Third Platform: Trends and Strategies
 
 
 
 
 
 
An in-depth discussion of the Third Platform, focusing on how businesses can use it to gain competitive advantage.

See transcription of video, below:

The so-called Third Platform, comprised of social, mobile, cloud and Big Data, is radically disrupting computing at many levels. This emerging toolset enables smaller, nimble companies to gain competitive advantage against larger rivals. It also decentralizes computing to an enormous degree.

To explore this trend, I’ll talk with Bernard Golden, named by Wired magazine as one of the 10 most influential people in cloud computing. He’s the author of the highly regard Amazon Web Services for Dummies. Golden is the Vice President of Strategy for Active State Software.

third platform, trends and strategies, talk by Bernard Golden

Bernard Golden, Active State Software

We'll discuss topics such as:

1. What is the Third Platform and why is it important?

2. What are some examples of Third Platform initiatives?

3. What kinds of tools does a company need to become a Third Platform company?

4. What is the right way to get started on a Third Platform journey?

Graphic courtesy of Shutterstock.

 

VIDEO TRANSCRIPTION: THIRD PLATFORM:

 

J. Maguire:

Third platform: I know IDC is all over this idea. It’s an upcoming idea. It’s emerging technology. What is third platform? What does it matter to people?

B. Golden:

Well, IDC put this concept forward at the beginning of 2014, and they said there is this new confluence of technology trends in the industry. They called it Third Platform. It’s a confluence of social, in other words the transparency of widespread communication, instant availability of knowledge, crowd sharing, whatever you might want to call it. Mobile, meaning mobile device, computing being in your pocket no matter where you are in the world. Analytics. Big data. So you can really sift through all this knowledge and data and gain operational insights that you never had to before. You can substitute analytics for opinion. And then cloud, which is sort of the operational or the computing infrastructure, which gives you the ability to have massive amounts of computing available on demand.

They said this confluence, they call it Third Platform, I’ve actually heard other people call it SMAC, Social Mobile Analytics and Cloud, which, you know, you’re talking SMAC. They said this enables new kinds of applications to be brought to the floor.

And what was interesting about them was they said so here’s the new technology trend, which is great and that’s what technology analyst firms do. The interesting thing they said was they made a business prediction and they said based on this new third platform capability every industry, the top twenty entrants, a third of them will face significant disruption by either some new entrant coming into the industry, or some revitalized. So basically they said no matter what industry you’re in, somebody is going to come after you with this third platform. So you’re choice is do I sort of gear up to be able to play in that space, or do I recognize that my business is going to be disrupted. That’s the third platform.

J. Maguire:

The interesting thing about that is that a lot of those elements, they can be very expensive, they can be done on large scale, or a smaller company that may or may not have enormous resources can also leverage those four things to take on a larger rival. So it’s the democratization of technology in many ways.

B. Golden:

Well, for sure. We were just talking before we came on, you know, we’re using Google Hangouts. This is unbelievable technology. I worked for a company a number of years ago that decided we’re going to do video conferencing. We had a number of locations. Oh my goodness, it was millions of dollars. And by the way, you could only do it when you went to this special room in the facility, and this room was always booked up. And yet Google now provides this thing that we’re communicating with however many people are on this thing. We’re talking. It’s free.

J. Maguire:

On a desktop, yeah. Log on and you’re there.

B. Golden:

And when I was running my own consultancy a few years ago I would basically meet and engage with and close business with people I had never met physically because I had this kind of connectivity to them around the world. It’s incredible, the democratization to this technology that comes forward.

J. Maguire:

Especially with Amazon Web Services, it’s one of the ways. I mean, way back in ’09, ’10, ’11, it was a lot of smaller companies getting on board with their cloud platform, and suddenly they could rent compute. You know, who needs a data center anymore? It made them seems like a larger company.

B. Golden:

Yeah. It’s funny. I ran across this company a few years ago called Hudl. And I had never heard of them, but I saw the guy give a presentation at the Amazon ReInvent Conference. I’d never heard of them. Didn’t know anything about them. He gets up and says we do football videos. You always hear about teams, they watch film and blah blah blah. Well, I had no idea. But they do that down to the high school level. And if you’re a high school player you need to have your roll or your thing that you can show off to a college that you want to go to. This is me making passes or tackles or whatever. Okay, makes perfect sense. Before it was like an individual assistant coach who would do that. And then it was like where do you store it and all that trouble. Basically, we’re going to make kind of a service to do that, a SaaS service to do it. Again, I’d never heard of them. Didn’t know anything about them. The guys says, yeah, we’re a company of, I don’t know, 15 people or something like that. He said we serve up 6 hundred million videos a month.

