Kat Cosgrove is a developer advocate at JFrog, makers of a robust platform designed to accelerate DevOps workflows. She’s also a legit cyborg, with an NFC chip implanted in her hand. Kat brings 15 years of experience to JFrog, having previously worked as a software engineer and lead teaching assistant at Code Fellows (where she also studied advanced software development in Python), a technical support coordinator at Online Holdings, LLC, and a business development professional at Remote Backup Systems, Inc., among other roles. She’s also the co-host of fsckdpod, a left tech podcast.
Join Corey and Kat as they discuss what it’s like to have an NFC chip implanted in your hand and how Kat uses hers, what it was like to get the chip installed, how Kat weaseled her way into tech from a gig at a video rental store, how working as a bartender at a strip club has helped Kat’s developer advocacy, how Kat is the same on Twitter and in real life and how she’s different, what it’s like to gain 7,000 twitter followers in 36 hours and the tweet that did it for Kat, why both Corey and Kat are keen on the liberal use of the Twitter block button, and more.
Announcer: Hello, and welcome to Screaming in the Cloud with your host, Cloud Economist 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 by our friends at New Relic
. If you’re like most environments, you probably have an incredibly complicated architecture, which means that monitoring it is going to take a dozen different tools. And then we get into the advanced stuff. We all have been there and know that pain, or will learn it shortly, and New Relic wants to change that. They’ve designed everything you need in one platform with pricing that’s simple and straightforward, and that means no more counting hosts. You also can get one user and a hundred gigabytes a month, totally free. To learn more, visit newrelic.com
. Observability made simple.
Corey: This episode has been sponsored in part by our friends at Veeam
. Are you tired of juggling the cost of AWS backups and recovery with your SLAs? Quit the circus act and check out Veeam. Their AWS backup and recovery solution is made to save you money—not that that’s the primary goal, mind you—while also protecting your data properly. They’re letting you protect 10 instances for free with no time limits, so test it out now. You can even find them on the AWS Marketplace at snark.cloud/backitup
. Wait? Did I just endorse something on the AWS Marketplace? Wonder of wonders, I did. Look, you don’t care about backups, you care about restores, and despite the fact that multi-cloud is a dumb strategy, it’s also a realistic reality, so make sure that you’re backing up data from everywhere with a single unified point of view. Check them out at snark.cloud/backitup
Corey: Welcome to Screaming in the Cloud. I'm Corey Quinn. I'm joined this week by Kat Cosgrove, who's a developer advocate at J. Frog, and an actual cyborg. Kat, welcome to the show.
Kat: Thank you for having me. I appreciate the mention of the actual cyborg. I feel like that gets glossed over a lot, and I'm not a fan of it.
Corey: It does. Can you tell us more about it? It sounds like if you're going to put that as the opening line to your bio, there's probably a story there.
Kat: Yeah, so I have an NFC chip implanted in my right hand. I got it at Defcon a couple years ago with my best friend; he also got an RFID chip. And I mostly use it when I'm at conferences and—you know, when we're allowed to be at conferences—to give people my contact info. So, if you tap your phone up against my hand, it will load either a link to my LinkedIn
page, or to my Twitter page, or a vCard for my contacts.
Corey: It becomes pretty obvious that if Cyberpunk 2077 have been delayed even further, it would have just been a documentary. My God.
Kat: True. Yep. It's pretty rad. They have ones with LED lights now, so that’ll probably be my next one.
Corey: Would you consider it a form of minor surgery at some level? Because what you're saying and—or at least what I’m hearing is, “Well, I had some minor surgery done in the back alley at Defcon,” which, all right, you know, I'm not one to judge out loud, but I look at that and I do have some questions that would be natural follow-ups to that.
Kat: Well. It goes in with a needle, actually. They come prepackaged in an injector that just goes a little bit under the skin between your thumb and your forefinger on the back of your hand. And it’s—doesn't require a stitch, doesn't require any glue, just a band-aid. It's really quick.
