Pivotal Conversations—Sam Ramji, CEO of the Cloud Foundry Foundation

May 18, 2015 Coté

sfeatured-podcastHow do you organize the work around building a shared cloud platform between vendors and actual users of that platform? That’s exactly what the Cloud Foundry Foundation was setup to manage and what Coté discusses with Sam Ramji in this episode. In talking about the mechanics of the Cloud Foundry Foundation we compare and contrast it to other open source foundations and also discuss what to expect over the next year.

PLAY EPISODE

SHOW NOTES

TRANSCRIPT

Coté:
Hey, how’s it going on this Friday? What does a CEO of an open source foundation do on a Friday afternoon? What’s the typical run-down? I’m basically, I balance these two schizophrenic things. I’m madly trying to scramble to finish things I haven’t finished. On the other hand I’m looking for an excuse not to go insane at the end of the day and start to transition into just relaxing a little bit. What goes on when you’re heading up a foundation?

Sam Ramji:
I think that is pretty much the same thing that I do. One of the things that we practice at the Cloud Foundry Foundation is Top Three Things This Week. We have a channel in Slack called Top Three Things This Week and the whole point of view is we work incredibly hard. Yeah, we’re going to do a bunch of stuff every single day.

There will be things we didn’t plan that come in, but somehow, even though all of us could put in 40 hour days and still feel like we’re not doing enough, how do you get to Friday and maintain your sanity and feel like, ‘I’ve got a lot of ambition to do more things, but somehow this was a good week.’ That’s the idea behind TTTTW that’s working.

Coté:
That’s a good idea. When I was in the throws of some past job just enjoying my corporate life, one of my friends gave me some advice. He said, “What I do is I keep a file where I just literally print things out of good work that I’ve done. When I get all upset, I open up that file and I cheer myself up.” Doing that on a weekly cadence sounds great.

Sam Ramji:
It’s really nice and Slack I think is just an amazing tool for the team like ours. We’ve got people in different parts of the country, different time zones, some here, some on the east coast, some up the west coast. We’ve got people who are fully on the team all the time and some people just help us out with particular things like the Cloud Foundry Summit coming up next week.

Being able to create shared state is important, but also shared mood. Today, just to give you a sense, first thing in the morning I’ve got an interview with a Wall Street Journal reporter who’s going to be writing about one of our new Wall Street-based members, which is kind of big news.

Then, running from that, tackling all the email, new things that are coming up. The Summit itself is provoking a lot of new actions from a lot of new members. I just found out that GE is committing to build an industrial internet dojo for Cloud Foundry, so amazingly good news, but it still has to get wrestled down.

We find out that IBM is going to contribute their .NET buildpack to Cloud Foundry. They’re announcing and open sourcing it and of course they want to land all these things on Monday or Tuesday. Just an enormous, those are just scratching the surface details.

Coté:
Yeah, I think I sort of, well I shouldn’t say sort of, the opposite of sort of, I joined the Cloud Foundry community back in January when I started at Pivotal and it is one of those be careful what you asked for. If you ask for excitement and fun things to happen, yeah, that happens.

Then, you have the delight of working through it and making them happen, but there’s a tremendous amount of fun stuff going on in the Cloud Foundry community, including … Correct me if I have my dates wrong, but it was back in January or so that the community formally launched the Foundation and everything, right? Then, when did you join and start heading up the Foundation?

Sam Ramji:
The Foundation was incorporated on January 21st, 2015, and that was the first board meeting and the first day that I attended and the board gave me my job. I have been here from the first second.

In fact, funny true story, there are five board positions. There are nine board members from the member companies, seven from the platinums, two from the golds. There are about 18 gold memberships and that class votes in the position team members.

Those votes were all done and then of those remaining nine people, there’s a chair, vice-chair, executive director, treasurer and secretary. In that opening meeting, only one officer role was something they could decide on at that moment, so it was me as executive director. For about two weeks I was actually the only officer of the Foundation.

Coté:
I love it how whatever type of organization you have, a legal entity, it still has the secretary and the treasurer.

Sam Ramji:
Yeah.

Coté:
It’s such a quaint notion.

Sam Ramji:
It is a quaint notion, but it’s interesting because there are things that we do in distributed software systems to say how do we track current state? What is the source of truth? How do things get initiated? How do they get terminated? Each of those roles has a real particular function.

Initiation is formally the role of the chair, and if the chair is not there, the warm stand-by is the vice-chair. That’s really the vice-chair’s only role. The executive director role is basically the run time, execute all the processes and make it all go.

