State of Spring 2020, survey

Episode 180 · 50:33 · SRT · archive.org · all Tanzu Talk

Hello, dear listeners. This is a nice episode where we go over the state of spring survey. A lot of good stuff on the topic of spring, but just software development in general. Now, this was broadcast live on Tanzu TV, which you can find everything you want to know about that if you go to Tanzu.tv. And this is part of my Tanzu Talk series that I do there, also a podcast. Now, I'll tell you a secret. If you go to Coté.io slash Tanzu Talk, you can see my special made page about it. But what I wanted to tell you about is that I do this show every week daily. It's usually me talking, but I do have interviews like this spread throughout it. And I'll be putting several of those interviews in here. That's, I think, going to be the primary channel that I use is the interviews I do for the streaming and the videos I'll put up there. Now, you don't have to watch it live on Twitch. If you wanted to, you could go to twitch.com slash VMware Tanzu. But I also upload them to the VMware Tanzu YouTube channel. And again, put them at Tanzu.tv and also Coté.io slash Tanzu Talk. But really, you know, I would appreciate if you check out those videos and you do get to see a lot more visual things, you know, charts. And I don't think I'll ever have demos, but the people talking. But hey, also, you can just listen to the podcast, whatever you like. So enjoy. This is a fun one. And it's always nice to have my co-host Rita on instead of me just talking to myself. And also in this episode, I remember that I had forgotten to take myself off of mute a few times. So listen, it's going to be a little weird, but I've just totally cut out the parts where I'm not talking. And you can infer the questions that I had for people. It'll be a fun game that you can play. But sorry about that. I've since learned my lesson. Enjoy. Anyway, you know, it's like I was saying, when you stream these things, I like have no idea when they start. I've done these a few times. And there's always a few seconds where we're there before they record. But I think I've talked long enough about stupid stuff that it's probably

started streaming. You see, that's the magic right there. You don't you don't have an earpiece with someone saying, OK, Coté Q. No, I should get one of those. That would be good. I instead I have these gigantic earphones, but they're very comfortable. They're like they're like my safety blade that I have big earphones. Makes me feel very good. Well, Rita, it's a Rita episode. I think that's what we'll call them. Yeah. And what? Hey, to two guests, why don't you introduce yourself? Sure. So my name is John. John, all right. I'm over here in Seattle, despite the accent. And I work in developer marketing and focus on Spring. And I've been around the developer beat for a while. So I used to work at Microsoft on .net and have nowhere. Lots of bodies are buried there as well. Whoa. Are there a lot of bodies? There are a lot of bodies. And a lot of happy and less happy developers. But that's that's the nature of the beat. That's fair. That's fair. But I love working with developers. Such a creative audience. Yeah. How about yourself? Other person? Hi. Yeah. Hi. I'm Ben, Ben Wilcock. I am a technical marketer for VMware. And I work alongside John in the realm of Spring. I don't know where any bodies are buried there. How about bones? Do you have a dog? No. Coté, I'm finding myself really distracted by your t-shirt because I really want to know what it is. Oh, well, okay. Look, since we're on video, this is one of my favorite t-shirts because it sums up a lot of my life. See, it's this hot dog walking through the desert and it says nothing matters. Okay. Is that not like some eggplant emoji thing going on there? Oh, yeah. You don't realize, but it's all been, you know, banned.

It could be. I haven't actually thought. That's a good point now that I look at it. It's pretty old. It's got little pills on it. I've worn it for a while. But it also, like, it's not quite on that line. But the other line, I forget if there's a name for it. But I also love those pictures that are things that are carving themselves up for you to eat. Like, pigs are very popular with this where there'll be, like, a pig who's, like, a butcher and he's, like, serving you pig, right? Or chicken sometimes. And, like, I mean, this is a hot dog that's squirting ketchup in its mouth, which is equally great. But it's very close to that thing of, like, you know, like, they're also, you know, I'm sure y'all obviously having something to do with Europe, you know, forgiving your locale. You've seen that big, and maybe off in California, but you've seen that big statue of a hot dog that looks crazy. And he's got, like, a big ketchup thing squirting it on himself or herself. But that's, you know, there's similar things. Like, that's a good example of the kind of, like, food preparing itself to be eaten, which I think is highly entertaining. I don't know. Maybe that's an American thing. It's really the ice cream cones that you see all over Europe, just everywhere. There's ice cream cones on the street. Enormous plastic ones. Yeah. Yeah. Yeah. Do you have the vans, Coté? You're in Amsterdam, aren't you? Do you have the ice cream van where a van will drive around? Now, if you would have asked me that, like, two weeks ago, I would have said no. But I actually passed by one recently. This is, like, three weeks, whatever, a month ago. I was in a neighborhood, and I heard this sound, and there was this van driving really slowly, and it took me a long time to realize what it was. But yeah, it was an ice cream truck. It was crazy. I've never seen that in Amsterdam, and yet there it was. Yeah, the kids are more into the raw herring in Amsterdam. That's right. They get out of school. They go listen to that, you know, what is it? I'm going to take myself to the cow town road or something, and then they slam down some herring. Yeah. Yeah. I haven't

