Confluent Developer ft. Tim Berglund, Adi Polak & Viktor Gamov
Hi, we’re Tim Berglund, Adi Polak, and Viktor Gamov and we’re excited to bring you the Confluent Developer podcast (formerly “Streaming Audio.”) Our hand-crafted weekly episodes feature in-depth interviews with our community of software developers (actual human beings - not AI) talking about some of the most interesting challenges they’ve faced in their careers. We aim to explore the conditions that gave rise to each person’s technical hurdles, as well as how their experiences transformed their understanding and approach to building systems.
Whether you’re a seasoned open source data streaming engineer, or just someone who’s interested in learning more about Apache Kafka®, Apache Flink® and real-time data, we hope you’ll appreciate the stories, the discussion, and our effort to bring you a high-quality show worth your time.
Confluent Developer ft. Tim Berglund, Adi Polak & Viktor Gamov
AMA with Tim Berglund | Streaming Audio Special
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Happy 100th episode of Streaming Audio! Thank you to everyone who has listened, subscribed, left a review, and mostly, for sharing our passion for event streaming. We can't wait for the next 100! To celebrate, Ben Stopford (Senior Director of the Office of the CTO, Confluent) hosts an AMA (ask me anything) with Tim, covering 62 questions in total—from his career, his time at Confluent, Marvel vs. DC, and what he looks for in a new hire, to how to nail your next conference talk.
We hope you enjoy this special 100th episode of Streaming Audio: a podcast about Apache Kafka®, Confluent, and the cloud.
EPISODE LINKS
- The Song of the Strange Ascetic
- Avoiding Lock-In
- Blogs by Ben Stopford
- Join the Confluent Community Slack
- Learn more with Kafka tutorials, resources, and guides at Confluent Developer
SEASON 2
Hosted by Tim Berglund, Adi Polak and Viktor Gamov
Produced and Edited by Noelle Gallagher, Peter Furia and Nurie Mohamed
Music by Coastal Kites
Artwork by Phil Vo
- 🎧 Subscribe to Confluent Developer wherever you listen to podcasts.
- ▶️ Subscribe on YouTube, and hit the 🔔 to catch new episodes.
- 👍 If you enjoyed this, please leave us a rating.
- 🎧 Confluent also has a podcast for tech leaders: "Life Is But A Stream" hosted by our friend, Joseph Morais.
For the 100th episode of Streaming Audio, I decided I didn't want to interview anyone. Instead, I thought it was time I sat in the other chair and answered some questions myself. Ben Stopford joins us co-host for a rapid-fire conversation on today's special episode of Streaming Audio, a podcast about Kafka, Confluent, and the cloud. Hello and welcome to the 100th episode of Streaming Audio. And today I am Tim Berglund, but I am not your host. Your host today is Ben Stopford.
SPEAKER_00Hello, everybody.
SPEAKER_01This is a lot of fun for me because I normally do the interviewing, and today I get to be interviewed, so I'm I'm excited.
SPEAKER_00Yeah, you do, and I have uh I have some questions for you.
SPEAKER_01I I have seen the questions and I'm a little apprehensive, but I'm ready. Yeah, 62.
SPEAKER_0062 questions. Yes. So let's go. So what time did you wake up this morning, Tim?
SPEAKER_01Okay, good question to get us get us kicked off. Uh about 4:50. Uh my alarm was set for 5, which is what it's normally set for, but I guess my body just knew it was almost time. So up and at him, time to go get some.
SPEAKER_00So you're a remote employee, so why get up at 5 a.m.?
SPEAKER_01Oh, good question. Um, there's a lot to do. Uh I mean, there's coffee to be made, there's there's uh reading to do. There I kind of like a little quiet time by myself in the morning. Um usually work out uh two, three days a week. Um like to be at the office at eight, even without a commute.
SPEAKER_00Do you like being a remote employee?
SPEAKER_01I love it. I love it. I is I've been almost exclusively remote for 20 years. Um I travel a lot, obviously, when we're not in a global public health crisis kind of situation. I travel about half the time. Uh, but when I'm not home, I I like to be home. Uh and it works very well. So how many countries have you been to? Uh how many countries are there? No. Um I think I think around 30. I've tried to do that. There's a there's a few countries whose visa applications, uh, I think Russia is one of these. They ask you to list all of the countries you've been to in the last 10 years. And I'm like, well, sit down. This is gonna be a bit. Uh but I think it's around 30.
SPEAKER_00So in all your travels, what's your best pro tip for developers who are traveling?
SPEAKER_01Ah, okay. Well, um, it's usually jet lag is kind of where that comes up, is the tough thing. And my secret is just be so exhausted all the time that you can sleep uh no matter when, you know, just fall asleep whenever. That may not be sustainable. That might be the die young path. I'm not sure, but that's kind of what I'm doing.
SPEAKER_00Yeah, that works. So uh who's the last person you emailed?
SPEAKER_01Oh, good question. I should be able to check that. That is something that is just an objective question. Let's go to sent items. And it was my boss, Mark Trang. Very good.
