At the IA Summit this year, Jared Spool's keynote questions the relevance of user-centered design. Putting People First summarizes some key points:

In fact, the UIE researchers found that design teams who tried to adhere to a set methodology and loyally followed a process often struggled and tended to blame the methodology for the failure of the design effort. ‘Finding a new methodology’ was often cited as a possible way of addressing this problem.

Spool placed heavy emphasis on the culture of the firm, suggesting that those firms that celebrated failures were most likely to see real innovation and impactful insights from research, betas and frequent ‘tweak, release & watch’ cycles.

He suggested a preference for ‘informed design’: design informed by a vision, research feedback and tricks & techniques

The link above has additional notes, summaries of responses from the audience at the talk, and Spool's slides from the talk.

It's interesting to watch evolving fields like IA and UxD struggle with complex issues like this: UCD can be useful but, as Spool suggests, it can also be extremely limited when applied dogmatically. And it's often applied just like that: the hammer that makes every design context look like a big, stupid nail. Getting beyond that mindset to something more robust (and, admittedly, more complicated) is necessary.

[via Putting people first]