had herring in a while. It's pretty good, though. When my son was little, I told him that when they when they played the music, that meant they'd run out of ice cream. Oh, wow. Don't forgive me. That is a, that's a good dad move right there. Always coming up with, like, let's call them untruths to avoid hassle. I like that. I'm not proud. I'm not proud. Yeah. It's actually carrying loads of change. Yeah. I like if you had a detailed analysis of that, you would think, but why? Like, that would be, like, it's like, it's like the wailing of the ice cream truck when it's run out of ice cream. All right. Well, we, Rita and I have been planning this for a while, and it's been, I forget how long the spring survey's been out. You'll have to tell us. But y'all two did a fair amount of work on the state of spring 2020 survey, which I think has all sorts of fun findings. So I thought it would be nice to have y'all on to talk about it, to go over it. And with that wonderful introduction of myself, I mean, not myself, of the survey. So what is this state of spring survey? What's the, what would you, like, what's the premise of it, John? So, again, I won't go back to the bodies, but, you know, Microsoft back in the day had this thing called the developer tracker. And it's, it's, when you're in marketing, everyone thinks, okay, you judge how well you're doing by how much you sell. And that's fine. But when you're working with an open source technology or a developer technology, where it's all about adoption, you know, that you need some other way to really figure out how, how you're doing and, you know, how, how, how well your, you

know, your message is getting across, what people think about your product and, and how, and how they're using it. And so when I, when I joined the spring team earlier in the year, I was thinking, okay, where are we? Bunch of questions about developers, like, what are they doing with, with spring? What do they like? What do they dislike? What do they think about it? You know, this kind of emotional thing. And, and so I thought, let's, let's have a, let's have a survey. Let's do some proper market research. So rather than, you know, me spin up a survey monkey, throw it out and 50 people answer it and everyone go, well, that's not really statistically valid. Let's do something kind of a bit more scientific. Who are actually in the trenches using spring. So they're either developers, architects, or their managers running teams that are using spring. Um, so that's the setup. And then, uh, for SpringOne, we actually kind of wrote it up and released the state of spring report, which everyone can go download on the, the Tansy blog today. Hmm. And I'll say, you know, as somebody who has to read a lot of research, as you know, um, from, from my role, uh, I really appreciated it because I think it's useful for, you know, when we think about the roadmap, right. What we, where we want to take this, um, this technology, where we want to take spring or the platform. So super useful. And I was, I was really glad to see it. Yeah. And it actually, so this isn't the first state of that VMware has produced. Um, Heptio before it was acquired by VMware started off the state of Kubernetes. So if you're interested in a kind of similar take, but with that, that technology, um, you can go look that up. That's been running for two years. And our intent with the state of spring is that we, we release this every year. So it's not, this is a good baseline to start with, but it's also good because it lets us kind of track, you know, where things are going, um, where usage is going, what, how people's tastes and opinions are changing as well. So when, when, when I look through, uh, the final presentation of these surveys, I always wonder like, uh, how, how many, how many questions did

y'all cut out? Were there like, do you have all of the questions listed here or did you, are there some questions that you asked and you were like, you know, maybe it wasn't a good idea to ask them like the last time they cut their dog's toenails. So we might want to cut that one out. Yeah. That's a, that's a great question. Cut straight to where the bodies are buried, I guess. Um, no body's buried here, but this isn't, this isn't everything. If anyone that completed the survey knows, we asked a bunch more stuff and, uh, you know, there's a TLDR piece. If we, sometimes when I read through the whole report internally, I see people kind of glazing over because, you know, some of the, some of the answers are kind of, well, yeah, duh, that's what we expected. And, and other answers, we look at it and go, well, we probably didn't ask the question in quite the right way. The results came back and, um, we're not sure that there, there is an insight to be gained from that. So, and there are a couple of questions as well that were really, um, questions that are interesting internally, but not, not, not generally. So we did cut a few things out. Yeah. Now that, that, that makes sense. You know, when I, when I've worked on surveys before there's all, most of the stuff that I would have cut out is just boring or, or sort of repetitive, right? There's just like, not really, uh, I don't know. There's, there's, you know, you ask people what they think about cheese and most people agree on it. So it doesn't really have that much of an answer. And, uh, it's, it's good to cut those things out. So, uh, now Ben, before this role, you, you were actually a spring developer. I don't know what that makes you now, if you still are, but you know, so, so if you were to put on your, uh, spring developer hat, when you, when let's, you know, let's get to the content, when you're looking through the answers here, like, like at, from a developer, like what's, what's stuff that, what are you looking for when you look at surveys, like as a developer? I'm really interested in trends, I guess, um, because, you know, part of, part of any developer's job is trying to identify where things are going. So you're, you're constantly looking out for new

