 Wow, so welcome to the Innovation and the Age of Multi-Cloud Panel. I'm Ron Miller and I read about the Enterprise for TechRunch. We've got a great panel today to talk about how multi-cloud hybrid and open source have come together to create an ecosystem of startups and established companies, and how this panel of venture capitalists decides where to place their bets. So just a reminder that we're going to leave some time for questions at the end. And if you have any, just if you could give me a signal at like five or ten minute mark. Yeah, because I don't have a clock or even a watch. So why don't you guys start by introducing yourselves and where you're from and then we'll get into the questions. I'm principal at Scale Venture Partners. I've been at Scale two years. I actually came to Scale by way of Google Cloud and AWS. I have my own little multi-cloud incarnate here as a product manager, mostly on data processing services. Hi, I'm Steve Herrod. I'm at General Catalyst Partners and prior to that I was the CTO at VMware where we were beginning some of the multi-cloud work. And I do early stage investing in sort of nerdy stuff in the infrastructure space. Hi, my name is Scott Rainey and I'm a partner with Red Point Ventures. I focus on enterprise investing with a particular emphasis on cloud infrastructure. So I just want to say, Steve, that maybe 2008 or 2009 or so you give a talk on hybrid cloud. It was the first time I actually understood what it was at the time. People talked about it, but it wasn't a very clear term in those days. Yeah, hybrid's gone, it's now multi. Right, oh, sorry, I'm saying I'm so behind. So cloud native is super hot right now as is the multi-cloud hybrid approach. And, you know, this wasn't always the case. And even a few years ago, people were thinking that maybe the public cloud would win and that, you know, you wouldn't have, you know, multiple places where you would have workloads or VMware would, and virtual machines would go away and none of that has happened. In fact, hybrid has taken hold even stronger. So how do you guys evaluate and foresee coming trends like that one? And decide how to invest where the puck is going instead of, you know, where it was? Maybe just a comment on the one you mentioned that we all thought public cloud was going to be this new thing and now we're seeing that that's maybe not the case. Maybe part of the reason that seemed to happen was that AWS gave us this thing that was actually a bundle of agility plus compute. And cloud native foundation has kind of decoupled that. Where now we can get the agility tools we wanted and we can have them wherever we want. And that can be on our own computer. That can be on somebody else's compute. Maybe more to your question of how do you spot trends. I try and keep in touch with users all the time. Every investment we make, we interview customers within our network and users of the product that the company will provide. Basically we want to make sure that it's not just our own judgment on the product but the market reveals what's valuable and what's not. Yeah, I think a lot of VCs are trying to use different forms of data out there these days to just see what's kind of hot. Everyone probably has, you probably have a red point, an algorithm for looking at open source projects, that sort of thing. But we also all have our own companies that we are investors in that are all using, using on the cutting edge of using the coolest new tools or getting into the cloud first. And for us that becomes a super ripe area for just asking all of our own companies what their favorite tools are and what they're thinking about in the next couple of years. Yeah, well I'd agree with all that and maybe I'll take a slightly different approach to the question which is talking about the general trend here. I think that you said it well where we're really primarily interested in helping companies become much more agile in terms of the way that they create and deploy and run and manage their software. And I think the evolution we've taken is initially that meant that we were focused very much on technologies that were facilitating the move to the cloud. I think what we're realizing now is we're just, we're interested in companies that are much more agnostic than that that really don't really care where the target is as long as we're helping companies make that migration. And fortunately what we're seeing are really interesting emerging trends that are popping up to facilitate that move. And a lot of that comes from exactly what Steve had shared, which is seeing what our existing companies or portfolio companies are using to actually facilitate their own work. So do you have companies that like come in the door and they aren't exactly aligned with what's happening in the market right now? And do you have those moments where you as investors say, wow, these guys seem to be onto something that hasn't been discovered before? Or is that something that's very rare? No, I think every Silicon Valley investor knows you've got to be super careful that what you're seeing there is not like the rest of the world on so many fronts. So no, I think everybody tries to make it out to other parts of the country or the world pretty often just to check it. But in general, I would say you see the cutting edge. It's somewhere amidst it all. Some of the other pieces might not kick in as much. But certainly amidst that usually is the signal of what the rest of the world is going to be moving towards. So one of the things that I find pretty interesting about Kubernetes and Cloud Native in particular is that Kubernetes sort of came out of nowhere and became this huge kind of system. And we were just talking in the hall about it. It's like 12,000 people here this week. And it was like 1,000 in the first one. And now all these adjacent technologies have built up around it. I think that's kind of hard to foresee that for every hot technology, a bunch of adjacent technologies will begin to start to pop up. When you see all those kinds of new areas, service meshes and so forth, new security measures around containerization, so many different things happening, how do you decide what's something that's worth investing in and something that maybe is just so nitchy that it's not really going to develop a larger market? I think there's been, there's a