Before any of the Bill Gates vaccine microchip conspiracy theorists jump up with their hands in the air, it is a very large needle. There is no universe in which this could be hidden in a vaccine, though I wish we did have chips small enough to do that because it would be rad.
Corey: Yeah. At this point, we pay way too much for our iPhones for people to have to chip us involuntarily.
Corey: But I digress. So, we'll talk a little bit—in a minute or two—about what you're doing now, but first, let's talk about how you got to where you are. This is one of the problems when you look at folks who are well-known in the space as you are, it's natural to assume that you sprung fully-formed from the forehead of some God. That is, I'm told, not strictly true. So, from whence did you come?
Kat: Yeah, not strictly true, while my dad is very cool. I didn't go to college for computer science. I barely graduated from high school. I went to college for biochemical engineering, and I also dropped out pretty quickly.
Actually, I then went to work as a bartender at a strip club. I don't talk about that a whole lot publicly because I was worried at first that people would look down on me. But now I've decided that I just don't care and it's not something you should be looked down on for. From there, I went to work as a clerk at an independent video store, a pretty big one. Like, an average Blockbuster had about 600 titles in store, we had, like, 41,000. And this was not decades ago; this was like ten years ago.
Kind of like, weaseled into tech from there because they needed their computer replaced, the one that was handling the rental database and it was running Windows 98 SE, in [laugh] 2008, 2009. And I replaced it, and from there taught myself SQL, became their database administrator, which was pretty rad, started teaching myself some other programming languages. Ended up doing some freelance web dev, mostly WordPress. And then I moved to Seattle and went to a coding boot camp.
Corey: And now here you are as a developer advocate at JFrog
Kat: Yeah. I went from zero to relatively popular on Twitter.
Corey: So, let's tie those two together: things from the very beginning and things at the end. So, yeah at first, I think that the, I guess, social shaming of folks who work either in sex work or adjacent to sex work is bullshit and I have no tolerance for it, but let's talk about what did you learn as a bartender that serves you well in DevRel?
Kat: It can be a difficult task to manage people who are needy and demanding, but you can't meet their needs, but you also have to keep them happy. It teaches you to deal with unruly people in a very specific way that keeps everybody happy. It respects my boundaries and the boundaries of my employer, or society without forcing me into a difficult situation with somebody that I don't want to piss off. Of course, frequently I do just piss people off if I've determined that they're not somebody I need to keep happy.
And that's also something you learn from bartending: where the line is; when to fire a customer, so to speak; when it's not worth engaging with somebody, and it's better off to just ignore them, or get management, or start a fight. It's a difficult thing to learn. It does also, unfortunately, teach you how to deal with tolerating a shitty situation, tolerating difficult people, tolerating somebody who is not going to respect you no matter what you do. I don't think that's something we should have to learn, but it is a lesson I learned and it has made doing my job easier and being a minor Twitter celebrity easier. [laugh].
Corey: What's fascinating is that so much of what you just said, is almost a foreign concept to my lived experience, almost as if not everyone is treated the same way. Namely, by which I of course, mean a cis-hetero white man. And that is a tremendous problem.
Kat: It is. It is a tremendous problem. It's wildly inappropriate. I hate that it's still an issue. But we are collectively being louder, and louder, and louder, and less tolerant of that kind of behavior if recent blow-ups on Twitter over the last few months have been any indication.
Corey: I can't tell if we're actually changing people's minds, or getting them to keep their stupid opinions to themselves. And if I'm being perfectly direct, I'm not sure I care. I like the outcome which is, if I have this misguided belief, that based upon what someone's background is, or what their appearance is, or how they express themselves, [00:07:00 of a] gender, or otherwise, winds up somehow invalidating or validating the legitimacy of their opinion, I kind of want you to shut up and keep that thought to yourself because it's not the people I argue with on Twitter that I care about so much as it is the people who watch that argument unfold, and what they take away from it.
