In this final episode, I reflect on things I learned by producing the show.

Photo by Diego Jimenez on Unsplash.

Show notes

Some show notes may include Amazon affiliate links. We get a small commission for purchases made through these links.

If you're enjoying the show, please rate or review us in Apple's podcast directory.

This episode's transcript was produced by an AI. If you notice any errors, please get in touch.

Transcript

Jorge: All right. So here we are. Last episode of the show. As I announced in my previous conversation with Peter Morville, I have decided to end The Informed Life podcast. I’m not going to rehash the reasons why. I went over that in my conversation with Peter. But I did think it’s a good idea to articulate some of the things that I learned by producing the show.

I’m going to take this opportunity to reflect on the things I learned, and I’m going to do it at two levels. The first level, I’m going to talk about what I learned about producing a podcast. So this is all about the mechanics of doing this, the kinds of tools I’ve been using, the things that have worked, haven’t worked, et cetera.

And then, the second level is about the subject of the show itself. So, this notion of organizing information to get things done. Obviously, it applies to a lot of different things in life. And I’m hoping to reflect a little bit on that as well. Although I expect that the bulk of this conversation is going to be about producing a podcast. Let’s get into that first part now.

Producing a Podcast

Jorge: Let’s start by recapping why I chose to do a podcast. There are lots of different media that one can use these days, and I specifically chose to cover this subject as a podcast. As I mentioned in the conversation with Peter, I did this because of two reasons.

The first is that I love podcasts, so it’s a medium that I myself find valuable. I listen to podcasts while exercising or when I have long drives. And I’ve learned a lot over time with this medium.

So that’s the first reason: I really like podcasting myself, and I thought to try my hand at it.

The second reason is that I thought podcasting would be easier than producing video. Even though I learn a lot through YouTube as well, I came into this experiment under the impression that producing a recurring audio-only show would be easier than editing and producing a video show. And I still believe that’s the case. I have done some video work, for example, with my IA: WTF? course, and producing an audio show is definitely easier.

But it’s not as easy as I thought it would be. So when I got into it, I thought it would be really easy to do this. And I vastly underestimated the amount of time that it would take. The very first conversation that I recorded for the show was an interview with my friend, Lou Rosenfeld. And that conversation, I think, was around 45 minutes long or maybe an hour long. I remember starting to edit that and being at it still after something like five and a half hours. And I started to panic. Because I thought, “My gosh, if this is what it’s going to take to do this, I’m not going to be able to produce this every other week, because if it’s going to take five and a half to six hours to do these things, each one of them, then I’m just not going to have the time.”

Fortunately, shortly after I released that first manually edited episode, a friend of mine recommended that I look into what was then a brand new tool called Descript. I think at the time I discovered it was still in beta, and I applied for the beta and got in. I started editing the podcast—I think that I edited the second interview using Descript or maybe the third. It was very early on in the run of the show.

And I can honestly say that that’s what made it possible for me to continue doing this, because Descript just made it so much easier for me to edit episodes. If you’re not aware of this tool, essentially it transcribes audio and then lets you edit the audio file by editing the transcript so that you can, for example, select an entire sentence in an interface that looks a lot like Google Docs or Microsoft Word. You select an entire sentence or a paragraph and delete it from the transcript, and it deletes it from the underlying audio. Or obviously, you can cut and paste or shorten it, or what have you. I’m a very visual person, and being able to edit these conversations visually made all the difference. So I went from taking six hours approximately for that first interview to around three hours for regular conversations. And that was much more feasible. If I could spend three hours or so every other week, then that was something that I could do. I could not recommend Descript more highly. Since I started using it, they’ve branched out and now also allow you to edit video. I have to say, this is not a sponsored read in any way. They don’t know that I’m saying this. But I’m really thankful to those folks because, without that tool, I just would not have done it.

Now you might be wondering, “Jorge, why didn’t you just hire someone,” right? That’s what most people do. I don’t know if most people do that, but I’m guessing that a lot of folks who have a long-running show will go off and hire someone to edit the podcast. And I could have done that except for a couple of things. The first is I was trying to do this very cheaply, in a very kind of shoestring manner. As I described in the conversation with Peter, the genesis of this show was as research for a book that I was planning to write, and which came out in early 2024, what ended up coming out as Duly Noted. I did not start it as some kind of explicitly commercial venture. So I was being very mindful about how much money I was going to be spending on it. Now that said, I did at one point hire someone to help me with the editing of the show. Because even with Descript, it became clear that I was not going to be able to keep the editing pace and also do my consulting work and my teaching work. So I reached out to an agency that matches virtual assistants with folks who need that kind of help. They’re called Freedom Makers, and I’ll include a link to the service in the show notes for this episode. Again, I could not speak more highly of those folks. They connected me with Sarah Clarkson, who helped me edit the show for a period of time.