J. Maguire:

Wow.

B. Golden:

And I was like, wow, massive. And not only that…

J. Maguire:

And all that from a company no one has ever heard of.

B. Golden:

And this is like a trivial customer for Amazon. And so the democratization is a really good point. Would it be useful to share and example of Third Platform to kind of bring that home?

J. Maguire:

Yeah, let’s look at some examples of what Third Platform is.

B. Golden:

I’ve got some slides here. Let me bring it up and show it off here. I’m going to see if I can switch over to screen sharing here.

J. Maguire:

Do we have the technology?

B. Golden:

Yeah, well, we’ll take a shot at it I think.

J. Maguire:

Okay. I mean, we put man on the moon. Let’s see if we can do screen sharing over Google Plus.

B. Golden:

Yeah, let me see if I can find my… What you should be seeing coming up is a slide that says ‘The Impact of Third Platform’.

J. Maguire:

Let’s see. If you speak it will probably come up.

B. Golden: Alright, can you see it’s a big shattered glass, and it says ‘The Impact of Third Platform.”

J. Maguire:

Yep. Go right ahead.

B. Golden:

All right. Well this is an example of the third platform. Let me just bring this down. I’ve got my slides here. And the one that we use is one that actually a lot of people talk about, which is Uber.

And, you know, let’s just talk about the taxi industry. The taxi industry has been around for a hundred years. Actually, I should put this picture in here. I had an opportunity to visit the Museum of London over in March. And they’ve got an actual London taxi from 1908. And the thing is, it’s a recognizable taxi. It’s not that different than a taxi you get into today. I mean, obviously, the car is quite different. But it’s an automobile, you get to sit in the back, there’s a meter you can look at which is just like the meter that was running on the taxi that I took last month when I went to InterOp in Las Vegas. It’s an industry that has been around for a hundred years. It’s gone through two massive wars. It’s gone through the Great Depression. And yet it’s been like a money-spinner. I mean, it’s just this kind of thing.

But if you look in San Francisco, over the last 15 months, 2013 through the end of 2014, San Francisco taxi trips are down 65%.

J. Maguire:

So it was an industry ripe for disruption, to be sure.

B. Golden:

Well, it was an industry that had been enormously successful for a hundred years. It was a life stream of money. And then Uber comes in and they basically take away 65%, two-thirds of the business in this industry in the San Francisco market over the course of 15 months. That’s how powerful the third platform can be.

J. Maguire:

They’re leveraging mobile. People call Uber with their smart phone. It’s cloud based, everything. And Uber is leveraging big data, as well.

B. Golden:

Well, I’ve got a slide here; let me just advance a little bit. I think this is my slide here. These are all booked up things here. I think this has got a slide, but I won’t be able to show it. Basically this is social, right? Because basically you rate the drivers and the drivers rate you.

And Maureen Dowd, who is a columnist for the New York Times was complaining… She took her column and was complaining that Uber wouldn’t come pick her up. And it turned out that she had a terrible rating.

J. Maguire:

Really? The drivers didn’t like her.

B. Golden:

Yeah, and so that really affects it, so the social, and this transparency. So if you’re a lousy customer you’re not going to get rides. And by the same token, if you’re a terrible driver nobody will use you.

So this social piece of it, mobile obviously, its like catnip using Uber because basically you pick up your phone, you say I want one of these things. It’s knows where you are from the GPS. It makes the connection. It tells you which driver is going to come and get you. You can watch the driver getting to you on a map because they have GPS in their mobile. And it’s completely mobile driven, right? And there’s no payment because basically your account has your credit card. So when they drive up and drop you off you just get out of the car and walk off.

J. Maguire:

There’s no comparison with a traditional taxi and Uber, really. It’s just true.

B. Golden:

They’re running analytics. I look at my account. I’ve used Uber in I think 7 or 8 different cities in 3 different countries. It’s got every ride. Well, guess what? They’re looking at that, too. They know a lot about me. And then it all runs in the cloud.

