The notion of building bridges to service providers and managing the interaction will become more commonplace in 2006 as we learn to accept that many services we leverage within an enterprise are services we may not host. The technology exists today. We need to define and refine our approaches now, including architectures, enabling technology, and use of standards. Most enterprises are way behind.
We are moving toward a day when most of our enterprise applications may be delivered as services, and thus provide a more economical way to approach information technology management with businesses going forward. This is also the great equalizer since businesses, both large and small, will have access to the same number and quality of services, much as they do with Web sites today. Shared services will create many opportunities, including better agility and the ability to operate a business with fewer IT resources. In essence, we're moving to Web 2.0 where service delivery over the Internet will be added to information delivery as the key strategic value of the Web to businesses, as well as extending the Web as a true platform.
All you have to do is to look around you. With the advent of SaaS (Software as a Service), guys likes Salesforce.com and NetSuite are cleaning up with soaring subscriptions. Moreover, the Web is getting the right interface with rich client technologies, such as AJAX, emerging to provide a much better dynamic user experience. Let's face it; the Web has grown from a simple information delivery platform to a grouping of many valuable exposed services with rich dynamic user interfaces. It's really the global SOA, and those who learn to leverage it now will be well ahead of those who ignore the trend.
However, in order to make this a reality, we must learn to how to bridge the gaps between our enterprise systems and SOAs and Web service providers that exist across the Internet. Special consideration must be given to connectivity, interoperability, security, and shared processes. Problems are easily solvable with the right technology and approaches, but I would say that most out there who are looking at this new opportunity don't have a clue as to how to make the new and old work and play well together.
The journey has just begun down this road, as we're looking to join our existing enterprise system with Web 2.0, and figure out how all of this works and plays well with the notion of an SOA moving toward a global SOA. Clearly, there are a few technical issues that you must address, such as:
- Semantic and metadata management, or, the management of the different information representations among the external services and internal systems
- Transformation and routing, or, accounting for those data differences during run time
- Governance across all systems, meaning not giving up the notion of security when extending your SOA to the global SOA
- Discovery and service management, meaning how to find and leverage services inside or outside of your enterprise, and how to keep track of those services through their maturation
- Information consumption, processing, and delivery, or, how to effectively move information to and from all interested systems
- Connectivity and adapter management, or, how to externalize and internalize information and services from very old and proprietary systems
- Process orchestration and service, and process abstraction, or, the ability to abstract the services and information flows into bound processes, thus creating a solution
It doesn't take a rocket scientist to figure out that the creation of an SOA on top of these applications, including process/orchestration layers, directory services layers, identity management, monitoring, semantic management, etc., would add a tremendous amount of value, considering the use of those applications and abstraction into real business solutions. Indeed, I find that many SOAs for many businesses actually exist outside of their firewalls, making their on-demand applications work well together. This trend is only accelerating as Web 2.0 becomes more valuable for enterprises.
What's cool about this is that businesses will have to change to remain competitive. As others learn to embrace Web 2.0 within their enterprise, as with the Web of 10 years ago, others in their community will have to do so just to keep up. There are many examples of this today, albeit it's still early in the cycle. Indeed, many small businesses might find that the majority of enterprise processing occurs by leveraging outside services - services they don't own and haven't created. Is that scary, or exciting? I think it's exciting!