SPEAKER_00New boss.
SPEAKER_01Yes, my new boss, Mark Trang. Welcome to the team, Mark.
SPEAKER_00So, what was your first software job?
SPEAKER_01First software job, I was writing embedded PC software. We called it firmware, you know, it ran from a hard disk, so that's that's a thin claim, I think. But embedded PC software for a VSAT startup in Melbourne, Florida, where I was uh going to school. This was while I was an undergrad. Uh VSATs are a uh like a small low-cost uh satellite terminal that back then you did kind of 64 kilobit, 120 kilobit data rates over them. Um yeah, that was that. Telecoms my first love.
SPEAKER_00So, what was the first programming language you learned back back in that uh good old days of uh VSATS and embedded software?
SPEAKER_01Somewhat before the VSATS, maybe 10 years before that, uh Commodore VIC 20 basic.
SPEAKER_00So meet Apple conference.
SPEAKER_01I well, that's a painful question right now, right? Because you can't do either one of them in person at the time of this recording. Uh and I can't, it's hard to pick. I can't pick between those. I love them both. I think they're both really, really important uh for me and and just for like the Kafka community and for for any technology community. You need to have both those things. I can't I can't pick.
SPEAKER_00So when did you give your first meetup talk?
SPEAKER_01First meetup talk would have been in 2008, I think Christmas time, in Colorado Springs, at the Colorado Springs Open Source User Group, I think is is what it's called. And they're still going, by the way. First conference talk? Uh first conference was in 2009, in October of 2009, uh Prague, Czech Republic. I love that city, by the way. Fell in love. Uh, that's 11 years ago now, and I'm still in love. I want to go back.
SPEAKER_00I know you've done some some pretty interesting and varied talks over the years. So, of all the talks that you've done at all the different conferences and meetups, which is the one you're most proud of?
SPEAKER_01Oh, good question. There's different ways to think about that, right? There's like the um uh which one in terms of like intellectual ambition um and um best execution. So let me kind of think here. Uh I had one called uh Then Our Buildings Shape Us. That's based on a Churchill quote, actually. Um and it was about um the the effect that that tooling and architecture decisions have on subsequent decisions, right? Like how you pick a tool, and that tool tends to shape the kinds of applications you're gonna build or the kinds of things you're gonna make. It was, I think, intellectually an interesting talk, and I liked giving it. In terms of execution, I think um this is disappointing because it's like eight years ago. I guess I peaked early, but I think my best delivered talk ever was a git talk in 2012 at JavaZone, uh, where I uh using a bunch of git plumbing commands, uh live coded a git commit without using git init or git add or git commit or any of those top-level commands that did it all by hand. It was just something I had fun with. You know, it's just like, hey, I'm gonna be on stage and we're gonna screw around, have a good time together. So do you think the best talks have demos in? Um that's hard for me to say. Now, I know again, it depends on the kind of talk. Um live coding is controversial among you know sort of the professional speaker set. Uh you have people who hate it and think you should never do it, and then you have people who do it, do it exclusively. Uh you have a guy like a Venkat Supermanium who he just live codes and tells jokes, and everybody does a great time, has a great time. I think if you if you divide talks into two categories, we'll call them API talks, you know, about the specifics of how to how to code a thing, and then big idea talks, just those two broad categories. I think among API talks, yes, the best talks have demos. But I think there are some very compelling big idea talks uh that would never have demos, but are still really awesome things. You know, the big idea talks, it's it's it should never veer into TED Talk territory if it's about technology, but it's it's more there than um, you know, here's how to optimize the Kafka producer or whatever.
SPEAKER_00So is there like a uh a goal, a speaking goal? Because you're like probably one of the best technical speakers in the world. I think that's fair to say. So is there like a a goal that you still have, or do you think you've uh have you done it all now?
SPEAKER_01Oh, I definitely haven't done it all. Um I um yeah, that's that's funny. I I feel like I should have a good uh quick answer to that. My goal is to keep doing it. Um and we're all in this in terms of live events, we're all in this situation of being paused right now. And uh, you know, people who are inveterate speakers like me are like kind of anxious to get back in front of people someday. Um I want to keep doing it. Um there isn't necessarily like an audience size that I want to hit or a venue or anything. And there are people I look up to. You know, I've mentioned Venkat's name before. Um I uh I I you know he and I have very different styles, and I don't want to become him, but I think, wow, could I could I be that good someday? So I have those aspirations, but I don't really have a goal as such.
SPEAKER_00Yeah, fair enough. So uh which do you prefer? Video or podcast?
SPEAKER_01Ooh, good question. Here I am on the podcast. You know, you're divided to be faithful to this microphone. I have to say podcast. I love podcasting so much. Um I was just reflecting earlier this morning. Wow, a hundred episodes. And that that's not all of me, that's Gwen and me together. Um, but that's we've been doing this for a long time, and I love doing it. I still think I love video more. I just the the studio lights and everything, the kind of the glow of the room with that you know blue daylight-colored studio light that that is typical in studios these days, it just makes me happy. You just want to be on TV. I just do.
