Crafting Code Podcast

[00:00:16] Allan Stewart: Welcome to the Crafting Code Podcast, where we discuss the importance of doing the right thing at the right time with the right tools. I'm Allan Stewart, a software architect, and lately I've been thinking about things which get lost in translation.

[00:00:30] Dave Adsit: I'm Dave Adsit, an engineering leader, and recently I've been thinking about sales and marketing when you are the product.

[00:00:38] Allan Stewart: Our topic for this episode is podcast reflections. This is our 64th episode, which in powers of two is significant. And so I got a suggestion from one of our friends that we should talk a little bit about the podcast itself and what it is, how it came about, and why we do it, how we do it. So here we are, 64 episodes in.

[00:01:04] Dave Adsit: Who knew? You know, I have read that typical podcasts don't last past five episodes. And so I feel like we've accomplished something. So how did we get started? It's been so many years. Do you remember?

[00:01:22] Allan Stewart: Well, I believe that I pitched the idea to you and to Matt in early 2019. So we had been working together, the three of us on the architecture team at a company. And I really enjoyed the team discussions that we had. Me too. We would often get together and talk through things. I remember we would do some architecture books and talk about what the teams that we were working with were doing. And just basically any time that we were having a discussion, it seemed like I was... I was learning something that there were good things about it. It was just enjoyable.

[00:02:03] Dave Adsit: Yeah, I really liked that team as well. I felt like it was a great opportunity to really level up some of my skills and learn more about software architecture. So yeah, that aligns very much with what I remember. And also at early 2019, I had just started a new job. And so it was very likely that we were going to lose contact, at least not have the same level of contact if we didn't do something intentional to maintain it. Right.

[00:02:35] Allan Stewart: And I think by May of that year, I had jumped ship and joined the new company you were at, leaving Matt behind. And the idea just kind of sat for a while. I don't remember what it was that caused us to say, hey, yes, we like this enough that we're going to do it. But the three of us did want to spend that time together. Kind of enjoying the conversations that we could have.

[00:03:00] Dave Adsit: Yeah.

[00:03:00] Allan Stewart: And so I think it was late 2019, maybe November or December. The three of us came together at your house. You had some audio stuff. I think your wife used to do a podcast.

[00:03:14] Dave Adsit: Yeah, my wife used to do a podcast about parenting with one of her friends who is a social worker and one of her friends who is a child psychologist. And so I had cobbled together. Some equipment that allowed me to record multiple audio streams and then do some basic editing on them. And so I think we just all got together at my place and set up all the equipment. And Matt has experience in music. So he was probably better at setting it up than I was.

[00:03:48] Allan Stewart: But you did have all of the plastic bags to keep all the cords from getting horrendously entangled. Yeah, that's true. That is true. I think we recorded one episode and then because of technical problems or laptop problems or something, that that was lost. But I'm pretty sure that our first episode, the one that we actually have published, was recorded from that equipment. Yeah, I think so. So after doing some of that recording, we're just kind of barely getting started when March of 2020 hit and COVID started. And then the pandemic sent everything into a spiral.

[00:04:29] Dave Adsit: Yeah, that certainly derailed us a bit. Well, it derailed everything. Yeah.

[00:04:34] Allan Stewart: Yeah, it really did. But eventually we picked the idea up again and we decided to switch to remote recordings, partly because everything had changed. The experience of the world had shifted. The remote things were not as, I don't know, looked down upon, I guess, as they were before. Yeah. It was easier to say, hey, we actually can do something remotely, even though we don't live that far from each other. Right. It might be different if we lived across the country and we were trying to do a podcast. That would kind of necessitate a remote environment. But that's not how we thought about it originally, because it really did come from that, hey, we got together in a room and we would just chat and have a discussion. That was kind of the inception idea of it.

[00:05:25] Dave Adsit: And at that point, I think that's how most podcasts were being done, was you get people together in a room. In fact, the equipment that I bought, I bought specifically to support multi-stream audio because I knew that I was going to have three people talking at once. The shift that allowed us to go full remote actually enabled a lot of things. I think some of those early episodes, I remember not even being at my home in Utah, but being up in Idaho recording remotely.

