 We're back live here at the Velocity Conference. If you want to contact us, go and use the hashtag Velocity Conference, not Fluent Conference. I've been using the Fluent Conference. That's over. That was the JavaScript show we did. That was a great conference. Man, we did a good job there. We had just dropped in and one day's notice. But this is the Velocity Conference. This is O'Reilly Media's Velocity Conference. This is the intersection of designing on the front end and back end dev ops, cloud ops, application performance, whatever you want to call it. It's really a hybrid show, really where the main alpha geeks are here. Really build on the next generation ops. I'm John Furrier, the founder of SiliconANG. I'm joined by co-host. Theo Schloss-Nagel is here. He's the CEO of OmniTI and is a tech athlete. We like to call him. Welcome back to theCUBE. Yeah, we like sports analogies here. And so good to see you again. Thanks, good to be here. So what's new since the last time we talked to you? Oh, wow. Still drowning in data. Still smothered by clouds and being rained on every day. I don't know. It's a lot of the same. I think the industry's matured a lot. So the conversations are better now, but a lot of the same problems remain unsolved. So you are obviously a big open source proponent and have made tons of contributions. Theo, it feels like the world is, you know, open source has obviously been here forever, but it feels like the enterprise world is starting to get it. You know, you're seeing, you know, IBM's are recovering alcoholic. You're seeing HP bet the farm on OpenStack. Even, you know, the likes of EMC, you know, companies that are known for sort of propriety are starting to hop on the bandwagon. What do you make of that? I don't know. It's kind of an interesting convergence. I guess I've been a fan of open source, but more than anything, I'm a fan of, you know, shit that works, right? And that's the most important part, is that you have actual solutions to actual problems that you face. And it turns out that when we started the consulting company OmniTI back in 97, probably 80% of the stuff that we did was open source. And then over the next 10 years, that went to almost, you know, probably, you know, one out of 100 solutions is now commercial. And it's just because the open source products scratch the itch better, right? There is a solution there. There's less technical risk in running open source. You know, you're not, you know, beholden to a vendor to provide you updates when things go wrong. And if you're on the bleeding edge of scalability, things always go wrong. You're the one to witness them. You probably have the best facilities to fix them. And if it's open source, you actually can. So open source makes a lot of sense there. But, I mean, even today we still run a decent amount of closed source software. You had said as complexity increases, operations are abandoned in a perfect storm. I think you wrote that somewhere. What do you mean by that? What was the quote again? As complexity increases at scale, operations are abandoned in a perfect storm. So yeah, you do the first one. I told us about, hey, it's not dev ops, it's ops dev. Yeah, I mean, I think that when complexity gets very, very large, right? And you're running things at scale. The discipline, the person that can actually troubleshoot those skills isn't a single person. That's not an ops person. It's really a person that understands the entire stack. It's a software generalist. So like the traditions of the siloed approach of operations versus development versus whatever, they don't apply very well when you have complex systems that are that large. So I got to ask you about some of the quotes you said. You said on theCUBE, this is a Stratas 2010, you said, it's not about dev ops, it's about ops dev because ops really needs to tolerate, can't tolerate downtime where devs you want to push and iterate the whole break stuff. And ops, ops people don't like that. But that was the first time on theCUBE. We've heard that counter-opinion, which we like because that really brings the reality to large scale enterprises, financial institutions. They need to have scale and they need reliability. So given then, next two years, what's stabilized in the dev ops world as it evolves? Has it stabilized as it's more the same? Has there any bright light with the cloud? Has it still trying to, as enterprises still want to do on-prem and do cloud, all this bursting and public and private? Is there any stability? I think that the industry is coming to grips with the need for more generalized talent. That stabilized that dev ops is no longer a term that you hear only at a conference. There are 35 books on it now, right? So I think the awareness has gone up, so it's not a novelty. So now there are mature conversations instead of just kind of rather immature, evangelical conversations around it. So I've said a couple of times at Velocity Conference that I serve as a body check to those people that want dev ops or ops dev, right? There's so many software engineers that try to explain why software engineering has so much to contribute to operations. Well, they're having that side of the argument. I feel necessary to mention the other side of the argument which is you need some operational skills in your software engineering because you just don't have them. And then the same thing is true when I hear people talking about instilling operational values in software engineering. I'm always trying to remind them, there's a lot of software engineering practices and workflows and approaches to problem solving that are really applicable in operations. So it's really just trying to be on the other side of the argument to make people get along. So I was talking to a CIO of a big insurance company who won't say the name, but they're a billion dollar operating budget on IT. You know, I'll see as what is a competitive advantage for them, you know, I'll see insurance company want to have the data. So they really want the big data. So he says to me, hey, I got some POCs going on with Hadoop, we're looking at everything. They have challenges. They need compliance, they have Red Hat here, they got this over there. So you have a lot of legacy baggage in these kind of rapid deployments. What's your experience been