The State of CI/CD in 2024 with CDF’s Dadisi Sanyika and CloudBees’ Sacha Labourey

This is Text Drunk tv. Hey everyone, welcome back here to Tech Drunk tv. I got a really great two guests here.

We got a really big topic I want to jump into. It's, it's, uh, it's something we've been discussing around Tech Drunk TV for a while. Uh, well not a while, but for the last week or two, and I hope we're gonna tie a bow on it today.

Let me first introduce you to my guest. First of all, he's becoming somewhat of a regular on, uh, on Techstrong TV lately. He is the, uh, current chairman for the CDF, that's the CD Foundation, actually, governing board chair is the right word.

Um, it's the DC Sonika and I, I think I'm almost starting to get your name right to dc. It only took me about six times on our show. That was perfect.

That was actually perfect. I'm, I'm, I think I'm ready to just hang out with you Reg regularly now. We, Yeah.

No, once, once, once I get your name right, you're a lifer man. Yeah. Um, speaking of lifers, let me introduce you to our next guest.

He's a good friend of mine. com in the 2013 going into 2014. They were one of the very first companies that got behind us.

He's the co-founder. He was the CEO at the time. He's currently, uh, chief Strategy Strategy officer of CloudBees joining us, I think from Switzerland back home today.

Sasha Lare, um, Sasha, And I have to say welcome. Say I you say my name wonderfully Well, it took you two years. It took me two years for that one, you know, know how I remember Labor Day.

That's what exactly, that was the thing that you told me that that got it to me. Um, Sasha Lare, but Sasha, by the way, is also a governing board member at the CDF. So we have two CDF board members here.

Um, guys, topic of discussion is the CDFs 2024 State of CICD report. Before I jump into it though, I just wanna quickly tell the audience, look, CD Foundation is a, a, I guess a daughter foundation of the Linux Foundation here on Tech Drunk tv. We do like a once a month show called the CD pipelines, uh, in partnership with CDF.

The DC's been on it a couple of times every month, we, we, we delve into a area of interest to the CDF and to our audience. Do check that out on Tech Drunk tv. CDF is also currently the c the DC himself, but the CDF is also, as we were taping this in Seattle at the open source summit running, I guess it's a summit within the summit kind of thing.

Yes. Right. The conference within the conference, yes, definitely.

Yes. Uh, and the, uh, CD con just wrapped up yesterday. Excellent.

And, uh, at that, you know, this report came out. We, we've discussed it on Textron gang, we've discussed it at, at live. We streamed live yesterday.

I think Mitchell was on with the dc but we're gonna discuss it a little bit here. D dc If you wouldn't mind, I'd like to kick it off by saying, what were you guys looking at in doing this report? What was the mission?

What was the goal of, of running this report? Uh, great question. The goal of the report is really to get a deeper understanding of how businesses are utilizing, um, CD DevOp practices, uh, and understanding if they're engaged with them and the impact of that engagement.

Um, it's highlights the usage. And so what you really have this opportunity to do is to, as a business, you get to kind of compare where you are, um, to your peers and then really see how that is impacting someone else's business to kind of refer to how it's impacting your business. It helps you find gaps, and then you can take action, I guess, those gaps.

Excellent. Sasha, CICD, who's been in it longer than you at in the CloudBees, right. com, I think Jenkins had, I don't know, maybe an 85, 90% market share of the CICD market.

I think the last CDF report that was done showed that Jenkins still had something like a 43% market share of, of the CICD market. Yeah, it's, it's pretty astonishing to think that we're close to 20 years, right? We're soon gonna, is it this year?

Next year. But it, it's, it's around the corner, essentially 20 years. How many open source projects are so active after 20 years, both in terms of usage, but if you could look at the health of the community, it's just, uh, amazing.

And, and, and so I, I, I, I, I wish well to all of the open source project to, to have such a, an amazing lifetime. Um, uh, I, I, you know, I was lucky enough to be involved in, in, in JBoss before that. It's also still active.

