What’s the longest stretch you’ve gone, away from your love of podcast creation?
For obvious reasons (in case you missed it) I’ve not been doing much in the way of podcasting this year. The last episode I published was March 28th, 2024— so about 8 months now. I’m at a point now, where I’ve enough health that it would be possible to resume . . .
I miss having the conversations, and I even miss doing the uncelebrated work (which we all know so well!) to get them published. But I don’t miss the grind… that treadmill feeling of always having some next thing that could be done.
What’s stopped you in the past? How did you get back on the bicycle? Why did you get back on the bicycle?
There is a huge difference between a job and vocation. A job is what we hold to earn money to meet economic demands. A vocation (from Latin vocatus, calling) is what we are called to do with our life’s energy. It is a requisite part of our individuation to feel that we are productive, and not responding to one’s calling can damage the soul.
The driveway moment: When a report or interview really works, you can tell and we can tell. We can, because the story hits the top of the most emailed list at NPR.org. You can tell, because the story keeps you pinned in your car, in a parking lot, in your driveway, or at the side of the road—as you wait to hear how the story will end. In letters and emails, listeners named these occurrences “driveway moments,” and say they look forward to them, even when it means being late for work or dinner. So that’s your goal: make some driveway moments.
I’m finally heading into NPR’s book, Sound Reporting, and this big of context included by Kernis in the Foreword got me thinking…
What are you doing so that you even know when you’ve put out a “driveway moment?”
It doesn’t matter at all if we feel it’s a driveway moment. It matters if our listeners think so. Are you paying attention to your listeners? Do you have multiple ways for them to connect back to you?
I do think about “driveway moments” when creating episodes. It’s difficult however, given the way that I create my work; They have to simply happen. If one wants to create them, that requires planning, work and editing.
For me, when I encounter a listener (virtually or in real life) the only question I ask them is…
Has any episode grabbed you? …any particular moment or image?
Are we restless and driven to explore, as Sagan says? Will going into space bring humanity together or will we simply bring inequalities and injustices with us? The idea of humans as benevolent explorers sits somewhere between two extremes: those who argue it is our “destiny” to “colonize” other worlds, and those who ask why we’re going into outer space at all.
We’re definitely going. Whether we like it or not, there are enough of us who are unable to not try to wander outside our little cave and over the next hill, sabertooth tigers or no. I think a much better question is: What have we lost, now that many (most?) of us are no longer in touch with the night sky. My answer: A lot. And if we continue and lose our curiosity entirely, everything.
I’ve had the insanely rare privilege of experiencing the real night sky on many occasions. (For one example, once in a very special place, on a moonless night, Mars cast my shadow.) In all my experiences, I still believe I’ve only glimpsed a part of my human heritage. What would we be like if we all were fully in touch with our heritage?
The sudden flashes of insight we have in states of meditative distraction—showering, pulling weeds in the garden, driving home from work—often elude our conscious mind precisely because they require its disengagement. When we’re too actively engaged in conscious thought—exercising our intelligence, so to speak—our creativity and inspiration suffer. “The great Tao fades away.”
I really dislike Open Culture’s web site—modal dialogs, moving thinguses, distracting whatsits… but then, that’s what Reader Mode is for. :) Meanwhile, this was an interesting read just for the nugget of: It’s the distraction, stupid. As I read the bits about the Tao, I realized that—if I had read the Tao—I would not have read into the Tao sufficiently to get this point. (And of course, I’m presuming that Jones’s interpretation—or his reporting thereof, at least—is correct.)
There’s a secret that real writers know that wannabe writers don’t, and the secret is this: It’s not the writing that’s hard. What’s hard is sitting down to write. What keeps us from sitting down is resistance.
People find your show because they are looking for something, or someone, very specific.
People are not just sitting around thinking, “I feel like a need a new podcast to listen to… maybe something that inspires me to move more…” And then they search for “movement inspiration” . . . and then they land on my Movers Mindset show. No that’s not at all how it works. People do not find our show.
People find ONE, SPECIFIC episode. That’s what Kleon did above.
Think of a guest, or a topic, which you did about a year ago…
Now search for that person or topic in your podcast player, or in a web search engine…
Did you find that one episode you were thinking of?
Because people do that. And only then does our show description, show title, show art, episode art, episode notes, and all our hard work gives them the chance to pick us.
[When you are asked to do the score for a film, what makes you say yes?] If it seems like it will be a challenge and fun, then of course I want to do it. Also if what is needed is not something that somebody else can do better than I can. There is a kind of more conventional soundtrack thing and if that’s what they are going for I’ll say, “You know, there are people that do this better than I do. You need to go to them.” But other people want to try something new. They want to try something maybe a little different.
The most important work we do is to make decisions. Decisions don’t seem effortful (turn left or right, say yes or no) but the apparent risk and emotional labor is real. Hard decisions are hard because of the story we tell ourselves about repercussions and responsibility.
