Developers are in fact the lifeblood of many of hte businesses today and how they operate.
We talk with Stephen about how RedMonk uses data to help companies understand developers and how developers are taking over the world.
Ben Newton: Welcome to the Masters of Data podcast, the podcast that brings the human to data. I'm your host, Ben Newton. This episode is in a series of interviews that I did on the floor of the Sumo Logic 2019 Illuminate conference. In this episode we talk with Stephen O'Grady, Co-Founder and Principal Analyst at Red Monk.
Ben Newton: Red Monk is an industry analyst firm focused on software developers and helps company understand and work with developers. In fact, Stephen wrote an amazing book called The New Kingmakers that explains the rising influence of software developers. So without any further ado, let's dig in.
Ben Newton: This is Ben Newton. Welcome you to another episode of Masters of Data podcast. I am, as always, I get to have a lot of really interesting guests at the Illuminate conference where you're meeting right here in the middle of it. You might be hearing a little in the background. I've got Stephen O'Grady with me here today. Welcome Stephen.
Stephen O'Grady: It's a pleasure.
Ben Newton: You are the Co-Founder of Red Monk, which has a lot of focus and a lot of people have heard of, I'm sure. Tell me a little bit more about your background. I mean, how did you end up, why did you start Red Monk? What led you that direction?
Stephen O'Grady: Originally I was a systems integrator and was running around flying everywhere, essentially part of teams that implemented software and didn't like to travel. I was basically it was full-time travel. I tried to find something that I could put those skills to work but travel less.
Stephen O'Grady: So happened into the analyst business, worked at a firm, unfortunately now defunct. I worked with my current business partner and James Governor and I had been working together for, I don't know, probably the better part of a year and just thought, "Hey, we could do some things differently. If we were running the show, we would focus on this and we would sort of run our business in a different fashion." So we did that ultimately.
Stephen O'Grady: We kind of looked around and said, "Hey, for example, developers are essentially being ignored by mostly analysts firms at the time." I thought, "Hey, why not take a shot?"
Ben Newton: Is that your core background is like computer science? Is that how you started out?
Stephen O'Grady: No, actually it's funny. Everyone assumes that my background is computer science. No, I was actually a history major in college.
Ben Newton: Really?
Stephen O'Grady: Yeah. I always tinkered with computers and so on, but then when I got out of college trying to get a job and the systems integration firm that I worked with first had a, what did they call it, it wasn't boot camp, but it was effectively boot camp. It was basically eight weeks of intensive residential training. Basically it's, I don't know that I call it the equivalent of a computer science degree, but basically you come out and you can program, you know what you're doing.
Ben Newton: What kind of history did you study?
Stephen O'Grady: Mostly European history, fair amount of American in there as well. Very, very little outside of those, because at the time everyone focuses on Western history.
Ben Newton: It's really funny. I came from a computer science background, but then I got out of grad school and I was so tired of it all for a while. Even though I did it as a career, I would refuse to read any science or science stuff. I only read history. It's like I was trying to catch up on all the stuff that you had.
Stephen O'Grady: There you go.
Ben Newton: That's awesome. I'm assuming that probably your kind of communication, writing probably came a lot out of that background, right?
Stephen O'Grady: Yeah. The history degree has actually been surprisingly useful because you have to write a lot. And as an analyst we break quite a bit. But also, and this is my bias probably shows at times in the sense that I tend to look at things in this industry from a historical perspective, how things have evolved because for better for worse we're so focused on, in many cases, building the future that we tend to forget about our past. It's been useful to have that degree to take a step back at times and see, "Okay, hey, we've been through something like this before or we've seen trends like this before. What happens and what does that tell us about what's going to happen moving forward?"
Ben Newton: One thing we were talking about a little bit, so Red Monk, you guys are really focused on developer culture and a better understanding of developer community. Tell me a little bit about more about how you interact with that community. You were telling me that you actually go out and learn about what they're doing and [crosstalk 00:04:02]? How do you actually find out what developers are doing?
Stephen O'Grady: When we started the firm way back when, a big part of our focus became developers because as I said, we looked around and things were being ignored. So in other words, we would go out and talk to developers and they were using tons of [inaudible 00:04:17] as one example. And you'd go talk to analysts firms and no one spent any time talking about it because in their view, these are developers, developers didn't have any money, no one was paying for it.
Ben Newton: Interesting.
