Probably everyone knows the math; I just think they're exaggerating the relevance of this model to real life.
I'm with the commenter (here) who said that real life is more likely to be an initial-value-dependent or path-dependent PDE. Someone else said having a high-IV person on your team will raise the slopes of everyone else which seems also right: the issues are multidimensional, not affine 1-D.
Ousterhout is taking sides (smarter > more experienced) which is fine; he can make that argument. But using y=mx+b as x→∞ doesn't count as an argument; it's rhetorical flair, not rhetorical substance. The substance of his reasoning seems to be "That's my opinion based on my experience in my past jobs".
Analogies are inexact by definition, which doesn't mean that the speaker is being intentionally deceptive by using one.
I don't think anyone took this beyond the anecdotal evidence provided. We can argue motive and hypotheticals all day but the fact remains that this is just a blurb thrown at fresh undergraduates.
It's a motivational speech (delivered via heresay on Quora) not a thesis.
If you read the full answer, you'll realize that he points out you'll pick the better slope, unless you think you'll be dead before the two lines intersect; so he clearly understands it.
If you're going to pick apart the math, we might as well pick apart that not everything can be modeled well as a linear function.