Screaming in the Cloud - Deserted Island DevOps with Austin Parker

Episode Date: June 8, 2021

About AustinAustin makes problems with computers, and sometimes solves them. He’s an open source maintainer, observability nerd, devops junkie, and poster. You can find him ignoring HN threads and making dumb jokes on Twitter. He wrote a book about distributed tracing, taught some college courses, streams on Twitch, and also ran a DevOps conference in Animal Crossing.Links:Lightstep: https://lightstep.com/Lightstep Sandbox: https://lightstep.com/sandboxDesert Island DevOps: https://desertedislanddevops.comlastweekinAWS.com Resources: https://lastweekinAWS.com/resourcesDistributed Tracing in Practice: https://www.amazon.com/Distributed-Tracing-Practice-Instrumenting-Microservices/dp/1492056634Twitter: https://twitter.com/austinlparkerPersonal Blog: https://aparker.io

Hello, and welcome to Screaming in the Cloud, with your host, Chief Cloud Economist at the Duckbill Group, Corey Quinn. This weekly show features conversations with people doing interesting work in the world of cloud, thoughtful commentary on the state of the technical world, and ridiculous titles for which Corey refuses to apologize. This is Screaming in the Cloud. This episode is sponsored in part by Thinkst. This is going to take a minute to explain, so bear with me.

I linked against an early version of their tool, canarytokens.org, in the very early days of my newsletter. And what it does is relatively simple and straightforward. It winds up embedding credentials, files, that sort of thing, in various parts of your environment, wherever you want to. It gives you fake AWS API credentials, for example. And the only thing that these things do is alert you whenever someone attempts to use those things. It's an awesome approach. I've used something similar for years.

Check them out. But wait, there's more. They also have an enterprise option that you should be very much aware of. Canary.tools. You can take a look at this, but what it does is it provides an enterprise approach to drive these things throughout your entire environment. You can get a physical device that hangs out on your network and impersonates whatever you want to. When it gets NMAP scanned or someone attempts to log into it or access files on it, you get instant alerts. It's awesome. If you don't do something like this, you're likely to find out that you've gotten breached the hard way. Take a look at this. It's one of those few things that I look at and

say, wow, that is an amazing idea. I love it. That's canarytokens.org and canary.tools. The first one is free. The second one is enterprise-y. Take a look. I'm a big fan of this. More from them in the coming weeks. This episode is sponsored by ExtraHop. ExtraHop provides threat detection and response for the enterprise, not the starship. On-prem security doesn't translate well to cloud or multi-cloud environments, and that's not even counting IoT. ExtraHop automatically discovers everything inside the perimeter, including your cloud workloads and IoT devices, detects these threats up to 35% faster, and helps you act immediately. Ask for a free trial of detection and response for AWS today at extrahop.com slash trial. Welcome to Screaming in the Cloud. I'm Corey Quinn. I'm joined this week by Austin

Parker, who's a principal developer advocate at LightStep. Austin, welcome to the show. Hey, it's great to be here. It really is. I love coming here. It's one of my favorite places to go. So let's get the obvious stuff out of the way. You're a principal developer advocate at LightStep. I know this because I said it a whole sentence ago, which is about the limit of my attention span. What is LightStep, and what does your job mean? So LightStep is an observability platform. We take traces and metrics and logs and all that good stuff,

throw them together in a big old swamp of data, and then kind of give you some really cool workflows to help you make sense of it, figure out, hey, where's the slow SQL query? Where's the performance bad? The way to figure out in most of my environments of where's the performance bad is get blame, figure out what part I wrote. But imagine there were like a thousand or a hundred thousand of you all working on this like massive distributed system and you didn't know how. It would snark itself to death before it ever got off the ground. Yeah.

