You might like the flexibility of Amazon Web Services or Rackspace, but not quite trust the cloud to keep your critical applications safe, available, and responsive. What’s a paranoid IT manager to do?
If you’re like many technology managers at big companies, you’ll take a look at hybrid cloud technologies that let you mix public cloud services with private clouds and even old-fashioned, on-site technologies in your own data centers.
Mike Maples, Jr. (@m2jr), is the founder and managing partner of Floodgate Fund, a small but fast-rising player in early-stage venture investing. With a background at broadband software startup Motive (which he cofounded in 1997) and enterprise software vendor Tivoli, he’s also got a keen perspective on enterprise technology. Floodgate’s enterprise portfolio includes Egnyte, DoubleDutch, MetaMarkets, SnapLogic, and Socialware.
We caught up with Maples recently to talk about how hybrid cloud solutions are helping ease the transition to the cloud for many big companies.
Want to learn even more about these topics? Come to CloudBeat, Sept. 9-Sept. 10 in San Francisco.
VentureBeat: It seems as though enterprise IT managers have become a lot more comfortable with the idea of the cloud in the past year. Where are we in the adoption cycle for cloud technologies?
Mike Maples: On one hand, it seems like the cloud is ready for prime time. But there’s another way to look at this.
You’ve got a lot of people in IT who would say that they’re never going to use the cloud. I think there was some study that came out recently that said something like 61 percent of files will never go to the cloud, because of security concerns. [A recent Microsoft survey of small and medium-sized businesses (SMBs) found that 60 percent cited security concerns as their reason for avoiding the cloud. -Ed.] There are other reasons as well, having to do with bandwidth and large file sizes and things like that.
So on the one hand, you’ve got very successful cloud companies like Salesforce and Workday. But if you step back further, you can say, “Well, EMC is twice as valuable as Salesforce.” And they just do storage, right? IBM’s worth more than $200 billion. SAP’s worth $90 billion. Oracle’s worth more than $150 billion.
The question that I find intriguing is: What will be the catalyst that causes cloud adoption to go from “interesting and high growth at the edges of the enterprise” to “mainstream in the enterprise?”
My belief is that it’s going to be something to do with this idea of hybrid infrastructure, or “cloud to ground.” The basic idea is that today, the enterprise IT person has this choice of 100 percent on the ground or 100 percent in the cloud.
Probably what’s going to happen in the future is that IT will say, “Well, my real job is to deliver digital services to the employees and customers and company. It shouldn’t matter where the bits live.”
VentureBeat: What’s making that hybrid approach work?
Maples: It reminds me a little bit of when Bill Clinton ran for president. There was this idea that there was a “Third Way.” It wasn’t all Reagan, 100 percent conservatism, small government, but it wasn’t a bleeding-heart welfare state either. You could be a pragmatic person and care about social good and economics.
I think that right now, it’ll be a combination of startups and big companies that bring about this hybrid infrastructure, this “cloud to ground” set of changes.
Egnyte has always said you can synchronize what lives in the cloud with your computer locally. [Floodgate has invested in Egnyte. -Ed.] The idea is to deliver the files you want on demand and to mask the decision about where those files have to live.
Above: SnapLogic’s architecture diagram.
Image Credit: SnapLogic
Another company that we’ve invested in that has embraced this from a different angle is Snaplogic. Gaurav Dhillon, who was originally the founder of Informatica, started Snaplogic. He has this term he calls “elastic integration.” If you go to their website and look at the front page and what they’re about, it has this diagram that shows how inside the company, you have SAP or Oracle or Microsoft SharePoint. Then, up in the cloud, you’ve got Saleforce and Workday.
The idea is that if you want to integrate across applications, you don’t want integration just among your cloud apps and separate from your on-premise apps. What you want to do is integrate a business process across all of the apps that you use.
VentureBeat: I’ve been hearing about integration in enterprise tech for a decade and a half at least. It’s always been the big challenge. In the pre-Internet days, it was getting your client-server systems to talk to each other. It’s a similar problem today, except with cloud applications.
Maples: That’s pretty insightful. It used to be, like in the ‘90s, applications themselves were stovepipes. Now, I think what’s happened is that what lives in the cloud versus what lives on-premise have become separate stovepipes.
If companies are going to redesign IT and advance it, they’re going to need to adopt cloud infrastructure in a massive way — not just at the edges. Someone is going to have to bridge those stovepipes.
Editor’s note: Integration and hybrid cloud approaches will be a big topic at VentureBeat’s upcoming CloudBeat conference, Sept. 9-Sept. 10 in San Francisco. Register today!
VentureBeat: Is there a particular strategy that these companies are using that makes that bridging possible? Is it about building APIs that talk to other applications’ APIs? Is it about syncing data? Is it some combination, or something else?
Maples: I think that you have to get two things right. One is, you have to have an approach that the incumbents, the IT infrastructure vendors, like. So for example, NetApp should think that you’re not trying to disrupt them somehow. Or EMC or any other big IT infrastructure vendor.
The second thing is that you have to have the type of infrastructure that masks the complexity of where the files live. As you point out, synchronization is an important part of that. But you also have to manage access control in a more sophisticated, fine-grained way than, say, Dropbox would. That becomes important as well.
The ultimate goal is to say: Whatever files or information you have an entitlement to, the goal should be to allow you to access it and not care where it is.
VentureBeat: Do you have advice for the IT guys who are considering implementing these cloud technologies, whether hybrid or not?
