#170 March 2, 2022

Kubernetes: The Documentary, with Josiah McGarvie

Hosts: Craig Box

Six years after its creation, Kubernetes is the subject of its very own documentary film. Job platform Honeypot has released. Josiah McGarvie was Honeypot’s head of video, and the lead filmmaker for Kubernetes: The Documentary. Join us for the director’s commentary.

Do you have something cool to share? Some questions? Let us know:

Chatter of the week

News of the week

CRAIG BOX: Hi, and welcome to the Kubernetes podcast from Google. I'm your host, Craig Box.


CRAIG BOX: Way back in episode 21, we spoke to Ihor Dvoretskyi. Ihor is many things. He's a senior developer advocate working for CNCF. He was a co-chair of the Kubernetes PM SIG. He's a Google Developer Expert. He's a great guy to have a drink with at a KubeCon.

Ihor is Ukrainian. He lives in the city of Lviv. And as of Saturday, he's a member of the Territorial Defense Forces of Ukraine.

Ihor should be holding a microphone, not a rifle.

I invited Ihor to guest host the show today if he could, but he politely declined. I asked him what our listeners could do. He shared links to the Ukrainian charity Come Back Alive and the fund of the National Bank of Ukraine, which you can find in the show notes. Donations to those charities will be used to fund Ukrainian citizens fighting back against the Russian invasion.

You may prefer to donate to the humanitarian relief effort, in which case, I've included a link to the International Committee of the Red Cross.

Google is matching employee donations dollar for dollar, so I've made my donation through our internal giving platform. Many other employers will do the same. Please take a moment to consider if there's anything you can do. Let's get to the news.


CRAIG BOX: The Podman project released version 4.0 this week. The headline feature is a new network stack with a tool called netavark to configure container networking. The new stack promises improvements with multi-network containers, IPv6, and performance. Other enhancements include the ability to mount the Podman socket on a Windows or Mac host, allowing the use of tools like Docker Compose. Check out our interview with the Podman team in episode 164 to learn more.

Signadot has launched a public beta of a Kubernetes-based testing platform for microservices. The Signadot SDKs allow developers to spin up virtual environments in the multitenant Kubernetes cluster, allowing testing at the granularity of pull request or commit. The company also announced a $4 million seed round led by Redpoint Ventures.

Also announcing funding this week, Okteto, who we spoke to in episode 125. Okteto provides remote development environments in Kubernetes clusters. Their $15 million series A round, led by Two Sigma ventures, further validates that there's both complexity and money to be made in pre-production.

Platform 9 surveyed over 500 people to put together their 2022 report on enterprise trends in cloud native. As is the case with most cloud native reports, this one suggests that Kubernetes has effectively subsumed all other alternatives for container management. As is the case with many multi-cloud vendor reports, it suggests that people are very interested in multi-cloud, with lock-in being cited as a concern, especially, at the executive level. You can have a copy for the price of your email address.

Finally, community storage and platform vendor Robin.io, has been acquired by Rakuten Symphony. Rakuten, who literally purchased a Super Bowl ad to make sure I pronounce that right, is a Japanese e-commerce giant, and Rakuten Symphony is their telco division. Rakuten used Robin's technology to power their Rakuten mobile product and were an investor in the company. The sale price was not disclosed.

And that's the news.


CRAIG BOX: Josiah McGarvie is a filmmaker based in Brisbane, Australia, who makes documentaries for Honeypot. Welcome to the show, Josiah.

JOSIAH MCGARVIE: It's great to be here, Craig.

CRAIG BOX: What is a honeypot?

JOSIAH MCGARVIE: Honeypot is a developer-focused job platform based in Germany. It's like a reverse recruitment platform. Basically, it's a platform where developers can create a profile, and then they get vetted. And companies pay to have access to the platform, and then they apply to the developer, rather than the other way around.

As a developer using the platform, you would get requests from companies to come and work for you. Basically, that's how it works.

CRAIG BOX: Online dating for nerds.

JOSIAH MCGARVIE: [LAUGHS] Online dating for nerds. That should be the tagline. That's better.

CRAIG BOX: Why does a job platform like Honeypot have documentary filmmakers on staff?

JOSIAH MCGARVIE: It all starts with one of the co-founders of the company, Emma Tracey. Emma had this idea about marketing that was taking a completely different angle to traditional marketing. Essentially, that angle was, marketing as a product. Most marketing is invasive. Most marketing is ads. It's pop ups. It's hey, come and do our thing.

CRAIG BOX: Punch the monkey.

JOSIAH MCGARVIE: Punch the monkey. Buy our product. Look at our thing. Look over here. Let's do an ad in front of your YouTube video that you want to watch. It's annoying, as all advertisements are. But if you approach marketing with a product mindset, and instead of creating ads, you create a product that people want to engage with, like documentaries. That's a different way of doing it.

Basically, the idea is, creating something that makes people want to approach your brand, rather than invading someone else's space. Emma Tracey's brilliant idea was, can we create documentaries around tech that tell these stories that no one's heard before?