trends where, where adoption is growing in particular areas. Things like, um, reactive is a really good example of that because it's been around for a few years. We now have it peppered throughout the, uh, spring, uh, framework and spring boot has lots of starters that include things that will work very nicely with reactive, like, uh, spring data, for example. So, um, it's seeing the results from that and then being able to think about, okay, how, you know, should I be considering adopting those things too? So it's a bit of a, it's a bit of a technology radar, really. It's sort of telling you what's on the horizon and what's, uh, what's incoming, but also it gives you a nice, warm, fuzzy feeling that the stuff that you're working on is similar to what other people are also working on. Yeah. You know, you're not alone. No one wants to be alone at the real raw kind of bleeding edge of this stuff. You want to be with a community and have a community going in. Right. Right. You, you have, you have sort of a, uh, soft confirmation of normalness, which, which, uh, you know, people, people are always telling me there is no normal or average when I come to them with my problems, which, which is, I feel like then I need some mathematical definition of like, all right, let's say you were to plot on a chart, everything, where are the clusters? I want to know if I'm that, uh, but whatever. But I, I think, I think, yeah, as, as far as like trends, let's say, it kind of lets you know where you are. And that's a, I guess that is an interesting thing. I, I haven't thought about as far as it's been a long time since I was a developer, but you can use it as like a discovery mechanism for like, what, what are directions and other things that, uh, people are using. And I, and I wonder Coté, uh, I was, you know, and to anybody here, you know, when you're seeing what your peers or other folks that are working with this technology that they're doing, if they're doing something, you might've been reluctant to check out, or we're kind of curious

about, does it empower you? Does you, do you feel like, okay, I'm going to try it because there's others who were there. I have this community I can tap into. I can ask them questions. Was there a feeling of that? I mean, do you think there's a feeling of that? I would imagine there, there is maybe feeling like you're more willing to experiment because others have been there before. I think it also helps you position where you're at, like in, in a wave of technology. Are you that early adopter or are you kind of, where are you comfortable? Maybe you're comfortable in that early majority. So I think this is going to start digging into the results, but reactive is one of those areas. I think there's a general perception that, you know, reactive, it's still very early. Who's touching it? And yet it looks, a third of the people we talked to are actually using reactive already. Look, man, Gartner wrote a report about it. So you know that it's for real. They wrote about it last year, right? For the CTO was like, Hey, get this on your radar. So that was really nice to see this further validate that or vice versa. Well, well, so, so to your point, let's look at that, uh, that reactive thing. Well, first of all, I think, I think maybe three years ago, I got Josh long to explain to me what reactive was. And I sort of remember, but, but it's one of you, some of you will have to clear it up because I feel like reactive at some point was a UI framework, but then it became a backend data handling something. And I never really disambiguated the two, or maybe there was another R thing that I confused with it. So what is this, what is this reactive business? I'm normally not terribly good at correctly describing. It would probably be better. But didn't you do a whole blog series on this, didn't you? I think, I think, I think, however, I think, however, compared to me, you will be an expert, no matter what you think of yourself. Um, what, what we're trying to do with, uh, you know, what reactive is trying to do really is, is, um, give you more efficiency for the compute power that you have.

Now there's a couple of ways of doing that. The, the kind of incumbent way, the existing way was to try and multi-thread to try and use as many threads as you can on your device, but that is very difficult to manage in code. You know, there's really not that many people who understand multi-threading very well. I'm, I'm certainly include myself in that number. It's kind of a nightmare. Um, so if you can, uh, come up with new and clever ways where you can get more, uh, horsepower out of what you have, then, um, generally you can do things like serving more requests. And that's what reactive does. It uses this, uh, technique called an, an event loop, which is kind of at the core of that. And it uses asynchronicity to make sure that you can throw your, uh, requests onto this, uh, event loop and the event loop will take care of managing what happens to those over time. And then when you have a response, you'll get a call back to say, okay, this, this thread is now ready to do some more work. And that just basically means you can queue up an awful lot more work than you can if you're just using sort of traditional mechanisms. I see. Yeah. Yeah. I, I, I kind of remember his explanation now that, you know, you would use this for handling multiple requests, like from a front end or something for like a, maybe a data data or process intensive thing. And like you're saying, you want to kind of spread it out and queue it. So it's a way of handling a lot of requests. Yeah. It's, it's basically increasing that request density. So, you know, if a regular server can manage, I don't know, 10,000, maybe you'd be able to squeeze out more sessions from the same compute power. So maybe you'll get 10,000 to 15,000 or 20,000. You really, you know, you really have to try it to know for sure what you're going to get.