And what’s interesting is I showed you this slide that shows you this, but just as interesting is this slide, which shows you what they’ve done in the San Francisco market. So, in the course of 5 years they have tripled what the taxi industry used to do in all of San Francisco. So all of the taxi rides in San Francisco, they out do three times that business because, as I said, it’s like catnip. All of a sudden it’s much easier to use, it’s very cost effective, and all of a sudden people start going, I’m going to take a taxi ride. Instead of me, you know, walking to this place or taking a bus or not going, I’ll just do this.

J. Maguire:

It’s a classic third platform play.

B. Golden:

It really is. It’s a great example of a third platform. And that’s the power. What’s what this third platform brings to the table? That’s why it’s so important. And that’s why IDC said this is going to come after every industry. So this isn’t like IT inside baseball. You know, which database vendor is going to do more or less, or what’s the next version of…

J. Maguire:

The SAP product or something.

B. Golden:

Yeah. All of a sudden this is now moving into real straightforward business, and a huge business impact. Just to show one more slide here in this little section of the discussion, and I apologize because I don’t know if we’re going to be able to get this slide to come up or not, it’s sort of a build… Let me see if I can just briefly do the slide show to do the build and then come back. Oops...

Basically, what this slide is that some people may go, well the taxi industry, it was ripe for disruption. It was run by a bunch of troglodytes. And let’s face it. It was a mess. This slide really builds and says it’s across every segment. So, if you were Hilton you’d like, ‘My main competition is Marriott or Hyatt.’ And now with Airbnb, you’re competing with every spare bedroom on the planet. Now, that doesn’t mean that every person that used to stay in a hotel is going to use Airbnb. There’s clearly some use cases that Airbnb can’t address. But for other use cases its actually more satisfactory.

I went to Madrid last summer. And instead of staying in a hotel, which probably would have been okay, I got a place through Airbnb. And actually, it was much more satisfactory. It was more like a small apartment, so I could bring food and have meals there. I had more flexibility. I could come and go as I want. And it was a block from the Plaza Major, which is kind of the main tourist area, the main cultural area, the old town. It was a phenomenal experience. There wasn’t a hotel nearby.

J. Maguire:

So you went right to the traditional industry then.

B. Golden:

Yeah, for that use case. So then that’s in the hotel industry.

Education, you’ve got traditional universities, you’ve got Udacity coming forward. Security, you’ve got DropCam coming up against ADT, so there’s all of these companies that are sort of saying, ‘how do I leverage this third platform and provide a use case that goes up against this?’ And so there’s enormous dislocation facing companies. So this isn’t all about, you know, how does the IT operate in-group work, this is all about what’s going to happen in industries.

J. Maguire:

Well, I know we’re going to get to this, but what sort of tools does a company need if they’re going to become a third platform company? Or actually, to compete with other third platform vendors?

B. Golden:

Well, let’s say that what you really need is that you need to operate IT differently. I mean that’s the fundamentals. IT has traditionally been a very slow moving organization, in part because that was the only way that you could do it. If it took you weeks and weeks and weeks to get a server installed, and then you had to go through a whole manual process and that took weeks to months, IT didn’t have to move very fast. It already couldn’t.

Now with this cloud computing and resources on demand, all of a sudden the barriers to moving faster are reduced, which means all of the other processes in IT have to improve. Plus, IT is now being looked at as, you know, you’re the way I’m going to compete. It used to be you’re a support organization. You’re going to help me process the invoices. But all of my business activity is really in the analog world, right? I send sales people out. Or I deliver a physical product. Or I might even run a report internally but then my sales person takes it out and shares the paper with people. Now all of that is changing and IT is now kind of really on the front lines of how companies do business.

J. Maguire:

But does this also mean that IT will be increasingly outsourced? You know, the cloud model is where it’s always going to be on demand, but we’re not going to own it as we did the data center. Are companies going to have their own data centers five years from now, ten years from now?

B. Golden:

Well, I think what you’re pointing to is that that functional IT used to be a very integrated hole, right? We owned the assets. We operate the assets. We build the applications and so forth.

I think what we’re likely to see in all this is that IT is going to be, in effect, delaminated. And it’s going to have to figure out what parts of that value chain does it deliver value, and which parts don’t deliver value.

So it used to be a benefit to owning data centers and all that, and I think that’s likely to be much less so going forward. On the other hand, a lot of organizations said, you know, our applications; we get no value from them. We’re going to outsource that. We’re going to go to an outsourcing firm and just have them operate it.