I mean, I think that's actually most large companies, right? We deliver shippable software only through inertia. Yeah, just because at some point it bounces off all the walls. There's nowhere else for it to go but to production. Yeah. But yeah, you have, you know, thousands of people, hundreds of people, however many people, right? I think the whole distributed workforce thing that most people are dealing with now

has really made observability kind of rise to the top of your concern list because you don't have the luxury of just poking your head around the corner and saying, like, hey, Joanne, what the heck? Why did things break? You can't just poke someone anymore. your head around the corner and saying like, hey, Joanne, what the heck? Why did things break? You can't just poke someone anymore or you can, but you never know what you're going to have to deal with. It feels weird to call them at home or bug their family members to poke them or whatnot. It just seems weird. It does. And, you know, until Amazon comes out with a minder drone that just kind of like hovers over your shoulder at all times and

pokes you, you know, when someone is like, hey, you broke the build, then I think we're going to need observability so that people can sort of self-serve, figure out what's going on with their systems. Cool. One of the things I'm going to point out is that I've had a bunch of people attempt to explain what distributed tracing is and how observability works, and it never really stuck. And one of the things that I found that did help explain it, and we didn't even talk about this in the pre-show while we figure out how to pronounce each other's names, but one of the things that has always stuck with me is the interactive sandbox on LightStep, which used to be prominently featured on your page. Now it's buried in a menu somewhere, but it's an interactive sandbox that sets up a scenario problem you're trying to solve,

gives you data. So it gets away from the problem of step one, have a distributed application where it's all instrumented and reporting things in. Because in a lot of shops, that's not exactly a small lift that you can do in an afternoon to start testing things like this out. It's genius. It shows what the product does, how it works, mapped to the type of problems people will generally encounter. And after I played with this, oh my stars, I get it.

We actually just recently updated that to add some new stuff to it because we shipped a feature called Change Intelligence, where you can take actual time series metrics and then overlay those on traces and sort of say, hey, I saw a weird spike and highlight that. And then we go through, look at all the traces for that service and its related services during that time and tell you, hey, we think it might be this, right? Like here's things that are highly correlated in those time windows. So if you haven't checked it out recently, go back and check it out. It's, yeah, a little more hidden than it used to be. But I believe you can find it at lightstep.com forward slash sandbox.

Yeah, and there's no sign up to do this. It's free access. It asks for an email address, but that's okay. I just used yours. No, I'm not kidding. I actually did. And yeah, it works. It shows exactly what it is.

It even has, instead of start, it says play, because that's fundamentally what it is. If you're trying to wrap your head around distributed tracing, take a look at this. Yes, definitely. I have a longstanding Jira ticket to add achievements to that. Oh, that could be fun. You could bury some, too, like misusing services as databases or most expensive query to get the right answer.

Yeah, and then maybe there's just one span kind of hidden there where it's using Route 53 as a database. I keep seeing that cropping up more and more places. That's something I get to own, and that's an awful lot of fun. Speaking of gamification and playing in strange ways, one of the things you did last year that I wasn't paying attention to, because, you know, there was a pandemic on, was you were one of the organizers behind Desert Island DevOps, which is a strange thing that I've only recently delved into, delven into, gone spelunking inside of. There we go. It wasn't instrumented for observability, but don't tiss. But it's fundamentally a DevOps days that takes place inside the animated world of animal crossings, new horizon,

which is apparently a Nintendo game, which is apparently a game company. Yeah. It is not really my space. I don't want to misspeak. No, you, you,

you hit it. Uh, deserted, deserted. Oh, deserted. Ah, got it.

And don't spell it as dessert either. As in, this would be a delicious game to play. I mean, it is a delicious and comforting sort of experience, right? If you aren't familiar with Animal Crossing, the sort of short 30 second explanation is it is a life simulator of building a game where you, as your character, you are on an island and there are relatively adorable animal NPCs that are your villagers and you can talk to them and they will say funny things to you you can go around and do

chores like picking up fruit or fishing and the purpose is to kind of do these chores get some in-game currency and then go spend that in-game currency on furniture so that you can make a pretty house or buy pretty clothing. And it came out at a perfect time last year because everyone was about to bundle inside for the... Well, we're still inside. But everyone had to go inside and suddenly here's this like, oh, it's just this cute sort of like, eh, you know, putz around and do whatever. It was community oriented. It was more of a building oriented game than a destruction game.

