Julian Wraith has started a discussion about the future of content management. There are a variety of responses to this linked to from the comments section, each with their own focus, but I recommend reading Laurence Hart for a longer-term view.

My own, brief take is that content management has to face a number of challenging questions over the next couple of years.

Will content need to be managed?
Content management currently focuses on providing tools for groups to create, review and retrieve content so that an approved version of that content can be made available to predefined audiences. User-generated content and the broadcast models of social networking challenge that focus.

1. Anyone can view content: most tweets go to everyone rather than direct to individuals.
2. Anyone can contribute content in a UGC world.
3. Distinguishing what’s your organisation’s content and what’s individual is becoming increasingly fraught; just take a look at any blogger’s site for disclaimers even though they’re blogging about their company’s services.

Will content need context?
Even in the least structured repositories (wikis, flickr, twitter) content is still tagged so that it can be retrieved. But the onus is on the user to find the right tag and on a search application to enable this. This is quite different from a CMS, where the software provides contextual models like folders and related documents to guide the user through an information architecture. As search interfaces and technology improves, there will be less need to provide those contextual models. I have my doubts that semantic mark-up will help people create more relevant content, but I do think that improvements to search will mean that content will be “find-able” and “relate-able” anywhere, even if it isn’t in the right taxonomical folder.

Will content need to be deleted?
As volumes of content continues to increase and contextualisation decreases, finding relevant content amid all the dross will become harder. I think that this will be an even bigger business driver than cost of storage for deleting content that’s irrelevant. But because distinguishing “approved” and strategic content will be harder, it will also be hard to identify which content is dross and what might be useful. Socially-driven records management is bound to take a stab at this problem, but whichever content management tool can help people to get rid of useless content is going to be a winner in the long term.