But the fact that Spring has reactive now throughout the stack means that at no point do you switch to IO blocking operations. You can keep your reactive use all the way down through to the database and back again. Okay. So, you know, Spring's pretty good at that. It's very, it's kind of unique in that, I think, I believe. So it's good to see that that is being rewarded with adoption. Okay. Yeah. Okay. That makes sense. And reactive isn't something that Pivotal or VMware now in Spring invented. There is a reactive foundation, part of the Linux foundation, that's kind of governing this and putting out specifications and we have some implementations of that. So it's, there is a general industry movement, you know, around reactive. New things like the R2DBC. So it's like a database interface that talks reactive rather than the regular, more kind of procedural code. So, you know, it's not just us. There's a lot of folks, Alibaba, just looking through the list. So, so I want to try something here. Do we, do we have some visual to show about what the reactive set, the survey says about reactive? I've never done this before. So we're going to see, I'm going to, I'm going to, I'm going to pin your video. I think you can share, right? Yeah, that's you. I'm going to pin your video, John. And then, well, and then I think I can transition to another scene. I don't know what I'm doing, but we're going to find out. You want me to share? Yeah. This is, this is live people. How's that? So, um, I go back to your early question about what we left out. I mean, the, one of the, one of the things we did actually was to take different questions. And when you, when you see the results, you start to put them together and go, well, actually these two things together, there's a kind of theme here, some interesting points. So what you're seeing on this slide are two really points around, you know, technology adoption, um, and, and new technology.

So on the left-hand side is, is the reactive point you see in the middle there, 33%. So we asked what, uh, development styles, spring boot. And when we say spring boot, we mean spring everything. So what development styles spring boot developers are using. Um, and it doesn't, they don't necessarily have to be using it with spring boot. It's just what, what development styles are, uh, are being used. And so as, as you can see, I don't know if this is a surprise, Coté, you've been on the microservices beat for a while. 83% said they're using microservices. You know, this is one of the things that we think spring boot made much easier. So it's great to kind of see that come back and everyone say, yeah, we're, we've got microservices kind of nailed pretty much. Um, and, and we'd like to take some credit with spring as to help make that easier. Um, and so projecting forward though, with, there are, there are a new development styles, um, kind of coming up reactive and serverless are two of those. So what we're trying to, excuse me, VMware is trying to take over my laptop and join domain, join it at this very moment. It's kind of inconvenient. Um, yeah, so we have these upcoming, um, development styles and it's interesting to see kind of how familiar they are to a spring boot audience. You know, looking forward, uh, what we can do, we have things like, um, like spring cloud function in the serverless space. So what spring can do to actually make, uh, create an environment where, where developers can use those. And one of the things we try to do is spring is, is to, to bring the, uh, bring new technologies to the spring developer rather than make them have to go somewhere else, learn a new language. You know, they can stay productive. Um, and, uh, as they do that, the platform just brings in the new styles like reactive serverless, um, and lets them take advantage of that. Um, so on the right hand side, um, it's in a similar way, spring cloud is, is bringing kind of new capabilities and new technologies to, to the spring develop spring boot developer.

Um, so spring cloud services has been around for quite a long time. Um, it was originally the, the, the Netflix, Netflix services they built to make their platform viable when those services didn't exist. Um, you know, they were pioneering microservices way, way earlier than, than a lot of folks. Um, but now microservices is a very mainstream. Um, and so then, then now they actually depend on the services we've built into spring cloud, um, rather than them kind of creating the services and us just kind of repackaging them. Um, but, but kind of new in the portfolio, we have spring cloud gateway, which, uh, leads everyone into that world of, uh, API internal API economies and, um, you know, uh, centralizing one point that a lot of, uh, a lot of API traffic runs through. Um, but we have some kind of cool tricks, which we could talk about now or in a later, um, session that, that are kind of make it more developer centric than, uh, your typical API gateway. And then spring cloud. Yeah. Because everything is code centric, right? They can do everything from, from the code. Right. The developer doesn't have to raise a ticket with the API gateway team to, to register their, their service. They just, they just create a little code asset, send it off to the gateway and they're self-registered. So, yeah, there's a lot of talk about like how something like this, a code centric approach to this kind of stuff is, uh, to, to, for example, creating APIs and building APIs or API management. And, and you said API Academy, uh, sorry, economy, uh, is that it, it mitigates the need for these big honking, you know, integration platforms. I mean, I spent years covering that space. So I have a soft spot for, for an iPad. I love me a little iPads, um, or before that. So, uh, so, uh, but, uh, yeah, the code centric approach has been, has been quite popular with API gateway and, and spring cloud data flow. Yeah. And it's a good segue. And then, yeah, the next one is spring cloud data flow, which is a, you know, a super powerful tool for building both batch and streaming apps. And, you know, so what the survey told us is that, um, your spring cloud services has been available for a while, pretty good use.