And the way that it would work is: I would record the conversations using Zoom. And by the way, I tried several different means for recording episodes. I know that there are several tools out there to do this kind of thing. But I always ended up gravitating back to Zoom for a couple of reasons. The first is that Zoom is ubiquitous. A lot of folks at this point out in the work world have used Zoom at one point or another. So it’s a familiar tool. And the other thing is, frankly, the couple of times that I tried using other tools, it did not go well for me. I either lost audio, or I had crashes, or they were not as reliable.

So Zoom, for all its faults, has always been fairly reliable to me at the expense, maybe, of a little bit of audio quality. And the trick in using Zoom was when you’re recording calls in Zoom, there are two ways that you can do it: Zoom lets you record in the cloud or locally. And I don’t know if they’ve changed this since then, but a few years ago I experimented with both, and there was a big difference between them in that if I recorded in the cloud, then I would get back one audio file that included all of the people in the call mixed in the same audio file. Whereas if I recorded locally, I would get separate recordings for each participant. And that was essential because when you’re editing interview-style shows or conversation-style shows, you want to have access to separate tracks for each speaker.

So I would record these conversations locally. And then I would import them into Descript, where they would be turned into a transcript. I would then notify Sarah that the transcript was up, and Sarah would then take a first pass at editing it, taking out obvious mistakes, some of the verbal hiccups, the oohs and aahs, that sort of thing. And then when she was ready, she would notify me, and I would do a final pass before producing the final version.

The other thing that Sarah did was correct the transcripts because, for some reason, Descript’s AI introduces a lot of weird transcription errors. I’ve done AI transcription conversion with other tools that are much better than Descript’s transcripts. So I had Sarah go and do a first pass at that.

Now, eventually, I stopped working with Sarah, not because of anything that she did, but because a) I had a bit of a low season in my consulting work where I had more free time on my hands. And I was like, “I have free time. I can do some of the editing myself.” And then, when I did that, I realized that Descript had integrated a bunch of new AI-powered features that did a lot of the editing in the tool. So a lot of the stuff that I was asking Sarah to do, now the tool did on its own.

So that’s a little bit of how the recording and editing happened. Now, let me tell you about the publishing part of this because that’s something else that folks ask about. There are many ways to publish a podcast. Podcasts are ultimately basically websites that serve audio files with an RSS feed. But beyond those basics, there are some nuances that you do want help with. And there are dedicated podcast hosting services. The one that I started with was Libsyn.

I always found it really complex. I never fully understood how Libsyn was structured, and it is a very complex tool, but it worked fairly well for me for a time. The main reason that I switched away from it is that at one point, Descript released a new feature where it did one-step publishing to different podcasting providers, and Libsyn unfortunately was not one of the providers that they worked with.

I looked through the list of providers that they did support, and I chose Buzzsprout. I don’t know that there’s anything extraordinary about it. There are several of these things. I did look at the websites of the ones that were supported by Descript at the time, and Buzzsprout seemed to be the one that had the best combination of features, price, and ease of use, which, after my experience with Libsyn, I valued. And I have to say I’ve been very happy with Buzzsprout and the Descript integration.

So that covers recording, editing, publishing. Those are the main things that you need when doing a podcast, but there’s a few other things that go into it. One, I have a website for the podcast. You can do this through Buzzsprout or Libsyn or any of the other providers. They give you the ability to host pages on the web that you can point people to. I chose to have a separate website for the podcast mostly because I like for episodes to have canonical URLs, and I like those URLs to be on a domain that I control and in a format that I control.

The other aspect of running the show is choosing what conversations to feature and reaching out to folks. And that ended up being a big part of what took up time for me. So deciding on the editorial schedule for the show was always complicated. I always tried to have at a minimum two, but ideally three, episodes already in the can whenever I released a new episode so that I had a little bit of a buffer, right? But even so, there were times when it really came down to the wire, where I was staring down at a coming weekend with no episode yet recorded. Maybe a conversation had to be rescheduled, or someone couldn’t show up or… When you’re depending on other folks’ schedules, it’s always really challenging to do this kind of thing.

Or sometimes it was because my own schedule fell apart. Like, sometimes I was really busy and I couldn’t make time to do the recording. And I did manage to release a new episode every other week consistently. I never missed a publishing slot. And it was very important for me to establish that cadence. That was one of the things that I wanted to do with this podcast. And I accomplished it, but it did happen at the expense of a lot of stress sometimes. And reaching out to folks and finding a balance of subjects that I thought would be interesting given the show’s focus, and novel, and frankly that I was excited about, that proved a challenge.

