The Future of Open Source pt 1 [Cloud2030 Aug 27]

Rob Hirschfeld: Hello, this is Rob Hirschfeld, and welcome to the cloud 2030 discussion on August 27. This was a spirited all our conversation about open source and how open source is going to impact the future or not impact the future depending on your point of view. Check it out, of course, put the discussion first and then our hallway track is at the end. So enjoy. If you want, please come and join these conversations be part of the community. The 2030 dot cloud. Christian you actually did some prep, right you have some some slides for us. Krish Subramanian: Ah, lovely.

Yeah, I just put together 00:40 - four or five slides just to get the discussion going. And then we can talk and don’t judge me for the team I’m using here because I was so bored with the usual team I use I am sort of like a thief Lawrence Hecht: okay. So Krish Subramanian: So So what it does is run by a few slides and then we allow open discussion that I had in mind. And I was expecting few others to join to discuss something on serverless data and join broadly you can have a follow up track on that follow up session on that. So this is the talk about open source becoming irrelevant does nothing new like IV I had all that discussion going on in 2010, with the CO funded saying that open source has no role in cloud and d’amore saying that architecture matters and protocol matters. License doesn’t matter.

Then Sam Ramji 01:42 - and Sam Johnston even started our open cloud initiative to drive open format and open protocol. They also argued that open source doesn’t matter. But what we are seeing in the last 10 years, leading to grow 2020 years open store still matters and the sectors of Kubernetes any indication, open source is still alive. And it’s sort of like continuing to drive innovation. So, go 2030 when you go towards coda 2030, I think open source is still going to drive all the all the innovation around the cloud. But there are issues which you need to discuss, that could make open source less relevant in the future. And we need to confront those issues now.

And I think that’s where Lawrence might get some interesting 02:34 - top chat there. And I also, I think, we need to define how open source is going to play a role in things like edge computing serverless, artificial intelligence, and a few other things. So I think these are some of the topics we could talk in today’s session, as well as some time In the future. So that’s my thinking. Let’s have a open session talk about some of these topics, I think, yay by us. It’s a big, big issue. Now, though there are many, many contributors to it, including data and all that.

I think some bias can be removed by embracing open source, for example, the 03:28 - implicit bias developers have, all those things can be removed, the code is out in the open and more people with more eyeballs could take a look at it, I think that we can reduce some of the bias out there. And similarly, like, we could discuss the topics like what is the role of open source Foundation, because they used to stop sort of brought into open some of the fault lines. So I think I should discuss that and dive in should sort of make sure that the open Source foundations or whatever, organization that go to sort of VIP there, whatever forms form factors that are going to be there in terms of supporting organizations for open source, I think we need to make sure that that is sustainable. And it puts code contribution at the center of getting any leverage right now, open source foundations our like focus more on getting the money. If you throw only $1, you could take advantage of that and build a service and literally demolish many of the startups out there. So I think that needs to change.

We need to put code contribution as the currency for open source. It should go back to the roots open source roots and we should make sure that Richard Stallman doesn’t laugh out loud that that whatever he said about having flexible was not friendly, open sources running or do we need to stop that and take it in a direction Makes open source more sustainable. So these are some of my thoughts. And I think I see this as a continuing conversation. So ping me on Twitter, we can take a session on any specific topic in one of the future weekly shows. So yeah. Now I’m open session from now on. Tim Crawford: I have a question about this Krish.

Have you thought through the why, like where open source kind of best fits in? And the reason I’m 05:29 - asking this question is to provide guidance that was squarely missing when we went through the Linux days and talks about open source of Linux versus Unix. And I’ve seen it kind of replaying itself again, where we get into these religious cesspools of discussion, which I don’t think benefit anyone. I don’t think it benefits those that are advocates of open source and I don’t think it it benefits those that are advocates of commercial software. So I guess my question to you is, is really one of the Do you have any guidance as to where open source really benefits and where it doesn’t? Because, you know, there’s, there’s kind of that yin and yang of maybe there are good places for it, and maybe not. But I was wondering if you have maybe some guidance that might help answer those questions or guide the conversation for us? Yeah. Krish Subramanian: It’s a great question. In fact, I’ve been thinking about it a lot.

So there are some discussions on where open source will fit in 06:35 - in terms of like, hey, it’s not going to make a difference in the case of infrastructure. So we saw the discussion coming up in OpenStack days. Of course, OpenStack didn’t really help help in that discussion. But if you look at Amazon, AWS itself, do you think Microsoft will keep the licensing terms to go from the software model to a services model, it works The weapon. So Lawrence Hecht: what why did he help at all? Unknown: In the case of Lawrence Hecht: any wide open stack up at all? Krish Subramanian: Okay.

Yeah, I mean maybe because that they were trying to go head on with Amazon rather than trying to be the next iteration 07:21 - of VMware virtualization. So that’s where they made a mistake. And of course, Rob and others have more thoughts on that. Yeah, Unknown: I think Rob Hirschfeld: OpenStack challenges are almost a session of themselves for this. Tim Crawford: Yeah, I’m not sure I would use OpenStack is kind of the the poster child of success or failure of open source specifically. Well, so I was asking a broader question of the value of open source specific and where it specifically would kind of fit in and very, maybe that’s where you were also kind of getting into Krish Subramanian: Yeah, so yeah.