When it comes to open-source software, especially big open-source software communities, there's huge numbers of people who are very passionately involved, who have their identities attached to these projects. Up until this point, no one's really paid any attention to it in terms of documentary filmmaking. The stories of these technologies haven't been told in this way before.

So Emma Tracey comes to me and says, why don't we do documentaries about open-source technologies? About the origins, about the people involved, about the stories, how do they come to exist? If we can tell the stories of these communities, that's going to create trust in our brand. That was essentially the idea. A marketing strategy that would create trust rather than annoy people. And so that was the idea. Documentaries, documentaries on open-source technology.

CRAIG BOX: How did you personally come to making documentaries? And how did you come to be the person telling open-source stories for Honeypot?

JOSIAH MCGARVIE: I've been interested in documentary filmmaking since I was about 18. As far as I know, there's no obvious way to get into documentary filmmaking asides from going and studying in University, but often that doesn't lead to any work. I was always really passionate about documentary filmmaking.

And in 2016, 2017, I wanted to travel. I wanted to go overseas, and I wanted to get an opportunity to make documentaries, and just see what's out there. And see what the opportunities were. And I was looking around in Berlin. I really wanted to move to Berlin. And Honeypot was a very young company at that time. And they were looking for a filmmaker. I applied, and they said, why don't you come out to Berlin and help us make films?

It was pretty crazy because I was in Brisbane. And I was just applying for jobs in Berlin, not expecting to hear anything. And then all of a sudden I get this offer, and they're like, how soon can you start? And I'm like, oh, shoot, can I start in four weeks? And so I think it was three or four weeks. My wife and I packed up like our whole life in Brisbane and moved to Berlin. I'd never even been to Berlin at that point, and we just moved straight away.

CRAIG BOX: Did you know anything about technology?

JOSIAH MCGARVIE: I knew absolutely nothing about technology. I still basically know absolutely nothing about technology, to be honest.

CRAIG BOX: I imagine, though, that the things that you need to do as a documentary filmmaker today are very technology driven. I'm sure that you're able to do all the recording and editing yourself. Whereas in the past, like you say, you would have had to go to school and then had a very small part of this. And only a small number of film studios would be able to put something like this together. It's very much been democratized now through technology.

JOSIAH MCGARVIE: Yeah, that's true, but I think there's a big difference in understanding how a camera works and microphones and all that, that's kind of technology, and then tech in the sense of software engineering. I know my way around Final Cut Pro, but I don't know the first thing about JavaScript.

CRAIG BOX: What are some of the other documentaries that you've made for Honeypot?

JOSIAH MCGARVIE: The first one we did was on Elixir. I'm not sure if you're familiar with Elixir. It's kind of a niche technology. We did that one. A really short documentary, I think it was like 12 minutes. That was our prototype, our first step into the world. It was kind of like, let's see. Why don't we make a documentary about Elixir? We'll make it about this open-source technology.

We have a connection with someone who knew the founder. And so we said, OK, let's try it out. We'll start a YouTube channel for the documentaries. We'll do this first project, and we'll see what the response is. We did that first one. Flew to Poland to visit José Valim, who's the creator of Elixir. We filmed with him for a few days. We went to the Elixir conference in Warsaw, filmed some stuff in Berlin, as well. There's some companies using Elixir in Berlin, put together a little show documentary. And we set up our YouTube channel, and we put it out there.

And in a few weeks, I think it got 70,000 views, something like that. We thought, oh, this is something that people are interested in. This is an audience. Because when I was first asked by Emma Tracey to come and work on these documentaries, she said, hey, what we're going to do is we're going to make documentaries about open source. We're going to tell the origin story. We're going to tell the come-up story of these technologies. And I was thinking, is anyone going to watch this?

CRAIG BOX: I would be thinking, there's only so many radioactive spiders that can bite programmers.

JOSIAH MCGARVIE: [LAUGHS] Yeah, yeah, well, I was thinking who the heck wants to watch a documentary about open-source technology? But OK, this sounds like a fun project. I don't know who's going to watch this.

CRAIG BOX: Exactly the sort of people who would sign up for your job board.

JOSIAH MCGARVIE: [LAUGHS] Yeah, good point. So I was skeptical. But I can't take any credit for this idea. It was really Emma had this foresight that I didn't have. We did the first one Elixir. That one for us, it blew up. Now it has 164,000 views. At the time, I remember I think once we reach the 20,000 mark, we were like, wow, there is an audience for this. That kind of blew me away. Then after that, we were like, OK, let's keep doing this. Let's keep creating these documentaries.

Then we did a documentary about Ember.js. Ember was the framework that Honeypot was built using. So we thought, OK, that's a connection. Now, let's go do an Ember documentary. So we did that. That was a little bit more extensive. We went to Amsterdam. We went to Portland, where the Ember people are. And we went and filmed with them. And that was really fun.

