Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

All indexing is wrong; some indexing is useful : social tagging in libraries Yousefi, Baharak 2007

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2007-0648.pdf [ 4.03MB ]
JSON: 831-1.0101032.json
JSON-LD: 831-1.0101032-ld.json
RDF/XML (Pretty): 831-1.0101032-rdf.xml
RDF/JSON: 831-1.0101032-rdf.json
Turtle: 831-1.0101032-turtle.txt
N-Triples: 831-1.0101032-rdf-ntriples.txt
Original Record: 831-1.0101032-source.json
Full Text

Full Text

ALL INDEXING IS WRONG; SOME INDEXING IS USEFUL: SOCIAL TAGGING IN LIBRARIES by BAHARAK YOUSEFI B.A. (Honours), Simon Fraser University, 1998 M.A., Simon Fraser University, 2003 THESIS SUBMITTED IN PARTIAL FULLFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF LIBRARY AND INFORMATION STUDIES in THE FACULTY OF GRADUATE STUDIES THE UNIVERSITY OF BRITISH COLUMBIA August 2007 © Baharak Yousef i, 2007 11 A b s t r a c t Social tagging, the activity and process by which users add descriptive tags to shared, digital content, is a socio-politically significant form of indexing. Adding social software to library OPACs challenges the legitimacy of traditional indexing languages and can enhance the information literacy - the ability to seek, find, and evaluate information - of library patrons, rendering social tagging a matter of import in the pursuit of critical librarianship. This thesis uses Sidney Tarrow's (1994) four basic properties of social movements - collective challenge, common purpose, solidarity, and sustaining collective action - to present social tagging as a social movement. It argues in favour of adopting an anarchist, and specifically Kropotkinist, paradigm for the future study, development, and implementation of social tagging. Most significantly, social tagging is carried out from the bottom upwards by means of patron contributions and not from the top downwards by means of authoritative rule. By choosing an anarchist paradigm, librarians and LIS scholars can make certain that social tagging continues to be maintained, developed, and studied as an anarchist social movement. i i i Table of Contents Abstract i i Table of Contents ; i i i Lis t of Figures iv Acknowledgements v C H A P T E R I O n Cri t ical Librarianship: A n Introduction 1 C H A P T E R II Tag, K e y w o r d , Graffito, Indexing Term, Metadata, Literature Review 7 Introduction 7 Social Tagging: Definitions and Advantages 8 Social Tagging: A Cri t ical Analys is 14 The Socio-political Ut i l i ty of Social Tagging 20 C H A P T E R III The N e w Order of Social Tagging 24 The Radicalization of Indexing 24 Social Tagging as a Social Movement 34 Conclusion 49 C H A P T E R IV The Structure of an Anarchist Paradigm 51 Introduction 51 Parad igm - Kuhn ian 52 Parad igm - Anarchist - Kropotkinist 58 Conclus ion 74 C H A P T E R V N o w We Can D o Big Things for Love: A n Epi logue 76 Bibl iography 81 iv List of Figures Figure 1.1 Social Tagging on Library O P A C s as a Type of Indexing and a Social Movement 5 Figure 2.1 L ibra ryThing Tag C l o u d 9 Acknowledgements I would like to express my deepest gratitude to the members of my supervisory committee for their patient guidance. I am indebted to Dr. Joseph Tennis, humbled by the anarchists and librarians in my life, and grateful for my lifetime membership on Library Thing. 1 Chapter One On Critical Librarianship: An Introduction The world has not to be put in order: the world is order incarnate. It is for us to put ourselves in unison with this order, to know what is the world order in contradistinction to the wishful-thinking orders which we seek to impose on one another. The power which we long to possess, in order to establish the good, the true and the beautiful, would prove to be, if we could have it, but the means of destroying one another. -HenryMiller (1939, 33) Critical librarianship refers to the position of being cognizant and involved in the social and political responsibilities inherent within the profession. Librarians help patrons find information. Seeking, finding, and evaluating information - information literacy - is a socio-political skill because access to information enables socio-political organization (Andersen, 2006). Knowledge organization - the lifeblood of librarianship - and individuals' ability to organize are intimately related. Librarianship is, therefore, not a socio-politically neutral profession (Samek, 1998). As part of their core set of values, The Canadian Library Association states: "effective advocacy is based upon understanding the social, cultural, political and historical contexts in which libraries and information services function" (CLA, n.d.). This thesis is a contribution to critical librarianship. The nature of librarianship as a socio-political endeavor confers equal responsibility on both scholars and professionals 2 in this field. It necessitates proactive participation i n matters and occasions of social, cultural and polit ical significance. The emergence of Web 2.0 technologies and tagging is one such occasion. Tagging is the process of labeling online content (Rainie, 2007). Sharing one's personal tags w i th other users is social tagging. The process of social tagging allows users to organize information intentionally, capriciously, and ideologically. It grants users the power to name. It is i n this very act of bestowal and practice of power that I am most interested in. In its broadest sense, social tagging is a social movement and a quintessentially postmodern archetype. A s an instrument of socio-political commentary and participation, social tagging has made the construction of meaning an inclusive and emergent pastime. In Library and Information Science, it is an atypical form of indexing. It is an alternative representation of reality - one wrought by those experiencing it. If traditional indexing languages are authoritative, theoretical representations of the w o r l d according to classificationists, social tagging is "a classic example of bottom-up bu i ld ing of categories instead of top-down imposi t ion of categories" (Rainie, 2007). Nat ional Information Standards Organization's (NISO) "Guidel ines for the Construction, Format, and Management of Mono l ingua l Vocabularies" defines indexing and indexing languages as follows: Indexing - A method by w h i c h terms or subject headings from a controlled vocabulary are selected by a human or computer to represent the concepts 3 in or attributes of a content object. The terms may or may not occur in the content object.1 Indexing language - A controlled vocabulary or classification system and the rules for its application. A n indexing language is used for the representation of concepts dealt with in documents [content objects] and for retrieval of such documents [content objects] from an information storage and retrieval system (ANSI/NISO Z39.19-2005, p. 6). We are witnessing the beginnings of a revolution in the act of indexing and the construction of indexing languages. Where available, the interpretation and representation of significant characteristics of content objects (Tennis, 2006)2 in our library systems is being performed by our patrons. Users are logging in and tagging the contents of our catalogues (e.g.; This contribution to libraries and other social tagging sites has generated a fair amount of excitement; people are participating. As scholars and professionals in this field, understanding the participation of taggers - amateur indexers - should be placed among our professional priorities. The freedom afforded us by lack of physical constraint ought to be harnessed for goals other than meeting the traditional objectives of bibliographic systems. As librarians, we ought to encourage and enable social tagging because, at last, we 1 This definition of indexing is used throughout this thesis as an example of traditional indexing. Although I do not specifically address other types of indexing such as natural language and free text indexing, the point still remains that until social tagging, indexing has been performed without public participation. Only in social tagging has the power to interpret and represent significant characteristics of content objects (Tennis, 2006) been shared with the end user. 2 This thesis uses Tennis' 2006 definition of indexing to comprehensively define both traditional indexing and social tagging. 4 can - the digital library has no shelves. We are able to address the discordance between how the world has been thus organized and how it is seen and experienced by marginalized groups and individuals in our society. It would be negligent to disregard this opportunity: The task for a modern industrial society is to achieve what is now technically realizable, namely, a society which is really based on free voluntary participation of people who produce and create, live their lives freely within institutions they control, and with limited hierarchal structures, possibly none at all (Chomsky, 1991). In Technopoly: The Surrender of Culture to Technology, Neil Postman (1993) invites his readers to ask of any new piece of technology: what is the problem to which this technology is the solution? Here, technology is considered pragmatically and towards human needs and not as an end in and of itself. How and what should we think about social tagging? Which of the many questions should we ask? Conceptually speaking, social tagging is of personal and professional interest to disparate groups of people. Though its novelty may not allow retrospection, its magnitude demands careful reflection. Privileging certain aspects and foregoing others, this thesis has explicit epistemic interests in social tagging as a social movement in response to indexing practices currently utilized in libraries. Though closely related, this work is not a study of technology and its place in human culture. The focus is on social tagging as a human phenomenon and not as an example of the transformative powers of the Internet or technology writ large. Human drive for solution finding is an 5 essential ingredient of technology (Ciborra, 2004). In libraries, what is the problem to which social tagging is a solution? This thesis does not assume to describe or capture the multiplicity of ways social tagging may be studied; rather it seeks to establish one possible conceptualization. As outlined in the following diagram we can see the act of indexing comprising social tagging as well as what I have called traditional indexing. The former, social tagging, I set out as a social movement that can admit to an anarchist conceptualization. I argue that this should be established and maintained by participants of critical librarianship. Figure 1.1: Social Tagging on Library OPACs as a Type of Indexing and a Social Movement (The size of the circles is not significant) 6 The structure of the thesis follows this conceptualization. Fo l lowing the second chapter's literature review of social tagging, chapters 3 and 4 take the fo l lowing positions: a) the act of indexing by social tagging on library O P A C s is a social movement, and b) it is recommended that librarians and LIS scholars adopt an anarchist (Kropotkinist) paradigm for the future study, development, and implementation of social tagging. The fifth and concluding chapter w i l l endeavor to summarize and discuss directions for future research. 7 C h a p t e r T w o T a g , K e y w o r d , G r a f f i t o , I n d e x i n g T e r m , M e t a d a t a , L i t era ture R e v i e w There is a precise line of separation between a nonrevolutionary and revolutionary situation. In a nonrevolutionary situation, one can solve the pressing immediate problems while postponing the big key problems; in a revolutionary situation, this strategy no longer works and one has to tackle the Big Problem in order to even solve the "small" pressing ones. - Slavoj Zizek (2006, 380) Introduction At this stage in the study of social tagging, we would do well to emphasize its shifting and emergent nature, rather than to attempt a comprehensive characterization. Like its subject, the material currently available on the topic of social tagging is amorphous: journal articles, web publications, conference presentations, podcasts, blog entries, listserv contributions, etc. Their form does not and should not discount their merit. This topic is being written about, and belongs to the masses rather than a select group of researchers. The literature is ever changing, "largely opinion-based" (Speller, 2007) and almost entirely published online. There are enthusiasts (Kroski, 2005; Shirky, 2005; Sterling, 2007) and naysayers (Blood, 2005; Lawley, 2005). There are spectators, soothsayers, and those calling for contextualization and politicization (Quintarelli, 2005). All and all, there is little doubt that social tagging sites such 8 as Library Thing (, Flickr (, Connotea (, and ( are growing in popularity. They have a combined user base of several million subscribers and new social tagging sites and novice users are being added everyday (Winget, 2006). Though the literature covers a wider range of social tagging sites than those belonging to libraries such as Ann Arbor District Library's O P A C ( and University of Pennsylvania's Penntags enabled catalogue (, the findings are applicable to the library environment, and where possible, this review's focus will be on social tagging software and their utilization in library systems. In reviewing the current literature on social tagging, several questions are addressed: What is social tagging? What are the advantages and disadvantages of social tagging? In what ways is social tagging a politically useful tool? Social Tagging: Def in i t ions and Advantages Social tagging describes the activity and process by which users add descriptive tags to shared, digital content. Social tagging is different from other forms of indexing in two significant ways: the tags are shared, and users and not professional cataloguers, indexers or authors add the metadata. Tagging or labeling digital content is not a new phenomenon. People have been labeling and naming their personal files and bookmarking their favorite websites for some time (Bruce, Jones & Dumais , 2004). These organizational strategies have served the function of f inding and re-finding digital contents for personal use. Social tagging makes users' personal tags publicly available. These contributions result i n a particular type of indexing language referred to by a variety of names: folksonomies (Vander W a l , 2004), ethnoclassification schemes, folk classification schemes, social classification schemes and distributed classification systems (Speller, 2007). Folksonomies uti l ize tag clouds for display purposes. 3 Tag clouds or weighted lists, as they are referred to i n the field of v isual design, are visual ly effective depictions of assigned tags. Commonly , the more often a tag is used, the larger its size i n a tag c loud (Kroski, 2005). Figure 1.1 is a tag c loud from my personal l ibrary on Library Thing. A s illustrated by the c loud, the "Fic t ion" tag is the most frequently used tag, fol lowed by "Feminist Studies". Figure 2.1: LibraryThing Tag Cloud**,« « * | iltHkl 11 *•*•> '*«<«* L i f c »21L!*' mm mmm i m • up **** ** Hnmnt An»umt SSUOMTS tttfWMMN Art A**»O Book imflBphltl Biography Comic Book Cookery & Entertainment Essays Evolution Feminist Studies Fiction History & Culture Knowledge organization ungMp ysHw* Memoir * * & MytMop Philosophy W*y» Poetry MNoi Pretty Pictures Psychology Reference Sex Travel 3 For the sake of consistency, I w i l l use "folkosonomy" throughout this thesis. 