TagsandFolksonomies

= = My suggested edits / comments are contained in the attached. some of it has to do with APA and others are editorial. Kem = = I combined Kem's edits with my own in the attached file. Not sure how concerned you are with the length (only came out to 9 pages with images and reference page when doublespaced).

=Section 2:= Tags and Folksonomies

Introduction
This chapter will provide an overview and understanding of the use of Folksonomies in the web 2.0 environment. It will inform briefly about the development of this phenomenon. Key questions or issues will be discussed, as well as potential solutions. Finally, we’ll end with a summary of what was covered, a list of references for further research, and definitions of requisite terms.

Background Information
So, to begin, one might ask, “What is a Folksonomy?” Many users of Facebook are already familiar with the term ‘tagging’ as it is applied to images either uploaded by users or their friends. This tagging is simply assigning the names of people that are found in an image, to that image. So a user is able to be tagged in multiple pictures from their own personal photo albums as well as from what other friends’ might post. This same idea is applied to anything found on the web. Whether websites, articles, blogs, images, video or anything, as long as it can be found, it can be tagged. However, with Folksonomies, the idea is taken further. Essentially, people create personal collections of things, ‘tag’ them, and then share with others. This is known as ‘social bookmarking’ which is a term that has only been around for a few years (Green, 2010, p. 4). Green (2010) elaborates: [Users can] store their bookmarks or favorites online, where they can be accessed from any computer. [They] create and assign tags, or descriptive terms, to each item they bookmark to help organize and remember the item. These tags are used instead of the hierarchical folders found in browser windows. [They also] share their bookmarks and see what other people are bookmarking. (p. 4) So the actual act of tagging an article, or selecting words that describe it, is left up to the user. This is where the ‘folks’ part of Folksonomy comes into play. Thomas Vander Wal, an information architect, is credited by Terdiman (2005) and others with coining this term. Vander Wal (2004), in describing this phenomenon, refers to it as a ‘bottom up’ type of categorizing instead of the typical ‘top down’ structure that U.S. society is most familiar with (as in the creation of the Dewey Decimal System.) As stated by Matt Haughey, the founder of MetaFilter, "To me, they're a great new organization tool for applications and large content sites.” "Tags are great because you throw caution to the wind, forget about whittling down everything into a distinct set of categories and instead let folks loose, categorizing their own stuff on their own terms” (as cited by Terdiman, 2005). Many websites are using this type of user generated content organizer, or Folksonomy, such as Delicious or Flickr. These are some of the most commonly used social bookmarking sites that allow for both aggregation of one’s personal collection of content, as well as using the tagging feature to organize it, and not only for yourself but for others if you choose. This tagging component then creates value for the object it’s assigned to. Delicious is primarily used for documents or articles and according to Green (2010) is the most widely used by librarians. Figure 1 below shows the bookmarking tool within Delicious. Flickr is also a commonly used Folksonomy and its content consists of user-generated photography. It is uniquely different from sharing articles or documents found on the web, as with Delicious, in that its content consists of original images. So when a user tags an image they’ve uploaded, say for example a picture of a beach in Oahu, they might tag it with words such as ‘Oahu_2008’, ‘white_sand_beach’, and ‘sunny_day.’ This tagging is essentially personal, but also allows for faster retrieval at a later time since there are more key word descriptions. Owners can then share with family or friends, the public, or not (see Figure 2). If shared, then others can find the image (and others tagged similarly) by running a search using any of those descriptive words. What is so remarkable about the use of Folksonomies, is that an actual person is responsible for assigning value to objects through this collective tagging. Search engines, such as Google (so far), “make no determination of the quality of sites it indexes” (Bauer, 2001, p. 41) and are aggregated without that inherent ‘whittling’ down feature. Therefore an individual can much more quickly find, share, and store sources that are potentially relevant. As Barsky, Pardon, and Bentley et al., state, “This architecture of participation, the generation of sharing digital artifacts by groups, teams, and individuals, ensures that social tagging is responsive to users” and “it thrives on the concept of collective intelligence, or wisdom of the crowds, when working cooperatively and sharing ideas” (as cited by Lee & Ge, 2010). Furthermore, as juxtaposed with traditional methods for organization that rely on a systematic structure of classification, Folksonomies are more about putting things into categories. These categories are broad and loosely defined. Jacob states, “It is based more on a synthesis of similarity than a systematic arrangement of materials” (as cited in Mathes, 2004, p. 4). So, while classification is known for putting all of an artifact into one category only, categorization through Folksonomies does the opposite. And by this very nature it opens something up for more potential use.
 * (Figure 1)**
 * (Figure 2)**