The treasurer’s job is to make sure that there is a source of truth about what is the set of resources available? What’s coming in and what’s been burned? The secretary is the source of truth for what has been actually decided. You can start to look at a proper corporate organization as being a thoughtfully designed actor-based distributive system.

Coté:
You got a regular ERP system of record there.

Sam Ramji:
You have to, right? Otherwise in the next meeting how do we agree on what we agreed on before? People’ stories change because biological systems don’t store data the way that digital systems do. We look back to the digital to say, “Yes, we actually did agree on that in these particular terms of language.”

Coté:
Exactly. You’re there in January at this board meeting of the Foundation. Obviously you didn’t just stumble Hugh Grant style into the room accidentally, you were there because you were interested in the community. When you were sitting there, I don’t know, what was attractive to you about the Cloud Foundry community and the Foundation in particular that made you mentally A-OK to accept that offer?

Sam Ramji:
Well, there’s this awesome moment at different decades in technology where you see one particular project is poised to solve a huge problem and it’s got all the right enthusiasm tailwind, all the possibility involved. I think there was a moment for each of us, we’ll probably each have a story about when that was true for C++, or when that was true for Linux, just that crystal moment.

That crystal moment had come to me before the board meeting for Cloud Foundry because the thing that I had been doing the prior half-decade, was I was chief strategy officer for Apogee, which just recently went public. Good testament that we at least got a few things right there. We solved some problems for some pretty big customers.

All of these non-IT companies, financial services, telecommunications, health care, heavy industry, we’re all becoming software companies. That transition means that they have to learn a lot of new stuff and they have to also bring in a lot of new infrastructure. A once-a-decade re-platforming process where you bring in a new set of capabilities.

As you pointed out recently on your diagram that you Tweeted, every IT story ever where you start by serving the technology, and then you start to wrestle with it, and then you land in it. Then, the technology starts to make you happy, and then you rest on your laurels, and then everything breaks and you go back to the beginning.

We were clearly at that moment for platforms as a service. There’s this Cloud computing that has become obvious. We all want an elastic infrastructure, we want our compute, we want our storage, we want our networking to all be available when we want it. We want on demand IT resources. A compute cycle and a buck won’t even get you a cup of coffee. You have to have an application to do that.

The abstraction level that’s been provided within cloud computing is basically horrible for developers. You have to know a lot about even building machine image. How do you load that? How do you make sure that the network wraps are balanced? There’s a lot of stuff that, yes, people are good at that, but not every developer is good at that.

There’s this normalizing function where you say, “Okay, how many people in the world can do this? Can 25,000 people do it? Can 250,000 people do it? Can two and a half million people do it? Can ten million people do it?” As we start to look at where all of these companies have to go, there aren’t enough developers on the planet for every global 2000 to become a software company, yet they all have to become software companies.

There isn’t enough money in all of the global 2000’s, let alone each of them, to completely re-platform everything on brand new proprietary stacks. Between those two inexorable forces, your Scylla and Charybdis, something’s got to give. Argos somehow has got to get past the clashing rocks.

In that moment I think Cloud Foundry already showed up as a ship with a tailwind that solves these problems fundamentally that’s open, that’s available for free, that also has a commercial ecosystem around it so it’s got sustainability.

The opportunity to build the Foundation is having this open, transparent, trustworthy governance model where both the business and the software project get governed in a way that users, vendors, and developers all get treated fairly. Their impact on the roadmap governance by contribution, all of these core principles of collaboration can come true.

If we get that right, this is the football that’s ours to either take over the goal line or to fumble, and I wake up every day worrying about if we’re fumbling it today. Then, I try to go to sleep saying, “We’ve taken it a little farther down towards the goal line.” There’s something amazing that we can all build together over the next couple of years. I think Cloud Foundry will still be default infrastructure in every enterprise 20 years from now.

Coté:
Yeah, I like the way that you framed part of, I don’t know, word like a mission is charged, but the mission of the overall community, which is one, there’s an entropy or a half-life to innovation. The things that were innovative decay and you’ve got to refresh them.

Sam Ramji:
Mm-hmm.

Coté:
At least polish the patina off if you want it to be shiny brass instead, or however you pronounce that. Then, there’s another thing that connects to another topic I wanted to discuss with you which is who is the Foundation acting on behalf of?

Open source is always on behalf, well, it’s always well-positioned to be on behalf of the end users, the developers and the businesses using it. Oftentimes, when a lot of developing organizations get together, they end up being vendors and they can often be shared code or it’s not necessarily focused on helping the end user become better software developers.

