I attended the Friday morning session on outsiders’ perspectives of the the public library catalogue. Representives from CBC, lavalife (an online dating service), and CIBC VISA discussed the challenges facing their web interfaces as they respond to user needs and new "web 2.0" options.
One observation about web 2.0 sites I found most interesting was that the successful web 2.0 sites tend to dominate in their space, and that adding web 2.0 features to sites with less traffic or already highly specialized services may be less successful. The dominance of social network sites FlickR in photo sharing and YouTube in video sharing may not translate to sites like CBC, lavalife, and CIBC VISA.
As a dating service, lavalife uses fixed value lists to create profiles, not user-supplied open content categories. Essentially this is part of the authoritative taxonomy vs folksonomy debate– the prescribed vs. the participatory approach– and while lavalife, as well as the other "traditional institutional" companies like CBC and CIBC, see some potential for more flexible services, there is concern about the effectiveness of creating web 2.0 communities around their sites.
A web 2.0 community needs to be a caring community where users invest energy in the maintenance of the social network site. But the users are not often experts, and in fact many do not want to invest in maintaining the site– they want the site to do the work for them. The speakers from the three companies all agreed on this one observation about web 2.0 technologies: "effectiveness depends on user engagement". Effectiveness was also more specifically defined for these companies as not a measurement of increased revenue, but something that drives user confidence which will help to increase market share. User confidence historically has been associated with the authoritativeness of the information sources (who wouldn’t want the news at CBC and the bank statements at CIBC to be authoritative?).
Given these observations, I found it interesting to hear the three outsiders comment on the library catalogue. The accuracy and specificity of the information in library catalogues are the primary added values in meeting user needs. The catalogue presents specific relationships between authors and works, and provides accurate details about where to obain the item. Even more specific functions along these lines were suggested, such as information about hours and geography for branch locations available from the catalogue. In addition, there was a request for more options to see relationships in the catalogue, such as a "more on this topic" style of interface as popularized by Amazon.
Web 2.0-like user participation is a possibility when users have the ability to supply more relationships between works based upon perceived groupings that may be missed by authoritative approaches. Whether this helps or hinders other people (by making search results too broad and imprecise) is still an open question. The suggestion that the additional "tagging" be done by professionals I thought was interesting, because I think this is where this issue dovetails with FRBR (the Functional Requirements of Bibliographic Records). Still very much a conceptual model, FRBR is all about establishing clearer relationships between entities in the catalogue (i.e., works, authors, topics) and tying these to the user tasks of finding, identifying, selecting, and obtaining specific bibliographic resources. There is a lot of space here for even more authoritative efforts in designing and building catalogues– in spite of the bandwagon for the web 2.0 participatory approach. Some combination of the two approaches for the library catalogue was the chief suggestion by these three outsiders.
Another challenge to the traditional library catalogue is keyword searching– not just keyword searching of the metadata, i.e the traditional MARC catalogue record, but keyword searching of the full text of the book. Ever more precise search algorithms could map out the content of books such as those being scanned by Google and Microsoft, and these in turn could lead to a new search mechanism that could augment, rival, or even replace traditional cataloguing. The availability of full text could also open a new door for the participatory model as users create links not just between the aggregate entities of books but sections within the books. Table of content services integrated into the MARC record has already moved the catalogue world a part of the way there. However, even considering all of these possiblities, there was a general reluctance from the outside observers from CBC, lavalife, and CIBC VISA to abandon traditional forms of bibliographic control as found in our current library catalogues. The ideal presented was one in which all methods of searching and organizing information could be combined to meet library users’ requirements.