Dr. Joanna Richardson [HREF1], Library, Bond University, Gold Coast, Queensland 4229 Australia. Joanna_Richardson@bond.edu.au
Janet Fletcher [HREF2], Library, Southern Cross University, P.O. Box 157, Lismore, NSW 2460 Australia. jfletcher@scu.edu.au
Alison Hunter [HREF3], Library, University of Southern Queensland, Toowoomba, Queensland 4350 Australia huntera@usq.edu.au
Philippa Westerman [HREF4], Library, University of the Sunshine Coast, Maroochydore DC 4558 Queensland Australia pwesterman@usc.edu.au
Many organisations conventionally offer face-to-face assistance to their clientele at an "Information Desk". This service has been adapted to the Web in part via a form-based service commonly called "Ask A . . .". This paper examines both the IT aspects and key organisational issues in establishing an electronic reference service of this type.
As the Web has become more pervasive--both in terms of people accessing the Web and of organisations establishing their own websites, it is a logical extension of the current face-to-face and/or telephone "Information Desk" in an organisation to provide an electronic method for the user to ask questions of staff. For purposes of this paper, the authors have defined "electronic reference" as a "reference service designed for remote users and identified by a specific link from [an organisation's] website. Electronic reference queries that come directly to individuals or via other [organisational] web links [are] not included in the definition" [HREF5]. In this context a "reference query" is simply a question that needs to be answered.
Lankes (1998) explains the relationship between "digital", ie electronic, reference services and AskA services:
"Digital reference services are Internet-based question-and-answer services that connect users with individuals who possess specialized subject or skill expertise. As opposed to static Web pages, digital reference services use the Internet to place people in contact with those individuals who can answer specific questions and instruct users in developing certain skills. Digital reference services are also referred to as AskA services, because of such service names as Ask A Volcanologist, Ask A Scientist, etc. ... AskA services normally respond to individual queries on a one-time basis."
Libraries are an excellent example of organisations which not only have a large body of information that they would wish to digitise but also have as one of their primary roles to match the user's needs with that information. Other examples would include government agencies, academic institutions, professional associations, and companies that specialise in a given subject area. These organisations can now offer their users the prospect of access to electronic resources at their convenience temporally and spatially. Users do not have to be concerned with the physical hours of operation, and users do not have to physically go to the organisation to access resources [HREF6].
However, despite this apparent ease of access, Fletcher [HREF7] has highlighted the importance for the user to be able to communicate with staff when assistance is required. So in a world in which organisations endeavour to remain relevant to their clientele, there is a shift toward "put[ting] a human face" [HREF8] on their information services. The "AskA" service is one method of establishing a human presence in the midst of a plethora of links to resources, collections, and facts.
This paper examines some of the issues associated with offering such a service; its objective is to present these issues in the form of a checklist for both existing and proposed AskA services. While the initial investigation by the authors was focused specifically on libraries, the results of that work apply to most AskA initiatives. In fact, many of the issues raised in this paper would apply equally as well to any site that invites users to provide comments/feedback via a web-based form.
As a result of a report by Janet Fletcher on a paper given by Anne Lipow [HREF9] at the Information Online & On Disc 99 Conference, the QULOC (Queensland University Libraries' Office of Cooperation) Networking Working Party discussed the perceived merits of member libraries offering an Ask A Librarian service. After all, if commercial web sites offer the viewer an opportunity to ask a question --either via e-mail or a form-- then why not libraries? The WP decided to investigate and report on the associated technical issues, blissfully unaware that at exactly the same time the Association for Research Libraries (ARL) had published on its web site a survey [HREF10] on precisely the same topic, followed by an Executive Summary [HREF5] in late 1999.
A subcommittee, comprised of Janet Fletcher, Alison Hunter, Joanna Richardson (Chair) and Philippa Westerman, devised a questionnaire (Appendix A) which was sent to 144 tertiary and public libraries in North America and Australia. The subcommittee has received 99 responses. The questionnaire goes beyond purely technical issues and attempts to gauge the level of success which such a service is deemed to provide in the respective institutions. The Networking Working Party was not interested in technology for technology's sake but rather in its integration within an existing key library service.
