Web 2.0 is going to catch a lot of IT organisations ill-prepared while they are looking in the wrong direction.
Web 2.0 is an umbrella term that encompasses a number of consumer-driven technologies, including collaborative tagging (or 'folksonomy'), virtual worlds, mashups, wikis, blogs, and more.
What they all have in common is the notion of electronic collaboration that is not dependent on mainstream IT. In fact, the uptake of Web 2.0 is showing a lot of the same symptoms as the initial adoption of the worldwide Web.
It is making the transition from being a toy for geeks into a useful business tool. People seeking to make a quick buck are 'cyber-squatting' on brand names in virtual worlds just as they did on domain names in the www, and the whole environment is journeying from recreational to commercial.
The immersive experience of virtual worlds is leaving the users of business IT wondering why they put up with clunky interfaces and out-of-date information.
Of most immediate concern to IT, mashups (and consumer-friendly mashup construction tools) are creating a way for users to build their own applications that bring together realtime data feeds, geographic data, locally-stored information, and - worryingly - corporate applications and corporate data.
There are two possible responses to this trend: suppress it or encourage it.
Suppression via a combination of security measures and corporate dictates is a possible short-term option, but is not likely to be viable in the long-term. Just as commercial realities made it necessary to provide web-based access to corporate front-office applications, over a short period, there will be growing business pressure to provide access to corporate applications from Web 2.0 tools, both for the organisation's own staff and for customers and partners.
It would be a great competitive advantage to have your customers build mashups around your product catalogue and sales order applications, together with their own purchase order application. In fact, it could become a corporate strategy to assist those outside the organisation to build just such an application.
The problem, of course, lies in managing the level of access that is provided to Web 2.0 consumers. The nature of the IT legacy is incredibly complex, with multiple applications and data sources residing in multiple physical locations on multiple platforms, using a multiplicity of architectures and technologies.
Equally, we must expect the world of Web 2.0 technologies and their application to continue to evolve for several years, and for users to grow in sophistication. The demand and usage profile is therefore going to evolve, and will be quite impossible to predict with any level of certainty.
The combination of these two elements - the server-side complexity and the client-side expectations - results in a management nightmare in the middle. It will be increasingly difficult to map user requirements on to the underlying applications, and to maintain security of access to those systems that need to remain private.
The solution is to provide an abstraction layer. Direct access to the applications should be blocked without exception. What users should be encouraged to exploit is an abstracted layer of virtual services. Service oriented architecture provides the ability to create this service layer, and SOA governance and run-time management tools provide the ability to define corporate policy regarding access to these services and to enforce it across all access points.
This serves two purposes. For IT, it makes the management tolerable, and provides the ability to throttle the usage of low-importance users to ensure critical transactions get the service levels needed.
For the consumers it provides simpler integration to more meaningful and useful information and functionality.
For most organisations there is still time to sort out the strategy for Web 2.0, but as always, the strategy is more likely to be effective if it is determined before the demand peaks, and before management becomes a tail-chasing exercise of closing stable doors after the horses have escaped. It should be a major consideration, and an important component of the enterprise architecture.