My my thinking is open source fits in, in cases where the traditional proprietary licenses didn’t give you 08:08 - flexibility to go beyond beyond what the company defines as the acceptable use. So, for example, that’s why I quoted the Amazon example. They couldn’t, they wouldn’t, they couldn’t convince Microsoft to change the licensing terms so that they could offer it as offer Windows as a service or Windows virtual machines as a service. So clearly like Microsoft, it seems that, okay, there is value to as a service model, they would have jumped in long back. So I think that license flexibility that allows people to go beyond normal acceptable ways of using software, I think that’s where open source really comes into play.

I don’t believe in open source, helping in terms of locking all those, all those things are okay, you 08:57 - can put checkmarks over there. But the real value lies in the flexibility of the license. And you could take it and you could sort of innovate beyond normal levels of iterative innovation you could think completely out of out of the box and probably build something that gives that flexibility gives you have the opportunity to build something that is solely bounded by your imagination. I think that’s where the value comes into picture. Maybe that this is not something that can be quantified and probably even given out as a recipe for others to emulate.

But I think the 09:36 - flexibility is where I see value for open tools, of course, other things like widing lock, and all those things are checkmarks. Lawrence Hecht: What are we talking about? Are we talking about specific projects, tests, civic services, in terms of open source? Basically, where’s the where’s the point where were you upstream collaboration between different cloud providers. Krish Subramanian: Oh, in terms of, I think collaboration between cloud providers, I would say, I think any layer above infrastructure is where I am seeing collaboration, for example, at the infrastructure layer, I think it was a total total failure. We saw with the OpenStack, code stack and all that Lika. They code for cloud providers see it as be something that threatens them.

So probably a layer above like Kubernetes, 10:38 - Kubernetes and maybe even k native and some of the open West, all those ideas about infrastructures where cloud providers will be interested, I don’t think they are interested in doing anything. Today use open source that day in particular, but at least I’m not seeing anything interesting at the layer of interest. Right, they want to. Lawrence Hecht: And I think, Rob Hirschfeld: go ahead my pointer hold. Milind Gunjan: No, I think I was just going to make a point like, at least even in Kubernetes, with introduction of like, cloud controller and trying to make it as like the API for infrastructure, right. I think tickle, typical work has started in that direction.

But I know 11:22 - this is where you you have I think we’ll have a better perspective. I just wanted to add that point to what Chris was saying. And I know like, at least for the all the major cloud providers, that’s not like, that’s their bread and butter, they would not like to invest a lot there. Again, from open source point of view, but then for us, like I come from, as Rob knows, I come from a telco background and we, we we dipped our toes with OpenStack. And how do I say now with edge and like distributed In the distributed nature of infrastructures we will have to build.

That is 12:06 - like one of the biggest question in any telcos mind right now how we are going to do that. Just what I was Krish Subramanian: trying to say was specifically responding to Lawrence’s question on where copro does play a role. But I totally agree with you open source at the infrastructure Layer Layer is more useful as you go towards the edge. And yes. Paul Teich: Once you get outside the big cloud providers, I think when you say, Krishna, when I when I hear you say cloud providers, I hear you say top five top 10. But there’s this whole tier two and telco space where folks want to be cloud providers.

They were they don’t have the funds to 12:52 - build their own infrastructure. And that’s where open source killed itself with complexity or sorry, OpenStack right. Yes. Wasn’t wanting to be this and the telcos really wanted it to be the solution. Right? They tried. Rob Hirschfeld: Still trying Paul Teich: three or four Asian. Built on OpenStack right now. Yeah. Lawrence Hecht: But then Milind Gunjan: telcos approached it with a very wrong mindset.

Like, 13:19 - Unknown: I think Milind Gunjan: I been in the industry being part of it and a big part of the problem, I can admit it. Like the way we approached it was very wrong. That was like, Okay, this is open source, but then we didn’t try to change the way we have been thinking around how we bring in software, we still thought that you know, we will do the same thing, what we have been doing with our infrastructure or with our applications, you know, bring in evaluate the application and bring a vendor with it and they will, an 800 number with it and they will solve everything for you which which definitely Mike Maney: that mindset changing in that industry or not really? Milind Gunjan: Oh, of course, like right now, if you look who the leader is right now, with regards to the cloud native 5g deployment that’s rocketed. They, they basically are approaching it with a very right mindset. They are trying to do the build their own, or like the whole OSS stack, they’re trying to own it. And that is giving other telcos some guidance. But as you know, like we slow to move. Yeah, like we are not like the So, but I see positive changes, like with OpenStack there was a huge lesson. Unknown: And yeah.

14:46 - Can you summarize that? Rob Hirschfeld: I want to I want to hear that. I want to hear the elevator last version of the lesson. Milind Gunjan: I don’t know. I think think that it was like it was for the spite of fear with license that like you know the the what they were asking for. For virtualization that’s why they said okay, this is it this this is where we will go behind it but then they thought you know like somebody else will can do it for them for cheaper but then people just forget to account for the optics. Rob Hirschfeld: Yeah, I was what I was thinking about the telcos because and I talked to you and I talked to a whole bunch of people, you know the telcos.

And they’re sort of like, well, we wanted you know, two things. One is free software. Because we hate we hate paying for boxes and appliance that has expensive software, but cheap hardware. And then we wanted we wanted the ability to be vendor vendor portable, and OpenStack didn’t, you know, accomplish those those missions specifically? I actually the point I was thinking of making is my experience with open sources. It’s sort of the David versus Goliath idea where open source allows the community, the little people to, you know, bypass the big behemoths. And the thing that’s weird to me about current current model to me and when I think about 2030 is open source has become this with foundations and giant companies.

