I agree, Bob is great. I think scaling a concept is also really really important. We need features but we need scalable frameworks and platforms to support these features. Bob's not going to write the distributed db system that makes his feature look good, but that also needs engineering. That too needs to be good enough to ship and no more btw, but it's more often the work of a team that has to slow down and think carefully about what it's doing, especially when evolving long-running platforms

Yes. I wished we recognized more often that hyperproductivity exists for these engineers with different strengths too. It just looks very different than Bob's twenty feature launches per quarter

This is something of a circular argument. Bob doesn't waste energy on on-call duties, so he implements more features, so he's a prolific engineer, so he doesn't have to waste energy on on-call duties.

Depends on what you compete on. If you compete on features and oncall is just some contractual obligation, then probably not, if you compete on customer service then perhaps yes