"http isn't even in the same network layer so I'd find it strange to even ask about the 'difference'"
Why? You just answered it well enough. There are people who could not say what you just said; you don't want to waste your time on them.
Actually UDP and TCP are the prototypical examples of a reliable versus an unreliable protocol.
Being familiar with protocol design is important no matter what type of programming one is doing, the ideas of ACKs, NACKs, and flow control pop up all over the place!
btree index is also something that can be learned conceptually in about 10 minutes.
It is far more important for me to see that someone can explain a project that they've worked on in the past with enough details to demonstrate sufficient technical knowledge than a litmus test based on somewhat arbitrary facts. 'don't know what lsof is? Not hired, I don't want someone who can't use shell'