The vast majority of podcasts feature talk or narrative-based content. Conventional best-practices for these podcasts holds the following:
• There should be a host. Someone to welcome and orient listeners.
• Guests should be named.
• Production should be clean. No audible edits. Everything nice and smooth.
• Ambiance or sound design in a story should have meaning and the meaning should be clear. No extraneous noise unrelated to the story.
And then there’s 10 Things That Scare Me, a podcast from WNYC. The name kind of gives it away, but each episode features a guest who ticks off a list of things that frighten them. Sometimes the guest just names the thing then moves on. Other times they offer an anecdote to explain the fear. It’s that simple. But the simplicity is also groundbreaking because they toss out convention.
There’s no host. Someone reads credits at the end of the podcast, but no one meets and greets and orients listeners. The person reading the credits isn’t even named. They’re just there. And their tone is the same each episode. They’re the anti-host.
Guests aren’t named until the end. Unless you recognize the voice or caught the guest’s name in the episode title you have no idea know who’s talking — often for several minutes.
Some production elements are rough. Edits of soundbeds are often clearly heard and purposefully glitchy. It’s disorienting. On top of that, some of the soundbeds seem random, although, I noticed, if you listen carefully to the narrative, the sound’s purpose is revealed. But, if you don’t listen carefully, the soundbed is weird wallpaper — that works.
I point to all of these convention-busting practices and then some on this episode of HowSound.