lesson that I've learned in investing in this space over the last seven or eight years, which is you want to see folks that are taking a very aggressive approach to thinking about next generation technologies, but also maybe have a pragmatic approach to what that's going to look like and how that actually ultimately gets rolled out. I think what we've seen years ago was a set of folks who had a very particular view on how they thought the world was going to unfold. But it required so much change from organizations and from the average enterprise that it really just wasn't all that realistic. And result has been, I think, some, you know, probably disappointments in the early days. What we're seeing now with a lot of companies is realizing that there's a journey that every company is going through. And by that I mean enterprises and their adoption of coordinated technologies. And they want to facilitate that journey. They're building products that can solve problems today, but also they believe position them well to solve really important strategic problems over the long run. And that's been something that we've been looking for in most of our investments more recently based on, frankly, some scars. Maybe another comment along the lines of how do you measure if something's going to be a bigger market and more willing to pay. A neat important trend for those here is just the fact that developers are now a buying party. They have budgets. I have a question about that. Yeah. Oh, I'm sorry. No, go ahead. But I think historically people have underestimated developers as a buying persona. And we've seen at least as evident by some of the sponsors here today that increasingly developers have an important say in what gets purchased in an organization. And they're now a viable market. Right. So yeah, I mean that's actually Mike. One of the things that I wanted to talk about was if you went to conferences, say in the 2011, 12, 13 time frame, you would have IT admins who were in charge of the computing at most companies talking about cloud and not the most glowing terms. I remember some really aggressive IT pros who saw their jobs changing in ways that they weren't happy about being very, very anti-cloud and saying like, over my dead body, we'll go to the cloud. That changed, of course. And part of that was the developer centricity that you just referenced. But so when you look at that switch to in buyer, how does that kind of affect how you think about people coming in the door and what they're trying to produce and sell? For us, it always starts with the product. What we're hoping is we're investing in companies that are building products that the developers and users are going to love. And you asked earlier what we look for in investments. More and more now, we're looking for these things that are actually developing some momentum within the developer community. I mean, either through open source adoption or some adoption of free service or whatever the case that they might be offering. But we're seeing these signals all around us. So it's easy actually. You're easier now probably to identify what these emerging things can be based on those signals. The biggest challenge is making sure that you're investing in teams that are absolutely 100% committed to building those products that are going to fulfill that that are going to maintain the kind of allegiance to the developer as an end user as they try to layer on features and capabilities that they're going to be able to charge for from others that maybe ultimately write the checks. Yeah, just maybe one extra thought of that. The general thought we've had most of the time has been most these tools you want developers to at least like it and you typically want operations to pay for it or to have IT to pay for it. With the exception of security, all you need is developers to not hate it. That's sort of the bar you set in that particular case. But it still seems the best thing about open source and cloud adoption all these areas has been companies can get aware of your product in ways they wouldn't have before. And so you see so many successful models which are looking at who's downloaded the product or who's using the free one and then obviously calling in probably several of you are involved with companies that way right now. And just that overall from a venture perspective it lowers the cost of acquiring new customers which is really what you hit when you hit any company at scale. And so I think everyone likes developer traction either directly leading to money as you're asking Ron or or just because it helps you get customers in a more efficient way. Maybe an additional thing into what. Seeing as your name is scale venture partner. What one thing we look for is just alignment between what you know when a founder comes in and pitches a company if they're going to plan to sell to developers we would expect probably that their channel aligns with that. That you know they're not they're not building a big direct sales team immediately maybe they're working on content that attracts developers and there's a path to buy by credit card. Whereas if you're a security company and you're telling us you're planning to sell to the head of security you may have a direct sales force and you're you're selling more long term engagements. I think alignment in your marketing and your sale go to market plan fitting with your buying persona is something we would be interested in seeing. I think it's just one last thought of mine which is like what the implication of all this which is really I think really profound which is that it used to be an enterprise the best product didn't necessarily win. I mean it was really about whether or not a company could build up the sales and marketing muscle to drive that software and organizations. Now it really is about it's much more of a kind of a democratization of creation where if you come up with a great idea you have the right product you have a real opportunity to ultimately capture a significant share of the market and that's that certainly makes it a lot more enjoyable you know I think from an entrepreneur's standpoint from the end user standpoint but from our standpoint as investors. Which sort of brings us to you know open source as a model and open source has the potential to create that market for you right you have those developers