Stephen O'Grady: So there was this sort of break between what was actually getting used and what analysts and firms and vendors alike talked about. We looked at it and said, This doesn't make any sense. If stuff's getting used it's important." So that's kind of how we got onto that track, so to speak.
Stephen O'Grady: Then in the early days people would come to us and say, using the example of MySQL, like, "How many running instances of MySQL are there?" We'd say, "We have no idea. There's no way that can't be measured."
Stephen O'Grady: In the early days when people came to us and wanted quantitative measurements, we would basically say we can't. There's no way that that data can be collected because anybody can download a instance of MySQL and run it in a thousand places inside of enterprise. We'll never know. There's no way to measure that.
Stephen O'Grady: What we ended up evolving over time was looking at, wait, are there sort of areas where developers are coming together and congregating that we can take a look at and see if there are patterns there? So GitHub became one. Stack Overflow became another, Hacker News. Any of these different sources where you can begin to look at, "Okay, I can't tell you, I still can't tell you how many instances of MySQL there are, but we can begin to measure performance and behavior."
Stephen O'Grady: In other words, probably the canonical example of this, and one that people may be familiar with is we do these programming language rankings. We got the idea from two folks, Drew Conway and John Miles White who are big data science people. They had basically as a lark measured the performance of languages on GitHub and the performance of languages on Stack Overflow and then correlated them.
Ben Newton: Oh, interesting.
Stephen O'Grady: So we thought, "Hey, this is a great idea." So we just repeated the process. We'd been doing it for, I don't know, probably close to 10 years now.
Ben Newton: Really? That long?
Stephen O'Grady: And so you can begin to see over time languages go up, languages come down. What does that tell us about what developers for example, are using?
Ben Newton: Are GitHub in Stack Overflow pretty well correlated?
Stephen O'Grady: Pretty tightly. You definitely see some differences. One of the examples, because two of the people who, Jeff Atwood and Joel Spolsky were very Microsoft focused people and they started Stack Overflow. That tends to, Microsoft languages tend to outperform. So something like C Sharp is higher ranked on Stack Overflow than it is in GitHub. But for the most part outside of some anomalies here and there, they tend to be pretty close.
Ben Newton: I mean I think this stuff is fascinating. What are some of the big trends that you're seeing in that area?
Stephen O'Grady: We've seen in sort of in recent years, it's very difficult to move up the rankings in any meaningful way, particularly to get into the top 10, because as you might imagine, if you're measuring either source code in form of GitHub or questions, these are a cumulative measures. So in other words, they tend to sort of accrue over time. More and more source code gets written. So somebody coming in, like if I invented a great new language today, I'm starting from zero where they're starting from, they'd been doing this for 10 years.
Stephen O'Grady: In spite of that, we've seen a couple languages that have really risen up the rankings pretty quickly. So Go had a pretty good charge a number of years ago. It's kind of plateaued in the 15, 14, 15 range. Swift exploded. What was that, two, three years ago, when Apple released it. Most recently, however, the fastest grower for us had been TypeScript.
Ben Newton: Really?
Stephen O'Grady: Yeah. So TypeScript has kind of languished in the state of 30s and 40s for a little while. And then I'd have to go back and look at the actual timing. I think it was like two years ago, basically just started up the charts and actually broke into the top 10 for the first time of the last rankings.
Ben Newton: What are people using? I'm not as familiar with TypeScript.
Ben Newton: No, that makes a lot of sense. I mean, what other kind of tidbits are you able to pull out by watching GitHub and Stack Overflow?
Stephen O'Grady: Languages obviously are one of the things we do a lot of. Like I said, companies will come to us and say, "Hey, I want to know again, like how many running instances there are." And we tell them, as I just said, "Look, we can't measure that." But what we do instead is you can measure things relative to one another much as we do for languages. In any given category of database, web server, application server and so on, we can't necessarily measure the instances, but we can see, all right, how are these or four or five competitive projects performing relative to one another. Are they higher, are they lower? We don't really care about any one property. In other words, if you're ahead on Stack Overflow, okay, this is interesting but it doesn't tell us all that much. Because as we just talked about, you do have some outliers and some anomalies like C Sharp outperforming because of the site was founded by some Microsoft random people.