It’s focused on creating tools that are shared in common in the industry, which is an overly-nuanced way of putting it. I like the notion of ‘we’re really trying to create the platform that allows you to write software on top of it’ and helps you become, as I like to think about it, software defined businesses, if you will.

Sam Ramji:
I think it’s a great form of reference.

Coté:
I think with that, I mean, now that you’re sitting at the helm with the other leaders of the Foundation, how are you thinking about doing things differently or different experiments that you’re looking to run when it comes to running the Foundation? How would you like to do things differently than other open source foundations have done them?

Sam Ramji:
There are a few dimensions of the experiment that we’re running here. I think one, just to reiterate what you said, and to put it in our frame, which is my favorite movie quote ever is from Tron. There’s this moment when Flynn has been put into the computer and he wakes up and he’s a prisoner.

He’s looking at this other combatant who’s just doing amazing stuff. He’s got his disc, he’s knocking out the computer components and Flynn goes, “Who’s that?” It turns out to be a program he created. That comes out a little bit later. “That’s Tron. He fights for the users,” which is just awesome, right?

That’s got to make you smile. That’s one of the big dimensions of our experiment here is to really re-balance software creation, and software roadmap, and contribution to core infrastructure that companies rely on, back towards the users.

We must continue to have critical mass and Pivotal has done extraordinary work in creating that critical mass. IBM, SAP, EMC, HP have all gotten on board and paddled hard. ActiveState. Mirantis has joined, Mendix, we have about 40 members. I’m not going to list all of them. There’s a set of contributors and Huawei of course recently announced and joined the Foundation as well.

They’re all creating a critical mass, which is required to have something that’s worth contributing to, but now that that’s there, we’re seeing JP Morgan Chase, we’re seeing BNY Mellon, we’re seeing AT&T, Verizon, Swiss Com, all of these different companies going, “Wow, there’s enough there and your process is clear enough.”

You go through dojo, contribute to a subsystem, commit your developer at 75% or more for a year or more. Continue to work on the subsystem. Make sure you do that because it’s delivering business value to you. It’s your opportunity to govern by contribution and influence the roadmap. All of that stuff now is the phase of experiment that we get to operate here. It’s the first dimension, I think, that is a bit different and is clear governance for clean contribution.

Coté:
Yeah, and that’s the part of the Foundation that I find fascinating is the way it’s attracting end users. I don’t know how else to categorize them. Just as you were describing, clearly there’s vendors involved who have all sorts of interests, commercial and otherwise and whatever.

There’s people who want to do that, which is perfectly normal and healthy, but then … I should stop being surprised, but I’m always delightfully surprised when someone like the actual non-technology companies you mentioned come on board. It’s not so much that I wouldn’t want them to do that, but the other thing you alluded to is it actually implies a pretty heavy commitment on their side.

One of the reasons I would pause it is that vendors have heavily been involved in foundations in the past is because that’s their job to write software and product. Whereas, from an end user perspective it hasn’t always necessarily been, ‘I’m going to have dedicated staff to work on infrastructure instead of end product.’

Sam Ramji:
Mm-hmm.

Coté: I think that’s where this out of the entropy of innovation that we’re in the middle of at the moment, that’s the great thing, it’s nice to see end user organizations realizing that, “Oh, we should start working on the platform and the infrastructure.

It will behoove us to be involved in that.” It’s great to see that they’re willing and that they can take this time commitment basically, time and money commitment to be coming part of the community.

Sam Ramji:
Yeah, and the funny thing is, the more conversations you have the more that you realize that we shouldn’t be surprised by that at all. The fact is that most global 2000’s have a really amazing core set of engineers within their IT groups. Now, these IT groups could be huge. They could be 1000 people.

In the case of Citi, it’s 40,000 people. You can’t have 40,000 A+’s because there’s just not that many in the world and they can’t all be in a bank. You’re going to have 40,000 good developers in a set of who have always had their heads on infrastructure.

You look at Goldman Sachs, when I was at Microsoft in 2007, we kept asking how can you run Windows as well as you can run Linux. We were talking to the Goldman folks and they said, “Yeah, it took us about ten admins to run a thousand Windows servers. It takes us four admins to run 20,000 Linux servers.

There was some infrastructure innovation there in a deep way and the more that you look through being used as a technology, what I call the one percent industries, where if you get a one percent improvement on your cost, you get a billion dollars to your bottom line, whether it’s telecommunications and banking are classic examples of those heavy industries coming on in that way, utilities, energy, oil, gas.

They’ve actually been building platforms. When we look at JPMC, BNY Mellon, just as members, and other banks that I’ve spoken with that I’m not going to call out by name, they’ve already built application platforms for the Cloud that work on heterogeneous underlying data center operating system infrastructure.

