LibraryThing: Social Groups Increase Book Tagging

There are many sites and organizations that help users search for books, and there are many ways to search a book. Some people search for books in the physical spaces, such as public libraries (e.g. Library of Congress, New York Public Library) or bookstores (e.g. Borders, Barnes and Noble). Some people search for books in the digital space (e.g. Amazon, Google Book Search, Web of Science). All these sites provide different purposes. For instance, a library provides different indices for books in their collection, but Amazon recommends a book they want to sell, and Google Book Search offers the actual content of the book. All of these sites offer valuable information, but LibraryThing tries to combine the objective of many sites. The site went live in 2005, and after one year, LibaryThing had “73,000 registered users who had catalogued 5.1 million individual books, representing nearly 1.2 million unique works,” (Wikipedia). To this day, LibraryThing has 190,324 registered users who have catalogued 13,041,058 individual books, representing 1.9 million unique works (LibraryThing Zeitgeist). Compared to Amazon, who has 60 million users and 1.3 million book tags, LibraryThing has 13 million book tags (Thingology Blog). Other social book sites like Shelfari lack the depth of LibraryThing, which aggregates information from Amazon, Library of Congress, and 78 other libraries (from different parts of the world and universities), Del.icio.us, Digg, and Yahoo. The members of LibraryThing have both an individual goal of cataloging their books, and a collective goal of tagging books, thereby collaboratively filtering, creating a successful recommendation system that is “library-quality,” and a fit social environment for its members.

LibraryThing is an extra large group. Many members keep coming back to the site to build their reputation, and many members use their real name, real photo, or keep their catalogs public, even when privacy is an option. The members and users of LibraryThing are often termed “thingamabrarians.” Members keep visiting the site to update their personal collections, reading lists, and wish lists, meet people who have similar libraries, and desire book recommendations with similar and polarized interests, all of which dynamically change according to other members. LibraryThing is easy to use, and its search variables and links are detailed, and relevant to members.

It is very easy to join LibraryThing, and registration requires just a username and password, and can be done instantaneously on the home page. After registration, the user’s profile shows up, and he/she can start adding books to the library by adding pre-existing sources (from Amazon, Library of Congress, and 78 other libraries), importing books from web sites, or by simply typing in the title, author, ISBN number or LC Control Number (Library of Congress). Once a book has been entered, results from Amazon is displayed, in particular, title, author, publisher, year the book was published, and format (hardback/paperback). Clicking on the title of the book adds it to your library and shows you how many other members have catalogued this book. Under the user’s profile, you can see a list of other members who share that book with the “weighted” statistic (number of copies of that particular book/number of total books in their library), to show “obscurity” and “library size.” This set of information can be sorted by users who have recently added this book to their collection (within two weeks), and “raw,” which gives the just the members who own the same book. This feature is also offered as an RSS feed. On the same page, the user can comment on the book (with the option of public or private) and other members can comment on the user’s comment. Once this book has been added to the library, the user can tag this book. Clicking on a specific group member’s username will give you additional information about the user, such as the number of “Books catalogued,” number of “Books reviewed,” the “Top tags,” “Groups,” “Author cloud” (data cloud of authors in that specific user’s library), and where the user resides. Most of this information is linkable to other web pages, and inter-linkable to each other.  

“Books catalogued” will lead to a member’s library, and displays the cover of the book, title, author, date, tags, rating, and the social aspect of the book: recommendations, reviews, members ratings, members, and citations. LibraryThing makes it very easy for members to transact. Icons that represent information about the number of members that own that particular book, book information, and an option to add that book, make it easy for the users to engage in updating books in their libraries or meet other people with similar books.

“Books reviewed” will link to a page of the user’s reviews of a particular book, but this page also displays any and all reviews of books in the user’s library, which are linkable to more book information, where reviews are listed. The number and content of reviews affects the reputation of the user, increases social capital, and encourages users to engage in conversation with each other.

