Sacred Trust: Libraries and Patron Information

Sunday, October 19th, 1:00 to 2:30 Presenter: Amy Benson of Radcliffe Institute for Advanced Study at Harvard <>
Amy Benson

  • How social networking and 2.0 technologies impact privacy issues and policies.
  • How libraries can use patron information to customize and individualize services.
  • What must be kept private? There are issues of trust with regard to customization and targeted adverts.

Discussion of libraries & privacy: legal aspects, technological aspects, mindsets, opportunities, expectations offline & online, on both the library’s website vs in the physical library

In the “Real” World, both online and offline, we leave footprints, often without knowing it: Consider requests for your zip code and tel #, the proliferation of discount/loyalty cards, online credit card transactions, cell phone use. Consider the dry cleaner who can ascertain your personal information from knowing your phone number, use of EZ Pass (individual tracking could get you a speeding ticket, while aggregate tracking could keep you out of a traffic jam by providing accurate traffic reports).

In the “Web” World, sharing personal information is a requirement Searches, email, maps -all require sharing personal information.

“Hanging out” on the Web Americans watched 558 million hours of online video during the month of August 2008, according to

Users should:

  • Check set-up applications carefully. Facebook shares your personal info with other “Know who I am and access my information” applications.   You can set categories of viewers and individual viewer settings as well.
  • Read Privacy policies
  • Note Trust/Privacy symbols on websites.

Users want control but not barriers to sharing & collaborating.

Library website use declined between 2005/07, the only segment of web search that declined- young people don’t find library systems intuitive so they turn to Google and Yahoo. They are used to having their info tracked, and it doesn’t particularly bother them. The barriers of log-ins and authentication checks irritate them. According to a recent OCLC survey, consumers believe that information found on the Web is as creditable as that found in a library.

User profiles and histories allow sites to supply targeted info – Amazon, Gmail, iTunes, Facebook’s newsfeed. Young users rely on this customization and don’t mind giving out personal information to get it. Users self-filter through customization. Examples: Pandora, the music genome project, allows users to their music preferences in order to hear “sound-alikes.” Downside: The user can be profiled into a cocoon and never challenged. Using Rollyo (Roll-your-own-search engine), a library could customize searches in preselected sites with a roll-your-own searchroll designed for a certain demographic. Use Google web history (tracks your own search history) so you can track your searches – any registered user of a Google product can use this.

Questions to keep you up at night Filtering narrows the information flow. Can libraries automate information delivery? Can libraries allow people to customize the flow of information from the library to them? Do we librarians want to go in this direction? If the trend is toward self-service, where does that leave libraries? What value can we add? What if the only economically sustainable model is to generate revenue based on content supplied?

What is data made of:

  • personally identifiable
  • Anonymized (anonymous)
  • Aggregated
  • Archived

Lack of control: Data is often hosted on systems and servers that we have no control over (Ex: Flickr – Note here the value of and difference between individualized tags rather than more formal objective aggregate ones) See YouTube: Supermarket 2.0.

What do they do with all that data? Personal information is a very valuable commodity, especially the preferences and habits of consumers. Analysis of this information is used for creating personalized suggestions for purchase (targeted adverts), as well as aggregated feedback for what is popular. Example: Gmail electronically generates ads based on content of emails in return for the free email service. Look at Google’s privacy policy. Note: If libraries offer customization based on personal profiles, patrons must have the option of opting in or out of this service.

Consider what aggregated data can produce, like Wikipedia, exemplifying individual personal tags versus collective wisdom. Google Image Labeler, Who is sick? Google floats beta tests out there all the time. Check out “Big Brother Pizza Shop” on YouTube posted by the ACLU.

It all comes down to trust: No such thing as a free lunch. We give up a measure of privacy for a measure of convenience. Users must be allowed to judge the trade-offs Steven Colbert “Wikiality” Libraries can bring more trusted and vetted resources to the table. is accessible to anyone, LibraryThing

Libraries can make use of personal information to customize, but patrons must have the option of opting in or out of these customized services. According to an OCLC survey, personal information on the Internet is more private now than 2 years ago; 52% of young people are less likely to feel the privacy constraints. 60% of respondents trust the library. Only 11% of respondents rate activity on a library website as private. On the other hand, Web searches were considered private by more than 25%. 54% look for security icons. Respondents had reservations about giving out credit card #s and phone #s. 24% are unsure if the library website has rules on how private data is used. Note from Jay the blogger: don’t take the above stats as gospel, “The hurrider I went the behinder I got” (paraphrased from AA Milne’s Winnie the Pooh).

Librarians should check their state’s RSAs, not just the legislation but also case histories, etc.

Patrons must retain control over their personal information. There is a balance between privacy and convenience, transparency vs restriction (log-ins, etc) which should be thoroughly explained to patrons.

In conclusion: libraries should:

  • Eliminate barriers
  • Strive for transparency by letting users know what to expect, and how to opt in and out of these personalized services.
  • Strive for ease of use
  • Serve as information guide and trusted source
  • Permit end users to contribute content on and off the library webpage.
  • Do more with the data you are currently collecting
  • Move from counting stats to watching users to understand their needs, habits and desires and to capture and analyze user behaviors.
  • Use that data to recommend, suggest, serve, and assess those services.
  • Offer users ways to contribute and collaborate (Use focus groups or bibliographic classes).

Privacy policies to peruse: Boston Public Library, ALA code of ethics, Danbury Library,


One Response

  1. What’s available around the web for library users interested in tracking their progress on library reference desk enquiries or letting others know of their progress on library reference desk enquiries?… anything at all?

Comments are closed.

%d bloggers like this: