Librarians properly show concern for the confidentiality of patrons' reading and other behavior. At the same time, some within the profession (and elsewhere) argue that patrons don't care about privacy and libraries shouldn't insist on it. In recent years, the latter viewpoint is frequently stated in terms of "keeping up with Amazon" and suggesting that libraries will become irrelevant if they insist on confidentiality.

The first section offers one of several recent discussions asserting that libraries should not protect circulation confidentiality to the extent that most now do--and a series of comments on the issue. The second section offers a perspective arguing that libraries should protect circulation confidentiality. The third includes portions of an earlier commentary on these issues.

This set of issues involves a number of subissues, among them:

1. Will patrons abandon public libraries if they fail to provide circulation histories and social-networking services?
2. Do patrons fully understand the consequences of abandoning circulation confidentiality, including the simple fact that libraries cannot assure that stored circulation histories will be kept confidential from everyone but the patron?
3. Is it reasonable for libraries to be more protective of patron confidentiality than patrons might themselves be?
4. In the interest of serving those patrons who do regard shared reading histories as more important than confidentiality, is it enough for libraries to facilitate transfer of current borrowing information to third-party services, or must libraries develop their own social-networking services and retain circulation histories?

 [Editor's note:] Possibly based on my own work history, I am strongly, even vehemently on the side of confidentiality and the need to enforce confidentiality by not retaining circulation history. That bias is based not on ALA's principles--with which I wholly agree--but on my awareness that circulation history can and will be used for various purposes without regard to state privacy statutes or even federal law. What the FBI did manually in the 1970s can certainly be done in a more systematic manner in the future. Saying "let the user choose whether to retain privacy" only works as long as the government choose to respect that choice--and history demonstrates that government has, can and will demand access to stored past-circulation records, regardless of state laws.

What libraries can learn from Facebook 

by Peter Bromberg. Excerpted and adapted from this February 19, 2009 post at Library Garden. Used by permission.

A colleague and I were discussing [a] recent Facebook TOS kerfuffle and she said she was fascinated by how much privacy people are willing to give away in exchange for a desired experience. I agreed that I am equally fascinated, and that it is vitally important for librarians to be on the vanguard of monitoring these trends and educating our customers as to the possible risks of sharing too much information.

But I also think that librarians, at times, can be too knee-jerk about privacy issues, and I wonder if while looking at one end of the Facebook dustup (big corporation trampling on privacy rights) we might be missing some important lessons on the other end (big corporation letting customers control their own information in exchange for a highly engaging experience...

A Quality Experience Trumps Privacy?

We all know that people (myself and probably you included) will share personal information in exchange for a quality experience. We share personal renting and buying habits in exchange for Netflix and Amazon recommendations. We share personal reading habits on GoodReads and LibraryThing to connect with others who share our interests and tastes. We share our credit card numbers with many online vendors in exchange for the convenience of "one-click" ordering.

Should Libraries Favor Privacy?

We know all this, and we personally experience the benefits, but librarians still seem generally loathe to let our customers share their personal information in exchange for anything. We don't just protect customer privacy, we paternalistically protect it from the customers themselves, rendering them childlike. Our privacy philosophy often reduces down to "We know better" or "You can't be trusted with that--you'll hurt yourself."

Our choice to disallow customer control of their own information means that their needs for connection and social networking go unmet, which in turn creates opportunities for entrepreneurial companies like Library Elf, GoodReads, and LibraryThing (created by frustrated library lovers, I wonder?) to come in and fill those needs. Which is great, but why aren't libraries creating and offering these experiences?

I worry every day about whether libraries will be relevant, three, five, or ten years from now. Unless we start allowing our customers to make decisions about their own personal data and start building systems that offer them a social networked experience based on their ability to selectively share their heretofore private info, I fear that libraries will grow increasingly irrelevant to our customers.