10 Social tagging allows for inclusive participation in the construction of indexing terms; it offers flexibility (Kroski, 2005). As demonstrated in Figure 1.1, the books that I have tagged "Myth and Mythology", for example, include religious texts. This is an idiosyncratic way of classifying, which other Library Thing taggers may or may not share with me. Fluidity and flexibility become particularly significant when certain concepts are not otherwise represented in traditional indexing languages. The term "queer", for example, was first used in the 1920s as a derogatory term referring to homosexual men (Allen, 1999). Over the course of the 20 th century, however, its usage changed significantly, becoming an antonym of normative heterosexuality (O'Rourke, 2005) and gaining emphasis in activism (i.e. queer rights) and scholarship (i.e. queer theory). Clay Shirky (2005) writes: Look for the word "queer" in almost any top-level categorization. You will not find it, even though, as an organizing principle for a large group of people, that word matters enormously. Users don't get to participate [in] those kinds of discussions around traditional categorization schemes, but with tagging, anyone is free to use the words he or she thinks are appropriate.... With flexibility comes inclusivity. Issues surrounding the digital divide aside, "metadata is now in the realm of everyman" (Kroski, 2005).4 Expert indexers and cataloguers have lost the monopoly on determining the organizational needs of 4 Digital divide refers to the gap between those who have access to and benefit from digital technology and those who do not (Digital divide: What it is and why it matters, n.d.) 11 the public. S imply put, the more people participate i n knowledge organization, the more varied and inclusive the categories become. Study of tagging habits and tag distribution (Guy & Tonk in 2006; Tonkin , 2006) found that few tags are used often whi le the majority of tags are seldom used. There are no attempts to r i d folksonomies of the seldom-used tags - the L o n g Tai l (Anderson, 2006), a l lowing for the inclusive representation of minority interests. Krosk i (2005) sees the L o n g Tai l distribution as evidence that "folksonomies include everyone's vocabulary and reflect everyone's needs without cultural, social, or poli t ical bias." When combined, the infrequently used tags may even outnumber the popular ones. A certain level of socio-political significance may be attached to the representation of the non-mainstream tags - a topic discussed under the heading of socio-political utility later i n this chapter. Currency is another advantage of social tagging. Digi ta l tags may be created as quickly as digital content. Being adaptive to changing vocabularies and emerging content has been recognized as especially useful i n the field of technology and the "blogosphere", but also more broadly relevant (Speller, 2007). That meaning is referential, context-dependent, and changeable, makes currency not only useful, but also necessary. A s definitions change over time, so can associations between them. Folksonomies a l low for an evolv ing flexibility and serendipity i n browsing not offered by traditional indexing languages. In theory, tag connections and discoveries may be made i n real time. A s two previously unassociated concepts become l inked i n the "real wor ld , " cyber space 12 may be updated to reflect and inform almost immediately. Bruce Sterling (2007) writes about folksonomy: It offers dirt-cheap, machine-assisted herd behavior; common w i s d o m squared; a stampede towards the water holes of semantics. There is room for scholarly smarts i n this approach - for instance, y o u might invent a really cool term like folksonomy - but mostly it's a new way to crowd-surf. It's as though y o u threw a kayak into a mosh pit and gl ided not just through Web pages but through labels, concepts, and ideas, too. The above quote emphasizes financial possibilities, but also, to some extent, serendipitous and artistic potential of social tagging. This particular advantage is elusive and especially subjective. What emerging patterns of knowledge organization w i l l we be witness to, and w i l l they be unusual, unexpected, and beautiful? Social tagging is associated w i t h a certain sense of romantic chaos and unpredictability, wh ich can be interesting as fodder for future discussions. Currently, social tagging is used far more extensively outside of the l ibrary, and the literature is reflective of this trend. What are the imp l i ed advantages of u t i l iz ing social tagging i n l ibrary environments? Winget argues: "not only does user-defined metadata give the Library and Information Science communi ty the opportunity to augment and refine our existing classification methods and schemes to be more user friendly, this method of description might also a l low for an enhanced human information interaction experience" (2006,15). G i v i n g users the ability to shape and influence the data wi th wh ich they interact enhances the human-information interaction experience (Winget, 2006). Act ive 13 participation on the part of the user, together with immediate feedback mechanisms (e.g. tag clouds and the ability to view and connect with other users with similar tags and interests) make for enhanced levels of interactive experience. Furthermore, chance discoveries may be made through exploring related tags, which may have otherwise been separated in traditional indexing languages (Speller, 2007). For example, The Sexual Politics of Meat: A Feminist-Vegetarian Critical Theory by Carol J. Adams (1990) is listed under the subjects: animal welfare, vegetarianism - social aspects, patriarchy, and feminist criticism in the Vancouver Public Library's OP A C . The provided list captures the main themes of the book. It does not, however, inform the user about an area of study, which brings together the formerly disparate areas of feminism and environmentalism: ecofeminism. If the O P A C of the Vancouver Public Library allowed for social tagging, the "ecofeminism" tag could be easily added thus allowing for the possibility of chance discovery of this relatively new concept. The patron could then explore "ecofeminism" using traditional subject headings or tags. Winget (2006) suggests that the desire to participate in and benefit from the social aspects of tagging has an effect on users' tagging behaviour. On Flickr, for example, users seem to adhere to a certain set of unspoken conventions, especially if they are interested in their pictures being found by other members and included in tag sets. Users tend to use multiple tags, provide geographical descriptors of their photos, and when appropriate, use multiple spellings and 14 abbreviations (Winget, 2006). Lack of explicit guidelines and instructions, therefore, does not preclude the observation of unwrit ten rules and standards. Postmodern poli t ical philosopher, Slavoj Zizek (n.d.), describes the weight and power of unspoken rules i n the maintenance and promotion of ideology and desired order: A n d is this not h o w ideology works? The explicit ideological text (or practice) is sustained by the "unplayed" series of obscene superego supplement.. . the explicit ideology of socialist democracy was sustained by a set of implici t (unspoken) obscene injunctions and prohibitions, teaching the subject how not to take explicit norms seriously and h o w to implement a set of publ ic ly unacknowledged prohibitions. Social Tagging: A Cri t i ca l Ana lys i s Simply put, the main criticisms of social tagging are that tags are not "good" enough. For a myr iad of reasons, tags assigned by the public may not be what the critic and the professional w o u l d have assigned. The fo l lowing points are samples of criticisms. Users may "try to game the system" (Lawley, 2005) by deliberate mistagging of content. Tagging gives "free reign to sloppiness" (Mejias, 2005). For example, tags can be misspelled or inconsistently assigned. Synonyms and homonyms are not controlled (Speller, 2007). Tags can be offensive to other users (Blood, 2005). In terms of indexing terms, there is little overlap between tagging and traditional content indexing languages such as the M e d i c a l Subject Head ing (MeSH) thesaurus (Lin , Beaudoin, Bui , & Dasai, 2006). Folksonomies are not particularly useful for f inding specific, accurate 15 information, but as Bruce Sterling (2007) argues "that's beside the point." Critics of social tagging ask the wrong questions and confuse the purpose of tags. Imagine, for example, that up until a year ago, the only types of shoes we were familiar with were comfortable walking and running shoes. Now, imagine a designer introducing a line of patent leather women's shoes with six-inch stiletto heels. To argue that these shoes would not make comfortable for walking and running would be nonsensical. To criticize a technology for not doing what it is not designed to do is a similar misdiagnosis. What, then, is the purpose of folksonomies? This is an important and necessary question with many emerging answers. Any given information system is designed with the purpose of satisfying one or a set of objectives. The purpose of something is its intention. Questions relating to purpose are best asked using the adverb "why." For example, "why are you peeling an orange?" or "why was Connotea created?" In asking such questions, you are inquiring about someone's or something's purpose - their intentions. In her text The Intellectual Foundations of Information Organization, Elaine Svenonius (2000) explores and emphasizes the importance of a full understanding of the objectives as the first step in the design and construction of any and all bibliographic systems. She states that purpose should determine ontology and not vice versa (2000).5 It is, therefore, not 5 The exact phrasing is as follows: "In the design of a database objectives should determine ontology and not vice versa, since for any given set of objectives, alternative models can be developed for alternative purposes" (Svenonius, 2000,18). Given an earlier assertion that purpose is "the objectives to be achieved by a system for organizing information" (2000,1), I have used purpose and objective interchangeably. 16 feasible to evaluate something without understanding its objectives. Accord ing to Svenonius, the fol lowing five functions are the objectives of a full-featured bibliographic system: finding, collocating, choosing, acquiring, and navigating. This list cannot be used to evaluate and criticize social tagging systems. W e must ho ld these systems accountable to a different set of criteria: one w h i c h is i n line w i t h their stated (and otherwise ascertained) purposes. Pr ior to discussing what folksonomies are designed to do, let us be clear that there is no evidence or c la im that they have been designed to replace traditional indexing languages. It is, therefore, not necessary to make tags behave l ike controlled vocabularies and folksonomies behave l ike k n o w n indexing languages, because though they may complement each other, their purposes are different. The problem of single-design focus i n Library and Information Science is addressed by G o o d and Tennis: There are differences here, differences that shed light on the nature of these initiatives [information organization frameworks]. Yet, not everyone i n the Information Sciences sees information organization as a diverse set of approaches. Some cla im that folksonomies are nothing new or that they are nascent structures on an inevitably teleological path toward discovering authority control (2007, 2). G o o d and Tennis (2007) studied the purpose, work practice, and structure of two example systems, Connotea (ht tp: / / and M E D L I N E (ht tp: / / to illustrate the intentional differences between the two systems. Connotea, a social tagging site w i t h the intended audience of 17 scientists, has as their motto: "Organize, Share, Discover." MEDLINE, an indexing system and bibliographic database that utilizes the Medical Subject Headings thesaurus, strives "to enable search retrieval by eliminating (or accounting for) the use of variant terminology for the same concept" (Good & Tennis, 2007, 3). I will review the "purpose" part of their discussion and findings here. The authors used a list of six different purposes as comparison points between Connotea and MEDLINE, and found that they overlapped on only one purpose - finding. Connotea's list of objectives also included management of personal collections, sharing of resources among peers, and interaction with the information organization system (e.g. tagging), while MEDLINE, along with the finding objective, listed collocation of resources and retrieval through precision and recall. Looking back at criticisms of social tagging discussed earlier, sloppiness and inconsistent tagging (Mejias, 2005), uncontrolled synonyms and homonyms (Speller, 2007), and little overlap between folksonomies and indexing languages such as MeSH terms (Lin, Beaudoin, Bui, & Dasai, 2006) are all related to two objectives, which Good and Tennis (2007) list as belonging to the traditional indexing as opposed to social tagging systems: namely collocation and high degrees of precision and recall in retrieval. As Good and Tennis (2007) demonstrate, there is something fundamental to be gained from studying the surrounding discourse of these information organization frameworks. However, an empirical investigation of taggers' 18 motivations w o u l d also add to this discussion of purpose and inform future design objectives. A s systems, wh ich rely entirely on user participation, the users' objectives must mirror that of the creators and, by extension, the information organization framework. These are not detached, independent systems, wh ich w o u l d function without the input of their users. A l though it is important to study the objectives of the creator (and the systems), it is necessary, in the long run, to ensure that users' objectives are also being met. The theoretical work presented on user and motivations offer the fo l lowing as possible incentives: future retrieval, contribution and sharing, attention, play and competition, self-presentation, and opinion expression (Mar low, Naaman, boyd, & Davis , 2006). The data seems to be largely gathered through the examination of systems and tags as opposed to direct user studies such as interviews or questionnaires. This gap i n empirical research and scholarship w i l l be further addressed in the concluding chapter of this thesis. To summarize, the research and findings on how social tagging and folksonomies work is varied and emergent. The fol lowing paragraphs w i l l address the question of why , ideologically speaking, social tagging and folksonomies may be of value to our l ibrary systems. Thus far, the critical issues discussed i n this chapter have been more procedural than ideological. One of the topics, wh ich was discussed as an advantage of social tagging is the flexibility wh ich folksonomies afford knowledge organization in general. The fact that users have the capacity to 19 organize and categorize according to their own purposes and beliefs is socio-poli t ical ly significant and w i l l be discussed further i n the fo l lowing section. However , critics may resist and condemn these same freedoms on ideological grounds. To protest against publ ic ly assigned tags is to question the phi losophy of social tagging at a very basic level. For example, one critic described being personally offended by a photo on Fl ickr , wh ich was tagged M L K (Mart in Luther King) : Unfortunately, for a few hours this morning the most recent tagged photo under M L K was a picture of a protester's sign that read "Setting aside our differences to focus on our common goals: peace, love, harmony, k i l l i ng Jews, and tolerance." Nice . N o w , that photo is perfectly appropriate on Fl ickr as part of an individual ' s collection, and as documentation of Sunday's rally. It's perfectly appropriate as an illustration for "protests", or even "Israel" and "Palestine", even though it surely w i l l offend some people wherever it appears. But it is not appropriate to illustrate a category tagged " M L K " . I was personally offended - these sentiments reflect the polar opposite to those espoused by Dr . K i n g (Blood, 2005). To engage in debate about the offensiveness or "appropriateness" (read "truth") of certain tags is to undermine the conceptual foundations of social tagging. The point of social tagging is for individuals to have their say without interference and censorship of any kind. This freedom forms the basis for the socio-political potential and efficacy of social tagging. 20 The Socio-political Utility of Social Tagging The freedom of library patrons to interpret and represent content objects according to their points of v iew is social tagging's pr incipal ideological and socio-political contribution. In the previous section, it was established that f inding, management of personal collections, sharing of resources among peers, and interaction w i t h the information organization system were four purposes of social tagging systems (Good & Tennis, 2007). It was also argued that social tagging and folksonomies are not to be thought of as substitutions for traditional indexing work performed by professionals and indexing languages produced by them. Social tagging software may be added to already existing classificatory structures on online public access catalogues of libraries to enhance rather than replace the already offered services. Though management of personal collections, sharing of resources, and the ability to interact w i th information organization systems are al l interesting additions to online public access catalogues, they are not ends i n and of themselves, but rather means towards what I see as the greater, more significant, and critical purpose of patron participation i n indexing. N o w that we have the technological capabilities to a l low for publ ic participation, we must ensure its implementation. A s I argue, our current indexing languages are arbitrarily selected and ideologically f lawed, but as they are the only systems we currently have that meet the objectives of full-featured bibliographic systems such as collocation and precision and recall i n retrieval (Svenonius, 2000), then we must content ourselves w i t h supplementing 21 as opposed to supplanting our current classification schemes and indexing languages. 6 Folksonomies provide alternative access to traditionally organized materials. M o r e importantly, however, they are the only available access point for alternative materials that are not accessible through or h idden by traditional indexing languages (Quintarelli, 2005). The most socio-politically important feature of folksonomies is that they are inclusive: They include everyone's words and vocabulary without leaving anything out. There is no central authority imposing a top-down view, and every voice gains its space. This aspect that the power law trend imply that by using folksonomies, we can also discover long tail topics: original , non-mainstream ideas can emerge from the interest of a small fraction of the populat ion to the attention of the mass (Quintarelli, 2005,11). Folksonomies are manifestations of user-generated aggregation of information, w h i c h a l low taggers to reclassify and regroup according to their o w n needs and belief systems. They are inclusive, rather than authoritative. A second espoused socio-political advantage of folksonomies is their flat rather than hierarchical structure (Quintarelli , 2005). Though this is true i n some folksonomies, it is not necessarily the case i n al l of them. Hierarchies can and do commonly exist wi th in folksonomies (Koine, 2005; Tonkin , 2006; Voss, 2006): Decades of research into human cognition and categorization activities have found that categorization is a fundamental human cognitive activity, 6 A detailed discussion of this topic w i l l be provided i n the third chapter of this thesis. 