Yeah. It's the sort of thing that is a great way of taking your mind off your troubles. It is accessible to a bunch of people that aren't generally perceived as gamers when you think of that subculture. It really is an encompassing, warm, wonderful thing by all accounts. And you looked at it and figured, all right, how can we ruin something? And the correct answer you got to is let's pour DevOps on it. Yeah, let's use this as an event platform and let's really just tech bro this shit up. And it seems to work super well. At the time of this

recording, I have submitted a talk that I live streamed my submission around, and I have not heard in either direction. To be perfectly frank, I forget what I wound up submitting, which is always a bit of a challenge, just because I make so many throwaway random jokes that, cool, we'll see how it plays out. I think you were even in the audience for that on the Twitch stream. Yeah, you found some bugs on the CFP form that I had to fix. To be clear, the reason I do those things is not because it's a, look how clever I am, but rather to instead talk about how it's not scary to submit a talk proposal. Everyone has a story that they can tell, and you don't need a big platform or decades of experience in this space to tell a story. And that was my goal. And I think I succeeded. You would have the numbers more than I do.

I hope people wound up submitting based upon seeing that. I want to hear voices that, frankly, aren't ours all the time. I think in like a week, we basically got more submissions than we did for the entire CFP last year. You know, one thing One thing that I think is interesting to bring up, because you bring up, oh, we don't hear a variety of voices, right? One thing I tell people, and I know that it's not universally applicable advice, but I got into DevRel as a, not quite luck, but everything in my life is luck on some level. It always plays some level of importance. But I didn't go to school to get into DevRel. I didn't do a lot of things.

I have actually been in tech maybe depending on how you want to count it. In terms of actually being in a software development job or primarily software development job. Maybe five or six years, give or take. And before that, I did a lot of stuff. I was a shorter cook. I worked at gas stations. I did like tech support for BlackBerry. And I did a lot of community organization. I was a union organizer for a little while. I like DevRel because it's like, oh, this kind of integrates a lot of things I'm interested in, right?

I enjoy teaching, helping people, and helping people learn, but I also like talking. I like to go and be sort of a public figure, and I like to build a platform and use that to get a message out. And I think what I did with Visitor to Island, or what the real impetus there was, we suddenly were in a situation

where it's like, hey, there's a bunch of people that normally get together and they fly around the globe in decent airplane seats and people come and see us talk. Because why? Right? Because they think we know what we're talking about or because we have something that shows we know what we're talking about or however you want to say it we know what we're talking about or However, you want to say it? but in a lot of cases, I think people are coming for that sort of community they're coming because Hey, I can go to a room and I can sit in some weird little hotel or conference center or whatnot and Everyone I look at everyone I see is someone that is doing what I'm doing on some

level, right? Like these are all people that are working in technology. They're building things, they're solving problems. And that goes away really quickly when you get into this sort of remote first world and when we can't travel and we don't have that visual aspect. So what I wanted to do with Deserted Island, and I thought what was important about it, is I was already sick of Zoom by the time everyone went to Zoom. I was already sick of the idea of, oh my god, a year or two years of these sort of events and these community things just being like everyone's staring at a bunch of slides and a talking head didn't sound very appealing. So what if we try something different, right? What if we do something where it's like, look, we're going to kind of take people out of their day. We're going to put them in somewhere else. And maybe that somewhere else is just, hey, you're watching

people run around an Animal Crossing island on a Twitch stream. But that sort of moment of just like, this isn't what you would normally be doing, I think takes people's heads out of their normal routine and puts them in a place where they can learn and they can feel community and they can feel like a kinship. I also think it's really important because it's that whole stupid New Yorker joke of like, on the internet, nobody knows you're a dog, right? Like, we have this really cool opportunity to craft who we are as people and how we present that to the world. And for a lot of people, you know, when you're stuck inside, you don't get that self-expression.

So here's a way to be expressive, right? Here's a way to like communicate who you are on a level that isn't just a profile picture or something, right? Or things that don't work as well over Zoom. It's a way to kind of help project your identity. And that, I think, gives more weight to what you're saying. Because when you feel like, hey, this is more of who I am, or this is a representation of me, I can show something about who I am. And that helps you speak, and that helps you deliver, I think,

an effective talk. And that, again, builds community and builds these bonds. I want to talk to you about that, specifically because you are one of those people that aligns very much with my view of the world on developer marketing. But I don't want to lead you too much on this, so why don't you start? Take it away. Where do you stand on developer marketing, and what do people get wrong? I think the thing that a lot of people get wrong is that they try to monetize the idea of community. If you go and you search, insert major company name here, you search Amazon community or you search Microsoft community or you search Google community. Well, if you do that, you'll get no results. But whatever, right? Like you get the picture that marketers in a way have turned the