Respondents were sent a follow-up communication (Appendix B), in which the Subcommittee Chair shared some preliminary observations, requested permission to quote the respondents, and enquired about the use of any automated acknowledgment script in conjunction with the service. This paper represents a report on the results of the responses to the questionnaire as well as any follow-up communication; discussion of these results has been expanded to incorporate non-library AskA services as well.
The objective was to find library sites which offer a service whereby someone can send a question to a reference librarian (desk) via the web. Entering the phrase "online reference" (plus "desk") in a search engine retrieved sites which offer links to online reference tools but not generally a librarian. However, by using the phrase "ask a librarian", one tended to retrieve exactly what the subcommittee was looking for. Many of the services actually call themselves "Ask A Librarian" or a close variant; in fact the majority do just that. Westerman used ""electronic reference desk" as her search term and although she retrieved a number of commercial services, she also retrieved some library-based ones, including NERD [HREF11], which combines a web form with an FAQ database and e-mail.
Target sites were identified from a list published by Sloan [HREF12] as well as by means of the search strategies already described above. The subcommittee members used the respective library's "Ask a Librarian" web form in order to submit the questionnaire. If either a web form was not available or a patron ID was required, then e-mail was used.
In writing this paper, Richardson used the phrase "AskA" in a search engine to retrieve a number of well-known, non-library sites. An excellent list of such services can be found at the Virtual Reference Desk AskA+ Locator [HREF13] project, which comprises high-quality question-and-answer services that respond to questions ranging from archaeology to online computing.
The original intention of the QULOC subcommittee was twofold: (1) identify any technical/IT issues involved in offering an "AskA" service and (2) assess its perceived usefulness at those sites which have implemented it.
As replies were received to the questionnaire, the authors quickly determined that the IT required to establish a basic non-interactive service was quite minimal. However, it was the actual process of linking to the target websites and then attempting to submit the questionnaire which first drew the authors' attention to a whole range of issues that appeared to have very little to do with technology per se. This was confirmed after reading questionnaire responses.
And so the authors' focus changed from identifying technological issues to identifying issues which would have an impact upon the communication between the user and an organisation in the context of an "AskA" service.
This paper is not about statistics. Instead the authors will raise questions for consideration and then present illustrated examples of various approaches. The objective is threefold:
The key to this paper, therefore, is the related file [HREF14] which outlines detailed questions and examples, but which is too large to include as part of this published paper. It is this additional file which differentiates the authors' report from others on the same topic. The authors hope that the reader will have examined this file and reflected upon the modus operandi at their own site before progressing to the discussion which follows.
"How/Where Do I Submit a Question?" looks at the context in which the service exists and "Let's Submit a Question" looks at the process involved from the perspective of both the user and an organisation's staff. "Let's Answer a Question" examines an organisation's role in the service. These sections highlight some, but not all, of the issues raised in the questions / examples file, which contains a far more comprehensive coverage.
The importance of where an organisation positions its electronic reference service is corroborated by the following comment from the University of Florida Libraries: "We receive somewhere between 50-60 questions a month now that we have moved our link to the front page of the UF libraries' website [HREF15]". A questionnaire respondent reported an increase from 4-5 queries per week to 10-12 queries per day when that organisation moved its "AskA" service link to a more prominent position on its web site.
Advertising such a service at the top level of one's web site does not, however, necessarily guarantee a high volume of use. One university library, for example, reported 24 queries in a 10 month period. Possible reasons for such low usage included lack of advertising, demographics, and existing instructional programs. It may also be that this is a culture in which students either do not have high computer usage at home or they tend to study on campus. Non-library sites need to focus on advertising, eg advising other subject-related sites of their existence, and knowing their client base.
