Hothouse

Welcome to Continuous Wave, your home for essays, histories, and reflections on modern media from the POV of audio — a project of story editor Julia Barton.

The term “broadcasting,” as you probably know, was borrowed more than a century ago from broadcast seeding, an agricultural method of scattering mixed seeds on a prepared field rather than planting them in the ground.

If you think of words and music as seeds, broadcast really does work as a metaphor. Sounds go out into the ether, some are heard, and a very few take hold and generate meaning in the minds of listeners.

With podcasting, we take packets of seeds and hide them deep within an RSS feed or streaming service, hoping sunlight (audience) will find them.

After a year of publishing this newsletter, I can compare notes between the media ecosystem I know best and newslettering. As I’ll get into below, some of energy around newslettering reminds me of podcasting more than a decade ago. But the two worlds are different in significant ways.

First, let me say that I really love writing this newsletter and reading many others. Most days, my inbox is filled with pieces that are well-reported, original, and/or useful. It’s amazing so many people are doing this work day in and day out, and we should all support the writers that we can. 

I’m a member of Project C, a community of independent newsletter writers, and I’ve learned so much from this crowd. Much of what I’ve figured out is thanks to them.

To extend the agricultural metaphor of broadcasting, publishing a newsletter is more akin to hydroponics, a technique the USDA defines as “growing plants using a water-based nutrient solution rather than soil [but] an aggregate substrate, or growing media.” (media!)

Let me try to explain what I mean.

Continuous Wave has been my attempt to articulate the connections between broadcasting’s past and media’s present, one post at a time. I want more people to read the excellent media scholarship about radio and podcasting. And to experience snark.

My long-term ambition is still to publish a book. The book market, I don’t need to tell you, is really daunting right now. It’s also possible I have acquired delusions after helping to build the audio annexes of many best-selling nonfiction authors. They make publishing look easy. 

But mainly I just needed to force myself to write on the regular. I am someone who works better with deadlines and at least some external reward. This project has worked out for me in that regard.

Podcasting did not enter my life as a public radio reporter and editor until after the financial crisis of 2008-9. Shows got cancelled, and I was among many radio people laid off.

With no other venue (RealAudio, anyone?) some of us started producing podcasts. I made one from my kitchen. Still, it took us radio orphans a while to shed the behaviors we learned under time constraints of the broadcast clock. 

When Roman Mars and I made this episode in 2011, his show 99% Invisible was still a four-minute+ radio module with an extra podcast version — and I was so excited to make that version. Maybe it could be eight whole minutes long. Eight minutes was the outer horizon of anything I’d made.

Our collective release from the broadcast clock evoked a burst of under-employed creativity and enterprise. In 2014, that got turbo-fueled when Apple Podcasts got baked into the iPhone and more consumers could find podcasts easily. 

In a cycle that unconsciously followed the beginnings of American radio decades earlier, no one thought much about how podcasting would be sustained in the long run. There was a ton of magical thinking and “dumb money,” which we gladly accepted.

I sense a similar creative burst is going on with independent newsletters now, and so is the magical thinking. As l navigate this online world, I have to make choices all the time about doing this the easy way or the hard way. It seems I’ve picked the hard way.

My first, and probably most consequential decision when starting out was to refuse to publish on Substack, even though that platform gives you a lot for free and will juice your numbers right away.

Partly I dislike the aesthetics of Substack. I wanted to feature all of the wonderful visual material in the radio archives, and that platform is not great for original design. But also I’d been reading dire economic and social arguments against being on Substack. 

And — this is petty, but also prescient given developments to come — I resented how Substack’s favored House of Journalism, The Free Press, never seems to give producers credit on its podcasts.

So I went with Beehiiv, which is the home of many newsletters by writers and journalists I admire. If we’re being honest, though, I have struggled with it. (I am a newbie to both web design and the email communications world, so the following complaints will surely show my ass.) 

Beehiiv’s site design features are complicated to use, and as a solution, they mainly offer the dreaded video tutorial. Instead I tend to just make mistakes, get frustrated, and call on customer service. I should probably house this publication on a funkier (and cheaper) platform, but the cost and time to learn a new thing keeps me here for now.

