A Social Networking Examination of Tagging On Gather

 ----------------------

In a previous article, I contemplated the cognitive science aspect of tagging - describing how people tag, and why they might find it easy. There is another, equally important aspect of tagging that I did not touch upon - the "why" of tagging. Why do people tag? For many, tagging is for sharing their own information and watching others. Even if you tag mostly to remember your own stuff, it is difficult to remain untouched by the presence of others - Have you ever surfed through Gather purely by following one tag after another? This article will explore how tagging lets us connect with others, how tagging propogates memes across information spaces, and some of the costs (economic and cognitive), or tagging as an information retrieval and search mechanism.

From solitary to social

Web browsing can be a solitary experience. Computers are individualistic devices. Many early mornings, I sit at my desk in our office, browsing the web, listening to music. I come across an article I want to remember. I tag it with Del.icio.us. That moment, I go from wandering the web alone to joining a group of others. This transition is important. In a moment, I am transported to a crowd of people with whom I have at least one thing in common. And best of all, I can enjoy their presence, but I don't need to converse. After being on many mailing lists for many years, let me say, conversation is often overrated. Often, I like to be in the company of others, without needing to follow threads and participate. It is the same reason that I like reading and working in a cafe - enjoying the presence of others without the burden of active interaction. Similarly, tags provide a companionable social hum that I enjoy.

As I think about it, the motivations to tag can be categorized into two high-level practices: organizational and social. The first arises from the use of tagging as an alternative to structured filing (formal classification); users motivated by this task may attempt to develop a personal standard and use common tags created by others. The latter expresses the communicative nature of tagging (Signaling), wherein users attempt to express themselves, their opinions, and specific qualities of the resources through the tags they choose."

A behaviorist would say that the I get reinforcement the moment after I tag. The social experience is pleasurable. It gets me hooked, it keeps me coming back.

Alternatively, on tagging the article, I might learn that I am the only one interested in that item, the one person who cared to tag that resource. Perhaps I am simply the first one. Perhaps I will become a trendsetter - my act of tagging will enable others to follow behind, discovering my footsteps. I can always imagine...

Notice how this works on Flickr. Someone takes a picture. They call it a squaredcircle. Another person is struck by that - they create their own squared-circle. Soon, there are many more. Squaredcircles become a Flickr trend. Like latteart.

This is the individual perspective. What is happening on the social level with tagging. I examine that next.

Ad-hoc groups or crowds

From the beginning, the World Wide Web has been a place for group formation, to find and connect with like-minded others. An important characteristic of tagging systems is that they lead to ad-hoc group creation, lowering the barriers to finding like-minded others, enabling social discovery and connections.

The basic social formations supported by tagging are more like crowds than true groups. I see the milling crowds and have some idea about what they are doing (reading, watching), but I don't know these people - they are not part of my network or members of my mailing lists or Gather group that I subscribe to. These are ad-hoc groups brought together by a particular tag or keyword. (Note. Systems like MyWeb2.0 are moving away from this by incorporating social networks.)

Four conditions for wisdom of crowds

Starting with Le Bon's analysis in 1895, psychologists have focused on the negative aspects of crowd behavior. Recently James Suroweicki has refuted this notion. His elegant analysis of four conditions that can lead to "wisdom of crowds" seems relevant for tagging systems. The four principles are (1) diversity of opinion - each individual brings their idiosyncratic perspective to bear on the issue. (this is definitely true for tagging. There is a long tail of tags). (2) independence of members from one another (that people make independent decisions. This is why mass copying of others' tags is not a good idea.) (3) decentralization (with tagging power does not reside in a central location, but it does seem to very influenced by the first few taggers..) (4) a good method for aggregating opinions (Tag clouds and simple lists seem to work well for this, though better methods are needed.

So far, tagging systems seem relatively free of negative aspects of other types of crowd behavior. For example, blog comments and track-backs often seem to encourage herd mentality and there are instances of mob-like behavior. So far, I am yet to see mob-like behavior on a tagging system, though there are examples of frenzied tagging behavior.

Stalking, imitation and gossip

What would a good social system be without some means of stalking, imitation and gossip? (Speaking of which - I was recently reading something about evolutionary psychobiology and the importance of Gossip in developing language and semantic maps in early humans, but can't for the life of me remember where -- need to come back to this - anyway, I will look into this and come back with some references). Part of social life is all the things we pretend we don't do when in polite company. Most of us, at some point or the other stalked someone (remember when you could "finger" people). Some report learning about others' personal lives using their me and craigslist tags. And of course, we can imitate people we watch (copy their items and tags). Recently, I have started noticing the watercooler type post-event conversations around photographs on Flickr (facilitated by specific event tags).

