This year was a big one for search as an interface. Apple announced that Spotlight would provide system wide search capabilities in OSX Tiger the same year that Microsoft announced it would have to delay that same functionality in Longhorn. The recognition of players like AppRocket and X1 rose as a result. Google released their own desktop search tool, followed shortly by the MSN Beta of their desktop search and the announcement that Yahoo release their own based on X1 (which they recently acquired). Not in the news so much was whether search was necessarily a good interaction paradigm for an OS, or whether the single box submit/respond model was the best. There was a little movement on this front, however, with players like Grokker(interactive data visualization) and DogPile (category based results) supplying their variations on the search and results interface.
In other news, with the bubble just a distant memory now, web development and web applications rolled on in 2004. I’ll leave it KC to talk about developments within the blogging world and web publishing generally, but it’s becoming clear that the web is going to be where a lot of significant HCI work wil be done in the next few years. It’s not so much that it is a wonderful platform for interaction design (actually - quite the opposite), but rather the ability to rapidly deploy, test, and iterate on design has made it a nexus for new ideas.