Um, so you have few projects that have this type of l longevity, longevity l I'm forgetting a few letters. I have the same problem with that word L longevity. Longevity.

Um, and, uh, yeah, so, so yeah, it is very exciting. Absolutely. So guys, all three of us live in somewhat of a DevOps, CICD bubble.

Probably most of the people we know know all the terms that we're throwing around here, all the acronyms. And they probably practice DevOps practices and CICD themselves. But there's a big world outside the bubble.

And every once in a while, it's good to remember that that world out there is bigger than this world in here probably. And, and these kinds of reports, I think serve to remind us of that. Right.

So the DC you're, you are the governing board chair. I'd like to hear from you, I mean, I, I know I've looked at the report too, but for our audience, what are the, what are the three big things that really kind of jumped out to you that you want to, you want our audience to be aware of here? I, I, well, the, the thing I'm gonna start with, the single biggest item in my view was the impact for new users and how new developers were unaware.

It seems to be an increasing number of new developers that are unaware of the tool sets, um, that weren't familiar with CICD. And for me personally, that that highlights that we need to, as the CDF, have a mission of how do we reach, you know, young people and students where they are. Um, how do we get and partner with universities?

How do we partner with, you know, educational OPOs, um, to prepare them to enter, you know, the enterprise workforce? And I think that as a service, as a group is the most impactful thing that we can do. Um, and then beyond that, that the usage itself seems to be kind of changing that there's this midst use that is reflecting cloud usage.

And so you have this mix of internal tools and external tools, and when it's a mixed use environment that, um, if you're using tools that are payment based and, and external based, that you seem to be having more success. And personally, that is really interesting to me because, uh, of my involvement in the CDF, of course. And, um, I don't know if that's, I'm still trying to like, figure out what that impact is going to be long term and how we can effectively, uh, help either on-prem systems or, you know, external pri.

'cause it's like this mixed use kind of of value that was, uh, impacted. That also was super, super, uh, strong to me in how we need to look at the problem and understand all of the things that people are challenged with. So those are my two biggest points.

And I think the, the third, um, was just the overall communication style. And I got some really interesting feedback about, um, the terminology that we're using to reach users. And I wanna be sure that it's really inclusive of people, uh, as you just mentioned, who are outside of our bubble.

Um, and that how we communicate this information to, and the impact of this information to business leaders who might not understand the value of, um, of, uh, continuous delivery and continuous delivery in DevOp practices. It is our opportunity to really, really highlight the importance of this information. And so we need to meet them where they are.

Fair enough. Sasha, what about, what were your impressions of the report? So, um, first, uh, uh, jokingly, uh, when sometimes I look at those numbers, uh, I, I can't believe some numbers.

Uh, I'm so much in the bubble when you see that you have a group of people who don't use, uh, code repositories and ci, I don't know. I felt the same way. Horrible.

You Know, I felt the same way. I know. I was like, come on, it's this real, I, I felt, I felt the platform engineering people were, were poisoning the surveyors.

I'm only kidding. Um, I'm kidding. I'm kidding.

But seriously, in this day and age, how do you not use a code repo? I know. How do you not version control your code?

Yeah. So sometimes I, I, I, well, but anyway, but I, it's, you know, independently of the big ones like source code and so on, you still have some tools that people clearly don't use. And that to me is very interesting because it means that the education, um, uh, that, that that is we talk, was talking about is, is very important.

And if you think about how you learn continuous delivery, DevOps ci, it, it's not one of those thing where you go to university and, and it's like a semester of ci, cd and DevOps. It's not any of that. Um, and if you join a large company, when you leave your studies, when you finish your studies, maybe you're not gonna even gonna ask yourself this question because you are part of a machine and you're being asked to, you know, your Lego block within a big system.

And, and it just happens to work. Um, if you work in a very small company, might not need all of that sophistication. And so you kind of learn by observation, you learn by failing, you learn by growing and realizing that it's a mess.

