Commentary: Shift happens: Cloud computing (and why you should care)

Listen to this article

While most technologies advance the state of best practices incrementally, there are some advances that are genuine game changers.

These changes create a paradigm shift that rattles the accepted norms of how computing systems are created and deployed. Smart phones, 3G and cheap laptops have all changed the way we work each day. Now cloud computing seems poised to have a seismic impact on the perceptions and expectations of any individual or business using any sort of software.

While the entire scope of what comprises “cloud computing” is still a perfectly good subject for debate, there are some workable definitions that give it some shape. The National Institute of Standards and Technology defines cloud computing technologies as those sharing the common characteristics of (a) being on-demand services that (b) are available from anywhere, (c) achieve economies of scale through shared resource pools, (d) are scalable on demand according to the needs of the customer, and (e) are metered or subscription based services.

The cloud technologies break down into three primary delivery models: Software-as-a-Service (SaaS), Platform-as-a-Service (PaaS) and Infrastructure-as-a-Service (IaaS).
Software-as-a-Service

SaaS applications form the “visible” outer edges of cloud computing. You’ve already used SaaS applications if you’ve used Google Docs or Gmail. Unlike traditional desktop applications that are installed on your computer, SaaS applications are delivered through a web browser and don’t have to be installed to use them. It doesn’t really matter where the web front-end and the supporting back-end services are, as long as they are available any time, from anywhere.

There are excellent examples of anywhere/anytime SaaS products. Salesforce.com is a very successful SaaS-based customer relationship management service. Salesforce’s popularity has been driven by its turn-key ease of use, quick implementation time and reasonable fee structures. Those factors have, in turn, been driven by the economies of scale that Salesforce has achieved with their multi-tenant architecture.

In a way, multi-tenant software is kind of like a multi-tenant building. The infrastructure is already in place, the rooms within a unit are already framed and each unit is plumbed and wired with necessary utilities. All of the hard work is already done. Instead of setting up a new server and installing new software for each new “tenant,” that tenant simply uses its own data with the servers and software shared by the other tenants. The new tenant simply has to “move in” and make minor modifications to suit its particular needs. Salesforce.com’s offering, for example, can be tailored and customized as needed by each customer, often without any sort of additional programming.

Platform-as-a-Service

Platform-as-a-Service (PaaS) provides the fundamental building blocks upon which software developers can build SaaS applications. The PaaS infrastructure provides mechanisms for security, storage, user administration and so on. The developers build new applications on top of that infrastructure and deploy those new applications to the servers hosting the PaaS environment. Salesforce.com’s Force platform and Microsoft’s Azure are excellent examples of PaaS.

If you are not a software developer, PaaS may not strike you as being very interesting until you consider that the availability of PaaS environments makes the prospect of specialized, highly tailored applications that are available anytime and anywhere not only feasible but affordable. Vendors of PaaS environments posit that because so much of the application support is already in place, developers can build, test and deploy custom applications in only twenty percent of the time that traditional applications might take.

Infrastructure-as-a-Service

Infrastructure-as-a-Service essentially virtualizes the server rooms of a business and outsources the physical server infrastructure to an outside vendor. As with the SaaS and PaaS models, IaaS customers can scale their computing needs up or down on demand as their needs dictate without have to build out new space for servers or install additional air conditioning.

Vendors and buyers of legal software have their eyes on the cloud computing space, and with good reason. For vendors, SaaS applications are an easier sale with a lower barrier to entry than back-office based systems. The metering/subscription revenue model provides a continuous, predictable revenue stream, and satisfied accounts are essentially an annuity. For customers, SaaS lowers the cost of entry and implementation, and since the maintenance and upgrades are handled by the vendor, on-site support costs are lower.

There are caveats, however. Because the computing environments in the cloud are shared with other tenants, believing that only you have access to your data requires a high degree of trust with your vendor. Your data can be scattered across servers literally around the globe (which in itself raises some interesting e-discovery questions). Integrating with back-office systems can be challenging as well. If the SaaS vendor provides PaaS-like services, you will likely need custom development work done to integrate with other applications. Without that service layer, integrating with other applications might be limited to importing and exporting data.

Cloud computing will change the popular conceptions of what software is, how we buy it and how we use it. It is not a matter of if, but rather when, cloud computing will change the way you work.

If you are interested in the legal ramifications of adopting cloud computing, here are some links of note:

Nerino Petro, Legal Implications of Cloud Computing, COMPUJURIST, Oct. 20, 2009,
Link

Tanya Forsheit, Legal Implications of Cloud Computing Part Two, LLRX.COM, Oct. 17, 2009,
Link

David Navetta, Legal Implications of Cloud Computing Part One, LLRX.COM, Sept. 12, 2009,
Link

Peter Mell and Tim Grance, The NIST Definition of Cloud Computing, NIST SYSTEMS & EMERGING TECHNOLOGIES SECURITY RESEARCH, Oct. 7, 2009,
Link

Ron Phillips is a self-described attorney-computer nerd with over fifteen years of experience as a software architect and technology entrepreneur. He has helped to design and build enterprise systems for large and mid-size corporations, developed commercial software products and authored several books and articles concerning software development, applications and technology. He enjoys helping fellow attorneys with their technology questions one-on-one and on the Practice 411 forum, and looks forward to sharing his technology perspectives in this column. You can reach Ron at [email protected].