SPEAKER_00I just I just want attention, really, Ben, is the thing. That's what it gets down to. So being such an accomplished speaker on all different media, uh, what's the best piece of advice that you would have for new speakers out there who might be listening to the to us now?
SPEAKER_01Awesome. So there's uh one thing I say there, and that is decide that you belong on the stage. Um because people uh even even people who are passionate about speaking and like driven to do this, um, they still struggle with this idea of am I good enough? Do I really know enough? Am I going to be able to answer the questions? What if it doesn't go well? What if they don't like me? Um, you know, there's that lonely little boy or girl inside their mind just wanting to be loved. Um like I get all that, and that's all very real. And there isn't anything that just makes that go away. But um, you have to make a decision that you are the one who belongs on the stage right now. And your reason for being on the stage is not to be seen, it's not to be famous, it's not to have people tell you you're cool, it's to serve the people in the room. Those are people who have voluntarily decided to be in that room. And uh your mission is to serve them, is to give them something. Uh so don't worry about whether you belong. You just have to decide, yes, you do, and you have this work to do, and that work is serving those people, and just do that work. And like you can figure out later whether you were good at it. But just get there, you belong there, do the job.
SPEAKER_00Thanks. That's good advice. So yeah, one of the uh big trends of the last few years now is uh microservices, and that's something that uh we kind of get involved in a fair bit of confluent. So what's your favorite thing about microservices?
SPEAKER_01Uh the hype. Um no, just kidding. Uh evolvability, if I could put it in one word. I think that's the aspiration. And um, I think microservices plus event streaming uh actually offer the promise of credibly evolvable architectures.
SPEAKER_00Least favorite thing?
SPEAKER_01Uh also the hype. And I this is very tendentious, and I won't apologize for that, but um, I think RPC-based integration, I I know people are successful with it, and and there are all kinds of successful deployments, but when you look at the you know the patterns that are evolving around how to deploy microservices at scale and the tooling and things like service meshes, um, and again, there are you know, there are there are good accounts of service meshes, they're not like this unqualified evil, but there's all this complexity that grows up around synchronous integration between services. So I think, I hope that we're learning that lesson, getting away from that, and we can have lots of little programs that talk to each other through events.
SPEAKER_00Command line or GBI? Uh command line. Wait, what about uh confluent control center?
SPEAKER_01Uh yes. Uh well, I didn't say command line for everything. I mean I use IntelliJ too. So I'd there it's it's perfectly fine to I have I have all kinds of graphical tools. I mean I I also make videos. I use Premiere Pro for that. That's not a command line tool. But I my I think my impulse is is when I want to get something done uh where there's a command line option, I I probably use command line.
SPEAKER_00Okay, Marvel or DC?
SPEAKER_01Marvel. Marvel, oh my goodness.
SPEAKER_00Thor or Iron Man.
SPEAKER_01The truth is I can't finish it. Um Iron Man, I'm pretty sure that would be anyone anyone who knows me would say Iron Man. Uh Infinity War or Endgame? Oh, good one. I am getting to the point where the feelings are not so raw and that I can start to talk about it. I still don't know. I've only been able to watch Endgame twice so far. Uh it's difficult. It's not for so we there are other movies that that make you cry. I mean, there's La La Land. I can watch that like every other month. Endgames, I just it's I'm working up to it. So I'll have an answer for that soon.
SPEAKER_00So on a more serious note, what's your favorite poem?
SPEAKER_01Ah, good question. Uh The Song of the Strange Ascetic by G.K. Chesterton, your very own countryman. Um, why is that? I oh wow, probably shouldn't get into that too much here. I know people like that. And uh you can Google, maybe we'll even link to it in the show notes, no promises, but you can Google the poem, The Song of the Strange Ascetic. And it's a it's a kind of a light-hearted but tragic story of a person who uh could love life but doesn't.
SPEAKER_00That sounds intriguing.
SPEAKER_01Can you recite it? I can. I I am not going to on this podcast, but absolutely that's one of the poems I've memorized.
SPEAKER_00Okay, well, thanks for that. Um so what's the biggest gotcha for people learning Kafka?
SPEAKER_01I think it's a mental one. Um uh thinking of it as a message queue, uh, because it it looks like a thing that they've seen before. And so you take all those uh mental categories and just import them into your thinking about Kafka. And so, oh, it's a message queue. Okay, I'll think about a message queue. So isn't it a message queue? Well, no, Ben, it's not. And I would say just you know, the the short answer to that is that queues generally don't offer persistence as a first-class feature. Um, and that's one of the one of the small number of architectural decisions that Kafka made that just have these broad-ranging implications. Uh, you know, it's a it's a system of record, and nobody has ever conceived of a queue that way. So, what's this Kafka tutorials thing? Ooh, glad you asked. So, Kafka Tutorials is this uh thing, it's a thing that my team uh contributes to, the developer advocate team contributes to uh quite a bit. It's a set of executable documented, and they're available at developer.confluent.io, which I think is fun to say. Developer.confluent.io. I like saying there. Go there after this podcast is done, go there. Uh but yeah, so uh it's good to have sample code, right? But you also kind of need to be walked through the sample code. Uh and so Kafka Tutorials gives you this step-by-step thing that says, well, here's the build file you're going to use. Uh, here's your KSQL code or your Java code, and here's your test data, and you know, just paste these things into files. And it actually makes you do that. So you're you're active in copying and pasting and creating this little program. And the back end is super cool. There's this CI uh integration setup and an automated test suite so that the tutorial code you're running can't be on the website unless its tests have passed. So like they're they're testable, nice ways to learn the simple things that you want to learn when you're just starting to kick the tires.