At the same time we were in the US, we were able to get in touch with the guys who started GraphQL. And we thought, oh, let's do a GraphQL one at the same time. So we put those two out. And really, the GraphQL one blew up. That got like 150,000 views in a really short period of time. Now it's almost half a million views, but we were like, OK.

At that point, we were like, I think we found a niche here. That there is a lot of demand for these kind of stories, and no one that we know of is making them. And so we thought, hey, this is pretty cool. So after that, we ended up doing Vue. That was super cool.

This is the story that I remember. Someone at Honeypot just tweeted-- I think they were at a Vue.js conference in Amsterdam. And they thought, oh, Evan You is on stage. I'll just tweet at him. I think they took a photo of him on stage, and then they tweeted at him. Evan, cool talk. Why don't we do a documentary with you. Have a nice day, or something like that. And then Evan got back to us, and said, yeah, cool, let's do it.

We went to Shanghai to film with Evan, and we went to the US and all this. That was really cool. Very much a benevolent dictator for life. So the documentary, that one blew up. That's got like 1.2 million views. I think that was what really put Honeypot on the map for a lot of people, the YouTube channel. There was that documentary because it was so popular.

And I think it was the success of that documentary that allowed us to win over the trust of all the people we wanted to interview for the Kubernetes documentary.

CRAIG BOX: How did you then select Kubernetes as the next topic to cover?

JOSIAH MCGARVIE: The way we picked Kubernetes was really interesting because we didn't pick Kubernetes. What happened was, I was working on the final few weeks of editing the Vue.js documentary, and I got an email from a guy named Chad Torbin at Speakeasy Strategies, a PR company in San Francisco.

Chad reaches out to me in an email, and he goes, hey, Honeypot. I don't know who should be receiving this email. I don't know what's going on. You guys have this really cool channel where you're creating documentaries. He was like, that's really cool. That's really interesting. I don't know of anyone else who's doing that but you guys. And he says, I happen to have the connections for all the people who created Kubernetes, and I can help you get in contact with all the people who created Kubernetes, just in case you're looking for another project, and you're not sure which to choose. Kubernetes, you may have heard of it. It's kind of--

CRAIG BOX: It's kind of a big deal.

JOSIAH MCGARVIE: It's kind of a big deal. Maybe you want to do Kubernetes. So I said, "what's Kubernetes?" I Googled it. I was like, OK. Again, I'm not a technical person, so I have to Google this stuff. And I look it up, Kubernetes. And I see all these people posting graphs of how it's just exploding. It's like the biggest open-source thing by far since Linux.

I was like, wow, OK, this is a big deal. So I wrote back to him. I was like, yeah, this sounds really cool, Kubernetes awesome. Once we finish Vue.js let's get started. So I emailed Chad back, and I go, this is really cool, yes, Kubernetes. Wow, big deal, we'd love to do a documentary about Kubernetes. We were eager to level up from our previous success with Vue.js. So we thought, OK, let's do Kubernetes. Chad Torbin is the reason why this happened.

CRAIG BOX: Well, thank you, Chad. You mentioned that leveling up, there are a couple of things that were different about this documentary versus the ones you've done before. You have sponsors for this documentary for the first time. What is the relationship between Honeypot and the documentary sponsors?

JOSIAH MCGARVIE: I got to back up a little bit to understand even why we decided to go with sponsors. Once Vue came out, we were like, OK, if you guys are familiar with your COVID-19 lore, these dates are going to be significant to you. Vue.js comes out on February the 25th, 2020. March, we're just dealing with promotion and all that March 2020, we go, let's start production on Kubernetes. As you may know, March, 2020, is when the COVID-19 situation popped off. And we all went into lockdown in Germany, hard-core lockdown. Now, the way that I keep track of all the COVID-19 dates, is how I keep track of the Kubernetes documentary project.

So we go into lockdown, Guillermo Lopez, and I, the filmmakers who are going to be working. He's a brilliant filmmaker. So Guillermo and I go, OK, we're in lockdown. We're working from home. We start getting together and going, OK, let's start mapping out the Kubernetes story. We start doing some research. We listen to a little bit of this podcast, maybe. We're researching. We're looking up people who've published timelines. We're trying to get our heads around the story of Kubernetes.

The first week, we just were trying to figure out what Kubernetes was. We would do some research, and then we'd like go, OK, you tell me what you think Kubernetes is and summarize it in your own words. And we'd be like, oh, it's this thing that does this. Oh it's like an orch-- oh, it's a container orchestration platform.

And then it was like, OK, well what's a container? Next job, is to figure out what's a container. After that, we'll figure out what's an orchestration. And then after that, we'll do research on what a platform is. We were really down to basics. Someone had made this really good animation, and they put it on YouTube. It's explaining Kubernetes to my like five-year-old daughter or something like that. Guillermo and I are like, perfect, this is the kind of content we need right now to help us understand this technology.

So we're doing that. Eventually, we get our heads around it. With the help of Chad, we start reaching out to all these people Craig, Joe, Brandon, Ville, Tim, Brian, Sarah Novotny. All the people, basically, who you see in the film, among others who didn't end up wanting to be in the film.

