I’m reading The Writer’s Notebook: Craft essays from Tin House. I like this book a lot, even though I don’t like the stories in Tin House much. (They’re good, individually, but they’re all meandering, inconclusive stories about helpless characters that end sad.)
Peter Rock’s chapter, “Telling that shows”, suggests some new ways to look at “Show, don’t tell”:

If you can tell it, tell it.

If all you have to say is that Ruth is upset, say that Ruth is upset. “Show, don’t tell” means that you should have more to say. Your characters’ thoughts and feelings shouldn’t be so simple that they can be summarized in a few words. (If they should, then the story should be told, not shown, like fairy tales.)

Show things that you’re not sure about.

Sometimes we slip into telling because we can’t really imagine our characters doing something, or how they would do it, or how it would happen. That’s a sign that that part of the story is fake, the writer pushing the characters along where they don’t want to go. Telling quickly plasters over that weak spot, but the joints show.

Tell the things that could generate confusion but not drama.

Scene changes can confuse readers, especially if you try to “show, not tell” time, location, and point-of-view shifts. Unless the shift is dramatic, which it usually isn’t, just say “Two weeks before Christmas,” or something like that.