SPEAKER_00So one of our favorite technologies at the moment is case db. Uh, what's something that most people don't know about case db?
SPEAKER_01Well, there are some recent changes. Like there's a REST API. Uh you can use that REST API to query tables. If you use KSEQL to say aggregate a stream, group group a stream on some field and compute an aggregation, the result is a table. You can actually pick rows out of that table through the REST API. What's a good book you've read recently? Uh recently read The Poetics by Aristotle. Um currently reading a book called Redemptive Leadership by Harve Powers. Good books. What values do you look for when you're hiring someone? Uh a lot of things, but if I had to summarize it briefly, I'd say smart, humble, and empathetic.
SPEAKER_00Good values. Good values, good values. So what do you think the biggest misconception is about being a developer? Ha, well.
SPEAKER_01Um it's funny who I'm talking to right now. Uh this I wonder if this one strikes you too. But there's a stereotype that uh we're all socially awkward and sloppy dressers and things like that. And it's it's a profession in which um you in fact can be socially awkward and can be a sloppy dresser and be uh 100% successful, right? You're not it does not select for attributes like that, unlike say sales, you know, where you need to be a sharp dresser and and you need to be great with people and all that. So like we don't care if you are, um, but I think that leads to the stereotype that uh you know those qualities never obtain. And I'm not a sloppy and you, Mr. Stopford, are most certainly not a sloppy dresser. Um so yeah. Thanks.
SPEAKER_00Um, do you ever watch your own videos?
SPEAKER_01Yes, yes, I do. Uh funny story here, it was so painful. Uh it was like eight or so years ago, a friend of mine named Matthew McCullough and I, we made a Git tutorial series on O'Reilly, and we had to watch all six hours of it. And it was just like, you know, don't you make sure that your cyanide capsule is in another room because you will take it. You will do what it takes to end the pain. Um, it's so so bad. But after like maybe 20 hours of that kind of thing, uh the pain faded. I kind of like it now.
SPEAKER_00Yeah, I can't do it. I think I've watched one that was good. I thought it was good. I'm never gonna say which one it was, but I thought it was good. All the others I thought were rubbish. Um but uh yes. So is that is yeah, is is your ability to re-watch or watch yourself on video, is that just because you're so good?
SPEAKER_01How can I answer that question? I don't think so. I think honestly, that has a lot more to do with there's just a familiarity, you know how people normally, oh, you know, I hate the way I sound on on uh uh my my recorded voice sounds. You know, you get used to that. I I hear my recorded voice all the time and it sounds fine to me. People hate the way they look because they're self-conscious, because they're thinking about themselves and they're not used to looking at themselves. Um, but I really think it's just a process of of just getting over yourself and realizing that this is the you that everybody sees and hears all the time, and most of them like you well enough. Um so I don't I don't think it's got anything to do with skill or me being better at it than I used to be. I think it's just uh it's a if you want to be a professional at this, you have to get over that. You have to get and it's a process of getting over yourself.
SPEAKER_00So um on that note, how did you get into developer relations?
SPEAKER_01Oh, yes. Well, um it was a gradual thing. And when it started, it we didn't even use that word very broadly. But um I was, I guess the transition was, you know, I was a full stack web developer, and I kind of operated as an independent consultant, sort of coder for hire, uh, and that drifted into coaching teams in a new stack, trying to do technology transition, and that drifted into training. Um and I thought, hey, I like that. I'm I'm actually a good teacher. And as a guy on my own, uh as a you know, independent sort of freelancer, I realized well I could make more money in a day training if you're helping 15 people be better, and that's more valuable than just writing code by yourself. Uh and then I I started to catch on to some of these conference speaker people that I knew and admired, mentioned Venkat, guys like Scott Davis and Ken Sype, and um just these folks that I thought, I want to be like that, you know? And so I started doing some speaking at meetups, and I got my first conference gig, and I realized I love this. It was just like electrifying. Um I eventually got a gig as a trainer. I worked for GitHub for a little while as a Git trainer and then uh moved on to another company where that training sort of began to evolve into a more full throated vision of developer relations, which is what kind of what I came to confluent with three years ago. You know, like let's let's uh not just from a training perspective, because I I never owned well, I didn't own training for long here, um, but from From a community perspective, uh advocacy perspective, like what are all these things that have to happen to do that? So that that that my thinking sort of evolved along those lines.
SPEAKER_00So what does a good developer relations team do?
