RETHINKING LIBRARY INFORMATION SERVICES FOR THE DIGITAL ERA.

The most essential element in library information services is having the right people in the right place at the right time. To aid this process libraries must rethink not just the value of individual positions within a service but the whole structure of work itself to remove staff from repetitive tasks and use their skills, experience and training. The technology is now mature enough for an assessment of its impact on positions and procedures. Edited version of a paper presented at Digital Libraries Technology 99--the third annual conference Transforming library information services for the digital era 22-23 July 1999

Back to the future is a movie title which has become a cliche. Yet in a very succinct way, this strange phrase describes almost perfectly the central theme of this paper.

It does not discuss new, leading edge technologies which librarians will confront in the first decade of the new millennium. Rather, it discusses the technologies librarians know or have heard of.

It is about rethinking what we know, and rethinking about how this knowledge can best be used to deliver better information services to our clients by better utilising our most important resource, people. Instead of looking forward, it goes `Back to the future'; to those mature technologies embodied in online databases and library management systems and cdroms and the internet, to technologies which have proven performance and endurance over the last twenty five years.

Why is now a particularly good time for rethinking the way we deliver information services? Partly, the answer lies in the natural tendency of humans to pause at a significant anniversary and reflect upon past achievements and future hopes. Partly too, the headlong rush of technological development seems to have paused, as if to take a deep breath. Perhaps the influence of the Y2K issue is responsible. Clearly many vendors have been forced to put significant resources into the resolution of the problem. Or perhaps it is a natural slowdown in an industry which has moved so far so fast.

Either way, those people you have heard wish `that technology would just stop for twelve months so we can catch up' may, for this brief moment, be getting their wish.

From the current proven technologies available, six essential elements have been chosen to help lay the foundation for quality library information service delivery into the early years of the next millennium.

* digital technical services

* electronic resources in collection development

* cataloguing the web

* digital document delivery

* email

* staffing

At first glance, few of these elements seem directly linked to the delivery of information services. However it is impossible to isolate any single library function without understanding its place in the overall service and the influences exerted upon it.

There is one single link between the variety of technologies to be discussed--people.

Information services are simply about getting the right people in the right place at the right time to meet the needs of clients, Digital libraries are not about how much technology can be jammed into a single building, but how well we can link the different abilities of people and computers.

In each of the elements chosen it is shown how current, common, mature technologies can help release staff from the mundane and repetitive current tasks, so that their talents and training may be better directed towards the delivery of quality information services.

The concept of information services is not limited to requests for factual data. Information is whatever the client requires, whether it be the name of Wilbur Smith's latest or the contact details for the local footy club.

Digital technical services

Three library functions are included under the general heading of technical services: acquisitions, cataloguing and end processing of resources.

Librarians have long been aware that technical services is one of the most labour intensive, and functionally repetitive, sections of a service. As a consequence, there has been constant investigation into the various means and methods to reduce the more mechanical aspects of technical services tasks. By far the most successful of these has been the catalogue record resource sharing functionality offered by ABN/Kinetica. However, faced with continuing economic and user pressures, libraries are looking at additional ways to automate technical services functions.

As an example a paper I delivered at the Digital Libraries 97 Conference was entitled `Re-engineering library services for the digital age'.[1] That paper spoke of an experiment by Stanford University to reduce both costs and staffing levels by automating or outsourcing a large number of technical services functions.

Advances in the technology of library management systems and data telecommunications are such that it is now possible for a library to have its resources selected, catalogued and processed by a vendor. The data generated by the vendor, at each of the various stages, is then automatically transferred to the library management system where all relevant databases are updated without the need for library staff intervention.

There are three notable implications for a library using this procedure in technical services. First, and most obviously, there would be a significant reduction in repetitive technical services tasks. This may provide an opportunity to transfer staff to information services functions, thus providing an improved service for the client. Secondly, such a procedure could benefit a library of almost any size.

