â€œWe will all be out of our comfort zone for a while.â€
Googling the Better Mousetrap: Cyber Resources on the Front Line of Reference
RUSA 2005 Presidentâ€™s Program
Sheraton Chicago Ballroom VI/VII
[Mere minutes late! Getting better at timing leaving the McCormick wifi teat and busing to a hotel. In my next life, Iâ€™m staying at the Sheraton. Itâ€™s right on the river, and I found the ballroom easily! Large ballroom, not quite full but crowded.]
Most complicated evaluation form ever seen. Eekâ€”forgot to fill out! Will mailâ€¦
John Dove, President, Xrefer
Chris Nasso, Gale Group
Bill Pardue, Arlington Heights Memorial Library
Marilyn Parr, Library of Congress
J. L. Needham, Google
Jimmy Wales, Wikipedia/Wikimedia
Abstract: How do information/reference sources live and grow on the web? A panel of librarians, publishers, and search engine designers will discuss:
Patterns of user behavior that affect resource design
Information quality control in distributed production environments (content vs. containers)
The development of finding tools for online resources
I donâ€™t think the panelists managed to cover all the points in the abstract, but it was a really good discussion.
Marilyn started things off nicely with a short discussion of LOCâ€™s history with computing since 1960. I was interested that she mentioned how digitization of photographs led to LOCâ€™s entry into electronic resource creation. Now, LOC pages are accessed 1 million times a day, many times through referrals from Google. She hears increasingly from the public that â€œthe Library of Congress should use Google for search.â€ (Well, American Memory results can be daunting.)
J.L. started off by noting that the program title was an example of improper use of the Google brand. He also admitted to only 6 months of exposure to library land at Google. (We must be gentle?)
He noted that previously e-books have been mostly invisible on the web to users and search engines. He said it is time to up the ante with publishersâ€”for both cataloging and search engine optimization (not his term). He said publishers need to reallocate resources to provide access to the document itself rather than to their home page.
Bill described how NorthStarNet uses distributed content creationâ€”it is a reference resource created and maintained by its usersâ€”mostly reference librarians. It is a directory, a calendar. Provides contributors with a blogging interface, forums, and a possible â€œcommunity in a boxâ€.
Chris described how as a regular public library user, he has observed patrons searching online for information he knows the library owns (doesnâ€™t he mean rents?), but they are looking in a place where he knows they wonâ€™t find it. Jokes that guards are finding his behavior suspiciousâ€”kinda have concerns there myselfâ€”privacy anyone? Anyway, he sees a big evolution in the industry. Gale/Google â€œaccessmylibrary.comâ€ aims to improve the content of the internet, and that users will get the â€œgoing through their libraryâ€ concept.
Jimmy gave a precise explanation of Wikipediaâ€™s history, philosophy, and process. He spoke a little about the influence of the four freedoms of free software. He considers Wikipediaâ€™s strength to be its strong, passionate community.
John had recorded comments from Terry Winograd and Amy someone, a reference librarian. Terry said we need to get to a mind readerâ€”and for that to happen, a search engine needs to know you. John said that if a search engine tried to know him via his shared home PC, it would go crazy. Amazonâ€™s â€œpeople who purchaseâ€¦â€ feature already gets pretty weird trying to reconcile him, wife, and son. Amy said we donâ€™t need a mind reader machine; we have reference librarians. John talked about reference needs moving to 24/7 and self service. Content + librarians + users need another business model, not advertising based and not F2F. Google Print is fabulous, but how will users make sense of the retrievals of the Bodleianâ€™s 18th century results? The librarian role in this has not yet been invented. Someday in the future, a 5th grade grandchild will ask him, â€œwhatâ€™s a results list?â€ So we need a good online self-service reference platform.
Highlights of the extremely lively Q & A:
How will Google do the ranking of hits from Google Print?
J.L.: already working with different algorithms for different formats.
What if a librarian who wants to teach evaluation plants bad info in Wikipedia?