And that part’s really changing. We’ve talked to many companies that have now realized that they need to insource their applications. So there’s going to be a big change, I think, in the portfolio of the overall IT operation.

J. Maguire:

Why is that moving back in-house?

B. Golden:

Well, let me talk about one of our customers, Capital One, which is a very interesting company, and everybody knows them, right? You know, ‘What’s in your wallet?’ and all that.

About, I think, maybe a decade ago, they went through the thought process a lot of people did, which is IT is a cost-center, what do we need to reduce it, and they outsourced everything. They turned it all over to outsource companies.

And then maybe six or seven years ago they went, wow, this world is really changing, probably with the rise of PayPal and those sorts of things. They went, wow, banking is going to be a very different industry. It’s going to be an IT intensive industry. And we’re basically; we’re dependent on low cost, not very responsive, not very creative things. They said we need to bring this in-house and have this become a core competence.

And the brought in a CIO. They said we need to set up some innovation labs. They brought in somebody from the outside to set up their innovation labs. They brought in a whole new set of skills. They started really experimenting and prototyping. And, because they said, the way people are going to do banking in the future is going to be an IT method, right? It’s going to be from a mobile device. Nobody’s going to a bank branch. Nobody’s even going to go to an ATM. So we’ve got to be, you know… and the very nature of what money is is going to change, right? And we’re really seen this with the ApplePay, and these wallets and Square. All of this stuff is really different. And so they said, we need a whole new range of IT competencies. That’s going to be a core competence. We need to bring that in-house.

J. Maguire:

You’re saying they actually bulked up their in-house IT operations.

B. Golden: For sure on the application side. I think we’re still seeing where it’s going to play out overall in their overall mix of their IT capabilities, but for sure on the application side. They’ve set up three innovation labs. One in San Francisco, one just outside of DC, and one in New York. They’re transferring some of their application development into San Francisco. They’re hiring new kinds of skills. That part is really growing a lot, so I think the nature of IT is going to change, as what it’s being asked to do is going to change. And that is happening really quite dramatically.

J. Maguire:

So, you’re talking about some examples, some tools the companies need to become a third platform player.

B. Golden:

Yeah. Absolutely. Well let me point out what we talk about. What you’re really sort of saying is if you’re going to be a third platform company what are the kinds of things or capabilities you need to have in place. Let’s talk about that.

J. Maguire:

How are you going to get started, in other words?

B. Golden:

Yeah, so I’ve got some further slides here that I’ll bring up.

So the first thing I think is, fundamentally and foundationally, this is about business agility. You know, the danger for IT…

I’ll tell you kind of a funny anecdote. I once talked to a friend of mine that did business strategy for big big companies. And I said how come there isn’t more focus on IT? And he said, you know what we think about IT? We think you show up, spout incomprehensible jargon, and ask for huge washes of money. We hate talking to you.

And really what he was saying was, IT was seen as being very self-involved, you know. We’ve got to get the land for this, and not really connected to the business outcome.

So the first thing is, this third platform stuff, whatever you do has to be connected to business outcomes, what we call business agility. So let me focus on this. And you should be seeing a slide come up now that says ‘Business Agility’. Is that true?

J. Maguire:

I see it.

B. Golden:

Okay, good. So a lot of people say one of the things you have to do is you have to bring applications to market faster, and that is absolutely true. That’s the foundation of all of this, what I would call DevOps movement. And there’s no question you need DevOps, right? You’ve got to be able to get something from the next rev of an application into production in a streamlined fashion.

You can’t have this kind of, well, it’s going to take me six weeks to get a new virtual machine, and then I’m going to have to configure it, and then I’m going to build this new rev, and then I’m going to turn it over to somebody else that has to build up a new environment themselves, and then they have to install it. You know, there are organizations; they couldn’t update an application more than a couple of times a year because of all of the manual steps. So clearly that’s a foundation. So I call that time-to-market acceleration. The whole DevOps, you know, you need a tool chain that supports that.

But really, business agility is much more than just, ‘ I can get stuff to market faster.’ Although that’s crucial, I don’t want to make it sound like that’s not. But you need to support these new kinds of things like what I’d call experimentation and rapid prototyping.