Excerpts from Comments

• Kathy Dempsey: I worry about relevance too. This whole social network/privacy overprotection thing is only one reason why... libraries have a lot of things to address to stay relevant. It's always a toss-up between falling into these new models vs "tradition." Tradition sounds great, till you start to realize that, eventually, those who remembered and valued those traditions will all be gone.
• "Andy W." on February 20 at 12:55: When personal information is treated as a legal liability, there is going to be an overprotective attitude to it... What you perceive as paternalism on behalf of the patron is really the library protecting itself from any potential security risks involving patron information...
     With that said, I don’t see the future of libraries as building social networks or creating ways to facilitate the sharing of patron information. I am more in favor of utilizing what currently exists. There are current third party entities (such as the ones you specifically mentioned in your post) that are able to take on the privacy liabilities and security concerns better than libraries can currently handle. The creation of bridge products that will allow libraries and these sites to interact with each other is the most viable and practical path at present. I think this is as an inevitable development as the web brings everyone and everything closer. I think with the proper outreach we can position ourselves as a facilitator and supporter to these outside sites and create mutually beneficial relations.
    Personally, I think there are bigger issues about privacy within the library community that really control the conversation. There is much graduate school classroom time dedicated to the value of freedom of inquiry and the various contemporary historical specters that have threatened this over the years (e.g. censorship, McCarthyism, the Patriot Act). This sets the tone and maintains the so-called status quo around patron privacy.
• "deepening": I often wonder about libraries trying to "keep up" with commercial enterprises. Are we really competing with booksellers, cafes, and video stores? I don't think we are. I think we complementarily occupy neighboring spaces.
    Which is background that makes it easier for me to say that libraries are different. There are businesses that connect folks with books, with information, with entertainment. But we do it for free. We do it to ensure that the opportunity to access learning, knowledge, media, exists for everyone.
    So, if you want to play with your own borrowing data, you have at least two choices:
  1. Buy your books online from a bookseller who will keep and track that data and won't protect it when you accidentally land on a no-fly list.
  2. Use the library and a third party add-on to play with your data.   Add your books into GoodReads and LibraryThing. Nobody is stopping you. Hell, the library will even provide all the data you need to add that title into Library Thing. It's your information, use it however you want. And whatever you do with it, it's your choice.
  Are we really wrong for committing over the decades to protecting user borrowing habits from anyone's eyes? I see it as bedrock, just like freely available access to information in the technology du jour. Just like not being in it for a profit.
• Peter Bromberg: Kathy, thanks for your comment. I think we're basically in agreement however, I wouldn't say that it's a "toss-up" between tradition and new models; I'd say it's about discovering new ways to express our traditional values. Certainly one of those core values is the protection of patron privacy--I just question whether it is useful expression of that value to protect patron data from patrons themselves...
    Andy, I think we have a few fundamental disagreements... I disagree that libraries adopt strict privacy policies to protect themselves. Protect themselves from what? Can you point me to any data, articles, etc. backing this up? I've never heard this suggested before...
    I would be very happy if libraries would open up their patron data and give patrons the option of sharing that data with 3rd party systems, but I don't see that happening too much outside of Library Elf--which doesn't provide a social networking experience, merely a convenient reminder service--something libraries really should be offering themselves IMHO.
    Deepening: Are we really competing with booksellers, cafes, videos, etc.? I don't know. I agree that we occupy a complementary space (well put.) But I also think it's helpful to approach planning and assessment our services with a good old SWOT (Stregths, weaknesses, opportunities, threats) analysis, and I would be surprised if bookstores, Netflix and Starbucks didn't show up on most library's SWOTs! Futurist Faith Popcorn says "time is the new money" and to the extent that people have less and less time, we are in fact competing for their time and attention.
    Deepening, you also say that "libraries are different," and I agree. But nothing about the ways in which we are different suggests that we can't also allow patrons to control their data and allow them to interact with each other. To suggest that doing so somehow conflicts with our mission to promote learning and knowledge is 180 degrees from how I see it. Learning and knowledge are not only, or even primarily about reading in a book. It's about connecting with others in conversation and debate. I learn more about great books to read from Good Reads than I do from my local library's ILS system for instance. GoodReads not only gives me great title recommendations, it also connects me with friends and acquaintances with similar interests and allows me to form relationships. Just think how useful it would be for researchers if the library allowed them to share and annotate their reading habits, share that information and connect with others doing similar research.
    You say nobody's stopping me from using my patron data and that the library will even provide all the data I need. But they don't! I can't easily access, manipulate, mix and mashup my library data. I'm pretty much forced to use GoodReads, Shelfari, or Library Thing. Most libraries don't even give patrons the option of saving their reading history, and those that do don't (to my knowledge) provide an API to get that data out.
    To suggest that if I don't like it I have the choice to go to a bookseller instead sounds to me too much like, "this is our library and if you don't like the way we do things you're free to leave." Libraries could maybe get away with that when we were the only game in town, but those days are over. My concern is that if we tell people it's our way or the highway, they'll choose the highway in a heartbeat.
    So to answer your final question: Are we wrong to protect borrowing habits from anyone's eyes? Absolutely, yes we are. We should be protecting borrowing habits from everyone except the patron themselves and whomever else they want to share that information with. I see that is the proper expression of our commitment to privacy.
• "Andy W.":...For me (and the ALA), library patron polices reflect a safeguarding of the First and Fourth Amendments and the freedom of expression that the entailing Supreme Court cases have defined within...
