American Libraries |
||
Site NavigationLeft Sidebar ItemsOnline Features
Follow American Libraries news stories, videos, and blog posts on Twitter.
|
||
The Technology of Patriotism
Head of Systems, North Carolina State University Libraries, Raleigh. Column for November 2004 By the time you’re reading this, the election is likely over. I won’t try to say that I predicted the outcome; suffice it to say I am either surprised or depressed. But I could not leave it at that, not in this political month. So, presidential politics aside, it seems like a good time for a brief discussion of political technology, which more and more these days seems to center around the technology of privacy (or the lack thereof) and security (or an inability to ensure it). Defy or complyWhile librarians continue their struggle at USA Patriot Act defiance in order to protect patron privacy, software companies outside the library market are making millions off government-mandated compliance to all those other sections of the Patriot Act that deal with increased information-sharing and searching for terrorist funding. These complex provisions make filtering look like (forgive the pun) child’s play. That’s right—if you think there’s something hidden in your data, you might take a look at solutions from companies like SearchSoftwareAmerica, providing products and software to comply with the Patriot Act that “solve the inherent error and variation in name and address data and deliver high quality data integration, duplicate discovery, and relationship linking.” Innovative Systems (not to be confused with Innovative Interfaces, the library-software vendor) offers a free Patriot Act compliance guide, as well as name-recognition software and data-warehousing solutions. These are just two examples of the kinds of companies that might have sought bids had the Orwellian Total Information Awareness effort—the ultra-database of financial, medical, communication, travel, and intelligence records scrapped by the Defense Department a year ago—ever made it off the ground. Truthfully, I am not trying to be glib. I am not in favor of hiding the identities of terrorists; but data mining casts a wide net, so I also don’t think we can let the civil liberties of our patrons be destroyed along with the targets of terrorists. Confidentiality or security?In the technical realm these days it seems like there are mutually exclusive battles between filtering and unfettered access, between anonymity and personalization, and between system flexibility and technical security against worms, spam, and pop-ups. The very security that protects us has the potential to work against us. There are dozens of companies like those named. The question for libraries is: What are you doing to safeguard the freedom of your patrons? Technically speaking, how’s your data? Can you identify the activities of individual patrons in your web-server access logs? Do you know the privacy policies of electronic database vendors who offer current awareness services through their own patron accounts? Does your integrated library system keep a borrower history or a fines history? Complete library surveillance might seem daunting to the Justice Department—there are a lot of libraries out there. But how many library automation and database vendors are there? Does your vendor have root access to your library servers? Does your library have a privacy agreement with its chat-reference provider? What is your library’s relationship with its Internet service provider (ISP)? "We protect each library user’s right to privacy and confidentiality with respect to information sought or received and resources consulted, borrowed, acquired or transmitted."—Principle III, ALA Code of Ethics, 1995 Look who’s listeningIf vended solutions above don’t seem scary enough, how about the government’s very own DCS1000, also known as Carnivore. The sophisticated FBI software can record almost any traffic from web surfing and file downloading to e-mail communication. Carnivore is the ultimate wiretap for the Internet. Carnivore’s unlikely presence on a library server is comforting; its more likely presence on an ISP or e-mail vendor is frightening. It has not been that long since the FBI’s Library Awareness Program (AL, Oct. 1988, p. 743) sent agents looking for information on suspicious patrons who looked like Eastern Europeans or had Russian-sounding names. Vigilance comes in political action, but it also comes in being good stewards of library data. Rhetorical protection is meaningless in the wake of technical inaction. Maybe by November 2008, I’ll be able to provide an update on software available to ensure the privacy, confidentiality, and even anonymity of library patrons. Would some of the billions going into data-mining development simply make my OPAC more searchable? I hope that libraries will have a privacy-audit boilerplate that they can send to their vendors to safeguard the data to which libraries have no proprietary access. I hope that libraries will broker more data for their patrons rather than maintain web pages that blindly link to third parties. I hope to see the sections of the Patriot Act most offensive to civil liberties sunset without much debate. I want to be surprised. Embracing Open AccessOn September 3, the National Institutes of Health released its plan for "Enhanced Public Access to NIH Research." Under the plan, peer-reviewed journal articles would be freely available through the PubMed Central database six months after initial publication (PubMed is maintained by the NIH). The Association of American Publishers is mounting opposition to the proposed plan. Whether reactionary or more of a one-two punch, BioMed Central (BMC) announced the following week that it would launch a service to maintain institutional repositories for colleges and universities. Based on the open source DSpace system developed by Hewlett-Packard and MIT, the system will store the intellectual fruits of research faculty labor. BMC will charge universities a fee to maintain the DSpace database. Announcements
|
Right Sidebar
|
|
