Now that on-demand computing has been adopted by the major enterprise software vendors, it's no longer a new idea. To be sure, there's plenty of work left to do in making on demand the dominant paradigm for business computing, and that work will continue for a long time. There is also a lot of money to be made by software vendors and others to bring the masses into 21st century computing, but as a cutting edge proposition it's time to look elsewhere. On demand is no longer what's next, it's what's now.

So what is next? Glad you asked.

I'm betting that we're entering the era of the platform, which might look a lot like the mainframe era before we're done. In the mainframe era, organizations more or less committed to computing platforms, as in System/370 architecture, CICS/VM, COBOL and all the acronyms I have conveniently forgotten. The new platform will be much more flexible, and it will not be hardware-bound or even code-bound. As a matter of fact, hardware, operating systems and even procedural coding languages will be largely hidden from the user. The platform is all about the application or the service and the user, not the technology as we are seeing right now from vendors such as salesforce.com, NetSuite and Siebel/Oracle.

The platform is a direct consequence of on-demand computing and effectively extends the benefits of low cost, fast deployment and simpler operation to the developer community. What's cool about the platform is that it consolidates many of the advances of the last few years - such as Web services, XML, SOAP and open source - into an overarching solution with critical mass.

My initial thinking was that individual software companies would mostly participate as developers who made their products available through a limited number of platforms owned and operated by a small group of platform retailers, but now I am not so sure. I already see many companies that can field true platform solutions. In my thinking a platform will need to have three essential elements to assure legitimacy:

  • A hosted delivery mechanism (naturally);
  • A "stack" of conventional enabling services such as operating system and database as well as a development environment for modifying existing applications or creating new applications;
  • A core application base upon which customers can make modifications (using the included tools) to customize the application to their specific needs.

The power of the platform will be in its ability to enable customers to develop solutions around the edges of the core application as well as to develop complete new applications, though it is not mandatory for new development to take place to validate the platform. The hosted delivery mechanism is not simply there as a passive component either.
Because most (but not all) hosted applications operate around the idea of having a single product version in operation at any one time, hosting will enforce synchronization of the customer base, making it easier for third parties, or ecosystem partners, to deliver additional functionality. Synchronization will make it possible for platforms to be employed collaboratively, too. I expect to see verticalized platforms in the not too distant future, not for markets such as pharmaceuticals, but more for things like business intelligence, demand generation, selling or customer communications (formerly known as the call center).

If the tools are any good, customers will buy into the building blocks and customize them at will for their unique business processes. Look for vendors to place a good deal of emphasis on their tools while partners sell domain expertise.

For a long time we've talked about and dealt with the issue of islands of information which was the natural outcome of buying separate applications for different parts of business processes. No vendor could support all of a company's application needs, and integrating multiple solutions was as close as we got to integrated processes. More than that, IT departments spent much of their time managing the integrations or trying to keep the resultant products operational.

In the little time left over, in-house IT tried to satisfy the unique requests of departments. When that was insufficient, the departments turned to capturing data in spreadsheets or developing small applications using separate tools and databases that frequently did not connect to anything else. The promise of platforms is standardization, which will enable an enterprise to have a single toolset for everything. If that happens, and I think it will, it could lead to a situation in which IT gets to do more for its customer because it is doing less for all of its vendors.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access