    Libraries develop subsequent privacy policies to protect themselves from the spectrum of legal ramifications...and in addition to (at best) patron information theft for use for junkmail and/or spam or revealing potentially embarrassing reading subject matter to (at worst) identity theft... Given the litigious nature of our society, this makes it a risk worth avoiding.
    ...What you are asking for is the ability to manipulate your own patron information how you see fit. I see the library's hands legally tied in this matter. However, I don't see it as the end of the road. I see room for development of third party programs that give patrons the interface that you are talking about. These programs would also need to improve upon the current bevy of sites that are available now. The alternative is a complex legal dance for people to waive their federal and state protections to share their library information directly; I cannot imagine any libraries eager to engage in this.
• "Anonymous": There's a huge difference between "open government" that allows me to view public documents and the idea that public access to records should extend to the personal information of patrons... In my state [library privacy] laws put strict controls on how personal information, including what I check out from a library, is handled and who has access to it.   These policies and laws exist to protect patrons from unauthorized access to their personal use of the libraries and the inherent threat to privacy that comes from allowing unfettered access to that information. If I was a system administrator of an ILS, I would have serious reservations about sharing patron data with 3rd parties. Who is "Library Elf" and what are they doing with that data? In my state, I don't even think the Library Privacy laws would allow that data to be shared with a third party service provider. I understand how that frustrates [some users] but in an area like this, I would err on the side of caution.
• Peter Bromberg: I did not suggest that library patron records should be open and accessible as some other government records are... There is nothing about my county government's approach to privacy that prevents my county library from giving me control of my own data.
     And that's the point that you and Andy and "Deepening" seem to be missing so I'll stress it again. When I talk about opening up the data, I'm suggesting that libraries give customers control of their own data.
    When I hook up with Library Elf for the benefit of text message reminders, my library has not given away access to my data. The have simply configured the ILS system to allow me to share my own data with Library Elf, in exchange for a service.
    See, I (as an adult) get to decide if Library Elf is trustworthy.
  I (as owner of my account) get to decide if Library Elf can access it.
    To sum up: I am only advocating that libraries give customers more flexible control of their own data so they can choose to share it with whomever they choose.
    It is also my wish that libraries would build some of these social networking features into our own systems so customers choose to play on our turf.
• Andy W. [omitting quoted material from New Jersey laws on library record confidentiality]:... It certainly helps to look at the actual law that controls in the state in which we both work in. Library patrons are within their rights to ask for disclosure of their information. However, inherent control of the record is vested with the library; it is up the patron to request disclosure.
    Now, this still leaves open an exciting conversation as to the nature of the disclosure. The social libertarian in me would say that disclosure should be anything and everything the patron wants, but the pragmatic side tells me that this is not going to be as easy...
    In a response to the...Patriot Act, some library systems scaled back the amount of information that a library automation system retains. In some cases...the library cut record keeping down to the bare bones to protect patron privacy from government intrusion. Even with revisions in the Patriot Act, I have not found any articles in which libraries have returned to previous record keeping policies.
    ...As a result of these draconic record keeping policies, there would be no borrowing history to disclose. You specifically mention in your original post how we trade shopping or borrowing history for enhanced services; what if there is nothing to share? The patron borrowing record would have to begin from the moment the patron exercised their library record rights; it is not for the library to maintain a borrowing record "just in case" a patron decided to keep one. In following your own example, it would take an action by the patron to ask that their record retain their borrowing history. So, there is the distinct possibility that whatever information is retained in a library record is not going to be much in terms of actual disclosure.
    With the exception of Library Thing, I don't see an automation or interface that allows for ease of disclosure with one of these third party sites. I would not put the onus on the library to meet the different demands of the various sites and services in order to comply with disclosure requests. This may be a case of waiting for the market to catch up and create an interface that would allow for a greater connection between library record and third party site or service. If the demand is great enough, it would certainly be in the best interest of the library to be the one attempting to bridge this gap...
    Bottom line: This now feels less like an issue of library policy and more about educating patrons on their rights. Beyond that, it is a matter of implementation.
• Peter Bromberg: Facebook just posted a draft of their "guiding principles" and they are asking members to comment on them--and then they're going to ask members to vote on them.   Now that's transparency! Sink your teeth into the first two principles (I'd like to see them adopted whole cloth into a Library User's Bill of rights...):
     Freedom to Share and Connect
    People should have the freedom to share whatever information they want, in any medium and any format, and have the right to connect online with anyone--any person, organization or service--as long as they both consent to the connection.
     Ownership and Control of Information
    People should own their information. They should have the freedom to share it with anyone they want and take it with them anywhere they want, including removing it from the Facebook Service. People should have the freedom to decide with whom they will share their information, and to set privacy controls to protect those choices. Those controls, however, are not capable of limiting how those who have received information may use it, particularly outside the Facebook Service.
• "Anonymous:" ...State[library privacy] laws err on the side of privacy and protection when it comes to patron information as compared to open access. Again, I can sympathize with your feelings that you "own" that information. But you don't. It's a government record about you and ultimately, it's the government, not the user, who is liable for the release of that information... Whatever Facebook does, they are not governed by the same laws of privacy nor I suspect, the same expectations of privacy...

