This is a very useful summary, thanks @RiskSC!
I’ll do my best to respond point-by-point.
- It’s important to first understand our line of thinking, which we reached to through countless of discussions with people internal and mostly external to the project. Essentially, as outlined in the blog post, we were looking for a number of nodes that is unquestionably decentralized, especially compared to other networks in the space, but that is still relatively manageable if swift decisions (such as an urgent hard-fork) need to be made.
The exact figure we reached in these discussions - 50, is already substantially larger (and therefore more decentralized) than many well-known networks out there (e.g., NEO, EOS, Loom, and many others…).
As to favoring large stakes - that’s (quadratically) discounted compared to commitment and participation, which feels like a reasonable balance. Additionally, it’s important to remember that proof-of-stake was invented with the premise in mind that higher stakers, having a larger stake, are less likely to attack the network. So from a network health perspective, ignoring that metric completely would be damaging to the network.
Finally, there’s a reason we call these Genesis Nodes - as the network grows and improves, we’d like to see tens of thousands of nodes participating.
There are ways to counteract that, and we’ve spend some time on it. For example, if selected, you’ll need to stake the same amount of tokens through your mainnet wallet. We’re open to getting more feedback and ideas here.
We’re open to feedback here as well. A quadratic relationship seems good to us, but we’re happy to consider others. The best way to do this is to back this up with simulation data.
I’m referring back to my explanation in #1. Once we all establish that the health of the network is the most important factor, then the logic of starting with 50 nodes makes sense. From here, it was a matter of finding the fairest way to select those 50 nodes, in a way that we don’t control (hence the formula and lottery) and can’t add bias to the results. It may very well be that 25K tokens will be enough for day 1 staking, provided you support testnet enough. Even if it isn’t, these tokens would be sufficient post hard-fork, so it’s not like that usefulness is lost.
Baseless. We gathered feedback from people who expressed interest running nodes, alongside industry experts that may not even hold ENG. There are much easier ways to favor large holders (for example, we could have simply set the formula to be based on # tokens and that’s it - or would have selected the list entirely ourselves).
That’s true for every open source project. I have yet to see one example where this was an issue. On the other hand, showing our work and hopefully - getting more outside contributors are two extremely important goals for the project, which led to this decision.
What matters is quality. Once people realize how much work has been put in the 6+ months (easy to ascertain through the code), I’m sure these concerns would be alleviated. I’m so extremely proud of what we’ve been building as a team.