The opportunity here is to say, “We realize that none of us is as smart as all of us. These capabilities are actually non-differentiating for each of you and we can give you a trustworthy way to collaborate on an open source project that is going to be mutually owned. You’ll probably want to take the lessons that you’ve learned building your own platforms and now instead get out of your offices and into the commons.

Let’s all do this in Central Park. Let’s make this a great place for all of us to benefit, learn, take something back, use it. Then you can put all of your extra effort on differentiating your business through new apps and services, but don’t be surprised of the fact that there are amazing infrastructure engineers in every one of these major companies.”

Coté:
Certainly. Now that I go on a lot more sales calls, I encounter pretty much the same thing. Everyone already has, or more than likely, has been building a platform for quite some time, in that they’re always looking to improve. Just like vendors have come together in the past to get the shared common code for GUI stuff and UI’s and things like that.

I don’t know if it’s self-aware in the industry, but there’s almost this latent desire to not duplicate that effort of building a platform across all those data centers.

Sam Ramji:
That’s exactly right and I think it is approaching the point of self-awareness there. I think another dimension of experiment we’re running here is can you get massive growth while also constraining the definition of what you’re doing? One of the great initial conversations that I had just in February was with Jonathan Bryce, who is the executive director of the Open Stack Foundation.

He said, “I’d love to spend some time with you, could be some lessons learned. Here’s some things to focus on, things that worked. Here’s some things that maybe I’d do differently.” One of the areas that everybody, including Jonathan has said is, “Make sure you focus on certification.

Make sure that Cloud Foundry means something and that when people say that and they have the trademark right to be able to put that on their product or assert that they’re Cloud Foundry compatible, that that means something.” What does it need to mean? It needs to mean you’re running a common set of bits.

You get a standard, you get portability as a result of that standard. You get interoperability as a result of that standard at the same time because as a foundation our mission is to see a sustainable commercially successful ecosystem. That’s what we have to do in order to maintain this vision of Cloud Foundry will be a mission critical technology in every enterprise in 20 years.

Still, it will be mission critical before that, but it’s still going to be running then. That sustainability has got to be grounded in commercialization. When we look at Linux there’s one company that’s really made Linux very commercially successful. That’s Red Hat. Everybody else has got it as a mix-in. Why is that?

The lesson we can take from Linux, and we take a lot of lessons from Linux because Cloud Foundry Foundation is actually part of the Linux Foundation and Jim Zemlin has a ton of experience, ans has got some great strategic models for us to borrow from. Linux had a certain approach, which Cloud Foundry is what’s called open core.

We are going to certify what does Cloud Foundry mean by making sure that people have a core set of bits and that those bits are the test for compatibility, rather than having a lot of API tests, secondary testing suites and everything else.

You’re going to have to stick with our trunk and as they say a strong main line is the best proof against forking and shipping quality code in the CF release every month is a great way to keep Cloud Foundry coherent and whole.

Been around that, IBM, EMC, HP, Merantis, everybody who wants to participate in this ecosystem, they could all add their proprietary or open source value add around that core Cloud Foundry.

What is Cloud Foundry? You’ll get it from GitHub.com/CloudFoundry. You pull down that repo, you’re going to have something that works and runs. If you want also to have access to Watson, then you’ll run it in IBM’s Cloud.

If you want to run it against Greenplum and Hadoop and you’ll run it in PC. There’s going to be a lot of differentiation, but we have to balance those two edges where there’s a meaning to the term Cloud Foundry and we’ll continue to create space and opportunity for a vibrant commercial ecosystem.

Coté:
Yeah, no, I think Jonathan’s advice and then obviously your internalizing and believing it is spot-on. To be the back-in-my-day guy, one of the things I’ve been lamenting over the past few years is that there’s no J2E spec. There’s no shared understanding of this is the application development architecture and language and things like that.

In fact, arguably, it’s great that there’s not because that means that there’s tons of innovation around figuring out what it is. There’s been all sorts of great exciting things going on. Then, on the other hand, when you’re doing whatever the opposite of entropy of innovation is, it becomes nice to have some sizable chunks of standardization or commonality.

I think it would be great if we can pull off having a standard for what … Again, the words will change and they’ll evolve because it would be different than a spec or a specification like back in my day as it were.

Sam Ramji:
Mm-hmm.

Coté:
It will be the commonly understood set of patterns and things and even API’s that one does to make Cloud native applications essentially. I think, I mean, based on the conversations I’ve had over the past few years, that will be incredibly helpful.