Responding to "What Libraries Can Learn from Facebook"

Walt Crawford's response to the blog post: While I did not take part in this conversation, I would note a couple of things:

• The move to eliminate circulation histories and to provide state protection for record confidentiality predates the PATRIOT Act by a couple of decades, at least in many libraries and states. You can trace much of it to the FBI's library program of the 1970s. At the time, I was working in a library that was visited by an FBI agent requesting prior circulation records--and, back then, there was no state law protecting them. The institution's lawyer informed the library that (a) the records weren't legally protected but also that (b) the library was under no obligation to retain prior records. Before the FBI returned, all circulation history had been expunged, a practice that continued from then on. This was never a phantom threat or library paranoia; it has always been based on actual incidents.
• Federal courts can and do ignore state laws (ask reporters about shield laws!). The PATRIOT Act and similar legislation ignore both state and Federal confidentiality provisions. It's fairly clear that, at least in some administrations, agents are empowered to ignore such protections as exist. As a result, for a library to say "we'll protect your circulation data from everybody but you" is dangerously misleading nonsense. If the circulation history is stored on library systems, it cannot be fully protected from government intrusion (or private hacking).
• With all due respect to Faith Popcorn, it seems reasonable to suggest that most people now recognize that money is not an infinite resource--that "time is the new money" is a saying particularly appropriate to the bubble economy. I, for one, believe that we will not see a fast return to a time of no limits.
• One thing curiously missing in this post and in other posts I've seen advocating that libraries abandon strict confidentiality rules: Evidence that some substantial portion of library users are demanding such changes.