The authors occasionally experienced technical difficulties in submitting their questionnaire via a web-based form. The presence of an e-mail address (either in the form of text to be noted later or a link to be activated on the spot) ensured that a query could still be submitted. On the other hand, the authors noted that the lack of an e-mail address on the input form (and even on the site's home page) was not uncommon. "Ask The Archaeologist" [HREF16] and "Ask a Linguist" [HREF17] are examples of very useful services, which have no e-mail option on the page with the web form.
From both a web design and public relations perspective, this raises questions about the image which an organisation wishes to project to its public. It is quite daunting to search an entire web site in an attempt to find at least one e-mail address. It is equally daunting to search an entire site and never find the name of a single staff member. It was almost as if the organisations in question have no staff! That, of course, is nonsense; nevertheless sites which give the appearance of being very streamlined and efficient (sterile perhaps?) might ask themselves whether they are conveying successfully to their users the impression that "we are [real people] here to serve you".
The authors' concerns are echoed by experts from the Library and Information Technology Association (LITA) as one of the top technology trends worth following:
"It's time to put a human face on the virtual library. What's the crucial factor in the success of the nonvirtual library? The people who work there and serve the user! What do libraries emphasize on their Web sites? Resources, collections, facts * with no human guidance or presence! On many library Web sites, the user is hard-pressed to identify the staff, whose names, if they're there, are five levels down. The human factor is still important." [HREF8]
The point is that regardless of where an organisation places its "AskA" service/pages in terms of its hierarchical web site structure, eg which unix directory, establishing a link to that service from the home/front (and other) page suggests to the user that there is a human face to the site.
Some sites in our survey complained that they received far too many "inappropriate" queries, eg complex, ill-defined, difficult or time-consuming questions which could not easily be answered using e-mail. This raises the issue of clearly communicating to users the parameters/limits of the service being offered.
Some sites use adjectives such as "quick" and "factual" to define the type of query which is appropriate. Others provide examples of these types of queries. And still others provide examples of the types of questions staff will not answer. The authors' favourite comes from the Indiana University Northwest Library: "What is the sound of one hand clapping?" - We have no idea." [HREF18]
Good examples of sites which help guide the user to ask appropriate questions include:
1. Ask a Linguist [HREF17]: "The field of linguistics addresses the the basic questions, "What is Language?" and "How Does Language Work?". This can involve examining something as specific as the actual sounds of speech or an area as broad as the link between language and culture. Examples of the kind of questions suitable for submission to Ask-A-Linguist include: [examples follow]"
2. Ask a High-Energy Astronomer [HREF19]: "Because of the volume of email we receive, we are not able to answer every question. Check the Ask a High-Energy Astronomer web page to see the types of questions we tend to answer. Students, please do not send us your homework or term paper projects. . . . We will not answer or check homework problems, do general web searches for term papers, or respond to "Tell me everything you know about ...".
The important objectives of an "AskA" input form from the library's perspective are (a) to collect as much information as is required to answer a patron's query; (b) filter out inappropriate users, queries or uses and (c) to set and manage expectations of service levels. Therefore form design is obviously critical to the process.
Along with determining what data needs to be entered by the user, organisations may wish to consider the option of making certain fields mandatory, eg name, e-mail address. While some scripting is required so that the user will receive an error message if they do not comply, the end result should be win/win: the user can logically expect to receive a reply and the organisation has the data required to actually send a reply.
Assuming that a user has filled in the "AskA" form and clicked on "Send/Submit", what might we hope would happen next in the process? An acknowledgment perhaps that the query has been received by the site? The authors feel that this is an important issue because of the wide variety of autoresponder messages which they received when submitting their questionnaire via a web form or e-mail.
In asking respondents whether their organisation uses some sort of automated acknowledgment for user queries, three factors have come to light:
The authors strongly recommend that staff send test messages to their respective "AskA" service and actually see what their users see. We are confident that some organisations will be quite surprised at what they are sent! They may well ask themselves whether this type of communication projects the image they would wish their users to have of the organisation. Some of the examples included in "Issues and Examples" [HREF14] provide ideas for the types of "helpful" information which could automatically be sent to the user, eg turnaround time for an answer, contact phone number, and the like.
