I often find telling comes about when there is no ground truth or confidence in a field. The author must advise because there's no certainty that if they just showed you what they knew you would arrive in the same place as they.

Mathematics does a lot of telling too, but the best books don't need to as much. They just present things and let you fill in the missing details. It's almost-but-not-quite the confidence of fiction, where a good author shows and doesn't tell because the things they're trying to share are just too large to be simply explained.

I've also found the best CS books don't have to tell. I was recently reading the Reasoned Schemer and I immediately came to love the style they wrote it in. The author and the "reader" are just exploring something together at whatever rate makes sense. It's one of the few CS books I'd want to read over again.