And I am fortunate to have a fairly large network of folks in the design space that I could reach out to and interview. And I have to say most people that I reached out to were super gracious, and the vast majority of them agreed to be on the show, and they are on the show. But the whole logistics of trying to think about who could be on, what’s the angle here, what’s the knowledge organization angle here, that was always stressful and always challenging to me. And frankly, it’s one of the reasons why I’m going to a different format with a recurring co-host.

That makes scheduling and organizing the editorial schedule much easier. All right, so that feels to me like a good segue to talk about the show’s subject.

The Show’s Subject

Jorge: So, what was this show about? This is a surprisingly tricky question. As I mentioned in the conversation with Peter, it’s actually one of the reasons why I’ve decided to end the show. Even though I started with what I thought was a pretty clear picture, it soon became muddied. The show’s slogan—this idea that the show is about how people organize information to get things done—was my attempt at building an umbrella wide enough to cover things that might not fit easily together. That might be the biggest learning for me from producing the show.

Like I said in the conversation with Peter, and as I mentioned earlier, I started this show as a way to do research for what became my book, Duly Noted. The idea was to apply towards personal knowledge management many of the tools, techniques, practices, and models that I’d learned in my work in information architecture.

In information architecture, we organize information so that people can find and understand it easily. This is something we do for ourselves as well. I had this hunch that there could be an interesting book in thinking about how the discipline of IA could be applied to this kind of work.

I wanted to write a book about it, but I didn’t want it to be just about how I organize information. I wanted this book to feature lots of other voices. I thought a podcast could be a good way to do that because it would give me a reason to reach out to folks to interview them, get their perspective, and bring diversity to the book. Also, it could serve as a way of marketing the book itself, to build interest in this idea, right?

And as I mentioned in the conversation with Peter, I had something like five or six conversations in the first few episodes of the show, where I did ask people about what Tyler Cowen calls their production function: how they went about doing things, writing books, managing their company, or what have you.

What ended up happening is many of them mentioned the same tools and techniques. They talked about managing to-do lists and using sticky notes. Keep in mind that at that point, I hadn’t yet landed on the focus that I ended up writing the book about. My sense was that if people kept coming on the show and saying the same kinds of things, this was not going to be a very useful or interesting show for folks. At that point, I decided to broaden the scope a little bit and do what I think a lot of folks were expecting me to do, which is to talk about information architecture. That’s been my focus for the bulk of my career and my writing, and I steered the show in that direction.

Even more so, particularly in the last year of the show, after the book had come out, I started hosting more conversations about UX design more generally. It’s been a very tumultuous couple of years for UX design. I just wanted to have conversations that were interesting to me, but also interesting to folks who comprise my usual audience, the people who are interested in information architecture.

At that point, I feel like the show had started steering away from its original intent. The second half of the year, maybe, we were not covering what I would consider to be conversations about information organization; we were talking a lot about user experience design more broadly.

The first half of the year, and especially in 2023 and 2022, ironically, we were hearkening back to the original intent of the show, which was more about personal knowledge management. Because at that point, I had landed on the subject for the book, and I was doing very active research for it. I hosted what I think were really interesting and valuable conversations with folks who are managing their personal information in novel or interesting ways.

The one that immediately comes to mind is Jerry Michalski with The Brain—and I’ll include links to these episodes in the show notes—or with folks who were focused on particular knowledge management tools. In some cases, the designers of those tools. The foremost episode there in my mind is the interview I did with Mark Bernstein, the designer and publisher of Tinderbox.

Interestingly, some of those conversations are among the most downloaded conversations in the podcast. I got the sense there was a lot of energy about them. But what happened is, first of all, I wrapped writing on the book. Then, when the book came out, it was right around the time when the language models, Claude and ChatGPT, were drawing a lot of attention. I think a lot of the energy that was present in the space in the last couple of years, around the release of tools like Roam Research and Obsidian, moved to the AI side.

As did my attention, frankly. I’ve been doing a lot of work in that space myself. So I can see why this would be. If you are somebody who is interested in how people work with information, you are going to be interested in language models and machine learning, just because it’s very clear that these are incredibly powerful tools.

Anyway, I’m not a good judge of the subject, utility, or value of the conversations in the show, because I honestly loved all the conversations I had. It was a tremendous opportunity for me to reconnect, in some cases, with friends, and in other cases, to reach out to people whom I had long admired from afar, whose work I followed, or whose tools I used. This gave me a perfect excuse to reach out to these folks and spend an hour with them. In every one of those cases, I got tremendous value out of the conversation and then shared it with you.