I’m not sure when I fully integrated the idea that making decisions is a creative act. But it definitely is a creative act. Making a decision is not simply choosing among options. Making a decision is not simply saying ‘yes’ to something. (And I’m not referring to the obvious corollary that a ‘yes’ to something is a ‘no’ to other things.) Making a decision creates a connection between the before and the after. Those were two things, and through our decision we create a connection; We create something greater than the simple sum of those two “parts.” The connection itself is something wholly new.
It’s important we actually think about this: what you’re unaware of controls you here. We can’t talk about continuous improvement if we don’t ask ourselves, why are we here as a team? What’s our promise to the organisation? Here, I think it’s vital that we embrace the fact that our hurry to get to say number three on a scale of ten is often what blocks us from actually getting to ten.
That’s from a July podcast episode of Ric Lindberg’s Results and Relationships which you can find wherever you normally listen. His is currently the only podcast I subscribe to.
Ric is usually showing up to lead others in the context of professional organizations… but not entirely. There’s plenty in his work that applies to us as individual creatives. Every episode, I find myself thinking: “Right! I already knew that,” and “thanks, Ric, for making me think about this!”
Showing up to lead is enough. You don’t have to break new ground for your work to be helpful.
Right! I already knew that. Thanks, Ric, for making me think about this!
ɕ
PS: About my subscriptions, there are many podcasts whose RSS feeds I follow in my feed reader app (along with hundreds of other things.) My podcast player is quiet; No new episodes appear creating that fear-of-missing-out. Instead, only when I open my feed reader app, do I see all the new episodes from the many podcast shows I follow. And just like everything else, I simply skim through, and I can add a podcast episode if I wish. This is an example of calm technology.
Instead of letting your head dangle forward when you’ve logged on, put some strength in your swipe and use a little muscle in your upper back to hold your head and spine up.
Over in the Movers Mindset project, Bowman is someone who has long been on my to-talk-to list. Some day!
Until then, you’ll just have to read everything she writes. It’s terrific. Large amounts of actionable stuff around bare feet. This article is about what you are doing to yourself through your habitual phone use—uh, it’s horrific. But then, maybe if you were empowered with some knowledge, then you’d change? (I know that worked for me!)
Your intellectual appetites might include knowing the answer to a mathematics problem; the satisfaction of receiving a text from someone you have a crush on; or getting a coveted job offer. These things won’t necessarily cause physical pleasure. They might spill over into physical enjoyment, but they are not dependent on it. Rather, the pleasure is primarily intellectual.
[…] But, for most people, such joy is fleeting. There is always something else to strive for – and this keeps most of us in a constant, sometimes painful, state of never-satisfied striving. And that striving for something that we do not yet possess is called desire. Desire doesn’t bring us joy because it is, by definition, always for something we feel we lack. Understanding the mechanism by which desires take shape, though, can help us avoid living our lives in an endless merry-go-round of desire.
Although I won’t be sharing specifics, I have stellar care and support, from my family, and from a huge team of the best healthcare professionals. My prognosis is excellent. If one must get cancer, you want to have the experience I’m having.
You may have noticed that I’ve not published a podcast episode since something like May. That’s when I started working through my diagnosis, and that’s when I intentionally pressed the pause-button on some of my current projects. I’ve been a guest on a couple of podcasts this year, and that has kept alive a warm ember of my passion for this wonderful art-form.
This is perhaps the greatest conundrum of our current technological era: the desperate need to connect with one another, because it is our only hope of survival; combined with the fact that nearly all the means of connection available to us are deeply—possibly irredeemably—fucked. Syndication, as I am currently experimenting with it, is then an effort to try and navigate that terrain, to find some productive way to play in the outskirts, to let the work out into the world while (hopefully) minimizing the misery that is reflected back.
Yes, to everything from Brown (and not just this particular piece.) Beautiful thoughts therein around why one should “publish own site, syndicate elsewhere (POSSE)”—my methodology since the beginning.
Unfortunately, the Internet went from “publishing your own stuff is difficult”, straight to “it’s easy to publish on platforms other people control.” To this day, it is still quite difficult to get your own domain name and begin publishing in a way that you control your own content. Worse, we went from people discovered and read your stuff (back in the “publishing your own stuff is difficult” era) to the now where no one can find or read your stuff regardless where you publish it (unless you pay money to the platform brunch-lords.)
Fortunately, if you have a little bit of time and a little bit of curiosity, you can still find everything that people are publishing.
Or another way of putting it, machines are very good at solving PROBLEMS, but not very good for solving THE PROBLEM.
THE PROBLEM of being alive. The problem of being conscious in the universe. That is what art is for. That is what connection is for. That is what leadership is for.
The future belongs to the people who can actually tell the difference.