16:22 - It’s like well, Amazon’s the leader in open source. I’m like, Wait a second, that’s counter to what open source is? Where the power where we’re trying to put the power if, right, it’s sort of like, I’ll go to a blockchain example. Right? If If 50% of the blockchain participants are owned by China or one company, then it’s not actually a blockchain, right. There’s just not a community thing. But, um, Lawrence Hecht: Rob, I think this guy’s sorry. I just I just think that reason why I like foundations and In this regard is because the cloud providers are the major source of the developers and the initiative to work on things. No, I, well, we disagree about that.

But that’s, that’s my thinking is that you get to other three main cloud providers collaborate, plus, have a big 17:20 - community, that’s what you need, and then doesn’t matter if everyone else has to join a board. That’s the theory of the case. No, but then I think this Milind Gunjan: great, sorry to interrupt. I had an example I will take issue as an example where I think too big to have the biggest cloud providers have been like, or like two or three big companies have been contributing it a lot. But then as an adopter, right, because they’re like so much of technical debt, I cannot just bring it in. Like in my like, how do I say in my production environment like that right.

There is called complexity there 18:00 - like, so many of technical know how is there now I cannot do with that easily I was just going to make that point, it may be like that, like that particular open source project can be easier for them to get to their production environment, but not not to like an adopter. So, just going to try to make that point here. Krish Subramanian: Okay, so I’m sorry to you, like I Tim Crawford: didn’t go. Well, I was just gonna add in here. I mean, I guess the the conversations are interesting. The the part that I’m having a hard time with is is kind of the why.

You know, if I look at this from 18:38 - the customer perspective, telco, accepting it, moving telcos out of the mix for a minute, but just looking at kind of enterprise, I mean, there’s a lot of question around why should I be considering open source and kind of going back to what you were saying Krish about licensing, if that’s really the basis and I can see some value there. That’s one thing but I’m just having a hard time following along as to where this fits in and why, and ensuring that it’s not just an academic conversation around the theory of where open source fits in or just a vendor centric perspective. Because otherwise, then you’re just you’re having a lot of conversation and putting a lot of energy into it, but to what end to what point if you’re not thinking about the customer and the value to the customer, which is that into enterprise. Lawrence Hecht: Like I work now for over 15 years with startups, and so on. I’m personally a big user of open source both professionally and privately.

And at least my experience has been that the the value that open source provides versus enterprise product from a startup 20:03 - perspective is the low entry cost. a startup, particularly when when you’re when you’re acquiring users, you, you cannot spend $10,000 a month for a single product. That is not infrastructure, because that just breaks the back. So and particularly with enterprise products, what what, what I found is that, that tends to be a gap in the pricing model between the open source or trial versions. And then the next base level tier. I’ll take, for example, Elastic Search, look at the x pack license.

20:52 - That that’s a big gap between the between the defeat features that are available on on the base subscription versus the next pay tier. And I think that is one of the things that drives towards adoption of open source or free slash open source products. And also towards the adoption of cloud services where you’re basically over a year your cost might be higher. But the cost to test things out on the short term is much lower. Because you don’t have that initial step of you need to pay for.

Yeah, but 21:47 - Krish Subramanian: but for enterprises, that cost is not a problem, right? Like, they don’t want to spend time or human resources trying to figure things out. They want something that works out of the box. So Would love to hear from enterprise arguments true? Tim Crawford: Well, the other thing I was going to add there is that it’s, if you’re starting from the standpoint that that open source is free, which a lot of folks do look at, I think that’s a misnomer, right? from the start. I mean, it’s the whole free like a puppy. Right? You might get the puppy free, but you got to take care of it. And, you know, the value of open source comes from the contribution back into the community. And And so again, I go back to, then what is the value if you’re, if the value is because you see open source is a cheaper version than commercial software? I think that’s a problem because it’s missing a lot of components even beyond the cost component, right? And so that’s, that’s why, again, I fundamentally go back to what’s the value then? Lawrence Hecht: I completely I read you like that the long term cost of maintaining a product yourself is Hire done buying something off the shelf, but the entry level cost is lower.

And then a lot of times you end up sticking with their nursery like already we already have this product. Why switch. And this is something that that we were talking in the track on Tuesday is that the trade the trade was saying that the North to switch from one product to another, there needs to be an order of magnitude difference in benefits. Rob Hirschfeld: an order of magnitude difference, right? It has to be a much better thing to go to I. Yeah, this the struggle I have with this conversation is that and that is that it’s really about sustaining. If you’re building a product and you build it on open source, and you don’t have a sustaining plan for that open source.

Then, you know, you’re you’re in deep trouble right? If I’m an open if I’m a startup, and I, I am and I built, 24:04 - I choose Open source technologies when I when I do that we need to, we’re aware that that’s something we’re taking on as a sustaining model or we pay to help sustain that that environment because if something goes goes away, then you’re in trouble. Right? Open Source, we have the same Unknown: problem with enterprises, Rob Hirschfeld: very much the same problem with enterprises. And so Lawrence Hecht: I’m like, yeah, go. So open source Tim Crawford: this, the codes always there. If if they if an enterprise company goes under, you might just lose that code forever. You might not be able to run it.

There might be DRM, there might be who knows what? Something that’s open source can be forked, like, look at TrueCrypt is a great 24:43 - example, where TrueCrypt even said it might be insecure, but yet enough people had enough momentum behind it. For people that keep pushing ahead with that code base. Even though Krish Subramanian: that’s a valid point. You know, not just the code you for the community to You know, the community, you have the court, but you cannot do anything with it. Right? Tim Crawford: So I’ll tell you from experience, it’s actually not a valid point, because it’s a theoretical point.