Peter Bromberg's Response

Thanks for posting my article, and the reactions to it. I just read your thoughts and my first reaction is that you're responding to points that I didn't make. I'm not sure if that was your intention--did you misread my writing, or are you setting up and knocking down straw men? I don't know. For example:

You write, "This was never a phantom threat or library paranoia; it has always been based on actual incidents." Well who said the FBI barreling into libraries and taking records was a phantom threat or paranoia? But that doesn't address my point about libraries, behaving paternalistically, it supports it. You seem to be suggesting that because there is a chance that the FBI might get it's hands on my library records, the library should not put me in charge of deciding whether or not to expunge my records (if I'm concerned about the FBI) or keep, manipulate and share my data in exchange for an experience I value. The library is in fact, saying, "We'll protect you--we know best." Aren't they? I suppose you could argue effectively that people can't be trusted with their own data; that they're not well-informed enough to make smart choices. But if that's your position just come out and say it.

You write that "for a library to say 'we'll protect your circulation data from everybody but you' is dangerously misleading nonsense." First off, who are you quoting here? The implication is you're quoting me. You're not. I didn't say that the library was protecting data from everybody but the customer. I implied (when I wrote that the library protect "[data] from the customers themselves") that the library protects data from everyone including the customer. Whether they want it protected or not. Again, if this isn't a textbook definition of paternalism, I don't know what is.

Regarding your comment on my quoting of Faith Popcorn (you wrote, "I, for one, believe that we will not see a fast return to a time of no limits."): I honestly don't even know what you mean. It doesn't seem that you are responding to my point which is that we are busier and busier. People value their time more than ever, and therefore we must realize that we are increasingly competing for the limited time of our customers. Convenience is becoming more of a premium, and what this has to do with a bubble economy is beyond me.

Lastly, you characterize my post as "advocating that libraries abandon strict confidentiality rules" and note you are not aware of library users demanding such changes. Well, you can characterize my writing any way you want I suppose. But I certainly don't think I'm advocating an "abandonment" (wow, that's a loaded word) of anything. I say that we should continue to advocate strongly for confidentiality laws that protect patron privacy, and protect their data from unwarranted government intrusion. At the same time, I suggest that we empower customers to make decisions about the deletion and sharing of their own data--and let them weigh the pros and cons, and decide for themselves if they want to share their data or delete it. As for whether library users are "demanding" it? I think that's an extremely low bar--so low as to be meaningless--as to whether or not giving customers control over their own records is a good idea or a bad one. Who "demanded" ATM's? Who "demanded" self-checkout. Who "demanded" Netflix? And let me ask you this Walt: If libraries implemented the changes I'm suggesting, would customers rise up and "demand" that their choice be taken away and that libraries go back to destroying all patron records whether or not the patron wants them? Here's my guess: No, they would not.

And Walt Crawford's Response...

1. OK, I will come right out and say it: Most people appear not to think through the consequences of revealing data very well. (That's one reason phishing and spam are as profitable as they are.) Yes, I do think agencies sometimes act to protect people from themselves, and I think that's a good thing. You could call the FDA's work "paternalistic" as well, also that of various licensing boards. I choose not to. Are libraries protecting people from themselves? Maybe so, and fine with me.
2. You are advocating a change to "we'll protect the data from everybody but you," and I say that's misleading nonsense: that libraries can't say that with any confidence.
3. What this has to do with a bubble economy is that, in my view of current reality, people are learning to trade off time and money: Convenience does not, for most people of moderate or modest means, trump cost in all cases. I believe that will continue to be the case.
4. When you modify confidentiality rules, you're inherently abandoning strict confidentiality.