Kat: Right. Every time I pick a fight with somebody on Twitter over some kind of gatekeeping, or judgment of another person's worth in tech-based on whatever about themselves, that argument is never to change the mind of the person I'm fighting with. Because you're not going to, probably. They're set in their ways. This guy's just an asshole, and there's nothing you're going to do to change that. But what you can do is broadcast to everyone who follows you, and everyone who follows them, and everybody who sees it, that this shit is not going to be tolerated anymore. And that's a little bit more valuable for me.
Corey: As someone with relatively recent, newfound Twitter celebrity, as you frame it, how much of a distinction do you draw between Twitter and the real world? That's what I wrestle with a lot.
Kat: I, in a lot of ways, am just me on Twitter. There are some aspects of my personality that are considerably louder on Twitter, and some that are considerably quieter. In real life and on my private social media accounts, I don't post that many selfies, and I don't dump my every single, half-awake thought on to my private social media accounts like I do on Twitter. That is absolutely an engagement thing and making sure people still see my stuff thing. But the one thing that is consistent between internet Kat and real-life Kat is being aggressively intolerant of intolerance. I just do not put up with that online or in real life.
Corey: One of the biggest problems I see across the entire industry is either explicit and direct forms of gatekeeping or backhanded, slowly subtle ways of gatekeeping. Either way, I can't stand it. This industry is difficult enough to master, and the technology is expanding, the surface area is geometrically exploding, and we need more people involved, not less. “Well, you didn't have the following educational credential,” or, “You didn't go to the proper school for the right kind of thing,” is just—this is awful.
I'm sorry, I look at the things I work with on a day-to-day basis, there is no academic program that tackles this sort of thing: the technology is moving too fast. And it comes down to learn how you learn best, but I don't think that there's any particular credential that is required to excel in this space. But let me also call out that I am talking about software, and technology, and evangelization of same. I am not talking about becoming an anesthesiologist.
Kat: [laugh]. Yeah, I think we got into some weird position where people in tech have started to think of themselves as super elite geniuses, and nobody could possibly do this, and we're so valuable to humanity. And yeah, we do a useful thing. We build useful tools. Most of our lives are touched by technology in some way, but we're not God, dude.
It's so, I don't know, gross, I guess, to think of ourselves as better than everybody else because we beep boop, make computer do thing. And it doesn't require a special education for most things, it really just super does not.
Corey: One of the problems that I perpetually run into is—how do I put this—folks who try to imitate aspects of what I do without the nuance, where it's, “Hey, you're out there insulting companies. I'm going to do the same thing.” But instead of, I don't know, a trillion-dollar company, they wind up going after a five-person startup or whatnot, it's no, no, you're just being a dick. But thank you for the attempt.
Kat: Yeah, that's reading the room wrong. There's a difference between being an asshole in a funny way, being mean to AWS, and being mean to some dude trying to start an app company out of his garage. That's not fair.
Corey: Or making fun of an aspect of AWS, versus, “You work at Amazon, therefore, I'm going to corner you, as some rando developer, and accuse you of war crimes.” It doesn't work that way.
Kat: It really, really doesn't. And it's something that, I think, I didn't expect to be shouldered with the responsibility of determining who I mean to on Twitter as early as I did because I went from 4000 followers to 11,000 followers in about 36 hours. And that's the difference between having a moderately popular niche Twitter account and being a minor internet celebrity in a particular field. And whoo boy, that has definitely changed the way I interact with people because I'm terrified of accidentally saying something mean to somebody with 200 followers who didn't realize they were stepping in shit, you know?