22 examples of category systems exist across cultural and lingual differences, and they share numerous traits including hierarchical organization (Kome, 2005, 2). Socio-politically speaking, the advantage of folksonomies is ideological and not structural. The discovery and existence of hierarchies within folksonomies does not discount the fact that they are non-authoritative. Here, then, "bottom-up classification" does not refer to the classiiicatory structures, rather the way in which the folksonomies - hierarchical or not - come to be. Having a non-hierarchical classificatory structure is a fortuitous rather than essential characteristic of folksonomies. What is important is that folksonomies allow users to be active participants in the comprehension and organization of information: "It comes down ultimately to a question of philosophy. Does the world make sense or do we make sense of the world?" (Shirky, 2005). Traditional classification structures imply that the world makes sense. Users' understanding of a given piece of information must, therefore, be reconciled and altered to reflect that of classificationists. Folksonomies, on the other hand, allow their users to make sense of the world. They do not privilege "one top level of sense-making over the other" (Shirky, 2005). Social tagging is a step towards leveling the playing field of knowledge organization; performed by taggers, it has allowed the users of these information systems the possibility of utilizing non-authoritative folksonomies alongside traditional, authoritative indexing languages. In his 1994 text, Power in Movement: Social Movements, Collective 23 Action and Politics, Sidney Tarrow presents the anatomy of a social movement: collective challenge, common purpose, solidarity, and sustaining collective action. In the fo l lowing chapter, I argue for the conceptualization of social tagging as a social movement given Tarrow's four constructs. The proposit ion to al ign the characteristics of social tagging w i t h that of social movements highlights the socio-political potential of social tagging and is, therefore, we l l suited to the aspirations of critical l ibrarianship discussed i n the introductory chapter of this thesis. 24 C h a p t e r T h r e e T h e N e w O r d e r of S o c i a l T a g g i n g The organization of the world is a task for realists. The poet and the workman will always be victims of power and interest. The world will ever be run by a mystic idea, because by the time it begins to function it ceases to be mystical... The realist always conquers the poetic as the human. Interest wins out. - Anais Nin (1936, 1995 ed. 341-342) The Radica l izat ion of Indexing Social tagging is a specific kind of indexing (Tennis, 2006); given content objects such as books, websites, or photographs, taggers interpret the objects and use tags to represent them in information systems such as online public access catalogues (OPACs) of libraries, and social tagging websites such as ( and Flickr ( Even though social tagging is a type of indexing, it has transformed it significantly. How has this seemingly simple act of tagging content on the open Web revolutionized indexing? How are folksonomies radically different from other forms of known indexing languages? To begin, knowledge organization, as embodied by social tagging, is no longer limited by physical constraints. As far as taggers are concerned, content objects need not be represented in one place only. The digital organization of content objects allows indexers previously nonexistent freedoms. To use Anais Nin's characterizations, realism and poeticism can coexist; the 25 w o r l d may be mystically organized. What is more, this act of indexing performed by taggers has liberated the users of these information systems from having to contend w i t h a single, authoritative indexing language. Weinberger calls this progression the "third order of order" (2007,19), w i t h the first and second orders of order being the physical organization of items (i.e. on shelves) and their representation i n information systems (i.e. l ibrary catalogues) respectively. In libraries, catalogues separate the information about the items from the items themselves, whi le point ing to the physical space where they have been stored. Weinberger (2007, 22) argues: "Second order organization, it turns out, is often as much about authority as about making things easier to f ind." The way i n w h i c h items are interpreted and represented is dependant on whose wor ldv i e w has prevailed: Classification is a power struggle - it is political - because the first two orders of order require that there be a winner. The third order takes the territory subjugated by classification and liberates it. Instead of forcing it into categories, it tags it (Weinberger, 2007, 91-92, emphasis original). Weinberger's (2007) third order organization (social tagging) has changed the ways i n w h i c h we think about indexing i n general but also, and more importantly, it has given us the tools to counterbalance the wor ldv iews and limitations imposed by the authority of classificationists: A s Dewey 's biographer Wayne A . Wiegand writes, the organization of knowledge Dewey produced solidified "a wor ldv iew and knowledge structure taught on the Amherst College campus between 1870-1875" - a 26 w o r l d v i e w and structure that assumed the West was the most advanced culture and that Christianity la id the foundations of truth (Weinberger, 2007, 53). Though we may have always k n o w n and understood these limitations, we have lacked the capabilities to work around them. Once the exclusive domain of professional indexers, establishing "aboutness," to use indexing parlance, is now also carried out by taggers. In l ibrary indexing, the choice between traditional indexing languages and folksonomies (Weinberger's second and third order organizations) need not be mutual ly exclusive. Social tagging is qualitatively different; it has a separate purpose. In the second chapter of this thesis a set of four purposes of social tagging was offered: finding, management of personal collections, sharing of resources among peers, and interaction w i t h the information organization system (i.e. tagging) (Good & Tennis, 2007). I further argued that the freedom of l ibrary patrons to interpret and represent content objects in accordance w i th their points of v i ew is social tagging's pr incipal ideological and socio-political contribution. W i t h mult iple indexing contributions from taggers comes enhanced alternative access to content objects. For example, the book Henry and June: From A Journal of Love: the Unexpurgated Diary of Anais Nin has the fo l lowing subjects listed i n the Vancouver Public Library 's O P A C : • Nin, Anais, 1903-1977 - Diaries. • Miller, Henry, 1891- relations with women - Anais Nin. • Authors, American - 20th Century - Diaries. 27 These access points serve the needs of some patrons some of the time. They do not, however, represent the book to everyone's satisfaction - they never can. I, for example, w o u l d add "feminism", "erotica", and "June Mansfield M i l l e r " as tags for this particular book, highl ight ing otherwise ignored or h idden aspects of the document. Even w i t h those added tags, the document can st i l l be indexed further to a different person's or group's satisfaction. The point here is that the more people have the opportunity to tag this particular book, the more aspects of the book w i l l be represented. More index terms added by the l ibrary 's patrons are bound to represent additional points of view. No t only w o u l d Henry and June: From A journal of Love: the Unexpurgated Diary of Anais Nin be more fully indexed - and thus, one could argue, the representation w o u l d be more faithful to the mult i tude of relationships between the document and its readers - but it could also be accessed using a larger variety of search terms. Let us, for example, imagine that I, as a user of the Vancouver Public Library system, have just recently read about June Mansf ie ld M i l l e r i n a book called Twilight Years: Paris in the 1930s (Wiser, 2001). I have learnt that she was Henry Mi l l e r ' s second wife and his inspiration for the character " M o n a " i n Mi l l e r ' s 1934 novel , Tropic of Cancer. I also learnt that she attempted, but failed to destroy Mi l l e r ' s manuscript of Tropic of Cancer before it was published. She was a taxi dancer, and a w o m a n w i t h "a neck out of a M o d i g l i a n i paint ing" (Wiser, 2001,182). 7 I am fascinated by this w o m a n and w o u l d like to read more about her. Unfortunately, I f ind it 7 A taxi dancer is a paid, professional dance partner. 28 difficult to f ind documents about June Mansfield M i l l e r using the Vancouver Public Library ' s catalogue. I try several search strategies: • K e y w o r d Anywhere : June Miller • K e y w o r d Anywhere : June Mansfield • Subject Browse: Miller, June • Subject Browse: Mansfield, June • Subject Browse: Miller, Henry, 1891 - relations with women None of my conducted searches result i n documents containing information about June Mansfield Mi l l e r . A t the end, I remain unaware of Anai's N i n ' s Henry and June: From A Journal of Love: the Unexpurgated Diary of Anais Nin, even though the title character is the same June Mansf ie ld M i l l e r I have been searching for. It is not possible for professional indexers to highlight every aspect of every content object. It is however possible, w i th social tagging, to a l low space for interpretation and representation by others who do not normal ly participate i n this process. Tags added to records i n library catalogues w o u l d a l low patrons to seek, find, and evaluate information from alternative viewpoints, us ing alternative paths. These points of v iew w i l l , i n some cases, d raw attention to aspects of content objects that are intentionally or unintentionally h idden by the particular indexing language being used. Therefore, social tagging w o u l d enhance the information literacy of our patrons. A s discussed previously, information literacy is a socio-political sk i l l (Andersen, 2006). W h e n ut i l ized i n Onl ine Publ ic Access Catalogues of libraries, social tagging w o u l d enhance our patrons' socio-political sk i l l sets: 29 Social organization matters as the way documents and knowledge are organized i n society is a reproduction of the social and ideological organization of society. But it is not a passive reproduction. The organization of documents and knowledge i n society also shape society's social and ideological organization (Andersen, 2006, 222). Society's social and ideological organizations, as reflected by ways i n w h i c h knowledge is organized, privilege certain groups and oppress others. For example, evidence of patriarchal bias in library classification schemes is w e l l documented (Olson, 2002; Foskett, 1971; Berman, 1984; Palmer & Malone, 2001). In Power to Name: Locating the Limits of Subject Representation in Libraries, Olson (2002, 9) summarizes an extensive body of research conducted on the existence of sexism i n the Library of Congress Subject Headings, Dewey Decimal Classification, and Cutter Expansive Classification, and presents three general findings: they treat women as exceptions to a masculine norm; they ghettoize women's issues by separating them from the rest of knowledge; or they omit women's issues altogether. Olson (2002) further invokes the myth of Procrustes, the br igand "who possessed a hammer, a saw and a bed. H e compelled travelers to lie on the bed, and those who were too long for it he w o u l d cut d o w n to size; those w h o were too short he w o u l d hammer out unt i l they fit exactly" (Morford & Lenardon, 1995, 459). She does so to elucidate the absurdity of the quest of universality pursued by classificationists such as Melv i l l e Dewey and Charles Cutter. She maintains that the desire for universality is one of the most significant barriers to effective change i n subject representation and access. "The 30 universal language itself is constructed on principles that marginalize or exclude the abnormal or unusual deviations - the Other - from the singular public 's no rm" (Olson, 2002, 80, emphasis and capitalization original). Through the act of naming, she argues, librarians select what is represented and what remains unnamed and i n doing so, involve themselves i n the act of information construction or creation rather than information representation alone: N a m i n g is the act of bestowing a name, of labeling, of creating an identity. It is a means of structuring reality. It imposes a pattern on the w o r l d that is meaningful to the namer. Each of us names reality according to our o w n v i s ion of the w o r l d built on past meanings i n our o w n experience. Each of us creates our own structure through naming. N a m i n g is, therefore, not a random process even though it is varied (2002, 4). Tradit ionally, indexers choose from exclusive and finite vocabulary to describe the subject content of a document; are l imited to choosing one and only one indexing term for each concept, and are further constrained by pre-existing structures that define the relationships between indexing terms (Olson, 2002). What does Olson recommend we do? She proposes three principles to guide the development of techniques for change and subversion (verbatim): 1. M a k e breaches in the l imit; make it permeable, rather than defining it or constructing a new limit . 2. M a k e spaces, rather than f i l l ing them: the spaces are for the other to f i l l should she desire to do so. 31 3. Be dynamic; address the relevant discourses i n a given context: they must be reflexive, changing responsively over time and space i n the broadest sense (2001, 21, emphasis i n original). The principles presented by Olson (2001) are admittedly nebulous. However , they fit w i th in Olson's (2001, 2002) more general plea for inclusion of "the other" in areas of significant decision-making i n knowledge representation. She warns against defining, f i l l ing spaces, and constructing limits; she asks instead for permeability, allowance of space, and responsive change: Techniques that fol low these principles seem risky to l ibrary and information professionals steeped i n the tradition of the presumption of universality in naming. The reason for this dis-ease is that making space for the voice of the other means that one must rel inquish power to the other - power of voice, construction, definition. Instead of possessing this power exclusively, we who are on the inside of the information structures must create holes i n our structures through which the power may leak out. T w o groups of people are obvious others, just on the other side of the l imit : authors and library users (2001, 22, emphasis i n original). Olson (2001) further argues that our discipline of Library and Information Science has, for some time now, been involved in discussing and researching the possibili ty of making information available through the natural language of authors. What is more complicated and, i n this case, far more interesting is the mak ing of space for the participation of l ibrary users. Olson (2001, 22) imagines 32 paths of a garden where subsequent visitors may choose to travel i n the footsteps of others w h o have come before them. She writes: Library users w o u l d be a l lowed to create paths through indexes and catalogues that other users could then follow as far as they found them useful or interesting. Instead of the one having power, power could be taken by one or the other - accompanied by responsibility" (2001, 22, emphasis i n original). Olson (2001) does not elaborate on the nature of the responsibility she speaks of, but one can assume that she is referring to the responsibility of the participant wr i t large. N o r does she give specific details about the ways i n wh ich this "garden" may be created other than to speak about "some k i n d of hypertext facility for creating connections betweens items in a database" (2001, 22). It w o u l d not be unwarrantedly hopeful to argue that the tone and language used by Olson (2001, 2002) is i n line w i t h current discussions concerning the deployment of social tagging in l ibrary catalogues (Quintarelli , 2005; Winget, 2006) reviewed i n the first chapter. Currently, there are no limitations imposed on taggers by controlled vocabularies or tag quantities. Taggers can choose their tags freely and use as many tags as they see fit to represent each content object. Furthermore, relationships between content objects are created spontaneously as other taggers use the same tags to represent other content objects. There are no pre-determined structures wi th in folksonomies. Unstructured and structured indexing languages serve different functions. Intentionally disparate from the serendipitously created paths anticipated by Olson (2001), traditional indexing 33 languages provide a needed service: collocation and h igh degrees of precision and recall i n retrieval. The cost of this procedural efficiency is ideological. Such indexing languages, through the use of controlled vocabularies, work to create and maintain the status quo: "These devices function