Luckily, tagging systems do not promote popularity lists the way blogs do. If they did, then this rich social tapestry might degenerate to popularity contests, and otherwise sane people would start behaving as in high school (specifically American high school. For a fascinating article on Why Nerds Are Unpopular - the importance of Gossip, pecking order, arbitrary hierarchies in social organizations and group flock behavior in american high schools - read this article by Paul Graham).

Tagging is malleable

Like all good social structures, tagging is malleable - it takes the form best supported by the content, rather than impose a rigid structure on the content. On Flickr, can lead to ad-hoc collaboration, collective self-expression that is very different than the type of tagging frenzy we witness for popular articles on del.icio.us. As tagging spreads we are likely to see other types of emergent ad-hoc collaboration. And as with any emergent, complex system - patterns begin to emerge - and those patterns serve to convey meaning themselves, but this is a line of thinking that deserves it's own article.

Social transmission of information

To summarize my previous article - tagging captures our individual conceptual associations, but does not force us to categorize or place into a hierarchical taxonomy. It enables loose coordination, but does not enforce the same interpretation of a concept. We could all tag items as "art" but mean very different things (not just Mapplethorpe or Monet). That would create chaos in a shared folder scheme, but works well in a social tagging system, because of aggregation, ranking, and comments - all means by which Gather's tagging system works as a variable within a function that also includes connections, page views, comments, and ratings. Not all articles tagged "Art" are created equal.

Tagging and Collaborative Filtering

By allowing loose coordination, tagging systems allow social exchange of conceptual information. Earlier I had written how collaborative filtering can be likened to a social process whereby like minded individuals share recommendations of books, movies etc. I watch a movie, I tell a friend that I liked it. In turn, she recommends a movie to me. Tagging facilitates a similar but richer information exchange. I comment that a movie is "romantic", or "a good holiday movie". Everyone who overhears me has access to this metadata about the movie. The social exchange goes beyond collaborative filtering - facilitating transfer of more abstract, conceptual information about the movie. (Note. the preference information is transferred implicitly - we are more likely to tag items we like than don't like).

Tagging enables social coordination that is simultaneously more direct and abstract than collaborative filtering. More abstract since we are exchanging conceptual information. More direct, since there is no algorithm mediating our connection. When we navigate by tags, we are directly connecting with others.

Flickr and del.icio.us both show that tagging helps in the spread of ideas, memes, trends and fashions. A related question - what role does it play in concept development, in social consensus building? Our concepts and languages are constantly in flux. If tagging systems allow a loose coordination of terms across people, then the question arises: "What role do tagging systems play in ebb and flow of concepts".

Concepts like squaredcircle and latteart are born and supported by Flickr. But what about more complex concepts?

To examine this question, let us consider two much-tagged concepts. The first one was tagged more than 230,000 times on del.icio.us (by October 2005), is the topic of many articles and blog-posts, the tagline of many applications and more recently the butt of many jokes and much ridicule. Yes, I am talking about Web 2.0. The second concept I am interested in examining is AJAX.

There is a fascinating difference between the two concepts, how they came to be, and their tagging trajectories.

A concept in search of a name

The term AJAX was coined by Jesse James Garrett in February 2005. People immediately got the concept AJAX. For many, it was a recognition of a concept they already implicitly understood, rather than a new concept they had to form. It demonstrates the power of naming something at the right time. The term AJAX took off - in the blogosphere, in the del.icio.us world. There has been no looking back. (Read Clay Shirky and Tom Coates for an analysis of the AJAX tag).

A name in search of a concept

Contrast that with the term Web 2.0. It was introduced by Tim O'Reilly, and first used on del.icio.us in March 2004. By October 2005, it had been used 230,000 times.

You might think that with so many people using the same word there must be some consensus regarding "Web 2.0". You might presume from this tagging frenzy that people understood the concept of Web 2.0. You would be wrong. I was.

First we had the conference. Then the experts starting defining it almost on a daily basis. But none of the definitions would quite stick. Then it became the butt of jokes. Web2.0 validator, BullshitGenerator, Supercilious, Web2.0 or not, Flocksucks, etc. And now, even the experts have started rejecting the term. This analysis and rejection of the term is mostly happening on blogs, comments and track-backs. It will be interesting to watch if and when this reverberates in the tag world.

Today, there is no consensus regarding what the term means. A valid question is - why were so many people tagging things Web 2.0 if no one understood what it meant. My hypothesis is that this is due to the nature of tags which only require you to make a conceptual association with a resource. You never need to say what something definitely "is" or "is not". It does not encourage you to ask the hard questions: "What is not Web2.0?"; "How is Web2.0 different than previous concepts?"