SPEAKER_01Yeah, I I I almost was kind of starting to answer that, wasn't I? Um a few things. So um fundamentally um they they teach, they inspire, they connect. Uh that's that's what the team should be doing. Now there's lots of channels that that can take place through, um and lots of forms, but uh connecting. This is why we have meetups under normal circumstances. We want people want a time to get together physically and share a meal and talk to each other and then learn some stuff. Uh that that connecting isn't. But that can happen online, that can happen in Slack, that can happen in Stack Overflow, there are all kinds of social venues there. So you you need to get people connecting, knowing each other, and helping each other. Um there's also teaching, right? And this is not not usually formal training. Um, that's a different and kind of a well more formal thing, but uh there are always the basics. And a devrel team is focused on getting you started. Uh people need to learn what is the mental model I should have, what is a Kafka, how should I think about it? Uh, and how do I twiddle the bits on these APIs? I need to know that. I need code. Um, but more than just here's the information, a good DevRel team should also say, you know, here's the information and here is the kind of person you can be. Like you've got aspirations as a developer, we can help you realize those. Uh, because we tie up our identities with technology choices. You know, you're you're not just using Kafka because you did a bake-off and it its figure of merit was the highest. It's because it's the thing you want to be associated with. And a good DevRel team uh builds a community that is a kind of a winsome thing to be a part of that people want to join.
SPEAKER_00So you you're also um you're also actually a pretty good, pretty awesome writer who's known for having a kind of uh a strong voice. Um that I am. Yeah, is there any kind of advice you would give to other developers out there? Because like writing's becoming um increasingly large part of the of the way that people actually do work, uh, particularly as we end up in uh situations like the one we're in now where everyone's remote. Um advice that you would give to people about that and developing a voice and that kind of thing.
SPEAKER_01Yeah, that's a toughie. Um because there's a lot of mechanics that you have to develop. Um, and that there is simply no substitute for the beatdown, right? Um, Mr. Gherkin in 11th grade, Phil, thank you for the beatdown. I've told him this on Facebook, by the way. I've been able to connect with him. Um that you know, sort of early formative influence I had on my writing, um, to beat the garbage out of it and make it something concise and good and meaningful. Everybody needs to go through that process. So the the mechanics of of just being good at writing stuff uh is step one, and that's long and painful, and just do it. Um, and and you know, submit yourself to the mentorship of somebody who can help you. Other than that, um you you gotta just, again, this comes down to a fear thing. Like there's a you in there, and there's a way that you talk when you talk to your husband or your wife or your friends that are close. Um, there's a relaxed you. And a lot of people think, well, when I'm doing formal business or technical or work communications, I have to put all that away so I can be taken seriously. And, you know, because somehow now my audience, well, there might be people in a boardroom reading this, and people in boardrooms are boring and don't want to laugh and are bereft of life and are these ghoulish husks of souls long dead. No, they're not. No, that's true. You know, so in to develop a voice, you have one, just don't be afraid to use it. And you gotta take chances sometimes. You might say things that that you might make jokes in in business communication where that's not normal. Um, but you'd probably say that in real life, and as long as it's not something that's like inappropriate for for work uh you know, on some other level, just be you. You know, you gotta be able to express yourself well in writing. Again, that's step zero, but don't be afraid to be yourself.
SPEAKER_00I think that's good advice. Um so for your like uh your talks that you do, obviously do so well. Um is there like a is there a structure? Is there a a secret source behind the uh the Berglin method? Uh how does how does a a great talk come together?
SPEAKER_01Uh so I'll again I'll try to be brief, but I'll take that two ways. There's so many ways a talk comes together. Number one for me, usually late. Uh I just, you know, my own attention deficit issues being what they are, I have to create the crisis of uh there's no way I can finish this in time in order to be able to focus. So I usually build things late. That's part of my process. And I that's not, by the way, if you also do that, uh, anyone listening, uh, don't let anyone tell you that that's a character flaw or that's procrastination. It could be, in your case, you know, you could be just a slothful person. I doubt you are. Um there's all kinds of reasons for doing things at the last minute that aren't that aren't sloth. But um the form that I'm trying to conform to in every talk is not mine. It's I mentioned Aristotle as a book I've read recently, The Poetics. It's an Aristotelian idea, which is that there's a three-act structure to any good talk. The first act is um, and I'm I'm applying Aristotle to talks here, but you want to talk to the audience about problems they have. You want to connect with them and say, hey, this is what's wrong with your life. Um, you know, in the context of what you're gonna talk about. But you gotta be saying, I know something is broken and hard for you. And then you exposit your material, you know, here's how you use Case Equal, or uh here's how consumer group rebalancing works, or you know, whatever it is, or here's how to think about microservice architectures and event streaming. Uh and then at the end, you take that technical knowledge that you've delivered and apply it to those problems you started with. And so you say, you know, your life was broken, I'm gonna teach you some things. Hey, wait, let's look at your life again. That broken thing is fixed, and you'll never be the same again. That's the the story formula.