Corey: Whenever that happens, and I find I've done it—which I'm getting better at steering away from where that goes—but I'm also looking at this through a lens of there are certain aspects of Twitter that will—nothing you do once you set a foot wrong is ever going to be enough. But I look at the folks who have become—how do we call it—the main character today on Twitter, and it's not just having a bad take, it's about doubling or tripling down on that bad take once people come back with, “Ehh,” moment. And I've gotten things wrong in the past and it turns out that a sincere apology is generally enough for most reasonable people. Now, there's always going to be someone for whom it's never enough. But those aren't the people you necessarily want to engage with and have these perpetual conversations with. And at some point, liberal use of the block button is the right answer.
Kat: It is. I was not aggressive with it when I had a small account, but now I get so many, like, crappy replies and weird DMs that I'm pretty aggressive with it. If you've got no profile picture, and you're following 350 accounts, and only have one follower and you say something weird in my replies, I am not going to gamble. I am just going to block you.
Corey: No, at some point, it doesn't make sense. And it's not just about Twitter followers and the rest, this also has to do with the real-world implications of when someone has positioned themselves as, I don't know, a senior engineer at Google, let's say—
Corey: —your words carry weight you may not realize that they carry. Right now, my shtick works because I am fundamentally, in the industry, a nobody. I run my own company, and that's great, but it doesn't have the gravitas of someone who's a distinguished engineer at one of the big tech companies, or someone who's been in the space and built a thing. When you're that kind of person, even if you don't intend it, there's no way, no matter what you say about, “Oh, these are my opinions, not my employer’s,” that's not how it's going to get cited in some BuzzFeed article or whatnot. It's always going to come back to, you have caused a problem for your employer. I don't have that burden. And you sort of do, in a different way.
Kat: I do.
Corey: Not that I think that folks necessarily take what you're saying on the internet as the gospel truth, according to JFrog, but there is the problem of having an actual company with a lot of people behind it that, in some ways, are going to see indirect consequences in the things you say, despite your best efforts to the contrary. Do you find that that winds up shaping what you say, and how?
Kat: Yes. Two weeks ago, I would have said no, but now I'm saying yes because it happened to me, and it could happen to you too. So, Kubernetes decided to deprecate docker-shim, and this is something that had been discussed for literally years within the community. This was open knowledge that it was going to happen, but I guess the users didn't know that. So, when the changelog released, a screenshot of it was tweeted, highlighting the fact that docker-shim was being deprecated, meaning that you could no longer use Docker as your container runtime. I tried to be helpful. And instead—
Corey: Oh, that’s where it all starts: “I tried to help,” and then look what happens.
Kat: Yeah, so the moral is don't try to help. That's not true. Do try to help but be aware that there still might be repercussions. And I tweeted a very helpful thread explaining what was actually happening to calm people down because people were rightfully freaking out. And at one point, I made a comment that was like, “Docker isn't dead, yet.”
And I had 4000 followers when I tweeted that, and I thought it was just a cute off the cuff thing, but then the thread took off. And two days and 8000 more followers later, that has weight. I don't know how to predict that. How do you predict that kind of thing?
Corey: Oh, you absolutely don’t. I find that when I do big threads that blow up from time to time, there's usually some offhand comment that I've made somewhere like around tweet 80 or so, but that attracts the sketchy rando vibe, where it's, “Wait, are you saying—” and they go in some completely different direction. And I'll respond to one or two with, “No,” but beyond that, like, I can always tell when a tweet goes beyond my typical readership because if I go out and I tweet right now that multi-cloud is a stupid idea, as it is implemented almost everywhere, most of my followers at this point are used to me taking that opinion. If that blows up beyond a certain sphere, then I start getting a lot of the, shall we say, “Well, actually,” in the response, and that sucks, where it's, “Oh, great. Now I get to clarify exactly what I meant.” And no sooner have I done that to one person asking me, then I have five more with variants of the same thing. And then it takes off, and at some point, you're playing whack a mole, and I just give up and start tweeting something different.