And the reason why I say that 25:11 - is because the commercial products, you have stability, and then you make decisions based on that if you’re working with a startup or working with a smaller company, it’s very common. And I’ve done this many, many times in the past, you have to require code to be put in escrow as part of your contract with that particular company. And so if anything happens to that company, you at least have rights to the code and then you can choose to either develop it or move to the enterprise stuff you’re talking about. That’s the world I live in is the enterprise. Lawrence Hecht: We do everything to start up, you know. Krish Subramanian: There’s started. Yeah, go ahead. Sorry. Lawrence Hecht: I was just saying like, that’s also why we have the two tracks for cloud 2030 and Correct me if I’m wrong, Rob, but to talk today is more geared towards the enterprise level discussion.

26:07 - Rob Hirschfeld: It To me, it’s more about the future. And so the simple thing, the thing, the thing that I would I would say is, these these questions are ones that have, you know, we need to get worked out. I don’t know, I think the enterprise who pays for the open source code is very real, because what we’re seeing is that open sources, especially at the infrastructure layers, is being funded by, you know, the infrastructure companies. I’d love to have somebody correct me if I’m wrong, but if what that means, fundamentally, is that if Amazon, Microsoft and Google and Red Hat, IBM, are, you know, doing the lion’s share of these open source contributions for core infrastructure projects, then they’re going to do it in ways that benefit what their infrastructure is. Lawrence Hecht: But you everyone just said that OpenStack was a failure is there a need for New OpenStack to basically to do it better.

I mean, 27:04 - that’s Rob Hirschfeld: part of part of what? Krish Subramanian: Krishna? Yeah, no, no, I was about to say that in the introduction doesn’t mean it has to be open source version replacing Amazon, it could, it could be at a slightly different layer to the problem with OpenStack. Because they were trying to claim themselves to be the open source version of what you can do with Amazon. That’s where they feel miserably. And the fact that they didn’t, even though at the time I supported them, like they didn’t make their API is compatible with the AWS API, CPA. I think these are some of the reasons why they failed, but OpenStack failure, as Tim pointed out, open tax failure. Doesn’t in any way, make open source irrelevant in industries. The infrastructure speeds. You’re muted, you’re muted dialogue.

And 28:02 - Lawrence Hecht: the reason why I bring it up is because the and this has to do with enterprise first nine prize, my getting my theory is that you need a significant enterprise adoption for any of these technologies to be viable long term. There’s a few technologies that were driven not by enterprise, but that’s the exception rather than the rule. So that’s that’s why I go back to enterprise use cases. So Dockers example where it wasn’t driven by enterprise in my opinion. Yeah. Krish Subramanian: I totally agree with you. Right. Rob Hirschfeld: But it was it was driven very much by and This to me is way interesting things. I just read doctor Paul Teich: was totally driven by enterprise. Okay, it was it was driven, driven by a large company wanting to put a code base out there so that they could attract developers and get folks coming in, who knew what that what Docker was all about. But it forked out of Google Docker didn’t just appear as a grassroots project, Lawrence Hecht: but actually it Krish Subramanian: was a grassroots project. Right? The container, the container format was? Tim Crawford: No, we disagree.

29:18 - Krish Subramanian: What I’m saying is that user interface which they, which they brought in and made it attractive for developers that initially gain traction among individual developers than enterprise developers, only later on like the ones that gained some momentum, then the enterprise developer started jumping in. So that’s when like I don’t even even started thinking about orchestration and everything like that. The Docker, Docker started going down. Lawrence Hecht: Hmm. I have a Paul Teich: different view watching this evolve as an outsider analysts that Tim Crawford: Paul, I agree with you, I just, I mean listening to this, remind Have the double edged sword conversation with open source. So I agreed John Cowan: to, because I think most startups make the classic mistake of open sourcing their code. And they, they they leave it for for folks like me to just take and make it into innovative stuff that makes money and they’ve left with nothing. And I I’m sorry about that.

But that’s, that’s the 30:20 - mistake of open sourcing code to begin with. So that leaves you with really the Googles and the Red Hats of the world that puts stuff out there. It’s almost like a baited hook, right in terms of how they they cultivate these ecosystems. But all they’re really doing is getting free development work, and they’re rolling it up into their commercial products. Maybe it’s because I live in Raleigh, and I’m next to Red Hat all day long. And I listen to these guys and they’re like, excuse my language about getting pirates. Okay, in the software development world, I wouldn’t touch any of their stuff and I wouldn’t open source anything I do, because I know how they rape and pillage Krish Subramanian: Yeah, that sort of brings up another question like our Which are robberies? How do you make open source sustainable? pointed out to the dad case. And then there is the foundation issue where you could just throw in our own million dollar and get rid of all the startups in the ecosystem. So how do we make it sustainable? That’s a that’s another question. I think we have got one question which Tim raise, which is what is the value for enterprise users? Then the next question we need to tackle is how do you make open source sustainable anytime? Tim Crawford: And I would just say that if you don’t answer the first question, it’s pretty hard to answer the second question. Unknown: Yeah. Yeah, Linux.