So, you’re Capital One, and you say we’ve got an opportunity to try out a new offering. Let’s just spin it up. Let’s get it out there. Let’s try it out with 5 or 10 thousand users in a controlled experiment fashion and see what kind of feedback we get. And the ones that go fast, you know, adoption and traction, let’s push on those. But the ones that don’t? Let’s let them fade away. So, how can we take advantage of resources, but also get rid of resources quickly. So, experimentation.

You need to be able to do rapid prototyping. And let me give you an example, again, from Capital One. They aren’t our only customer, I don’t want to make it sound like that, but they’re a great, I think, a great exemplar of this third platform transformation. I was talking to one of their folks and they said, you know…

Actually, I’m sorry. I’m wrong. This is another company. Another customer of ours, which unfortunately I cannot name. I apologize for that. But they’re a large wireless company that are well known in the technology industry.

J. Maguire:

They’re a telecom. We know the name.

B. Golden:

Well, I didn’t say they were a telecom, but they are in the wireless industry.

J. Maguire: Okay. All right.

B. Golden:

Anyway, the point is, so you said, you know what’s one of the great things about is, it used to be we would have a product manager, and, you know, obviously this mobile world, which is part of the third platform, is a huge transformation. A huge change. Lots of business opportunities. Lots of changing business relationships. Lots of partnership opportunities.

So he said, you know, we would have a product manager who would come to us, IT, and say we’ve got an opportunity to do something with a partner, and we kind of want to envision what that would look like. We want to put together a prototype of what the application would look like. And he said, you know, we’d sit down with a developer and sort of whiteboard it and the developer would say great, I’ve got an idea. They’d go back and they’d request some resources to work on it. It would take six weeks. They’d have to build up the environment and get it working and all that. That would take a couple weeks. They’d put together the prototype that would take like a day, you know, and then they’d come back two or three months later and say, is this what you had in mind, right? So you can….

J. Maguire:

Far too long, really.

B. Golden:

Yeah. So he said, you know, what’s great about using your product is, today we do that whiteboard session, the developer goes off and he or she is back the next day saying is this what you had in mind.

So this kind of rapid prototyping, rapid iteration. That’s another aspect of this agility. That’s really critical because this is a much faster changing world and you need to be able to respond much more quickly. So, overall DevOps, you know, faster time to market of applications, but also experimentation, rapid prototyping. I just talked about flexible partnering and gave you an example, but, for example, turning back to Uber, they’re starting to put together things like delivering products, and I think they’ve had some discussion lately with Amazon or someone like that. So, you know, how do you hook those kinds of things up? So, flexible partnering. Your systems have to be able to encompass that, develop it quickly, prototype it and support flexible partnering.

And then internet of things or also internet of service. I mean, Uber is not an internet of things, it’s a service. But really it is an exemplar of internet of service, right, because they expose APIs, you could probably start building Uber support or functionality into your product, right? So I’ve got a service, that maybe you order a product? I can deliver it to you in four hours. And I make a call into Uber to come pick the thing up and deliver it to this address. So the internet of things, internet of service support, you know, the ability to expose and support APIs, consume APIs, put them into your applications, all really critical for business agility.

So, this is the first leg of three legs. We said what does it take to be a third platform company? You’ve got to have business agility. So, this is kind of the output, the outcome, what you want to do.

The next thing is, that’s fine, but what about the technical side? What do you need from a technical solution to support that business agility? And here’s the stuff that we would talk about. Polyglot Enablement. It’s a world of, an explosion of languages, and, you know, frankly, an organization that says we’re a Java shop, and that’s what we use, and if you want to come work for us or if you want to have your code integrated with us or whatever, it’s got to be Java. That’s like Henry Ford saying, you know, you can have any color Model-T you want, as long as it’s black. That’s just an outmoded world. So, the ability to use and support multiple languages, languages that are appropriate for the problem set that you’re solving, really critical.

And by the way, what I’d say about our customers, we’ve had customers come to us that say, we’re a Java only shop, we need to get better. Within a year a couple of different ones that we’ve run stats on, they’ve gone from single language shops to being ten different languages or ten different framework that they’re using, that they have production apps running on. So, you know, the power of this polyglot stuff is really pretty amazing.

J. Maguire:

Yeah, I think the point on that is, I mean, the modern IT world is very much a heterogeneous world. I mean, the conference is called InterOp for a reason, because things really do need to interoperate with each other.