Sam Ramji:
Yeah, it’s crucial, so on top of what you’re saying, Yefim Nadis from Gartner, he’s responsible for looking at application platforms as a service, which is basically the gross category that the Cloud Foundry occupies.

He used Java and J2E as an example of standardization and a good standards process and certification over and above that. You couldn’t call it Java if it didn’t pass the test, as a way that you start to create a coherent ecosystem. Certainly Java created a tremendously successful commercial ecosystem.

On the flip side, another very widely adopted technology, but there was no standardization and no commercial bet to, was PHP. That created almost no value for any software company ever. It created value to other people, maybe to consultants and the companies that use them, but the sustainability there ended up being a lot weaker.

As a result we’ve seen a lot of movement beyond PHP to other languages on the web. Not to be a language-centric person. As you know I’m a Buddhist, so I like a platform that makes me one with everything. Whether it’s Java or Go or Ruby, you don’t want the platform making bets on the particular product, that certification is crucial.

I think in 2015, Occam’s Razor for specifications, spec combat-abilities, reference tests, it’s actually running our code. One of the things that has

Coté:
Yeah.

Sam Ramji:
… Struck me about Cloud Foundry is how mature the development practices are for it. Everything from simulation testing, which is really breaking new ground about how you make sure that complex distributive systems actually work at scale before you call it ship-able to internal tests to make sure that you’ve got API code compatibility over time that all goes into the open source.

If you’re contributing through the Foundation to the project, the intellectual property that is owned by CloudFoundry.org, you are participating in a process that generates known good backwards compatible code by design all the time. We don’t need that to be an area where commercial players add additional value.

That’s remarkable maturity and that’s one of the things that gives me the best confidence that this is going to be a tremendously successful enterprise technology. It’s already been through a couple of major versions. Compatibility has been good.

We’re coming up to another significant revision with Diego, which is going to greatly expand compatibility with things like Docker, while also improving performance and manageability. The thought that’s already gone into making sure that that is a clean, smooth experience for every enterprise that is already adopted Feed 2, it’s pretty amazing.

There’s guidance on how to do a blue/green deployment and shift your code load over time, and how to do that without having to require a lot of application developer engagement, and do all this manual recompilation, and going through that whole process. That’s pretty extraordinary.

Coté:
Yeah, no, the notion of escaping the irony of not dog-fooding your own thought technologies, which is to say we should be using Cloud to basically do all this certification and things like that, instead of having it be a manual process or not taking advantage of … I mean, frankly the idea that we could, in the community, run multiple of our own Clouds to make sure things are compatible by whatever rules we come up with, right?

Sam Ramji:
Mm-hmm.

Coté:
It certainly seems like it would give us the ability to run much more complicated tests than we’ve been able to before.

Sam Ramji:
Exactly right

Coté:
… Quickly.

Sam Ramji:
That’s right and the solution is the simple solution. Again, plugging Occam’s Razors, run the same code.

Coté:
Exactly.

Sam Ramji:
Run our code. If you’re pulling code from a repo, regardless of exactly what snap moment in time you got it, that’s the code that you should make your commercial product around. When you need to rev that, rev that based on code from a repo. That’s how we’re going to certify you. It’s not going to be an API test and all of this stuff because when you get into those you actually do enable ecosystems but they end up being frequently very inefficient.

A good example of this is to look at NFS before. The specification was thorough, thoughtful, complex, took years to build, building NFS before compatible implementations was hard to determine because it was so hard to build the reference test for that.

Yes, it enables a lot of the different companies to build integrations to it, or implementations of it, but you end up with this very long, drawn out process that is pretty difficult, to the extent that I do have biases. I’m certainly biased towards implementation, open source implementation as the standard, as opposed to specification as the standard because it actually cuts through so many problems.

We have the opportunity to do that because of a layer of technology that we sit at. We are not managing Blue Tooth. Our stuff is not getting put into Firmware. You don’t have to have the specificity and precision in saying, “Once we ship this it’s going to be in the field for ten years.”

Coté:
Just two more questions here before we wrap up. One of them was in that we have a lot of, again, end users or whatever you want to call them, non vendors involved in the community, and I would presumptuously assume you talk with several of them regularly.

When you compare the types of concerns and the questions and the things that they have with what you’ve typically encountered over years with the open source vendor world, what are the topics that are unique to end users and being involved in an open source foundation? What are the things that they either stumble with or don’t understand and then delightfully have epiphanies about, or the core issues that tend to come up over and over again that they need help with?