that download those things and start using it but you know and it certainly has a big role I think in this open in the cloud native and hybrid world but it's not a guarantee of success and it really depends a lot on how it's implemented and how you know savvy the people who create the project are. So when you have a company that comes in with an open source project is that like an advantage to you or does it really depend on the you know how they're packaging and presenting it. I'll go first so it depends on the area obviously there's open source in every aspect of kind of everything you sell these days certain parts of the stack it's a non-starter to not be open source so that's just table stakes to get in but I think you then you know once you get past that you pretty quickly shift to you know everyone probably knows there's probably four or five different ways you could ultimately make money from open source projects whether it's a supportive subscription or running a cloud service or open core or some blending of those things so I think every every venture firm at least starts to ask how would you think about that how would you do it over time most people now ask what would Amazon do to you that's a very common open source question actually it's a common everything question probably and then I think the other thing you really quickly the beautiful thing about open source from at least from an investing standpoint is that there is actually data out in the public that you wouldn't otherwise get and again whether it's directly looking at stars and forks and that sort of thing or whether it's looking at number of discussions happening on one of the various boards that are out there it just becomes an immediately useful thing to to measure and see if there is a scouts that if there's traction taking off and then lastly you know especially in some of these areas recruiting is one of the hardest things on earth so on one hand you you'd love seeing an open source project that is mostly run by that company but if you see one that's really taken off where there's a broad community you've also just found a way to recruit some of your future employees which can be something that we look for as well and hopefully customers right I'll add to that that as Steve mentioned in many cases open source is almost a requirement in some parts of the stack and it's a normal try before you buy type pattern that users in many cases expect and then there's the other extreme open source where it in some cases projects become a ubiquitous standard for that part of the stack and in very few cases have companies who've tried to monetize ubiquitous open source project failed I mean once you reach broad adoption across all developers monetizing becomes a little more straightforward and that becomes very exciting I think to the venture community so if you if your open source project is going well I think that's very attractive to investors I agree with everything that's been said here I think one of the things that might be interesting to note though is that in just about every one of our open source businesses the business model is becoming cloud services so we're moving away from service and support that was you know I think something that was much more prevalent a five ten years ago open core became a big part of what we sing but now in just about every one of the companies within our portfolio and many of the leading open source companies that have that have taken funding that are trying to build business around it are rolling out cloud services so they look they look more like you know a cloud resource you might get from amazon then they do a traditional open source company it's just that there's a maybe a little bit of a safety net that's there for people that are using the technology to be able to fall back upon if they don't want to leverage the service over the long run so well it's just a matter of finding that that whatever magic percentage it is that's willing to pay right and that's what it comes down to you're going to have thousands of users but you know you have to have some that are willing to subscribe and you know make that payment and there are plenty of companies who are making big money doing that but we've seen examples of companies who haven't been able to take advantage of that and we saw a doctor for instance last week sell off its enterprise business to marantis and i think that a lot of people including me were pretty shocked by that and we've seen Hadoop which was a huge project it was doing great it had three big companies come out of it two of which have now combined one with which was sold for pennies on the dollar to hpe so you have these successful open-store technology you know with tons of promise but they both had significant turbulence long after a ton of money had been invested out of those kinds of experiences kind of affect your investment strategy and i mean they they must be the kind of things that keep you up at night i'll take i'll start um so how do i think about it i think every case is you know it's is unique and rather than going into the specifics of those ones i think what you have to make sure that you're doing with your open-source projects is ultimately solving a set of problems that the people that that are going to be making the business cases for those investments that you know you're making a strong case for them and in the case of docker obviously i think what ended up happening was the real problem ultimately became you know managing the containers at scale and that's something that obviously kubernetes did much better than docker was able to do and as a result i think that put them in a difficult spot where they have a lot of developer adoption but really weren't solving many kind of business level issues of problems i think the thing with adoop is the reality is there's been a lot of value created in that adoop ecosystem it played out a little bit differently than i think we all thought part of that is because adoop itself became i think a little bit less important really quickly it began to be displaced by other emerging technologies and approaches to solving that problem and i think that was probably inherent or less complex right yeah i think it was probably more inherent with the adoop technology than it was with with you know the execution of any of