SUBSCRIBE NOW!

Vol. 30 No. 4 — Jul/Aug 2006

A MINEFIELD OF OBSTACLES

There is a virtual minefield of technology obstacles when implementing virtual reference. Many VR software programs do not work well with certain browsers and operating systems—or with slower Internet connections. This is particularly true of co-browsing, one of the most important value-added features that VR software offers over instant messaging (IM).

Libraries must also contend with the increasing popularity of features intended to protect computers from intruders, such as pop-up blockers and firewalls. These often interfere with co-browsing but can also prevent users from accessing the service in the first place. If students can’t get to the VR service or if their session is unsuccessful due to frustrating technology problems, they will not be inclined to try again. Plus, the inconvenience of having to disable pop-up blockers and/or security features may not be worth the effort.

As if this were not enough, Microsoft released Service Pack 2 (SP2) for Windows in autumn 2004, which compounded the problem by adding additional security features that interfere with VR software. In response, libraries have started to explore alternative technologies for delivering virtual reference, such as IM.

THE LIBRARIANS’ VIEWPOINT

According to the literature and discussion lists, many librarians agree that, in terms of being able to offer the level of service we would like, the technology is simply “not there yet.” Libraries using co-browsing technology in particular have been frustrated with problems encountered on both the librarian and the patron sides. There is a wide gap between what we envision and what is actually possible, given our level of technology.

Steve Coffman and Linda Arret, once considered “pioneers” of virtual reference, wrote a two-part article in Searcher (“To Chat or Not to Chat—Taking Another Look at Virtual Reference,” July/August 2004, pp. 38–46 and September 2004, pp. 49–56) that called into question the value of this kind of service. They say that users have not embraced chat as a means to obtain reference services and point to the demise of commercial services, such as Ask’s Answer Point, Askme.com, Allexperts, and ExpertCentral, once considered a challenge to library reference desks. Nancy Kalikow Maxwell wrote a much discussed article (“The Seven Deadly Sins of Library Technology,” American Libraries, September 2004, pp. 40–42) in which she claims that virtual reference has been plagued by technical glitches and low enthusiasm. She thinks that librarians can provide better service though traditional methods, such as in-person or telephone reference. This disillusionment is reflected on discussion lists such as DIG-REF. A review of the DIG-REF archives [http://groups.yahoo.com/group/dig_ref] reveals that many posts are focused on problematic VR software technology and how libraries are trying to cope.

While some of the reasons for these problems seem obvious, it is useful for librarians to understand technical issues that prevent VR, and particularly co-browsing, from working. Inability to co-browse or use other advanced VR software features is often related to issues beyond the library’s—or perhaps even the vendor’s—control.

BROWSER INCOMPATABILITY

Remember, all browsers are not created equal. Microsoft Internet Explorer (IE) is generally the preferred browser for co-browsing—collaboration may not even be possible if patrons use other browsers. While IE is the most popular browser by far, it is certainly not the only one. Many patrons have a strong preference for other products, such as Mozilla or Firefox. In order to co-browse, patrons using browsers other than IE must either temporarily switch or be handled in “chat mode” only. Furthermore, users are often required to configure their browsers to work with VR software. For example, Tutor.com mentions in its Q&A section that “to participate in a co-browsing session … the patron must be using Microsoft Internet Explorer (IE) under Windows and IE must be configured to use the Microsoft VM on their computer.” Docutek patrons using IE must also download Java Virtual Machine if it is not already installed on their computer.

While many libraries post instructions on their Web sites, it is difficult to determine how many users lack the technical expertise, patience, and/or inclination to go through a configuration process or download a piece of software onto their computer. It is reasonable to assume that this has an important impact on usage. Some vendors have attempted to address these issues as upgraded versions of their software are launched, but as more Internet users discover browsers other than IE, it will become increasingly difficult for vendors to keep up with making their products compatible. Libraries must be aware of how their VR package interacts with different browsers and decide on how best to accommodate their patrons.

