By Eric Van Buskirk
We sometimes forget that at the dawn of Web 2.0, in 1995, the leading Websites in the Social space were about user shared content which were virally pushed to homepages. They used more crude systems than we see with the user interfaces that makes social media sharing easy today. Soon individual trust and authority will be determined by “author rank,” a status calculated by search engine algorithms which look at individuals interactions using social media. Both Google and Microsoft are clearly putting a lot of effort into what is also referred to as “agent trust.” Google’s patent which discusses its method of determining a reputation score for an agent. Microsoft just submitted their patent on the topic: Ranking Authors in Social Media Systems, and their filing indicates a strong emphasis on shares which show the authority of individuals.
The leaders of early socially online content ranking systems, Netscape (Propeller), Digg, and Reddit, were based on users’ approval of content shares. What happened to these sites where people judged who became the most important curators, often with simple sentiment input via clicks? They failed, for the most part.
I have a list of over 100 sites from Mashable and Wikipedia about bookmark sharing and voting which were in the some of the hottest space in 1995-1997. At least 90% of the companies are completely out of business. The biggest problem was the top people found ways to game the system to maintain their high status and visibility. The history of Digg is especially interesting here. In the absence of an editor to keep an eye out, the “most popular” users conspired with each other to keep their top status.
On today’s social media platforms, thought leaders in their fields on sites like Twitter often have an offline reputation which contributes to authority on the networks. This is a big step in the right direction. Furthermore, the sharing that takes place is more feature-rich, and less easy to game with automation software or “conspiring” to artificially climb to a position seen by a user base as thought leadership.
We’ve seen that people are not great at judging the best content curators when an online platform is not robust enough to match the complexities of offline communication, but can computer algorithms 10x more complex than Digg or Propellor’s do better when author rank is rolled out across Google and Bing Search?
