocurrency is the primary focus and not a tool to incentivize user participation, are unlikely to move away from mining models.

“There are many technical challenges,” said Narayanan, an associate professor at Princeton. “Even if those are solved, the question remains as to whether all the existing mining-based cryptocurrencies will switch… my view is this is unlikely.”

Mark Toner
Mark Toner is a Washington, D.C., writer and editor. He has covered business, technology, media, education, and healthcare for a wide range of trade and industry publications.