Jimmy: in his experience, librarians are more ethical than that. [Laughter.] Described how new posts are the most scrutinized. The constant review process has made bad behavior rare.
Wikipedia and controversial posts, i.e, George Bush entry during election?
Jimmy: media reports were completely wrong. The Bush and Kerry entries were not controversialâ€”it was just vandalism from outside the community that made them lock the entries. They are about to go live with new software that will allow a time delay mode to keep vandalism at bay.
What do Google and Wikipedia know that the library community does not? (wow, this spark ignited things!)
Bill: Rely on the community to generate reputation and usefulness. Marilyn: LOC uses Google to find LOC stuff. The LOCâ€™s own metadata doesnâ€™t talk to each other, leading to incomprehensible results. J.L.: Currently the open web has no competition. Libraries are walled gardens. Speed drives decisions (milliseconds matter.) Comprehensiveness is key. He used Wikipedia to get info on a Chinese province recently. Search engine indexing of our walled gardens is key. John: Google speaks to Ranganathanâ€™s Fourth Law. Speed, yes. But relevance and authority need context. Chris: Search engines have such sheer breadth of coverage. Individuals information needs morph so quicklyâ€”within one search session even.
Lowered expectationsâ€”Google gives sufficient answers. Students going for full text convenience of lesser source rather than go for better print resource one floor away.
Jimmy: Funny line about regular emails from college students getting Fs for citing Wikipedia. He says, “You are in college! Learn to use sources!” [laughter.] Itâ€™s a problem that wider digitization efforts may solve. Marilyn: Google draws students into things they wouldnâ€™t have found.
Which product is fun? Which offers community? It has upped the burden of scholarly resources. Chris: new name for Thomson-Gale? [laughter.] John: new name for Xrefer that does not include â€œThomsonâ€! [much laughter]. Then gave example of how Xrefer content map on Armageddon led to paper on Pink Floyd.
Users donâ€™t care about the container! LIS schools spend so much time on encyclopdediae, almanacs, gazetteers, etc.
Chris: we want to be fun for the user, but we must offer authority to the librarian customer. We will all be out of our comfort zone for a while.
LIS educator: we are trapped! Cri de couer; we donâ€™t know what is going to happen, butwe have to keep all segments happy.
This started devolving into a wonderful conversation between vendors, publishers and audience–impossible to credit who said what, plus questioners would amplify, translate, perform! Ideas:
Jimmy: Wikipedians (my term) get to be quiz show winnersâ€”exercise serendipity by following links.
John: we canâ€™t prejudge what users need. Publishers need to present hundreds if not thousands of results on one page to show context. J.L.: Serendipity abounds on Google. Now using format icons to present context.
Expanding Johnâ€™s comment that â€œads arenâ€™t it; free isnâ€™t itâ€?
John: donâ€™t libraries need more than one encyclopedia? Need best thoughts of best thinkers. Need editors. What is the business model? Look at extrapolations. Ads donâ€™t meet academic approbation. But academic journals often have ads. And open access publishers are using AdSense to generate some revenue.
Google Scholar is looking to aggregate all emanations of an article. Thomson Gale wants any content of value. Expansive search vs. focused search. Learning styles. 8 modalities of reference inquiry. Sometimes source matters but sometimes it doesnâ€™t. Creating content is a form of service. The classics were gatekept. Librarianâ€™s role is getting to the right question. How is 24/7 reference mediated? We need to build bridges. Need diversity of sources and contexts.
I think this session might have gone on happily until all were completely exhausted. But someone else needed the room. The moderator mentioned the panelists had a lively lunch beforehand. The give-and-take on the panel and with the audience was wonderful to see. Information=conversation indeed! There was agreement and disagreement but there was listening on all sides. Hmm. How can the range of conversation, including the front-line librarians, LIS educators, info literacy, etc., continue? Maybe a wiki? Kudos to panelists and questioners!