That’s really the point. I keep saying we don’t yet have artificial intelligence. The things we have definitely lack agency—that’s a requisite. We already know of non-human intelligences such as other mammals and cephalopods. We’re just spoiled because they’re inferior in at least one dimension (such as “that intelligence lacks our level of language”). But one day, sooner rather than never, there will exist non-human, human-level-and-beyond, intelligence. We won’t call that AI. We’ll call it whatever it prefers; He, She, Them, Finkelstein, Bob… whatever. And Bob will be an artist, or a writer, or a philosopher— or all those things and more. And it will have curiosity and questions, just like you and I.
With great power, comes great responsibility. Large language models (LLM) like Chat-GPT are powerful tools. How do I use it responsibly?
I want to find and present great quotations from guests on my podcast episodes. What happens when I try to get Chat-GPT to do it? Following is a really deep dive into exactly what happens, along with my best efforts to work with this power tool in a way which accurately represents what my guests say, while showing them in the best possible lighting.
The transcripts the LLM works from
For this to make any sense, you need to know that I start from a machine-generated transcript. I get them from the recording service, or from another service where I upload audio (for older episodes before machine-generated-from-the-service was available.)
Below is a screenshot.
In the left-margin are line numbers. Line 107 and 123 are too long to fit, so my text editor has visually, “soft” wrapped them for ease of reading.
The files have time codes in them (the format does vary somewhat too). They have silly amounts of precision: 16:51.57 is 16 minutes and 51.57 seconds. Hours appear in the front as another number with another colon.
The LLM understands which person is the guest, because it understands who introduces the show, and introduces the other person.
I break the very long transcript text file into chunks, because there’s a maximum amount of text you can paste into the LLM interface. The screenshot above is from the 2nd or three chunks from my conversation with Martin. The whole chunk is 144 lines and about 8,000 characters.
Imagine having to read through the whole transcript to find the best part to quote. That’s very hard for me to do (nevermind I don’t have the time to do it), but the LLM can do it in a blink. LLMs are tireless and patient.
The prompt
There’s much discussion about “prompt engineering.” It’s an art. The best clues I can give you are: Explain it to a 5-year-old. And, the 5-year-old does not get confused by ordered lists, even if you nest them.
I give the LLM all the transcript chunks. Then I give it this prompt:
Select 5 direct quotations (not from Craig) from the conversation. I prefer longer quotations which include more context. For each quotation you select, do three things: First, show me your selected quotation. Do not rewrite the quotations. You may remove verbal tics such as “ah”, “yeah” and “um”. You must leave the speaker’s false-starts and restarts in place, ending those with an em-dash and a space. Second, show the nearest time from before the selected quotation. Show that time exactly as it appears in the original transcript. Third, show the exact original transcript from which you selected the quotation. For context, show several lines of the original, unedited transcript before and where you selected the quotation.
Oh! Reviewing this post, I even found a problem in the prompt above. Can you see it? Below, you’ll see a complaint about the LLM response. Now I think it’s not an LLM error, but an error in my prompt above. :slight_smile:
(This entire post about quotations is actually just part 2 of a much larger prompt which starts with, “Perform the following 5 tasks. Include a numbered heading before your response for each of these tasks:”)
It spits the result out in one long stream of text. I’ll break it apart…
Quote 1
There are many things that I have to consider as I look at that:
Is it really a good quote, based on what I remember of that conversation? Meh, it’s okay. But that’s why I as it for 5 selections.
I don’t love that it refuses to give me context after the part where it selected the quote. I’ve tried, but after hours of work, I’m done prompt engineering and want to start this post. (As I mentioned above, I think this is because there’s an error, above, in my prompt.)
In this particular conversation, Martin talks a lot about “parkour”—thus “park order” (and many other variations I see a lot)—is just an error in the raw transcript. Ignore that for today.
Most importantly: Is it hallucinating?? Well, it’s easy to use that timestamp. 1 minute 22 seconds is definitely in the 1st chunk… a moment of scrolling…
Here’s the actual, original chunk I uploaded, and the LLM’s output side by side…
Okay, that’s sane. If I was going to pick this quotation, I’d have to work some form of my question into that quote, so his quote has some context…
Moving on, I’m just going to give you the screenshots for each.
Quote 2
Quote 3
Quote 4
Quote 5
Closing thoughts
I use LLMs to write my show notes. Getting a quote or two is just one part of that.
“Write show notes” is not “the work only I can do” (as Seth would say.) And, I simply do not have the time to do show notes from scratch.
Yes, I’ve spent hours today on prompt engineering, but I have 319 more podcast episodes from 2022 and earlier (!) that I want to have show notes for. Those episodes would be better with show notes. A few hours spent here, enable me to copy-and-paste… wait a few minutes (the LLM is not instantaneous) and I have a really good starting point for show notes.