Key Issues
It might be helpful, before moving forward, to clarify just what is meant by tags and tag clouds, and how they have developed into the folksonomies that we are seeing today. Humanity has been organizing the information available to it from time immemorial, and tags and folksonomies are simply the latest evolution of that attempt at organization. Most everyone knows someone in their circle of friends who has or had a huge collection of cd’s or cassette tapes, or even vinyl records. Most likely they were not simply in a pile on the floor. Instead some form of organization had been imposed on the content. They might have been organized by artist name, genre, favorability, or heck, even date of recording. This made retrieval by the collector relatively easy, but what about the level of difficulty for a visitor unfamiliar with the classification scheme? Or, when people used brick and mortar libraries they might have encountered a physical card catalog that defined the location of a book or resource by author, subject, title, etc. Again, familiarity with the defined classification hierarchy was needed to make these systems useful. These were all efforts to organize the content for retrieval. Tags are simply keywords that define content according to the vocabulary of the user instead of the predefined vocabulary of the system creator. Tag clouds develop around content as multiple users contribute their own tags to related material. A folksonomy is basically the system within which these tags and tag clouds are used for defining and retrieving content.

One of the most important aspects of folksonomies as classification systems is the bottom-up nature of the classifications. Rather than establishing folders by subject matter or some other arbitrary hierarchy that divides content according to the perspective of the creators or administrators of a system, folksonomies develop more organically as users themselves apply multiple, varying tags to content. In this way, a system developer’s perspective is actually subordinated to that of the users of the system. This defining characteristic is crucial to understanding both the potential power and weaknesses of folksonomies. For one thing, a more user-defined experience can result in substantial increases in the utility of classification systems. The system of tagging content by keywords generated from multiple users results in a greater likelihood that one will successfully search for content because of the higher probability of common terminology between the searcher and at least one of the tagging users. However, there is the potential that relevant content will be missed because it is poorly tagged or the tagger does not reflect the vernacular or mindset of the searcher. Another issue along similar lines is the propensity for irrelevant material to be summoned because of problems with synonymous, ambiguous, and similarly spelled tags with differing meanings. It is hoped, and studies have shown, that the power of the crowd in terms of large numbers of users tagging within a system will result in more efficient tag clouds centered around a developing shared vocabulary (Robu, Halpin, Shepherd 2009). Over time this may overcome the deficiencies related to so many different users with their own individual vernaculars and perspectives tagging content in their own ways. In essence, both the strengths and weaknesses of folksonomies as classification systems stem from the power of the crowd to define content.

It is interesting to note that even with so many different users responsible for the classification of resources via tagging in a folksonomy in contrast to a more traditional, expert or administrator-defined taxonomy, a consensus around the best tags or classifications of a resource does seem to emerge over time. Like many user-distributed systems, consensus in a folksonomy develops according to a power law distribution in which a few core tags become associated with a resource and are reinforced as more users demonstrate their “agreement” with those tags as appropriately descriptive for the content (Mathes 2004). Radiating outward in the tag cloud are a far greater number of tags linked with that resource that may be helpful for some particular users, but their utility is not widely enough shared for them to become closely associated with the content via frequent and repetitive use (Mathes 2004). Taking this concept further is research being conducted to increase the effectiveness of search engine results by applying these concepts within a community of practice as a sort of second layer of search in which the initial results of the engine are refined based on tags applied by experts in the field whose tags are given added weight. In this way the results are re-ranked to prioritize content likely to be of particular import to users within that community of practice as opposed to general users (Boydell and Smyth 2006).

It is likely that tags and folksonomies will play a large role in the evolution of the internet into Web3.0, known sometimes as the Semantic Web (Berners-Lee 2001). The great leap forward in this regard will be tags and metadata that are written in machine-readable format so that computers and other devices can actually begin to understand the content of the web rather than just report on the text it contains (Herman 2008). If realized, this could lead the way into the artificial intelligence and semi-autonomous machines of science fiction past (Unknown 2007). Wouldn't it be something if the next generation could trace the roots of their computers that process knowledge instead of just text back to the humble photo sharing site Flickr, or to Facebook, or Delicious?
 * Solutions**

Summary
From card catalogs to social bookmarking to tags and folksonomies and into the future, organizing the vast and ever-increasing content of the web into usable resources is an undertaking of enormous scale. Yet the simple tag that lets one retrieve all pictures of their beach vacation is laying a foundation for exponential leaps forward in realizing this bit of potential for the web. Releasing the collaborative power of the masses to tag and organize content on their own without being pigeon-holed by formulaic, administratively-defined categories is giving rise to folksonomies--truly, the people's taxonomies (Vander Wal 2007)

Terms & Definitions
__Folksonomy__: Wikipedia states, “A folksonomy is a system of classification derived from the practice and method of collaboratively creating and managing tags to annotate and categorize content; this practice is also known as collaborative tagging, social classification, social indexing, and social tagging.”

__Social Bookmarking__: Wikipedia states, “Social bookmarking is a method for Internet users to organize, store, manage and search for bookmarks of resources online. Unlike file sharing, the resources themselves aren't shared, merely bookmarks that reference them.”