idea of developer community into something that you can just throw a KPI or throw an OKR on and squeeze it for money. And I don't like that. I'm not very comfortable with that idea of community because I think community in a lot of ways is like family and the families that you like the best are the ones you choose, right? The family you choose is an important concept. Right. And for the most part, so much of human experience and activity is built around finding those people you choose and those communities develop out of that. I use AWS sometimes. I don't necessarily know if I would put myself in a

community with every other AWS user, right? Oh, I certainly wouldn't. This is the problem. Everyone thinks when you talk about community or a group of people doing something, there are other people that are in some level of otherness. And that's, like, there are entire communities around AWS that I do not talk to. I do not see, I do not pretend to understand. Yeah. Even at LightStep, right? We're not a massive, massive company by any means, but we have a bunch of different users that are using our tool in different ways and they all have different needs and

they all have different wants. Right. So I could say like, oh, you oh, here's the Lightswap community, but it's not a useful abstraction. It's not a useful way to abstract all of our users because any tool that's worth using is going to be this collection of other abstractions and building blocks. I don't know, look at something like Notion

or look at something like Airtable or the popularity of low or no code stuff where someone built a platform and then other people are building stuff on top of that platform. If you go to those user groups or you go to those forums and it's just like there's a million, million different varied use cases and people are doing it in different ways and some people are building this kind of application or that kind of application or whatever.

So the idea of, oh, there's a community and we can monetize that community somehow, I'm uncomfortable with that from a base level. And I'm uncomfortable with the idea of the DevRel industry or the developer marketing industry moving towards this idea of, we're going to become community marketers or whatever. I think you have to approach people as individuals. Individuals are motivated by a lot of things.

They're motivated by, can you solve this problem? Do I like you? Are you funny? I believe that if you're a developer tool and you are trying to attract developers, then it works a lot better, I think, to have just individuals. To have people that can help influence the much broader superset of all developers that might have an interest in what you're doing by being different, I guess. Being something that's like, hey, this is entertaining or this is informative

or this is interesting. The world is not a meritocracy. The world is governed by many, many different things. You're not going to win over the developer industry simply by going out and having the best white papers or having one more ad read than your competitor. You need to do something to get people interested and excited and a way that they can sort of see themselves using it, right? It's like, why did Apple go and do ThinkDifferent ads, right? Because you using a Mac, that's kind of like being Einstein, or that's kind of like being Picasso, right? This is basic

marketing stuff that I feel like a lot of technical marketers or developer marketers sort of leave at the door because they think the audience is too sophisticated for it. I'll even soft launch it here because I haven't, at this point in time, talked about it in public. But if you go to lastweekinaws.com slash resources, we wrote our own developer marketing guide because I got tired of explaining this same type of thing again and again and again. It asks for an email address and it sends it to you. I know, I'm as guilty as any. And I, of course, called it dev reliper, which is absolutely a problem with me and how I talk about things. But I'm right. And it goes through an awful lot of what you're

saying. An example that you just talked about of giving people something rather than trying to treat them as metrics. One of the best marketing things I've seen you do? For example, is you wrote O'Reilly's Distributed Tracing in Practice, which means if someone has a question about distributed tracing and how it's supposed to work, well, that's not a half bad resource. And okay, I've read it and I have some further questions. Let me track down the author and ask them,

oh, you work at a company that is in this space, huh? Maybe I'll look into this. And it's a very long-tail story. And how do you attribute that as far as, did this lead come from someone who read your book or not? It'll drive marketers crazy. Oh, it's super hard. And it does drive them crazy. Yeah, my answer is, I don't know, and I don't care. One of the early sponsors of this podcast sponsored for a month and then didn't continue because they saw no value. A month goes by, they bought out everything

that held still long enough, and thank you for your business. Can you explain to me what changed? Oh, we talked to some of our big customers and it turned out that two of them had heard about us for the first time on your show. And that inspired them to start digging into it and reaching things out. But big companies, corporate games of telephone, there was no way

to attribute that. My firm belief is on some level that if you get in front of an audience with a message that resonates and, and this is the part some people miss, something that solves an actual problem that they have, it works. It's not necessarily predictable. And it's hard to say that this thing is going to go big and this thing isn't. So the solution on some levels is to keep publishing things that speak to your audience. But it works long term. I'm living proof of this. Yeah. I think that it makes a lot more sense to, rather than to do sort of, I don't want to say vanity metrics, but kind of vanity metrics around like, oh, this many stars or this many forks or whatever. There's a lot of people, especially sort of in this OSS proximate world,