OPERATING SYSTEM ANOMALIES

Many VR programs also run into trouble with users who are not using a PC with the Windows operating system. For some libraries, it is impossible to co-browse with Macintosh users or to use other features such as form sharing. With some VR software, it is possible to co-browse Web sites, but librarians cannot access licensed electronic resources with Macintosh users. Due to the relatively low numbers of Mac users, some VR vendors claim that they have no plans to make their products Mac compatible.

Another issue that has always plagued VR services relates to Internet connection speed. Obviously, a high-speed Internet connection on the patron side is ideal for receiving VR services. Patrons using dial up may be unable to co-browse even if they are using a properly configured, preferred browser. There is little that libraries or vendors can do about this. Although more and more people in North America have access to high-speed Internet service, many still do not. Again, this has a clear impact on usage.

FIREWALLS AND POP-UP BLOCKERS

Incompatible browsers and different operating systems have always plagued real-time VR, but the increasing use of pop-up blockers, firewalls, and security patches has compounded the problem.

According to Webopedia, a firewall is a “system designed to prevent unauthorized access to or from a private network.” Firewalls are often used to “prevent unauthorized Internet users from accessing private networks connected to the Internet.” In other words, firewalls are designed to protect computers and networks from unwanted intruders. It is easy to understand how firewalls interfere with features such as co-browsing, which involves one computer taking over the browser of another computer. Firewalls work to prevent others from doing just this sort of thing. Firewalls have been common for some time on computer networks to protect institutions and companies but are becoming increasingly popular on home personal computers as well.

Pop-up blockers are fairly self-explanatory. Internet users are assaulted on a daily basis with pop-up windows attempting to sell them everything from bogus university degrees to medication. People love pop-up blockers because the blockers prevent these annoying windows from being displayed. The problem for virtual reference is that many of the software products use pop-up windows. If a library is using software that generates pop-up windows, staff and users may not be able to access the service or may be required to disable their pop-up blockers beforehand. Many users may not know how to do this, and even if the library posts instructions, patrons may consider this to be more trouble than it is worth.

SECURITY PATCHES

The prevalence of viruses and worms being transmitted over the Internet has encouraged more people to install various security patches on their computers. Again, while this practice is not new for large organizations, it has now become very common for personal users at home. In 2004, Microsoft introduced Service Pack 2 (SP2) to the Windows environment. SP2 is a patch intended to provide “better protection against viruses, hackers, and worms,” according to the Microsoft Web site. It includes a firewall, a pop-up blocker, and other security features. While SP2 is intended to protect computers and networks, its security features have wreaked havoc with some VR software products.

Libraries have reported a series of problems related to SP2. While vendors have worked hard to address these problems, some have responded more quickly than others. Still, libraries continue to experience problems—and the introduction of new security patches will likely challenge the vendors’ ability to keep up.

SOFTWARE TESTING

The University of Guelph, which has offered virtual reference since May 2001, was one of the first academic libraries in Canada to do so. In 2004, we switched to a different VR software product and experienced a significant drop in usage, losing 30 percent of our users.

This drop coincided with the launch of SP2, which was installed on all computers in the university library. SP2 is usually installed on new personal computers running Windows as well. Thus, it was difficult to determine if the decline in usage should be attributed to SP2 or to the software itself. This question required further exploration. However, before deciding to switch back to our original software vendor, we wanted to thoroughly test how different software products interact with SP2, pop-up blockers, firewalls, and other security features.

Reference staff at the University of Guelph Library conducted an evaluation of virtual reference software products. We also conducted a series of interviews to find out what other libraries had been experiencing. We decided to test three products: Docutek’s Virtual Reference Librarian, Tutor.com’s Virtual Reference Toolkit, and OCLC’s 24/7-Question Point product. We had originally planned to test Question Point and 24/7 separately, but the two companies merged during our testing period, and the 24/7 platform was adopted with some modifications, so we concentrated our testing on this new product.