Last, and most importantly, a library may implement either the full procedure or just the parts that best suit local circumstances. For example, the library may wish to retain control over resource selection. Staff could then place an electronic order and the vendor would complete the procedure as described. Or the library could catalogue inhouse and implement the remainder of the procedure.

Before implementing such a procedure there are some practicalities to be decided. First is the acquisitions profile of the library which is delivered to the vendor. The profile would contain such details as the type of library, breakdown of the collection and budget. The vendor would need to be able to prove that it could meet all the procedural needs.

For logical reasons, it is also better to implement this procedure with relatively few vendors. As the profile is likely to change (and will, at the very least, require close monitoring), the more vendors involved the greater the library administrative overhead. Most libraries which have implemented such a procedure have used no more than three vendors.

Similarly, the resource budget would not be entirely devoted to the vendors involved. There will always be material required by a library which cannot be supplied by a given vendor. This may be particularly true with collections such as local studies and languages other than English.

The benefit of such a procedure for information services is not only that additional staff or resources may be able to redirected, but that technical services staff may be able to better provide the base data for information enquiries. In other words, instead of a cataloguer assigning two or three subject headings, he or she may assign ten or fifteen, thus providing a greater level of access to the client.

There are also financial issues to be considered. Such a procedure cannot be implemented without some initial costs during the early evaluation stages. Library management system vendors are going to require additional funds to build the automated links and resources.

Whether these costs are offset by staff reductions or additional funding, measured against improved service, will be a matter for each organisation.

Such a procedure to `digitise' technical services will provide longterm benefits by allowing precious library resources to be moved from mechanical tasks to professional duties which better help meet the information needs of the client.

Electronic resources in collection development

Libraries have been including electronic resources in their collection development policies for many years. However as the millennium approaches we are confronted with even more types of this material. Decisions as to what to include then become even more critical as increasingly sophisticated client requirements are met by limited budgets.

There are two primary issues to be addressed when collecting electronic resources. One is space and the other is user preference. Unfortunately the outcomes seem to point in diametrically opposed directions.

One of the great myths of electronic resources in libraries is that they require less space. This situation seems much worse when it is perpetuated by librarians themselves. At first the argument seems obvious. At Blacktown, the World book encyclopaedia in hard copy needs 41,580 [cm.sup.3]. The equivalent cdrom needs 175 [cm.sup.3]. However, 585,000 [cm.sup.3] of hardware is needed to read a cdrom, and more if you want to print.

To this space debate, the question of the number of users should also be contributed. A pc can access multiple cdroms, but only one at a time. A print edition could theoretically be used by as many clients as there are volumes. So the library must decide whether it has the space, hardware, and technical support, to provide, and continue to provide, the resources in a satisfactory method to the client. Against these complex issues is the simple fact that many users prefer to use electronic resources.

At Blacktown it is not uncommon to see users who prefer to wait to use an electronic resource when a print resource could satisfy their information needs. This issue is also exacerbated by the perceived status of the intemet, where another myth exists: that the intemet contains all human knowledge and it is all free and it is all easy to access. Refuting this myth will be one of the great challenges of libraries in the next millennium. As with any resource in collection development, the question of electronic resources is one of balance. Librarians must be willing to continually monitor the situation and change as required. After all, any resource which is not used to its full potential is a waste far greater than its purchase price.

Cataloguing the web

The internet is here. It is a mess. It is not all rubbish!

   Having spent the better part of a weekend reading through cyber flotsam and
   virtual jetsam ... I went to a library [and] within 10 minutes I had eight
   books selected from more than 100 available.[2]

Perhaps the greatest challenge librarians will face in the new millennium is the classification of information on the internet. The difficulty will be distinguishing between that very precise information generated by a sophisticated search engine, and which should remain accessible in this way, and the more general access required by many users.