I think your post and the comments provide a good and interesting position; that's why I expanded the extract from the original brief form. I also disagree with the position you're taking (and have done so for years--after all, the second part of the article is four years old now), and say so. Doesn't mean I'm right, to be sure.

Peter Bromberg, concluding the discussion:

Walt, thanks for putting our responses side-by-side, I appreciate that. I also appreciate your clarifications. I don't mind disagreeing; we have an honest difference of opinion. My concern was that I thought my points were being characterized inaccurately. Interestingly, I think that we basically hold the same values, but come to different conclusions about how to structure library policies and services when these values come into conflict.

Technology, Privacy, Confidentiality and Security

By Walt Crawford. Excerpted and adapted from Chapter 3 of Policy and Library Technology, the March/April 2005 Library Technology Reports. Used by permission.

With few exceptions, all libraries claim to protect user privacy and circulation confidentiality...

The fundamental principal of user privacy means that a user’s reading (listening, viewing) habits are strictly their own—that librarians don’t concern themselves with those habits and strictly protect that information from others.

Circulation confidentiality is the same principle, but in reverse and on an aggregate basis. It’s a relatively recent principle, at least in practice—after all, many public and other libraries used to use signature book cards, where past readership could be observed simply by reading the card...

Balancing New Technologies with Privacy and Confidentiality

Innovative librarians keep on the lookout for new technologies that can improve library service. Companies develop new technologies and uses and pitch them to libraries, pointing out the problems that the new device can solve. That’s as it should be; libraries have long been leaders in effective use of new technology, and should remain so.

Problems arise when new technologies and uses are implemented without considering the policy framework. Every technology, even seemingly minor ones, should receive at least a cursory policy scan.

If your library proceeds with a new technology that does affect privacy and confidentiality and you haven’t addressed those issues in advance, there’s a good chance someone else will address them for you...

When your users raise questions, you need to have answers. “We didn’t think about that” generally doesn’t serve very well as an answer.

Collaborative Recommendations and Similar Services

Why can’t library catalogs be more like Amazon? Variations of that cry have risen in various quarters. Depending on what “more like Amazon” really means, one answer is that many of them already have—catalogs showing book covers, including tables of contents, linking to reviews.

What some people mean by “more like Amazon” is a collaborative filtering and recommendation technology that suggests new items for your consideration, based on some combination of your own buying patterns and combined patterns of other purchases. “People who purchased x also purchased y” represents a simple form of collaborative recommendation; the technology can go much further.

Since this isn’t a discussion of Amazon, there’s no point in considering whether Amazon’s collaborative recommendation engine is unbiased. Some similar engines do appear to operate without bias (and to serve the company’s aims in doing so), with Netflix being one of the most widely-used. Netflix invites you to rate as many movies as you can. Based on those ratings, the records of what you’ve already viewed and liked, and similar records for a couple of million other viewers, Netflix can offer surprisingly apt suggestions for movies you might never have considered but will probably enjoy.

Wouldn’t it be great if a library catalog could do the same—offer a personal reader’s advisory that suggests some books (or CDs or DVDs) that you might really enjoy, based on your past borrowing and related borrowing records from other library users?

The technology’s feasible now, given cheap disk storage and high-speed computing. As far as I know, it hasn’t been implemented in public libraries.

Confidentiality Issues

The problem with collaborative recommendations is that to work really well, they rely on stored knowledge of your past history and that of others. How do you provide such stored knowledge without compromising confidentiality?

There may be answers to that question, but those answers require testing and thought. At first glance, it seems problematic. You could achieve one level of collaboration by only coupling items taken out at the same time and storing those links with codes that can never be linked to an actual borrower. Thus, you could say that “someone took out book a, book b, book c, and DVD d at the same time.”