Sam Ramji:
Yeah, I think the delightful epiphany is going from, “Hey, we thought it would be a good idea to join the Foundation to demonstrate support.” The same reason that I write a check to the Sierra Club every year or to the Yosemite Conservancy Fund, we’re with that. Certainly that’s where we see donations coming in to the Drupal Foundation or the Linux Foundation.

The conversation that we’re having that produces that epiphany is, “We’re so delighted to have you as members, we want to make sure that you’re getting as much value out of this as possible. We know that you’re taking a strategic dependency on this technology and we want to make sure that you both understand the road map and can contribute to it.

That you have another path of influence on how this gets built as opposed to just the traditional path of going through your commercial vendor and then being dependent on the amount of money that you have to pay them to influence them and hopefully the things that you want happen over time.

Then, they start seeing this, “Oh, hey, I don’t want to become the software distributor. I do want to have my commercial relationship.” Having this other path of influence, it’s like the big ah-ha effect, especially when you’re talking to people in IT.

I don’t think it shows up to the line of business quite the same way. They’re more like, “Hey, we’re innovative. We want to put our logo on the website and then we can seem shinier,” which is great the effect we have on marketing benefit to those who want to contribute.

That’s great, but we’re deep pragmatists, those of us in the Foundation are primarily Silicon Valley, start-up as well as public company veterans. We’re taking this very pragmatic view of what’s valuable to you that’s going to make you keep renewing your membership? We want to make sure that you’re interaction with the product, the project and the developers and the Foundation are all in a direct line of sight between what you do every day and how you succeed.

Coté:
Yeah, it’s like not only can you be supportive, but you can actually get involved and start to control the roadmap if you will, start to, as with everyone else you can get involved and start working on projects.

Sam Ramji:
Exactly. Exactly right. The other nice is there’s a harmony across a number of different open source technologies that exist today. Some of them are single vendor open source. Some of them are multi vendor open source. Some are more commercial. Some are less commercial, but those are the technologies that have taken over the enterprise data center.

Having those conversations with Mongo or with Redis or Docker in making sure that these things work together. An open source foundation is a really natural place to align to that where we have clearly no axe to grind and we have a tailwind of goodwill being able to step into those conversations where they might feel out-matched or threatened or pressured by a very large successful commercial vendor.

We are already on their level. We’re a small group of half a dozen people at this point. We’ll probably be no bigger than a dozen a year from now. Our mission is pretty simple. Our finances are public. That makes it easy to connect with other open source projects and make sure that we’ve got some technical harmony between them. That’s really the place that I light up.

I’d like to see us have great compatibility between Cloud Foundry, Docker, Open Stack, Node, Open Daylights, to have all those things just work really naturally together because it will delete a lot of inefficiency out of pretty much every enterprise that’s still competing today.

Coté:
Yeah, no, there’s a nice chance to avoid not only being mono-vendor and therefore mono-technology, but also being mono-company type.

Sam Ramji:
Yeah.

Coté:
There’s a much different … I imagine it’s happened already, but I imagine it’ll happen many times over where there’s some critical decisions to be made and it’ll be fun to have vendors and end users in the room who have different priorities and they’ll have to sort it out.

Sam Ramji:
Exactly.

Coté:
In a harmonious way.

Sam Ramji:
Right, and then also vendors that are powered by open source communities, whether it’s Docker or whether it’s Cloud Soft. Cloud Soft is the inventor of the Apache Brooklyn Project.

Coté:
Right.

Sam Ramji:
Yes, they’re commercializing it, but they’re not venture funded looking for a billion dollar payday. They’re, “This is really useful technology. We want to see this work. Can we continue to make something out of that?”

Having, as you said, a non-homogeneous membership is super important from the big vendors who are going to drive big adoption from the smaller innovative companies who are driving new ideas.

Then from this whole expanse of different enterprises who didn’t ever think of themselves as being software companies before, but now are just being pushed there by digital transformation. There’s no choice. You become a software company or you go out of business.

Coté:
Right. You’ve already previewed … Well, I should say, you’ve already answered some of this question. We’re talking about certification and obviously getting more people involved and things like that, but as you look out and plan over the next year or so, not only for the Foundation but for yourself, what do you think the possibilities are, your hopes and dreams that will happen over the next year? What are some little presents you’re setting yourself up for?

Sam Ramji:
Good question. I’d like to see us become the acknowledged gold standard for Cloud platforms. I think the way that we get that acknowledgement is not just from the tremendous weight of vendors that we have working on this, IBM, Pivotal, HP, EMC, VMware, Huawei, Intel, SAP, it’s an embarrassment of riches.