Libraries have never displayed the marketing power to challenge the perceived status of intemet information. So libraries will need to prove their worth to current clients by including quality web resources as part of the information services they deliver to their clients, while continuing the marketing push to nonusers.

There are a number of different ways for libraries to approach web classification. One is to emulate the Yahoo style of subject hierarchies and build their own guides. Any brief survey of library home pages will show that this, in this early stage of classification, is a fairly common approach. However the contents of these guides vary significantly in both size and scope. Significantly also, there is a large amount of duplication between the work being undertaken by different libraries. This is somewhat analogous to using original cataloguing records rather than a shared resource such as Kinetica.

Another method is used by the Dublin Core Project. In this case, the onus for cataloguing is placed on the author of the web page. They include in the metadata section of the html file which makes up each web page, title, author, subject details and so. This data, although it cannot been seen by the user, may be indexed by internet search engines. The interesting fact in this solution is that it attempts to amalgamate some of the best of what librarians know, while not overlooking the abilities of search retrieval engines. In the long term, such a solution will be the best outcome.

However at this stage the search retrieval software does not display an ability to `understand' user requirements and to list only those documents which are truly relevant. Experiments with expert systems, and holistic software which `learns' from each enquiry, have been somewhat successful in very controlled environments, such as legal databases, where the subject area is relatively narrow and the documents comparatively small. These experiments have not yet transferred well to major information resources.

The third and best method is to catalogue a website in exactly the same way as one would with any other resources. In other words, look at the site contents, assign an appropriate number of LC subject headings, add a Dewey classification number and file it into the database.

Modem Gui based library management systems offer the ability to place the location of an electronic resource within a Marc tag. The client then clicks on the link, the appropriate software is launched and the resource displayed. The advantage of such a system is that the electronic resource need not only be a web address, but could be a directory link to a sound file, word processing document, image or other computer file.

By building a web catalogue in this way, librarians are able to address a number of issues raised by internet classification.

First, such a system offers some consistency of interface and search technique. The client learns one method of searching, via the local library opac interface. Additionally, the implementation of the Z39.50 communications standard into library management systems means that users are able to access other library service catalogues from within their local opac interface.

Secondly, making the web part of a library's normal collection of resources promotes the idea that the intemet is just another source of information. The client may in fact be able to find another resource which better meets their information needs. Thus all library information services are improved.

Finally, the existence of a web link on a library catalogue should offer some sort of validation of the information contents of that page. Validity of information is always going to be a contentious issue with the internet, as indeed it sometimes is with print resources. However by developing a set of criteria to measure the value of a page's contents, librarians should be able to direct a client to those sites which are likely to provide accurate information.

Linda Bruce[3] suggests asking the following questions about intemet information

* does the site cite the author of the material?

* does the site supply an email address for contact, and if so, does the author respond to your email

* does the site list the author's credentials or experience or do you need to accept expertise on face value

* many quality sites link to other sites related to the issue. Does the site you have located link to others or does it standalone?

* can you find any other information sources, whether they be web pages, journals or encyclopedias to confirm this information?

* beware of bias. What does the site aim to do? Why is the web master publishing the site? Which cause, if any, does it promote?

* is the information objective? Does it ak both sides of the discussion? Be suspicious of information that relies on emotion rather than debate

* how well informed is the material compared with sites you can vouch for?

* has the site been recently updated? If not, it may be dead or outdated 

* has the veracity of information on this site been checked by others?

By asking these questions a library will be better able to assess the contents of each site and whether it is capable of meeting the client's information needs.

As noted earlier, one of the concerns in these early days of web classification is the amount of duplication being undertaken by different library services. Perhaps the time has come for the information services industry to consider the establishment of a web resource database similar in purpose and scope to Kinetica. Oder[4] discusses the various attempts to establish such resources in the US and UK, by organisations such as OCLC, the Interact Public Library and Infomine. The problems, however, appear to an inconsistency in approach ie what format should the classification use, and funding.