If that pattern happens often enough, then you could suggest that someone else who checks out book a and book c might find book c and DVD d interesting. But that’s a weak database—and it will keep recommending books a user has already read, which is likely to be more annoying than useful.

You’ll have much stronger recommendations if the engine can track borrowing habits over time. I don’t know how you could do that while maintaining confidentiality.

There is a way to avoid the problem of recommending an already-read book over and over, but it involves significant overhead. If records of a user’s past circulation are only maintained on that user’s own PC (or better yet, on a flash USB drive), stored in some encrypted manner that only the library database can relate to actual items, those records could be used on the fly to provide new recommendations without necessarily endangering privacy or confidentiality, assuming a secure link is used for the process.

These aren’t trivial problems. They shouldn’t be solved by asking users to acknowledge that their reading history may not be private if they want new book recommendations. Library users don’t generally have or need the same background or depth of awareness of privacy issues as librarians.

It’s the job of librarians to maintain library principles, not to attract users to waive those principles by offering shiny new toys. I’m sure very few PC users want adware or spyware installed on their machines, but millions of them “signed” forms consenting to add such adware or spyware, so they could achieve some desirable end.

...It’s not at all clear that the supposed benefit of automated reader’s advisory outweighs the dangers, or that the dangers can be eliminated at reasonable cost. Before any such system comes into play, those issues need to be studied and resolved.

Online Access to User Records

This doesn’t require much discussion. You probably offer web access to your online catalog: most libraries do. There’s a good chance you also allow library users to view their current records—to see what they have out and renew items online. Many libraries offer that service.

Are you sure you’re not compromising privacy in the process?

Do you require that users register and create passwords before showing them their current item list? Probably not. Does the circulation information operate over a secure link? Again, probably not.

Does it matter? Possibly. If, by some chance, you allow users to login with only their card number, and if (worse) you then show them their record including name, then it certainly does. All an interested party needs to do is get a library card, figure out the range of numbers your library is using (and the check digit methodology, usually easy enough to determine), and the party can set up a harvester to associate all current circulation with the people holding the items.

But you probably don’t make things that easy. My library doesn’t use a secure link and doesn’t use passwords, but it does require that you enter your name and card number. If anyone else gets that information, they can check on your current reading any time—but at least that’s a smaller risk. Assuming, of course, that the database that links card numbers to patron names is truly secure. Is that a safe assumption?

Conclusion

Most new technologies don’t raise major policy questions. Many new technologies raise more policy questions than the average librarian wants to consider. The first step in making sure that technology doesn’t undermine policy is to consider the possibility.

Electronic Services and Security

By Glen Holt. Excerpted from a 2006 commentary in LLN.

Being a relatively heavy user of computers and electronic communications, I am about as paranoid about personal electronic security matters as anyone I know. But my paranoia is based upon my knowledge of electronics and my assessment of my risks. When I make financial transactions online, I take a good deal of care, minimizing my risks. However, when I make most of my information-searching efforts on the open Internet or even in databases I pay for, I could care less if anyone is looking over my electronic shoulder. My point here is that I set my level of security based on my tolerance of risk, and most of the time the work I do involves little risk for me or my clients.

A different situation prevails for my two scientist sons and one financial-analyst son in law when they search. The electronic masks their companies provide for them as they search electronically in open or specialized for-hire databases are far superior to any disguises that Western heroes and our nation’s spies ever wore. That doesn’t mean they have absolute security, but their security is as close to absolute as corporate money, the latest expensive technology and company policy can create and still do business electronically.

To summarize the current situation, information abundance and new electronic methods of searching have made security an even more relative discipline than it was previously. Most libraries still mandate and control all aspects of their patron security, many of them without having asked their users if that is what they want. In a world in which we know that different customers like different things, library security systems mostly appear to send the message, “Our way or not at all!”