Um, so, so to me that that's, uh, that's, uh, uh, that, so, so it, it, it's why those reports are very important because they show the benchmark essentially, of where we stand as a, as an industry. Um, I also feel sometimes, um, like some numbers were, were less good. And, uh, when they're less good, it, it, it's a source for concern, obviously.

But sometimes there is also a part of me that think, are we being more honest about it? Because if I look at some of the data I remember at, you know, in the early DevOps era, you had all of those companies coming up on stage and talking about the amazing results they had. But if you were double clicking under, you realize that you had a, a couple of leading edge teams doing amazing things, uh, traveling across the globe to talk about it and a market their way.

But if you were to really look at the broader range of, of teams, uh, you know, the thousands of other teams around you would see that it's not quite there yet. And so maybe it's also this, right? It's people, people speaking up being more visible and saying, no, we don't do that.

We don't do that yet. Uh, because, uh, yeah, we're, well, well into DevOps. But, uh, reality is that a lot of companies are still very far into their, uh, their processes.

I, I don't disagree at all. Um, couple of interesting things that I, I noted is, and, and to me it was almost counterintuitive. The longer you have been a developer, the more years of experience you have as a developer, the more likely you are to use CICD.

Right. You know, if you looked at the, the charts, so back to the point of, yeah, maybe people aren't being taught CICD when they first come into the industry, when they're graduating school or they get their first developer jobs, they, they, they don't have that CICD foundation, no pun intended, right. Um, to, to, to use it.

But certainly as they progress along their, their career path, it's almost like, I guess they get introduced to it and they never go back, right? That's the way they do software from now on. Do you think that, you know, where I would think CICD, okay, it's 20 years old, but it's still a relatively newer thing.

The younger people are the people who are embrace, the younger developers are the people who are embracing it. But it, it's opposite. What do you, what do you think the DC Well, I, this is going to, this, this makes me wonder what's happening with platforming, uh, in general, um, as you are young and you're engaging in a new company and you come to a, a group that has, you know, you're in a large company, like you said, it's a machine.

And when you go in and you push a button and everything's set up and you don't see what happens behind the scenes, you just know you do your part, push the button and go away. Do you really understand that what's happening behind that? Or are we giving the education about all of the steps that go in to get that this new piece of code that you've written into the test environment, into the production environment?

And are we making it fully aware of that process and those tools? And so what I am curious about is the gaps in information of how we're communicating to young people about where, what they're using. And, you know, any newcomers to, um, development, you, you really need to be aware of the tools that you're using and the impact of those tools to be able to contribute to them, uh, and make changes to the system and, and prove it.

Uh, and I'm also con not concerned, but wondering how that impact for new developers and that engagement and that knowledge is being transferred. Um, so one of the things that even as I took this role as the, uh, the chairman of the board for, uh, the CDF, I started reaching out to educational hospitals to say, to ask that question, like, how are in your universities are you teaching these, these fundamentals and continuous delivery? I was surprised that most of the colleges that I talked to were like, wait, why would we do that?

Um, and so it's an opportunity. I see the opportunities. It was great to see that in the report, to say, oh, daddy, see that work?

You already started. Yeah, there's a need for it. Let's do something with it.

And so, Um, I I agree. Yeah, Go ahead. No, no, no.

That did DC That is, That was the point I was making. Yeah. Yeah.

Yeah. So I mean, I agree. I, you know, maybe I'm the boy who looks at the glass half full.

But, um, I think it's a tremendous opportunity. And Asha, I'm gonna ask you to put your CloudBees hat on here. As a, as a commercial provider in the CICD space, that has to be a tremendous opportunity.

How do you exploit it? How do you fulfill, how do you, how do we get more people you to doing CICD? Well, I'll tell you, um, in any project we have with a company, uh, the adoption phase is, is the most important.