ROLE PLAYING

In order to reproduce the user’s experience, the staff members assigned to “play the patron” during some testing sessions were asked to log in from a public computer rather than from their desks. Knowing that pop-up blockers and SP2 were installed on all computers in the library and on an increasing number of off-campus computers, we decided to test how each software product functioned on these computers, as well as on our staff computers without SP2 or pop-up blockers installed. Testing was conducted over a period of several weeks. Staff members were asked to rate the quality of individual features such as co-browsing, ease of use, and user friendliness. The testing allowed us to develop an inventory of problems encountered. The problems noted were confirmed by our interviews with other libraries, most of whom were experiencing the same problems we encountered while testing these three products.

Many of the problems fell into two categories—access and co-browsing. In terms of access, firewalls and pop-up blockers seem to be major culprits. Users attempting to log in from public computers or browsers with pop-up blockers installed were often confronted with a message advising them to disable their pop-up blockers. They were required to do so in order to proceed. However, even some experienced staff members were not sure what to do upon receiving this message, so it is clear that many users would be confused and perhaps simply give up.

Other problems encountered included students not being able to log in and being “turned away” from the virtual reference service. Other sessions crashed before they even began because pop-up windows were blocked. Another common problem has been referred to by some librarians as the “disappearing student syndrome.” The student logs in, sends a question, and then seems to disappear. Some libraries have responded by posting instructions on how to disable pop-up blockers, but it is unclear whether students are willing to take the time to do this.

CO-BROWSING PROBLEMS

In terms of co-browsing, firewalls and security features such as SP2 are very problematic. We found that the software packages worked only sporadically on our public computers. In some sessions, we were able to successfully co-browse, but in most cases, co-browsing features did not work at all. Nothing was visible on the user side. This was generally not apparent from the librarian side, as the interface indicated that co-browsing was functional even when this was not the case. Therefore, we found it necessary to use the chat feature to ask our colleagues on the user side what they could and couldn’t see. It should be noted that we also tested the patron interface on staff computers without SP2 or pop-up blockers installed. We were forced to remove these features in order to use the staff side interface. Relatively few problems were encountered and co-browsing sessions went far more smoothly, confirming that the various security features were interfering with the VR software.

The problem is difficult to resolve. It would be unreasonable to ask patrons using VR services to remove security features from their computers as they would be at increased risk for viruses and worms. Our testing determined that while different software products do seem to interact differently with firewalls, pop-up blockers, and SP2, no software package is immune from these problems. Contrary to what we had hoped for before the testing, no product stood out head and shoulders above the rest.

INTERVIEWS

We also felt it was important to determine what kinds of experiences other libraries have had with each of these products. Approximately 15 to 20 U.S. and Canadian academic and research libraries were contacted, mostly by e-mail, for each of the three products. Libraries were asked specifically about problems with SP2, firewalls, and pop-up blockers, but also about any other general access and technical problems with their software. Comments from the DIG_REF list were also recorded as a part of this process. The results seemed to confirm that while VR software products behave differently with features such as firewalls and SP2, all products encountered problems.

Libraries using Tutor.com seemed to have fewer problems attributable to SP2. Most libraries who had installed SP2 reported that their usage remained steady, and that they were not aware of any problems on either the user or patron side. What problems were reported seemed to have occurred on the librarian, rather than on the patron, side and have been resolved by a patch released by Tutor.com early on. With regards to pop-up blockers, most Tutor.com libraries agreed that pop-up blockers were not an issue on the patron side, as the user operates within a single window. Some libraries reported that staff members must deactivate pop-up blockers on their side, but this is much easier to deal with than forcing patrons to do so.

The responses from 24/7 users were mixed. Some libraries reported that while they have not noticed any problems due to SP2 and pop-up blockers, they were frustrated with the software for other reasons. Others claimed to have experienced significant problems with SP2, firewalls, and pop-up blockers. Some of these libraries have dealt with the situation by posting messages on their Web sites advising students that virtual reference is not compatible with SP2, and by providing instructions on how to disable the SP2 firewall. While following these instructions does seem to work, it is generally acknowledged that it creates an important barrier to using the service, as many students will not bother to do this.

