Proof of Stake - Was Algorand Really the First Implementation?

Proof of stake is hitting the news a lot right now, as ETH 2.0 gets ready for its long-awaited shift to the new consensus model. However, Redditors have been debating proof of stake for other reasons. There seems to be a point of contention emerging over the fact that Algorand is claiming to be the first proof of stake blockchain - a claim that it's unable to back with any convincing evidence. Let's look at the facts. 

Proof of stake emerged in 2012, linked to the Peercoin project. The project founders, Sunny King and Scott Nadal, proposed the idea in the Peercoin white paper and went on to implement it as part of their build. However, that iteration was a hybrid proof of stake, operating alongside a proof of work consensus. 

Soon after this, a few other projects took the idea of proof of stake and ran with it as an alternative to proof of work. Bitcoin's energy consumption and inefficiencies were already becoming apparent, so both Nxt and Blackcoin incorporated proof of stake as their consensus method of choice. Of the two, Nxt was first to launch, putting its mainnet live in late 2013. Blackcoin swiftly followed, releasing its own proof of stake platform in early 2014. 

These days, there are many different variations of proof of stake. Perhaps the most well-documented and debated is Daniel Larimer's delegated proof of stake, used by Steem and EOS, and often widely criticized for being too centralized. 

Algorand - The First Proof of Stake? 

Now, there appears to be a conflict arising between Nxt developer Lior Yaffe and the Algorand team. Yaffe went on to co-found Jelurida, the firm operating both the Nxt and Ardor blockchains. Algorand has been publishing press releases in which it claims to be the first proof of stake blockchain. Yaffe, having worked on the Nxt implementation back in 2013, has accused the project of misleading the community for marketing purposes. 

The confusion made it over to Reddit, where several users engaged in a lively discussion, with a few pointing out that Nxt and Peercoin came along way before Algorand was even dreamed up. 

It's pretty self-evident that Algorand wasn't the first proof of stake given that more than one project came before it. So, what's the game? 

Pure Proof of Stake = Different Proof of Stake? 

Algorand calls its consensus model "pure proof of stake" abbreviated to PPoS. Therefore, it appears to be clutching at the word "pure" to differentiate between its own proof of stake and those that came before. This may simply be a matter of poor word choice. 

Arguably, Larimer did something similar when he created delegated proof of stake. However, taking the word "pure" appears to suggest that Algorand has implemented a proof of stake closer to what King and Nadal proposed in the Peercoin white paper that the versions used by Nxt and subsequent projects. But on closer examination of Algorand's definition of pure proof of stake, that doesn't stack up. 

Algorand's definition involves "true" versions of decentralization, security, and scalability. True decentralization means that influence on block generation is proportional to stake, which holds true for most PoS implementations. The same can be said for true security, which they define as no small subset of tokens being able to endanger the system. 

The statement also appears to link adoption to its achievement of these properties. Although adoption is clearly important for any project, it has absolutely no relevance to the technicalities of an implementation. 

So Algorand's strategy seems to be to take the idea of proof of stake, put its own spin on it, and call it pure proof of stake. There doesn't appear to be any clear reason for doing this. The Algorand team could have called it something else - PoS 2.0, for example - and claimed that like Dan Larimer, they were iterating on the original version of proof of stake. 

Instead, they chose to implement a variation of proof of stake and call it "pure," when nothing that they claim is unique has any significant relationship to what King or Nadal appears to have deemed important when they first proposed the idea.  

Sharing ideas and iterating on them is central to the entire ethos of cryptocurrencies and blockchain. It's how we've come to such a thriving ecosystem today. It's a shame that Algorand has chosen to ignore these principles in a quest to paint itself as a trailblazer.