And we started talking to them. We're recording all these calls. We'd talk to them on a video call for an hour or so. And we'd record that and take notes. And that was the process of really figuring out, OK, where are these people coming from? Where's their mindset? How does the story fit together? What role does this person play in the story? We're hashing that out.

And we're going, this is great. No distractions, we're working from home in lockdown on this story. It's really coming together. By the time this pandemic is over in a few weeks, we'll be ready to shoot. So we're like, yeah, that didn't happen unfortunately. [LAUGHS] When is this COVID-19 thing going to be over? It's really starting to get annoying.

Ended up being, I think it was in September, 2020, sometime around then. It was the end of the first wave, and there was a lift of all the restrictions. And we thought, brilliant. There's a few people in Europe that we want to interview for the documentary. Arnaud Portier; I think I'm pronouncing his name right. Great guy, he used to work for Docker. Really, really, interesting guy. He wanted to talk to us, do an interview in Amsterdam. And then Ville Aikas, who was in Finland at the time.

So we went to do those two interviews. We went to Ville. We went to his island and everything. That was awesome. Ville is an absolute legend. After we'd filmed those two interviews, we had the lockdowns again. And anyway, this project was dragging on. COVID was really putting a spanner in the gears of our plans, and it was frustrating.

We had this huge list of people that we wanted to go and visit in the US. We wanted to fly out with all our gear and travel around the US and meet all these people, get their story, film at all. Our project was becoming more ambitious than we originally thought, as we talked to people the story grew. And we thought, we want to do two parts. We want to do three parts. At one point, I think we were even considering doing four parts, based on all the different people that we wanted to interview, and we wanted to talk to, and we wanted to have an episode on this section, and an episode on that. There were plenty of people that we really wanted to talk to, but we didn't have to.

The project was growing, and we realized our budget for the documentary was not going to cut it. We know there's companies out there who really care about the Kubernetes community. They want to see a documentary about Kubernetes as much as we do. Why don't we reach out to them, and say, hey, can you spare some change? Can you give us a little money, so we can pay for this trip and cover the costs of creating this documentary?

So we reach out to a bunch of companies. The ones that ended up getting back to us who are interested were, Cloud Native Computing Foundation, Red Hat, and Google. We ended up reaching out to these sponsors. It was after we had already gotten everyone on board. And we'd already shot some of the interviews. So this was kind of a later development. Basically, our pitch was, hey, we need a bit of money to finish this project. We know you want to see it made as much as we do. So can you give us some money?

And so we ended up getting a bit of funding from Google, a bit of funding from CNCF and a bit of funding from Red Hat. And that really was the rocket fuel we needed to get over to the US and pay for that trip and pay for the documentary to be finished.

CRAIG BOX: At this point you've done a bunch of pre interviews. You've learnt what Kubernetes is, and presumably, how kubectl is pronounced. And you've started interviewing people on record. At which point have you decided what the story that you want to tell is? Have you decided it beforehand based on those conversations, and then you're just filming interviews to get the information that you need? Or is the story constantly evolving as you meet and film people?

JOSIAH MCGARVIE: There's two phases to that. We wanted to do as much research as we could to have a pretty solid story of how we wanted to tell the documentary. From my experience with making documentaries, you have to expect that that story is going to change at least 30%. Even if it's just some scene that you imagined isn't as important as you made it up in your mind, and so you just scrap it altogether. Or you learn something new in the actual official interview, that you didn't find out until then. And you go, oh, wow, this is big. Let's move some stuff around and put that in there. So it changes a bit.

There is a huge challenge with this kind of project. There is a huge challenge with creating open-source documentaries or technical documentaries, which is, the stories often aren't obvious. There isn't an obvious cinematic story in there. The likelihood that you're going to create something dry that's just a series of interviews and hardly a documentary, you're fighting that the entire time.

Other stories, there's much more obvious villains. There's much more obvious obstacles. So much of tech is very abstract. What we're trying to do, is get an idea of what happened, and then fit all those elements into a traditional story structure into a traditional story arc. The villain becomes not getting permission from the managers to open source. That's one villain, or competition in the industry.

CRAIG BOX: It kind of felt like the bookshop was being painted as the villain at one point.

JOSIAH MCGARVIE: [LAUGHS] Oh, yeah, the thing to remember with documentaries is, documentaries aren't necessarily the truth as you would hear it in a court of law. A documentary is someone's impression of the truth, someone's version of the story. You have to kind of get inside the mentality of the characters, really, and have these roles filled.

For example, everyone has the villain in their life. Everyone has the challenges in their life. Everyone has the doubts. Everyone has the hesitations. It's all these universal story elements that you have to identify. So you've got to get into the mind of the people who were trying to make Kubernetes a success, and go, how did they see the landscape ahead of them? How did they see the obstacles? How did they see what the villains were?