Stephen O'Grady: What we try to do instead is look for are there patterns between different properties. That's why we correlate, for example, GitHub and Stack Overflow. You can sort of go in and try to correlate property after property after property. We tell people all the time is one of these is not terribly interesting. If I see the same thing twice or the same thing three times, or the same pattern four or five times, that tells me something. That'll tell me something about usage. It'll tell me something about interest. It'll tell me something about traction.
Stephen O'Grady: I mean we do a lot of sort of looking at all kinds of things like frameworks and languages and databases and what are they doing, how are they performing and what do they look like next to one another.
Ben Newton: Back to your My Sequel case, today you would figure out, I don't know, like the buzz around My Sequel about. Are people asking questions about it? Are people asking questions on Stack Overflow that they're using it?
Stephen O'Grady: Yep.
Ben Newton: So it's a proxy to usage?
Stephen O'Grady: Exactly.
Ben Newton: Do you feel like in that way, do you feel like you get a lens earlier than maybe some other analysts yet to what's going on doing that?
Stephen O'Grady: I think so. Certainly in some areas, but it's kind of always been the case for us in the sense that going back to the [inaudible 00:10:24] example. If you at the time, this is less true now, but certainly was true when we got started. If you went in and talked to businesses, they would all say, "Yeah, we're not using that."
Ben Newton: Because they actually were.
Stephen O'Grady: They actually were. The classic MySQL sale, in fact, like if you're a MySQL salesperson, what you used to do is walk into a large bank or whatever and they'd tell you, "Hey, we're not using any of that. We don't need to talk about sort of licensing." And you would come back and say, "Well that's interesting, but [inaudible 00:10:52] MySQL has been downloaded from IPs registered to your company a thousand times. So odds are pretty good there's some of it, you're just not aware of it.
Stephen O'Grady: The menace is, is that for us, we have spent more time talking to the people who are the users, people using MySQL, then we have talking to the CIOs. We need to talk to CIOs. We need to know what their priorities are. But when you actually talk to the people in the trenches who are actually using the technology, you tend to see things a little earlier. The cloud is a classic example.
Stephen O'Grady: We were talking to people. So Amazon essentially created the cloud market in March and August of 2006 so by early 2007 we were talking to developers and developers using a lot of this stuff. We running around saying, "Look, this is a thing. You need to pay attention to this." All of the big companies consistently come back and say things like, "Hey, that can't do what this piece of physical hardware can do. My stuff's more reliant, more reliable, more faster performing," and so on. We were saying, "Hey, that's great, but if I'm a developer, I can have this thing up in 90 seconds." That's a big difference.
Stephen O'Grady: I think in terms of just where we sit and a constituency that we talk to, we tend to see things a little bit earlier than some other firms.
Ben Newton: Yeah, no, that makes a lot of sense. One interesting thing is, as you and I were talking before, just had a conversation about kind of cultural changes in companies right now. Kind of the flip side, we were talking a little bit more about before this, I was talking a little bit more about centralized or decentralized and kind of grass roots and in some sense with that in [inaudible 00:12:25], I don't know, code for is really developer led. I'd be really interested to hear what you're seeing because in some sense I've been doing this for almost 20 years. And to your point earlier, any developers where like you put them in the back room before, they write some code and then, whereas now it feels like the developers are actually leading the charge [crosstalk 00:12:46] forward.
Stephen O'Grady: For sure.
Ben Newton: What kind of things are you seeing in that way?
Stephen O'Grady: So it's interesting. So if you go back to, I want to say it's 2004 maybe, somewhere in that timeframe, a gentleman by the name of Nicholas Carr wrote a book called Does IT Matter? Essentially the general gist of the book essentially was, look, this is not differentiating for you. This is not essentially a strategic initiative. And this accompanied, in many cases, businesses outsourcing all their development saying like, "Hey, this is not important to me. Whatever, I'll just send requirements to oversee shops, because I can pay a developer in the Philippines, for example, a fraction when I have to pay somebody sort of in the US," or wherever. This was essentially, that was the way things were done and certainly when we were getting started.
Stephen O'Grady: We just looked around and said, again, "This doesn't make a ton of sense," because by that point, obviously the internet was a thing and digital businesses were becoming more and more of a thing.
Stephen O'Grady: So anyhow, what's happened over the last say decade is that digital business has become just business. It's very difficult in many cases to have a purely offline business and what a lot of these businesses have figured out is that, okay, if I'm going to be a digital business, that stuff that I thought I could outsource, that wasn't such a great idea, and therefore developers are important. Developers are in fact the lifeblood of many of the businesses today and how they operate.