So what I thought would be fun to do, given the work I’ve been doing with AI, is to have AI draw up an outline of sorts of what the show was about.

I’ve done this a couple of times during the run of the show. At the end of 2023, I released an episode where I used AI to do a year-end review. Before that, I think two years before, I had done another one of these year-end review episodes, but I had done it manually, which was a lot of work.

In 2023, the AI helped me do it much more quickly, but it was still a primitive technique that I used for that. For this conversation, I built a Graph RAG instance. I had an LLM basically build a knowledge graph from all of the transcripts of all episodes of the show. Then I could ask it questions.

The main question I wanted to ask in preparation for this epilogue was basically: What was the show about? What were the themes or topics that came up over and over again? The AI came up with seven themes. Some of them, I think, are obvious and I would have come up with these myself. A few of them surprised me, and I’ll just tell you what they are.

Seven Themes of the Show

Jorge: So the first theme is one of the not surprising ones; it’s user experience design. I think this is self-evident. It’s not in any way surprising because so many of the guests that I invited on the show are my colleagues in the UX space. For me, the conversation that epitomizes this theme was the conversation I had recently with Joe Natoli about The User Experience Team of One, the second edition of that book that Joe collaborated on with Leah Buley.

I’d like to think that the conversations and the show that centered on UX focused primarily on information architecture, although that is not universally true. It’s certainly not true of the conversation with Joe. So that was the first theme that the AI highlighted.

The second theme did surprise me. It said that the show is in part about collaboration and community engagement. And that’s curious, but I can see why it would say that because I hosted several conversations in the show that were about organizing information but at the scale of communities. And for me, the episode that epitomizes this was a conversation I had a couple of years ago with Brian Breslin about the work that he was doing in South Florida at the time, organizing the community there.

And the reason why I hosted conversations about this is that it is such an important part of what we do online these days, or at least some of us do online. So we can organize information for ourselves, which is what we do when we’re doing something like organizing our notes. This is the subject of Duly Noted. Or we can do information architecture, which is in the more traditional sense about designing systems that are meant to be used by others through structuring the navigation schemes for an online store or defining taxonomies for a repository of documents or something like that.

But there’s this middle ground where you’re trying to organize a shared information space that is going to be used by other people. And this is something that we do, for example, at work, if you’ve ever had to set up something like a Slack instance or a Discord server. And I thought that would be interesting territory to explore. And that’s why I reached out to people like Brian to talk about how they did this.

So another topic that the AI highlighted was effective communication. And you can’t see me, but I’m smiling when I say that because, again, this kind of surprised me. I think it’s a very broad subject, but I can see why the AI would say this, because a few of the conversations that I hosted on the show were about people who were trying to do things like write books or write a thesis in the case of Beck Tench. These are all conversations that hinge on organizing information to communicate effectively.

And the other reason why I think that this is an appropriate theme or topic to highlight is because even in the design space, even if we were just focusing on the conversations that centered on user experience design, communicating intent is such a central part of that work. So I can see why communication would come up as one of the main topics of the show.

Another topic that the AI highlighted with no irony is the integration of technology. Or, to put it another way, the use of technology in organizing information. And again, this is not something that surprised me. There are many episodes on the show that focus on the use of technologies for effective information management, particularly AI. And there were several of these that focused on personal knowledge management. And immediately, the conversations with Alice Albrecht and Lorenzo Bernaschina come to mind. So that was definitely an appropriate topic to highlight on the show.

The next one is ethical considerations in design. And again, this did not surprise me. There were several conversations that focused exclusively on this. Two that come to mind are my conversation with Sheryl Cababa on her book about the role of systems in social structures and the conversation that I had with Chris Chandler on ethics. Again, it’s a topic that a couple of episodes tackled directly. But it was also a topic that surfaced in several other conversations, particularly the ones around experience design.

The next topic, and this is the second to last, the AI labeled as humanitarian efforts and innovation, which frankly surprised me. I don’t think that this was a major focus of this show. We did have one conversation that I would say fits into this description pretty neatly, which was my conversation with Jason Ulaszek about the Rwandan genocide.

That was a conversation that I would say was explicitly about a humanitarian effort. But I can’t think of many more in the show that fit this description. So I’m going to chalk this one up as a hallucination of the AI. I did want to include it here as part of the record, right? To say, as amazing as things like RAG based on knowledge graphs can be in helping you make sense of a large corpus of content, such as the transcripts for five years’ worth of podcasts, it still is not perfect.