Kat: Yeah. And that's what happened to me in that case. It was a bunch of people rapid-fire asking the same question once the thread took off about, “Well, what do you mean?” People assumed I worked for Google. People assumed that I worked for the CNCF, or was actually a Kubernetes maintainer, people assumed that I worked for Docker.
And my employer is right there in my bio, so they were not even bothering to check. And I was fielding the same angry question a dozen times over the course of a couple of hours. And I just muted the thread because I answered it three times, and then, come on. Just read. Just read the thread, please.
Corey: Yep. People never read as much as you think they will. I've had people who are amazed that AWS hasn't fired me yet. I'm not sure you understand exactly the nature of my employment situation, but that's all right. I'm still waiting for someone to reach out to my business partner, who is the CEO of The Duckbill Group—Mike Julian—and attempt to have me fired over something I have tweeted. That's going to be a glorious day. A spoiler: he can’t.
Kat: You know what, that has happened to me once. Somebody tried to get me fired because of something I said on Twitter. It did not work. But it did happen. And it was funny. So, it's a weird thing to do.
Corey: 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 percent faster, and helps you act immediately. Ask for a free trial of detection and response for AWS today at extrahop.com/trial
Corey: The thing that bothers me is when I wind up doing a tweet that inadvertently winds up smacking at people I didn't intend for it to smack at, that's the stuff that keeps me up at night. Not that I'm going to upset some alt-right dingbat, but rather the fact that I'm talking about a product decision being crappy, but I don't want the people who built the product to think that I'm crapping all over their hard work. So, it's a weird problem where things like that tend to focus on one or two specific people, but I want to be nuanced and careful in how I criticize the product—“Because that's how they improve—without ruining someone's week where, wow, that loud jackwagon on the internet thinks that this is awful.” “Well, yes, but I didn't mean for you to take it that way. I mean that there's an opportunity to improve here.” But you never get to add that nuance in 280 characters.
Kat: Yeah, you don't. And it's one of the reasons why I try to avoid aggressively talking shit about other products unless it's something I genuinely, really, really do not like, which I think has only happened once within the last year. I wrote a really nasty blog article about an AWS product, actually, that I now do not remember the name of because they announced it, and I genuinely was like, this is so incredibly stupid.
Corey: Oh, I say that, like, three times a week. But normally, I've started bounding it to making fun of the service name because no one spent 18 months naming Lookout for Metrics like it's a warning sign. And if they did, they should feel bad.
Kat: Yeah. They obviously do not spend time on naming things. And I get it; naming things is hard. That's a long-standing joke in tech that will probably never go away, that the hardest thing in tech is naming things, and off-by-one errors. But it feels like they are not trying, Corey [laugh].
Corey: They're not. At some point, naming things is actually super hard because making fun of the name: relatively easy, but what are you going to call this thing? And, that exists, that's a trademark problem, et cetera, et cetera. I mean, I just launched the DuckTools SaaS offering that we're doing and I'm waiting for someone to draw the Ducktales comparison at Disney and start yelling at us for it. It's hard to name things.
And… we'll see. I'm not terribly concerned. But there's also this problem of how do you wind up having something still tied to a unifying brand, but it's clearly its own thing. And when you start launching something like Systems Manager, great, you don't necessarily know in advance that it's going to have 20 different sub-services and that calling one of them ‘Session Manager’ becomes a valid thing. But at the end of it, of course, the name is dumb in hindsight, but how do you go back and fix it because naming is hard; renaming is worse.
Kat: I don't think you can. I think you just have to own it at that point. But AWS has had a long time to figure this out. And a lot of product families where they have made the same mistake over, and over, and over again, nothing stresses me out like looking at my AWS dashboard. I just don't do it anymore.
Sorry, AWS. I'm actually not even vaguely sorry. But yeah. They’ve made the same mistake repeatedly with naming things. They released a thing called AWS CodeStar that lets you spin up a skeleton of some project in whatever framework and language you—
Corey: And tie together a bunch of CI/CD services on the AWS side that don't have any actual customers here in the real world.