All of those companies are delivering products this year and revving it next year based on Cloud Foundry, making sure that everybody is aware that those products are all based on Cloud Foundry. That’s something that we absolutely have to get done.

If we’re sitting here a year from now and all of them have products where they’ve got Cloud Foundry right in the name, just like Red Hat Enterprise Linux or SUSE Enterprise Linux or Hoopla to Linux. You know Linux is right in there. I’d like to see Cloud Foundry get that kind of name recognition. If we do that then the next several years get much easier.

Coté:
Yeah, it’s the old Intel inside marketing stratagem.

Sam Ramji:
You are exactly correct. I think that the second big thing is I’d like to be able to count a dozen ISV’s of various sizes that have made strategic bets on Cloud Foundry that they’re shipping Cloud Foundry based product because they see it as a great market. They see it as widely adopted in very important prospects, customers for them and the deployment of servicing costs are going to be so low that they’re just compelled to ship on top of Cloud Foundry.

Obviously it’s near and dear to my heart because I’ve been working as an ISV off and on for the last 20 years and trying to build a network enterprise software company has always been hard. If there were a fabric that could be deployed to, we could create so much new efficiencies and so many new opportunities for a field of software vendors that could be really valuable to everybody.

The second thing I’d like to see is 2016 is the year that we go from enterprise adoption of Cloud Foundry which has already taken a huge up-tick, it keeps growing strongly, to ISV adoption of Cloud Foundry.

I guess the third thing is I’d just like to see a real broad base of contribution to the codes. I’d like to see a wide range of different companies for different purposes, but doing it in a sensibly aligned way that’s tied into both Cloud Foundry roadmap and also to the Cloud Foundry development process.

In order to do that I think we need to take more credit for the opinionated way that Cloud Foundry gets built. It’s built on para-programming. It’s built in a fundamentally actual way that has produced better software. We’re going to continue to grow the basis of how that’s a great software model. I think a number of contributors, committers that we have is going to be a great measure for that.

I’d like to see us double that twelve months from now where we say, “We’ve got twice as many developers who are A, trained and B, committing in a consistent way to make Cloud Foundry a great platform.

Coté:
Yeah, I mean, that last point reminds me of something … When I look at the machinations of the community and the Cloud Foundry Foundation, one of the things that novel from the perspective of, I don’t know, studying the IT industry as a somewhat pathetic hobby of mine versus woodworking or whatever else one might do for a hobby. I think the Cloud Foundry community is unique in the process that it’s trying to undertake, right?

Sam Ramji:
Yes.

Coté:
You were mentioned pair programming and things like that and looking over the past 15 years or so of open source development, getting really orthodox and specific about the methodology hasn’t really come up that much. Every now and then it does, but it’s not that big of a deal.

Sam Ramji:
You’re exactly right.

Coté:
I think our community is a lot more specific about wanting to be agile.

Sam Ramji:
Mm-hmm.

Coté:
Particular types of agile and practices. It’ll be fascinating over the next few years to see how that works out. We see this at Pivotal quite a bit. There’s a Pivotal way of doing software development that has great results, but it’s dependent on doing things in that way, having para-programming and sitting at the same desk, and all of these kinds of things that doesn’t always work for people.

I think there are ways of making it work if you force the function of, just to pick one example, having to be remote. It’s almost like we’re giving ourselves the task of not only trying to sort out a Cloud platform, but also monkeying with the core methodology for writing software, which it’s a nice side effect to have if you can improve that.

Sam Ramji:
I think that’s right and I think it does a couple things. It forces real commitment in order to contribute to Cloud Foundry. You have to leave your previous biases behind, you have to open up to the idea that this project has been built this way and that it is the right way, and that if you want to contribute you need to learn and become a student again.

We have people who are in their mid-40’s who are distinguished engineers at giant public companies and they have to interview. They’re like, “I’m not going to interview for a software job. Are you kidding?”

We’re, “No, actually you do have to pass the RPI and then you have to do pairing. Then you go through dojo where you’ll come in as a white belt and you’re going to leave as a brown belt. That’s going to take you probably eight to twelve weeks, but then you’ll have commit rights and you will have made meaningful impact on the code and then you can go on and do a lot more. Then, you can start doing this remote stuff.”

Every filtering function I think is valuable because it separates those who we really want as part of the Cloud Foundry contribution community, someone who’s cares about it enough to put all that effort in, and will continue to be a part of our development community for another couple of years versus some open source projects are open to drive-by upgrades.

“Hey, here’s my PR. Please take my poll request,” or, “Hey, I fixed a whole bunch of bugs. I’m going to mail you a zip file and you should just sort of do a dif merge on that and it’ll take care of your out of memory condition.” Well that’s not really the way to communicate with us. We’re putting our money where our mouth is.