SPEAKER_00Yeah, that's great. So, other than your writing and your speaking and your videos and uh your general you know running of a devrail department, do you have another secret talent? You know, I really don't.
SPEAKER_01Um I don't have a lot of talents, and the ones I have are pretty public. So, no.
SPEAKER_00What about like uh we were talking about your your presentations? Do you have any uh pre-presentation rituals that you kind of go through to prepare yourself for being, you know, uh owning the stage?
SPEAKER_01Right. Uh getting the demo to work, uh finishing the last slide, it's usually things like that.
SPEAKER_00So what runs through your mind at that time?
SPEAKER_01Is the demo going to work? And if it if it's a talk without a demo, uh nothing. Uh at that when I'm getting on the stage, I it's this game time thing. I just let go.
SPEAKER_00Uh so what's your worst ever on stage experience?
SPEAKER_01Oh, um, you know, I've never failed. I've never had a demo just not work where I had to walk off. Um I don't know. I don't know. I had a very narrow miss last uh December. I won't mention the event, but uh was literally uh fixing a Docker file two minutes before we began. That was pretty scary.
SPEAKER_00So it's the biggest stage that you've presented on.
SPEAKER_01I think that was actually Kafka Summit San Francisco last year, 2019. It's about 2,000 people.
SPEAKER_00That's uh that's a lot of people. So what did that feel like?
SPEAKER_01I love it. Um it's as far as I'm concerned, the more the merrier. And Summit historically is like one of the my favorite things I do all year long.
SPEAKER_00Um I heard you made a builder bear with the confident community catalysts. Uh, how is he doing?
SPEAKER_01He's doing well. And if you're outside the United States, I I don't know how international that brand is, but it's a retail store where you go and there are these blank stuffed animals, not stuffed with anything, and you go pick the one you want, and they have this machine where they stuff it and you put like a little heart in there that beats when you squeeze it or plays a recorded voice message when you you hug it. So I when my girls were younger, I spent untold hundreds there. But yes, we did that, went to build a bear with the catalysts. Uh, we're gonna do that again the next time we can meet together. And uh you may see him in the background of some of my videos on YouTube, the ones filmed in my office.
SPEAKER_00Ah, that's what it is. Yeah, I was not being American, I was struggling with this one a little bit. Um so so uh yeah, how would you describe Kafka Summit?
SPEAKER_01Uh again, the best. And I realize I'm highly biased there, but I love that conference and I love being there, uh seeing so many friends, looking forward to when we can make that a thing where we get together again.
SPEAKER_00So, what's the scariest part about building systems in the cloud?
SPEAKER_01To me, lock-in to a single cloud vendor. Um lock-in is a complex topic. Um, there was an entry on um Martin Fowler's blog a few months ago. I think it was written by somebody else, not Martin himself, but it was on his blog. There was this really good analysis of lock-in. We should put that in the show notes. Um But I think cloud vendor lock-in is is a risk. Uh, it scares me. Um again, that one could consider that answer to be tendentious from someone who is a cloud vendor, but uh that's that's a risk.
SPEAKER_00So open source, proprietary, cloud, is it the same?
SPEAKER_01Cloud is changing that game in so many ways, right? Because anymore uh you want an API and and it hits your credit card, you know, you just want to do stuff. Who cares what the license is? I still think open core is uh the most meaningful model. I think there's a lot of security that people have in knowing that um even a proprietary product is built on an open source foundation and the the core APIs and the core semantics of the whatever it is, uh you know, you have that mental escape hatch. If I just need to go back to open source, I can. Uh and so you you you're it's easier to make the adoption decision. There's less lock-in. Um but there's always proprietary features. And when it comes to cloud, uh who cares? It's a cloud service.
SPEAKER_00Yeah, so because I know you're you're a big uh open source advocate. You're uh one of the sort of louder members of the of the of that community. Um do you think so do you think the shift to cloud is gonna sort of damage open source? Or do you think it's something that will just it's sort of unrelated and it'll just continue anyway?
SPEAKER_01Oh wow, that's such a potentially radioactive question. I I don't I it matters less, okay. It matters less. Again, what you want is an API where you put your credit card in and things happen. And I understand for larger enterprise engagements, these things are not built on credit cards, but um that's that's license matters less in the world of the cloud. I also think uh, as you and I have spent quite a bit of time on uh in in public and private conversations, um, the cloud changes uh how we think about open source licenses and what all that means. That that equilibrium developed 20, 25 years ago pre-cloud. And now in the presence of the cloud, I think we are in the process of rethinking what all that means. But don't hear me saying open source doesn't matter. It matters profoundly.
SPEAKER_00So from one uh slightly controversial question to another, uh, is Kafka a database?
SPEAKER_01Yes, yes. We're we're well into the interview now. He's tightening the thumbscrews. Uh no, I mean, anybody who asks me that, I just say, have you ever really used a database? Uh I mean, no, okay, it's it is not. Um I think of it as a database construction toolkit. And honestly, Ben, a lot of my thinking on this has been shaped by stuff you've written. And I know you've been shaped by um by you know other thinkers in the space, but um, I think database construction toolkit is a good way to think about it, and we should link to some database inside out articles by you.