Kat: That's a separate product family. They did CodeStar years ago, and then they just released CodeArtifact and CodePipeline—or whatever it is—within the last year. So, they already had a thing that was called ‘code something,’ and then they went and released a whole family of CI tools called ‘code something,’ and they're not related. It drives me absolutely bonkers.
Corey: And one of the things that drives me nuts is, take a look at Cloud9, I really was interested in that when it launched, it was oh, cool, almost like—like, they focused on the idea of having it as a hosted platform—as opposed to Visual Studio Code, which is focused on building an IDE—and Visual Studio Code evolved to the point where you can now run it online with GitHub Codespaces, and that is freaking incredible.
Kat: Yeah, it's rad.
Corey: And Cloud9 basically then kind of sat on it’s ass for two years, and everyone forgot it was there. And it wound up launching a new API at re:Invent this year that I'm still not sure what it does because I don't care because neither I nor anyone else I know uses Cloud9. And it's not because I don't want to; it's because it's bad. Whereas every problem I have with it is solved, by and large, with GitHub Codespaces. So, rather than screaming about wanting a first-party thing from AWS, I'll use the thing that's actually good.
And I don't understand why AWS just likes to let things sit and marinate for years. I keep expecting there to be some giant release that modernizes all of this, and they don't. Instead, they just do an iterative improvement a few years later. It's like, I don't know if they disbanded and reconstituted a team for it or what, but God, it drives me crazy.
Kat: I would love to know why they do that, actually. I would really, really love to know what the business logic is behind that. Is it somehow beneficial to just keep cranking out a ton of services and just sit on them so nobody else can have it. Or when somebody else releases a better version of that thing that actually takes off, they can take the moral high ground and be like, “Well, we did it first, just worse.” What is it?
Corey: Well, then you have the other side of it: CloudShell, which just came out, which people have been wanting for years. They were late to the market in a rather severe way. Google came out with this five years and two months beforehand, and it was transformative. And since then, Microsoft came out with it, in 2017, for Azure. And then at the beginning of 2020, we saw that Oracle Cloud came out with it, and then, holy crap, IBM Cloud came out with it in June.
So, when you're trailing IBM to market on something, really stop and evaluate what's going on. I love the service, but at this point, I'd mostly mitigated all the things I would have wanted it for. The idea that you log in as an IAM user on an account and have the access to run various things in that account is huge. You don't have to teach beginners to configure their local environment for half a day first. I've been stuck in that particular trap.
And it's still obnoxious because to have access to this, you need the power user admin-level tier in your account. So, okay, that means it's going to be challenging to wind upscaling that down at the moment. But ugh, I want to see a better answer, and I worry this is going to be one of the things we never see an update for, ever again.
Kat: I didn't know that they had finally released that, and the lack of a service like that is a huge reason why I just hard committed to GCP because I don't want to spend half of my day messing with my environment to make sure that I can actually connect to my AWS environments. I can't believe they finally released one. I assumed they were just going to die on that hill.
Corey: Yeah. I don't for the life of me understand it, but here we are.
Kat: Well, good for them. Welcome to 2017, I guess. But if IBM and Oracle beat you to something, you need to really, really look at yourself because nothing makes me think old, slow, and stodgy, like Oracle's overall brand. That's a product of growing up the way I grew up and having a dad who's an engineer and just deeply does not like Oracle either. So, maybe there's some personal bias involved, but oof; big oof on that.
Corey: Yeah, it just becomes a difficult thing to do because, again, as soon as you get into the cloud space, everyone wants to start smacking the competition and the rest, and it's harder to do than you think because there aren't too many people who are going deep into multiple cloud providers because why would they?
Kat: You don't need to. You don't need to. And honestly, for the overwhelming majority of the stuff I do in my personal side projects, I just chuck it on Heroku.
Corey: Yeah, absolutely. No problem of that. I use Heroku myself for a few things.
