Long story short - hype and broken promises. About security. About scalability. About proof of stake. About code is law. About the world computer. About deadlines and timetables. All the way to the very beginning of the project, although the author doesn't go there. Make the thing so damn complicated few can reason about it or see the time bombs hidden behind the marble pillars.
That's a recipe for a high priest class.
Still, the hoards eat it up. It's a testament to the power of complex, flawed ideas, marketed to the hilt, and the promise of riches if you stick with the leadership.
My prior with respect to the ability for ethereum to deliver "the merge" to switch to proof of stake was recently reevaluated dramatically upward, after the recent London hard fork went through without a hitch. I was following it for awhile and there is (in my view) really no way to look at it except as an example of an extremely well run very large update to an enormous and complex system. The communication was good, the cadence of test rollouts was good, the testing setup with multiple testnets was good, the upgrade itself was a total non-event, which is honestly unbelievable to me, as someone who has worked on upgrading complex (but significantly less so) systems.
I still have a lot of questions about ethereum as a project (like: what are blockchains actually good for?), but this software project management stuff is way down the list of concerns for me. I certainly don't understand why people care about arbitrary deadlines and timelines. One advantage of open and permissionlessly developed software like this is not being beholden to a boss breathing down your neck about deadlines and timelines. Just build the software, it will be done when it's done.
That's a recipe for a high priest class.
Still, the hoards eat it up. It's a testament to the power of complex, flawed ideas, marketed to the hilt, and the promise of riches if you stick with the leadership.
It's really a recipe for a technology religion.