(e.g., how stealth works, how the engines on the SR-71 worked, the details of the Navy's version of GPS)None of these are "ideas", they are implementations/executions of ideas. They are the result of trial and error (even thought-experiments) that ultimately resulted in real world impact and observable changes. Ideas are unexpressed potential. How the Navy's version of GPS works isn't valuable until after the Navy starts using it. Before that, it's just conjecture. But at the point where the Navy deploys GPS, while it is information, it's not solely an idea; it's pragmatic.
Even knowledge of a newfangled GPS system the Navy is working on or going to deploy and start using on a certain date isn't an idea anymore: it's actionable information that can be traded on. An idea is a kind of information, but not all information is an idea. And while thoughts can be information, not all thoughts are ideas.
That being said, I think you are partially right: the concept of an idea not having value is unintuitive, and a lot of other industries are commonly in the habit of presenting ideas as having value. You'll see this when people talk about patents, as if having a patent somehow conveys value on the thing that is patented beyond just the government enforced exclusivity. A patented idea that can not be brought to market in some fashion isn't that valuable (and actually has negative value: not only is it not brought to market; without bringing it to market, you can not reap the benefits of exclusivity; and you've paid for obtaining the right to a patent you are unable to exploit. So it's a net negative).