Maples: The problem the cloud guys have now is that they’ll go to an IT person and they’ll say, “Do this in the cloud.” Well, one of two things will happen. They’ll either say, “Do this in the cloud,” or they’ll go around them.
VentureBeat: Right. They’ll go right to the corporate officers.
Maples: Sometimes that’s the appropriate course of action. But the idea that IT is never going to participate in implementing cloud-based infrastructure, to me, just sounds wrong.
IT is a huge constituency in any corporation. They play a huge role in driving the competitiveness and the technological ability to compete of any company.
VentureBeat: But the role of IT is different, isn’t it? Last year, you and I were on a panel together, where you talked about how the sales process for enterprise software is different now. You’re no longer trying to get the IT guy in a “headlock” and make him commit.
Maples: I think that is true. In the late ‘90s, there were some very heavy-handed tactics. Customers were not really allowed to try products before they were put into production. Professional services costs were really high, to make these things work and to implement these products. Only a fraction of customers even went live with a lot of products. So customers got really tired of being oversold and strong-armed.
Then what happened is that the vendors who started to win began to borrow a page from consumer companies. They began to realize that the customer needs to be able to try the product. They need to get value from the product quickly, all on their own.
I think that is a fairly durable change. The decade of the 2000s shifted power from the vendors to the customers. I think that’s still true.
VentureBeat: One of the things that we’ve heard recently, though, is about the increasing importance of the role of the chief marketing officer (CMO) and how marketing guys and business guys are driving a lot of IT decisions. Do you think that’s a real trend?
Maples: We’re investing in that game as well. For example, we’re working with a company called DoubleDutch, and Genwi. We definitely believe that the CMO will have more of a digital spend. No doubt.
Marc Andreessen’s theme of software eating the world is right. Every individual who is part of a company is now finding that their function is becoming more and more digital. If you’re the chief marketing officer of a company and you think about what you spend money on, you spend money on ads, you spend money on campaigns and marketing programs and events. All of those things are becoming digital. That’s all happening independent of IT. I think all departments of a company now have a digital strategy.
IT still has a really critical role to play as a service provider for the enterprise. They have to keep the servers serving and the storage storing and all of the infrastructure — the pipes and plumbing — working all the time.
The discussion about how cloud impacts the department has more to do with how the department can go digital without having to wait for IT’s blessing.
But as far as how cloud plays a role in IT, I believe it will play a role in IT by allowing it to implement a hybrid infrastructure.
VentureBeat: There’s a term that you’ve used in the past: the hypernet and the hyperweb. What are your thoughts on that today? Is that still part of your investment thesis and the way you look at the world?
Maples: Yeah, for sure. Roger McNamee and I started riffing on it in late 2010 and early 2011. We had this palpable feeling that the iPhone and the iPad and the mainstreaming of social networking were causing a whole bunch of new forces to gather. We also noticed that in 2011, it was going to be the first year that Windows-powered computers would be less than half of Internet-access devices. They had been like 95 percent in 2005.
Our big conclusion was that 2011 was a turning point in the computer industry because it was the first year where it quit being primarily about computers.
In the future, you were going to have billions of nodes connected to millions of clouds, and computers were no longer going to be the center of it. They were going to fade into the background in terms of mobile devices and an Internet of things.
That led us to this idea of the hypernet and the hyperweb. The hypernet is the new architecture that results from combining the traditional internet with cellular and Wi-Fi and other future standards that allow digital technologies to communicate and interact with each other.
Then the hyperweb is more about: What are all the new user experiences that result from these billions of nodes and millions of clouds?
The mobile phone and the touch interface is an early indicator of that, but not exactly. Mobile phones still feel a little bit like computers. They still have interfaces that look like computers and things like that.
But self-driving cars, for example, would be a farther-out idea. That is, the car is basically gathering data from the cloud, simultaneously munching the data in certain ways, and then using it to direct the car or tell the driver where to go.
And so more and more we think we’re going to see interfaces that don’t resemble the computer interface. The Nest thermostat would be another example. One company that Roger and I are involve with as investors is Sonos, the guys that make the home stereos.
We believe that, over time, the fact that it’s computerized is going to become less central. Everything is going to have computers in it. You’ll have all these different types of new natural interfaces.
VentureBeat: It seems like there’s a huge role for cloud services to play as a back end to all of that.
Maples: Yeah, no question. One way I like to look at cloud services in the future is that if you look at the original HTML 4 web, what you basically had was a bunch of hyperlinked web pages.
Now, we’re going to go up a level of abstraction, from web pages to the true realization of web services. You’ll have all these cloud services — everything from Dropbox to Egnyte to Evernote. All of these services are going to be programmable and accessible via APIs.
Rather than just focusing on linking between pages, you’re now going to be able to create experiences that link between these cloud services. For example, one of the reasons Floodgate invested in If This Then That is that we thought that the capability create APIs and programmable links between all these services is going to be extremely relevant.
VentureBeat: What’s the takeaway, then?
Maples: The high-order bit for me is that clouds will continue to proliferate into corporate departments as software eats the world. The cloud will be the vehicle for how it happens, the cloud and mobile.
And then IT is going to have a different frame of reference for the cloud and where they’re going. They’ll look at it as part of the infrastructure they need to manage, their core mission.
The hypernet and hyperweb are going to transform the digital landscape from a computer industry to an industry of billions of digital nodes and millions of clouds. The hyperweb will be all about how those things connect and deliver an experience.