31:40 - Paul Teich: Linux is the poster child, right? Because it didn’t have the licensing model of Unix, or Windows. Right. And so it got used, and a large community huge community evolved to make sure Linux was sustainable but who benefited mostly Red Hat Tim Crawford: Red Hat red. I didn’t get into. Paul, I think I would agree with this is what I would say about that. I think that that red hat is a is an anomaly in the course of history. And you’ve been spending 20 years people trying to been replicating that model thinking that if I just open source my stuff, I’ll be red hat. I’ll be the billion dollar open source company. It’s just not true. Unknown: Right Place right time.

Yeah, 32:23 - Rob Hirschfeld: well actually opens. One of the things that Red Hat did very successfully was they actually took a whole bunch of open source projects, bundled them in and made them accessible, and they actually lived off this vision at Tim Crawford: Philly. It was more simplistic than that. So if you go back to if you go back to the Linux wars, back in the 90s, when we were fighting with something other than BSD and Solaris, and we were looking for an alternative solution, you could go with the core Linux spin offs. But the problem was, you needed something in the middle you needed one of the corners Linux spin offs, but you needed support because Linux wasn’t as well known or understood me enterprise. And so Red Hat started out really with basically providing support commercial support more a Linux product. And that’s the core of where it started. Lawrence Hecht: Yeah, I agree. 100%.

But unfortunately, 33:20 - Tim Crawford: you know, there’s a lot more to those Linux wars. I know I gloss over it. And those of you that, remember those days, understand the pain and a lot of the lessons that came from it. Unfortunately, most of the people that are working in open source today, were never in the workplace during that time. And so a lot of those lessons have long been lost. And we’re frankly revisiting them once again, or having to go through it once again, because those lessons were lost. But I think again, fundamentally, you have to go back to the question of why, you know, what is the value? So, as a student of history in all of this, I agree with you that I That’s exactly what that’s exactly how the roots got started around Linux and Red Hat and what have you.

And if you think about the potential for the a, an equivalent perfect storm, if you will, you know, between like, you 34:12 - know, Microsoft and Unix and then all in there now there’s an open source operating system out there. You know, I think that that the idea of cloud is not going to be it I don’t think there’s going to be the kind of Gemini that was that was put in in the operating system world that precipitated the need for Linux to exist and the the pent up desire amongst developers and operators and users to support an open source operating system. I don’t see I don’t see cloud as as that perfect storm because I don’t think that cloud is the panacea that quite frankly, it was when we all got together in 2013. Not it’s not it’s maybe it’s it’s just not it. I think there’s there’s something else that The fourth generation of the internet and which this perfect storm could exist, in which case, the desire to break down the oligopoly might be big enough to support an open source movement equivalent to Linux. Rob Hirschfeld: And so this this was in the in the chat for those people find the chat, I want to escalate, elevate this point, because fundamentally, when we’re thinking about cloud 2030, in the future, the question I think he in front of us is, is open source going to disrupt the current incumbents, is it right? Oh, yeah.

And and 35:37 - Krish Subramanian: that’s the wrong question to ask that we I think that’s the mistake. OpenStack did, is it going to disrupt AWS? No, that’s the wrong question to answer. I think that they need to look at it as how open source is going to play an enabling role for future disruptions. So if you look at open source as a way to disrupt the incumbent, probably we are losing out. So we need to look at it as is enabler for future disruptions? That’s the way I see it. Duncan Johnston-Watt: If you want the example of No. project, I would posit that, first of all, I would say, absolutely, you know, it’s not going to disrupt. I mean, you know, a lot of successful open source projects become services. Funny enough. We’ve seen that plenty of times. But I think if you look at one that has succeeded, I would say it’s Kubernetes. Actually. And, and it’s why we base a lot of what we do on that.

And it’s not because it’s open source, it’s because it’s actually ubiquitous or near standard. And it’s achieved something which OpenStack clouds that none of those things ever got close to. It’s useful, and it’s available. I think the other thing I’d add is that I do believe in foundations because they help with governance, so you don’t have the single company, open source project, which is actually a huge one. For enterprise, if you think about what Tim is saying, What’s the value of open source, open source on its own has very little value to an enterprise. But open source, it’s property governed, where there is one or more company that’s providing a distribution, which means something that’s consumable by enterprise does have value.

The challenge, of course, is, as john was saying, I mean, yeah, how do you if you’re 37:22 - even if you’re the original inventor of an open source project, how do you monetize that? That’s, that’s really hard. And very few companies have succeeded. Even Docker arguably has failed in that wouldn’t be Krish Subramanian: a valid point. Right in funny. Tim Crawford: Well, I was just gonna say, Duncan, I would say on Kubernetes. Specifically, I think you I agree with your points. But then take it a step further. Why has Kubernetes been successful to this point? And will it be six will it be equally successful into the future? I think it’s solved a problem a specific problem that exists And it was a clever way to be able to solve it.

Whereas there are open source projects that either 38:08 - are very Michie or kind of one off. And so again, it goes back to the what is the problem that it’s trying to solve? And what is the audience that that is solving it for? Shane Gibson: So I agree with that. It has to solve a real problem at the right time in history in space. But it has to be more than that. I think that Kubernetes took off because it did that it was novel and new. And it was replacing something failing, in concept which we can point to OpenStack as the poster child for that, that failure.

So there was a ripe moment in history for a transition, but more importantly, a community 38:51 - formed and coalesced around it. And that allowed it to sort of be spread to the wind and become ubiquitous. more available and more people using it and more people contributing to it. And I think that that community aspect is the biggest part of what open source provides is people that can come together and coalesce around a technology that they can see and touch and be a community. Now, I’m not talking about any of the enterprise or the commercial aspects of things, those are follow ons or adjunct to that.