Uh, and and that's because even though you might sell, like a company might have 2000 developers and they're gonna buy 2000 seats, and then when you actually count how many seats are being used versus how much they vote, you, you see a a a big difference initially. And the ramp up phase, uh, can be very slow. And so, uh, we're obviously doing everything we can to promote internally and to help customers, uh, be successful.

But by and large, it, it needs to be a very strong movement within the company to constantly educate, uh, and, and make people realize that, uh, the continuous way is the best way. And, and so, um, I'm just talking about within a company here, right? But I guess that's my point.

It's pretty obvious that outside on the market with everything we discussed, right, it's, it's a talent that must be learned, uh, not a talent. It's, it's a knowledge that must be, uh, learned. And, and that is, is is doing an amazing job, uh, uh, driving the continuous delivery foundation towards, uh, more education.

We, we tend to see the CDF through the lens of, of the projects that are hosted at the CDF, those are absolutely critical project like Jenkins, obviously Spinnaker, Tecton and, and others. But the educational aspect is absolutely critical. The normalization of effort and trying to define, uh, uh, the best ways of working.

That's why organization should be joining. That's why it makes sense to promote this, because then as a user of that technology, you yourself, you are gonna be facing the problem of adoption within your own company. And if you can work, uh, uh, with, uh, the CDF, for example, uh, to, to accelerate the adoption internally, I think it's, uh, it's a win-win.

Absolutely. The DC I, I, you know, I've been covering the CDF role since the CDF started actually, um, plans for ne, I don't want to jump the shark, but plans for next year's state of CICD report. 'cause what I'd love to see is be able to take a five year slice of this and see, you know, to see trends.

'cause sometimes one year, you know, there could be a few things that throw that throw the whole, you know, thing out of outta whack. But you know, when you start looking at it over a three to five year period, you, you really can't see chance what, what plans going forward that we can continue following some of the things that kind of stick out to us on this year's report. Yeah, don't, well, um, I, I liked that you're putting me on the spot, by the way.

That's, that's great. I'm sorry. No, no, it's perfect.

It's perfect because it, it is, as I read the report and as I, uh, was giving my comments, one of the things that I I, that is going to happen is, you know, I will sit down with, uh, slash data and really analyze what we want to achieve by the report long term. Um, because that is really, really important and it should be lined up to be super specific to those goals, and being sure that the report is valuable to the community, um, in all of our community, right? Because it's gotta really clearly reach, as I was saying in the beginning, that's really clearly reached the people outside of the continuous delivery ecosystem who we want to understand the value of this, this work.

And then it also needs to, you know, both help the projects understand the places that they need to improve. And finally it helps us, it needs to help us understand how to build that educational path. Um, for those who didn't go to the CD con, uh, there was a great talk about it was, I think the talk was titled, do We Need another, um, CICV tool?

And it looked at the ecosystem of tools and when they came into the market and what tool the speaker, and it was clearly from the speaker's perspective of course, and what tool that was most valuable to them on that journey. And what was really clever about the talk is they listed the first tool that they used all the way up through the tools that they use now. And the one that they used the most was the very first tool that they used, because that's the one that can rely on.

And so if you start this educational process in that, in that vein, and of course that that tool was Jenkins, um, and you know, and be sure that people are educated, young people are educated about how and why that's important to understand what those two stoops they can actually contribute and build towards it, right? Because you, you start foundationally, you look at what's out there, and then you understand everything that's in between. And so when you're sitting in front of that IDP and you're looking at that platform, you know what's going on underneath, and then you can, you can make some judgements about what you wanna improve, how you wanna work.

Um, but that does start with educating ourselves and understanding that information as it comes in on the report. You know, what I find interesting too, when CICD historically, traditionally was a space that was kind of dominated by open source, whether it was Jenkins or, or, uh, Spinnaker Deton, you know, there was some, there's some superstars in that lineup. Um, even on the cloud native side, if you look at Argo as a GI ops slash CICD tool, it's the third most popular project in CNCF.

