High Low Piccolo with Ian Steyn

In this episode, Ian discusses Week 8 and 9 at Transistor, where he’s whittled down the large video podcasting waitlist and completed the somewhat arduous task of finding dozens of old listing URLs for a customer. He also reflects on how such work can be helpful for getting acquainted with systems and problems. And finally, he looks back on last episode’s production process to consider why video podcasts take a little bit more to make.

Timestamps:

Links:

Creators and Guests

Host
Ian Steyn
Computer Science Student

What is High Low Piccolo with Ian Steyn?

Every episode, Ian shares a high-low-piccolo about his learnings as a software intern: a success, a challenge, and something interesting.

Ian:

Have you ever thought about the fact that the medium of video is actually two mediums? It is both audio and visual, which means it can engage two senses, but it's not surprising that it's a little bit more work to produce. More on that later.

Ian:

Hello, and welcome to the fifth episode of "High Low Piccolo with Ian Steyn", where I share with you a success, a challenge, and something interesting I learned in the past couple weeks. I am, of course, the titular Ian Steyn, a curious guy and computer science student doing a summer co-op at Transistor.fm, the company that hosts and distributes this very podcast.

Ian:

This episode is releasing a little bit behind schedule. It's about weeks eight and nine, which were a little shorter because of Canada Day and a few days of vacation. But I still did and learned a ton of stuff, so let me tell you a bit about that.

Theme:

High! (You're so out of sync)

Ian:

I have a workflow figured out for waitlist emails. So I've been whittling it down, whittling that waitlist down.

Ian:

So last episode, Justin mentioned this big waitlist of what is now over 1,800 people waiting to hear updates about video podcasting. Since then, I have sent out updates to nearly 1,250 people on that list in big batches of around 250 at a time. We're doing it in batches just so we don't get overwhelmed with responses and customer support.

Ian:

But we were doing, like, five a day manually before - five of these emails. So this new process in these big batches is way better. Justin set up the core of it. So there's a spreadsheet that automatically updates as people sign up to the waitlist. You grab a chunk of those emails, import them to Kit, and add them to an email sequence in Kit that sends them that update.

Ian:

But in week eight, as I started doing this, I realized that about 15 to 30 people were getting missed in every batch. If they ever unsubscribed from our newsletter, they couldn't get imported to Kit. So we wanna respect their decision to unsubscribe. Right? But we also want to respond to them since they signed up on this waiting list.

Ian:

So I went and I figured out a process to reach those people. So what I do now, I export the people we contacted through Kit, and I add that list back to Google Sheets. And then I use a special Sheets function that I set up to extract the group of people who are in the 250 for that day, but not in Kit. And then I export that list and add it to Crisp, which is our customer support software, and that has a campaigns feature that I have set up to send them the same email.

Ian:

So, kind of the takeaways from this - or things that I'm getting to experience - are just that I get to have ownership of a business task. It's something that, you know, is gonna bring the company more business potentially. It's an exercise in making sure people don't get missed. Just doing the extra effort to care for our customers.

Ian:

And also, it's a chance to use a little bit of technical skill to improve a completely manual process. It's still not an ideal process. It's not perfect. But it's kind of temporary, so it's okay. It's good enough now. And I'm quite happy with what we have done to improve it. So great success. Great success.

Theme:

Low!

Ian:

I manually back filled all the listing URLs for seven imported shows.

Ian:

So to give a little bit of context here, on the distribution page of a Transistor podcast, we have a bunch of fields where you can submit to major platforms. And then you can also add the listing URL for platforms that you've submitted to manually.

Ian:

So we help people submit automatically just with that button click to Spotify, Apple, Amazon, Amazon Music, Pocket Casts, Podcast Addict, and a couple of others. And then there's a whole bunch of other ones that don't make it so easy to submit automatically on a customer's behalf. But if the customer wants to go and do the work of submitting their show on that platform, like iHeartRadio, for example, then they can come back and put their show's URL on their page. And what that does is if someone goes to your podcast website, they can find the link for their podcast player of choice and just listen to your show there. So you're making your show more accessible.

Ian:

Now, normally, when you import a show from another provider, our system will automatically find and fill out as many of these fields as possible. But we had this one customer who had a real difficult time importing their shows from WordPress. It wasn't really their fault. There was just some nasty caching stuff going on.

Ian:

You know, as the famous saying goes, there are only two hard things in computer science: cache invalidation and naming things, and off-by-one errors. It's true though. It is. It's hard.

Ian:

So anyway, Michael ended up finding a workaround where we copied their feed and then imported the copy. But because of that, we weren't able to automatically fill in all of those listing URLs. Also, it took, like, two weeks to import when it should have taken less than a day. So we wanted to help them out. I was tasked with helping them backfill all of those listings manually.

Ian:

So why is that hard? Well, it's this arduous manual process of searching for a podcast on various platforms and then adding the link, adding the link for Spotify and Pocket Casts and Podcast Addict, and so on.

Ian:

There are some ways we can make this process easier. For example, if you search for a show on Podnews, they will list a bunch of platforms that it's on, which is really helpful. But Podnews lists way more platforms than we support. So you still have to sift through all of the ones that they're listing. Also, sometimes Podnews incorrectly links to other shows with the same name.

Ian:

On top of this, not all of this customer shows were all on the same platform. So they had seven shows. Some were on Podcast Addict and some weren't, for example. And the customer had a website with links to some listings, but some of the platforms on there don't exist anymore or they've been acquired by other platforms. Man, this is complicated.

Ian:

So what is the problem? Well, I haven't seen the technical side of how we actually do the automatic listing filling in our app. But I imagine that we probably have a bunch of different processes for different platforms because it seems like listings are much less standardized than feeds. So your feed is a single document, an RSS feed, and we send that out to a bunch of podcast platforms or rather they pull episodes from this feed. We host the feed for them.

