posted 5 Sep 2007 in Volume 11 Issue 1
Feature: Social networking
Jan Wyllie explores the potential consequences of the rise of social-networking – and how taxonomies can help bring order to chaos.
Sometimes, it’s so slow it is almost imperceptible. We’ve always known the mass media and the mass market that it enabled. The thing about the mass approach today is that it is the few pushing their goods, services and thoughts on the many in return for money. It wasn’t always so.
The mass market economy is not yet 200 years old – a tiny fraction of human history. Before that markets were overwhelmingly local, supplied by local farmers, artisans and ‘cottage industries’. Of course, trading was sometimes undertaken over long distances (think of the Silk Road – or roads, to be more accurate) but only for exceptional items, such as spices, silk and other exotica, as well as limited quantities of valuable raw materials, such as marble and gold.
When the media, technology and capital behind the mass market became widely available, very significant changes happened, very quickly: people’s lives changed radically within their lifetimes, whether they liked it or not.
Ever since then, advances in mass production, transport, power and communications technologies have been changing people’s lives at an ever-accelerating pace. As markets became increasingly global so, too, did the mass media and its various creatures, such as celebrities and brand names.
Until recently, the internet has extended and amplified the expansion of the mass market at the expense of local markets. How many bookshops have been closed as a result of the success of Amazon, for example? The internet enables it and a handful of other, established retailing behemoths to compete in many more sectors than any bricks-and-mortar store ever could.
Despite all the change in the past 200 years, the direction of change remains the same: from smaller to bigger, from fewer to more, from local to global, from labour intensive to capital intensive, from steady-state economics to eternal growth economics. But these trends could be beginning to reverse.
Take the mass media, for example. It has lost the advantage that once favoured broadcasting and print media – very few people could afford to own a TV station or a printing press. But with broadband everyone can blog, video and podcast to the rest of the world at low cost to themselves and very little to their audience.
Meanwhile, the mass market as most people have known it, especially in the UK and the US, is hurtling towards the twin brick walls of unsustainable debt and rising energy costs – and those energy costs are not going to drop with the populous nations of China and India becoming more prosperous.
The change in question is a reversing of the polarity of the marketplace: the power of the small is on the increase, with the social and economic advantages starting to swing back towards local, rather than global, trading.
This new form of many-to-many markets is still small. But it is growing exponentially, led as ever by ‘geeks’ and ‘techo-glue-sniffers’, and driven by powerful market forces.
Instead of consuming media passively, anybody and everybody can now air their views to the whole world – virtually for free. Life-and-work-changing wikis, blogs, photo sites, video-chat and a host of other new free software services appear in this market on an almost daily basis.
The opportunity is there for everybody (with a broadband connection) to be able to broadcast or narrowcast their thoughts, voice and video, as well as to evaluate information for themselves and for free, rather than being mere passive receivers of stories from the mass media – the pillar on which the global mass consumer-market rests.
Multimedia document sharing and video conferences are effectively free, enabling people to be with each other in what used to be called virtual space. Who, in the future, will want to waste all that money, time and effort travelling to the office and back? The only aspects of the social experience which cannot be shared through the new medium are touch and smell (which could be an advantage in many offices).
The implications are enormous, significantly affecting the way people live and work. This new broadband ‘@home economy’ is being brought to you by Skype, Technorati, Flikr, Del.icio.us, Netvibes and a host of others serving the ‘long tail’ of the many-to-many marketplace, which make up the social technologies of Web 2.0 in its myriad forms.
One thing all these new software applications have in common is that they are personal and social, not technical, which means that virtually anybody can use them – both to develop and enhance their knowledge of what interests them, as well as to engage in new types of human and economic relationships.
When everyone has their own globally accessible, two-way multimedia communications medium, they will need to know how to manage and organise information and interactions. They need to agree on ways of doing things (protocols) and ways of describing things. Otherwise, there will be a Tower of Babel-style chaos as they desperately scroll up and down endless lists, or are left to the vagaries of free-text searching to find the information they need.
The good news is that we are not entering completely uncharted waters. Librarians have had to deal with similar problems for hundreds of years. So have all large organisations. One of the reasons that science has been such a successful endeavour is that these kinds of common protocols and terms have been at the heart of the scientific method since the 1700s, when Swedish botanist, physician and zoologist Carolus Linnaeus developed the first taxonomy to describe living species in the natural world. Evolved versions of this first taxonomy are still used today in the bio-sciences.
