 Welcome back everyone, theCube's live coverage here. Mongo, local, New York City's packed house. End of the day, we're doing a wrap-up keynote review analysis, we've got an analyst panel here. We're going to break it down. Nothing's, everything's on the table. Stephen O'Grady, Principal Analyst, founder at Redmonk. Deep in the developer ecosystem, open source coverage for over 15 years now. Too long. Congratulations. We have a Lovers founder and principal analyst, Lopez Research. Guys, we've been around the block on these events we've seen in the movie before. We're kind of at this new inflection point, Maribel. We're seeing kind of like this next wave come and it's kind of building on the wave we've been in. Like open source, you know, cloud native, cloud growth, but it feels a lot different with the AI stuff. It just feels more, I almost say dot-com bubble-ish in a way it does, but like, it's got legit legs. I think, what's your opinion? Hey, you know, data's all the rage right now. We've been talking about it for a while. AI has really got everybody enthused about the opportunity with data, but the fact is, if you're looking at AI, you're looking at large volumes of data, you're looking at the need for really performing systems. We're talking a lot about integrating that with the cloud, but also being able to make sure that you can connect with your on-prem data sources. So it's a really complicated environment, but it's nice to see companies like Mongo try to step it up and add things like stream processing, add vector. We saw a lot of good announcements today that are really related to helping the developer out, helping make this process of actually moving to real-time actioning work. Steven Open Source is the real driver here. I mean, I was, they have a vector database. Everyone's bolting on all this stuff. I'm streaming natively in there, but Milvus is a project in the Linux Foundation that does the same thing, no mention of that. So you've got other Open Source projects, probably like, what, seven vector databases out there in the market. It seems like everyone's kind of like in this new world from a, on the fence standpoint, like, which community's going to get all the AI? LangChain's certainly doing great. It feels like college, you know, kind of like, you know, like, who do I want to hang out with? Right. Yeah, it's, as you say, it's crowded field, right? And AI has been probably the most transformative sort of macro trend that we've seen in a couple of decades at this point, right? Because we've seen, we've been doing this for a while and we've seen lots of things come up and sort of come and go and, you know, you said it yourself, like it has legs, right? You know, because that was, AI has been through multiple cycles, you know, dating back to the 60s. This is sort of the, I think one of the first waves where, you know, when we go talk to companies now, when we talk about what they're doing, it's actually real, it's real work, right? You know, you're like, oh, that's actually really interesting, hadn't thought of that. Whereas in years past it'd be potentially jumped up, you know, linear regression or, you know, it was not what we today think of as AI. So in the vector space, you know, because that, you know, is a facet of that market, there's a lot of interest, there's a ton of talk, a ton of chatter, and now it's just going to be, you know, to your point, like, who hangs out with who? And, you know, it's one of the reasons, like, we like what we do. We get to figure out and track this stuff and try to understand what are developers using and why they're using it. You know, one of the things, oh, good. Just a way in on that though, right? I think it's really exciting that we get to have a lot of companies, because you get to have a lot of companies, you can't make a market with one, right? So you need a lot of competition, you need people to push the ideas forward. I think what both of what we're seeing is it's moving really fast, like faster than it's ever moved in our careers. And I just think that's really good for enterprise IT buyers. I think it introduces a lot of confusion though over how to figure out who to work with and, like, how long they're going to be ready to work with them. Out of control. Yeah, totally. Well, the thing that gets my attention, I agree with you guys both, it's that, but the applications are impact. You're seeing apps being built. And so to me, what gets me excited, I want to ask you guys how you see and how you're framing it in your research, because I don't think the puzzle's complete, but you can put the borders together, is that the people are building apps. Yeah. And you know, Amazon's like, we have bedrock. No, we don't care what to run on yet. Right. Or what the GPU, I want to build my apps. Right. So building and then running them kind of going on. It feels like cloud early days, where I had EC2 and S3 and SQS, that was good enough. Yeah. Do my SaaS app in the cloud, and then I was good, got better obviously, but similar moment here. And what's different though is that there's really use cases that people can see. Yeah. And you know, the old expression, beauty's in the eye of the beholder. Right. Every company has