in the past year or two where these guys need to go there, they need to look at this material, they can buy books and whatnot. But you know, as you're going there and consulting with these kinds of guys, what's the approach? What are you seeing working? What's not working? They obviously are want to get the emerging technology like Hadoop and store some batch and or use other technologies. But they got to scale up and scale out at the same time. So what are you seeing there? I mean, anything you can share? I mean, there's two strategies to that. And I think it really depends on an honest conversation about how well their heads are screwed on. So the bigger companies, a lot of times they have some really good talent internally that are progressive thinkers that realize that technology is there to meet business requirements. And they don't have a kind of a silly commitment or loyalty to a specific technology. They're willing to shift around. And those companies, the advice really is, yeah, you should explore all of these things. You should really look at them, you should evaluate them. You have the technical talent to evaluate these things from a good perspective. Then there are the other companies that are just like old, tried and true, IBM are from the old days. Not IBM from today, but the IBM is from the old day. It's all, I have these business processes. I know these business processes. And the honest truth is they really shouldn't try everything. Someone needs to tell them to kind of stop and reset and you realize, you have to realize and come to grips with in order to go to the future and be able to use software as a service platforms and infrastructures as a service platforms and then deploy things quickly. You have to actually rid yourself of your previous workflows. You have to get rid of the processes that you had in place and you have to pull in new ones. And from a consulting perspective in those situations, it's almost always better to be highly prescriptive where you go in and you just tell them what to do. How does that relate to the business process piece of it? Because if you've got a business processes connect into those IT development processes, so how do you sort of recommend that organizations deal with that? They're fundamentally related. If they don't relate, you're going to fail. If there's no reason to even have technology unless it's going to enable a business process. The problem is technology is not nearly as flexible as people think it is. Technology is good at some things. It works in some ways. So understanding those and being willing to change your business processes to adapt to the strengths of technology is key, right? I mean, otherwise your competitor's going to do it and blow you out of the water. So historically, you've got this database and then you get these processes that are these technical processes built up around that database and then you sort of wire your business process to them because you have no choice. And what you're essentially saying is, okay, we get that. We understand why that's historically the case, but you're going to be roadkill if you don't change that. But there's a lot of, I would imagine, friction in making those changes. So how do you facilitate that change? Well, I mean, as consultants and being rather small, being rather small consultancy, we have about, we're just under 50 people. We can't take on all the work in the world. So the customers that have their heads squarely. You just walk. Where the sunsides are. Tell them, you know, you're not ready yet. You run, you don't walk. You're not ready yet. We're not here to convince you that you need to do this. You need to know you need to do this. We're here to help you not fall off the cliff while you're doing it, right? Those business processes, the big difference today is that technology enables business in a way that it has never done so before, right? You have big data solutions that can answer crazy questions now, but you have to ask the right questions and it's good at answering specific questions. So if your business process that you have maps onto the technology in a way that's really awkward and not very efficient, like there is a great reason to have those technologists, a technologist that's highly versed at the table helping design those business processes because suddenly you can make some small, very surgical changes to those and make them highly efficient and highly, much more powerful. Yeah, you know, we were at the Service Now conference, a couple, actually last month, Jeff Frick and I were doing it, and it was amazing to see what they've done. I mean, it's a small-scale implementation of this, but they've got a single CMDB and they're able to completely change business processes around that to your point. And it was real enabling. You get these IT guys jumping around, making cakes, having parties because they're like the happiest IT people you've ever seen. So are you seeing that kind of enthusiasm in shops that you've touched? Yeah, I mean, it depends on the culture of the company. I think, I mean. No, these are hardcore IT guys. Yeah, yeah, I mean, so who are the students that act out most in college? It's the ones that went to the restrictive Catholic, you know, girl schools, right? I mean, so I think that- That's Frank Zappa. That the IT organization where IT was always a cost center and they're always the underdog and they're always the servant to that. Like when they turn around and add value and everyone in the organization recognizes it, they have a reason to celebrate like no one else does. Because the difference, the contrast between where they were in their place and where they are as a player at the table, like it's so radically different. The other more progressive organizations where IT's already having a good time, I would say the celebrations aren't quite that big, but they're very rewarding. Awesome. Theo, always great to have you. We're going to try to get you on tomorrow as well. Thanks for dropping in. We've got our next segment coming up. Always great to have you on colorful, knowledgeable and always having all the sides of the angles covered. So appreciate it. We write back, this is theCUBE. We are at Velocity Conference, instructing the signal from the noise. We write back after the short break.