System iNetwork

May 14, 2007

Gettin' SaaSy With It

If you take a peek at IT publications outside of the System i market, one term that seems to be all the rage these days is "Software as a Service," or SaaS. Basically, this is the practice of offering software applications via the Internet instead of licensing them to run on a client enterprise's own machines.

In our market, the concept isn't really new. Back in the 80s and into the 90s there were "Service Bureaus," which offered green-screen apps running on someone else's AS/400 and largely accessed by private connections. Later came Application Service Providers (ASPs), which added the twists of offering client/server apps over public networks, eventually with HTML interfaces. Then came Managed Service Providers (MSPs), which tossed some networking equipment and services into the mix, as well as the ideas of also being a host to multiple web sites or providing an enterprise with general Internet access.

Yes, I realize I'm oversimplifying. The point I'm trying to get to is that conceptually SaaS is largely a recycled idea in the System i market, and one that has always found no more than limited acceptance. The number of successful Service Bureaus, ASPs, and MSPs in the System i world has never been large. Partly, this was due to the expense of connectivity solutions before Internet use became widespread and the inherent insecurity and relative unreliability of network communications in decades past. Even more, I think, was the prevalence of what used to be called the "not invented here" syndrome, the egocentric idea that one's enterprise was too unique to use software that wasn't custom-built to handle those home-grown eccentricities that were each company's "secret sauce."

Nevertheless, SaaS is a term that's spreading throughout the Windows part of the computing world and becoming trendy because Microsoft uses it. But I wonder if, after all the years of knocking around in its SB, ASP, and MSP forms, SaaS isn't an idea whose time has finally come even in the System i market.

SaaS makes some economic sense, for SMBs particularly. After all, it's a way that promises to offer browser-based access to proven applications for System i end users while bypassing many of the IT resource-consuming activities that are often hard for SMBs to fund: guaranteeing service levels to the users, managing backup windows and media for applications and data, introducing software upgrades without causing at least minor trauma, and in some cases, even making custom modifications to apps. And with SSL/TLS and other forms of encrypting confidential business data moving across the Internet in place and performing reliably, and the ubiquitousness of ISPs offering reasonably priced access to the web, the instability, insecurity, and expense that hampered the ASP business model in the past is disappearing. It's still not perfect, of course, but so much better than it was. And with the system management tools becoming available, it's more economical for a company that offers application software to start thinking, "hey, we could handle keeping separate from each other the transactions of multiple client businesses running our software on our machines."

I think there's a new psychology at work among users as well. Even many SMBs are feeling the heat of needing 24/7 uptime to support computer-based consumers who don't follow the mindset of only doing business doing "business hours," or to support potential markets in more than just the U.S. time zones. The idea that only custom software can support "business done our way" is starting to give way to the thought that, particularly at small companies: "well, if the software companies successfully selling these apps seem to expect us to run our inventory system a particular way [for example], maybe we should be changing our inventory system to conform to the software. How hard can that be?" Granted, larger companies can't do this without considerable pain. But small, nimble companies can take advantage of such expediencies when they don't have the resources to develop custom alternatives on their own. And in an era of increasing business consolidation that often requires merged companies to conform their business practices to an application software standard that may be unfamiliar, even large companies are finding it necessary to change their internal methods, sometimes drastically, to fit software rather than the other way around.

In the past few months there have been some signs of a move in a SaaS direction by some companies that include System i support in their repertoire. In February, Voltage Security offered Voltage Security Network, a managed encryption service for e-mail messaging that includes Domino servers like the System i. In March, Esker unveiled Esker Fax Services for SAP applications, a SaaS offering that lets users send faxes directly from their SAP applications, and Lawson Software announced Lawson Total Care Platinum, a managed hosting solution for its ERP applications that lays the groundwork for those apps to be offered on a SaaS basis across all platforms Lawson supports, including System i. Particularly in the case of Lawson, this is a move by an application software heavyweight that could offer it some traction with smaller companies. Little guys that may not be able to afford to purchase Lawson software licenses for use on their own machines might be able to afford SaaS services for the same products if backups, service-level guarantees, and help desk support were part of the deal.

Do you think SaaS is an idea that makes sense for the System i market? If you're an application software vendor, are you thinking about offering your products on a SaaS basis to smaller companies than those that are currently the bulk of your user base? If you're at an SMB user company, would apps on a SaaS plan be affordable even when an outright software purchase might not be?

Posted by at May 14, 2007 2:53 PM