Stephen O'Grady: If you don't have good software development organization, you're behind. What that means is that we've written, I wrote a book on this a number of years ago called The New Kingmakers. So essentially the idea that developers and using that term loosely.
Ben Newton: I love that book, by the way.
Stephen O'Grady: The developer though, it could be anybody. It could be any kind of practitioner. In some cases it's a [inaudible 00:14:24] man, it's an operator, whomever. It's the people in the trenches. They are more often than not the ones who are setting the agenda moving forward. Businesses that don't do that tend to fall behind.
Ben Newton: When did you write that book? It's been a few years now.
Stephen O'Grady: It was 2013 I think.
Ben Newton: What pieces of that do you feel like are born out? Is that, I mean, is it pretty much, is it coming about as you predicted or are you seeing, [crosstalk 00:14:43]?
Stephen O'Grady: It's basically what we expected, honestly. I'm sure if we go back, I'm sure there's some things, some hits and misses and so on. But the gist of the idea, which was hey look, you treated developers for years as this sort of fungible resource, this non-strategic sort of resource that you could farm out to wherever. It doesn't matter. And in fact they are absolutely the lifeblood of your organization. That's been picked up so many times. We have businesses you've never talked to who are on stage giving talks about hey new kingmakers and developers are the most important resource we have and so on. It's sort of largely played out as we expected.
Ben Newton: With that idea of new kingmakers, because I definitely say I've read the book a couple of times and I think it's been really helpful for me to do that. Because when I started out, like I said, I was in that weird before DevOps was a thing, I was one of those people that got trapped in between, as like I could write code but I also was doing operations so I was like the person who got abused in the middle.
Ben Newton: A lot of what you were talking about kind of connected with me on a personal level because I saw that. Did you or what you're seeing in that context you, do you feel like the development organizations, are they more distributed or are they becoming, are they centralizing [inaudible 00:15:58] in a sense like they give you a perfect example. Are they more closer to the business itself? Is in some sense like for example, if there's a business unit that does acts, like they make this product, are the developers now sitting right there close to that GM to that business unit?
Stephen O'Grady: It depends on the business obviously. It's a classic consultant or analyst answer. It depends, but it really does. There are different ways to do it. What I would say is, is that there are a lot of reasons for this. Open Source is one. Clouds another, softwares and service. Basically, it's become much easier for lines of business to essentially run their own show. It used to be, if you go back 10 years ago, 15 years ago, 20 years ago, if you're a line of business and you want to do anything, you need to go out and find some hardware. You need to find somewhere to put that hardware. You need to go out and buy licenses for things and get all that stuff up and running. That's why most businesses have centralized IT.
Stephen O'Grady: Today, in many cases you see lines of business who will say things like, "Well I can just go get this on cloud." He goes, "We'll go to IT and IT will say, "Yeah, we'll get to that in six months." And they say, "Yeah, no, I'm not going to wait six months."
Ben Newton: This is not going to work for me.
Stephen O'Grady: Instead, you know what, I'll go to Amazon or whomever and I'll have something up in seconds and I can get working. So you definitely see a lot more technology autonomy within the lines of business. That's driven again by, obviously, technical resources, developers, operators [inaudible 00:17:20], DBAs in some cases and so on. You see a lot more of that.
Stephen O'Grady: But there really hasn't been a, I don't think there's no sort of consistent pattern that was like IT. Everybody tends to do it a little differently, but sort of largely, what we've seen is that businesses that want to be successful and want to move forward, their IT has to get better. The days of, hey, I'll get to that in six months, if those aren't gone, they're on their way out because that's just not an acceptable answer for businesses today.
Ben Newton: That makes a lot of sense. I remember my first couple of jobs I was like in, like I said, I was kind of in that DevOps ring. I had a pager. I didn't want to [crosstalk 00:17:58] , but I had a pager. None of the developers had a a pager. I remember I was actually talking to one of our younger developers as Sumo Logic that I was working with when I was in product management. He found it really strange the developer wouldn't have a pager. That's the world he's grown up in. It's really interesting how it's changed because it's a cultural change because now he feels responsibility for his code. Whereas, I remember working with developers. It's not like they didn't care, but they felt no responsibility for it.
Stephen O'Grady: No, for sure. That's definitely changed. I come from that same sort of era. In other words, I was a developer for these SIs. I was on call, but other words, the operator was the one who would make the call. In other words, they had run books and hey, we're going to sort of figure this stuff out. And if something doesn't work in the middle of the night, we end up getting paged, it sucks. I don't want to go back to that.