l ike D N A ; they enable the current system to replicate itself endlessly, easily, and painlessly" (Delgado & Stefancic, 1989, 208). Social tagging w o u l d al low us to be cognizant of beaten paths and, insofar as we may, work around them. 8 To summarize, social tagging has indeed revolutionized indexing, and folksonomies are radically different from other forms of k n o w n indexing languages. What they have offered does not amount to significant quantitative differences in interpretation and representation of content objects, such as more efficient results lists. The change is qualitative. A d d i n g social software to l ibrary O P A C s and enabling social tagging is a socio-politically significant decision, w h i c h challenges the legitimacy of traditional indexing languages and enhances the information literacy - the ability to seek, find, and evaluate information - of 8 Tangentially, i n light of Olson's denunciation of universality and sameness as desirable goals, it is important to note that her discussion of the existence of sexism i n library classification schemes is used here as a case i n point and not as a catchall i l lustration of the systematic treatment of other "other" categories (i.e. race, class, sexual orientation) w i th in traditional indexing languages or classification schemes. Us ing the existence of sexism to illustrate all other exclusions or misrepresentations w o u l d be akin to the process of marginalizat ion i n subject representation, wh ich we have thus far condemned. Sexism refers to a very specific form of discrimination. It does not assume other forms of prejudice and should not be pr ivi leged over them. A n indexing language or classification scheme may not be sexist, but stil l be biased i n other ways. We w o u l d not want to lie i n our o w n Procrustean bed. 34 l ibrary patrons, rendering social tagging not only a particular k i n d of indexing, but also a type of social movement. Library and Information Science's conceptualization and commitment to the ideological import of social tagging is related to the discipline's role as a socio-politically relevant profession and field of study. Social Tagg ing as a Social Movement Reasons for an individual ' s participation in social movements are numerous. A m o n g them are group interaction and solidarity, personal gain, and pr incipled, genuine commitment to a cause (Tarrow, 1994). This diverse list of motivators makes it difficult to predict future participation or exercise any meaningful levels of control over the behaviours of current participants. Though inconsistently secured, people's desire to participate i n collective action does to some degree depend on social networks, shared cultural understandings, and the availabil i ty and strength of external - often polit ical - opportunities (Tarrow, 1994). W h y do people engage i n collective action? In the case of social tagging, w h y do taggers tag? It is important to ask about motivation because although the result w i l l be a heterogeneous list, it is useful to gain some understanding of the parameters. A r e taggers tagging for personal purposes alone or is the social aspect of social tagging important to them? Do taggers who tag on library O P A C s such as the A n n Arbo r District Library (ht tp : / /www.aadl .org/cata log) and Univers i ty of Pennsylvania Library (http:/ / tags.l do so 35 because they understand the ideological import of their actions, or is dis turbing the status quo and prov id ing others w i th alternative access points an opportune byproduct of their actions? A t the time of wr i t ing this chapter, I am not aware of any research projects conducted on taggers' motivations. This is an open research question w i t h significant implications regarding the commonali ty of purpose and solidarity in social tagging as a social movement and its viabi l i ty as a complement to traditional indexing. Currently we do not know w h y taggers participate i n social tagging and once we ask, it is l ikely that there w i l l be many different reasons. W i t h that in m i n d and for the time being, the fo l lowing paragraphs w i l l focus on the fact that taggers are motivated, and they do participate, and that the results of their participation is socio-politically significant. Whether taggers are tagging content objects i n library catalogues because they are aware of the subversive nature of their actions remains to be studied. What is of note at this time is that their contributions have la id bare the inadequacies inherent i n traditional indexing languages. Un l ike traditional indexing languages, folksonomies are not restricted by exclusive and finite vocabularies and structural limitations; rather they are flexible and relevant to the needs of taggers w h o create them. Social movements are defined by contentious collective action and their formation occurs when polit ical opportunities arise for social actors who, generally speaking, lack them: "Movements ," Tarrow argues, "are better defined as collective challenges by people with common purposes and solidarity in sustained 36 interaction with elites, opponents, and authorities" (1994, 3-4, emphasis i n original). Though both contention and collectivity come i n many forms, they are the shared components among social movements. A t a very basic level, social movements are a group of people mobi l iz ing against a more powerful agent. Extreme or violent action, often associated wi th social movements, is not a necessary part of contention; rather such movements are manifestly contentious because they are the collective's most significant, if not only, recourse against their more dominant adversaries (Tarrow, 1994). The collective aspect of social movements is self-explanatory: the challenges faced affect groups of people rather than individuals . In his 1994 text, Power in Movement: Social Movements, Collective Action and Politics, Sidney Tarrow provides the basic properties of social movements: collective challenge, common purpose, solidarity, and sustaining collective action. Social tagging, I w i l l argue i n this chapter, meets the criteria presented by Tarrow (1994) and may be reasonably viewed as a social movement. Collective Challenge Collective challenge is the first of four basic properties of social movements enumerated by Tarrow (1994). It is an example of collective action mounted by social movements and is characterized by "interrupting, obstructing, or rendering uncertain the activities of others" (Tarrow, 1994, 4). Collective challenges are not the only type of collective action ut i l ized by social movements; 37 rather, they are quintessential. Social movement groups employ disruptive and challenging action, as they do not, generally speaking, have access to financial resources and socio-political platforms required for other forms collective action, such as lobbying and negotiation. It can be argued that social tagging is an example of collective challenge against traditional indexing practices. W h e n enabled i n library O P A C s , social tagging has the capacity to interrupt, obstruct, and render uncertain the dominance of top-down, authoritative indexing languages. W h i c h properties of traditional indexing languages make them suitable targets of social tagging as a collective challenge? Social tagging as a particular k i n d of indexing is, i n essence, postmodern whi le traditional indexing and indexing languages are based on a modern v i ew of the wor ld . M a i makes the fo l lowing distinction: Classic tradition of classification theory is based on a modern v iew of the wor ld . This includes the idea that classifications can be a neutral and objective mirror of an already there universe of knowledge. A postmodern theory of knowledge organization rejects this assumption and instead places focus on the social praxis and the language of the communi ty for wh ich knowledge organization is created (1999, 548). M o d e r n indexing languages are assumed to be descriptive and objective, whereas postmodern indexing regards objectivity and neutrality as an unattainable goal. In this context, the point of divergence between modernism and postmodernism is postmodernism's commitment to the social constructionist v iew of knowledge. The shift between modern and postmodern 38 perspectives on indexing w o u l d necessarily include a change "from a (intended) reflection of the universe of knowledge to a pragmatic tool i n the mediat ion between author and user" (Mai , 1999, 554, parenthesis original). For postmodernists, then, all indexing is wrong, but some indexing is useful. 9 A postmodern v iew of indexing challenges the objectivity of the practice. It holds that a given indexing language is not a "true" reflection of content objects and their relationships to each other rather a particular v iew of such objects at given times by given persons. Folksonomies, as indexing languages, are openly and manifestly postmodern and confess their social construction. They, therefore, posit a challenge to modern indexing languages. In his 1999 text, The Social Construction of What? Ian Hack ing provides a framework of social construction, w h i c h I use here to examine indexing languages as socially constructed phenomena. H e further lists six gradations of commitment w i th in the constructionist framework, implici t wi th in w h i c h are six grades of response categories. Faced w i t h M o d e r n Indexing Language X , social constructionists w o u l d ho ld that: 1. M o d e r n Indexing Language X (MILX) need not have existed, or need not be at al l as it is. M I L X , or M I L X as it is at present, is not determined by the nature of things; it is not inevitable. Often the social constructionist w o u l d go further, and argue that: 2. M I L X is quite bad as it is. 9 A homage to statistician George E . P. Box's famous quotation: " a l l models are wrong, but some models are useful." 39 3. W e w o u l d be much better off if M I L X were done away wi th , or at least radically transformed (Hacking, 1999, 6). In a physical sense, it is obvious that M o d e r n Indexing Language X is socially constructed. Hack ing (1999) argues that emphasis must be placed on the difficult distinction between object and idea. Obvious ly objects (in this case, indexing languages) are not inevitable, whereas the idea of indexing languages as objective and accurate representations of the w o r l d can be v iewed as something that is unavoidable. It is, therefore, necessary to be explicit that by using Hacking ' s stipulations, I am referring to the ideological structure of indexing languages and not indexing languages as physical objects created by classificationists and indexing committees. What is being argued here, is: the modernist discourse holds that indexing languages are structured the way they are because they are objective tools, wh ich mirror the w o r l d of knowledge as it exists i n "reality." Modernists take the ideological structure of traditional indexing languages for granted; they v i ew it as an inevitability given the structure of knowledge i n the "real w o r l d " . In response to establishing that M o d e r n Indexing Language X is socially constructed, a social constructionist w i l l have different recommendations and reactions to the three statements mentioned above, resulting i n the six grades of constructionism: 40 1. His tor ical : A social constructionist presents a history of M I L X and argues that M I L X has been constructed i n the course of social processes. Far from being inevitable, M I L X is the contingent upshot of historical events. A historical constructionist could be non-committal about whether M I L X is good or bad. 2. Ironic: M I L X , wh ich we thought to be an inevitable part of the w o r l d or of our conceptual architecture, could have been different. We are nevertheless stuck w i t h it, it forms part of our way of thinking, w h i c h w i l l evolve i n its o w n way, but about wh ich we can do nothing much right now. 3. Reformist: This grade of constructionism takes the second step, M I L X is quite bad as it is, quite seriously. W e should do something to change some aspects of M I L X i n order to make M I L X less pernicious. A n example of a reformist step towards reducing the harm caused by M I L X is to submit new indexing terms to the committee in charge of M I L X . 4. Unmasker: Once one sees the extra-theoretical function of an idea, it w i l l lose its practical effectiveness. M I L X is unmasked not so much to disintegrate it, but to strip it of its false authority and appeal. A n unmasker may or may not be a reformist and vice versa. These two categories are placed on the same plane. A n unmasker w o u l d be interested i n exposing M I L X and removing its epistemic authority. A t this stage i n the development of indexing languages and given that social 41 tagging and traditional indexing languages have been argued to serve different purposes, the position of this thesis w o u l d be that of a unmasker social constructionist. 5. Rebellious: A constructionist who actively maintains that M I L X is not inevitable, M I L X is bad, and we w o u l d be better off without M I L X . A rebellious constructionist is committed to all three steps on an intellectual level. For example, a rebellious social constructionist w o u l d wri te a paper recommending social tagging as a replacement for M I L X . 6. Revolutionary: A n activist constructionist who moves beyond the w o r l d of ideas and tries to change the w o r l d in respect to M I L X is a revolutionary. A revolutionary social constructionist, for example, w o u l d be a l ibrarian who introduces social tagging on her l ibrary's O P A C and removes al l other indexing terms (Hacking, 1999,19-20). In the last two steps of social constructionism presented by H a c k i n g (1999), there is no desire to work wi th in , rework, revise, or improve existing indexing languages; nor is there an aspiration for creating newer models and enhanced replicas. There is no commitment to maintain any of our presumptions or to work wi th in our existing frameworks. The challenge, wh ich is currently mounted on library O P A C s by social tagging against traditional indexing languages does not, at this time, fall w i th in Hacking 's (1999) last two steps of social constructionism. Social tagging added as a feature alongside traditional 42 indexing exemplifies the coexistence of postmodernist and modernist indexing practices. Due to this coexistence, the social constructionist level of commitment remains at Hacking 's (1999) fourth step - unmasker. To summarize, social tagging meets Tarrow's (1994) first basic property of social movements - collective challenge. Social tagging interrupts, obstructs, and renders uncertain traditional indexing practices. M o r e specifically, folksonomies, as postmodern indexing languages, challenge the dominance of modernist, traditional indexing languages. A s this challenge is brought about by large numbers of participants, social tagging is a collective challenge. Common Purpose & Solidarity C o m m o n purpose and solidarity, the second and third properties of social movements, w i l l be discussed together as solidarity among a group of people is the result of their acknowledgement of common purposes and interests. "It is participants' recognition of their common interests that translates the potential for movement into collective action" (Tarrow, 1994, 5, emphasis original). W h i l e certain social movements have been marked by l ively demonstrations or mob-l ike behaviours, spectacle is not often the primary purpose or a necessary part of social movements. Rather, the group seeks "to mount common claims against opponents, authorities or elites" (Tarrow, 1994, 4-5). Effective leaders of social movements are often those who emphasize the ties that exist among the participants. The distinction between unacknowledged and acknowledged 43 commonalities in purpose and interest is important because though unacknowledged commonalities may exist, they do not translate into collective action, w h i c h is needed for social movements. H o w are unrecognized commonalities elucidated? Tarrow argues that engaging i n "framing work" is necessary for the mobi l iza t ion of social movements. H e offers the fo l lowing explanation of "framing work": .. . it is no simple matter to convince t imid people that the indignities of everyday life are not written i n the stars and can be attributed to some agent, and that they can change their conditions by taking action collectively. Inscribing grievances i n overall frames that identify an injustice, attribute the responsibility for it to others and propose solutions to it, is a central activity of social movements (1994,123). Furthermore, framing work involves the bu i ld ing and advocating of a discourse of injustice, wh ich names and defines the targets of the collective act ion. 1 0 Most aware of the limitations of traditional indexing languages are librarians and scholars i n the field of Library and Information Science. They can choose to partake i n framing work and become active participants i n the mobi l izat ion and maintenance of social tagging as an alternative k i n d of indexing and a social movement. They can choose to engage i n critical librarianship and assume the social and polit ical responsibilities inherent w i th in their profession (Andersen, 1 0 The work of authors such as Clay Shirky and David Weinberger may be seen as "framing work" in relation to Web 2.0 technologies writ large. The work, which is required now, is similar contributions in relation to the implementation of social software on library OPACs. 