Secondly, Beehiiv is very enthusiastic about giving users the ability to monitor subscribers and communicate with them on a granular level through something called “segmentation.” 

If I wanted, I could write different versions of this post, tweaked for every subscriber. I could A/B test subject lines and send times to maximize my open rates. These practices are standard for newsletters that have their shit together, I gather.

But for me, an immigrant from the land of broadcast who just wants to write, having the ability to see who subscribes and who opens my posts, and when? That power is both addictive and disturbing, and largely a waste of time. 

When you open the back-end of Beehiiv, as with many products these days, it offers you a dashboard, like a cockpit. Or the automated controls for a hydroponic garden.

I am trying to learn, and not just resist, newslettering’s nuanced delivery system. But there is one place where I really hit a wall. And that’s the type of writing that this ecosystem seems to encourage.

Newsletters are often scanned in a hurry by people on their phones. I have been known to do this, too. 

So a lot of newsletters offer a certain prose style I call BEAT. Which stands for:

Bossy. 

Earnest. 

Algorithmic and

Tiresome. 

BEAT.

It’s the writing that gets your limited attention. Direct. And it’s earnest. Truly.

It’s the telegraph. We collectively re-invented the telegraph, y’all!

And yet: BEAT often repeats the same thing over and over again. This is partly for the search engines (RIP). Once BEAT makes a point, it goes back around and does it a few more times. Saying the same thing. With many CTAs. Bossy yet earnest Calls to Action.

👉 Like its prose-cousin, “LinkedIn Cringe,” BEAT favors the use of emojis and bullet-points. So it’s easy to scan. Especially on your phone. Oh yeah, I already said that. No matter. You probably missed it the first time. 😈

I have tried BEAT writing. And LinkedIn Cringe. They do seem to get results!

As a podcast and radio editor, I am well aware of the fact that you adjust your communication style to what works best in your given venue.

But also, you know what? This is my newsletter, and that style is not me.

So beat it, BEAT.

Last week we hothouse farmers got an announcement: Beehiiv users can now plug an AI system, Anthropic’s Claude, right into that dashboard via what’s called a Model Context Protocol, or MCP.

I don’t have a subscription to Claude, but my partner (CW’s art director) does. I have been avoiding any AI use — do I need military, drone-targeting technology to write a quirky newsletter about radio history? 

Then again, I’m supposed to be exploring this strange new land. Like many humans, I hate asking for money. Claude wouldn’t care. So I plugged it in and asked for help planning a fund drive.

Claude, needless to say, was thrilled to be of assistance.

I gave it access to my newsletter. Within seconds, Claude was praising the whole thing: my writing, my incredible research, my unmatched reporting, my independence and spunk. Unlike the recriminating Beehiiv dashboard, with its many knobs and dials, Claude knew exactly what I should do.  

It offered a fund drive strategy and then it asked, “Would you like me to draft an appeal letter?” Sure, Claude, why not? It generated a letter, which I set aside.

Here is the worst thing about that encounter: I felt great. Like, for the whole day.

Some time prior to this flirtation with Claude, I’d been having calls with subscribers who volunteered to give me feedback. No surprise, but CW readers are super-talented, original and creative people who all share an interest in audio history. They said so many nice things and had great ideas. 

But only Claude, a non-sentient agent, had me walking on air. Why?

The next day I took a closer look at the appeal letter Claude drafted. The structure of it was OK, actually useful. But the prose was 100% BEAT, just coated with an eerie sheen of Barton.

I know I should have given Claude more specific prompts to avoid that outcome. I know AI could be a publishing buddy who could help me get organized, allowing more time to focus on original writing and research. I can’t really shame anyone else for trying it out, and I think its use will become widespread as newsletter authors run out of steam. The head of Beehiiv expects Claude will become the sole interface for many newsletter publishers on that platform. 

So yes, I now have access to a sophisticated tool to run my hydroponic publishing system, adjusting the nutrients and light and temperatures until every single flower blooms in flawless color. 

But hothouse flowers are grown to be cut. I’ve put Claude back in its box. I’m a broadcaster at heart, so let’s see what spring can do without it.