One thing I want to make clear from the outset is that every project approaches governance in their own way that is, ideally, custom-tailored to their system and so any attempt to generalize is going to miss the mark when applied to any specific project.
My views are generated from my narrow perspective and are more a reflection of the system we are developing (Koinos) than anyone else's, so please don't interpret anything I say as being an attack on anyone else's implementation. I've been in this space a long time now and just like anyone else who has been around "the block" that means I have a lot of scars and personal experiences that seriously color my analysis.
In our governance discussion I saw the convergence of two trends that I have been observing over the years;
Don't get me wrong, tokens are amazing and can be used to solve an infinite variety of problems, but that doesn't mean they are the right solution to every problem.
Creating complexity, however, is almost never the right solution. Unfortunately, tokenization creates an incentive to "complexify." The whole point of tokenization is to allow for price discovery. If any positive price is discovered then this can create a windfall for the creator of the token. That's why how one launches their token is so important, and why there was no ICO or pre-mine for Koinos.
This relates to governance because it creates an incentive to design a solution instead of UN-designing a solution. When building a system, there are often ways to solve problems not by adding components, but by removing them ("undesigning"). In general, removing components is better than adding them because it creates a simpler system which usually translates into superior efficiency, performance and scalability. More importantly, it makes the system easier to understand which is critical for allowing developers maintain and upgrade the system.
When governance is solved by creating an additional system, that system can be tokenized which can be used to enrich the creators of that system. This therefore means that there is an incentive to add complexity rather than remove it. That doesn't mean that this is never the right solution, just that the first thing people should do when analyzing a given tokenized solution is to examine the incentives of the people creating that token and the nature of the launch to ensure that there are no perverse incentives.
The example I give in this episode of an undesign solution to governance is the modular upgradeability of Koinos. When people are talking about blockchain governance they are almost always talking about how hard forks get approved. But in those discussions they rarely ever talk about the solution being to remove the need to hard fork!
On Koinos, thanks to modular upgradeability, a majority of the fundamental blockchain logic is implemented in smart contracts which means that the community can upgrade these contracts and propagate them through the p2p network thereby modifying blockchain behavior on the fly and avoiding the need for hard forks in the first place.
In other words, the single biggest focus of most governance solutions is a non-issue on Koinos. We solved the problem, not by creating a new system, but by eliminating the cause of the problem in the first place.
Of course, Koinos governance is extremely important, but the beauty of modular upgradeability (aside from the technical benefits) is that it will "free up" blockchain governance to focus on improving the features of the blockchain, allowing Koinos to get better, faster.