I enjoy Twitter (and other feed/social interfaces like HN) but many aspects of these designs are abusive/exploitative of attention, creating a false sense of novelty and urgency.
I expect as this becomes better understood, intelligent summaries and filters, at longer controlled intervals, will become more prevalent. The 'Nuzzel' service, which builds a TechMeme-like summary from your own network, is one step in that direction.
A feed service that's loyal to users/developer, rather than advertisers – like App.Net, even if not App.Net itself – could also be an important part of the future. Why? What's needed is client/UX innovation that actually calms the interactions, via reordering/rearrangement/aggressive-filtering leading to less (but higher-quality) engagement... and Twitter/Facebook have strong incentives in the other direction.
I just started with the assumption that getting real-time quality news is the shining light at the end of the news tunnel.
Which seems like a reasonable assumption considering that every innovation/breakthrough in news/information industry has been heading in that general direction.
There are only 3 contexts by which news is relevant time(no one is interested in what happened years ago), location and category. Blogs have clearly favored category and applications like twitter have greatly reduced the time (although I believe twitter is not streamlined to get the most important news out quickest,different discussion). The only way to further reduce that time seems to be by truly democratizing news as in crowd-sourced news.
Who right now is paying for "real-time quality news"?
I believe future news/social applications will have to harmoniously marry the global with the local. In all humble honesty I sorta envision these really cool applications that'll be aware of where we are and what we want and serve us relevant content.This idea is exciting to me because we don't have to wait for superior technology to build them.