The vast majority of sites responding to the questionnaire indicated that user queries --whether submitted via a "mailto" link or a web form-- are sent to a generic e-mail address, eg "reflib@library.xyz". What happens to the query after this depends very much on the individual organisation. In general this is based on whether the "AskA" service is
The authors' impression is that staff attitude plays a big role in determining workflow procedures for this service. Whether the service is viewed as a nuisance or as part of an organisation's pro-active role in its respective community is a matter of attitude, not of technology.
In examining replies to the questionnaire, the authors noted that most organisations advise users that it will take between 24 and 48 hours (or two working days) to reply. Some sites clearly indicate this on the same page as the web form/mailto link; others include this information as part of the automatic acknowledgment message sent to the user (as discussed above).
The authors cannot comment on what data is gather by non-library AskA services; however, any organisation which receives requests for information may wish to consider how it handles related statistics. Having received a user query, most libraries responding to the questionnaire record nothing more than the number of queries received. Other libraries, however, take a different approach and record data which may include:
Very few libraries keep a database (in database format) of queries and answers and other data, and even fewer direct the form input directly to a database. While volume may not require such infrastructure, it is likely to be an efficient mechanism to consider as volume grows, or if detailed statistical data is desired, or if access requirements are very distributed.
It is the authors' impression that in many institutions surveyed, the volume of traffic --be it actual, anticipated or historical-- seems to have governed the choice of "back-end" processes.
One of the areas addressed in the authors' questionnaire was the possibility of a 24-hour shared reference service, which might include cooperative arrangements with other libraries around the world. In other words, how does Fred Nerk working on a project at 11 pm get an answer to a brief but important question if his local library is closed? If, however, his local library --be it public, special or academic-- has a cooperative arrangement with another library in the UK, for example, then Fred does not have to wait 11 hours for someone back in Australia to receive his query.
Many respondents to the questionnaire have expressed the view that their AskA service is in fact available 24 hours per day because a user can access the relevant web form 24 hours per day. In addition, there is an assumption that users in general do not expect instant answers when they send e-mails. An Internet Public Library [HREF20] staff member observed: "I also think that patrons get better answers when librarians have time to think and research and collaborate on their e-mail reference questions." This is certainly a valid point.
Another challenge is that some questions may have a local, regional or even national context which would be unknown in another country. In theory any "Ask A Librarian" service should be able to answer "What is the capital of Spain?" and any "Ask A Geologist" service could answer "Why does California have so many earthquakes, and not New York?" However, factual questions relating to a specific organisation or geographical locale might prove considerably more difficult.
Shared services consist of both "24 x 7" services and collaborative approaches to sharing the work of answering questions, such as is done at Ask a Geologist [HREF21]. When a question is posted, the questioner must submit where they live, and the question is then automatically submitted to the branch United States Geological Survey Library that serves their region. This approach has helped spread out the obligation among the four branches, but the USGS is still overwhelmed by public requests.
More recently the Library of Congress has initiated a project to provide a collaborative remote reference service [HREF22] with the objective of "using a variety of Internet communications technologies (e.g., email, chat, etc.), toward a potential goal of reference service for users any time any where." Three pilot phases have been planned for this Internet-based interactive reference service which would be offered both nationally and internationally 24 hours a day, seven days a week [HREF23].
The authors of the paper feel that 24 x 7 reference services at a basic level are possible with current technology but would require collaborative service agreements and would have to fit in with customer service policies. For instance, institutions around the world could arrange to provide 24 hour services by connecting the enquirer via a web browser and some standard applications with the provider that is currently "on" the desk. A recent joint initiative between Macquarie University and the London School of Economics [HREF24] offers a possible model. Clear boundaries would have to be drawn up for services provided and attitudinal changes on the part of both parties to the communication would be necessary. Technology brings out the worst in human behaviors: if phone users will barely wait for 5 minutes, Internet users will not wait 60 seconds for a response!