44 2005). This participation may take the form of wri t ing, action research, discussions, formal lectures, advocacy for social tagging or any other forms of communicat ing and exposing the current state of affairs in indexing. [For] social movements are deeply involved i n the work of "naming" grievances, connecting them to other grievances and constructing larger frames of meaning that w i l l resonate w i t h a population's cultural predispositions.. . (Tarrow, 1994,122). A s is the case w i t h established social movements, ind iv idua l taggers have different reasons for tagging. However those aware of social tagging's potential as a collective challenge can be instrumental in br inging about the recognition of a common purpose and solidarity. For example, a tagger w h o tags pr imar i ly for personal reasons may also tag for reasons related to tagging as a collective challenge against the indexing language used i n her public library. Personal motivations and common purpose w i t h i n social movements need not be mutual ly exclusive. G i v e n the historical endurance and preeminence of authoritative indexing languages, framing work i n favour of social tagging w o u l d not be easy or uncontested. Opposi t ion is l ikely from both librarians and the public. Tarrow (1994) discusses two ways i n w h i c h resisted mobil izat ion can overcome such obstacles. The first is the s low process of formation and mobil izat ion of consensus. Consensus formation is the passive and eventual acceptance of change, wh ich takes place when o ld and new social structures coexist for a prolonged periods of time. Consensus mobil izat ion is the active promotion of 45 that change through agents such as the media. Both passive and active consensus bu i ld ing require less organizational effort than the second option of collective action, but are far slower i n br inging about change. For example, without framing work on the part of librarians and scholars the adoption of social tagging by libraries can take a long time, as this form of indexing may be v iewed as unnecessary in comparison to paradigmatic indexing languages. G i v e n enough time and w i th the populari ty of social tagging on non-library websites, libraries may come to accept and adopt social tagging even without the framing work and encouragement of supporters. However , as Tar row (1994) argues, the realization of the process of change w i l l be delayed. In situations where more immediate change is desired, framing work and advocacy is required. For example, a l ibrarian interested i n introducing social tagging to her l ibrary 's O P A C may propose a tagging trial period to her l ibrary's board in order to promote the concept wi th in the institution. Due to the weight and credence of current cultural practices and understandings, the framing of new movements is often difficult, yet unl ikely movements are constructed and can be successful given that movement leaders are able to take advantage of presented opportunities (Tarrow, 1994). In social tagging, for example, librarians can take advantage of technological advances in social software, public 's demonstrated interest i n non-library sites such as Fl ickr and Library Thing, and the existing willingness of taggers to tag. Changes i n opportunity should be seized and expanded though framing work and collective 46 action if they are to be sustained and result i n socio-political change. G i v e n the lack of empirical data on motivational factors behind social tagging, it is difficult to discuss common purpose and solidarity amongst taggers. However , though it is true that we are currently not i n a posit ion to comment on the reasons w h y taggers engage in tagging, we can nevertheless attempt to sanction a strong sense of common purpose and solidarity given auspicious circumstances. A s discussed i n the first chapter of this thesis, l ibrarianship is not a socio-politically neutral profession, and there is no pretense of neutrality i n the arguments presented in this thesis. Librarians and scholars i n the field of Library and Information Science who are supportive of social tagging have a significant socio-political opportunity presented to them. The technology is available and the crowds are wi l l i ng . A s contended earlier, social movements are formed when poli t ical opportunities arise for social actors who lack them. Social tagging is a chance for the public to be active participants i n the interpretations and representation of information. This occasion is, i n m y opinion, a ra l ly ing cry par excellence. Sustaining Collective Action Social movements require sustained collective action. "The social movements that have left the deepest marks on history have done so because they sustained collective action against better-equipped opponents" (Tarrow, 1994, 6). Though there is a tendency to compare social movements to 47 contentious episodes such as riots, this analogy is not always accurate. Most riots do not translate into social movements because rioters may or may not have a common purpose and may disband as abruptly as they emerge. The tendency to a l ign social movements w i th riots may arise from the desire to d imin i sh the sense of purpose of the former by point ing to the capriciousness conjured by the latter. For example, those opposed to social tagging on library O P A C s may use this characterization of tagging as a social movement to point to it as an unreliable and short-lived trend. It, therefore, is necessary to emphasize that social movements, by definition, require sustained collective actions. Movements are sustained both by external forces such as movement organizers and internal forces such as social networks. The role of movement organizers was discussed in the previous section i n relation to the cultivation of a common purpose and solidarity amongst taggers. To the extent that organizers and those involved i n framing work are instrumental in br inging about collective action, they also play an important role in its continuation. The second factor i n the sustainability of social movements is the social network. Onl ine social networks have considerable power as motivators and enforcers of cooperation (Donath & boyd, 2004). They are analogous to real-life social groups i n myr iad ways, as by using social ne tworking sites, users are able to create descriptive profiles and invite others (or are invi ted by others) to connect w i t h them digitally. These connections take many forms: leaving comments on a user's profile (Fl ickr) , adding another user 48 to your "watch list" (LibraryThing), "pok ing" someone or asking them to be your friend (Facebook), etc. The point is that "people are accustomed to thinking of the online w o r l d as a social space" (Donath & boyd, 2004, 71). Belonging to social networks online has similar implications as real-life situations: users can display their connections w i t h others in the same way they w o u l d if they hosted a party, other members can infer certain information about a user's poli t ical leanings, taste i n literature, music, food, etc. from their profiles but also from their online network of friends (Donath & boyd, 2004). Similar levels of profile sharing and online networking may be available to taggers on library O P A C s . For example, as a tagger on the Vancouver Public Library O P A C , I w o u l d be able to create a personal profile, w h i c h tells other taggers or browsers information I choose to share w i t h them: my user or real name, books I have borrowed, books I am reading right now, books that I have tagged. Other users w o u l d infer a certain level of trust or distrust i n my tags from other information that I have p rov ided to them. For example, they may trust tags provided by taggers w i t h s imilar taste i n literature. In this case, as w i t h other manifestations of social signaling, the accuracy of conclusions d rawn may be questioned. The more significant point is that users are given the choice to use this socially provided information to remain engaged and to make decisions related to their requirements at any given time. Onl ine social networks mirror other social networks i n some ways, but surpass them i n terms of convenience. U s i n g the Internet to keep i n touch may 49 be less personal but it is easier. I w o u l d , for example, be far more l ikely to keep up w i t h m y membership i n a particular social movement collective i f m y contributions could be made digitally. It is more manageable and more conducive to long term sustainability if I can participate on my o w n time, rather than having to make real time commitments to meetings and rallies. The simple fact that tagging is easy is imperative to its success as a sustainable social movement. Furthermore, tagging as a k i n d of information seeking behaviour, w o u l d not be a novel activity: "Humans have for al l times sought information i n order to support their activities" (Andersen, 2006, 217). The fact that tagging is, to some extent, a manifestation of an everyday human activity - information seeking - adds to its qualification as a sustainable collective action. The already inherent elements wi th in social tagging, such as its uti l ization of social networks and convenience of use, favour its candidacy as a tool for sustained collective action and social change. C o n c l u s i o n Folksonomies, performed collectively by the members of the public, are contrary to the ethos of traditional indexing languages and a challenge to their preeminence. Whether or not taggers can be described as sharing a common purpose is, as of yet, undetermined. However solidarity - the recognition of a common purpose - can be harnessed by scholars and librarians w h o are interested in the mobil izat ion of social tagging as a social movement through 50 framing work and other forms of advocacy. Librarians and other interested parties i n LIS are faced w i t h an unprecedented socio-political opportunity: through the ut i l izat ion of online social networks and given the ease and convenience of tagging, social tagging once introduced, can be effectively sustained on Onl ine Public Access Catalogues of our local libraries. Andersen (2005) argues that knowledge organization systems (of wh ich O P A C s are an example) are "the professional tools of librarians. Due to this fact, we should expect that librarians have a lot to say about the roles and doings of these systems i n the mediation of society and culture.. . ." Advocacy for social tagging capabilities on library O P A C s is a significant step towards counterbalancing the documented lack of critical l ibrarianship i n LIS (Andersen, 2005; Andersen & Skouvig 2006). The selection of a socio-politically engaged paradigm for the understanding of social tagging on library O P A C S - the specific focus of this thesis and the emphasized area of overlap between social tagging and social movements in Figure 1.1 - is needed to realize the full potential social tagging as a social movement. The fol lowing chapter w i l l propose an anarchist - and specifically Kropotkinis t - paradigm because by choosing an anarchist paradigm, librarians and LIS scholars engaged i n critical l ibrarianship can make certain that social tagging continues to maintain and develop its pr incipal socio-political contribution: the freedom and agency of library patrons to interpret and represent content objects according to their points of view. 51 Chapter Four The Structure of an Anarchist Paradigm There are periods in the life of human society when revolution becomes an imperative necessity, when it proclaims itself as inevitable. New ideas germinate everywhere, seeking to force their way into the light, to find an application in life; everywhere they are opposed by the inertia of those whose interest it is to maintain the old order; they suffocate in the stifling atmosphere of prejudice and traditions. - Peter Kropotkin (1880, 2002 ed. 35) Introduction The future of social tagging i n libraries is at best uncertain; at worst, librarians, classificationists, and scholars w i l l reject it as either unworthy of serious consideration or a dangerous idea to be deftly evaded by those guarding the intellectual integrity of traditional indexing practices and, by extension, our discipline. W h y w o u l d the integrity of Library and Information Science be seen as threatened by public participation i n library indexing? Can social tagging be regarded as a new order of order without imposed, authoritative rule or is it fated to be condemned as disorder? C a n social tagging add value to the ways i n w h i c h we currently represent content objects i n online publ ic access catalogues of libraries or is it an unacceptable aberration i n the long tradition of l ibrary indexing? Therein lies the difference between the acceptance and rejection of 52 social tagging as a Panacean cure or a Pandoran threat, respectively. 1 1 We are at a critical juncture i n our understanding of social tagging; the ways i n w h i c h we choose to frame, study, and discuss public participation in l ibrary indexing n o w w i l l have a direct and significant impact on its future uti l ization or imminent disregard. In addition, it is not enough to discourage and refrain from direct attack on social tagging, rather we must also ensure that librarians and LIS scholars are actively engaged w i t h ideas such as social tagging - ideas that are of value to the pursuit of critical librarianship. The socio-political potential of social tagging i n libraries is far too significant for it to be consigned and relegated to non-library indexing alone. Furthermore, as an issue of considerable socio-poli t ical consequence, it is fitting to seek a polit ically informed paradigm for its consideration and study. Paradigm - K u h n i a n Author Michae l Ondaatje describes storytelling as the process of shining a spotlight on a dark stage. A t any given time, the narrator has the power to i l luminate specific aspects of the story and to conceal others. There may not come a time when the stage is lit in its entirety and some parts, though present, 1 1 To borrow from the title of the 17 t h annual Special Interest Group/Class i f ica t ion Research (S IG/CR) Workshop: "Social Classification: Panacea or Pandora?" presented at the A n n u a l Meet ing of the Amer ican Society for Information Science and Technology (ASIS&T) i n Aus t in , Texas. 53 are never revealed. 1 2 Spotlights and paradigms are both selective narration tools; they guide our attention. Since the 1962 publication of Thomas S. Kuhn ' s The Structure of Scientific Revolutions, it is common to read about paradigm-directed research in the scholarship of conventionally disparate disciplines. O u r o w n field of Library and Information Science has summoned his name and contributions i n discussions on the establishment of research frameworks (Ellis, 1998; Hjor land 1998). Here, fo l lowing a brief overview, I base the parad igm part of the proposed anarchist paradigm on the Kuhn ian characterization of the concept. K u h n describes paradigms as gu id ing and even determining the directions of research. D u r i n g periods when scientists are involved i n solving the problems of their paradigms, referred to as "normal" science, the values of that particular paradigm shape the scientists' o w n sense of "reality" and "truth." A n y given scientific community, then, operates under a set of wel l-guarded beliefs and assumptions. These core assumptions are maintained and communicated to future generations through education. K u h n writes: "because that education is both rigorous and r ig id , these answers come to exert a deep ho ld on the scientific m i n d " (1962, 5). Furthermore, there are interrelations between the "descriptive and normative dimensions of such communit ies" (Bernstein, 1983, 78). Whi le descriptive guidelines provide epistemological tools wi th in such communities, normative values have a regulative impact on scientific activity, research, and inquiry. Whi le involved i n research, or as K u h n 1 2 This description was given by Ondaatje dur ing a live interview w i t h H a l Wake at Christ Church Cathedral i n Vancouver, Bri t ish Co lumbia on M a y 3, 2007. 