By contrast, the blog-world and the new Web 2.0 services are in the early stages of development. However, people recognised early on that blogs, bookmarks and links need to be described, so they tagged their items with descriptive words and phrases to help themselves and others to find them.
If blogs, wikis, and audio and video podcasts are to realise their potential, then they will have to master some of the basic tools of knowledge management (KM). Consistent classification (ie. taxonomies) is, as it always has been, at the heart of the matter. Although there is a large body of professional experience and knowledge in the taxonomy field, classification is a natural faculty that people use instinctively.
Taxonomies, at their best, provide a conceptual framework to help people improve their understanding of the new meanings that come with the multimedia, two-way broadband-internet explosion.
From tagging to folksonomies
Blogs, photo-libraries, podcasts and video clips – of which millions are being added to the web every month – all require some kind of description in order to have any chance of being found and seen.
‘Information overload’ was a hot issue in the 1980s, when sources were limited to books, journals, magazines and newspapers. Early online-information-publishing services, such as Dialog and LexisNexis, were regarded as making a desperate situation worse by making the full text of documents searchable by the words they contained.
There was even a heady moment when many information professionals believed that the ancient arts of indexing and classification had become obsolete. It did not take long for user dissatisfaction to push for more reliable and meaningful ways to access what, at the time, was overwhelmingly textual information.
Since then, information professionals and professional information users have shown an increasing interest in taxonomies and thesauri as a way to make their knowledge storing and gathering processes more useful and satisfying. Much money and effort have been invested in software intended to automate classification and thesauri creation in an attempt to make this kind of information structuring easier to use, more economic and therefore more widespread.
Millions of people have wanted to both retrieve and share; first, their text documents with blogs, then their digital photographs with services such as Flickr; then their audio podcasts and, most recently, their videos with YouTube. As personal library-systems and descriptive words (known as tags) became public, certain words inevitably became more commonly used to describe similar sorts of content – except for the equally inevitable occasions when the same words are used, but their intended meaning is different.
So, personal tags soon evolved into what became variously known as folksonomies, folk classification, ethnoclassification and social classification. Meanings are not defined because users apply their personal preferences within a changing sea of social meaning. This kind of user power and contribution is central to the Web 2.0 ethos of social networking.
Folksonomies simply comprise a flat list of terms with none of the hierarchy or the multi-faceted relationships between terms that are found in the most useful taxonomies. In other words, folksonomies are a list of descriptive terms which a group of users happens to use. There is no thought about design, consistency or quality in the creation of ad hoc tagging systems.
The great thing about the process of informal tagging is that anybody can do it quickly and effortlessly. It requires no discipline and brooks no authority. Furthermore, flat, one-dimensional word-retrieval is simple to realise.
As many commentators have pointed out, searching using folksonomies is good for finding things unexpectedly and also provides rich social-networking opportunities, too. At Del.icio.us and other sites, such as Flickr, there is seen to be a strong social benefit to tagging: people contribute to, and benefit from, the tagging done by others.
The advantages diminish sharply when it comes to people who are using tags and folksonomies to look for something specific. Users can be overwhelmed by the sheer number of terms and inconsistent usages presented in the form of endless alphabetical indexes, without the benefit of any organising schema at all.
People inevitably have to spend their precious time scrolling up and down long lists of tags and articles of possible interest with no way of knowing whether the content to which they point is pertinent to what they are seeking. Perhaps even more troublesome is that there is no way of knowing whether the items that are being sought have been tagged at all, let alone in a way that is useful to the knowledge seeker. And then there is there the difficulty of looking for older material, as tagging usage and fashion changes.
At present, the vast number of taggers have no idea of how using taxonomies could enrich their blog and wiki experience. As taggers and searchers drown in vast (and very messy) indexes of words, yesterday’s key insights and items will, to all intents and purposes, be forgotten – lost in the index – to anybody who doesn’t already know (or cannot remember) the exact tag that the tagger happened to use. A culture that continually loses its memory is unlikely to survive long.
It is ironic that while all the various Web 2.0 media make it possible for people to record their thoughts and pictures for posterity, many (if not most of them) could be unfindable, lost forever in massive inaccurate indexes. Searchers simply have to guess at the words that taggers might have used.