those companies and so taking a step back what does that mean we need to ensure that you're you know the the thing that we spend time on is that we investing in companies that are solving things that we think will be aiding enterprise at scale to build business cases around adoption that software and are investing in the software that we think ultimately solves problems that are going to be you know with with a lot of runway associated with them so but you certainly must have thought i mean i don't know if you guys invested in any of those companies or ecosystems at all but i mean just as an outside observer of that and who wrote about it it certainly seemed like these these were companies that were you know primed to be successful long-term companies and then just hit a wall that i don't think anybody can expect and certainly one of the reasons was that you know docker got displaced by but i've also heard people argue that they just gave too much away and that there wasn't enough value left in what they were selling and i mean i just don't seem comfortable talking about the specific company but if you have an opinion i think you know the audience would be interested well i would just say there's so many reasons companies don't work out in the end and right like we all have opinions on each of these companies and we could probably go into them but i think the bigger thing that i think healthy thing to do with investors and with startup founders we actually try to write the negative case on every investment with the founders and say like let's figure out the two or three things that could go terribly wrong here and whether it's you uh your community goes astray because you don't give up enough whether it's you're gonna miss the boat on on cloud hosting you know whatever it is i think it's just really healthy to at least postulate up front of the many many ways you could fail which is the most dangerous one and then uh and then start working on that kind of immediately so that's actually an exercise we go through before we invest decisions i answer it a little bit it's it's sort of game theory and other pieces as well but it's at least having your eyes open when you start on where have other companies like this had troubles i maybe make a controversial statement here which is i think that very few open source companies are trying to commercialize the business fail because they gave too much away i mean there's generally a pretty bright line between what is you know in the open source and what people are charging for enterprise and people very rarely cross that i think that really what we're saying there is that they're not solving a big enough problem and that is probably much more inherent in the challenges that any one of those companies is facing so do we want to do some questions now from the audience okay you guys can share thanks guys are you seeing like a clear winner in the kind of the anti-cloud licensing world like comments clause business source license like the confluent license the stuff that lets the clouds not sell the your open source i you know i think we all looked for a winner early in this process and a couple of people you know picked one the comments clause and they didn't work out and i think in many cases some of us have just kind of moved on we'll just ship our own license for now and and wait for a winner to emerge and and it doesn't seem to be a lot of urgency for a winner to emerge and so i don't know that we'll see one anytime soon i i i personally haven't seen one you see a lot of the database companies though in particular trying to recapture the and control their product because companies like ad w s are kind of taking advantage of the open source and undermining their business models what do you guys think about that i'll just say that uh the the companies that we often cite in question elastic mongo are doing really well um the the idea that amazon has killed database companies hasn't played out in its fulfillment yet uh so it's a little bit of wait and see um it certainly seems like the threat should be there but i'm glad to see mongo and elastic for example doing well any other questions so i have one um as basam keynoteed in the morning today he said that the clouds the hyper clouds particularly are master integrators so given that how does um how does that impact your opinion on potential investments in this space are you more excited less excited what do you think well i wasn't here for the keynote but i i get the the core point of how you have to have multiple products working well together and that being a core competency i would say maybe this hits hits on the licensing question a little bit too what we look for the team to be as scott said a lot of them are really going to be monetizing as a cloud service themselves either through partnership or directly so we actually look at the operational capability of that piece of software can you run this thing at scale and can you provide the efficiencies that by owning the software base at least being the primary contributor you should have an advantage over the public clouds who would just sort of take it and tweak it a little bit more so certainly you do have to have the skill to make it work with everything else depending on what category you're in but um but i would say it's more the operational capability for that software stack is where we can see some real advantages happening over time and we play that one up as well i think it's also the operational capability of the team because if you think about your executing against an open source roadmap in many cases you're executing against um a open core set of features that are sitting on top of that and you're being asked now to run a cloud service and the reality is that's incredibly difficult for startups to do um and the dna associated with running and scaling um a service versus those that are associated with delivering enterprise software they're actually quite different and so um that's the thing that we've spent a bunch of time or their own companies on trying to help them think through that and we've seen folks who have taken pretty extreme views on what that means which is that they don't roll out anything in the open core until they roll it out in the cloud and so you know you treat that like a first class citizen um you know it's others who basically