Hmm, validators may have other incentives, but I’d be concerned about potential DoS with just a PoW puzzle
I actually had to look that up. The way Nano was explained to me, the main mechanism is batching transactions into separate bins based on the amount that is being transferred. Spam is averted because small transfers have to wait until the bin is full. The PoW puzzle is just the cherry on top I was not even aware of. Nano has been around for 10 years, so they have a track record.
To me, staking one token to get another token which you spend to send transactions is a fee, just a fee in a different token.
The crucial difference is time.
To casual users transactions cost “money”, which they loathe, while Hive’s Resource Credits and Koinos’ Mana replenish over time and are considered “free” (despite the adage time is money). To make a comparison with game economies, there are consumables which are permanently expended upon use, and repeatable abilities with cooldowns. Players tend to hold onto consumables since they do not replenish and this makes them not fun. So spending consumables, or “money”, or tokens, is a negative experience, while abilities are fun.
Of course, abilities being fun to use does not exactly help to lower the amount of transactions by itself, but their cooldown prevents spam attacks, and in addition it makes the user experience better.
I see, this could be helpful from a mechanism design perspective indeed, but it’s still incompatible with shielded transactions (and I worry that users with small transfers would be annoyed at having to wait).
Yes, I can see that there’s a psychological difference. We certainly could have staked NAM generate some “FEE” token which can only be used to pay transaction fees, as you propose here. It works well enough from an anti-DoS perspective. If others think it’s compelling we can investigate this further from a technical perspective.
Seeing your github issue, I got back to this. Are you proposing in fact that also nam native fees are burned? (or is there a conversion step missing in the ordered list?) what happens with the foreign assets held in the treasury?
at the risk of repeating myself, if you are proposing burning all nam native fees that used to go to validators, how does this not make the validator set worse off than before?
sorry if I’m missing something here, I feel I may be.
The native token base fee, yes. Tips would still go to validators.
Nothing yet. In a later upgrade we could elect to add a buy-back-and-burn mechanism as described here, and potentially auto-convert non-native tips for validators as well.
I’m proposing a split between the base fees (which would be burned) and tips (which would go to validators). I suppose that this might lead to a decrease in some fees for a few validators short-term in “direct” terms, but burning NAM just channels the same demand into making NAM more valuable (in terms of other assets), and validators receive NAM as staking rewards, so it probably balances out – fees right now are pretty insignificant anyways.
I can accept the insignificant amount argument (if indeed the amounts are insignificant). However, I think the raw numbers reality suggest that burning tokens will benefit 100% of the remaining holder base, of which validators receive as a collective group around 0.5% p.a. if not mistaken (at a 0.05 commission rate). Thus I would consider it clear that if these amounts were not insignificant, it would leave the validator set as a whole much worse off than if the tokens are burnt (a remote secondary and diluted effect from burn deflation) rather than receiving these base fees directly.
If the fees were significant, I think that your argument holds if we take the validators to be a group totally in isolation, who don’t otherwise hold or stake NAM and just sell whatever fees they receive for some other token (e.g. USD). I’d argue that one of our network incentive design goals, however, is just the opposite: to encourage validators to hold NAM, and to encourage the folks who hold lots of NAM to operate (or support by delegating) validators – and to do so in a way oriented towards the overall success of the network, not just a minimum commission rate or such. The security of the network doesn’t come from fees – it comes from values-aligned validator operators who want the network to be secure, and take action to make it so.
Isn’t the value staked also key for the security? Back when polkadot was launching I had several conversations with Gavin the founder of Polkadot about this topic, and agreed that a key advantage of PoS is that if a malicious actor tries to buy enough tokens to attack the network the token price would increase exponentially making the attack difficult. However, this only holds if there is ample liquidity and a large marketcap. Precisely for this Polkadot needed to have a large marketcap since as the relay chain had to secure the parachains and hence a large marketcap was key. Same for the Cosmos hub with replicated security, stride and other projects always mentioned how they were secured by the Cosmos hub with a large marketcap in the billions. If Namada wants to be taken seriously, a security in the billions and large liquidity is a must. Otherwise Namada can have the best tech in the world but without security it doesn’t matter. Bitcoin is valuable just because of its security, measured by the hashrate. In PoS the security is measured by the value of staked assets. I have been criticized multiples times on the Namada discord when I mentioned about this saying that I’m ‘trying to pump my bags’, let’s try to be a bit less naive and understand the pillars and fundamentals of PoS. Yes, Namada brings MASP, built from scratch but at its core it is a blockchain secured by PoS and hence having enough security is critical, what brings the value to the network and what makes it trustless and a true decentralized blockchain
I think that economic security, understood in the sense of how difficult it is to acquire a large fraction of the token supply, is absolutely an important factor as well – for sure. My sense is that this proposal would be at least directionally helpful from that perspective, but curious what you think as well.