where you have a lot of businesses that are implicitly or explicitly built on top of an open source project. Not everyone that is using your open source project is going to, one, be capable of converting into a paid user or two, be super interested in it. And I would rather spend time thinking about what is the value someone gets out of this product and even if that only thing is that hey, we know what we're talking about because we've got

a bunch of really smart people that are building this product that would solve their problem. If you want to go out and build your own internal observability solution using completely open source tools, you know, Grafana's and Prometheus's of the world, great, go for it. I'm not going to hold you back. And for a lot of people, if they come to me and say, well, this is what we've got and this is what we're thinking about, I'll say like, yeah, go for it, right? You don't need what we're offering. But I can guarantee you that as it scales and as it grows, then you're going to have a moment where you have to ask yourself the question of, do I want to keep spending a bunch of time stitching together all these different data sources and care and feeding of these databases and this

long-term storage and dealing with requests from end users, or I just want to pay someone else to solve that problem for me. And if I'm going to pay someone else, shouldn't I pay the people who literally spend all day, every day thinking about these problems and have had decades of experience solving these problems at really big companies that have a lot of time and effort to invest in this? This episode is sponsored in part by our friends at Lumigo. If you've built anything from serverless, you know that if there's one thing that can be said universally about these applications, it's that it turns every outage into a murder mystery.

Lumigo helps make sense of all of the various functions that wind up tying together to build applications. It offers one-click distributed tracing so you can effortlessly find and fix issues in your serverless and microservices environment. You've created more problems for yourself? Make one of them go away. To learn more, visit lumigo.io. Oh yeah, we're doing some new content experiments on our site. And what we're doing is we're having some folks write content for us. Now, when people hear that, what a lot of marketers will immediately do is dive down the path of,

ah, I'm going to go ahead and hire some content farm. Well, that doesn't work. I found that we wound up working with individual people that work super well. And these are people who are able to talk about these things because their day job is managing a team of 30 SREs or something like that, where they are very clearly experts in the space. And I want to be very clear,

I'm not claiming credit for our content writers. They get their own bylines on these things. And it turns out that that over time leads to good outcomes because it helps people with what they need. There's the mystical SEO juju that I don't pretend to understand, but okay, I'm told it's important, so fine, whatever. And it makes for an easier onboarding story where there are now resources that I can trust and edit if I need to as things change that I can point people to that isn't a rotating selection of sketchy sites. I think that's one thing that I would love to see more of. Not in any one particular part of the tech industry, but overall.

One thing I've noticed, at least in the pandemic, during this whole work from home, whatever, whatever, we don't talk enough. And it sounds maybe weird, but I think this actually goes back to what you were saying earlier about, you know, everyone having a story to tell, right? Like people don't feel comfortable, I think, putting their opinion out there or saying like, hey, this is what worked. This is what didn't work. And so if you want to go find that out, right? Like, if I wanted to go write something about, hey, these are the five things you should do to ensure you have great observability, right? Then that's going to involve a lot of me

going around and sort of Sherlocking my way through Stack Overflow posts and forums and reaching out to people individually for stories and comments and whatever. And I would love to see us get to a point where we're just like, actually, no, this isn't, you know, we should just be sharing this, right? Like, let's write blogs about it. If you're sitting there thinking, like, no one's going to find this useful, right?

Like, you solve a problem or you see something that could have worked better and you're like, eh, no one else is going to find that valuable. Like, I could almost guarantee you that someone is going to find that valuable, right? Like maybe not today, maybe not tomorrow, but go ahead and like,

write about your experiences, write about the problems you've solved, write about the things that have vexed you and put that on the internet because it's really easy to publish stuff on the internet. Yes. Which is a blessing and a curse. That is very much a double-edged sword. That very much is a double-edged sword. That very much is a double-edged sword. But I think that by biasing towards being more open,