And so then you use their perception. You use their eye as best you can. And use that to create the landscape of the documentary. And so in this case, the villain, in some ways, was Amazon. That was the big threat. Not to say the managers at Google were villains, that's not what I'm saying. But in the context of the film, you've got to create these elements. In the world of the film, you've got to have your villains. You've got to have your adversaries that characters can overcome if that makes sense.

CRAIG BOX: Something I'd like to ask about filmmaking, perhaps. You said there that documentaries do have a story arc, and they do have villains. You hear a lot about biopics, which are a movie made about a real person, but that actually does have drama and it is obviously made to fit that mentality. I know, for example, in the recent movie "Bohemian Rhapsody" about Freddie Mercury, there was a lot made about the fact that he had a relationship with his personal assistant, and that person was going to blackmail him and so on.

And then a lot of people said, well, actually, that didn't happen that way. And the band didn't nearly break up over it, and so on. But it was also manufactured for the story. And if it had been a documentary, it would have just been telling the story. These things happened beat, beat, beat, and so on, but it might not have been so interesting. I'm interested then to hear that you as a documentary filmmaker feel like it's a little bit more like a biopic of these people, where you do have to have that conflict and tell that story with a more traditional arc.

JOSIAH MCGARVIE: That is the key. The key is, if you're going to make a good documentary, a good documentary is going to have more of an arc like a traditional film. A boring documentary is going to have less of an arc. The exaggeration element of things, like how much do you exaggerate? I don't think we exaggerated anything beyond any of the real perceptions of the people who were in the documentary.

It's hard to do that, especially, when you're not using a voiceover. You're just using the words of the people in the documentary. But it would have been a lot more exciting if we had done that, and we had more villains.

CRAIG BOX: Did you consider doing a voiceover?

JOSIAH MCGARVIE: I would have done it if there were big holes that we had to fill. But generally, that's not my style of filmmaking. I don't like to do voiceover. I like to see everything told directly from the horse's mouth, if you will.

CRAIG BOX: You generally only have one chance to sit down with each of these people. It looks like you didn't go back and speak to people, or if you did, you made sure they were dressed exactly the same way as they were before. Are you asking the same questions of everyone or have you set something out specific to say, hey, we want you to tell this part of the story? And how do you tell who answered them best, and who gets put in to answer any particular point?

JOSIAH MCGARVIE: I would say, everyone got 80% of the questions, and then there were certain people that we were able to identify, who could only tell a certain aspect to the story. And so we would say, OK, Craig and Eric were the only people who can really tell the story of the bus. For example, Sarah Novotny has this really great perspective about community building. She's got a real intuition and a real gift of insight there, that I think it would have been difficult for anyone to tell. So there were certain questions we asked Sarah that we didn't ask anyone else.

And there were certain specific questions, but mostly, it was, we took the elements of the story that were completely crucial, that we couldn't possibly go without–we asked to as many people as possible because then if someone says it in a way that isn't so articulate or they don't cover all the points, you've got all these versions, then you go, oh, OK, they didn't say it quite right. We'll just use someone else to say that story. So most of the important parts of the story, the key points that we couldn't live without, we asked everyone, so we had as many versions as possible.

And then, yeah, we did just go, this person tells it in a more funny way. This person tells it in a more exciting way. Let's just use that instead.

CRAIG BOX: I will say that the B-roll footage that you used for the bus ride, I think, glamorizes perhaps what it's like to travel on a Google offsite bus.

JOSIAH MCGARVIE: [LAUGHS] There's a call out here, call out to Google, whoever's organizing the buses.

CRAIG BOX: The American highway system is never that cinematic and picturesque.

JOSIAH MCGARVIE: The bus is on a European road, so that's why it looks so nice. [LAUGHS]

CRAIG BOX: In terms of the footage that you use to tell some parts of the story, especially from the past, most of the Docker story, for example, is told through video. Was that all public footage or were you able to get behind the curtain to get access to some of those videos?

JOSIAH MCGARVIE: We used all public footage, but there was some very obscure footage that we came across that was just on YouTube randomly. We found a video from, I think it was Docker's one year anniversary, that someone who worked at Docker at the time, had just recorded on their phone and put up. And we thought, wow, this is a really cool video showing this little time capsule, so we used that. We used a lot of footage from different events that we tried to coordinate based on the timeline and put that in there.

CRAIG BOX: Was there anyone that you would have liked to have talked to but couldn't?

JOSIAH MCGARVIE: Yeah, there's lots of people that we would have liked to talk to. The top person that I would have loved to talk to, would have been Solomon Hykes. He seems like such an interesting guy. We corresponded a bit on email back and forth. He seemed like he was initially interested in telling his side of the story, but in the end, he decided he didn't want to be in the documentary.

There were a couple of other people. Benjamin Hindman, who's behind Mesosphere. That would have been such an interesting element. We had mentioned a little bit about Mesos in the documentary, just very briefly. We really wanted to go into this competition between Mesos and Kubernetes much more. That was a very, very interesting element of the story to us because it was like this arms race during that period of time.