[00:06:22] Allan Stewart: We had built up a small buffer of recordings and felt good that we could start putting them out and then continue to put out some episodes. And so our first episode was published on November 5th of 2020. And the title of our podcast that we came up with was Crafting Code Podcast. I think we have explained fairly well what we've meant, right? Our first episode covered this. Also, our previous episode just recently talking about craftsmanship and engineering and what do these words mean for us. Yeah. I remember that we felt like there were things that we had learned that were valuable to our careers and that they were worth talking about.

[00:07:08] Dave Adsit: That's right. Well, and also we were at a point where, you know, I really like listening to podcasts and I like listening to engineering podcasts. But I had. Kind of run out of engineering podcasts that I really enjoyed. Back in the day, I used to listen to the Ruby Rogues and I got a lot of value out of that. But that podcast fizzled and went away. And, you know, there were a few other podcasts that came and went that I found valuable for a while and then they just stopped recording. And so I was like, you know, it would be nice if we had something that we could point people to that helps explain in more depth and more detail. Some of the things that we value have come to value through our work in software and some of the things that we think people should understand and consider when they're making technical decisions for their teams and their products. Yeah.

[00:08:08] Allan Stewart: I remember setting up the podcast. There was definitely a bit of a learning curve. There are some things I had never done and getting it going. So. Well.

[00:08:20] Dave Adsit: And I remember when we first got started, you really took the lead on doing a lot of the logistical work of getting our documents hosted, getting our website set up, figuring out how to get us syndicated in Apple and Google and all of the various podcasts syndication places.

[00:08:44] Allan Stewart: Yeah. I kind of felt at the time. And I really appreciated that. Well, thanks. Thanks. I kind of felt like at the time that if we're going to do a engineering podcast about code and these kinds of things, that there was a certain amount of we ought to be able to publish this. Like that. I don't know. There are many podcasting services out there and some of them I'm sure are great. And many of them help with a lot of the logistical things, recording and whatnot. But I kind of felt like if. If a couple of people who know how to write code can't put out a podcast and host it in their own S3 bucket, then what even are we? Do we do we have the credentials to be presenting a podcast? And so that's. This is true. Kind of where that's kind of where I went with it and said, let me let me just do this, which did mean that I had to come up with some things that I have no skill for. I like I remember trying to figure out and this was before you could ask an LLM to generate you an art. So I had to figure out there's going to be something that shows up as a album artwork kind of thing. Right. And so what was that going to be? And it took a while, but eventually I happened to be playing around with some ASCII art stuff for a completely different reason. And I thought I could just make ASCII art of the words of the name of our podcast, which ended up being what the logo is. I think it. Turned out, okay. I am definitely no artist, but from that idea, terminal code, this stuff, that's also the inspiration that I use to create the website as if it was this is the terminal because I knew I mean, how many people bother to go to podcast websites? It's a necessary thing. I think you ought to have one. You ought to have one. But it's not like people like I always listen to all the podcasts I listen to from a podcast app. And it helps me find feeds and it keeps track of all the episodes. And so I very rarely go to the websites. And so I figured, you know, it would be good enough. Build a static site. Again, the kind of thing that somebody who writes code ought to be able to host a static site. And so that's what I did.

[00:11:09] Dave Adsit: Yeah. With our combined 60 years of experience doing web development, putting up a static website with an RSS feed was well within our capabilities. Right.

[00:11:21] Allan Stewart: Another thing that I recall early on, we recorded and we started publishing. And so now we've got a website. We've got a little piece of artwork. I mean, technically, it is, in fact, a image file. So we were able to publish that. But we didn't have any intro or outro music. And I thought about that a lot. And I looked at a lot of. Things online. There are various places you can go online and purchase some music, pay for that one time fee, and then you can use it on your thing. But I just couldn't ever find anything that seemed quite right. Yeah.

[00:12:03] Dave Adsit: Same. Well, and I was constantly trying to pressure Matt, who is a musician, to create some music for us. But he was very busy with a startup at that point and didn't have the capacity to. Yeah. Go authoring music for us. So. Yeah.