But I think you have to have community around it, because there are some 39:27 - amazing open source technology stacks, services, platform solutions, applications, whatever you want to call them out there that never gained traction, because they didn’t gain community. Krish Subramanian: Yeah, that sort of brings to the next point which Duncan raised but that is the relevance of foundations. Yes, I agree with Duncan, that you need a good foundation to ensure governance. So the question is, do you want to partake in the foundations that will ensure governance and maintain the community and focus On Deck, or do you need the next foundation based trade associations that manages the trademarks does a lot of marketing and also the management handles governance. So, what is the right approach to sort of like drive the community? So that open source is sustainable in long term? So my question my dog x, I see that Yeah, Duncan Johnston-Watt: Linux Foundation become or what is under the Linux Foundation have become more fashionable, if you want to call it that.

40:36 - Precisely because they allow marketing and, and yes, Krish Subramanian: exactly. Unknown: So Mike, my Rob Hirschfeld: first experience on this one, yeah. Krish Subramanian: My explanation was I don’t I’m okay. That’s fine like that. But the problem is like they don’t put tires on contribution as the currency for leverage. That is my biggest problem like, for example example, let’s take the case of There’s an Amazon been caught into denic much code. I won’t say any code.

I didn’t, it didn’t contribute much code compared to other other players in the community. But they leveraged it to the maximum they are making money. But it’s no one else’s making money. I don’t care if Google doesn’t make money. But I care about startups in the ecosystem, who are who never had an opportunity to really make money. Amazon didn’t come and take away everything. That’s my problem that the Linux Foundation and Duncan Johnston-Watt: so arguably the counter to that is that it is very hard to make money out of Kubernetes because it is something where common cause is more important. And you know, what would make you money out of Kubernetes really mean? It’s hard to see what that would mean? I mean, doing something makes sense, but Krish Subramanian: we’re making money out of Kubernetes doesn’t mean directly like I like making money directly, but the fact that a startup couldn’t Gain level or any leverage inside the community that hurts in making money outside of Kubernetes maybe using Kubernetes at the core but innovating on top of it, but once bermad comes and occupies the space, I think that’s gone.

But 42:16 - Unknown: this comes back to is our cause a huge amount of value. Duncan Johnston-Watt: And, you know, I’m very comfortable doing that, knowing that I can find Kubernetes pretty much everywhere but the idea of, you know, being the next head to the next whatever it makes no sense you know, that that’s, that is past and longer video. Rob Hirschfeld: But even and this was your, I guess, your I want to build on that point because I don’t even the next step to heptio was, you know, effectively a VC investment it was it was a an investment strategy. It wasn’t, you know, what, what, what they did was they take and they did some good things, but they really took rock stars, they pulled them together and then they spun it out as an exit to that The idea that, you know, a startup with open source is going to disrupt something in market. I think is and the key is with open source AI, to me part of this discussion is we all expect the incumbents to be disrupted. The question in this this No, you we don’t, okay. Um, so it. All right.

So maybe we need to say 43:26 - could be what’s what, what’s, what’s the flyer here for the long odds, because if the incumbents aren’t disrupted, then 2030 is just 2020. With the temperature turned up to 11. If, you know to me the interesting conversation becomes how does how does that direction change? Or maybe we all love it, and we’re just like, Hey, here’s some features I’d like to see Amazon ad. But I don’t I don’t have this conversation. I’m not hearing that open source is going to deliver Some breakthrough technology that is going to get mass adoption and and disrupt Amazon, not in its current incarnation. It’s not. Then that’s then I think the follow up conversation for this because I’m watching the clock is what would open source need to be in that follow up incarnation to actually create a new market rather than Krish Subramanian: an example I don’t see you open source disrupting a commodity like Amazon. But if Amazon was not in the market, probably a startup could have taken something like Kubernetes and built a kind of abstraction that fargate provides.

Maybe if Amazon was not there, maybe offered an 44:48 - opportunity for a startup to provide that kind of an abstraction that is missing today because they must come and if Amazon contemplated code and then did this, I have no problem. Got a go? Yeah, I, Rob Hirschfeld: I actually think that for somebody to disrupt what’s going on today or create something new, there’s got to be a profit motive and open source except through acquisition into big companies has not been showing the same type of build company opportunities that that’s my promise but hold on because I we actually need to wrap it. There’s plenty of conversation for additional comfort for this to have a follow up. Tim, you and I can we work out enough so that we can talk major themes for next next week. There’s an email in your inbox. Awesome. You’re awesome. Thank you. So next week, we’re going to go back to topics tracks and themes and we’re going to do some logistics and we’ll I’m sure we’ll have interesting arguments and fights about that because that’s the meat of what we’re discussing anyway.

Please somebody if you had trouble finding 45:53 - the zoom link, I want to put the registration and the zoom links in an easier to find place so send that to me one on one. You can do it through cloud 2030 or however Twitter however you want to do it. Please don’t post the direct zoom link because that’s going to invite zoom bombers in and that’s problematic but I do want to make it easier for Unknown: for Rob Hirschfeld: it your careers. Duncan Johnston-Watt: Tim, you’re a one man bomb. Rob Hirschfeld: Yeah, I this is this is why this is why green screen PJ’s are the next conference tchotchke for this.