right, by biasing towards transparency and sharing what works, what doesn't work, and having that just kind of be the default state, right? I'm a big proponent of things like radical transparency in terms of incident reports or outages or hiring or anything. Right. The more information that you can kind of put in the world is going to it might not make it better, but it at least helps change the conversation. It gives more data points. There was a whole blow up on Twitter this week where someone posted like, hey, this is a salary I'm looking for. Right. I think you. Oh, yeah. She's great. Yeah. She's worth it. Right. And the thing

that got everyone's being a bond, it was like she's saying like, oh, I want one hundred eighty five K. And it's like, well, why don't we just publish that information? Right. Like, why isn't everyone just very open and honest about their expectations. And I know why, because the paucity of information is a benefit to employers and it works against employees. There was a lady that left. Gosh, where was it? I forget the company, but she left because she found out she was systemically underpaid compared to her male peers. Having these sort of information imbalances don't really help the people at the bottom of the pyramid. They don't help the little guys. They really only help the people that are in

the very large companies with a lot of clout and ability to control narratives. And they want it to stay that way. They don't necessarily want you to know what everyone's salary is because then it gives you, as someone trying to get a job, a better negotiating position because you know what someone with your level of experience is worth to them. It's important to understand the context behind these salary negotiations and how to go about getting interviews and the rest. The entire job hunting process is heavily biased in favor of employers because, especially at large employers, they go through this multiple times a week. Whereas we go through this as employees basically every time we change jobs, which for most people is every couple of years. And for me, because of my mouth, it's every three weeks. Yeah, I'm not saying it's a simple solution, right?

Like, I am advocating for sort of societal or just like cultural shifts. But I think that it all comes full circle in the sense that, hey, a big part of observability is the idea that you need to be able to ask arbitrary questions, right? You want to know about unknown unknowns. And maybe that's why I like it so much as sort of a field. Why do I, why I like tracing? Why I like this sort of idea? Because, yeah, you know, a lot of things in the world

would be interesting and different and maybe more equitable if we did have more observability about not just, hey, I use Kafka and I use these parameters on it and that gives me better throughput. But what if you had observability for like how HR runs, right? What if you had observability for how HR runs? What if you had observability for how hiring is done? And that was something that you could see outside of the organization as well. What if we shared all this stuff more and more and more

and we treated a few less things as trade secrets? I don't know if that's ever going to happen in my lifetime, but it's kind of my default position, right? Let's share more rather than less. Yes, absolutely. Especially those of us with inordinate amounts of privilege, and that privilege takes different forms. There's the usual stuff people are talking about in terms of the fact that we are overrepresented in tech in many respects, but there are other forms of privilege too. But there's a privilege that comes with seniority in the space. There is a privilege with being a published author in your case. There is privilege in having a broad audience

like I do. And it just becomes this incredibly nuanced story. The easiest part of it to lose sight of, at least for me, is I tell stories about what has worked for me and how I've done what I do, and I have to be constantly conscious of the fact that there is that privilege baked in and call it out where I can. I've gotten much better at that, but it's an ongoing process because what works for me does not work for other people across a wide variety of different axes. And I don't want people to feel bad based upon what I see. Oh, yeah, absolutely. I mean, I'm in the same boat.

I tend to be very irreverent and shitposty and I don't have much of an explanation other than I learned at some point in my life that I would rather go through live shitposting on Twitter rather than be employable. It's just who I am, right? I'm sure some people think I come off as rude. I don't know.

I also agree. You never punch down. You only punch up, right? But you never know how other people are going to take that, and I don't think that it always gets interpreted in the spirit it was meant, and I can always do better, right? As can we all. The hard part for, I think, a lot of us

is to suppress that initial flash of defensiveness when someone says you didn't quite get there and learn from the experience. One of the ways I do that personally is I walk away before responding sometimes. I want to be a better version of myself, but when I get called out of, like, this tweet thread is the whitest thing I've seen

since I redid my bathroom walls. And I get a flash of defense. Excuse me, that's not accurate. And then I stop and I think, and then sanity prevails, where it's, yeah, there's a lot of privilege baked into my existence. And if I don't see it, that doesn't mean it's not there. I have made it a firm rule of not responding defensively to things like that ever. And there are times when I get called out for aspects of how I present that I don't believe are justified, to be very honest. But that is a me thing. That is not them.