[00:12:22] Allan Stewart: And I wasn't even sure what I wanted. Like, nothing seemed right. I wanted something that had kind of a tech feeling, but I didn't know how to articulate that. And so I couldn't.

[00:12:33] Dave Adsit: And the LLM music generators weren't, hadn't been created yet. So you couldn't try 15 different styles.

[00:12:39] Allan Stewart: And I'm not even good enough at something like GarageBand. Right. Like, there are ways to make music, but not by me.

[00:12:47] Dave Adsit: Yes. The way that I produce music is by hitting play on the CD player or rather on my MP3 player. Right.

[00:12:54] Allan Stewart: But luckily, friend of the show, Todd Fisher, offered up his musical services. And so he took some pretty minimal direction. This vague, I don't know what I want, but it goes at the beginning and the end of the episode.

[00:13:09] Dave Adsit: Yeah.

[00:13:10] Allan Stewart: And he gave us what we have now. It incorporates a robotic voice that says crafting code. So. What more could you want? Yeah. I'm pretty happy with it. Yep. In retrospect, I think I could have, I didn't know what I wanted, but some kind of synthesizer, maybe a theremin like you get in the Loki series music. That might have been what I've been looking for. But the music that Todd created for us has definitely grown on me. And now it feels, especially when we get to the, when I'm doing the editing and I get to the end of an episode and I hear a lot of like, oh, this is a great song. I'm lining things up and I hear the music starting to come in as we're wrapping things up. It was like, oh yeah, there you go. This is the right. This is our podcast.

[00:13:54] Dave Adsit: Yeah. This is that this is it. Yeah. Definitely added some depth to the, to the individual episodes and made them feel more complete. So I'm very grateful that Todd volunteered and produced that music for us. Absolutely. So after a while, we kind of. Hit a dry spell. And around 2023, we just stopped meeting, stopped recording. Didn't get any new episodes out for quite a while. Yeah. It seems like we had a few different problems, few different challenges. One of them was we were doing some pretty heavy editing or rather I should say you were doing some pretty happy editing and Matt and I were asking you if you were done yet. And the amount of editing, I mean, it, it made the episode sound very clean and that was kind of the style that a lot of podcasts were following at that point. Right. And so it felt like the thing that we needed to do. But also around that same time, Matt's schedule started to get really busy. I think he moved. Actually, I think he moved several times because he was looking for the right place to kind of set down roots and homestead. Yeah. And so he was. Harder to meet with partially because he was traveling and partially because, you know, we never knew when he was going to be available. And so we just kind of let it fall apart.

[00:15:28] Allan Stewart: Yeah. I can't remember how many episodes we were in at that point, but we had recorded a decent number.

[00:15:36] Dave Adsit: Yeah.

[00:15:36] Allan Stewart: But then with everything going on in life, we just, we stopped. It was easy to kind of. To a, to a halt. We weren't as consistent back then anyway. And like you say, I was probably spending too much time editing. I wasn't using the tools like audio editing. Isn't something that is a passion of mine. And so I learned how to do it and I had learned some things, but I, I'm no expert in those tools. I also have a tendency to hesitate. We will say things like, um, I was spending a really long time trying to go through and listen for every time that we stutter or stay in the same place.

[00:16:44] Dave Adsit: off again. Let's make this happen again. Both of us really enjoyed the conversations that we have while recording and the forced cadence of making sure that we do it. So we decided it's time to restart. And as part of that, we decided to review and simplify. Matt wasn't really available. He hadn't come back, hadn't made himself available to record with us, but podcasting had changed in the year that we hadn't put out an episode. And a lot of podcasts have gone to video, meaning they're recording while streaming a video and you're not going to edit out all of the language hiccups. And so... Some do a little bit still, but... Some do. Yeah. Like that style has kind of shifted. But a lot of them do live streams. A lot of podcasts are recorded as kind of an A live stream and they will stream to YouTube or Spotify while they are doing their initial recording. And the recording goes up after the fact for people who are not available to watch it live. But the podcast is what it is. It's like they could go back and edit, but often don't. And so kind of the expectation had changed. And so we adopted the new style or adapted to the new style, right? Where we... Yeah. Are doing a lot less editing.