Ian:

But when it comes to your listing on a specific platform, what that is is the representation of your show on that platform. So for example, you have an Apple Podcasts listing that has its own link, and if you give that link to people, that will take them to the Apple Podcasts version of your podcast. And so the ways that those links are formed are not standardized. There's not a standard way to refer to a show on a given platform. Some websites use numerical IDs. Some will use a form of the show's name. Some will use some combination of that.

Ian:

So yeah. That's just something interesting. It's kind of a problem. Luckily, this is quite a rare task that we have to do this for someone. And it's rarer still that we have to do it for so many shows. So it's not like it's something where we really really need to implement an automated process here. But, you know, if it was something that we were doing a lot, obviously, we would want to improve it.

Ian:

A positive out of this or like a good takeaway is I got to know our distribution details pretty well. I've now been on pretty much every podcast app we support. I can tell you all the details of "Okay, there's five apps that pull their links from Apple Podcasts. And if you submit to Listen Notes, then you'll get your listing on Fountain." And I can tell you that Stitcher is now called Pandora.

Ian:

So there's all these little details that I've learned. I know the ins and outs a lot better than I did before this task, as annoying as it kind of was to do.

Ian:

And so as often I find, I found that a manual slog like this can help me get to know a problem or system really intimately. This is one of the reasons that I'm wary of how I use generative AI as a budding software developer. I think sometimes just getting in there and working through it and getting your hands dirty with the details can provide a really solid foundation.

Ian:

A solid foundation for two things. One, for when you need to be able to own the system or the problem. If you have to be able to really answer questions about it later on, and figure out edge cases and stuff, then get in there.

Ian:

It's also a solid foundation for later building tools or abstractions or automations to hide the complexity of the work that you are doing. Doing some of that work yourself first allows you to appreciate the value of the tools that you then build or use. You get to understand what they're doing and when you should or shouldn't use them. At least that is my experience. So I'm gonna keep trying to learn stuff that way.

Theme:

Piccolo! (I'm just gonna use it just like that)

Ian:

Video is two mediums in one.

Ian:

Last episode was called "Video! Video! Video! with Transistor CEO Justin Jackson" because it was a video episode. And also much of what Justin and I talked about was related to our rollout of these video podcasting features at Transistor. Well, having now released that episode, I have some reflections on video from the perspective of a podcaster.

Ian:

Namely, video's value comes from the fact that it is two mediums. It's audio and it's visual. That means that it can engage two senses. It can stimulate different parts of your brain, engage with you in, like, a more holistic way. Right? That's why video is a great medium for explaining things and communicating things and for growing your audience. There's more people who can see different parts of a show or of a creative project you're working on.

Ian:

But my experience was that it also takes more work to produce. To clarify, this is certainly not the first time I've ever produced a video, but I can now directly compare audio podcasting to video podcasting on a show with a consistent format. So here's some things that I had to consider when I was working on this video episode.

Ian:

One is it just took longer to edit. I spent all week on it. One reason for this is partly just because Justin had a lot to say. It was longer than my typical episode. We had a nice long conversation. But also, there's a lot more factors to consider, right?

Ian:

Audio cuts are more invisible. So if I cut out a sentence cleanly, you'd never know. But with video, you will see that cut because people will be in a slightly different position in the frame. That's fine. I think people are used to seeing some jumpy cuts in an edited video like that. But I can't be cutting out every "um", and every, like, intake of breath in between stuff because I don't want it to be super choppy. So I kind of have to pick and choose my battles of where I'm gonna cut stuff and where I'm just gonna leave it more natural.

Ian:

Even the recording process had more for me to consider. For example, I held my mic in my hand instead of putting it on a stand because the stand is too low for me to be framed normally in camera while talking into the camera. Also, as someone with ADHD who has a really hard time sitting still and making eye contact for prolonged periods of time, I found myself having to work pretty hard not to be, you know, contorting my body into weird positions and be looking around everywhere as I tend to do.

Ian:

Also, your recording location might have to change. So with audio, I could be inside of my bedroom closet. I'm not, but I could be. And I could get great acoustics because there's lots of sound dampening material in there, such as my clothing. But it's a little hard to film video inside your closet, and the backdrop might not suit your show.

Ian:

So this is not a value judgment on video podcasting. I'm not saying video podcasting is bad because it's harder. In fact, I think there's lots of really great things about it, like I already said. It engages more senses. It can broaden your audience. If you're using it to explain stuff for, like, educational podcasts, especially, it can be really helpful. But for me as a first timer, I'm just reflecting on the fact that video is different from audio podcasting, and it takes a little bit more prep, little more work afterwards to do

Ian:

Also, it is up to the creator how much polish they wanna put on it. I made some visual graphics for the intro and transitions because that's something that I thought about. I didn't have to do it, so that's another reason it took longer.

Ian:

In conclusion, there's lots of cool things about producing something in two mediums at the same time. But because you're getting two mediums out of it, it shouldn't be all that surprising that you have to put a bit more work into it.

Ian:

This has been High Low Piccolo with Ian Steyn. Thanks for listening, and tune in next time to hear about more interesting stuff that I am learning. Bye bye.

Ian:

Bla bla bla bla, mee mee mee mee, moo moo moo moo, ah-ee-ooh. I think I've come to the conclusion that you need vocal warm ups sort of even when there's no one here with you. You know, I kinda just need to get used to talking around, talking out loud. Take another sip of water.

Ian:

I'm running out of breath. Probably because I turned off the air circulation in this booth, so it would be quieter.