And we were on the phone with Benjamin a couple of times talking about it off the record. And it was very interesting. But in the end, he didn't want to be in it. We didn't feel like it was fair to go into that story any more than we had to, without him in it. So if he was in it, it could have been a three-part documentary.

CRAIG BOX: I want to dig into that. Where do you decide to stop a story? Talking about Kubernetes, someone said, "we did a rewrite in Go". I think there's a documentary in that entire sentence.

JOSIAH MCGARVIE: That's the thing like, there's certain impulses that you have where you want to indulge in one area of the story or the other. And one of the rules of filmmaking is, you want to leave people wanting more rather than leave people having seen too much. When things start to get complicated in the edit. When things start to get unclear about what should be included and what shouldn't. It's always best to stick with the main story beats. It's always best to stick with the abridged version.

The thing that really annoys me about documentary filmmaking these days, although, we are in a golden age of documentary filmmaking like I've never seen before. We've never seen such good documentaries. But you do have this tendency from like, Netflix documentaries, I'm not sure if you watch them. Well, they'll take a story that could be an hour and a half documentary, and they'll turn it into a 10-part series.

I was just watching the Kanye West documentary the other day, and I'm in shock. There's a scene where they start going into the back story. The cameraman starts going into his own back story about his early life growing up, and how he got-- and I'm like, isn't this a Kanye West documentary? It's frustrating because at some point, you just got to get to the point. Tell a concise story and not drag it out. It was that energy that we tried to bring to Kubernetes. Because it could have been five hours at the end of the day, but we wanted it to be consumable.

CRAIG BOX: Your documentary is in a wider aspect ratio than you normally see on YouTube. It's more what you'd expect for 35mm film. Why that decision?

JOSIAH MCGARVIE: Film nerd alert. The aspect ratio that we uploaded in the film in, is 2.35:1. Most of what you see on TV now is 16:9. And the only reason that 16:9 is used, is because when TV manufacturers started to make TVs, they didn't know whether to do 2.35:1, in order to watch films, or 4:3, in order to watch TV shows, which were all 4:3 at the time.

So they just said, well, why don't we just split the difference and make it 16:9. So the film aspect ratio doesn't look too bad with too much black at the top of the bottom, and the 4:3 TV ratio doesn't look too bad with the black on the sides, so we'll just invent 16:9. 16:9 only exists because of TV manufacturers. And I think, and maybe, perhaps you would agree, 2.35:1 just looks on the surface way more visually pleasing, essentially because 2.35:1 looks cooler, so [LAUGHS] that's the answer.

CRAIG BOX: It definitely has that "I'm aiming for old-school cinematic" feel about it.


CRAIG BOX: You gave a warning of a film nerd alert there. You have to realize that when you're making a documentary like this for software nerds, that there will be people who pause on any piece of source code you put on the screen.


CRAIG BOX: The first piece of source code that comes up is actually from some PHP Bitcoin mining software, as far as I can tell.

JOSIAH MCGARVIE: Oh, jeez, this is like my biggest nightmare whenever I create. Whenever I'm working on these documentaries, it's like my worst nightmare is people pausing on this code. Because I'm here going, OK, you need to make these things as visually stimulating as possible.

CRAIG BOX: Of course, get the prettiest code, put that in there, doesn't have to compile.

JOSIAH MCGARVIE: Yeah and often the challenge is, you don't have hardly any footage of what's going on. And actually, if you want to get into this later, we faked a lot of archive footage for the documentary. This is going to be a big scoop. No one knows this.

CRAIG BOX: Craig McLuckie did not have that beard in 2014.


CRAIG BOX: Don't think you slipped that past me. I wasn't going to mention it. I was going to give you a pass on that, but--

JOSIAH MCGARVIE: Yeah, well, you have an eye for detail. When we're in the edit and, we're like, OK, how can we make this visually interesting? Let's get some footage of some of the code that they're working on. How do we do that as people who are not software engineers and don't know how to identify the difference?

So we just go on Google or YouTube and type in Kubernetes code or whatever, PHP code, Docker code, and try and find something that's convincing enough. And then flash it quickly on the screen and hope no one pauses. Here's some code for you. Don't pause, though. [LAUGHS]

CRAIG BOX: I apologize. One other thing I did see in the background of one shot, which I wasn't specifically looking for, but does talk to the current realities, was someone wearing a face mask. How did you manage to pull off the filming during the worldwide COVID restrictions?

JOSIAH MCGARVIE: This was just a nightmare. I already told you how we filmed the European stuff already. So we had to go to, mostly, we were in San Francisco and Seattle, to film with all the people that we wanted to film. And the project had been delayed, and it had lots of setbacks you couldn't even imagine, that I can't even get into.

So basically, we couldn't put it off anymore. And we had to pick a time, and we said, we're going. We're going in August, 2021. And we're going to film it all rain or shine, pandemic or no pandemic. It's going to happen. I was flying out of Brisbane. Guillermo was flying out of Berlin.