Interestingly, the daily expert definitions and the Web2.0 spoof sites (Web 2.0 validator and Web 2.0 or not) attempt to do what tagging systems do not - put boundaries around the concept in a more definitive manner.

This highlights both the strength and weakness of tagging. On the one hand, tagging allows social coordination which never forces a decision, so you can reach a tagging frenzy even if the concept is ultimately rejected.

But Tagging Is Not Free

On the surface tagging seems to offer a new paradigm of organising information, one that reduces the cost of entry and so enables a long tail of participation to emerge. I've come to realise that the cost isn't removed, instead it's displaced, transferred and possibly increased. Tagging bulldozes the cost of classification and piles it onto the price of discovery.

There's a saying I've heard once or twice (I wish I could attribute it): "The cost of metadata is in its application, but the value of metadata is in its use."

Not exactly something you'll be quoting at dinner parties (I have tried - it's a good way to kill a good buzz), but it nicely captures the cost/benefit gaps of metadata.

The arguments against professional classification have almost always worked on the cost side of the equation. Automated indexing, search and now tagging are seen as ways drive down classification costs. But classification costs are only one part of the system.

In my view the total cost of an information retrieval system is the cost of classification plus the cost of discovery (In Peter Morville's latest book, Ambient Findability - Information Retrieval and Human Information Interaction are discussed in depth and provide a fascinating exegesis on the state of IR and HII). In the formal classification world you have a very small number of people incurring a high cost in order to reduce the costs incurred by a very large number of people. In contrast the tagging world has the unit costs reversed: it's cheap to classify, expensive to find. But the numbers of people involved are large in both cases so you end up with a lot of people paying a tiny cost to classify added to a lot of people paying a high price to discover. I think it's pretty likely that the total cost is going to end up much higher than in the classification scenario.

Here's an analogy. I visit a lot of used-book stores. The true cost of an item in a used book store is a function of the time it takes me to find it, not the price (which is always cheap). A very large book store is more likely to have what I want, but at a greater discovery cost. Like del.icio.us, a book store is great for serendipitous discovery but not so good for known item retrieval. Put another way, del.icio.us wouldn't be your first choice if you needed articles on Rousseau and the French Revolution, just like the the Salvation Army Thriftstore wouldn't be your first choice if you needed a smoking jacket, size 42S (I do - by the way - so if you come upon one - email me).

Where I think this might be wrong (or at least need some revision) is suggesting that all the discovery costs are shifted back to the user. In fact, most of the costs are shifted to search, blogs and other more efficient discovery tools. In large part this is because the domain of tagging systems has been the "big messy" web.

In that case, the "classicial" economics of information retrieval don't apply because there are often multiple ways of finding things. Or because Google can radically lower your discovery costs by selling keyword advertising to offset their infrastructure. Or because algorithms can do much of the heavy lifting. Or because users' expectations are for "just good enough" results. Or because users are not interesting in finding so much as tracking. And so on.

But I'd argue that once the domain is constrained--by subject, by context, by user population, by privacy/security, by business goals, or by those things in combination--the economic principals of classification and retrieval come back into play. Because other discovery tools are either not available or not optimal, poorly designed retrieval systems do shift the burden back to the user.

In that middle ground--and the "big messy" web contains probably millions of cases where local structure is valuable, not to mention information systems that aren't part of the "big messy" web--I think there's a large area where a mixture of emergent, algorithmic, formal and now social classification systems will make for optimal retrieval. And perhaps this is where Gather falls down right now. The social classification system of tagging is in place, but the search functionality is circa 1999 (there are, as far as I can tell - no contextual search algorythms at play with Gather search), and Gather removed their formal taxonomy over a year and a half ago - all meaning that tags are more important than ever, and there is a huge discovery cost shifted to the user.

Conclusion

I strongly believe that all good social systems need to serve the individual motive. Tagging works because it strikes a balance between the individual and social. It serves the individual motive of remembering, and forms a ad-hoc social groups around it. If you are designing a tagging system you need to understand how it serves the individual and what sort of social formations it supports.

Finally, writing this piece reminded me how much I admire good designers. It took me a while to gain clarity regarding how tagging works on a social level and a cognitive level. And I am simply deconstructing. What came before are the people who dreamt this up in their minds. Hats off to them. 

--------------------------------

Note: Will Evans is a software information architect for a risk modeling software company in Boston. Previously he was the information architect responsible for designing the Gather user experience (UX). He has published articles about Information Architecture, User Experience, and Interaction Design. He has taught User Centered Design and Building Usable Enterprise Architectures to both small and large corporate audiences.

He enjoys publishing his musings, ideas, poetry and pre-Simulationist critiques of modern culture, technology and aesthetics. He drinks way too much coffee and needs more sleep but is really trying to change that.