that look, like I see three things I could do tomorrow. Right. And this is the dynamic. Will open source drive that? Will proprietary models leak? Because they got the GPUs already secured? Yeah. Or does it matter? Who cares? I guess my question is, how do you guys see this market developing? Because apps are involved. That means that's going to impact the business model of the company. Well, I'll just all throw it quickly that I think one of the biggest issues that I see currently is that in the AI space in particular, companies are misusing and misapplying the open source label. Right. So in other words, we have a bunch of models that have come out. The source of these models is available, but the license is under a non-commercial license. Right. That is essentially not, that is fundamentally not what open source is about. And the difficulties is that you now see people going off and building apps on this. It's been billed as open source to them. They feel that they have the same rights that they've always enjoyed with open source software. And yet, at some point, somebody's going to look at the license underlying and say, uh-oh, I can't build a business like this. Right, so the first thing to me that has to happen in all this is that- GPL had that same problem early days, remember? Well, the GPL, it was difficult to understand. But in other words, we've had 30 years. And even the most conservative enterprise license people now, they understand, okay, this is what it is. And the challenge now is you see these brand new licenses come out that prohibit things, they ask for royalties and everything else, and it's like, all right, we need to sort that out first. That's interesting, because the people who usually break those licenses, so the bomb throwers, disruptors, who don't give a shit about licenses- Well, it's companies that should know better. They're young innovators. Meta did this last week. So Meta released a model, and they themselves on Twitter call it open source, it's not. It comes out under a non-commercial license. So like I said, that needs to get sorted first. Yeah, that's a head fake, by the way, too. Big time. And it's one of many, right? So the other side of that, if you take a look from the enterprise buyer side, one of the things that we're starting to see happen is, you know, there's open source that has more commercial grade security and manageability around it, and then there's just open source, right? And understanding what you're using, how that's going to be supportive moving forward, if you're going to be able to have all the regulatory compliance and other things you need, do you have to go buy a service from like Azure for Azure services for open AI or something like that? That's really complicated. And I think after everybody went all like chat, GPT, GAGA, happy and thought that AI could do everything on the planet, right? It went from though we can't do anything with AI to like, oh yeah, we're going to do everything with AI. And then it came back to hallucinations of people were like, oh, put the brakes on, right? And now people are putting their proprietary code into a system that is open and they can't get it out. So all of a sudden there's a lot of nuancing and I think this is the opportunity for technology vendors to come in and say, okay, here's what you actually need to make AI work in a secure and compliant way that leverages open source, right? And that's still really fascinating. What's interesting about MongoDB, I'd love to get you guys thoughts around the economics. Okay, they're about a billion in change right now. They want to get to, I think he said 1.5, or he said, I think two is the stretch number. I mean, to get to $10 billion in revenue, they got to actually be commercial. Okay, and they're an open source company right now and Atlas is doing great. So if they're going to be a data platform, they got to solve these problems, right? So that means they got to be, they got to, I won't say relinquish their open source roots, which is organic bottoms up. They have a really good bottoms up mojo going on, but they want to be $10 billion in revenue. You kind of got to get up to the ivory tower and work on some of those top down issues, compliance, licensing, I don't want to get sued. What code are we using? This is kind of the, I think the dilemma or my IP's being, who's using that? Kill ChatGPT, no one, no more open AI. I mean, seen that right now. So I see this top down, bottoms up, not challenge, but more of, this is serious. We have to kind of rein it in. Let the chaos reign, reign in the chaos as Andy Grove's famous line is. So the question mark, what do companies do? Because you don't want to foreclose the innovation and then more creativity. So are you guys seeing that, are you advising companies on how to think about it? Is there any kind of visibility into a playbook? Well, you know, one of the things that happens is the customers are going to take Mongo to where they need to be, right? So Mongo's already got, I think in the keynote, they were talking about 43,000 companies. They had a lot of really impressive logos. They also talked about working with the startups on the AI side. So it's not like they don't have