54 sees it the "strenuous attempt to force nature into the conceptual boxes suppl ied by professional education" (1962, 5), scientists doing "normal" science tend to overlook smal l anomalies. Nevertheless, there comes a time when particular anomalies can no longer be ignored because they undermine the core suppositions of the paradigm. They become problematic enough that a change i n assumptions becomes necessary. This change or "shift i n professional commitments" (1962, 6) ushers a paradigm shift. Once the shift has taken place, the path to a new normal science begins again. The paradigmatic execution of science does not concern K u h n . In fact, "the successive transition from one paradigm to another v i a revolution is the usual developmental pattern of mature science" (Kuhn, 1962,12). Paradigms are useful to scientific inquiry because they al low scientists to agree on fundamental assumptions of their fields, making both pure and applied science possible. They provide frameworks w i th in wh ich to ask questions and seek answers and provide a myr iad of mop-up work, al l w i th in the paradigmatic framework. Paradigms are of great worth to scientific inquiry because as K u h n writes: "no natural history can be interpreted in the absence of at least some impl ic i t body of intertwined and methodological belief that permits selection, evaluation, and cri t icism" (1962,17). Paradigms help shape scientists' questions, methods, and answers as without them "a l l facts that could possibly pertain to the development of a given science are l ikely to seem relevant" (Kuhn, 1962,15). However , operating wi th in a paradigm prevents researchers from being 55 cognizant of the whole picture - the "truth". G i v e n that there is no human alternative to operating wi th in paradigms, the "truth" remains at large. Though Kuhn ' s contributions are about the sciences specifically, they are invaluable to other disciplines of study. Since 1962, Kuhn ' s work has problematized claims of neutrality in research frameworks wri t large. It is commonly understood that through paradigms, research becomes value-laden. By selecting a particular paradigm, researchers choose to be taught, learn, and espouse certain viewpoints and disregard others; it involves va lu ing one explanation over another. Traditionally, the physical sciences were understood to provide a God ' s eye v i ew of the w o r l d and the social sciences were seen as str iving to get as close to this posit ion as possible. W i t h the foundations, assumptions, and practices of the physical sciences under scrutiny and spotlight, the biases inherent i n the social sciences are even further pronounced. For our stated purpose of the study of social tagging, what w o u l d this proposed anarchist paradigm look like? A s Chomsky writes, there have been "many styles of anarchist thought and action. It w o u l d be hopeless to try to encompass al l of these conflicting tendencies in some general theory ideology" (1970, vi i) . A singular theory of anarchism does not exist. That being said, the ethos of the anarchist movement, i n general, together w i th that of social tagging, i n its current form, demands further exploration. For what brings all threads of anarchist thought together is their denunciation of external authority (Ward, 2004, 3). N o t only are there different anarchist schools of thought, but also each 56 thread, each concept, and each movement wi th in the movement has been "reinvented or rediscovered continually" (Ward, 2004, 31). This evolution is a sign of relevance, a l lowing for its continual application without anachronistic reservations. M a r k Leier writes: The anarchist critique of the state, of capital, of power, is a compel l ing one, and the lesson of anarchism is constantly relearned through experience: people who do not benefit from the system w i l l organize to create alternatives (2006, xiv) I believe that the above quote highlights anarchism both i n content and form. N o t only are critiques of state, capital, and power the contents of anarchism, but also the ideology is expressly carried out by people who ascribe to anarchism's critical content and who, not benefiting from the system, organize to create alternatives for themselves and others. Us ing anarchism - both content and form - as a point of departure, it is important to make a parallel dist inction i n its application to social tagging. Whi le I see social tagging as anarchistic in its inherent rejection of authority and the way i n w h i c h it is carried out on l ibrary O P A C s , the choice of paradigm does not necessarily extend to taggers' poli t ical inclinations. The critical point in social tagging, the point that makes it anarchistic, is that tags are not monitored or controlled by an external, imposed authority. Taggers are free to tag content as they see fit even if their tags are ideologically opposed to other taggers' contributions or to anarchism. Tradit ional indexing languages are generally speaking racist, sexist, and classist 57 (Olson, 2002) and as such, it is l ikely that tags contributed by patrons on l ibrary O P A C s w i l l generate folksonomies that are poli t ically and ideologically opposed to these prejudices, however, this prediction is not a necessary part of the consideration of social tagging as an anarchistic social movement. Regardless of the ideologies represented by traditional indexing languages, social tagging can work to undermine their authority. The strength of social tagging does not lie w i t h the content of the tags per se but rather w i t h the fact that it is carried out from the bottom upwards by means of patron contributions and not from the top downwards by means of authoritative rule. What is important is that the tags provide alternative access points different from traditional indexing terms and a l low space for viewpoints that are, generally speaking, unrepresented. The lack of emphasis placed on the content of tags and folksonomies may seem contradictory given the contentual critique of traditional indexing languages presented i n this thesis. This inconsistency can, however, be addressed given the way i n wh ich each act of indexing is carried out. Consider, for example, a situation where a group of ten friends are ordering dinner at a restaurant. If one friend assumes the authority of ordering the same dish for everyone present, then the k i n d (content) of the food being ordered is significant and open for critique and discussion. Alternatively, if each person were given the opportunity to choose and order his or her o w n meal, then a critique of each ind iv idua l d ish w o u l d be nonsensical. A l though the most significant differentiating factor 58 remains the authoritative vs. non-authoritative form of ordering, the former al lows for a critique of content, whi le the latter does not. Paradigm - Anarchist - Kropotkinist In the previous chapter, using Sidney Tarrow's four basic properties of social movements - collective challenge, common purpose, solidarity, and sustaining collective action - from his 1994 text, Power in Movement: Social Movements, Collective Action and Politics, I argued for the conceptualization of social tagging as a social movement. Here, I propose that not only is social tagging an anarchist social movement by nature, but that it can remain so by nurture. The two positions - first: social tagging is an anarchist social movement and second: it can continue to function as such - are related but distinct. Currently, social tagging is formulated and conducted without the interference of authoritative, central governance such as the Library of Congress and, as such, it is an anarchist undertaking both i n theory and praxis. In fact, social tagging is proof that "organization without government [is] both possible and desirable" (Ward, 2004, l ) . 1 3 For future considerations, the use of an 1 3 In Anarchism: A Very Short Introduction, Colin Ward (2004) describes anarchist Pierre-Joseph Proudhon's position: "Proudhon argued that organization without government was both possible and desirable." I use this quote about anarchism in relation to social tagging to highlight that, interestingly, I have found descriptions of anarchism and social tagging to echo each other in the literature of their respective fields. In response to common concerns about chaotic and ill-intentioned tagging practices on social tagging sites, Winget writes about taggers' motivations and behaviour on Flickr: "the strength of the social relationships of players within the system provides a mechanism for ambient and informal policies and procedures, which although not official or strict, perform the function of enforcing appropriate tagging behavior" (2006,15). 59 anarchist paradigm w o u l d have implications regarding both scholarly research and practical implementations of social tagging in library settings. A s far as social tagging is an anarchist practice, it is useful to understand it as such, but for reasons beyond practicality - reasons of ideology and paradigm choice - it is further recommended that we do so. By choosing an anarchist paradigm for the study of social tagging, librarians and LIS scholars can make certain that social tagging continues to be maintained, developed, and studied as an anarchist social movement. To summarize, the fact that social tagging, i n its current form, has anarchist characteristics does not necessarily justify the adoption of an anarchist paradigm for its future study, rather an anarchist paradigm is suggested because it w i l l work to preserve the already existing advantages of social tagging discussed i n the first chapter - advantages such as inclusivi ty and flexibility. A s a gu id ing paradigm for research, anarchism w o u l d preserve and further advance social tagging as a socio-political tool and ensure that the interests of taggers are considered at every stage of development and implementation. A n anarchist paradigm of social tagging, for example, might preclude the introduction of tag lists, wh ich w o u l d either recommend or require taggers to choose from pre-approved tags. The authority assumed by people or committees responsible for creating controlled and acceptable tag lists w o u l d be rejected on anarchist grounds and the ethos of choice and voluntary participation emphasized. I believe that librarians and scholars i n the field of Library and Information Science w o u l d do we l l to look outside the traditional boundaries of 60 our discipline for more robust conceptualizations of this relatively new phenomenon; I propose anarchism as a viable choice. In reading discussions about social tagging, it is increasingly difficult to overlook the distinct and powerful ideological confluence between this emergent publ ic action and the theories of anarchism. No t only is social tagging anarchistic, but also "anarchy is a theory of organization" (Ward & Goodway , 2003, 62) wor thy of our collective and unremitt ing attention. M a r k Leier writes that the w o r d anarchism itself "has long been separated from its real meaning of ' rule by no one'" (2006, x). From the Greek anarkhia, anarchism's literal meaning of "wi thout a ruler" or "rule by no one" is thus employed here. Though this definition is often clearly stated in anarchist texts, there is also the need to clarify and defend anarchism against its more popular and misleading understandings: a socio-political ideology espousing violence and chaotic lawlessness (Guerin, 1970; Leier, 2006). This argument is we l l made by these authors and w i l l not be reiterated here. 1 4 To expand beyond the literal meaning of the w o r d , Peter Kropotkin ' s 1905 definition of anarchism w i l l be adopted for use throughout this thesis. In the eleventh edit ion of the Encyclopaedia Britannica, Kropo tk in wrote that anarchism is: 1 4 In Chapter 3, the inclination of a few to inaccurately equate social movements w i t h riots was discussed. The point made was that detractors, w h o w i s h to discount the significance of social movements, rely on its occasional correlation w i t h incidents of r ioting to make a causal and blanket statement: al l social movements involve rioting. A similar fallacy is at play i n l ink ing violence w i t h anarchism; though violence and anarchist actions sometimes coincide, violence is not a defining characteristic of anarchism. 61 The name given to a principle or theory of life and conduct under wh ich society is conceived without government - harmony i n such a society being obtained, not by submission to law, or by obedience to any authority, but by free agreements concluded between the various groups, territorial and professional, freely constituted for the sake of product ion and consumption, as also for the satisfaction of the infinite variety of needs and aspirations of a civi l ised being (In W a r d & Goodway , 2003, 25, spel l ing original). For our purposes here, anarchist Peter Kropotkin 's writ ings are used as illustrative of anarchist philosophy wr i t large, for the existence of varied schools of thought wi th in anarchism, precludes its discussion as one, unified whole. It is inevitable, therefore, that what w i l l be said here w i l l fall short of a satisfactory treatment of anarchism as a whole. However , by concentrating on one of anarchism's most significant theorists, something of value can nevertheless be contributed. In fact, Kropotkin 's contributions to anarchist thought are broad enough to include most anarchist tenets, even when they differ on smal l or specific points. Far from detracting from its usefulness, anarchism's many-thronged philosophy allows the theory malleability and relevance to common and practical situations. C o l i n W a r d writes that Kropotk in , "the most wide ly read on a global scale of a l l anarchist authors, l inked anarchism both w i t h subsequent ideas of social ecology and wi th everyday experiences" (2004, 8). W a r d further (2004, 31) argues that even though there exists a great deal of reinvention and evolution, it is possible sti l l to talk of an anarchist theory of 62 organization as consisting of the fo l lowing four characteristics: voluntary, functional, temporary, and small; these four points w i l l be revisited later in this chapter. A l t hough Kropo tk in devoted much of his life to science, the most significant of his contributions were to the anarchist cause. A s an anarchist and a revolutionary, Kropotk in is k n o w n as the father of anarchist-communism: Kropo tk in argued that anarchism and communism combined two currents of thought - radical l iberalism, w h i c h placed an emphasis on i n d i v i d u a l freedom and the negation of state (anarchism), and liberal socialism, which , stridently anticapitalist, opposed both private property and the wage system (communism) (Morris, 2004, 22, parentheses original). Broadly speaking, where the central issues of the anarchist-communism are property, land, natural resources and means of production (Ward, 2004), this thesis is concerned wi th the organization of knowledge as represented by indexing, and as far as naming can be seen as an act of ownership, knowledge as property. To some extent, then, social tagging on library O P A C s is an act of c la iming communal rights to the knowledge contained wi th in the library and its content objects; folksonomies are created by voluntary contribution and are used freely by patrons as they see fit. Cr i t ical l ibrarianship - the position of being cognizant and involved in the social and poli t ical responsibilities inherent w i th in the profession - can be informed by Kropotkin 's contributions because anarchist-communism is, first and foremost, preoccupied wi th action and not theory. Of 63 course, the tenets have to be communicated, and they often were through propagandist writ ings, but Kropo tk in wrote w i t h the intention of mobil izat ion (Cahm, 1989). Describing Kropotkin 's anarchist-communism, George Woodcock writes: W h e n Kropo tk in says that everything must return to the community , he does not mean this i n a vague and general way; he means specifically that it must be taken over by the commune. This is a term familiar enough to the French, w h o m he was pr imar i ly addressing; it describes the local unit of administration that is nearest to the people and their concerns, but it also carries revolutionary connotations of the Paris communes of 1793 and 1871 (2004,167-168, emphasis original). Similar ly , social tagging returns the holdings of libraries to their patrons both i n a figurative and literal sense. B y tagging the content objects, the patrons reclaim them symbolical ly and by actually f inding and accessing the materials, they are given literal, communal rights to libraries' holdings. For Kropotk in , communal associations were formed voluntari ly, uni t ing groups of individuals concerned w i t h a l l levels of social interests and functionality. In practice, everyone w o u l d contribute voluntari ly what they could, and take freely what they needed. Moreover , unions between the communes produce "a network of cooperation that replaces the state," (Woodcock, 2004,168) for al l central authority is unequivocally opposed. Kropotk in argued "that a further advance in social life does not lie i n the direction of a further concentration of power and regulative functions i n the hands of a governing body, but i n the direction of 64 decentralization, both territorial and functional... (1887, 2002 ed., 51)." Folksonomies are a consummate example of decentralized, functional, and cooperatively produced indexing languages. They can be created and continually adjusted to serve the needs of different communities of library users at any given time. Ward (2004, 31) argues that anarchist organizations are characteristically voluntary, functional, temporary, and small. Voluntary engagement is seen as necessary by Kropotkin (1887, 2002 ed.), and is reiterated by Ward: "There is no point in advocating individual freedom and responsibility if we go on to set up organizations in which membership is mandatory..." (2004, 31). Social tagging is obviously a voluntary undertaking. A 2006 survey by PEW Internet and American Life Project found that 28% of people who use the Internet have, on their own accord, tagged online content (Rainie, 2007). In fact, groups of users have embraced tagging - voluntarily - at a rate that has demanded the attention of technology and Internet researchers: "there are even reports that some web users now have made tagging sites their home page, making these sites at least nominal competitors to big media companies that hope users will start their online experiences on their main page" (Rainie, 2007). Ward advocates a functional anarchist theory of organization. He writes: "there is a tendency for bodies to continue to exist after having outlived their functions" (2004, 31). We would be hard pressed to imagine a situation where taggers are involved in tagging online content even