This naturally leads us to a discussion of interactive services.
Given the lack of a face-to-face interview in the web form/e-mail methods, some sites have experimented with various technologies, eg videoconferencing and chat, in order to offer an interactive service. Obviously such initiatives are considerably more complex and labour intensive to administer than the form/e-mail approach.
In a recent paper Folger [HREF25] provides an overview of desktop videoconferencing technology, discusses a University of Michigan project, and considers the future of this technology in libraries. While some libraries responding to the questionnaire expressed an interest in offering some type of interactive service, they also expressed concern about the lack of available resources (staff and money) to support such an activity. In the case of chat-based options, the necessity for users to cope with another layer of software was highlighted as a potential barrier. A summary of the responses is available at the end of "Issues and Examples" [HREF26].
Temple University and Cornell University are examples of libraries which have addressed the latter issue by providing chat sessions which require no additional client software. The TalkNow service at Temple uses software which has been specifically developed for its use [HREF27]. Cornell uses the LivePerson [HREF28] software as the basis for its new experimental service [HREF29].
Interactive services offer a decided advantage over asynchronous services: immediacy, ability to refine, ability to create a relationship, and the "human touch". Therefore companies that have invested millions in e-commerce are trying to turn people who browse into buyers and retain their custom by providing some of the experience of in-store customer service with services such as chat. Web-based reference services are competing with other Internet search services in providing answers to questions. Like other service industries they will have to continue to evaluate the new web-based customer services, such as Voice-over-IP, [HREF30] as they develop.
An article by Daniel Lyons this year in Forbes [HREF31] highlights the recent spate of commercial "online advice" sites which have emerged:
"Iexchange.com, launched in October with $35 million in venture funding, offers investment advice from 3,000 "experts", including day traders and housewives. Expertcity.com, launched in December with $34 million in capital, has 150 know-it-alls dispensing computer support. For $10, the site lets you chat live with one of its experts, who use special technology to connect to your computer and "share" your screen with you."
As potential "competitors" to library-based e-reference services and non-commercial AskA services, the dot.com companies will bear watching, particularly in their use of interactive technologies.
Truly interactive reference services that can vie with face-to-face contact are dependent on three technological developments (together with convincing reasons for using the service):
Until such time as these components are present, interactive reference services will continue to be limited to specialised environments.
One of the objectives of administering the questionnaire was to determine the perceived usefulness of an "Ask A Librarian" service at those sites which have implemented it. As has been suggested above, it is the role which an organisation determines that an AskA service will play in its overall mission of providing a query service which seems to influence how that service is viewed. In her annual report, Borisovets [HREF32] outlines the types of decisions which Rutgers Libraries had to make when its "Ask LIAC" link unexpectedly blossomed into a popular, full blown service. As we have already seen, the US Geological Survey [HREF21] is also having to struggle with a service which has become "too successful".
Some libraries have assigned a low degree of success because of a low volume of queries. Other libraries have taken a different approach; rather than look at the number of queries received, they have focused on the fact that if patrons are actually using the service then it must fulfil a need. Santa Monica Public Library best sums up this approach: "... anytime we can extend the reach of our services to our clients wherever they are and whenever they need them we think it is a success." [HREF33]
For those organisations planning to implement an "AskA" service, they should decide who their target audience is and how this service will fit within their overall mission/provision of services before they tackle the technological implications of providing that service.
The authors have outlined a number of issues which institutions will want to address in both implementing and maintaining such a service. The basic issues raised include:
The authors believe that the challenges in designing such a service really have very little to do with technology, although an organisation should have access to support for database management and web issues. One needs to focus instead on the relationship between an organisation and (service to) its users. The fundamental issues, we believe, are communication and attitude. If an organisation has no direct verbal communication with its users through this service, then how does that organisation
Once a user's query has landed somewhere in the system, how does an organisation
And so, factors such as appropriate web design, on the one hand, and workflow procedures, on the other, become critical to the potential success of the service. The authors hope that readers will use the discussion in this paper along with the examples provided to determine whether they have structured their "AskA" and other information/comment services in a way which maximises communication between users and their organisation.