enterprise buyers using them. So the question now becomes is, there's that kind of co-innovation of like where we want to take you next. And if you look at some of the stuff that was announced, clearly it was things people asked for. So I think that Mongo's at that point in their journey where every company has to like cross their own enterprise IT chasm, right? That's where they are right now. And they have a lot of opportunity to do so. They have a lot of license to do so. And I think that the good news about that is there seems to be enthusiasm around adopting the platform and making it happen. If we were talking a year ago, maybe there was slightly different discussion around it. But once you get to that point, you kind of think it's a go and I feel like it's a just don't screw it up, right? Like you can make it happen, just don't do anything to screw it up. What do you think, Steve? For me, I think it's just, we've been advising open source companies for a long time. And what we always tell them is that, particularly in this day and age, the simplest model for you is as a service, right? In other words, Dave said it in one of the sessions today where when the company IPO'd, Atlas was 3% of the revenue, it's now 65% of the revenue, right? So it's no secret, the money from all these services is coming, running them as managed services. And if you are the company who is best in the world at producing a particular service, guess what? People will pay you for that. And Albongo has had two companies, two very large competitors clone the API and yet they're still here and they're still growing. And it doesn't matter at the end of the day that their source code is available. So that's the thing. If you focus on what customers want, customers don't want the source code, they don't want the product. In most cases, what they want is somebody else to run it for them. Yeah, and they have the experience on this trajectory. They've been doing it, the platform for multiple years since they took them, what? Four years to build Atlas. And then it's been in marketing now for what? Six years, roughly so, give or take. So they have a little bit of an increase of scale there. But it brings up another good point. If open source is going to continue to thrive and you're seeing more software being built, I mean, you look at some of the things that the Prometa Llamas stuff got, it's getting repurposed in other open source projects that's lightweight, super fast. Open source is kicking ass. Now they don't have any data. The data holders have the value, right? So back to the app integration, I see this whole programming model where data is like, feels like the security shift left kind of mindset of, get the developers something quickly. So they can code at the point of coding in the CICD pipeline. And I don't know how to call it yet. I can't put my finger on it, but it feels like it has to happen fast. Your reaction to that? Nothing, nothing burger? You can say it. You can say it. I don't know. I guess I got a couple of different reactions. The first one back on your last point is I think the multi, you mentioned as a service and the multi cloud thing is actually really important there as well. If you look at anything, they're trying to create a platform. So what do you need around a platform? You need things like app marketplaces and integration and other stuff to go with that, right? Because the more people that are leveraging your platform, the more powerful and strategic you are to an organization. So I feel like that's another iteration of talking about what has to happen for them to be a real big enterprise player. And we've got my attention from the conversation I heard with some of the people today was the CTO said to me, yeah, we have a sub millisecond failover on our databases distributed. And I could roll my workload right over to BigQuery. And I'm like, hmm, that's interesting, moving workloads around that fast and to take advantage and consume other services. Yeah. I mean, technically it's possible, but it's not like real companies sit there and they're like, being, I'm going to do that to BigQuery. I know, but the idea of it, but remember in the edge, we had the big debate about move compute to the data. Data's strapped right now. Moving data is your egress killing the people. But the point that's really well taken there is the point that you've taken that whole set of concerns off the table, right? So they're not worried about like the ability to failover. They're not worried about like specific cloud lock-in to use stuff, right? So once you get to that point, you know, the customer can focus on the innovation and they don't have to focus on some of the things that are really important, but aren't the things that are going to differentiate the business, right? And the numbers show as a service works. So integrating services, whether it's consuming new things from other people, it becomes an integration game. It feels like, it just feels, I think it feels hyped up for sure, but I just feel like it's the gen AI stuff is legit value. It's legit. Well, AI is happening. I mean, there's just no doubt about that. AI is happening. And the only question for organizations is, how do you