SPEAKER_00Yeah, or Martin Kletman or uh, I think uh Jay was the original uh um sort of thought leader behind that.
SPEAKER_01Um I thought it was Martin Klepman. Okay.
SPEAKER_00I think it was I think Jay, I think uh yeah, technically it's Jay.
SPEAKER_01Um uh that thing where you disrespect the intellectual contributions of your CEO on his company's podcast.
SPEAKER_00I did it.
SPEAKER_01Ladies and gentlemen, you just heard it.
SPEAKER_00I think uh Martin actually gives Jay that credit in his his book, as I recall.
SPEAKER_01Nice, okay. Well, Martin, you are a better man than I am. Jay, I'm sorry.
SPEAKER_00We'll get that linked. Martin's definitely better at sharing the idea with the world. Well, in some ways. In some ways. Um But yes, okay. So what's uh let's say that you're uh a new developer and uh you're getting started with Kafka. Uh what's the best what's the best place to start?
SPEAKER_01Ah uh without question, developer.confluent.io. Uh it's a fairly new, I mentioned it before, fairly new site, but that is uh Confluence Developer site, where basically what what the Confluent Developer Relations team does is reflected there. Do we have meetups? Well, that's where you're gonna go find them. Do we have a podcast? Well, okay, you're listening to this, you already know about that, but uh there'll be the index of podcasts there. Tutorial videos, they're there. Kafka tutorials, same thing. So wherever you are in your journey of being aware of Kafka, trying to under build a mental model of it, trying to play around with it, getting help actually building things, um, that developer.confluent.io.
SPEAKER_00That sounds great. It is actually very cool. I think it needs a pattern sign just to like round it off. But uh who would write such a thing? I don't know. It's almost like someone tried and then ran out of time. Um we're gonna get there, don't worry. So uh if you had an infinite amount of time, uh, what's like your dream Kafka demo that you'd want to build?
SPEAKER_01Yeah, oh, this would be great. Everybody, you know, uh talks about how they have all this extra time because we're all uh in lockdown and I don't have any extra time at all. I don't know what that life is. But if I had that time, I would I do this movie demo thing where I say that you know there's this mobile app where you can rate a movie while you're watching it, and you know, there's millions of people using it. You have all these movie ratings streaming into Kafka, and then Kafka Streams or KSQL doing interesting things to process that data. Uh I would actually build that with a front end and everything.
SPEAKER_00That sounds pretty cool. So um yeah, I think and you also see like a lot of you meet a lot of customers on your travels and uh a lot of people at conferences. Uh, what's like the coolest KSQL DB application that you've come across?
SPEAKER_01Uh yes, that would uh have to be Children's Hospital of Atlanta. Uh they were processing neonatal and pediatric intracranial pressure monitor data in real time. It turns out that those things throw off a lot more data than you'd think. You might think like, well, how often do you want to sample that? But uh there's a lot of data there, and they were processing it to predict uh bad outcomes. And apparently uh, you know, morbidity and mortality were actually reduced through that application. Um we'll link uh the the guy who ran that uh has been a previous episode on streaming audio. We'll link to that.
SPEAKER_00So uh yeah, you've been at Confluent now for about three years. Um what's your favorite part about working at the company?
SPEAKER_01Um this is a cheesy answer, but my team, honestly, they're really great people.
SPEAKER_00Yeah, they're pretty cool, uh have to say. Um okay, so like uh you're also like a bit of a photographer. Uh you know, you've got you like to get your camera out. I know you like a bit of video too. Um what kind of photography do you like most?
SPEAKER_01My uh so yeah, when it comes to photography, the funny thing is uh I'm I'm still trying to make videos. So time lapses are my favorite kind of actual still photography. Um I I count that as still photography because you use a still camera for it. But some people call them Tim lapses because I do them a lot. But uh just two days ago in what I think was a uh uh a lawful uh sortie, I was uh very highly socially isolated from every socially distanced from everyone. Um I took a time a drone time lapse of the sunrise over downtown Denver. It was very pretty.
SPEAKER_00Yeah, I saw that was very cool. So, what's the one thing you wish you knew at the age of 22? Oh man.
SPEAKER_01Uh well that question could get dark. But let's go with the I I wish I knew what I was actually good at.
SPEAKER_00That's a pretty good answer. Uh what are you actually good at? No, I'm just kidding.
SPEAKER_01I think we've uh ask me the question of uh what I wish I knew when I was 47. It might be the same answer. Uh so who's a person that you'd love to get a coffee with? Uh Fred Brooks. Um, if it could be anybody, uh he is the uh lead of the OS 360 project back in the 60s when um IBM was developing System360, author of Mythical Man Month. Yeah, that would that's a pretty good one.
SPEAKER_00So let's say uh you're you're you're a young person and you want to get into software development. Do you think you need to get a college? That's a pretty controversial one. There's a controversial question.
