The permissionless nature of the Bitcoin Cash (BCH) network is the diamond on which all the other properties depend (censorship resistance, etc.). If that is lost, "decentralization" and other things become just a show. The network will be captured and typical financial regime rules will be imposed over time, neutering the coin of its unique value proposition and ultimately of its ability to deliver real change.
"Fast", "low cost" and a number of other properties are necessary to have at some level, but they will never be enough to achieve global adoption. Digital fiat will decisively wipe the floor with any coin that goes head to head on these features.
2) Regarding nodes, what needs to be done to ensure they do not become a weak link in the maintenance of "permissionless"?
The industry standard is that a coin has a reference implementation. Pools and exchanges use that implementation. Even though BCH actually has multiple node implementations, it was not enough because there was a de facto coupling between the coin and the reference implementation.
This is what must be broken to ensure that the properties of 1) are preserved. In the specific scenario we are in today, it means that BCH needs to work with industry to show that we have a viable way for multiple nodes to represent a coin through very high interoperability confidence. When this reality is established, it will be a meaningful flavor of decentralization - avoiding the entrenchment of a weak link.
Is that going to be easy? No of course not. We should expect the creation of a global money network to be hard. We may even have to pull off some heroic feats and outside-the-box solutions on the way there.
Are there other potential ways to solves this coin + reference node coupling problem? Probably! I would love to have an easier way forward on this issue if someone can come up with one.
I hope stakeholders will take this issue to heart when they start to think about establishing a similar coupling between the coin and yet another Bitcoin Foundation. We have learned this lesson and there is no need to repeat potentially fatal mistakes. Any attempts to establish weak links need to be resisted and better solutions sought out even if they require more work up front.