Kat: Yeah, it's rad. It's easy to use, it's CLI is sane.
Corey: It hasn't really changed much in ten years, and that's kind of a benefit.
Kat: Yeah, that's kind of the benefit. I don't have to learn something new every six months, when I get a harebrained idea and decide, I will start another side project that I will definitely finish this time. I know it's going to be the same deployment process than it was the last time I did this six months ago. With AWS, I have zero guarantee of that.
Corey: Yeah, trying to predict the future in the time of Cloud is always weird. Like, you have these companies that are just smacking at each other, they're trying to find their next big customer as a customer of a different cloud provider. But there's so many folks still moving to Cloud and so many workloads that are still on-prem that, fight for those instead. That's the opportunity. Alternately, focus on getting a bunch of folks to start building their ideas on your cloud provider.
And that's where the big companies of ten years from now are going to come from. It just—it's a longer term game. I think that Microsoft Azure is going to be a behemoth in the space, not because Azure itself is awesome, but because if they play this right, their integration story for puttering around with Codespaces on GitHub and GitHub Actions during the CI/CD testing pieces, and if you can just have a single-click deploy to Azure at that point, well, why wouldn't you? And they stand to more or less capture an entire segment. But Azure has to get better than it is right now, too. So, I'm optimistic about the future there, and I think that something like that would be fantastic for customers, I don't want there to be a one right answer to what cloud provider someone should use. I want people to have different conversations.
Kat: Yeah, Microsoft has done a really, really good job of turning it around lately. Like, obviously, Microsoft has problems politically and with respect to some contracts they have. But they aren't the evil company, I remember growing up with, anymore. Intel still gives me the same vibe that they did when I was a teenager and hated Intel, but Microsoft has really—I don't want to say that they're the good guys now, or whatever, but they don't give me big evil corporation vibes like they used to.
Corey: No, they've done a fantastic job of turning it around. It is the story that I'm seeing right now of corporate turnaround, where there's this incredible value of what they've been able to build and how they were able to achieve it. The other side of it, though is that, cool, that means that the things that they do that upset me and don't align with that, are ever more annoying. And okay, that's great. But there's also the painful part of what are they trying to build toward and what things, corporately, are in their way of really taking it all?
Because they've done such a good job at brand rehabilitation in some areas, but they haven't touched it at all on any of their licensing story. So, great, how do you unify all of the company behind this particular, I guess, addressing of that transformation story? It can't just be piecemeal anymore. And the idea of being able to do this, from a perspective of something that is across the board rather than on a business unit by business unit process is sort of their next obstacle. And I haven't got a clue on how to solve for that thing, but I am curious to see how it works out.
Kat: Yeah, I don't know how they're going to handle that either. And they're not the only company with that problem—of that size. Amazon overall definitely has a reputation for being a place to burn out as a software engineer. You go get hired at Amazon on some AWS team, fresh out of college. They pay you gobs of money, mostly in stock.
You work your ass off for a year, maybe two, and then you burn out and leave and you go to Microsoft. But that's not the reputation Twitch has, and Twitch is owned by Amazon. So, why do they have this cultural divide between teams? Microsoft has the same problem. I assume a similar issue exists at Red Hat, and IBM, and Oracle between some teams being, “Cool—” You can't see me because this is a podcast, but I'm doing air quotes—and being miserable to be on. I have no idea how that happened at any of these companies, and how they correct it, but it's a problem that needs to be addressed, I think.
Corey: Yeah. And I wish them well. And I think that there's going to be an awful lot of stories where this becomes better for everyone across the board. Was it a Bill Gates quote that said something along the lines of, “Do you underestimate how much you can get done in a year, and underestimate how long you get done in a decade.” Or something very close to that. Don't @ me. But the sentiment is there. Like, we take a look at where all these companies were ten years ago, and where the state of technology was ten years ago, and we're living in a completely different world. But on a day-to-day basis, we don't see those sweeping changes.