make it happen fast and secure? And how do you pivot it if you got it wrong, right? That seems to be the- We got the beer out here, rings rocking, we're getting into the day. Okay, final question. Your two-part question. Give them a grade, letter grade for the show. They're positioning, they're messaging what they're doing. And then what advice would you give leadership on what to do, change, double down on? I'm going to let you get it for sure. I can take that. I think for me, the show is definitely, it's an A, they get high marks, because in other words, the key insight I think that they've had, which a lot of vendors struggle with, is that it's not necessarily a marketing event, it's about developers and bringing people in and getting them using their software. And that's been the focus, it's been the focus on the keynotes, it's been the focus here, it's sort of down on the floor. So that is something, I don't want to say that, you know, sort of Mongo lost that, because that certainly was its roots in the early days was easy to use, easy to pick up and get going with. But as they built up as a company, it sort of went, you know, sort of after enterprise customers, you know, it's harder to hear that developer message and they've refocused on that here. And as far as a piece of advice I'd give them, you know, honestly to me, it is really sort of more of the same and trying to, so if your market is developers, how do you make their lives easier, right? How do you put these pieces together? Because the database is a, it's an important piece of the tool chain, but it's just one part, right? So in other words, you know, you talk to them about the work that they're doing, integrating with partners like, you know, for sales and so on in the world, who are doing things with the developer experience, like more of that, right? Smooth out those rough edges, make developers' lives easier, and if you do that, they'll be happy and they'll use your stuff. I'm going to give them two grades, because I'm going to break this up into two parts. Part A, developers. You know, we've talked forever about like developers are dead, we're killing developers, you know, we've had low code, no code, all this other stuff. The reality is, is developers aren't going anywhere, you need like really good developers for certain things. I think that, I'm with Steve, we're going to give it an A for the developer, because there was a lot of talk about what the developers care about, how they're making things more simplistic, there was actual news around that. I'm going to give it a B for the second category, which are the check signing people. I think Mongo needs to do a better job of appealing to the check signing people, because right now there's a lot of organizations that are out there trying to figure out what sort of things do I need to make this AI transition happen. And you know, Mongo's not going to be the first person that comes up to the list. They had some good, in the innovation awards, they had some good insight there. I think more of those customer case stories really needs to be told to kind of bring it to life for everyone. You're a tough teacher, you're bringing down written composition, different sub-categories of the topic. Why make it easy? I'll give them an A minus, mainly because I do agree with, I think they get an A for the developers, but I want to see more meat on the bone on cloud native as it relates to the enterprise. Application, fidelity on what the delivery is, what they were going to bring to the CXO level. Security, we heard reinforce, authorization is a big part of scaling out, privileges, access has been a problem. So these are a little in between the toes details that I didn't hear much of here, so I give them a little ding on that. Other than that, I love the developer angle. If they can nail this developer-led innovation, I truly believe that developers are going to come up and influence purchase decisions, and I think they have to connect that with either the business case or manage some of the compliance stuff and just create global entry. Love your developer, how many people say it, but don't do it, so it's nice to actually see it, yeah. Is it good or bad? This sounds like, I think it's good, I think it's really good. We talk to people all the time, they say they want to build a developer community and then they don't have anybody, right? You talk to them a year later and nothing's happened, they didn't do anything. Who's that targeted towards? Developer or the CXO? Both, in other words, you're talking about, you're trying to sell the CXO on that as a vision, and you're trying to advocate on behalf of the developer. Absolutely, love it. It's a lot of band bottom, you got it. Developer, developers, developers. Yeah, developers are the standard. Steven, Maribel, thanks for coming on. We're getting the hook. Thank you for breaking down all the action with the analysts here on Mongo. They're getting good grades on this panel, A's across the board. There you go. The database guys all gave them a B. No surprise. Different perspective. There you go. All right, we got all the angles covered here. We'll be right back with our last interview here after this short break.