Um, and, you know, most of the folks we talked to are interested in it. Um, but if you're thinking about gateway and data flow and you go, well, I'm not sure who's using that. Then you can, you can refer to the survey and, you know, see that there is a hard core people that have really adopted those technologies. Um, but also interesting to us is that there is a lot of interest. There's an appetite to hear more about it. And so, so it's on Ben and I to really produce, you know, interesting demos and, uh, starters and, and reference architectures and things that will help them on that journey and, uh, feed the interest that's already there. You know, I was going to say something you said sparked my, my, my, my, I have a question here. I wanted to ask is as a developer, right? As a spring developer, how can I use this, this survey data? What's the value of it for me? I'm just, I want to think about other audiences and they're saying, and then let's take it. Like, what about app dev managers and app dev leaders? What would, what should they be taking from this report? And then maybe we think about product development folks, you guys want to crack at that? I mean, one of the things that we wanted to take the temperature of was, uh, was just January with spring. Where is it going? Um, you know, everyone's very familiar with Java. It's been around for going on like 20 years. Um, and so there are a lot of assumptions about technology that's been around that while, uh, sorry, that long. Um, um, and so I'm going to, I'm going to try this. I'm going to flip. Come on. Yeah. So we, so we wanted to kind of take the temperature of, where does it sit within your constellation of options? Both as a Java. I'm busting out Duke to say a little distraction. Okay. You know, where does it sit? Uh, our folks, is it something that's kind of there, but there are a lot of other options. Um, or is it really the, the number one thing? Um, and what we heard in some of the early interviews was, well, it's kind of the corporate standard.

Um, it's what everybody uses. Um, and, and that's great because when we onboard a new developer, uh, they, they instantly have, um, like a lot of processes and tools to plug into and we, and they can get productive very, very quickly, uh, with spring. So yeah, having it as that kind of primary development platform is really important. Um, and I think, well, that's great, but maybe that's just a sign of just the amount of time since that investment decision. You know, the, an organization makes a strategic decision to, to build on a platform. And once they do that for a few years, it kind of sticks, but that it may be that everybody dislikes the platform and it's just there and they're just stuck with it as the incumbent. So, um, but what you'd expect, if that was the case, then everybody would be running away from the platform. Uh, and, and every time a new app comes up, they go, oh, let's put that on, I don't know, Python or, or something else. And so built kind of building off that first stat of, you know, over half the people we surveyed have it as their primary, um, their primary development platform. Um, these numbers just blew me away. So 75%, like three quarters of the folks we talked to expect their usage of spring to grow over the next two years. Um, and of, of, of those folks that said it was going to grow 82% said that growth was going to come from new apps. So this is, this is not the sign of a, of a platform that's kind of stagnant and everyone's running away from. Um, and there's other data that, that talks about how folks really actually like working in the spring. Um, but it's, it's thriving really. Yeah, that's, that's, that's interesting. I was, I was, uh, I hadn't thought of that use for that, right. To confirm that, uh, people actually like it, that they're, they're, they're not like running away from it. But that's, that's a, that's a good point to be made with that, that information. I mean, I was just more looking at it from the, uh, looking at this question from the, um, I don't know, the momentum angle of like, you know, what's the planning for it?

And therefore like people thinking about if it's going to be used, but you're right. Like if, if they're, uh, planning to use it for new projects, then that makes sense. There's, there's some affinity towards it. Yeah. And, and I don't, I don't know if it came through yet. It's, it's in the text. Um, also they're migrating 56% said they were migrating apps to spring. So you, you don't typically, you wouldn't migrate an app to a platform that you were trying to move away from. Right. Right. So this was really helpful in, in just, and, and to go back to our early point about kind of what, what are other people doing? Some of that security, um, you know, you may feel in your organization that spring is great and we're doing a wonderful, you know, it's, it's amazing, but are we headed in the wrong direction is the rest of the industry using something else? Uh, how do we, with the reference point. And so in the report, we just wanted to kind of provide, um, some data and, you know, I'll be honest when we put the report out there, we didn't know this data was going to come back like this. So, uh, pleasantly surprised. And, and so, I mean, since it's up, I think, I think the other on the, uh, the right side there, like that's the, it's essentially going over what people use, uh, spring boot for. Which I think, I think, uh, as most people pointed out, it's kind of a unexpected sort of finding that there's a lot of API usage. Um, I mean, it's, would you say head and shoulders? It's, it's above the other, the other usage, right? So it's like just over 75% or 76% of people say they use it for, uh, internal and external APIs. Now, I guess, I guess you could pick multiple ones. This wasn't just your, uh, your top pick. You could pick, you could pick multiple and you could pick at this and say, well, actually, when I'm building a website, I'm doing that and I'm building the backend as an API. But so it's a, it's a little peek into the insight that it's sorry, uh, a peek into the internal way that developers are thinking about what they're building.