At that time, Australia was hard-core lockdown, and we had basically, zero cases. So I could get into the US easily without any problems. There was a travel ban from Germany to the US. So in order to get into the country, Guillermo had to fly to Panama, and stay in Panama for two weeks because you weren't allowed to enter the United States if you had been in Europe in the previous two weeks. So he had to stay in Panama for two weeks. Thankfully, his parents live there.

CRAIG BOX: There are worse places to have to stay for two weeks, I'm sure.

JOSIAH MCGARVIE: [LAUGHS] Yeah, can I stay in Panama for two weeks, please? Then he was able to fly to San Francisco. So we arrived in San Francisco the very beginning of August, could have been the 31st of July. I fly out of Sydney, and Sydney just had hit a spike of COVID. And I was worried that they were going to lock down the city. The night that I flew out of Sydney, I landed in San Francisco, and I checked the news. The Australian government had just sent in the military to lock down the city of Sydney. Then, while I was flying out, so I got out just in time and got into the US. And they said, oh, you're Australian. You can come in. That's no worries. So we were in. OK, that was one hurdle. We were able to get in. That was a mission just by itself.

Then we had a challenge of where we were going to shoot all these interviews. At the time, we were talking with some people at Google, who were helping us. They said, we think we can probably arrange for spaces on the Google campus in Palo Alto, I think, is where the Google campus is. We can arrange for spaces on Google campus for you to film, film all the interviews. And we said, brilliant, that's awesome. They said give us a little while. We'll get back to you soon.

It ended up taking a little bit longer than usual. And as we were dealing with other hiccups, we were just sort of neglecting to follow up on that. So the day we arrive in San Francisco, we get a message from Google saying, yeah because of the whole COVID situation actually, we're not allowed to give you permission to film on the campus. We go, oh, oh, no, what are we going to do? We have no way to film this.

So we're kind of freaking out a little bit, me and Guillermo. Then Chad comes through with an absolute save. Chad Torbin from the beginning, this is like a throwback character now. So Chad comes through and he goes, oh you, can use our Speakeasy offices in San Francisco. Everyone's working from home, so they're going to be totally empty. So we go, oh, thank god for that. We organize to pick up the keys from Chad, and then the next day, I think, we started filming.

So we went and checked out the location and figured everything out. We ended up shooting all the interviews in San Francisco. All the ones in the Bay Area were all at that office. So that was like a huge save. So it was totally empty offices. There wasn't people walking around in the background for the most part with masks, except for you spotted one. So that was huge. We were able to get that.

The other challenge was, some people who we wanted to film, ended up dropping out because of COVID-related things. So we were going to film with Brendan Burns. We had it all lined up, and he needed to cancel the interview because of COVID concerns. And so that was an added challenge. Thankfully, no one else ended up canceling, but that's to be expected when you're filming in the middle of a pandemic. Lots of challenges in COVID.

And then you're in the airport flying around everywhere, and you're in crowds every day. And you're just wondering, when am I going to get COVID?

CRAIG BOX: How am I supposed to tell Tim Hockin and Brian Grant apart, when they're wearing the same t-shirt?

JOSIAH MCGARVIE: Yeah, that's so funny. We tried to switch up the backgrounds every time we filmed, to make it dynamic. It was day one, and I think the first person we filmed with was Tim. So we're in the offices. We set up this cool background. Tim shows up, and he goes, check out my shirt. It was a special Kubernetes shirt from a special occasion. And he sits down, and we do the interview like, oh. very cool.

We didn't have time that day to change the setup. So we were going to film the second interview in the exact same setup. And then Brian comes in, and he's wearing the exact same shirt. And he goes, check it out. I'm wearing this shirt for the occasion. And we're like, oh. [LAUGHS]

CRAIG BOX: They are very exclusive shirts. Even I don't have one of those.

JOSIAH MCGARVIE: Yeah, so we were like, oh, Tim was wearing the same thing, and they're like, oh, cool. So it's kind of funny seeing both in the same location, both in the same shirt. It looks like we planned it that way. People might have been like, couldn't you have asked them to change the shirt or something? We didn't ask them. That was their Kubernetes pride.

CRAIG BOX: I do happen to know that Tim Hockin has several other Kubernetes t-shirts he could have chosen from.

JOSIAH MCGARVIE: Yeah, well, that was his favorite. You can't deny that.

CRAIG BOX: I think he's got about 100, he said.

JOSIAH MCGARVIE: [LAUGHS] About 100. My goodness. Everyone who we filmed with were such great people to interview and really were generous with their time and with their answers. It was really a good experience.

CRAIG BOX: I will say, I was pleased to see that the Kubernetes Podcast sticker made an appearance at 22 minutes and 7 if anyone's checking out the time code. I think it was probably on Brian's laptop.

JOSIAH MCGARVIE: Brian's laptop, OK, so we know who listens to the podcast.

CRAIG BOX: Everyone listens to the podcast.


CRAIG BOX: Just not everyone has the exclusive sticker. What are you most pleased about now that people have seen the documentary?