Nevertheless, such a resource would not only save a vast amount of task duplication. It would also provide even those services which do not yet have the appropriate library management system at least access to a website's details in electronic format. Once information is in such a format it can easily be manipulated to meet the requirements of other software.

By adopting this process, libraries will be able to firmly cement a particular niche in the delivery of internet resources to the client. It is not suggested that cataloguing websites replaces the need for search engines and other subject guides, but there are serious shortcomings in the ability of these to deliver some forms of information, notably general subject data.

Libraries, on the other hand, are experts

   ... after thousands of years of experience, librarians have developed a
   sophistication of categorisation that isn't sexy or able to deliver
   thousands of vaguely applicable hits within two milliseconds, but can
   actually give you the information your looking for within a few minutes.[5]

Digital document delivery

You can order a meal by phone; you can shop via the interact; you can pay bills without ever leaving home or talking to another human being. So why do libraries find it so difficult to `home deliver' information?

Librarians have always prided themselves on their Commitment to client service. But that has almost always been in a situation where the client comes to them. Certainly there are mitigating circumstances. Libraries are institutions which exist for the public good and so remain committed to concepts such as free service and equity of access. Unlike for profit enterprise, there are no additional funds automatically available if more services are delivered to more clients. Consequently the number of client services delivered by a limited number of library staff severely restrict a library's ability to commit staff time to such a method of information delivery.

There is also a sense of the library's user education role, whereby it is actually to the longterm benefit of clients if they learn how to access information. Nevertheless, libraries retain a reluctance to serve clients who are not physically present. Such a situation, however, cannot continue. It is no longer compatible with changing social expectations.

Traditionally document delivery has meant the supply of material from one service to another before it is passed onto the client. Advances in digital technology now allow libraries to include the client as part of the document delivery process, the term document here covering all formats of information.

As discussed earlier, the Z39.50 telecommunications protocol allows the local library to search external databases. In practical terms, this means that a user at Blacktown could search the catalogue at Sutherland, assuming both library management systems are compliant--they do not need to be the same system.

An alternative to the library opac is interact technology. Many libraries offer their catalogue via the web, which in mm could be linked to a Z39.50 session.

However the shortfall of this technology has always been that for the client the retrieval software stopped once the information had been found. Too often the client is then forced into some other method of actually obtaining the document. Usually they either have to ring their local library, or the library which holds the document, and make a formal request.

To overcome this problem librarians must be willing to implement a system which will allow the client to directly request a document. Such a system would be able to distinguish between the hold of a local member, a common functionality in library management systems, and the request of an external client. A method of accepting payment would also need to be included for external users. A library could, for example, take credit card details to cover delivery costs to the external user. Should the material not be returned, the full cost could then be applied against the credit card.

Not all documents need to be handled in a similar way. There is already a number of document delivery services specialising in journal articles. Libraries could offer access to these services for the clients.

The advantage of offering the client direct input into the document delivery process is that it meets the increasing need of clients to have information delivered to them in the way they want. The process puts the onus on the client to decide how much information they want, who is going to access it and how much it will cost.

Sophisticated clients could have information delivered directly to their home or business by searching library databases, deciding which documents meet their specific needs and paying for delivery, all without the need to involve library staff. Inexperienced clients could pay for the library to research and deliver the documents to meet their information needs.

Meanwhile, the library would offer an improved free service to clients, because staff have been removed from some involvement in the document delivery process.

Email

The use of email in libraries will, perhaps more than any of the elements, depend upon the type of library service being discussed.

A library with a closed membership, such as an academic institution or most special libraries, has the ability to incorporate email as a form of communication with much more authority than that of an organisation with an open membership, such as a public library.

For example, one of the suggested uses for email is the distribution of overdue notices for circulated items. This is obviously much easier at a university where all library users have at least one email address provided by the organisation. In a public library with tens of thousands of members, there will be, in the foreseeable future at least, someone without an email address.

