> But that is only a blunder if the counterfactual is that China would not have achieved that military aviation as fast. The picture drawn is that the Chinese had a sophisticated and intelligent organizational apparatus that knew to get key players and empower them to create successful organizations.

I think the potential difference here is that the contemporary Chinese government was still pretty new at the time; they only fully took power in 1949, a year before his house arrest started and five years before they were able to have him begin working for them. It isn't that implausible to me that a government less than a decade old might not have a consensus around how to plan longer-term projects, and I don't think that knowing who the right person for a job is can be equated to the knowledge to do the job itself. My reading of the article is an argument that the national security policy of the US was not compatible with the type of planning that would be needed for technological superiority in the long term; they're saying that while they might be aligned for a short time due to the priorities during a time of outright war, in the long term the priorities of the way the US strives for national security change much more due to immediate circumstances.

> Integrating him after imprisonment would surely have been hard. So the counterfactual is that we don't do the prisoner exchange and find a way to hold him indefinitely?

I'm realizing now that I think I might have been conflating this article with another one that seems to have been flagged and the comments moved here (https://news.ycombinator.com/item?id=48207251), which made the argument that the imprisonment was the unsalvageable mistake. (I guess it was LLM generated? I've suspected for a while that I'd probably be bad at recognizing whether content is "real" or not, so I'm not totally surprised that it wasn't obvious to me when I read it before anyone had flagged or commented on it).