He ends with a good list of five key barriers 1. Many problems require new fundamental knowledge based on new physical measurements and experiments.
2. Many problems require situational knowledge to create and maintain complex infrastructure.
3. Moving molecules around
4. Feedback loops
5. Chaos and complexity make many phenomena hard to simulate.
The assumption that most HN readers wouldn't even bother skimming the article is kind of amusing despite how correct it likely is.
If you don't want to wait till they get to the comments here they are
>seize control of the lightcone forever.
I found the article compelling and the conclusions insightful. My assumption was that that sharing them would encourage more people to read it.