SPEAKER_01Um there's a I mean that's that's complicated, and like I'm not your life coach, and you shouldn't be taking this kind of advice from me in a podcast completely divorced from the knowledge of any other circumstances of your life. This is stuff that you work out over coffee or beer, not uh over a podcast. But um I think it's very possible not to. And uh it's funny, we you know, you and I work at a company with a tremendous number of PhDs, uh, and they they never they don't flaunt it, right? You like find out you work with somebody for a year and a half, like, oh, you're a PhD too. Oh, okay. So nobody wears that at Confluent, but there's all these very highly educated people. And here I'm saying, eh, whatever. Um, but I I do think it has become a live option uh to become a developer without a college degree.
SPEAKER_00So what do you think the route into that then is? If it's not like comp si at a university, then what is the route in?
SPEAKER_01Um I I kind of think the coding boot camps are are pretty promising. It seems like they're working out. Um and and you might say that there's a particular kind of developer that they're they're creating. You know, that's like the full stack or front-end web developer, enterprise web developer person. Uh and I think they all very much are oriented towards that highly employable set of skills. Um so if that's what you want, like you want to earn, you're a young person and you love writing software and you do it on your own, and you think this is amazing, and you're trying to make this decision about life, I think that that looks like a pretty promising option. Like you're gonna get your foot in the door and probably get a job writing code, and you know, you'll go from there.
SPEAKER_00Okay, so let's say you've had like a really rough day, you've written some talks, you've done some coding, you've had some some demos that just you know, just keep not working. Bunch of calls. Yeah, uh too many meetings, you know. You just need to sit down and and relax. What is your favorite, what's your favorite comfort food just to sit back and relax with?
SPEAKER_01Ah, well, that depends on the country I'm in. Um, but in the United States, it is a burger and fries. And um I uh usually try to eat low carb. You know, I travel a lot, so you gotta manage that kind of thing. But uh so I'll go no bun on the burger and sweet potato fries if possible. So that is my version of burger and fries, and it's my go-to.
SPEAKER_00What about in the US? What about in the UK?
SPEAKER_01Uh fish and chips and a pint. Um carb abatement is a fool's errand in the UK. Uh at least in the big cities? Uh just go for it. But no, it's I've spent enough time in London that um fish and chips is absolutely a go-to comfort food for me, and it just makes me happy.
SPEAKER_00Okay, what's the strangest thing you've ever eaten on your travels? Uh-huh. Um I ate fried crickets in Chiang Mai.
SPEAKER_01Wow. What did that taste like? Uh strangely like potato chips, really, or crisps, as your people would say. Uh, except crisps don't have legs that get stuck in your teeth, and fried crickets do.
SPEAKER_00So, Tim, did you see that thing where Conan crashed that Zoom meeting at Tibco the other day and mentioned Kafka?
SPEAKER_01I did. I did. I just saw some tweets about that this morning.
SPEAKER_00So if he cat if he crashed a confluent meeting, what would you say?
SPEAKER_01If Conan, Conan the Sumerian, black-haired, sullen eyed, sword in hand, a reaver, a slayer with gigantic melancholy, gigantic mirth, came to a zoom meeting to tread the jeweled thrones of earth under his sandal feet. Uh no. I don't mean that Conan. I mean uh That's the Conan Okay. Conan O'Brien. So all these times when people tell me I look like Conan, do they mean Conan O'Brien and not Conan the Barbarian?
SPEAKER_00Uh yeah. So I think the penny's finally dripping.
SPEAKER_01That is potentially very disappointing. Okay. Well, if it I mean if uh Conan uh the barbarian, I that that would be easy. Uh if it were Conan O'Brien who crashed a Zoom meeting, I would uh boy, what would I say? Uh I'd ask him if he'd heard any jokes, and then I would tell him some. I would I would give him some one-liners. I'd say, uh, like, hey Conan, do you hear the one about the sandwich who walked into the bar? And he'd probably just stare at me. And I said, the bartender said, I'm sorry, we don't serve food here. And like, I wonder, I wonder if he'd laugh. We'll never know.
SPEAKER_00Uh well, thank you very much. My guest today has been uh Tim Berglund. Tim, uh, thanks for being part of Stream Your Audio, and thank you for letting me ask you your 67 questions, and thank you also for dealing with me, changing a few of them as we went through. Um, so yeah, my name is Ben Stutford. My guest today has been Tim Berglund, and this has been Streaming Audio.
SPEAKER_01Ben, thank you so much. It's always a pleasure to talk to you, and this has been fun. And there you have it. I hope this podcast was helpful to you. If you want to discuss it or ask a question, you can always reach out to me at TL Berglund on Twitter. That's at T L B E R G L U N D. Or you can leave a comment on a YouTube video or reach out in Community Slack. There's a Slack sign-up link in the show notes if you want to register there. And while you're at it, please subscribe to our YouTube channel and to this podcast wherever fine podcasts are sold. And if you subscribe through iTunes, be sure to leave us a review there. That helps other people discover the podcast, which we think is a good thing. So, thanks for your support, and we'll see you next time.