Anyone who has ever been served with multiple subpoenas, as directors of big libraries often are, knows how little information many judges require before giving legal permission to a law enforcement officer to examine the library-use records of a particular patron. This legal shield exists in state legislation protecting most public libraries, however. That shield, minimal though it may be, is far less protective in school and university settings, as students of different ages have found after some hacker’s stunt using a library computer that they signed into with their student identification.

When the CIPA dragon came across the library landscape, we discovered just how divergent institutional security policies and practices were even in libraries that served similar social, economic and ethnic constituencies. Many of these differences exemplified below-the-radar attempts by local librarians to give regular constituents more room to make choices about their relative exposure to risk within their perceptions of the existent legal and policy frameworks.

I always worry when I find a single librarian or a library informally stepping around written law as upheld by the courts or institutional policy as adopted by governing officials. That is especially troublesome in public service delivery, especially security. I’d like to see libraries of all kinds make open attempts to find out what kinds of security their various users want when they are involved electronically with the library. Is a mother willing to be identified easily in some user group to receive RSS feeds about toddler-storytelling times at all system branches? Is a business man reading a Clancy spy e-novel willing to leave behind an electronic trail that shows he downloaded Chapter 7 on August 13 in order to receive Chapter 8 on August 14?

Security is relative because risk is relative. If libraries are really going to be self-conscious about improving customer service, they will have to recognize that different customers are willing to tolerate different levels of risk--and that they want different levels of institutional security applied to their various transactions.

An essential commentary on the issues I have raised in the paragraphs above can be found in an excellent article, Andrew Richard Albanese, "Google is Not the Net," in Library Journal, September 15, 2006. In this elegant summary, Albanese writes:

In just a decade, the evolution of the Internet has upset the balance in an information ecology that had served libraries for centuries... Blogs and RSS feeds have changed the way we gather and look at news. Wikipedia, the living, free online encyclopedia that anyone can edit, is changing how we think about reference. Services offered by Amazon.com and NetFlix are winning the hearts, minds—and habits—of our users.

“All of these things,” [library columnist Joseph] Janes says, “are opportunities.” Libraries today, he observes, cannot afford to be paralyzed... [I]n a world drowning in information, libraries should be more vital than ever...

“I’d love a NetFlix thing at my library,” Janes says. “If I could put ten or 15 things on a list at my library’s web site and when I return one item I get another one sent to me, how great is that? How hard is that?”

At present, very hard. Aside from the technological challenges, most libraries, fearful of hackers and subpoenas, won’t keep circulation records for more than a few days much less use that information to enable social networking services like NetFlix or Amazon’s book recommendations, services users have increasingly come to enjoy—and expect. “Since we refuse in libraries to create the minimum level of links you need to create social networking, we can’t create social networks,” [SirsiDynix electronic policy guru Stephen] Abram says flatly. “And it’s always easier to blame the network than actually look at ourselves and say we won’t allow even adult patrons to choose their own level of privacy.”

“From a library’s perspective,” Janes says, “they just don’t want the responsibility, and I understand the burden and can respect the professional ethic.” The Microsofts, Googles, and Amazons that do collect such information, he notes, do so not without controversy and have “huge enterprises” dedicated to protecting user information. But the fact remains that users readily hand over the same information to commercial ventures—to Google when they search, to Amazon when they buy books, to NetFlix when they rent movies.

“I think people are willing to do the same at the library,” Janes says. “My take is let the user make the decision. If I, as a user, want to give my information knowing it might be subpoenaed or hijacked, in return for the enhanced service I might get, it’s my choice."

Like so much else about libraries, issues about...social networking and database security need to be very much on the minds of library professionals. These are issues where we ought to be trying to lead, not follow, in answering the hard questions about improving electronic services and security options.

Related Articles

• Reading Matters - Jamie LaRue on literacy and reading.