BRINGING USERS BACK TO BASICS

In addition to upgrading their software to more “SP2-compliant” versions, both of these products have dealt with some of these problems by allowing users to be “brought down” to basic (non co-browsing) mode. This allows librarians and patrons to interact even if the patron has security features installed or is using a noncompatible browser, but loses the value-added features of co-browsing.

In the case of Docutek, most libraries reported technical problems related to SP2, firewalls, and pop-up blockers. Librarians claimed that a growing number of users had difficulty accessing the service in the first place, and that co-browsing had become almost impossible. SP2, in particular, was a major problem for many libraries, and some libraries reported a significant drop in usage since its release. One library that had recently switched to Docutek noted that it had lost almost 30 percent of its users since making the change, a statistic remarkably similar to the drop experienced by the University of Guelph and its partners. Again, some libraries have responded by posting instructions on their Web sites to help users deactivate SP2 firewalls and pop-up blockers, but noted that this is hardly an ideal solution. Docutek has since released a SP2-compliant patch, which seems to have addressed the major access problem for some libraries. Co-browsing continues to be problematic, however, and many users must be handled in basic mode.

It is important to point out that in terms of general satisfaction, few libraries were completely pleased with their VR software, and over 90 percent of libraries contacted reported that they are experiencing technical problems.

ENTER INSTANT MESSAGING

With dissatisfaction on the rise, a number of libraries have started to shut down their virtual reference services, while others are considering this action. Another option many libraries have been exploring is using IM software to offer virtual reference. IM software supports Internet-based synchronous text chat. Some of the more popular programs include MSN Messenger, AOL Instant Messenger, Yahoo! Messenger, and ICQ.

For many, the benefits of using IM outweigh the downside of losing value-added features such as co-browsing, with its related technical difficulties. First of all, people (particularly young people and students) are already using IM. They use it regularly to socialize and are both familiar and comfortable with it. A Pew Internet & American Life Project study, “Teenage Life Online: The Rise of the Instant-Message Generation,” conducted in 2001 [www.pewinternet.org] found that 74 percent of online teens use IM and 69 percent of these individuals use it regularly. The authors conclude that teens have embraced IM in ways that adults have not.

Librarians who offer IM feel that by providing IM reference, they are aligning their services with the preferred technology of an important user group. IM allows patrons to use the technology that they prefer rather than forcing users to communicate with librarians using “library technology.” Particularly vocal IM advocates include authors Aaron Schmidt and Michael Stephens (“IM Me,” Library Journal, April 1, 2005, pp. 34–35) and Aaron Schmidt and Sarah Houghton (“Web-Based Chat vs. Instant Messaging,” ONLINE, July/August 2005, pp. 26–30).

There are various other reasons that make IM an attractive alternative. Unlike VR software, IM is generally free. It is easy to use, from the patron’s perspective as well as on the library side (which has training implications). Unlike VR software, IM works with most computers, operating systems, and connection speeds. It does not require the patron to download software or configure a browser. It is also faster, as patrons who already use an IM system can simply add the library service to their list of “buddies” and message a librarian when they require assistance. VR software requires patrons to fill out a form, log on, and in some cases pass through a series of configuration tests. Of course, IM does not offer the value-added features of a more complex VR system.

RESPONDING TO VR PROBLEMS

Technical problems have always been a barrier to providing real-time virtual reference. Users of the three major VR software products reported significant technical problems. While users of certain software packages have experienced more access issues related to firewalls, pop-up blockers, and other security features than others, frustration with all VR software remains high. As more libraries adopt IM as a means of communicating with patrons, it is likely that more research on its value will emerge. It would be particularly useful to study patrons’ perceptions of VR versus IM. If given the choice, would patrons prefer the ease of using a familiar technology such as MSN Messenger, or do they see an important value added service in the features offered by VR software?