parseerror.com / pizza / doc / brave-news-world.html
Making complex, far-off predictions is a narcissistic and futile exercise, so here goes: Someday soon, the 20th century-style newspaper company as we have known it will be extinct, a historical relic gone the way of the buggy whip manufacturers. But news won't be dead.
Instead, editors will work like DJs or video editors, spinning and remixing bits and pieces of existing content from blogs that cover timely topics of the day. Content aggregator sites such as Drudgereport, Huffington Post and Reddit will be considered the fore-runner for these new papers; but as more and more money is involved larger sites will be orchestrated by professionals.
Most of the top reporters will be professionals, like today; but they will initially compete directly with a large pool of amateurs from around the world, until a proper hierarchy forms which will increase the barrier to entry. One type of story that will not be covered by professionals ever again is the "breaking news" type, as people with cellphone cameras and blogging/tweeting/whatevering on handheld devices will have already recorded it and published it before the pros even know it happened. Professionals will need to stick to predictable news, on which they can offer their expert opinion: politics, sports, the arts, editorials, etc. Reporters will self-publish all their stories, within existing online applications. Given the average consumer's disregard for quality, it is unclear how reporters will be paid... possibly via advertisers on their own self-published sites.
Likewise, circulation will change from being synchronous to asynchronous. There will not be a daily paper per se, but a "current revision" analagous to software development, which is updated continually. Transfer will be electronic; people's reader devices will pull down and cache the latest revs whenever they are in radio range, or on a schedule, or on demand. Some highly-regarded papers may be able to charge small subscription feeds; smaller and/or lower-quality ones will be free and ad-laden. Efficient, aggressive content-based networking algorithms will be implemented to minimize traffic. Content will be signed, and possibly encrypted, but only methods that are breakable by first-world governments will be popular.
Many traditional issues today will still exist in this new model, such as falsified and/or planted information. Optimistically, software may be developed to identify and classify article authors based on previous articles, but like today it is unlikely the average reader will be interested in such things. Furthermore the initial plethora of specialized papers will mean people will choose ones that already cater to their existing beliefs, delivering news that meshes with their world-view. In a system of even moreso subjective news than today, the reader will be less willing to question the source.