B. Golden:

Yeah. So you need DevOps. I’ve talked about that. You need automation, right? You’ve got to get manual steps out of the process and you’ve got to use common artifacts across all that. So whatever a developer creates, that moves through the entire thing. We happen to use Docker. That’s on the next slide. I’ll talk about that.

But, you know, DevOps Automation is really about getting rid of manual steps and using common artifacts, and you need to have automation. If you don’t have that, you don’t have a third platform company. You’ve got a company that’s going to be disrupted by a third platform company.

You need API support, and that’s both consuming and exposing APIs, because that enables integration with other kinds of applications. That enables that partner, flexible partnerships and all that. API support is really really important.

Mircoservices architecture really supports the rapid iteration, rapid prototyping and so forth. The monolithic application architectures is really breaking down in the face of the kind of new world. So you need to be able to support creating, managing, marketing microservices architectures. Super important.

J. Maguire:

Actually, Bernard, can we stop for just a second. You know, the term microservices, I think in terms of my buzzword watch, that particular term is going straight upward in the buzzword cycle. What exactly is a microservice?

B. Golden:

Well, simply put, a microservice tends to be a much smaller unit of functionality being exposed. So, a monolithic service is sort of all of the functionality of our application gets bundled into one executable.

J. Maguire:

Like the totality of Uber, for example, is the full service.

B. Golden:

Yeah, and that would be put into one big application deliverable, right. So kind of like a websphere or weblogic application, typically you bundle all of the code into a single executable. And the shortcomings of that are that, one, every time anybody wants to change anything about the application, you’ve got to reintegrate all of this stuff to make sure it’s still operates properly, right. You put in a change. Does it still work with everybody else’s code? So, no matter how small the change, you’ve basically got a big integration overhead. So that makes it difficult.

But also, it’s very difficult to manage that, because, you know, I want to scale my application. Well, I’ve got to scale all of that. Even if only one little part of it’s being… You know, maybe if people are just checking to see what the availability of an Uber taxi is, they’re going to go through all the code. All the code’s going to be exercised. So it’s very difficult. It’s also quite difficult to update the individual services, because again, you’ve got to do all this integration.

So, microservices says, instead of doing that, what I’m going to do is I’m going to break all this functionality; I’m going to decompose this large monolith into many many collaborating services. And that way I can update the display, what the taxi availability is service, without affecting everything else. And I only have to worry about that one operating properly.

J. Maguire:

So, it’s much more agile, then.

B. Golden:

Well, it definitely enables agile stuff. By the way, we have a whole thing on microservices, and if you ever want to do a return engagement of this thing I’d be happy to go through an entire session on microservices. A very interesting trend. Very important. I mean, it’s sort of now maybe being a little bit oversold, or kind of it’s consolatory problem.

J. Maguire:

It’s very hype worthy at this point.

B. Golden:

It brings it’s own challenges, no question. So it’s kind of a two-edged sword. But I would say that to be able to be a third platform company, you’ve got to have that in place. You’re not going to do it with these monolithic things, I believe.

Blue-green deployment is a way of deploying new functionality and making sure that it operates properly. So, I talked about Capital One. They might bleed something out and expose it to 10 or 20 thousand customers. That’s the mode that they do it with, with this kind of blue-green deployment. It’s a way to validate functionality and then gradually grow it and expose it to more and more of your customer base, and do that in a controlled fashion.

The old monolithic thing was much more, sort of, it’s either version 1 or version 2. I can’t say I want to run version 1 for 90% of the users, but I want to expose version 2 to 10% to make sure it works properly, and to make sure they’re happy and so forth. So blue-green deployment is very important.

Application scaling and elasticity: This is absolutely crucial, because guess what? A lot more people are using Uber, you know, right after work, than are using it at 2 in the afternoon. Well, their application has to be able to grow and shrink. So, application scaling and elasticity.

Cloud is typically a component of this, I would say, a very common underpinning for that. So you need to have that.

And I would say, really, this is platform as a service capability. And you and I were talking before we got on about platform as a service. And, you know, there’s people who go, well, you know, I don’t want to PaaS, but I want to have the following 8 things available in my application. And kind of, at the end of the day, what they really describe as platform as a service.

So really, it’s sort of saying I need to have an application framework that supports rapid development, quick deployment, all of this functionality above here, versioning and so forth. You need platform as a service functionality. Maybe you don’t call it PaaS.

