Relevance or Timeliness? The Real-Time Search Dilemma
What is your priority when you are searching for information on the web? Quality or authority? Relevance or timeliness? Or all of the above?
At any moment, thousands of messages pass through social networks. Every second, new articles are produced. When you do an internet search, you want real time, updated content that takes into account all the recent developments
Until a few years ago, internet search was carried out by visiting a page, clicking on a link - usually a newspaper or a content aggregator website - and then visiting another page. Before crawling became the main information aggregation technique, search engines focused on building directories and portal-like spaces, carrying out an assessment of quality in the process.
Launched in 1994, Yahoo holds the position as the web's oldest "directory" based on a traditional library model, where human editors organized web sites into categories. In 2002, Yahoo made the shift to crawler-based listings.
Altavista started off as a search engine, and at the end of the nineties changed its main business model into that of a portal site. This model, however, didn’t stay relevant for very long and quickly lost ground against powerful real time crawling algorithms like Google’s.
Now search engines are releasing features that provide regular search results combined with a dynamic stream of real-time content from across the web: the online stream of info nuggets - tweets, facebook updates etc - released through social networks. A typical Google or Bing or Yahoo search will therefore be able to include info nuggets straight from social networking websites.
There are already a few search engines exclusively dedicated to real time search and analysis across social medias (try, for example, a search on socialmention.com.)
Real time data can be presumed to be timely, but how do you assess its relevance? The amount of real time data that is available is increasingly growing with the proliferation of mobile applications for real time updating, and accessing data directly from these sources has been compared to drinking directly from a fire hose.Continued on the next page