JOSIAH MCGARVIE: For the most part, everyone seemed to have received it very well. My biggest fear when creating something like this is, I'm coming as an outsider outside of the Kubernetes community. My fear is people saying, this was not portrayed in a fair way. This was not included. You really totally miss this important part of the story, or you didn't do the story justice.

Storytelling is so incredibly important psychologically for people. It's hard to overstate how important it is, actually. When you have an opportunity to tell someone's story, they're putting a lot of trust in you. Storytelling is framing the reality around your life. Something bad happened and you're getting therapy, often the therapist will say to you, you need to change the story that you tell yourself about yourself.


JOSIAH MCGARVIE: Because the story that you tell yourself about yourself, totally shapes your reality. It shapes who you are. It shapes how you see yourself in relationship to other people. It shapes your goals, completely shapes your goals. And it shapes who you think is in your way. Who you think is your adversaries. It shapes your whole mindset. That's the natural way human beings are. Human beings aren't logical, necessarily. That's an unnatural thing. We as storytellers. And we exist in our own story.

And so when you expand that out and you have a community, that you're telling the story of a community, you have an incredible power in your hands to shape the reality of this community. Sarah Novotny, I think, really gets this and probably on a much deeper level than I do. She really felt that, very deeply that, a community, just like a person, needs to tell a story about itself to itself in order to define itself; define its past, define its future, define its present, and define its values and its character.

As a filmmaker, failing to do justice to the story of Kubernetes. That has huge ramifications. And I really felt while I was doing this, that was my biggest fear. I didn't want to fail at telling the story of Kubernetes.

CRAIG BOX: Someone has told the story of Josiah by way of a LinkedIn recommendation, which I'll read to you now, and ask if it's the story that you want to be remembered by? "Josiah is a real ripper who works chocka for his dingo dollaroos."

JOSIAH MCGARVIE: Oh, my [LAUGHS] The story behind that is, I worked just outside of London. It was like a fintech magazine. Digital fintech magazine. And I was doing video for them. Because I was an Australian living in England, they used to tease me all the time about my accent, tease me about being Australian, and all the things you get. They would also ask me, do you want to get paid in dollaridoos or dingo dollars? It was good fun. The English people love to riff.

CRAIG BOX: That's a Simpsons gag, for anyone who doesn't get it.

JOSIAH MCGARVIE: That Simpsons episode about Australia, they would be constantly quoting that to me. Prime minister, forky spoony, all that. It was so funny.

CRAIG BOX: What's next for Josiah, and what's next for the Honeypot documentary series?

JOSIAH MCGARVIE: I'm freelance now. There isn't really anything set in stone that I can talk about. But I know Honeypot are very interested in continuing this adventure into open-source documentaries, tech documentaries. Maybe, you'll see me making documentaries for them again in the future. This has really started something. Anyone who's listening who's interested in tech documentaries, pay attention to this space.

Because Honeypot has created waves. It's created a domino effect. There is lots of companies out there that are starting to see the success of the open-source documentaries. And they are putting money down. They are looking to have their own documentaries made. I know what people have reached out to me. I know through my circles there's other people who are doing it through this splash. That's massive shout out to Emma Tracey. She created this splash with this idea at Honeypot.

Now, you're going to see more tech documentaries, which I think is amazing. If you're a tech person, wouldn't you want more tech documentaries out there? Wouldn't you want every technology you could possibly work on or use, there's a documentary. There's an origin story. There's going to be a cinematic universe of tech documentaries, where they all connect, these characters, overlapping characters and all that. So it's going to be big. I think in the next few years, we're going to see a real revolution of documentaries in tech.

CRAIG BOX: Do you think we'll get a Linux documentary? And do you think it'll cross over to Netflix if we do?

JOSIAH MCGARVIE: Oh, man, that would be cool, wouldn't it? I don't know. Would some of these big streaming platforms pay attention enough to be like, oh, we want a slice of this cake. We want a piece of this pie. I don't know, maybe.

CRAIG BOX: They famously will greenlight anything.


CRAIG BOX: Stereotypically, at least.

JOSIAH MCGARVIE: I think we've all noticed that they'll greenlight anything, yeah.

CRAIG BOX: All right, well, thank you very much for joining us today, Josiah.

JOSIAH MCGARVIE: Awesome, Craig. Yeah, thanks so much for having me. It was great.

CRAIG BOX: You can find Josiah on Twitter @josiahmcgarvie, and you can find Honeypot at honeypot.io.


CRAIG BOX: Once again, we find ourselves at the end of another episode. If you enjoyed this show, please help us spread the word and tell a friend. If you have feedback for us, you can find us on Twitter @kubernetespod or you can reach us by email at kubernetespodcast@google.com.

You should also check out our website at kubernetespodcast.com, where you will find transcripts and show notes, as well as links to subscribe. Please also consider rating us and your podcast player, so we can help more people find and enjoy the show. Thanks for listening, and we'll see you next week.