As a first step to fulfilling that mission, Google’s founders Larry Page and Sergey Brin developed a new approach to online search that took root in a Stanford University dorm room and quickly spread to information seekers around the globe. Google is now widely recognized as the world’s largest search engine — an easy-to-use free service that usually returns relevant results in a fraction of a second.
Making the information findable is one step, but it’s not enough. At some point scanning in physical data from books, magazines, etc. will have covered most of what is available. The next step is aggregating or centralizing information, sorted by relevance. This is likely when information in data sources is roughly stable and blogs, research, current events, etc. are mostly the new sources.
At first the internet had little relevant information. Then people joined the network and contributed their local information and participated in the greater whole. The initial problem of the web was lack of contributors. The method of finding data was just as successful using URLs or human evaluated directories (Yahoo). Then the next problem became too much information and finding or sorting by relevance. This is Google.
The next problem is organization or aggregation. To go beyond finding one or two or ten sources to finding a one reliable source. I would suggest a combination of computer generated search results of Google with human filtering like the Yahoo directory of 1997.
This then transitions like Yahoo to Google. Imagine a Google result page based on search results, but with a good degree of stability. The page itself contains content, 10 lines instead of 2-3, plus links out to more or the original sources. A sort of computer generated Wikipedia.
Content is generated from analyzing the sentences across many sources and generate a consensus. Also detailing pigeon holes, link outs, and differing opinions. This can be updated from references like research, gossip, or analysts/experts.
New content can still come in from a number of sources like photos from Flickr and video from YouTube. This is where image recognition software could be of use. Evaluating the relevance of the photo to the content. For example, a photo of Scarlet Runner beans is thus marked and there is space for 1-2 photos. The computer can tell the difference and pick a flower, profile, or seed pod image to place on the page.