We believe, my company, we’re a cloud foundry distribution. We think cloud foundry is going to be the victor in all of the PaaS stuff because it’s got the biggest ecosystem, it’s open source based, and so forth. You know, we would say that’s the right…

J. Maguire:

You’re claiming that cloud foundry is going to overtake, say, the Azure PaaS offering?

B. Golden:

Well, yeah, ultimately. I think Azure, well, .net is a fantastic environment and Microsoft has a huge developer base, but it certainly isn’t the totality of the developers. And there’s an increasing affinity in enterprises for open source based solutions. In fact, I wrote about that in my most recent blog on CIO. And I don’t know if you do show notes or something like that, but, you know, you can find that blog post. Enterprises are increasingly preferring that.

And so I think that cloud foundry is going to have a phenomenal presence long term inside enterprise IT because it provides all this kind of capability of all these bullet points up above that I just went through. And it’s open source, so, you know, there’s less of a proprietary lock-in. It’s more cost effective. So yeah, I think cloud foundry’s got an incredible future. And that’s what we believe, so that’s why we support our product on it. That’s why we released our product on it.

So, let’s talk about the third aspect. So, we just talked about technical innovation, in terms of the way you write, deploy, manage applications and so forth.

The third leg of the stool is infrastructure choice. The reality for enterprises is they run in multiple environments. Back in the day maybe you found people saying I’m an IBM shop. Nobody does that anymore. They’re an everything shop. And so you’ve got to expect your application to be running in many environments. Some on-premise, some in a public cloud provider, and probably across multiple cloud providers and maybe multiple on-premise clouds, right? It’s kind of a hodge podge.

And so you need a third platform solution that can support all those environments.

You know, before we got on you said, you know, what about PaaS? Maybe somebody would just use Amazon. And Amazon’s a fantastic product, no doubt about it. You know, I wrote a whole book about it, so obviously I have enormous respect for it. But no enterprise is going to say every application we have runs on Amazon. So you need a solution that can span all of these environments.

We think containers and, in particular, Docker, is the right foundation for that. It’s an application execution environment that can be supported across all these environments. It’s sort of the lingua franca for execution.

J. Maguire:

Do you think, by the way, that the rocket container technology will give Docker a run for their money? Or will Docker be predominant?

B. Golden:

Well, to me, product victors are based on the size of the ecosystem. And right now Docker has a very large and growing ecosystem. I don’t know if Docker will be the one that will ultimately win, but right now it’s clearly in the lead. That’s why we’ve chosen to use it.

We were the first commercial software solutions shipping with Docker support, and we use Docker, so we think it’s great. And it solves that problem of how do I have a common execution environment across all these different environments, and that’s why we use that. So that’s why we use a Docker foundation. You could say container foundation, and absolutely, that’s the case, it’s got to be container based, I think. Docker, right now, is the right solution for that.

You want it to be language and stack neutral. It can’t dictate, you know, you can use our PaaS as long as you use our lock into this environment. You need to be able to support any language, any stack.

It’s got to be lightweight hybrid cloud. And this is a little bit more complex, technically, but one solution for hybrid cloud is kind of down at the infrastructure layer. And you basically use the APIs of the cloud providers. And you use them to build up their individual virtual machines. And then you load in the code for those things, all in an automated fashion. And you can do that. I mean, I came from a company that did that. The challenge is that’s a lot of heavyweight lifting to really get to what you really want which is application portability. And if you use the container based solutions and a PaaS layer, it provides, what I call, kind of lightweight hybrid cloud because you can basically take the same application.

I mean, for our product, you can take the same thing you developed on a developer’s laptop and deploy it into all of the environments we support. So it’s very lightweight. And you don’t have to do a lot of rework. You don’t have to create new virtual machines to be able to run it on Amazon or VSphere or something like that. So, lightweight hybrid cloud.

J. Maguire:

To clarify on that one, when you talk about the hybrid cloud, I assume you mean the public and private united. But then there some people who say that private is going to fade away as public becomes evermore dominant. Do you buy that one?

B. Golden:

Well, in my opinion, enterprises are always going to have a mix. And so the question really is about what’s the portfolio makeup. You know, what proportion of applications run internally? What proportion run externally? I don’t know what the answer to that is. You know, is it 80/20 one way or the other? Not sure. Is it 60/40?

I think the right answer, if you’re an enterprise, is, you know, I want to be able to make that choice based on what’s right for a given application. And, so you want a lightweight hybrid sort of solution.