when their efforts serve no 65 purpose. Ward ' s second criteria is closely l inked w i t h his first, as voluntary efforts w o u l d not be made if the volunteers d i d not gain something from their participation, either ind iv idua l ly or as a group. Of course, the functionality of tagging is varied across different social tagging sites and for different ind iv idua ls and communities, but for taggers to continue tagging, the pursuit must have a purpose. Even i n situations as specific as l ibrary O P A C s , not a l l taggers w o u l d tag w i t h the same intentions, but they tag w i th intentions nevertheless. F r o m an indexing point of v iew, social tagging can be offered as a solution to archaic and inflexible indexing terms i n traditional languages. Social tagging lets libraries stay current and relevant; tags can meet the changing demands of language and culture more efficiently than traditional indexing terms, thereby mak ing social tagging a useful and functional activity. Anarchist organizations are temporary; social tagging is, by definition, transient. Tags are not maintained and tolerated s imply because they have been used before. W a r d (2004) emphasizes the temporary nature of anarchist organizations to warn against the acceptance of permanence for its o w n sake. A g a i n , his point is closely related to the one mentioned previously, because organizations should be abandoned once they cease to serve their stated functions. Tags, folksonomies - the aggregation of tags - and taggers are ever changing. They are responsive to change, and easily removed. Taggers can remove their tags as we l l as themselves easily and quickly, thereby changing the makeup of their respective folksonomies. Furthermore, social tagging as a whole 66 does not require large financial or infrastructural commitments on the part of libraries i n question, and may be easily abandoned if and when it no longer serves the community. The opposite is true i n case of traditional l ibrary indexing. The time and money invested i n the creation and maintenance of traditional indexing languages is often regarded as a reason for its continued use (Dykstra, 1978). Once organizations are maintained for reasons of tradition or financial investment, functionality is not valued as highly or questioned as regularly as one w o u l d expect i n dynamic institutions such as libraries -institutions that hope to serve their communities in the best and most efficient ways as possible. Social tagging, as implemented on l ibrary O P A C s , is a small and communal undertaking. The folksonomies that are created reflect the tags contributed by the patrons of that particular l ibrary system. Especially i n relation to traditional indexing practices and controlled vocabularies whose a im is often universality, social tagging can be more immediately and cultural ly relevant. Where folksonomies reflect the culture of a particular community , members of that community are better able to interact w i t h the information that is presented to them; that is they are more l ikely to find what they are look ing for using search terms and keywords that are culturally current, relevant and appropriate. A t the same time, social tagging is not a bureaucratic process. Individuals can contribute tags directly and without interference from other members of the library community, making the process more functional, 67 immediate, and relevant. B y va lu ing the needs of the users over institutional bureaucracy, social tagging may be seen as tool that encourages patron participation and interest i n libraries. The fol lowing example of an information seeking incident at the Vancouver Public Library demonstrates that the system of representation currently provided by traditional indexing i n our libraries may work to impede the information literacy - the ability to seek, find, and evaluate information - of certain groups of people i n significant ways. D u r i n g a reference interview, I assisted a patron who was look ing for the book Where People Feast: An Indigenous People's Cookbook by D o l l y and A n n i e Watts. The patron had forgotten the title of the book, but had attempted to use the subject browse function of V P L ' s online public access catalogue. Unfortunately and to her surprise, the cookbook was listed w i t h the fo l lowing subjects: • Indian cookery - British Co lumbia • Indian cookery. • Cookery - Brit ish Columbia . The book, w h i c h was catalogued on A p r i l 30, 2007, was not found because the patron d i d not consider "Indian cookery" as an indexing term for the cookbook she was searching for. 1 5 It is true that no content object w i l l ever be represented to everyone's satisfaction and it is true that anarchism means many different things to many people, but there is something critical at stake: i n the year 2007, a 1 5 A s an aside, books on East Indian cooking is listed under: • Cookery, Indie [sic]. 68 patron at the Vancouver Public Library cannot f ind the cookbook Where People Feast: An Indigenous People's Cookbook because she searched under the subject headings "native cookery" and "aboriginal cookery" instead of "Indian cookery". In social tagging we have at our disposal a technology that can provide other access points to content objects and help patrons w h o cannot decode our current indexing languages to f ind their books. Neither she, nor I, w o u l d have attached the tag "Indian cookery" to the book i n question. If it is our goal to help patrons find the information they are looking for, then librarians and LIS scholars must carefully consider this opportunity. Biases inherent w i t h i n traditional indexing languages have been previously discussed. The example prov ided here highlights an instance where the indexing language used at the Vancouver Publ ic Library had a direct impact on the ability of a patron to f ind a book. A s critical librarians, we need an evaluative paradigm to assist us i n recognizing, formulating, and presenting addit ional access points than currently offered by traditional indexing languages. Anarch i sm can help us frame, understand and discuss the necessity and value of public 's participation in l ibrary indexing. Contrary to the aforementioned threat to our discipline's intellectual integrity, an anarchist analysis of social tagging can help librarians advance the information literacy of our patrons. There is an interesting parallel between two particular critiques of the two movements - social tagging and anarchism - and the structure of the response prov ided by their proponents. A s discussed i n Chapter 2, though never 69 declared as a purpose, a l o w degree of precision and recall i n retrieval has been v iewed as a shortcoming of social tagging and a point i n favour of traditional l ibrary indexing. This criticism, better directed at systems that do list h igh degree of precision and recall i n retrieval as one of their purposes, highlights the system's failure to provide this functionality as demonstrated by the patron's search for Where People Feast: An Indigenous People's Cookbook. Similar ly , anarchists have been criticized for their proclivities towards chaos and disorder when in fact, anarchists support order as long it is sustained through "collaboration, deliberation, consensus, and common coordination" (Vaidhyanathan, 2004, 4). In the abovementioned definition of anarchism given by Kropo tk in i n 1905, he emphasizes free and harmonious activities "constituted for the sake of production and consumption". Folksonomies are not universal and do not aspire towards "truth" i n indexing, rather they provide a functional solution to irrelevant, albeit consistent indexing; they are pragmatic tools that are created and used for the purposes of ind iv idua l and communal naming and access to information. Inevitably, questions of noncompliance arise. Detractors of social tagging have crit icized and questioned taggers w h o do not observe common tagging conventions - taggers who deliberately engage i n "sloppy" tagging practices (Lawley, 2005; Mejias, 2005). Anarch ism offers an interesting answer on this inquiry. Kropotk in argues that there are two kinds of agreements. The first is the contract, w h i c h is agreed upon freely; given other, equally viable choices, the 70 parties i n question have chosen this particular arrangement. The second is an "enforced agreement, imposed by one party upon the other, and accepted by latter from sheer necessity; i n fact, it is no agreement at al l ; it is a mere submission to necessity" (Kropotkin, 1887, 2002 ed., 69). Kropo tk in makes this distinction, as he does not "see the necessity of force for enforcing agreements freely entered upon" (Kropotkin, 1887, 2002 ed., 69). W h e n taggers engage i n social tagging, they do so out of their o w n vol i t ion and desire to reach the level of functionality that they are looking for. If common tagging practices can help them achieve their purpose, then it w o u l d serve them to observe these conventions and they w o u l d do so voluntari ly. For example, if it is a tagger's intention to help other women i n her communi ty by highl ight ing the local l ibrary 's holdings on women's health, then it w o u l d be counterproductive for this tagger to tag the books "womyn ' s health" as there is a good chance that other patrons w i l l not use this unconventional spelling of the w o r d to conduct their search. The next point of contention is the v iew that without external authoritative enforcement, long-term contributions to society are not possible; i n social tagging, people are unl ikely to continue participating of their o w n accord. This Kropo tk in sees as an essential ideological disagreement concerning human nature that is not easily reconciled. H e believes that w i t h humans, "work is a habit, and idleness an artificial growth" (Kropotkin, 1887, 2002 ed., 70). Kropo tk in argues that it is overwork and work under conditions of uncertainty 71 and unfair remuneration that is objectionable and not work per se. W i t h freely entered agreements and work arrangements, there w o u l d be decreased incidents of anti-social and anti-communal behaviours because "freedom, fraternity, and the practice of human solidarity is the only way of dealing, humanely, w i th antisocial behaviour" (Morris, 2004, 23) both before and after its occurrence. G i v e n the choice, people w o u l d choose dignified participation over nonparticipation i n work and society. Thus far, the contribution of taggers to various social tagging websites illustrates this anarchist principle: they have shown they are prepared to take up the work. Furthermore, Kropotkin ' s ideological position regarding workers ' will ingness to participate freely may be u t i l ized i n social tagging research. The above discussion is of particular interest to future research in social tagging because, as discussed i n Chapter 2, questions regarding tagger motivation are yet to be asked and systematically studied. W i t h no "central authority imposing a top-down v iew and every voice gain[ing] its space" (Quintarelli , 2005), are taggers motivated by freedom w i t h w h i c h social tagging sites currently operate? A r e taggers spending their o w n time tagging books on Library Thing, photos on Fl ickr , and websites on exactly because "free agreements need not be enforced" (Kropotkin, 1887, 2002 ed., 70)? A r e taggers devoting their o w n time to seemingly laborious tasks because they are cognizant of the benefits - the greater good - of sharing both the efforts and the benefits of social tagging? The answers to these questions are yet to be determined, but the ways i n wh ich researchers choose to frame and ask these 72 questions are inevitably influenced by their choice of paradigm. For example, researchers operating wi th in an anarchist paradigm w o u l d be more l ikely to study non-traditional library indexing and advocate a pol icy of non-interference in social tagging environments; they might for example, be against the implementation of pre-approved tag lists. Social tagging, as an area of research, is particularly we l l suited to the anarchist agenda; after al l , i n Kropotkin ' s o w n words , anarchism originated "from the demands of practical l ife" (1887, 2002 ed. 154). In fact, social tagging exists as a theoretical construct because of publ ic participation - because people tagged. A s argued throughout this thesis, information literacy - the ability to seek, f ind, and evaluate information - is a practical sk i l l that librarians must foster i n the patrons of their libraries. Thus far, traditional l ibrary indexing has not been successful i n adequately representing content objects contained wi th in our libraries, leading to decreased rates of information literacy. Us ing the anarchist paradigm, social tagging may be offered as a practical - albeit imperfect - solution to central and authoritative representation and misrepresentation of content objects i n l ibrary O P A C s that l imi t communi ty participation and use of l ibrary resources. The choice of anarchism as a paradigm for the study of social tagging on the open Web is particularly apt; interestingly the Internet has had a history of being theorized as an agent of anarchy (Graham, 1999; V a n Aelst & Walgrave, 2004). Of course, here again, one must sift through the misuse of the w o r d anarchy w i t h added qualifiers such as "good" and "bad" for purposes of 73 distinction. For example Graham (1999) uses "good anarchy" to refer to the use of the Internet as an instrument of publ ic access and participation and "bad anarchy" to refer to the Internet as a m e d i u m of misinformation and a place of scant moral pol icing. It is his use of "good anarchy" that is of interest here, and is henceforth referred to as anarchy. Graham (1999) argues that internationalism and popu l i sm are two important factors contributing to the Internet's ut i l izat ion as an agent of ind iv idua l participation and creation of a space w h i c h functions independently from and, in some cases, despite the state. Graham's (1999) internationalism goes beyond the idea of connecting people across borders to refer to a profound disregard for national and international boundaries. Notwi ths tanding governmental attempts to regulate and control its use, the "Internet is who l ly indifferent to international boundaries" (Graham, 1999, 86, emphasis original). Therefore, communities - both large and small - can form using the Internet regardless of individuals ' physical location. Similar ly , the popul i sm of the Internet goes beyond the observation that the Internet is s imply popular. Graham (1999) is referring here to the fact that access to the Internet is increasingly available to those w i th l imited financial and technical capabilities. 1 6 No t only has owning computers become more commonplace, but also places such as Internet cafes and libraries offer the public more affordable access on a l imited time basis. Furthermore, personal laptop computers and free wireless services offered on university campuses, coffee 1 6 This is especially true in the "first w o r l d " . 74 shops, and other venues are adding to the Internet's accessibility and popul i sm. The Internet is also progressively more user-friendly. A considerable amount of technical know-how is not required to use and contribute to it (specially since the advent of Web 2.0 technologies) and places such as public libraries offer introductory skills workshops to their patrons at little or no cost. In short, given its functionality beyond international boundaries and ease of use and access, the Internet has been, and w i l l i n al l l ike l ihood continue to be, a technological ally of the anarchist movement. Certain utilizations of this technology are, of course, more pertinent than others; social tagging is one such application. W i t h the expansion of social media on the Internet, we have the tools we need for i nd iv idua l participation i n activities such as l ibrary indexing - activities that were previously inaccessible and kept separate from the public domain. Conclusion A t the 2007 Supernova Conference 1 7 i n San Francisco, C lay Shi rky spoke of a 1995 meeting about programming languages where a group of C++ engineers question the level of commercial support available to Perl users. Perl , an open source programming language, does not have commercial , contracted support. Instead, Perl users util ize free community-based online platforms where they post questions and receive answers and support from other users. To demonstrate their point, the Perl users at this meeting posted a question on one 1 7 Supernova is an annual conference on technology, decentralization and connectivity organized by K e v i n Werbach and The Whar ton School. 75 of the support websites and received an answer to their inquiry by the end of the meeting. Shirky says that even given this evidence, the C++ engineers were not convinced, and noted: "they didn ' t care that they had seen it work i n practice because they already knew it couldn't work i n theory." 1 8 Today, that same support website is still operational and fully functional. Participants diagnose problems, offer solutions, and even write code for each other. " N o contracts are written. N o money changes hands. The work goes o n . " 1 9 People w h o care about the existence and longevity of an idea w i l l contribute to it - freely. Likewise , l ibrary indexing through bottom up patron participation is realized through social tagging because taggers tag. The introduction of social tagging i n libraries compels us to recognize and re-examine our assumptions about indexing, critical l ibrarianship and information literacy and also to ask fundamental questions about the role of cooperation and authority i n human society; questioning and opposing the status quo is no small aspiration and no small accomplishment. 