Interesting. My statement was based on the experience of three different computer science professors (all tenured, all grads from top 10 computer schools worldwide) - two are tenured professors in the Boston area (but not Harvard) the other is a professor at a top school on the West Coast.
All three do very theoretical stuff, but you’re right, all three also have a lot of industry contacts - one has done quite a bit of consulting with a major SVcomputer company, and the other two regularly work with INRIA