[00:18:17] Allan Stewart: Which saved a lot of time for me, for sure. And I also got a little bit better at the tools using features of Audacity, which is the tool I use to like truncate silences and do some things like that. But we just kind of stopped worrying and trying to make it present very clean, very polished and give the podcast a little bit more of the, well, this is raw. This is actually, what we really said.

[00:18:45] Dave Adsit: We're having a conversation and you get to listen in.

[00:18:48] Allan Stewart: Yeah.

[00:18:48] Dave Adsit: And this is how the conversation went.

[00:18:51] Allan Stewart: Unless I say edit and we cut out an entire chunk where my dog came in or my daughter came in or my son wanted to get on the computer or whatever noises are happening around here.

[00:19:07] Dave Adsit: Or I started to tell a story and completely lost my train of thought and had to stop and rethink about it and start over. Not that that has ever happened.

[00:19:17] Allan Stewart: No. The other thing that we did that simplified things quite a bit is that I got rid of the timestamps that I was trying to do. Because it was one of those, again, it was one of those things I'd seen on other podcasts and it's like, oh, well, people put up and it is nice. And there were options for transcribing and now they're even better than ever before. And so maybe one day I'll consider using AI to build the... transcript of our episodes. But for now, getting rid of those made it a lot easier to spend less time doing editing, doing post-production work and more just get it out there. That and fixing up our schedule, our cadence for recording has helped a lot.

[00:20:03] Dave Adsit: Yeah. I was going to say, we should not understate how valuable it is to have set a standing recording time. Every Thursday afternoon, we get together and either prep an episode or record an episode. And we've even been known to record two episodes in an afternoon. But setting that fixed time, as opposed to saying, hey, I can do this or this or this, making that part of our recurring schedule has made it a lot easier to be consistent. And even with just two of us, there are a lot of times when we still miss. Scheduling is super... Super hard. Yeah. If we worked at the same company in the same office and we could just make the time happen during the workday, maybe it'd be easier. But man, scheduling is a challenge.

[00:20:59] Allan Stewart: Scheduling one person, just yourself, your own schedule is hard enough. You add a second person and then a third person and it just goes off the rails real quick.

[00:21:08] Dave Adsit: Yeah. But because of the consistency of the scheduling and the consistency of the recording, we've been able to... We've been able to keep a healthy buffer and still increase our release cadence to twice a month. Is it twice a month or is it every other week now?

[00:21:25] Allan Stewart: So as of 2026, we've started doing it every other week, which is close to twice a month. But some months... It's close to twice a month. Some months get you with that fifth week. Yep. Principles of product development flow. Yeah, that's true. He was right. He was right about the meeting cadence stuff. It's true.

[00:21:44] Dave Adsit: But yeah, it's been really good. It's allowed us to have a consistent delivery cadence for new episodes. I will say that we have enough of a buffer that by the time the podcast episode shows up in my podcast app, I'm always excited to hear what we had to say several weeks ago.

[00:22:07] Allan Stewart: That's right. Because you didn't get to hear it. I got to hear it at least once more. You got to hear it several times. I didn't.

[00:22:12] Dave Adsit: Yeah.

[00:22:14] Allan Stewart: So let's transition for a second and talk about how we make the podcast. You mentioned that we're using Zoom. Yep. Zoom has a nice feature when you're recording that allows every participant's audio to go to a separate file, which that is great. And so you will always... Technically, I have a copy, right? Whatever got sent to my machine as we were recording through the internet, I've got a copy of that. But just to make sure that we get the best quality possible, I take my audio from my machine and you send me your audio from your machine.

[00:22:49] Dave Adsit: Yep. And I think it might be worth mentioning, I use a Focusrite Scarlett 18i8 audio interface to connect to my laptop. And I have a Shure Cardioid 58 microphone on an arm that I try to keep close enough to my face that we... Yeah. ...can get a good, loud recording.