Lankes, R. David and Kasowitz, Abby S. (1998). AskA starter kit : how to build and maintain digital reference services. Syracuse, NY: Syracuse University.
Sloan, B (1998) Service perspectives for the digital library: remote reference services [HREF6], Library Trends, 47(1): 117-143
As a member of a small, Australian statewide working party looking at implementing an electronic reference desk service across a number of university libraries, I have looked at your site and the service which you offer. I am particularly interested in the IT aspects which lie behind it and wondered whether you would be willing to answer several questions which I have in that regard. [Or would you please forward this to the appropriate person in your area.]
Q1: Do you use or have you considered using a commercial system to manage requests? (eg. call centre or helpdesk software) If so, which one?
Q2: What is the "destination" of the form which the user completes? Does the form go to (a) a database such as FileMaker, Access or Oracle, (b)a generic e-mail address ("reflib@xyz"), or (c) some other destination?
Q3: Do you run an FAQ database built from previous replies or does the person answering each query rely on memory, library resources and the like?
Q4: Who actually answers users' queries and how is this determined?
Q5: What statistics, if any, do you collect about the service?
Q6: What were the costs in setting up the system and what are the ongoing costs of operating the service? [general indication only, please, of dollars and personnel]
Q7: How successful do you think the service is?
Q8: Is your library (or do you know of any library) currently using or considering methods for answering queries 24-hours-per-day? (This might include cooperative arrangements with other libraries around the world OR rostered shifts).
Q9: Is your library also offering, or considering offering, interactive reference services via methods such as ICQ/IRC/Web Board, NetMeeting, videoconferencing or some other mechanism? If so, what problems have you encountered?
Thank you VERY much for your comments. My interest could end up becoming a project to implement a state-of-the-art online reference service, perhaps for 24-hour service. Would you be interested in pursuing this further with me (and my working party)?
It would be appreciated if the following questionnaire were completed and returned to me by Friday, June 11, 1999
As you may recall, in late May/early June you received a questionnaire regarding "Technical Aspects of Your 'Ask a Librarian' Service." As chair of the subcommittee which designed that document, I thought I would share with you a brief summary of the outcomes to date as well as planned future activity. I would also like to use this opportunity to request clarification (if applicable) of aspects of your response.
[Preliminary Observations have been deleted to avoid repetition since they are addressed above in this paper.]
Your Permission
I wondered if I might have your permission to quote your response(s) to our questionnaire. You would not be named as such; instead the quote would be attributed to your institution. For example: University of X: "We keep our statistics online, set up in an Excel spreadsheet ·" If you cannot remember what wonderfully witty, insightful comments you made, I would be happy to e-mail you a copy of your response.
AutoReply
At the time we designed the questionnaire, we did not think to enquire about this aspect of communication with library patrons. If you do use some sort of automated acknowledgment when a patron submits a query via the Ask a Librarian service, then
If you do NOT use an automated script of any type, then do you (your staff) individually acknowledge patrons' queries?
Other Issues
[This part of the communication was customised according to the reply received from the respondent to the original questionnaire.]
Thank you once again for your interest and assistance.
Joanna Richardson, Janet Fletcher, Alison Hunter, and Philippa Westerman, © 2000. The authors assign to Southern Cross University and other educational and non-profit institutions a non-exclusive licence to use this document for personal use and in courses of instruction provided that the article is used in full and this copyright statement is reproduced. The authors also grant a non-exclusive licence to Southern Cross University to publish this document in full on the World Wide Web and on CD-ROM and in printed form with the conference papers and for the document to be published on mirrors on the World Wide Web.
[ Proceedings ]