Um, it's like an API first mentality. That would, that would match up with the, uh, the microservices finding, right? That people are doing a, uh, a lot of microservices. Now, now Ben, you were, you've been involved in some modernization projects here and there over the years. Is, is, uh, like do, do a lot of the older ways of doing things, do they use a lot of APIs like this? Like, are there microservices or things that feel like microservices in use? Or is that like a new phenomena amongst the, the corporate estates that you've, you've seen over the last few years? Well, I would say that, you know, one of the beauties of Spring is that it has an enormous legacy. You know, there's, there's all sorts of applications being created over the last 16 plus years, all relying on Spring for their internal structure. Um, what, what you can see from this survey is that folks are modernizing some things, other things they might be planning to retire at some point in the future, but haven't done so yet. And so we need to, um, be respectful of the thing that they, um, that we saw recently in, uh, Jürgen's keynote, for example, at SpringOne was that, you know, we do have a very broad church of users for Spring. And we have plenty who are doing APIs and microservices, but we also have some who are doing SOAP still, maybe they're using SOAP based, uh, web services. Uh, we have folks who are just doing regular web apps as well. And you can, you know, you can see that there in the survey at 39%. And so we are always very cognizant of the fact that we need to bring those folks along with us on a journey. So when we innovate, we also need to have a very clear structured plan about how those folks can come along on the journey or where they need to plan to, um, you know, release themselves from that journey.

So, um, I think we're, I think we're really good at that. And, uh, I think we rarely break stuff, which is very, you know, if you're an architect or a development manager, the fact that it's very rock solid, that it's trustworthy, that it's been going for a long time, that it has this, um, you know, uh, uh, uh, uh, legacy behind it. And, uh, and this, uh, history behind it, and that we take folks along gently along, along the path that we're going. I think people really appreciate that. And, uh, that you can see that in the number of production apps that we have, you know, the production side of spring is enormous. And that's what I saw working with clients out in the field. They had every type of spring, every version sometimes. Uh, so, you know, some teams are working on very modern stuff. Other folks were maintaining older stuff. Um, but the, the way that they were doing that was, was broadly similar. And they really appreciated that. They didn't need multiple skill sets, for example, to be able to both maintain and innovate. And so, uh, you know, also looking at the demographics, like it, what, I mean, how would you describe the demographics? Like, I, the, the one, it seems like maybe like, uh, around 40% or so work at large organizations versus smaller ones. But, but when y'all were thinking about the size of the company, who, how does it kind of break out? I mean, and, and, and, you know, are, are there some slices of it that you didn't put in the report? Like, you know, big companies act more like this versus small companies act like this. Like, are there any differences in company size? Yeah. I mean, one of the wonderful things you can do with all this quant data in a spreadsheet is do what we call cross tabs. So you can, uh, you can mash up or you can take slices. So, I mean, one of the, one of the ones we, we took were, um, what do we, um, I'm sorry, my mind's gone blank.

We talked about this already. Um. Oh, was it the age thing? The age, uh, one that you don't have here is age. Looking at the demographics and that was interesting, I think. Yeah. No, it was, it was actually about primary development platform. Oh yes. Uh, maybe this isn't a surprise, but you know, smaller organizations are more likely to have just one, one development platform. So when we talk to smaller organizations using Spring Boot, that's likely it, that's likely all they use. Huh. And as the organization gets bigger, um, then, you know, maybe it's a departmental thing and we see this, Coté, you must have seen this. You know, there's some, some folks choose .NET, some choose Java. And so as an organization, they have more, more choices. Yeah. I was going to, I saw that in one of the previous slides, right. And I had, I had a scratched note for myself to ask you all about the concept of, you know, tool sprawl or platform sprawl and the dangers of that. If there are, uh, it's not so much spring focus, but what's, what's our thinking on that? Uh, is that a real problem? Is that something that large organizations need to deal with? Or is it just do the thing that's the best thing to get your job done? So how do we, how do we, uh, balance that out? The tools that are for the jobs to be done versus like, you know, uh, um, consolidating onto one platform. Yeah. I mean, one of the things that we, we, we surveyed was a bunch of technologies that are kind of around spring. And I'm, I'm not sure we, it made it into this, into this report. Um, but there is a huge fan base for Python among spring boot developers. That doesn't mean that they're, you know, they're becoming, um, not just loyal, but they're, they're, they're moving away from spring boot, but it's the right tool for the job. And while they're using spring boot to actually build corporate applications, you know, for when it comes to data science or, um, or maybe some tactical apps, Python is really productive. Um, and so it's, you know, it's another, it's another tool they have on their belt. I mean, so I have a little bit of a controversial question here.

