Listen to my conversation with Jeff Kaplan, managing director of strategic consulting firm THINKstrategies, and one of the foremost analysts tracking software-as-a-service, on-demand and cloud computing.
In this podcast, learn why it's not enough to simply port an existing software package to the cloud without rearchitecting it, and hear about some of the ways enterprises will deal with hybrid environments that mix on-premise and cloud assets.
Listen to or download the 10:05 minute podcast below:
PW: So Jeff, we've obviously both been tracking SaaS and cloud computing for quite a few years now, and they've come on quite a way, haven't they, since those early days?
JK: They sure have. The world is coming in our direction, for a variety of reasons. Most importantly of course, is the economy.
Well, yeah, absolutely, because I think that makes the economic proposition of SaaS or cloud much more compelling, and certainly makes people look at it much harder than they have in the past. And also, I think we've gone past the days when this was a minority interest. Everyone knows that there's people out there doing this, and therefore you're not sticking your neck out by going for it.
That's absolutely right. It's become mainstream, in part because we're so familiar with these kinds of services in our personal lives, converting that idea into something we feel comfortable in the corporate environment has become a lot easier. But, as you point out, seeing a lot of other companies doing it large, small and mid-sized businesses as well also gives us confidence that this is not a bleeding-edge technology any longer.
So here we are at the beginning of a new year. I thought it would be good to get together and have a look at what we see happening next. One of the things I think we're seeing is that a lot of people now are moving to cloud computing, on-demand computing, and perhaps they're doing it without a full understanding of what's involved. And I worry that maybe sometimes we take a step back, because people are embracing the model but without fully understanding where they're going.
Well, I think there's a lot of truth to that concern, because what often happens and we're already seeing this happen is, when you get a topic like software-as-a-service, or now what people are generally referring to as cloud computing, [being] popularized and a lot of companies beginning to not only investigate it but adopt it. Of course, that attracts a proliferation of players who want to capitalize on that market opportunity and exploit it for their own proprietary purposes. So as a result, we're seeing and you refer to this as "SoSaaS" so well
Yeah, which is that's my, Same old Software (but) as a Service, and I think now we're seeing
we're seeing people, again, taking the same old software with all the faults of the old, on-premise, software model. Someone that I spoke to earlier this year described it or last year rather described it as 'Soviet-era software'. And just by putting it on an Amazon instance or in some cloud platform, they think they've taken it to the cloud, whereas they haven't, they've just taken all these same old problems into the cloud.
Well yeah, and it in a sense, first of all, they've bastardized the terminology. Secondly, they've taken us 'back to the future' of the old ASP model, where all we were doing was moving the cheese. And smelly cheese was being taken from within our own data center to someone else's data center. Now, there's some incremental benefits possibly to that, but there's no way of achieving the overall goals that a software-as-a-service or a true cloud computing approach has to offer and can promise. And those, as you well know, have to do with the scalability of a multi-tenant environment, the ability to roll out innovations across a population of users simultaneously so they don't have to worry about those disrupting their operations on a day-to-day basis.
But what I've always felt is the ultimate goal of all this is, it creates not only a different vendor-customer relationship, but it actually creates a new kind of customer community where the aggregation of the activity data across that community can now be aggregated and analyzed, and redistributed as true benchmark statistics that people can actually utilize to gain better insights into what the real best practices are in those business arenas or business areas. So there's some tremendous opportunities there that are only compromised if we're not really using the best platform, which is a multi-tenant platform.
Yeah, and I think the community angle is often overlooked, but I think it's so important. Because the fact that you're actually in the cloud means that it's much easier for that community to come together to be using the same reference architecture, to be able to exchange ideas, to be able to swap extensions that will be ready to run but don't have to be redesigned or re-architected before they're put on someone else's instance.
All of those factors I think are often not taken into account when people are doing a straight like-for-like comparisons between, 'I've got a piece of software here in my data center' or 'I've got a piece the same piece of software over there in the cloud, and what's the TCO comparison and the feature comparison?' and so on and so forth. The community angle is a whole dimension that you also need to look at, and really I think gives the SaaS or the cloud model a lot of advantages that people are not aware of until they've actually got into it and started using it.
Well obviously, the good news for guys like you and I is we have a lot more opportunity to educate people about these things.
Right. Okay. And yes, absolutely, we're certainly taking all those opportunities that we can, aren't we Jeff? One of the issues that I think, though, is going to become quite a lot more important this coming year is the fact that, as more people put IT into the cloud, a lot of those enterprises are still going to have IT on-premise as well, so they're going to have a hybrid environment. And I think that muddies things even more than in the past. Because it makes it more difficult to understand, well, how do I take advantage of all those cloud-side benefits when I've still got to connect into an on-premise IT infrastructure which isn't multi-tenant, which doesn't have that community angle and which is cut off from all the other aspects of cloud computing?
Well, I absolutely agree that this muddies the waters and certainly confuses things. But unfortunate truth is, that that's the reality. We're only talking about a very small percentage of the overall market opportunity that is in fact a green field, where we have the privilege of being able to go entirely to the cloud with our business operations and IT operations.
So it is incumbent on cloud vendors to help customers understand how to integrate the growing array of resources that are available via the cloud with their legacy systems and software so they can take best advantage of it. And over time, of course, migrate as much as possible to a cloud-oriented infrastructure and set of services. But until that time comes and it's not going to be for a long time before that happens the key to success is, giving customers the best integration, giving the customers the best options, and giving the customers the best assurances that they're going to be able to govern that hybrid environment on a day-to-day basis in a cost effective way to meet their business objectives.
And do you think you said that hybrid is going to remain fairly consistent over the coming period but do you think that people are gradually going to move more and more into the cloud or are they going to stop in this hybrid condition?
Well, I think there will always be a mix. But here's where I know your views and mine may be diverging even more severely. That is, I actually think that the technology and the technological evolution of the cloud as well as the subsets of things like SaaS and maybe even PaaS are evolving so quickly that they can in fact be shrinkwrapped and delivered behind the firewall, in such a way that it doesn't violate either the conceptual framework of a true cloud or SaaS solution, or the operational integrity of that. Because it's being delivered in essence as an appliance which may be locked down in such a way that the customer cannot corrupt it, but they can feel more confident from a security standpoint, that data and that function stays behind the firewall.
I think it's already happening. We're seeing a growing number of SaaS- and cloud-oriented companies with an appliance option available to their customers. There's some major vendors, who will remain nameless for this purpose, some who are doing some experimentation in this area, mostly driven by their major customers. And I think that that is actually going to expand the addressable market opportunity, as well as give customers a greater number of options to take advantage of.
Now, there will be some vendors who, again, bastardize that idea and use it to their own benefit, and in essence really are only talking about a different kind of hosted service. But I do think there's some pioneers in this space who are also delivering something that will be very exciting and liberating in a sense, because it overcomes those inherent obstacles that so many companies feel about security, privacy, reliability or accessibility, and all the rest.
That's interesting Jeff and we'll have to leave that debate for another day. But I think, yes, can we take the cloud, box it up in an appliance, put it on-premise, and still get the same benefits that we get from that application if it's running in a cloud data center? I'm not so sure, Jeff, that that's going to work out the way people want it to. But as I say, it's a debate we'll have to save for another time.