are always kind of trying to play catch up with the cloud service and you know my you know i'm firmly in the in the camp right now that i think open source the future of open source is going to be a managed services and you know i think what's going to happen with a lot of these is just a really interesting collection of talent that's going to happen from day one in these companies where there's going to be a set set of folks who are joining these open source companies who are basically saying that you know we're going to think about this less about the enterprise roadmap that we're building and more about the cloud services that we're going to be launching and that'll be i think a big change we see in the makeup of these companies over the next couple you see those as mutually exclusive i mean building a cloud service and being enterprise i i don't see them as mutually exclusive i see it as you know well first i may be at the early stage of a startup it might be mutually exclusive it might be too hard for a company that scale to do both of those things really really well you know what we have seen for the most part is companies that are starting on you know maybe open core and then rolling cloud service out afterwards and it's then just a question once you get to that stage of rolling out do you do you think yourself now as a managed service company do you think yourself as a company that has open core has managed services i would argue for you know until you're up into well into hundreds and millions of revenue it's hard to do all that really really well maybe your question can also be interpreted as like the cloud providers as master integrators are going to offer a lot of products and pretty much all the and so when we make a company we just assume we no longer ask is Amazon going to do this we just assume Amazon is going to do all these things and the buying decision becomes almost like buying generics versus brand products at a grocery store and in certain categories people buy generics all the time they don't really care what kind of beans they get but cookies you're going to get your oreos or something and so we think about is this startup in a category where you can be the best and it's a category people care about that they're not going to be considered a generic is one way of thinking about how to compete with the integrators and clearly one of the bets we're all making here if we believe in the future of multi-cloud is that customers are going to want aren't going to want to be tied to anyone individual cloud and their associated cloud services and so the opportunity exists to create cloud services around any of the next generation database technology whatever there will be cloud agnostic and that you can architect and build around one cloud service and it doesn't matter which of the clouds it's actually running on and that's a big part of our investment thesis I think you have to keep in mind that just because a big company does it whether it's AWS or anybody else doesn't mean they're going to be very good at it it just they can throw resources because they have lots of resources it doesn't mean they're going to build the best product in that category and very often they don't we have one of our companies who would argue that it's actually been beneficial for them that they they launched a service that competed against theirs it raised the visibility of that people go on they're using the Amazon service they express some level of frustration based on some of the limitations of that and are now finding them and so it's actually been a lead source for them may I just follow up question to that so do you folks see the activity in the startup ecosystem declining compared to two years ago because there are now these additional nuances that you all investors are thinking about or do you think we'll stay as frothy as we were with the Kubernetes ecosystem exploding two years ago I mean clearly if you just if you come here you see how many startups there are so I think the good thing is like when there's a big disruption of any kind you get all kinds of companies being built around to to either augment or rethink that new layer and so I think it's been great to see obviously Kubicon and Kubernetes coming on board first of all you tend to get a ton of companies that are doing things just for that platform and then you kind of think multi-cloud or multi-platform and so I think this kind of a year we're seeing more and more that are adopting yeah we'll work with this and we'll work with this and so I think anyway change has always been good for startups in the enterprise space and I think this big of a change has been great we started we we started the cloud native landscape with the CNCF a few years ago we've actually turned it over because it's become this massive project that is just extraordinary in terms of the number of projects that are out they're both open source and and and companies and I guess what I would say is any big disruption like this presents massive opportunity we're seeing lots of really interesting companies and really excited there's also naturally going to be some sort of reckoning that will happen because there's you know any kind of big disruption there's lots of companies that flood in there's going to be some winnowing down of of the technologies that are going to make the most sense and you know I think that's healthy because I think we'll see this kind of reallocation of resources towards the things that I think will probably propel the kubernetes and cloud native ecosystem for the next 510 years so I think as a journalist who covers startups you know or maybe two years ago you saw a lot of kubernetes services and those have kind of sorted themselves out now you're starting to see some you know adjacent services like I had asked about before you know you see a lot of service matches you see a lot of security this morning I wrote about gremlin releasing chaos as a service for kubernetes so you know you can do this kind of scale testing on the service and make sure your clusters are as hardened as they can be you know those kinds of interesting things are still happening and there's still plenty of activity around that awesome other questions from the audience sorry you hugged so much time all right well thank you so much panelists and moderator that was amazing