Maybe it's not so controversial, but let's talk about support for other languages in spring. I mean, well, could we ever see a day that spring will be polyglot? I mean, don't get me wrong. Again, I have my little dukes right here. I'm not going to go too crazy. I love Java, but, uh. So to a certain extent, you know, we already have parts of spring that are polyglot. So really good examples would be spring cloud gateway and, uh, spring cloud data flows. So, you know, the idea of a, of an API gateway is that it doesn't, you know, your APIs are, for example, delivered via HTTP, which is, you know, completely agnostic of any one particular development environment. So the things that the spring cloud gateway can do around routing and filtering and, uh, you know, dynamically, uh, helping you to provide these APIs, it doesn't really care whether the implementation of those is a spring implementation or something else. Similarly, with, uh, with data flow, you know, the, the data that's flowing through the application and the microservices that are dealing with the processing of that data, it too does not care so much about how they are implemented. It just cares about what the API is that they offer. So we already have parts that are very cognizant of the fact that it's a very fragmented world. You know, there's an awful lot of stuff going on and it would be churlish of us to, uh, to, to be too prescriptive about how these things should be. So, uh, so, uh, just a couple more things that we can wrap up, but the, you also asked about, uh, containerization and, uh, you know, if people are running things on Kubernetes or not. So what was the, what was the state you determined there? Yeah, again, shocking numbers, uh, which point I think I'm, I'm giving away the, uh, the punchline here, but the, the state of, uh, of Kubernetes to coin a phrase again, but just, you know, just the plans.

And the, the rapid migration there. So we asked about this in two steps. Um, we first asked about, you know, are you putting your Spring Boot apps into a container? Um, and then we asked about plans to put those containers, deploy those containers to Kubernetes. Um, and, and so, um, the number here is 95% of the folks we surveyed said they either were, um, containerizing or they plan to containerize. And I can break that down. Uh, yeah, it's in the text again. 65% of Spring Boot apps are already going into containers. So there's another 30% that, you know, are on the plan or on the backlog and are going to be moved into containers. So, you know, as a working assumption for the Spring Team, it's like, yeah, we're, the apps are going to be containerized. Um, that's, that's, that's what we should optimize for. And did you, did you, uh, did you also ask around like where else things run? Like, it seems like Tomcat always comes up. Was that, I mean, that, that must've, uh, shown up here and there or other things, but what are other methods, other things people deploy on? I'm going to defer to Ben. Well, yeah, we did see, uh, we did see, uh, we did see in some instances of that kind of traditional style of usage where you would take an application server, you would package up into a, into a wall. And then you would deploy that alongside other applications, also sitting on the same application server. We did see that. Um, but what's, uh, interesting is that that number didn't, you know, it wasn't, it wasn't huge. Um, we see a lot of folks using the built-in packaging, uh, jar packaging of spring boot. So a lot of folks like that because it packages up the Tomcat with the application. So it becomes embedded. So you don't have to then go and find another Tomcat and provision infrastructure for that. You already have it. And your application can sit in isolation, which is great for microservices.

And then of course, uh, with, with Kubernetes, it's the containerization of that, that really opens up the door to it being deployable anywhere with any cloud provider. You know, all of them offer a very good, um, Kubernetes substrate. So you could, you could see, uh, how folks were starting to leverage that. And that's what surprised us the most was that the, the amount of folks leveraging Kubernetes was higher than we'd expected. But also it kind of, you know, if you're reading this survey for the first time and you're looking through it, it's also an indication of, you know, are you on the right path? If so many other people are doing this, um, you know, should you be considering it, um, as more of a priority perhaps than it currently is? Because, uh, it's, uh, it's clearly moving in that direction. Yeah. Yeah. And spring has some great tools to help you with that. You know, we can, you can build a container image in a few seconds with, uh, the build image, um, uh, tasks that we have inside the Maven and Gradle, uh, plugins for building spring boot apps. So you can get a container image without even using a Docker file, which I genuinely much prefer. If you've ever had to maintain 50 or so Docker files, you wouldn't really wish that on your worst enemy. Right, right. And so I'm very happy that, you know, there's an alternate trustworthy mechanism now built into, to spring boot. Yeah. You know, you know what I'm, what I've, when I was looking over this, this part, what it made me remember that one of the things I'm interested in, I don't know, I guess I could be excited about it. I mean, you know, that's not embarrassing, but what I'm sort of excited to see over the next, I don't know, let's say year or so, or some short amount of time is, you know, I've, I've like kind of gone out to the, and, and, uh, kind of picked up all these little threads about like how spring works well with Kubernetes.

Like there's all these little hooks and things like you were saying, but I'm still, I'm still really interested in seeing sort of like the, uh, you know, the, the comprehensive writeup of how it integrates together, like a book sort of form of like, you know, spring for Kubernetes or whatever. And I think, you know, like, like I've talked with, uh, a couple of people in this forum, as it were about, uh, the programming model for Kubernetes. And I think, you know, it made me realize how much time has been spent in spring to just make it easy to map in that with something as easy as just like, you know, every time you deploy a container in Kubernetes, you need to have a health check to tell you if the thing should be, uh, restarted or not. And that's just like in spring as, and, and then also handing in the configuration. So it seems like there's like this marrying up of like, here's what the architecture of a Kubernetes application looks like. I don't know if you're supposed to say that in the Kubernetes people would be all upset that you're being that specific about things, but that's fine. I think, I think as they used to say, the street finds its own uses for enterprise architectures. Prescription is the new block. That's right. And so you've got like that architecture and it would be interesting to see, see how like, you know, here is, uh, here's how like a spring thing, like maps into that architecture really easily. Cause all the little bits and pieces are there. But I think to your point, that's what makes it, that's what makes using Kubernetes easy. If you've already like bundled up your stuff in spring, cause it's, there's not a lot of, uh, there's not a lot of glue that isn't dry already. I guess you don't want the glue to be dry, right? You want it to be flexible, but, uh, well-defined. Yeah. It's both, uh, Kubernetes ready, but also Kubernetes friendly. You know, you can, you can use as much or as little of it as you like. Yeah. And the shout out to our friends in Wavefront or Tanzu Observability, as it's known now, there's some free starters there that, that when you deploy on Kubernetes really, you know, give you a, um, a lot more insight what your apps are doing.

