1/ So many have asked me my comments on @el33th4xor's statement that "Chainlocks is only a “temporary” fix and can’t be “considered safe or sound.”" I reached out to him and had a friendly and productive discussion with him. https://twitter.com/decryptmedia/status/1352024355149111299

2/ I would like to preface this that I consider @el33th4xor one of the foremost experts on blockchain consensus algorithms and really like the @avalancheavax protocol and therefore was surprised and concerned on his comments.

3/ After clarifying with him, his definition of chain locks is different. The chain locks he was referring to was merely on a max reorg depth which we all know and agree is only a stop gap. https://twitter.com/el33th4xor/status/13525308126175477763/

4/ We agreed that utilizing @Dashpay's LLMQ chainlocks does effectively protect the network however his question was then, why use PoW then if it's a consensus that uses masternodes? PoW as a security mechanism is actually not great and its weaknesses are well understood.

5/ PoW adherents rightfully defend that PoW consensus is truly objective as anyone can verify the 'correct' chain without relying on any external information. However is a chain realistically 'correct' if it allows reversals of days, weeks or months of history? Face reality ppl.

6/ However PoW is not just about security. It's also about distribution. How do you decide how coin emissions are allocated? You can ICO, pre-sale or airdrop but all of these have its own set of distribution problems. PoW despite its flaws is one of the fairest ways to distribute

7/ provided that the hardware is easily available. Hence CPU/GPU mining. But being CPU/GPU mineable also makes it easier to attack given hardware can be rented to attack and the loss of the coin doesn't affect the value of the hardware unlike ASICs whose sole purpose is to mine.

8/ The other argument is that masternodes are centralized which depends how you view it. ANYONE with 1000 $FIRO can acquire and host a masternode. There's no 'cap' on the number of masternodes. Its share of the reward pool is fixed and therefore if there are too many masternodes,

9/ its profitability drops. An equilibrium based on supply and demand is achieved. ASICs PoW is also similar but maybe arguably less fair. Miners buy ASICs to compete on a fixed block reward pool. The number of miners depend on supply/demand of coin value vs hashrate.

10/ But you can't always buy ASICs. ASICs also may be sold to you at a markup as manufacturers can choose to self mine too! ASICs also can have different efficiencies or secret r&d that make some players more efficient than others. You also have different electricity costs.

11/ Masternodes on the other hand are completely equal to each other. They can be freely sold at any time. Running costs of masternodes costs around the same among the major VPS providers. With chainlocks, they contribute to consensus. So what's the problem?

12/ Many of the criticisms of masternodes stem from many projects using it for governance. It was perhaps one of the earliest DAOs. There, masternodes vote on allocation of its treasury and on important decisions. This has many challenging issues that are beyond the scope of this

13/ But we at Firo realize this and therefore intentionally do not have masternode governance, making them equal actors in the ecosystem. In short, we believe that PoW + LLMQ chain locks are a great way of combining security and fair distribution while remaining permissionless.

14/ That being said, this doesn't rule out us changing to a different consensus algorithm. While the team prefers to focus on privacy tech, there is no reason to not adopt new ones as needs change. @avalancheavax is definitely a strong contender!

15/ I hope this clarifies @el33th4xor's comments to @decryptmedia :)

16/ Also shoutout to @Dashpay for massively improving the masternode system from the original masternode lists. Their innovations in deterministic masternodes and LLMQ chain locks are a huge step forward.