“Top tags” list most popular tags used by the specific member, and are linked to web pages in “catalog” style, and displays linkable and statistical tag information, specifically, the frequency of tags, and the frequency of users using that tag. For instance, “the tag ‘awesome’ is used 379 times by 118 users.” Book tags are successful collaborative filters in a large population (“When tags work and when they don’t: Amazon and LibraryThing” by Tim Spalding, Thing-ology Blog). The reason why scale is important is because large numbers give particular importance to relevant and important tags. In the data cloud, tag frequency is visibly prevalent, therefore, tags that are used frequently appear to be bigger, while irrelevant tags/opinion tags/commercial tags appear smaller, and are more likely to be used again and again. Tagging helps LibraryThing’s recommendation system have more “library-quality.”

“Groups” will link to a page specifically about the book group. For example, “Byzantinistik” is a group interested in “history, culture, religion, art, science, economics and politics of Byzantium and the Byzantine Empire.” On this page, “common tags,” “other key terms,” “related link,” “most recent members,” number of books with similar tags (i.e. “Byzantium,” “Byzantine,” and “Constantinople”), and “most commonly shared books” are listed.

“Author cloud” aggregates the authors of the users library in a data cloud. “Author cloud” links to a data cloud with authors that are tag-sized by frequency of books written. Linking to the author will give a title list of books authored, link to conversations (leading to a forum of topics related to the author), user ratings, list of related tags, and options to link to global author clouds. “Author clouds” is helpful to filter out reviews made by the author him/herself to promote the book commercially. For example, when a new book comes out on Amazon, publishers and authors typically post positive reviews of their books to enhance their books’ commercial value.

User’s specific “Author cloud” from personal library:

Information about author:

Conversation about author:

Small population sample of global author tags:

Amazon’s suspicious author reviews. In my own opinion of this book, I had a low opinion of this book, which is why it is quite suspicious that there are only four customer five-star reviews.

Additionally, LibraryThing has posted links of popular statistics under their “Zeitgeist” page, which engage members just to explore. Some links include, authors who are members of LibraryThing, members who are top taggers, recent users, top 25 languages, 50 top-rated authors, 50 lowest-rated authors, 75 top data tags, books that were added within the last 24 hours, and top largest groups to name a few. LibraryThing has also added a lot of tools to increase data facilitation and social capital. Such examples include blog widgets, RSS, import/export data, barcode scanners, mobile applications to view browsers by the web (the user can check to see if it’s in their library), and the newest feature is their promotion of sites that allow members to swap a book.

The success of LibraryThing’s book tagging system relies on extra-large groups, and their members to be involved. Although they have less registered users than Amazon, and 10-fold more tags, indicate that maybe there is a Tragedy of Commons effect, but since there are more members who care, and they revisit the site to constantly update their personal libraries (adding more tags), this tagging system demonstrates a successful collaborative filter. Last December, I had started three memberships at Shelfari, which is also a book social networking application, but I haven’t visited it since to update my profile, even though some Shelfari members invited me into their network. LibraryThing is easy to use, and since all the pages are interrelated, finding relevant information about a book is easy as well, and promises more revisits from loyal members to their site. In class today there were other reasons for why LibraryThing was successful that I overlooked. The main difference between Amazon and LibraryThing is the prospect of acquiring books. Amazon offers members a service to add books to their “wish list.” Once a book is purchased on Amazon, members are not required to update their profile on the site, but LibraryThing offers “wish lists” and a space to show of their acquired books (books members own), which is why members of LibraryThing frequent that site. Come to think of it, out of all my book purchase, Amazon had only requested my reviews for two different books after the first month of purchase, and under that time constraint, I only read five chapters of the book, and didn’t feel adequate to give the books a fair review.  I believe that LibraryThing is pretty successful at designing the social aspect and collaborative filtering system. The only thing I can think of to improve the site is for them to bridge the digital and physical spaces. They are trying to implement the tools with the “swap-a-book” program and mobile application. I would like to include some of their tags into my thesis project, which is a physical shelf that will offer more relevant information.