To part of it, but chess is generally played one against one, there are well understood rules and a clearly defined goal, and every win is someone else's loss.

When building software, if you can state an unambiguous goal and what rules apply you are more than halfway done. It's not uncommon to work on something for a year and discover you have been building the wrong thing. Navigating that ambiguity is where all the value in software engineering is.

Armstrong did not mean chess players.