[00:23:19] Allan Stewart: I've gone through a couple different mics. The one I'm using right now that I like, because it's USB-C, is an Audio-Technica one, but I don't know the specific model. But yeah, so we have our mics. A lot of this equipment we've got in our home offices anyway, just because we've been working remotely ever since the pandemic. Yeah. So as far as recording, we... We'll discuss some different topics ahead of time. Sometimes the standing recording time that we have, if we don't have something ready, we will spend the time, instead of recording, planning and figuring out, well, what it is that we want to talk about.

[00:24:00] Dave Adsit: Yeah. So we have a big list of potential topics, some of which we've done a little bit of discussion and expansion on, and some of which are just an idea in a list. But we'll pick one of those and then use a shared Google Doc to expand out a basic outline of what we want to make sure we cover when we talk about the topic. I will say when we first started recording, we were doing a lot of writing and almost reading our script as though it were

[00:24:39] Allan Stewart: the prepared script. Yeah. I think it kind of went back to what we were talking about. We were talking about before where there was a kind of a presentation, like there was a quality bar that we were trying to set for ourselves that we wanted to get. And so now we're less worried about the presentation quality, but hopefully the content quality is there.

[00:24:59] Dave Adsit: Yeah, that's right. Yeah. Now it's a lot more like an outline of these are the points that I think are critical that we touch on. And they may just be some bullets that we'll just expand. And don't forget to tell this one story about... Yeah. ...this one job where you did that thing, you know, that kind of stuff. And so the prep is a lot less overwhelming. Yeah. And the recordings themselves are a lot more fluid.

[00:25:28] Allan Stewart: Ironically, getting back to that, we're having a conversation. Yeah. Like, I think what we're doing now is much more akin to the conversations that inspired the podcast than the scripted version that we tried to... Yeah. ...tried to do to make it feel polished. Yeah. So yeah, it's pretty simple. We get on a Zoom call, we hit record, and everybody's recording to their own computer so that we get nice quality. And we've got the Google Doc that we use, which is just nice for being able to see, like, if I remember something in the middle, I could type it in. If you've got something that you're thinking about, I can see if you're highlighting some text or whatever. And we just have that open while we record. Hmm. Then once the recording is done, then I go in and do the editing. I think I mentioned before, I'm using Audacity. It has some tools for loudness, normalization, for truncating silences, kind of piece together our two pieces of audio and then cut out some chunks if we had to stop and edit a piece. But I do a lot less editing of listening to every little bit and try to say, say, Hey, was there an, um, was there an, ah, and most of my time is mostly just looking. When were we talking at the same time, talking over each other, or I said something and then you said something real quick. And then I said something again, and it was going back and forth and just making sure that it doesn't overlap in a way. Yeah. Making sure it sounds clean, especially with the zoom recording. If you mute during the recording, sometimes it like things get off just by a tiny bit and just making sure that that's smooth. Um, which also makes it easier because zoom does actually a really terrific job of getting rid of loud sounds. So when we first started, we would try to indicate places where we wanted to have an edit by putting, by clapping or making a loud sound. Like, uh, I mean, that's what they do in movies, right? They've got the clapper thing and you get an audio spike that helps them synchronize the start of the frame with the audio, except for that zoom filters that all out. And so I make a nice big clap and, and it disappears, nothing appears on the audio stream. So instead we just both say edit. And then I see that we're both saying something in, in the audio feed. And I, I go to look at it. Why are we saying something at the same time? Oh yeah. It's because I cut this part out. Yep. So yeah, just do kind of that minimal editing. Then I throw, a little bit of audio metadata onto it. And then I have an MP3. I've got a couple of scripts that help, uh, read the MP3 and create the feed, right? The, um, XML file. That's the feed for the podcast. It has certain requirements for it and I can't remember them. And so I wrote a script that reads it out and tells me how many bytes was this file and how many seconds and all of those things. And, and then I just slap in the, the title and the description and run another script to upload the feed in addition to the MP3 file. And, and then it's done and we've got an episode published. I guess I also do go over to the website, you know, we're using GitHub pages to host for right now. And so spend a couple minutes to create a new page specific page because the, the page links to the, the MP3 and to the feed and the feed links back to the page. And so it's nice to have the all, all copacetic. So, but it's, it's made the publishing a lot easier when I was doing a lot less description, just, just a brief summary of what did we talk about in this episode and what

[00:29:25] Dave Adsit: was the name? And it feels adequate. It feels like it covers what we intended when we first set out to do this podcast. So I guess that's the question. Why do we continue to do this podcast?