The last topic that the AI highlighted, though, I think is one that is obvious. It’s knowledge management and personal development. And this is, like I said, the reason why I started this show. So I was not in any way surprised that this came up. There are several conversations in this show that are explicitly about this. And one of them I’ve already mentioned, my conversation with Mark Bernstein. Another one is the conversation that I had with Kourosh Dini. And I don’t think it’s surprising that those are among the most popular episodes of this podcast. I think they are the ones that perhaps hew most closely to the original intent of the show.

And I think that these topics that the AI suggested highlight that. I have to say also, I was surprised that there is what I think is an important topic that was a recurring theme throughout the run of the show and which the AI did not mention, which is systems thinking. There were several conversations in this show that were explicitly about systems thinking. A few that come to mind are my conversation with Jeff Sussna about cybernetics, the conversation with Sheryl Cababa that I’ve already mentioned, and especially my interview with Hugh Dubberly, which went pretty deep into systems thinking and modeling. So I was really surprised that did not come up. And again, I think it points to some of the limitations of language models when doing this kind of work.

Closing

Jorge: All right. I’ve been going long enough. It’s time to bring it to a close here. So, to recap, I’ve learned a couple of things. One is, I think I’ve built certain chops for producing a show. I could’ve been more effective at a few things. One I’ve already mentioned is focus. I think that focus is super important. And I would say if there’s been anything that I really have not done well with The Informed Life, it was having a clear focus. The idea that I could do a show that encompassed both personal knowledge management and information architecture, I think, did not pan out. There was a point where the show was growing, and then it hit a plateau. In retrospect, I attribute that to the fact that the show did not have a clear focus. This notion that it was a show about how people organize information to get things done is just too broad. I think it’s too abstract. I don’t think it’s a meaningful phrase to a lot of folks. It’s certainly not as clear as, “This is a show about how you can use AI to become more productive,” or “How you can manage better task lists,” or “How you can remember everything using Obsidian,” right?

Like, those are all positionings that might be a lot clearer and perhaps evoke a clearer audience in someone’s mind. The notion of organizing information to get things done is just too broad, right? So, that’s the first lesson. I think that having this broad remit made it interesting for me because I like high-level thinking and exploring things broadly. But that made it difficult to market the show.

So, that’s the other thing: I like to think that I could have done a better job of marketing. I guess I was under the impression that producing a show would be a… What’s that line from Field of Dreams? “If you build it, they will come.” So, if I produce a show about information management, then people will show up. Sure, they will, but it’s going to be a relatively small number of people. And clear positioning helps with marketing. So, that’s one thing that I will definitely try to do differently the next time around.

And that’s a good segue to repeat something that I have already mentioned on this show and that I discussed in greater length in the conversation with Peter: that I am working on a new podcast, and I will be announcing that when it’s ready. It’s going to be early in 2025. So, not quite ready to talk about it yet. But do stay tuned because I’m hoping to take the things that I learned from producing this show and apply them there.

I do plan to leave the recordings for The Informed Life up on the internet. I’m proud of the work that we did. I think that the conversations are, for the most part, evergreen. And I think that they have value, and I will keep referencing them. I have transcripts for everything, and that might be fodder for future projects. Now that we have AI, we can process these things in various ways. And I think it’s a corpus that is worth keeping up. So, you can expect to find them on the internet for as long as I’m around, in any case.

I do want to end by acknowledging the efforts of the folks who contributed to making the show possible. First, I’ve already mentioned Sarah and the folks at Freedom Makers. Sarah’s help was very important for a considerable part of the show’s run. I’ve already mentioned the folks at Descript, although I haven’t met any of them, I’m a very happy user of the tool, so I want to thank them as well.

I have to thank my family because they’ve put up with my sometimes obsessive publication demands. I’ve really said that I haven’t missed a single publication slot in the five years that I did this. That required that I sometimes focus my attention on the show rather than on other things. And my family has been very patient with me, so I’m very grateful for them making it possible for me to do this and often encouraging me.

Of course, I’m very grateful for all of the guests, everyone who appeared on the show. This would not be a show without them. I think that this epilogue is the time that I’ve spoken the most in the entire run of the show. So, even though I’ve moderated the show and published it, really the bulk of the conversation hasn’t been on my end. It has been highlighting the knowledge, the expertise, and the talents of the wonderful guests who’ve agreed to be on. So, I’m very grateful for them.

And, of course, I’m very grateful for you. I have ended every episode in the show’s run thanking you for listening. And this one is no different. I am very grateful for every second that you’ve spent listening to these conversations. I hope that they have been valuable to you. And I hope that you can stay tuned for what’s coming next. Because, like I said, I’m excited to be spinning up a new show soon. So, until then, I wish you the best. Thank you again for listening – and I hope this was valuable.