The Cloud Foundry Foundation is opening the first dojo that is not a Pivotal dojo in San Francisco on June 1st, 2015, so that’s just a few short weeks from now. We’re going to have engineers from different companies coming and dojo-ing.

That’s our opportunity to do this very thoughtful and opinionated process to ramping people up and giving them hopefully a deterministic path to becoming a committer and a dedicated committer to Cloud Foundry.

Coté:
Yeah, no, I think that’ll be great to see how it pans out. Then, for me the exciting part is always once you build up a bunch of mini masters, then they start reinventing things, right?

Sam Ramji:
Mm-hmm.

Coté:
Coming up with new ways of doing it, which hasn’t really occurred in methodology land for awhile at large scale. It’ll be fun to see what happens there.

Sam Ramji:
Yeah, we’ll pay careful attention.

Coté:
That’s right. Well, thanks for taking all this time to go over the Foundation stuff. I get asked a lot of questions about it, so many of these things were just me parroting what people ask me all the time. I’ll have to go back and see if I answered correctly.

As ever, I think with things like this, it’s good just to get to the spirit of the manner and go over the goals and give people a sense of why an organization like this exists and what it wants to accomplish, to know it, if you will.

Sam Ramji:
Awesome. Yeah, we want to create a world with Cloud computing that is ubiquitous, portable and vibrant and if we can get ubiquity then we can give everybody portability of applications. If we give everyone portability of applications then we can create a vibrant ecosystem of applications and developers that are just constantly driving more innovation for all the things that we try to get done in business every day.

Coté:
Where would you point people to if they want to start drinking some Koolaid or whatever the funny Japanese word equivalent is?

Sam Ramji:
I don’t know the funny Japanese word equivalent, but I would go for a nice cup of Genmaicha. Go to CloudFoundry.org, https://CloudFoundry.org, dot org, that’s the foundation’s website. That’s got some core information, blogs, updates, Tweets, dreams, pointers to everything else.

Frankly, then go to Github and go to Cloud Foundry repo on Github and that will give you the entrée into signing up for google groups in the development, building us everything that you could possibility hope to find there.

By the time people listen to this podcast they should also go to CFSummit.com because by the time you hear this you should see almost all of the videos, talks, presentations that the community of a hundred plus presenters should all be available.

Our job as an organization is to harness the wisdom of all of the community practitioners and then reflect that so that it’s available for the whole world. Hopefully that will all be available at CFSummit.com, at CloudFoundry.org by the time you listen to this.

Coté:
Also as a hopelessly biased plug on that, I mean, I think before I joined Pivotal, watching those recordings of past CF summits was part of what won me over to not only working at Pivotal, but being part of the Cloud Foundry community. I think it’s a good conference.

Not only the technology talks, but finding the actual end users get up there and talk about all the excitement they’ve been up to is a rare treat. It’s not the usual computers are awesome stuff. People are pretty brutally honest about the problems they had and how they sorted them out.

Then, my favorite end to a talk like that is always, “Well, things seem to be going okay. We’ll see how it pans out.” Just the brutal honesty of, “Now all the hard work begins.” I think the tone of a lot of those talks are in the line that I find useful.

Sam Ramji:
I love that and it’s very much what made this a great place for me. A privilege to be a part of this community is that it’s just a good personality fit. I’m deeply pragmatic. When we have a really good day and people are, “That was a great day.” I say, “Yep, great day. We’ve got more work tomorrow. Just keep going.”

Coté:
Well, on that note, I’m going to go scrounge around the floor here for three exciting things that happened this week and close it out. Thanks, as always, for listening. We’ll see everyone next time.

If you want to hear our back catalog of exciting Pivotal conversations podcasts, you can go to Pivotal.IO/Podcasts, plural with an S at the end. As always, you can also email us at podcast@Pivotal.IO or harangue me through whatever channels you want. I appreciate you taking the time to listen in.

About the Author

Biography

More Content by Coté
Previous
By The Numbers: A Cloud Foundry Summit 2015 Recap
By The Numbers: A Cloud Foundry Summit 2015 Recap

Coming off the heels of the Cloud Foundry Summit, the Pivotal team is excited about all of the activities t...

Next
Build Newsletter: Open Source, Data Platforms, Mobile, Containers & PaaS—May 2015
Build Newsletter: Open Source, Data Platforms, Mobile, Containers & PaaS—May 2015

The May issue of Build brings some of the latest, most important events for developers and architects who a...

Enter curious. Exit smarter.

Register Now