Email, as a form of communication offers libraries both a significant challenge and an exceptional opportunity. The challenge lies with the concept of the email reference question and the opportunity is in the marketing of the library service.

The problem of using email for exchanging reference information was touched on in the previous discussion on document delivery. At the core of the issue, however, are two problems which must be addressed by each library offering this service.

One is the definition of a reference question. We have all had seemingly innocent looking questions which turn out to require significant time and effort in answering. Or the question may require the use of several different resources. This is not a problem if you are directing a user inhouse, but is of significance if a qualified librarian is now forced to do the work.

Related to this issue is the question of the mechanics of delivering information via email. For example, what if the information is stored on microfilm? You go and load the reader, print off the data, scan the printout and then email the resultant file--this could take a while.

The other problem in an email reference service is that of speed. We live in a world of fast food, instant coffee and live sport. The near instantaneous delivery of email engenders a belief by the user that the response should be equally as fast. How will libraries react to the demand for information `in the next two hours'? Drop everything? Charge? Ignore time limits? This could be a hazard in itself. You could end up doing work that is pointless because the information really was needed two days ago.

Libraries have always prided themselves on their service ethic. The email reference question demands a defined set of criteria simply because the user is not face to face. They cannot see the queue of other people waiting, or the printer waiting for repair.

Some institutions will also face a third problem: delivering services to nonresidents or nonusers. How is your parent organisation going to feel about funding the information needs of an email user who may be anywhere in the world? Is charging for such service inevitable? On the other side of the coin, email offers libraries an exceptional ability to market their services at minimal cost.

Listserv software is widely available and may be installed on a library server. Users subscribe to a chosen service and receive via their email a kind of electronic bulletin board of ideas, comments and information exchange. The service list may be moderated, in which case an administrator may refuse the submission of certain messages due to unsuitability or irrelevance. Or the list may be unmoderated, in which case whatever message a subscriber submits is available to all other subscribers.

Libraries could take great advantage of such software to market their product. The lists may obviously be about any subject: new books, user feedback, reviews etc. The library may use the technology as a promotional tool for library events, new services or to push a certain part of the collection.

For example, marketing people use the concept of distressed inventory; in library terms this may best be described by resources which are not being used to their full extent. A library could use a list to promote a `did you know about these?' list of resources to clients.

The library industry could also look at using marketing companies who can teach us how to best use current technology for the promotion of our services. Earlier, Jeremy White was quoted on the effectiveness of the Dewey system over a web search engine. What is of concern is that while we are busy telling each other how good we are, we are not doing nearly enough to tell the people who need to know.

As people are confronted by even more information choices, we need to do whatever it takes to promote our value and worth.

Staffing

Quality library information services can only be delivered to clients through people. The technology can be state of the art, the software super sophisticated and the telecommunications lightning fast and it will still make little difference.

It is people who are responsible for the accuracy of the initial data, it is people who evaluate the quality of the data and it is people who can best deliver the data. Human communication is 65 percent nonverbal.[6] How can technology emulate that?

For this reason, the single most essential element for library information services in the new millennium is having the right people in the fight place at the right time.

To aid this process libraries must rethink not just the value of individual positions within a service but the whole structure of the work itself. We all went through some form of organisational restructuring in the recent past. However the technology is more mature now and libraries may be in a better position to assess its impact on positions and procedures.

The theme of this paper has been to remove staff, wherever possible, from repetitive or mundane tasks and use their skills, experience and training. Why are we using professional or paraprofessional staff for data input duties? Would not we be better hiring specialist staff for specialist duties? In the case of data input, which may be a catalogue record or a membership form or an interlibrary loan request, it is surely preferable to use a skilled typist to process the work quickly and accurately and use the qualified librarian to make appropriate decisions as the need arises.

Systems librarians

A specific example is the position of systems librarian. These are the people who are qualified librarians who have drifted or moved or been seconded into information technology duties in the library. Typically, in small and medium size organisations, they retain some library duties as well as supervising or maintaining hardware and software.