Stephen O'Grady: But there was this idea that, hey, I've written this thing, now it's your problem. That has absolutely changed because we talk a lot about sort of DevOps. We talk a lot about these sort of cultural changes and what DevOps implies, just by the words itself is a conflation, this coming together of developers and operations.
Stephen O'Grady: You have extreme examples like a Netflix for example, where it's, hey, we have conflicts between our developers and operators. So all the operators are gone and now the developers [inaudible 00:19:21]. That's an extreme example. Most organizations don't take that path, but basically what you end up finding in most cases is that as these two sort of groups come together, well guess what, the responsibilities change. So if you grew up in the school of like, hey, I'm a developer and I just sort of write this stuff and then somebody else's problem, that's not going to really work in sort of an area where the developers have operational responsibilities as well.
Ben Newton: No, absolutely. So kind of back to the thing you were saying before, you guys basically have an insight that others wouldn't have. You're kind of in a workflow a little bit. You get to see what's coming. What are you watching now you don't think other people are watching?
Stephen O'Grady: It's a great question. I mean there's so many different things. Honestly, one of the things that we frankly we struggle with to some extent and businesses definitely struggle with, is just how to make sense of everything. So in other words, I talk about this a lot, but when I was a systems integrator, there was basically one way that everybody built applications. You had a three tier architecture, there was a presentation tier, a business tier, business logic tier and a data tier. You were probably going to pick from two or three application servers, two or three databases. There weren't that many choices to make. And you certainly didn't have to make choices as far as the approach. Like I said, this was kind of an industry standard. These days there are tons of options, tons of options. So how do you begin to sort through them? What makes sense?
Stephen O'Grady: That's one of the things that we're trying to sort of understand in terms of basic questions like, "Okay, we have ..." for example, you could go and sort of build and wire everything together yourself on top of a cloud or you can go the complete opposite end of the spectrum and purchase something that is all in one. Hermetically sealed environment, which you don't have any control over, but you also don't have to try to scale yourself.
Stephen O'Grady: What's the relative performance of these? How are they being used? Because it's not likely to go back to the world I came from at least, which was like, hey, there's one approach, but it's also probably not the case that we're going to have a dozen different ways to do things. There has to be some consolidation or has to be some standardization.
Stephen O'Grady: So when you take a step back, where do containers fit? Where does functions and service fit? Where do you use one? Where do you use another? There are questions that we're trying to sort of understand better in terms of just what's trending up, what's trending down and who's using what and why.
Ben Newton: No, that makes a lot of sense. It makes a lot of sense. I can pay you for one. I'm pretty confused by everything that's going on.
Stephen O'Grady: There's a lot going on.
Ben Newton: Yeah, there really is. I don't know, it sounds like you've been around as long as I have. I mean it's a lot faster than it used to be.
Stephen O'Grady: It's funny you say that, because I gave a talk on, was it Monday, I guess, and one of the things I was pointing out, this is to a bunch of executives is the rate of essentially technical changes accelerating. In other words, if you go back and look at things like the Java middleware era to platform as a service to open source platform as a service to containers to serverless. The timeframe in between those is something like, what is it, 10 years, six years, two years, one year, something like that. In other words, the point is whatever the actual numbers are, the point is essentially that not only do we have more options, more options are arriving more quickly. There are a lot of developers actually who are in the boat of this is too much. I can't deal with this and I can't cope with this amount of change. We try to help with that where we can.
Ben Newton: Developer psychologist.
Stephen O'Grady: There you go.
Ben Newton: Well Stephen, this has been amazing. I love what you guys do and I appreciate you taking the time to sit down with me.
Stephen O'Grady: Not at all. My pleasure.
Ben Newton: Thanks everybody again for listening to another episode of Masters of Data. Check us out in your favorite podcast app and rate us and review so other people can find us. Thanks for listening.
Speaker 3: Masters of Data is brought to you by Sumo Logic. Sumo Logic is a cloud native machine data analytics platform, delivering real time continuous intelligence as a service to build, run, and secure modern applications. Sumo Logic empowers the people who power modern business. For more information, go to Sumologic.com. For more on Masters of Data, go to mastersofdata.com and subscribe and spread the word by rating us on iTunes or your favorite podcast app.