[00:29:40] Allan Stewart: For me, I continue to have enjoyment in the conversations that we have. Oftentimes topics that we discuss are a direct mirror of things that are happening to us in our day-to-day jobs. And we take the opportunity to say, Hey, what are some of the ideas? What are the principles that are relating to these problems? That we're having? And sometimes it's idealistic, right? Where we're going back to first principles and saying, how should things work? If we could make them how we want, what actually happens at work is not right. Not always quite the same, but, but it is nice to have that, that time to, to converse. But more than that, I, I've always enjoyed making this because I have learned a ton. In the software craftsmanship community. I think I've shared this before, but getting started in my career, I did a lot of pulling myself up by my own bootstraps. I didn't know what I was doing, but I wrote some code. I did go get a computer science degree. So I learned a lot of theory, but the academic theory of computer science and what you actually need to do to build products in the industry are different. Maybe a little less different now than they used to be. But when I went to school, those were very different things. I've never used the pumping lemmas in my, in my, in my work that I've been doing only, only on tests at school. And so I spent a long time learning by myself. I taught myself how to do certain things like unit tests, just by going out there and finding out like, Hey, what are people sharing? And I say by myself, but honestly, even then there were books, there were articles online. There were things that people were giving away that they were trying to inform other people about. And once I got connected with the software craftsmanship community, it exploded for me. And I had many more people who were struggling with the same kinds of things that I was struggling with. I was able to learn a lot more. And so why do I do this podcast? Partly. It's to give back in some small way to that community, things that I think are important that I have learned that have helped me in my career, things that I think continue to be helpful in the AI age, right? Like a lot of our episodes aren't about how, what is the syntax to write a particular piece of code, but it's like, well, how do we interact with each other and what strategies and patterns and things we use? And those are still very applicable. And it's also the reason, part of the reason why we don't have the time to do that is because we're not, we don't have any ads. It's because I want to give this, I want to give this back to the community and there's nothing that bothers me more than when I'm listening to a podcast and an ad shows up.

[00:32:44] Dave Adsit: Yeah, I agree. I, I understand why people have to monetize their podcasts, especially if they have a huge subscriber base and have to actually pay for bandwidth and hosting and things like that. But I am, I am also very happy to continue to have these conversations on a regular, uh, regular, basis and share them with people. There's been a lot of things that you and I have done and learned and discovered working together and building software together that I want to share with people. And as we have built up a library of episodes, I have been able to point people at certain episodes and say, Hey, go listen to this episode about API categorization or about microservices. Or about testing. And let's use that as the starting point for the conversation, because then you'll understand what, where I'm coming from. And then we can build from there. And I found that that's been super helpful for me in a leadership role to be able to point members of my teams to specific conversations that you and I have already had and recorded in the past that, I mean, that's one of the things that's valuable about it, but just like having a regular cadence of, having deep technical conversations with each other has been very valuable for me as well. Um, and I agree. I don't, I don't want to have ads now if, if people wanted to just give us money, that would be one thing, but no one's volunteered yet. So I guess we can just keep putting it out as a, as a contribution to the, uh, to the benefit of the software craftsmanship community.

[00:34:31] Allan Stewart: And I hope that it is a benefit. Yeah. For people that, yeah, at least at the time of this recording, we don't yet have massive numbers of people downloading each episode, but there is a slowly growing number. And for those of you who are out there who are listening, I hope it's helpful for you. Whether it's you're listening and you're saying, Oh man, it's nice to be a fly on the wall of this conversation. And I'm learning something, or maybe I am just agreeing with what, is being said or shouting at Dave and Allan that, wow, you guys are so wrong. Whatever it is that keeps you coming back. I hope that it is enjoyable. I hope that it is useful in, in some way, and we will keep trying to make it be so.