But yet I think we're seeing some company and, and Sasha, quite frankly, even cloudy, move away from that open source, um, foundation where, you know, the newest cloud based platform, quite frankly is not Jenkins based, or you don't need to be running Jenkins and, and so forth. They, you know, they clearly say that is, are we seeing a, and I'm wondering if it's in these numbers somewhere, uh, like a waning influence of open source, a diminishing influence of open source within the CICD space. And I know that's controversial in the DC men, you're making all kinds of eyes at me, so I'm gonna let you go first on this one.

I, I don't think so. I, I, I hear what you're saying. I think that the, the business of doing business is, is always a good business to be in.

And, um, that is also a concern of mine with the Continuous Delivery Foundation, that we have a great community of vendors, uh, that support us and we want to continue to support them. Um, there is a lot of opportunity in the space, because what's happening is, as we go more along the lines of like MOPS and DataOps, there's more things to do and more ways to look at the tooling. I think foundationally what happens is that, you know, something happens great in the open source community.

People either draw into it in a way that, you know, their intent is, I wanna build a business, or a community comes up around it and says, Hey, we really love this tool, and then there's an opportunity to build business around that. Um, that's what I see in the space. Um, but of course, I am not, uh, in the business of like CloudBees of DevOps.

And so I would really, really be interested to hear, uh, Sasha's take on that. So I, I think in, in, in any space, and that includes CICD, open source is absolutely everywhere. The internet is running on open source.

Uh, Linux is a, a, a mo most widely used operating system for from embedded to, to, to, to multi-core systems or AI learning tools. Um, look at Kubernetes, that's, that's, that's a cloud, right? Uh, like a decade ago, everybody was thinking, how can we, uh, abstract us, uh, be abstracted away from the A-W-S-A-P-I and we're trying to mimic this, wasn't that right?

It was a, a calipers back in the days, wasn't that it was a completely new thing that came out of open source. So everything tends to be redefined. It's, it's, it's, it's really the baseline for pretty much any infrastructure technology in the world.

Um, but, um, as that is said, uh, uh, it's great to be in the business of making money. And so I, I think, uh, SaaS in particular, uh, created a new wave of, uh, uh, challenges for, for organization on how they can make money. And we've seen for alpha decade now, all of those open source being shifted to pseudo open source licenses, business licenses and what's not.

And what, what's that's saying is that the, the, the business of making money with open source has shifted. Um, I've been, uh, doing open source business model since 2001, right? Uh, from consulting training in initially to the subscription business model, allow Red Hats that serves them pretty well, um, to what we're seeing now with SaaS.

And this model needs to be redefined constantly. Uh, that's part of the challenge. Uh, but uh, you have to do with it because open source is here, is strong, is here to stay and is not going anywhere.

Uh, and for the record, the cloud is platform is based on tech time. So here we go. There you go.

And that, that's the first time I've heard that on the record. So interesting. You heard it here first on text, drug, tv, and exclusive.

Anyway, hey guys, we're, we're, we're way past time, but I, I enjoyed the conversation, so I didn't want to add, but did DC for people who want to maybe download the report and dig in for themselves? 'cause there is, there's an awful lot in this report. If you're watching this, do download the report, go do down, download and, and take a good look at it.

There's a lot of nuggets in there. D dc how can they get it? Well, we actually have it on the front page of our, um, website at CD Foundation, and, uh, they'll link be right there for you on the page.

Fantastic. Um, Sasha, my friend, good to see you. I, I feel like we haven't seen each other in person in way too long though, so let's work on that.

The DC we got your name right now. I expect to have you on here at least every other week. You know, I'll be back, I'll be back, I'll be back.

Famous words. Um, thank you both and thank you for actually thank you both also for the work you do on CDF, right? This is not, you know, it's a labor of love.

It's certainly not in it for the money. And, um, you know, volunteering and open sources is a great thing. And, and on foundations like this.

We're gonna take a break on text Trunk tv. We'll be back in a moment. Bye-Bye.