An important aspect of a folksonomy is that it is comprised of terms in a flat namespace. That is to say, there is no hierarchy and no directly specified parent-child or specified relationships between the terms. That means that natural linguistic and formal hierarchies cannot be used to enable people to look at material at different levels of generality. As already noted, the experience of using tags as search terms is the same as using an alphabetical index. It is necessary to look up many different terms to find a set of items which could be accessible with just a single category in a taxonomy.
More people are beginning to see the advantage of encouraging taggers to collaborate on employing common terms. Ways employed to facilitate convergence include the following:
‘Suggest tags for me’;
‘Find synonyms automatically’;
‘Help me use the same tags others use’;
‘Infer hierarchy from the tags’.
To succeed in attempting to improve the performance of tagging at least two conditions must be met. First, the community of users needs to be ready to set rules, and agree upon a set of standards for tags. Second, users need to be made aware of, and agree to, follow these rules.
Once there is a preliminary system in place, it becomes possible to use the most common tags to develop a controlled vocabulary that truly speaks the users’ language. One of the attractions of these kinds of value-free, purely practical conventions is that they quickly become virtually unconscious to users.
Another approach that has been widely discussed (and, in the case of Del.icio.us, even implemented as ‘tag bundles’) is the tagging of tags, which could result in the creation of hierarchical folksonomies. This is an area that is well worth investigation by taxonomy designers as a first step to introducing taggers to the benefits of working with a taxonomy.
Advantages of working with taxonomies
Until recently, taxonomies were something that only librarians or specialists in scientific disciplines seemed to care about. Indeed, part of the reason for the great success of the scientific process is the use of agreed classifications among collaborating researchers.
For the past five years, the ins and outs of building taxonomies have been becoming an increasingly important topic within business and government.
Why the surge in interest? Because taxonomies are an important way of helping organisations to structure their information so that it can be more easily shared. The implementation of enterprise portals, content-management systems and business-to-business e-commerce has highlighted the inadequacy of the way that many organisations classify their knowledge so that users waste time trying to find what’s relevant to them when they need it or, worse, fail to find it at all. A good taxonomy, effectively implemented, will significantly improve knowledge-worker productivity and satisfaction.
If financial assets lose value, knowledge assets are likely to become a relatively secure store of value because they can be turned into production and traded for mutual benefit. As a result, the husbandry of knowledge assets should become a critical strategic objective. The building and implementation of corporate taxonomies is a duty of care for any enterprise today with vast amounts of information that may need to be instantly retrieved.
And if the internet transforms into the semantic web, where users, suppliers and technology systems share common languages to enable more meaningful and effective interaction with a global ‘soft machine’, an understanding of taxonomies will be essential. Taxonomies are at the core of the ontologies out of which the semantic web is being constructed. An understanding of how taxonomies work in this context will be necessary to take advantage of the new market opportunities they may offer.
As collaborative Web 2.0 technologies begin to spread within and between businesses, a choice has to be made about how to manage the vast collections of text, pictures, sound and video being put online everyday. Will random collections of personal tags be the norm for knowledge organisations? Or, will a more navigable structure and a degree of accuracy become more popular?
There is some scope for meeting in the middle. Tag lists and clouds can be analysed. The resulting classification schemes will immediately help users see categories of interest to them which, in turn, should help them form collaborative groups. As people learn a classification system at the most general level, they are primed to learn a more complex classification scheme to further investigate categories of interest.
Of course, the other way round is to make quality, easy-to-learn taxonomies available to information users, along with online tutorials and expert assistance, if the money is available…
Either way, classification and taxonomies will play a vital role in shaping the new knowledge universe being ushered in by Web 2.0 technologies, which will inevitably lead to ‘Enterprise 2.0’ applications.
And anything that helps us to navigate around the humungous amount of information on the internet today – which will double in volume in the next 12 to 18 months – has got to be most welcome.
Jan Wyllie is managing director of Trend Monitor, experts in the application of faceted-classification schemes (http://trendmonitor2.wordpress.com). He is also the author of Taxonomies: Frameworks for Corporate Knowledge, which is available from the Ark Group in PDF format. Please e-mail Adam Scrimshire for details, firstname.lastname@example.org. Jan Wyllie can be contacted by e-mailing, email@example.com.