Kat: Oh, for sure. I found my hard drive from high school. And it's a Buffalo brand TeraStation. This was the first one-terabyte hard drive I bought, and it is enormous. It's, I don't know, around the size of a loaf of bread. And also, I remember paying, like, $500 for that.
And now I've got six terabytes of solid-state storage in my desktop, and I think I spent 400 bucks total across that. But it happens so slowly, I didn't notice. And all of technology is like that, I think. Every once in a while there's an outrageous breakthrough, but for the most part, it's a slow, slow revolution until we wake up one day and realize, “Oh, shit, this changed a lot in a decade, didn't it?”
Corey: Yeah. I take a look at my typical workflow on a day-to-day basis, and, “GitHub? Why would I be going to GitHub? I'm not much of a developer.” And weird, that’s kind of a lie we tell ourselves. I spend more time writing code these days than I would have expected, but I still don't think of myself as a developer. “Amazon? You mean the bookstore? Oh, that Cloud thing took off.” And look at how the world changes in a relatively short period of time. But there's no one day I woke up and thought, “Oh, my God, Cloud. This changes everything.” It's a creeping realization.
Kat: It is. And it has taken over a huge portion of our lives. I remember it being a big deal, and everybody was very confused and very concerned the first time an AWS region went down and it took down half of the internet. And now that happens on a semi-regular basis. And every time we're like, “Oh, what is it this time? Is it Cloudflare? Or is it us-east-1—” or-2? Which one was it this time, that went down?
Corey: 1 is generally the one that experiences issues, and it's so—the problem is, that's such a big region that a blast radius is enormous.
Corey: And people say, “Oh, us-east-1 is just a constant tire fire of downtime.” But looking at it for the past few years, it’s really not. The days when it has issues, which there have been a couple of them, but it causes massive disruption. But then you look at the data center that you're running for your company, and the reason that it doesn't have that reputation is that your customers don't care because you're not that big.
Kat: Yeah. They run a huge portion of the internet. And it feels like that happened overnight, but it totally didn't. And I never would have thought that we would be in a point where [laugh] an AWS region going down would make it literally impossible for me to do my job for a day. But here we are.
And also, I'm not complaining because I do need the time off every once in a while. Burnout’s real. So, anybody at AWS wants to pull a plug occasionally, just, like, strategically, that'd be cool.
Corey: Yeah, I think that would be worth doing in some respects. Ugh, there’d be days. So, thank you so much for taking the time to speak with me. If people want to find out more about your hot takes, where can they find you?
Kat: You can find me on Twitter at @dixie3flatline
. If you don't know what that's a reference to, and thinks I just crammed a bunch of words together from a Google search, it's not nothing. It's not nothing. It's just a much, much deeper cut than I expected. It's the name of a character from a book, Neuromancer
by William Gibson. So, if you don't know what that is, you should also read that book because it's really good. If you don't have Twitter, first of all, why?
Corey: That's an excellent life choice, is my counterpoint, but please, continue.
Kat: Yeah, you're right. Sometimes I regret it. You can email me at [email protected]
, but I will warn you I am not great about responding to emails. So, best of luck to you if that's the route you choose.
Corey: We will of course include links to that in the [00:34:33 show notes]. Thank you so much for taking the time to tolerate me, and suffer my slings and arrows, and ridiculous questions. It's appreciated.
Kat: Yeah. Thanks for having me. It's been lovely.
Corey: It really has. Kat Cosgrove, developer advocate at JFrog. 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 topic, please leave a five-star review on your podcast platform of choice and a comment including a link to your 200 follower Twitter account where you tell me I'm an idiot.
Announcer: This has been this week’s episode of Screaming in the Cloud
. You can also find more Corey at screaminginthecloud.com
, or wherever fine snark is sold.
This has been a HumblePod production. Stay humble.