Have they reached their use by date?

In 1989, when I moved to Blacktown as systems librarian, we used the Utica library management system on a Pick minicomputer via text based terminals over serial data lines. We had two pcs, one with a cdrom drive, and two Apple IIs for public use. By the end of this year, it is expected that the library will operate on a combination of pcs and WinTerms, over networks connected by ISDN and ethernet backbones, via servers operating on unix, Netware and NT.

The interesting implementation will be that of a WinFrame/WinTerm solution and telecommunications infrastructure which ironically suits the `Back to the future' theme of this paper.

WinFrame/WinTerm is a thin client option where software resources are located on a central server or servers. The WinFrame software directs the client request to the appropriate server. The resources are then accessed by the WinTerm, a thin client colour terminal which has the ability to run Windows via an internal rom chip. It has no local hard drive and so all software upgrades are done on the central server. In other words, basically an update of the old style central server-dumb terminals configuration, but one, which is able to reap the full benefits of technological advances.

This infrastructure will position Blacktown Library to move towards the fully networked and graphical system which will be the foundation of all digital technologies in the new millennium.

The repercussions of such an intense level of technology within the service are that there must be an ongoing commitment to system upgrades of both hardware and software as well as appropriate technical staff support.

While it has been reasonable in the past to have a systems librarian maintain inhouse library information technology, that is no longer the case given the diversity and complexity of modern hardware and software systems found in most libraries.

Instead systems librarians will return to the tasks they were initially employed in libraries for: to maintain the library management system and to assess information technology developments and how they may relate to library staff and client needs.

Due to the complexities previously mentioned, other tasks will need to be directed to specialist staff. Whether such specialists are employed within the library service, as part of the parent organisation's IT section or are outsourced to consultants, will be a decision appropriate to each individual service. That such specialists will be needed is not in doubt.

The example of the position of systems librarian is given as an indication of how flexible management of both the library and the parent organisation will need to be if the library is to meet the challenge of delivery quality information services to the client. All positions and procedures should be investigated to see where technology can best be utilised to ensure that staff resources are used in the most efficient manner possible.

Conclusion

Whether you are a technophile or technophobe, the digital library service is an inevitability. However digital does not exclude human. Instead it provides us with an unparalleled opportunity to rethink how we can best link the skills and talents of people resources with the speed and functionalities of computer technologies.

Only by taking advantage of the best of both worlds can we perform the basic duty of any library service which is to deliver to clients the information they want when they want it and in the way they want it.

This paper began with a movie reference and closes with one. As the millennium changes, we must take this opportunity to build technical and staffing infrastructures now. For if it is technically the case that the new millennium actually begins on 1 January 2001, then libraries are about to take the first steps on an odyssey into an unknown future. We must be prepared for whatever may happen.

References

[1] Grosvenor, M Reengineering library services for the digital age Australasian public libraries and information services 10(4) December 1997 p224-234

[2] White, J Editorial in Australian personal computer September 1998

[3] Bruce, L Editorial in School PC Australia September 1998

[4] Oder, N Cataloging the net: can we do it?' Library journal

[5] White, J op cit 

[6] Adler, R, Rosenfeld, L and Towne, N Interplay: the process of interpersonal communication 3rd ed New York, CBS Publishing 1986 p96

Mark Grosvenor is Operations and Community Information Services Librarian Blacktown City Council Library and has spent eighteen years in public libraries, starting as a library assistant. The last twelve years have been as a system administrator. As a member of the senior staff at Blacktown, Mark supervises staff with a broad range of responsibilities, including community information, circulation and general administration. He has also acted as Manager Library Services. Address: Max Webber Library PO Box 227 Blacktown NSW 2148 tel [phone] fax [phone] [e-mail]

Mark Grosvenor Operations and Community Information Services Librarian Blacktown City Library NSW Received August 19993