And by late binding deployment, what that phrase means is you don’t even really want to be able to have to choose at the day that you start designing an application, oh well, this is going to be an Amazon application. Well, what happens if all of a sudden Amazon starts competing with you and you go, I don’t want to run my business in Amazon Web Services. You don’t want to be in a place where you go, well now I’ve got to rewrite the whole thing. You want to have a solution that says I can design and build my application, test it, have it ready to go, and then decide where I’m going to deploy it. And that’s late binding deployment.

And so I think organizations are going to have a mix of portfolio where they deploy things. It’s unclear today what the proportions of those will be. I’m pretty sure that they will always have both on-premise and off-premise. And you want a PaaS solution that supports all those, and doesn’t restrict you from that.

And then you need a common application design and operations. I already talked about the application design. You know, you want to be able to design at once and have it operate in all the environments where you’re supported. And you want to have a common operational model.

One of the things we didn’t just mention about hybrid is there are applications that will probably run in both, right? And so you want to have a common operations model, because you don’t want to say, well, half my application’s running in Amazon, so all of my logging and all of my monitoring for that part’s in the Amazon wing of my noc. And all of the stuff that runs internally is in the VSphere wing of my noc, and now how do I figure out, how do I cross correlate, how do I define root causes?

J. Maguire:

Well, I think the ideal model is going to be a seamless world between those two. So there’s really not going to be much of a difference, looking way ahead.

B. Golden:

Right. And to do that, you need common application design and operations. And that’s really what you need from an infrastructure choice.

So, sort of summing up, to be a third platform company, you need business agility, you need technical innovation, you need infrastructure choice, and you need some kind of environment that makes that all seamless and fast. Now, obviously we’d love it if you chose Skatako, which is our product, but you need some environment.

And so the choice for enterprises is this is the direction I’ve got to move in. You know, I can’t stick with what worked perfectly well a decade ago for 15 years ago. That’s way outmoded in a third platform world. If I’m going to be competing with Uber or Airbnb or Udacity or whoever it might be, I’ve got to be a lot faster. I need something that walks and talks and, what’s the word about the duck thing?

J. Maguire:

If it walks like a duck and talks like a duck…

B. Golden:

Yeah. It’s a duck. You need something that’s like a PaaS.

I had an interesting conversation with a guy who is pretty well known in this area, a guy named Jeff Susna at Velocity last week. And the said, you know, the thing is, a lot of people go I’ll just start with Docker. I’ll just use Docker containers. And he said, you know, the thing is you start with Docker, and then you say, well, I need monitoring, and you start finding a solution for that. And I need management and I need a solution or that. And I want to do some kind of auto-scaling and I want a solution for that. He says by the time you’re done doing that it kind of looks like a PaaS. So whatever you call it, you need that kind of functionality. And that’s what we’d say.

J. Maguire:

Bernard, that’s a lot of good stuff. I mean, in closing, anything about third platform? What’s it all mean? What are we going to be talking about a few years from now in terms of third platform?

B. Golden:

Well, I think that we are going to see enormous change. I think if you’re an enterprise IT organization you’ve got to… yesterday was too late to start. You’ve got to get going today. This is like existential threat kind of stuff.

And I would encourage anybody who’s interested in this or interested in our product, you just go to ActiveState.com/Stackato. We have free versions of our product that you can download and use. No controls at all. We have what we call a micro-version, which is 4gig. You can put it on a laptop. We also have a 20gig version that you can actually deploy production apps onto, multi-node, cloud structures and so forth. And we would encourage everybody, you know, just experiment.

The thing to do for organizations is get started. These trends are only going to get more and more powerful, and every day you wait is kind of like you’re falling a day behind. So you really need to sort of say what’s my plan going forward? I need to get going.

J. Maguire:

The time is now.

B. Golden:

Yeah, the time is now. Let me stop sharing so I can bring back my picture.

J. Maguire:

There we go, all right. A lot of good stuff. I really appreciate it. I’ll send you the link. We can tweet about it.

B. Golden:

Fantastic.

J. Maguire:

And great. I will talk to you soon. Thank you very much.

B. Golden:

Well, thank you so much for the invitation today. I’ve really enjoyed it. It’s been a fantastic experience.

J. Maguire:

Alrighty, bye bye.

B. Golden:

All right, cheers. Bye now.


Latest Videos



Upcoming Hangouts
See All Videos