Many of John “Searchblog” Battelle’s predictions for 2005 seem like near certainties, e.g., a fractious year for the blogosphere and trouble for those who expect major revenues from blogging.
Two trends I hope 2005 proves that Battelle’s predictions missed:
Metadata-enhanced search. Will be ad hoc and pragmatic, pulling useful bits from private sources and people following officious Semantic Web and lowercase semantic web practices.
Proliferation of niche web scale search engines. Anyone can be a small-scale google, crawling the portions of the web cared about and offering search options specific to a niche. The requisite hardware and bandwidth are supercheap and the Nutch open source search engine makes implementation trivial.
The Creative Commons search engine is a harbinger of both trends.
Battelle’s look ahead spans the web, not just web search. Possibly the biggest trend missing from his list is the rise of weblications. Egads, I have to learn DHTML, and it isn’t 1997!
A few of my near certainties: lots of desktop search innovation, very slow progress on making multimedia work with the web and usable security, open source slogs toward world domination, and most things get cheaper and more efficient.