1 8 h t t p : / / w w w . y o u t u b e . c o m / w a t c h ? v = X e l T Z a E l T A s 1 9 Ibid. 76 C h a p t e r F i v e N o w W e C a n D o B i g T h i n g s For L o v e : A n E p i l o g u e With love alone you can get a birthday party together, add coordinating tools and you can write an operating system. In the past, we could do little things for love, but big things - big things required money. Now, we can do big things for love. - Clay Shirky (2007) Debates concerning the advent and proliferation of Web 2.0 technologies make manifest our most candid and deep-rooted presumptions about human society. Inadvertently or otherwise, they divulge trust or distrust i n our ability to br ing about change; they reveal our faith i n humanity. They strike a chord so close to our hearts that our emotions run h igh and we speak of love, as C lay Shirky does, i n order to emphasize just h o w deeply invested we are i n this phenomenon. Librarians who are actively involved i n the social and poli t ical responsibilities inherent wi th in their profession ought not shy away from the language, the politics, and the discussions. In a debate about publ ic collaboration i n library indexing, we cannot choose to ignore what people are saying. Every tag, every article, every user-generated comment is, by definition, of value. Some may be inclined to dismiss this discussion because it is indeed daunting to sift through tags, blog entries, podcasts, etc. Others welcome it as a harbinger of anarchy and collaboration heralding the end of authoritative control 77 in domains previously inaccessible by the public. Most sit somewhere i n between, l istening still to arguments, discussions and verdicts whi le t rying to better understand user-generated content on the open Web and its role i n their lives. O n July 18, 2007, The Wall Street Journal Online published an exchange between A n d r e w Keen, the author of The Cult of the Amateur: How Today's Internet Is Killing Our Culture and one of Web 2.0's most vocal critics, and D a v i d Weinberger, the author of Everything is Miscellaneous: The Power of the New Digital Disorder. The style of the discussion is forthright and informal: no punches are pu l led and assertions are made boldly and without sufficient attempts at corroboration and case making. Nevertheless, Keen vs. Weinberger is representative of the ways in wh ich we - on blogs and at dinner tables - discuss collaborative technologies; it reveals what is in fact at stake when we debate the wor th of activities such as social tagging. To illustrate his points, Keen sets up a dichotomy: Disney's Cinderella vs. Kafka's Metamorphosis. H e argues that collaborative technologies have not contributed something of great value; they have not empowered us as Cinderella 's glass slippers d id , but what they have done instead is doomed us to "stare at our hideous selves i n the mirror of Web 2.0." H i s emphasis on the Kafkaesque elements of Web 2.0 reverberates throughout the entire piece: the aggregation of user-generated content is grotesque, dangerous, unfathomable, and horrifically surreal. We, the taggers, bloggers, and contributors, are the architects of Keen's nightmare. Web 2.0 is 78 "enabling anyone to publ ish anything on the Internet." W e have "forgotten h o w to listen, how to read, how to watch." W e are jeopardizing the future of broadcasting, music and publ ishing industries. We do not know the value of "truth" and have lost "interest i n the objectivity of mainstream media." Without the accountability and authority of experts "everything becomes miscellaneous. A n d miscellany is a euphemism for anarchy." In Keen's v iew, instead of va lu ing publ ic participation we should seek "arbiters of good taste and critical judgment", wh ich he associates w i t h traditional media outlets. "C i t i zen media" or Web 2.0 has sacrificed "the impartial i ty of the authoritative, accountable expert" replacing it instead w i t h the "anonymous amateur." After reading and watching M r . Keen's many contributions to the Web 2.0 debate I, l ike Gregor Samsa, have a distinct feeling that I have awaken from "uneasy dreams" and found myself "transformed into a gigantic insect" (Kafka, 1952,19). We are indeed dealing i n absurdities regardless of w h o m or what we analogize as Kafka's cockroach. Keen holds that the media is objective, the publ ic has little of value to contribute to our o w n intellectual and cultural domains, such as they are represented on the open Web, and our priorities should l ie w i t h keeping traditional media alive and prosperous. D a v i d Weinberger, his opponent in this debate, disagrees. Weinberger argues that as w i t h other information sources i n our lives, it is up to us to determine the quality and source of the information we are consuming in accordance w i th our needs -pragmatic, intellectual, aesthetic or otherwise. The Web, Weinberger writes, is 79 "far better understood as p rov id ing more of everything: M o r e slander, more honor. More porn, more love. More ideas, more distraction. M o r e lies, more truth. M o r e experts, more professionals. The Web is abundance, whi le the o ld media are premised - i n their model of knowledge as we l l as their economics -on scarcity." The open Web generally, and user-generated content specifically, help us better understand a w o r l d "richer and more interesting than the constrained resources of the traditional media let on." They a l low space for voices and contributions from people who, customarily speaking, do not have opportunities or outlets for participation as they " w o u l d not, could not, or d i d not make it through the traditional credentialing and publ ishing systems i n the areas they're wr i t ing about." Public involvement on YouTube, Fl ickr , Blogger, and on library O P A C s has the ability to generate content that is richer and more "frui tful" and "st imulat ing" than the contributions of authorities, professionals, or experts alone. The disagreements between Keen and Weinberger are indicative of larger conversations about user-generated content. Is Web 2.0 "f lushing away valuable culture" or is it "our culture's hope?" I am hopeful. Social tagging, a social movement against the authority of traditional l ibrary indexing, provides alternative access points to the content objects w i th in libraries and i n doing so contributes positively to the information literacy of patrons. In this thesis, I have argued in favour of adopting an anarchist paradigm for the study of social tagging because anarchism w o u l d preserve and further advance social tagging as a socio-political tool and ensure that the most critical characteristics of tagging -inclusivi ty and flexibility - are considered at every stage of development and implementation. In future research, it is vi tal to engage taggers and other Web 2.0 collaborators i n conversations about their involvement. User-generated content on the open Web has reached critical mass. The incentives and motivations behind this grand partaking w o u l d inform future theoretical as we l l as practical considerations: perhaps they do it for love. 81 Bibliography A d a m s , C. J. (1990). The sexual politics of meat: A feminist-vegetarian critical theory. N e w York: Cont inuum. A l l e n , R. (Ed.). (1999). Pocket Fowler's modern English usage. N e w York: Oxford Univers i ty Press. Retrieved M a y 21, 2007 from Oxford Reference Online, http: / / v i e w s / E N T R Y . h t m l ? s u b v i e w = M a i n & entry=t30.e3139 Andersen, J. (2005). Information crit icism where is it? Progressive Librarian, 25, 12-22. Retrieved June 11, 2007 from h t tp : / / l ib r .o rg /p l /25_Andersen .h tml Andersen, J. (2006). The public sphere and discursive activities: information literacy as sociopolitical skills. Journal of Documentation, 62, 2, pp. 213-228. Andersen, J. & Skouvig , L . (2006). Knowledge organization: A sociohistorical analysis and critique. Library Quarterly, 76, 300-322. Anderson , C. (2006). The long tail: why the future of business is selling less of more. N e w York: Hype r ion A N S I / N I S O Z39.19-2005. Guidelines for the construction, format, and management of monolingual vocabularies. Retrieved M a y 21, 2007 from http: / / Bernstein, R. (1983). Beyond objectivism and relativism. Philadelphia: Univers i ty of Pennsylvania Press. Berman, S. (1984). Out of the kitchen - but not into the catalogue. Technical Services Quarterly. 2,167-171. Blood , R. (2005). Unti t led blog entry. Retrieved A p r i l 20, 2007 from http: / / /archive/2005/01.html#ll technorat i Bruce, H . , Jones, W . & Dumais, S. (2004). Information behaviour that keeps found things found. Information Research, 10,1, paper 207. Cahm, C. (1989). Kropotkin and the rise of revolutionary anarchism 1872-1886. N e w York: Cambridge Universi ty Press. Canadian Library Association (n.d.). Miss ion , values & operating principles. Retrieved A p r i l 25, 2007 from h t tp : / /www.c la .ca /abou t /miss ion .h tm 82 Chomsky, N . (1970). "Introduction". In D . Guer in (Author). Anarchism: From theory to practice. N e w York: Mon th ly Review Press. Chomsky, N . (1991). Economic round table. In M . Alber t & R. Hahne l (Eds.), Looking forward: participatory economics for the twenty first century. Boston: South E n d Press. Ciborra , C. (2004). Encountering information systems as a phenomenon. In C. Avge rou , C. Ciborra & F. L a n d (Eds.), The social study of information and communication technology. N e w York: Oxford Universi ty Press. Delgado, R. & Stefancic, J. (1989). W h y do we tell the same stories?: L a w reform, critical l ibrarianship, and the triple helix di lemma. Stanford Law Review, 42, 207-225. Digi ta l d ivide: What it is and w h y does it matters, (n.d.). Retrieved A p r i l 20, 2007 from h t tp : / /www.d ig i t a ld iv ide .o rg /dd /d ig i t a ld iv ide .h tml Donath, J. & boyd, d. (2004). Public displays of connection. BT Technology Journal, 22, 71-82. Dykstra , M . (1978). The l ion that squeaked. Library Journal, September, 1570-1572 El l i s , D . (1998). Paradigms and research traditions i n information retrieval research. Information Services and Use, 18, 225-241. Foskett, A . C. (1971). Misogynists al l : A study of critical classification. Library Resources and Technical Services, 15(2), 117-121. G o o d , B. & Tennis, J. (2007). Speciating information organization frameworks us ing framework analysis. In press. Graham, G . (1999). The Internet: A philosophical inquiry. N e w York: Routledge. Guer in , D . (1970). Anarchism: From theory to practice. N e w York: M o n t h l y Review Press. G u y , M . , & Tonkin , E . (2006). Folksonomies: t idy ing up tags? D-Lib Magazine, 12,1. Retrieved M a r c h 10, 2007 from http: / / w w w . dlib. or g / d l ib / j anuary 06/ guy / 01 guy .html Hack ing , I. (1999). The social construction of what? Cambridge: Ha rva rd Univers i ty Press. 83 Hjor land, B. (1998). Theory and metatheory of information: A n e w interpretations, journal of Documentation, 54, 606-621. Kafka, F. (1952). Selected Stories of Franz Kafka. N e w York: Random House. Keen vs. Weinberger. (July 18, 2007) Wall Street Journal Online. Retrieved July 19, 2007 from Kome, S. (2005). Hierarchical subject relationships i n folksonomies. Retrieved M a y 6, 2007 from ht tp: / /e td . i ls Kropotk in , P. (2002 ed.). Anarchism: A collection of revolutionary writings. N e w York: Dover Publications. Kropotk in , P. (2006 ed.). Mutual aid: A factor ofevolution. N e w York: Dover Publications. Krosk i , E . (2005). The hive mind : Folksonomies and user-based tagging. Retrieved A p r i l 2, 2007 from http: / / 07/ the-hive-mind-folksonomies-and-user-based-tagging/ K u h n , T. (1962). The structure of scientific revolutions. Chicago: The Univers i ty of Chicago Press. Lawley , L . (2005). Social consequences of social tagging. Retrieved M a r c h 29, 2007 from h t t p : / / m a n y _social_tagging.php Leier, M . (2006). Bakunin: The Creative Passion. N e w York: St. Mar t in ' s Press. L i n , X . , Beaudoin, J., Bu i , Y . & Dasai, K . (2006). Explor ing characteristics of social classification. Retrieved M a r c h 12, 2007 from ht tp: / /dl is t .s i r .ar M a i , J. (1999). A postmodern theory of knowledge organization. Knowledge: creation, organization and use: proceedings oftheASIS Annual Meeting. Medford : Information Today. 547-556. M a r l o w , C , Naaman, M . , boyd, d., and Davis , M . (2006). Posit ion Paper, Tagging, Taxonomy, Fl ickr , Art ic le , ToRead. Retrieved A p r i l 2, 2007 from http: / / www.danah .org /papers / 84 Mejias, U . (2005). Tag literacy. Retrieved A p r i l 2, 2007 from http: / / ideant. teracy.html M i l l e r , H . (1934). Tropic of Cancer. Paris: Obelisk Press. Mi l l e r , H . (1939). Henry Miller On Writing. N e w York: N e w Directions. Mor fo rd , M . P. O. & Lenardon, R. J. (1995). Classical Mythology. Whi te Plains: Longman Publishers Mor r i s , B. (2004). Kropotkin: The politics of community. N e w York: H u m a n i t y Books. N i n , A . (1995). Fire: from a journal of love. San Diego: Harcourt Inc. N i n , A . (1986). Henry and June: From A Journal of Love: the Unexpurgated Diary of Anais Nin. San Diego: Harcourt Inc. Olson, H . (2001). Patriarchal structures of subject access and subversive techniques for change. The Canadian Journal of Information and Library Science, 26 (2/3), 1-29. Olson, H . (2002). Power to name: Locating the limits of subject representation in libraries. Boston: K l u w e r Academic Publishers. O 'Rourke, M . (2005). O n the eve of a queer-straight future: notes toward an antinormative heteroerotic. Feminism & Psychology, 15,1, pp. 111-116. Palmer, C. L . & Malone, C. K . (2001). Elaborate isolation: Metastructures of knowledge about women. Information Society, 17,179-194. Postman, N . (1993). Technopoly: the surrender of culture to technology. N e w York: Knopf. Quintarel l i , E . (2005). Folksonomies: power to the people. Retrieved A p r i l 15, 2007 from http: / / www. iskoi .o rg /doc / fo lksonomies .h tm Rainie, L . (2007). 28% of online Americans have used the Internet to tag content: Forget Dewey and his decimals, Internet users are revolut ionizing the w a y we classify information - and make sense of it. Retrieved M a y 3, 2007 from h t tp : / /www.pewin te rne t .o rg /PPF/ r /201/ repor t_d isp lay .asp 85 Samek, A . (1998). Intellectual freedom and social responsibility: A n ethos of Amer ican librarianship, 1967-1973 (Doctoral Dissertation, Universi ty of Wisconsin-Madison, 1998). Shirky, C. (2005). Ontology is overrated: categories, l inks, and tags. Retrieved M a r c h 13, 2007 from h t t p : / / w w w . s h i r k y . c o m / writ ings/ontology_overrated.html Shirky, C. (2007). Supernova conference presentation. Retrieved July 30, 2007 from h t t p : / / w w w . y o u t u b e . c o m / w a t c h ? v = X e l T Z a E l T A s Speller, E. (2007). Collaborative tagging, folksonomies, distributed classification or ethnoclassification: a literature review. Retrieved M a r c h 21, 2007 from http:/ / / lsjSearch.php?kw=Edith+Speller Sterling, B. (2007). Order out of chaos. Retrieved M a r c h 21, 2007 from h t t p : / / w w w . wired, c o m / wi red /a rch ive /13 .04 /v iew, h tml ?pg=4 Svenonius, E . (2007). The intellectual foundations of information organization. Cambridge: The M I T Press Tarrow, S. (1994). Power in movement: social movements, collective action and politics. N e w York: Cambridge Universi ty Press. Tennis, J. (2006). Social tagging and the next steps for indexing. Retrieved M a r c h 10, 2007 from http: / /dl is t .s i r .ar Tonk in , E . (2006). Searching the long tail: h idden structures i n social tagging. Retrieved M a r c h 10, 2007 from h t tp : / /www.s la i s .ubc .ca /users / s i gcr/events.html Vaidhyanathan, S. (2004). The anarchist in the library: How the clash between freedom and control is hacking the real world and crashing the system. N e w York: Basic Books. V a n Aelst , P. & Walgrave, S. (2004). In W i m can de Donk, Br ian D . Loader, Paul , G . N i x o n , Diete Rucht (Eds.). Cyberprotest: New media, citizens and social movements. N e w York: Routledge. Vander W a l , T. (2004). Folksonomy. Retrieved February 2, 2007 from http: / / vanderwal .net /folksonomy.html. Voss, J. (2006). Collaborative thesaurus tagging the Wik iped ia way. Retrieved M a y 6, 2007 from ht tp: / / . IR/0604036 86 W a r d , C. & G o o d way, D . (2003). Talking Anarchy. Nott igham: Five Leaves Publications. W a r d , C. (2004). Anarchism: A very short introduction. N e w York: Oxford Univers i ty Press. Weinberger, D . (2007). Everything is miscellaneous: the power of the new digital disorder. N e w York: Hen ry Ho l t and Company. Winget, M . (2006). User-defined classification on the online photo sharing site F l ickr ... Or , H o w I learned to stop wor ry ing and love the mi l l i on typing monkeys. Retrieved M a r c h 10, 2007 from http: / / /1854/ Wiser, W . (2001). The twilight years: Paris in the 1930s. London: Robson Books. Woodcock, G . (2004). Anarchism. Peterborough: Broadview Press. Zizek, S. (n.d.). Tolerance as an ideological category. Retrieved A p r i l 23, 2007 from ht tp : / /consel lodacul tura .org/sent idos/wp-content /uploads/2007/03/zizek_santiago.pdf Zizek, S. (2006). The parallax view. Cambridge: The M I T Press. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items