And I welcome the feedback regardless. If you make people feel like a jerk for giving you feedback, they stop giving you feedback. And then where are you? Yeah. Funny anecdote. I wrote a blog for my personal blog a little while ago about, oh, togetherness, community, something like that. But I wrote the intro was something like talking about how white people love Sweet Caroline, right? Favorite song in the world. Da-da-da, da-da-da.

Yeah. Yeah. I'm not allowed to play with that song here at the Duckbill Group because one of our employees is named Caroline. And firm rule, don't make fun of people's names. They're sensitive about it. And let's not kid ourselves here. I own the company. Even if she says, it's fine, I love it,

that doesn't help because I own the company. There is a power imbalance here. I don't know that she would feel that she had the psychological safety to say, that's not funny. I hope absolutely hope she would. Cause that's the culture that I send significant effort on building, but I can't depend on that. So I don't go down the path of making those jokes,

but I, yes, I love the intro to the song. It's great. Everyone loves it. So the intro, my, my initial paragraph was ruminating on that, right?

And this post went around enough that it got submitted to Hacker News a few times. And the only comment it got was some mendacious, busybody, Hacker News type going on about why I would be so racist against white people. And I was just like, and this is why I don't come to this website at all. Yeah. There are so many things on Twitter that are challenging and difficult and obnoxious, and it's still the best thing we have for a sense of community. This has replaced IRC for me, to be perfectly honest.

Yeah. No, I used to be big on IRC, and then I left because, well, a couple of reasons. One, I really liked being able to post GIFs. Yeah, that is something where the IRC experience is substandard. I was Freenode network staff for years, and that was the thing to do.

Now, it turns out that the open source dialogue and the community dialogue have shifted form. And I still hang out there periodically for specific things, but by and large, it's not where the discourse is. Yeah, it is interesting. It's something that concerns me kind of in a long-term sense about not only our identity, but also sort of the actual organic communities we've formed. We've sort of put onto these extremely unaccountable, privately held platforms, right, whose goal is monetization and growth so that they can, you know, continue to make money. And for as much as anyone can rightfully say, hey, Twitter's missed the mark a lot of times,

it is a hard balance to strike, right? They don't have simple questions to answer, and I don't necessarily know if the nuance of their solutions has really risen to the challenge of answering those well, but it's a hard thing for them to do. That said, I think we're in a really awkward position where suddenly you've got the world's collection of open source software is being hosted on a platform that is run by Microsoft.

And I am old enough to remember embrace, extend, extinguish, right? Oh yeah, I made an entire personality out of hating Microsoft. Yeah, and I mean, a lot of people still do. I read Mac rumors sometimes, and they're all posting there still. Or Slashdot.

I wondered where they'd gone. I didn't think everyone had changed their mind. I had just a very out-of-body moment yesterday because someone replied to a comment of mine about Slashdot on it, and then the Slashdot Twitter account liked it. And there exists a photo of me from when I was a teenager where I owned a Slash dot ball cap. And that picture is somewhere in the world. Probably not on the internet, though, for very good reason.

I'm mostly just still reeling at the discovery that there's a slash dot Twitter account. But I guess time does evolve. It does. It makes fools of us all. It really does. Well, Austin, thank you so much for taking the time to speak with me. If people want to learn more about what you're up to, how you view the world, etc., etc., etc., where can they find you? So you can find me on Twitter, mostly, at Austin L. Parker.

You can find my blog with various musings that is updated infrequently at aparker.io. And you can learn more about Deserted Island DevOps 2021 coming on April 30th this year at desertedislanddevops.com. Excellent. And we will put links to all of that in the show notes. Thank you so much for taking the time to speak with me. I appreciate it. Thank you for having me. This was a lot of fun. It really was.

Austin Parker, Principal Developer Advocate at LightStep. I'm cloud economist Corey Quinn, and this is Screaming in the Cloud. If you've enjoyed this podcast, please leave a five-star review on your podcast platform of choice. Whereas if you've hated this podcast, please leave a five-star review on your podcast platform of choice, and then a giant series of comments that all reference one another, and then completely lose track of how they all interrelate and be unable to diagnose performance issues. If your AWS bill keeps rising and your blood pressure is doing the same, then you need the Duck Bill Group. We help companies fix their AWS bill by making it smaller and less horrifying.

The Duck Bill Group works for you, not AWS. We tailor recommendations to your business, and we get to the point. Visit duckbillgroup.com to get started. this has been a humble pod production stay humble