It's not just Kubernetes, but generally, yeah, yeah, there's, there's, there's, there are more goodies coming out of that, that relationship. Yeah. I'm excited. Yeah. I think we're going to try and figure out that, figure out, share some of the insights we grok from Wavefront, uh, with folks. Yeah. So if anybody has any, uh, questions there about that, uh, I don't know, figure out, you could find us, the four of us on Twitter and say, Hey, I want to know more about this. What can Wavefront tell me or whatever. But actually, as we wrap up, uh, good segue there. I, I, uh, set myself up nicely. There is, you know, I want to ask John and Ben, uh, what questions do you wish you had asked, um, that you didn't? And, uh, you know, maybe we'll see next year. Oh my goodness. Where to start? You know, one of the challenges, and if you've ever done a survey, one of the challenges is that there are, so you want, once you have an audience, um, or folks that are willing to, to give you answers, you just want to keep them answering forever. Um, but we've all been on that survey where you get to a certain point and another screen comes up and you go, yeah, I'm kind of done here. Um, so we were, uh, kept very honest, I guess, with, uh, our research partner in terms of what was realistic. How many questions can you ask? Um, so they're trying to, and some of this is about, well, when you get the answers back, I wish I'd asked that question in a different way. Um, so trying to get specific, uh, there was stuff around cloud, you know, we had, we did, we, we got one slide on where people are deploying and it's kind of interesting to see public cloud versus hybrid on-prem. Um, but we didn't ask which clouds they were deploying to, you know, and we have a great relationship with the folks in Azure, with Azure Spring Cloud. Um, and maybe, um, and maybe kind of drilling into some of that space would be, would have been interesting. Um, drilling more into the, the kind of workloads folks are deploying on Spring, uh, figuring out what's, what's beneath those APIs.

Um, those would be kind of my, um, those would be kind of my, my go-tos. Ben, how about you? I think there's lots of things I'd like to do. I think John's absolutely right there. There, there, there are some, you know, itches that we want to scratch next time around asking some questions differently. But also if we can maybe do it for longer and get a wider subsection, that would be great, um, as well. Cause you know, we, uh, we would really benefit, I think from, from even greater data and, uh, probably sort of drilling into Spring a little bit, you know, finding out more about the APIs that folks are actually using and how they're using them in combination with each other. So, yeah, it's the art of the possible. John's absolutely right. We didn't want to bore people to tears with a super long survey. Yeah. Um, and we, we got exactly one meg of respondents. I mean, for our first time, this is pretty good. So. Yeah, we, we, I think we did okay. Yeah. I think that was very good. So, uh, clearly we had a buffer overrun. When, when, when, uh, when Diane from dimensional came back and said, we had a thousand and 24 respondents. I'm like, that's, that sounds perfect. It's destiny. I think, I think we, you know, we, we could have gone either, either route because we had a lot of content. Uh, I've seen some competitors in slide mode that looked pretty thin and I'm thinking they didn't have much of a choice. So, so I'm going to go with, you know, we had a choice, but we went with kind of slide mode. And also, you know, these, these graphics tie in very nicely with the new spring website that spring.io. And I'll, I'll give you the truth, the real story. Ben's being very diplomatic here.

And like there was, uh, so the state of Kubernetes, they took five weeks from having the results to actually publishing the report. Um, and I, and I, and I, and we wanted to publish this before SpringOne. So I had two weeks. So I basically, I got the same folks engaged that did the state of spring, the state of Kubernetes and said, let's do that again. Um, with as much reuse, let's not try and, you know, reinvent the thing so we can be agile and get this thing cranked out. And so the state of Kubernetes chose, uh, slide mode. So that's, that's why. No Europeans were harmed in the making of this. I still, I still class myself as a European. That's a whole other discussion. This is true. We had some good folks like Stefan Nicole in, in Belgium. And so he was very involved, you know, um, the whole spring team was kind of read into this to, you know, to make sure that they, uh, they concurred with it. And a lot of those folks. Yeah. Congratulations, you guys. Thanks. Bye. Bye now.

Tanzu Talk · 🍕 Get more pizza.