LIS 9763

 
One of my major projects here at McMaster is conducting collection analyses using OCLC’s WorldCat Collection Analysis Tool. Working with a liaison librarian, I have been analyzing our current collection numbers for subject heading: “People with Disabilities.” In order to focus my analysis, I asked the liaison librarian to consult the faculty members studying this subject matter. As a result of this request, I was bewildered! One of the faculty members sent me a four page response detailing her research areas and desires for collecting. In the OCLC Collection Analysis Tool I cannot narrow my results by a specific call number. Instead, I must contend with ranges (i.e. HV 1551 – 3024: “People with Disabilities”). There are other pertinent call numbers that would match the faculty member’s research interests (i.e. NX 180.H34: People with disabilities and the arts). As you know, subjects are multidisciplinary, now more than ever, and even controlled vocabulary has its limit. 
 
When I was in cataloguing I struggled with this very dilemma. How do I classify this text using one subject heading when it is equally applicable to another subject heading? I do not think in binaries. As you can see from my example above, collecting for faculty can be daunting when every call number under the sun seems relevant! When I came across Ellyssa Kroski’s article “The Hive Mind: Folksonomies and User-Based Tagging” I was elated. I realize that there is a lot of contention surrounding the term and practice of “folksonomies,” but for the first time I felt that there was hope at the end of the controlled vocabulary tunnel.
 
Carol Ou argues that “ideally, distributed classification will prove to be more than merely functional and provide librarians with yet another means of discerning user preferences and responding to user needs” (3). As noted by all the readings, controlled vocabulary isolates user groups. In the PsycINFO thesaurus, for example, “human females” is used for women. The majority of psychology students and faculty might know to use this subject heading, but students from other disciplines may not. Based on my experience at McMaster, this is the case. Folksonomies are created by users so they reflect the users’ needs.  Kroski notes that “we are moving away from expert-dictated, exclusionary models of information organization and toward, inclusive participatory ones” (4). I agree with Kroski’s observation, but like Liz Lawley I am still hesitant to full heartedly join the folksonomies “band wagon.”
 
Kroski notes that folksonomies are:
1)      Non-binary
2)      Democratic
3)      Inclusive
Rebecca Blood (qtd in Lawley) argues the deliberate tagging can occur to misdirect people. I think that the bias attached to folksonomies skews the inclusiveness argument. Mathes argues that “you can use the most common tags to develop a controlled vocabulary that truly speaks the users’ language”. What if that language is exclusionary, biased and binary? Bias is not necessarily a conscious act. These are thoughts that are embedded. If tags reflect users’ language then they reflect the users’ biases. Tag clouds, for example, highlight the most used tags. If a new term or group emerges will their tag be highlighted? I realized that controlled vocabulary is not all encompassing, but I do not think that communal tagging is completely inclusive, or non-binary, based on individual biases and communal cultures, practices, and perceptions.  
 
It is an establish code in librarianship that we are objective when classifying. Even if bias does leak into expert forms of classification at least the overarching purpose is objectivity. This is not the case with folksonomies. There are no rules, guidelines or practices governing this collaborative practice. Based on my first blog post, I feel there is a panoptic form of surveillance. The dominant thoughts of society will govern tagging especially if individuals compare their tags to others. As noted earlier in this post, I see the limitations of controlled vocabulary but I fear the “peer pressure” factor that skews folksonomies. 
 
In the online chat last week, I noted that I would support a hybrid of these classification forms. The best of both worlds: where control would meet chaos, and produces a user-friendly and ordered system of retrieval.  Various libraries are integrating tagging into their OPACS (as I noted in my last blog post). I like Ann Arbor District Libraries use of the right-hand menu that notes the “Top 10 Tags” or “10 Most Recent Tags.” I did click on one title in the catalogue: “Children's literature: a reader's history, from Aesop to Harry Potter by Seth Lerer” which someone had tagged “Harry.” Rather ambiguous considering this text does not only focus on Harry Potter and “Harry” does not exactly equate the infamous “Harry Potter.” When I clicked on the tag itself the only results was this one. Also, I am not sure what “10 Random Tags” means on the home page? Moreover, I did not see a post or anything explaining the tagging system.
 
In the Blais Catalog, I found the library thing tags helpful because I could click on the tag “Harry Potter” and retrieve more texts on this subject. Again, I did not see an explanation of this tool anywhere on the site. So, I think that integrating user tagging into an OPAC requires a lot of pre-assessment and pilot work. Also, suggesting similar tags might assist users to counteract ambiguous tagging.  I believe that folksonomies and taxonomies can co-exist but that these hybrid catalogues are still grassroots projects. They require ongoing assessment. I think this assessment piece will ultimately produce a hybrid that forces taxonomies and folksonomies to work in an integrated manner for the betterment of users information needs and librarians’ sanity.