46:35 - Unknown: You mean Chris Zubov and this whole time? Lawrence Hecht: Exactly. Rob Hirschfeld: So please, please let me know I want to make this easier to find. Unknown: Thanks, everybody. Rob Hirschfeld: All right, everybody. Thank you. See you next week on Tuesday. We’re talking ipv6 at DevOps. So it’s gonna be good with Lawrence Hecht: the six Rob Hirschfeld: this fantasy that You keep chasing along with open source profit by Unknown: crush Krish Subramanian: it up, how are you? Unknown: Good, how are you? Lawrence Hecht: Good doing good. Krish Subramanian: So this is a casual discussion, right like Rob Hirschfeld: that’s my expectation. Yeah, there were will be doors. Yeah. Did you have specific items that you wanted to Krish Subramanian: know I just thought I will kick started discussion like I have four or five slide deck, just discussion and from there I will really open it up and we’ll have a more casual kind of discussion on various topics. Rob Hirschfeld: Perfect.

And then my expectation is that we use the first usually 15 minutes as hallway track. Unknown: Okay, sure. Rob Hirschfeld: Just how things are going. Absolutely, absolutely working. It’s been working actually really nicely for DevOps, the DevOps days. Nice. People come late anyway. And so what you do is you end up with a nice chance people, their early birds get to get to talk a little bit more, huh? Unknown: Yeah, and then say Mike Maney: hello, Krish Subramanian: Andrea. Hi. Mike Maney: Hi crash. Hi, Paul. I don’t know that I’ve ever seen you before crash.

This is 48:39 - Krish Subramanian: the first time I guess. Mike Maney: Yes. Usually we just had on Twitter. Unknown: Yeah, that’s awesome. Good morning. Mike Maney: And I don’t even think I know Paul. He’s not a Twitter friend yet. It’s an Austin friend. You should definitely Oh my God. One day. Paul Teich: Yeah, I’d say despite my shirt, this is actually a south by shirt. Rob Hirschfeld: I actually actually wore specialized special where you can see it. It’s my finger on my camera open. It’s my OpenStack San Francisco, Santa Clara Santa Clara.

49:26 - Mike Maney: I like your old Dell crowbar t shirts. Rob Hirschfeld: Oh, that was awesome. We did we have some really nice swag. Yeah. Some of them I want to recreate. We did that’s what we that’s what we’re like somebody like sceeto advisor was sending out masks. He has he has masks as his conference. Paul Teich: That’s great. Yeah, Rob Hirschfeld: I was thinking PJ PJ Actually, I heard somebody doing PJ bottoms. Mike Maney: Oh my god. That’s very funny. Paul Teich: not sending out swag yet. We’re just sending out sample data. Rob Hirschfeld: sample data. So it’s useful, it’s fresh, and Lawrence Hecht: we’re going to talk about how foundations are just for marketing. And that’s all. Krish Subramanian: Yeah, it doesn’t look like it. Let’s see.

50:22 - Rob Hirschfeld: We’re no sacred cows or Lawrence Hecht: Yeah, I’m Paul Teich: orange. Good to meet you virtually. Lawrence Hecht: Oh gosh, yeah. Rob Hirschfeld: Sorry I was and I’m fully I’ve put together a document for us to take notes and I will put it in the chat Mike Maney: by Rob Hirschfeld: and actually, I will put it in the cloud 2030 if too many windows. Don’t open up my Cameras blocking. Andrea Kalmans: Is anyone know what Chairman Powell said this morning? Whatever it was, it was really good apparently. Mike Maney: I don’t know what it was, but apparently it was great. Lawrence Hecht: Next Chairman pal, right. Rob Hirschfeld: People coming in Hold on a second. Mike Maney: Welcome to the party. Rob Hirschfeld: Love friends.

51:38 - Mike Maney: Krish Where do you live? Krish Subramanian: I live in Seattle. Mike Maney: Oh, nice. Lawrence Hecht: Yeah. Mike Maney: Actually, I understand downtown. Seattle’s pretty tricky right now. Yeah. My parents were telling me that the riots were back in Minneapolis last night when I grew up. Unknown: Crazy Tim Crawford: Good morning Brian was Unknown: john on Joe. Rob Hirschfeld: Oh, I have a cloud. Unknown: Hey there, how are you doing? Mike Maney: Oh my god, what’s on your ear? Andrea Kalmans: Not Tim Crawford: a lot of history.

I 52:27 - Unknown: got to go mountain biking fall and rip my ear lobe off on the side of my head. Mike Maney: Oh my god, I’m so sorry. I kept Unknown: riding. What? Mike Maney: Walter had his wisdom teeth out and well, I was in the nurse’s office getting picking him up. I was on the weekly call with Rob. The nurse loved me that I was so compassionate. Tim Crawford: I’ve got stories that go like that in capital raising. Unknown: Oh, that’s Tim Crawford: just that’s just another month. And raising capital for a startup. Here low. Exactly. Unknown: Give you a left leg. birth. Okay, you’ve got two of them. That’s highly available your loads at it.

I’m having trouble I’m here and everybody in mono 53:20 - Manny, next time do something cool like crash going downhill not uphill. I know that was that was a while I was talking to my buddy I was riding with and I was like, I don’t really remember the crash. And he goes oh, I do. He goes half a mile earlier. I could tell you were getting tired and I heard you put your foot down and I heard you heard something else and then I heard clumpiness. Oh yeah, it was time. Rob Hirschfeld: That’s brutal. Yeah, go in too long when you’re tired. is Mike Maney: just great. Nice, nice early dawn ride before everyone else was up. Just wake up. First people on the trail. It was great. Unknown: Until the crash, it was great to crash.

