Thought leader

Jan Wyllie asks if the 'follow the leader' trend is sending Web 2.0 down a blind alley

In the early days of Web 2.0, when the internet first became a publishing medium open to the public, people tagged their material with their own descriptive words. They discovered that this was a pain because people used different words to label similar items, or forgot which words they had used, so they could not use them effectively as search terms.

An attempt was made to address both these problems with folksonomies, where people were informally encouraged to reuse and share tags from common alphabetically sorted lists. As the lists lengthened, another pain threshold was reached, and folksonomies faded into the background of academic discourse.

Now, in the new era of social media, massive information flows are no longer organised in terms of ‘about-ness’, but in terms of ‘who-ness’. It’s not what you know, but who you know. This trend started on social networking sites, with the concepts of friends and feeds.

Twitter embodies the latest form of knowledge sharing, where ‘leaders’ win ‘followers’ for their 140 character tweets. Retweets are second order recommendations that ‘leaders’ make to their ‘followers’. The great thing about it is that it requires no effort or thought on behalf of users, other than to hit the retweet button. The idea is that somehow through the ‘wisdom of crowds’ this system of recommendations delivers the best knowledge, fastest, to the people who want or need to know. Leaving aside the question of what useful knowledge it is possible to glean from ‘following’ thousands of ‘leaders’ who tweet many times a day, my experience suggests that without some form of content organisation, follow the leader actually gets in the way when connecting people to the best information flows for their purposes.

One of the cultural dynamics that reduces the value of information streams based on follow the leader is that many people follow others as a kind of thank-you for following them. Many others follow people principally to boost their own follower numbers. So, who is following who is a very inefficient way of finding anything out in advance about content or its quality. Then, there’s the difficulty that people quite naturally seem reluctant to recommend unpleasant news to their followers, which creates even more distortion in feeds, potentially eliminating the most important items.

The other problem with managing knowledge in this way is that most people are interested in a wide variety of topics. What doesn’t happen to overlap adds to the overheads of useless hits, which waste time and electricity.

However, the most serious drawback of the follow the leader paradigm may be the tendency for early members of a group who, as a consequence, have the greatest number of followers, to more or less dominate the proceedings thereafter. Then, of course, there is the massive distortion caused by the celebrity factor in early 21st century culture.

The difficulty is not about recommendations being a measure of value. After all, that is what Eugene Garfield’s Science Citation Index is all about – the more an author is cited, the more he is deemed to have contributed to the subject in question. What is lacking in the follow the leader paradigm is context, in order to give at least some idea of where people are going with social media contributions. Some kind of directory of interests based on questions in user profiles would be a start.

Directories of interests would be an improvement, but there is no escaping the fact that in order to make sense of a body of material, there must be initial (and subsequent) pointers to what is there, beyond shooting words into the dark with Google-type search. Librarians and booksellers have known this secret for thousands of years. Now, the next generation of Web 2.0 social media enthusiasts, which seems to have forgotten them, needs to learn them again. (Alas, ‘twas ever so.)

Web 2.0 and social media need to develop common knowledge frameworks to provide the context necessary to make sense of the growing torrent of information streams. It is the next big job to do: turn information chaos into collaborative intelligence. The technology of the semantic web might be useful for implementing such frameworks, but only humans can make them up because they depend on purposes and feelings, as well as creative skills.

And surely the goal is to amplify networked human intelligence, not try to substitute it with semantic rules and algorithms, not least because there is so much unused human intelligence lying about the place.

Nobody said it would be easy, but it is possible and it is worthwhile.