54:10 - One more cup of coffee and I probably wouldn’t have crashed. Rob Hirschfeld: Oh, I’m gonna mute for a second. Mike Maney: How does everyone like cloud 2030 thus far? Krish Subramanian: I think it’s fun. Mike Maney: Okay, Krish Subramanian: we need to make it more fun like that’s early today we are talking about what? Paul Teich: Okay, more fun. liking it. Like this question. Okay, gotcha. Mike Maney: Can you think about fun I’m we’re fun.

54:49 - Unknown: After I mean, the first the first one was amazing. Like I told Rob off offline. These are conversations that aren’t happening in the industry right now. That sort of need this face to face conversation back and forth? There’s, you know, in this group, there’s just so much respect among everybody to be able to mix it up and actually have those conversations and did you say respect? Yeah, hold on, hold on. Still may be 80% accurate. Tim Crawford: I mean, you all may have respect but nobody has respect for me Come on. Unknown: I still may be concussed from the fall Mike Maney: Okay, so Rob we have a request for more fun Rob Hirschfeld: more fun, I will try to deliver it definitely.

And so the format and bring good topics but um, go here is 15 minutes of hallway 55:47 - track every you know as our as our forming and then before we dive in, so please bring topics bring rants. I was we you know, bring the of the day. And we can we can, can do, we’re doing the same thing on the DevOps sessions, by the way, and last, the DevOps session, we just sit down with charity majors. And she talked about observability for about 20 minutes. And then she started ranting on management and fundraising and other stuff like that. And it was entertaining to say the least. She was very inspiring Tim Crawford: things and always make it.

56:27 - Rob Hirschfeld: Yeah, that was she was she was she was good and talking about organizational accountability and things like that. So if you have tactical topics, please bring them to the Tuesday sessions. We’re, we’re, we have a good stable base, just like I’m hoping we’ll have here which it looks like we do. And you know, so if you bring an idea or topic and you want to just bet it through, that’s there’s always an agenda time like this. I’m talking logistics Rob, what do you need? Where’s, where’s where’s the invite for the Tuesday session? It is in.

So if you go into Cloud 57:03 - 2030 there’s a there’s a topic thread, hold on a second. I’ll actually show you some here. And then I got the domain. So now it’s at 2030. Cloud. Let’s see if I can do this in a fancy way. Unknown: That’s, that’s the way it Rob Hirschfeld: is. You can just do 20 2030 cloud as a as a thing. Let’s say desktop. This one. Yeah, there we go. Um, and from here, I think that that’s so cool. If you go into the topics, there is a topic for DevOps Lunch and Learn and the invite is at the top of this, sorry to drag you through the whole thing, but I’m trying to, I’m trying to consolidate it down. So we’re putting And inconsistent places. I think I have it on the home for 2032.

But this is where like the recordings I’m doing the same thing for the this these sessions. So the recordings are here and then I’m getting fancier on the transcripts. So there’s actually real real like the transcript I did for this one is time stamped and has speaker speaker names and fancy like crazy. fancy, fancy. Yeah, it’s fun. Alert. It’s not that it’s not a big lift to add real good transcripting Mike Maney: are you using red Rob Hirschfeld: calm I would want to use I’m using otter Mike Maney: cool. Rob Hirschfeld: And it’s pretty it’s pretty sweet.

So for like for, for posts, I’ll reorder it so like for this part of the conversation will show 58:43 - up at the end of the recording. So people coming in after the fact they’ll jump into our the right to the discussion, not to the hallway. And so all that stuff, I have to pass back through the transcripts. Lawrence Hecht: That’s pretty cool. That’s fun. Unknown: All right back to Rob Hirschfeld: that help on the signups and please help you know promote this promote this the since we’re not doing this corporate sponsorships all gonna have to come through social media and people are getting referrals and and stuff like that so if you want somebody in the conversation, please invite him in. I do not want this to become a sponsored event. Yeah. Unknown: We’re pretty clear on that. Rob Hirschfeld: From that perspective, Paul Teich: I don’t have to write a check.

59:41 - Rob Hirschfeld: You can you just don’t have to write it to the cloud 2030 I’m, I’m right here. Ready. Actually, I think Rob Moussa would be would be happy to just say right, you can go direct to them that bypass meal Got it. So California people doing okay with the fires? It’s like Tim, Tim was dealing with the door. Krish Subramanian: I think Tim was in salt, right like it is in San Diego somewhere. Yeah. Tim Crawford: And you’re muted Jim. So Krish Subramanian: again, so quad was talking about booms getting pretty close to the zone.

00:27 - Tim Crawford: Sorry, folks, I had to step away for a minute. Yeah, no, I’m in Los Angeles. And the The air was actually pretty bad for the last week or so. But in the last couple of days, it’s cleared out pretty well. Like I couldn’t see across to downtown LA and the San Gabriel Mountains before and as of yesterday, is kind of the first time in a week and week and a half that I’ve been able to see them. Unknown: Yeah, here in Silicon Valley. I’m in Mountain View where 15 miles from three major groupings to the south to the north and to the east surrounded by them in the last week and a half It was really bad.

It was pretty much raining ash 01:16 - wake up in the morning there be a layer of ash over everything yesterday though was beautiful it was 72 and blue skies and you couldn’t smell the smoke. So I’m it’s cleared out some nice little break but the fires are still going so I imagine will be covered back in smoke here soon. Lawrence Hecht: Stay safe .