York Information Connections: An Attempt to Catalogue the Internet

LibWeb [1] is the University of York Library’s web information service. LibWeb was designed right from the start as a comprehensive site which would provide as much information as possible on library services, staff and collections, and which would include electronic versions of all printed library guides.

One of the first questions that arose in designing the site was how to deal with Internet resources. The Computing Service had maintained a campus information service for several years before the Web became the standard method of access to the Internet. The Library already had a collection of bibliographies and other resources downloaded from other sites and made available on the information service which we did not want to lose. On the other hand, there were already other national collections of resources, notably at BUBL [2] and NISS [3] which we did not want to duplicate.

As a compromise solution, we decided to provide a convenient collection of links to resources that would be of particular interest to the staff and students at York, including links to other more comprehensive sites where available. This developed into the York Subject Tree and, ultimately, York Information Connections [4]. The development of this service and the rationale behind it is outlined in the first half of this article.

The cataloguing of electronic journals involves many of the same issues which apply to other Internet resources. However, they also pose some of their own problems and these are highlighted in the second half of this article.

York Information Connections

The York Subject Tree was part of LibWeb from the launch day. The aim was to collect together Internet resources which reflected the teaching and research interests of the University of York in order to provide a starting point for staff and students. As well as pointing to specific resources, the Subject Tree included links to other more comprehensive sites for those people who wanted to explore further. It was, if not the first, then one of the first Subject Trees developed in a UK university library and has been much imitated!

York Information Connections introductory menu

At that time, the Computing Service maintained a page of links on Exploring the Internet. As the Subject Tree evolved it began to overlap more and more with this page and to include a broader range of information than was covered by the title Subject Tree. It was decided to merge the two sections and this led to the development of York Information Connections.

Why did we decide to have our own Subject Tree? The rationale was that since we do not send students to the British Library Catalogue when they want a book for their essay, why should they have to plough through the BUBL collection of resources to find Internet resources? The idea was to develop a “collection” of Internet resources in the same way as we have a collection of books and journals.

Selection and cataloguing

Selection of resources is distributed between the LibWeb Coordinator, Subject Librarians and members of academic staff in departments. As is always the case, some people are more enthusiastic than others and this is reflected in the subject pages. An example of a well developed page is that for Women’s Studies [5]. This is due to the enthusiasm of one member of academic staff who sends information on many resources to add to the page. She has also advertised the page widely to researchers throughout the world with the result that usage statistics for this page are relatively high, justifying the time spent. Most of the links on all the subject pages have a brief annotation explaining what the resource is about, but we go no further in cataloguing or evaluating the resource.

Subject arrangement

The subject arrangement on York Information Connections is by University department. We decided at the start that we would leave the major information gateways (at that time BUBL and NISS were the two main examples) to provide fully classified collections of resources and that we would concentrate on broad subject headings. It seemed logical to relate the subjects directly to the departments (there are no faculties at York). Since the whole site was maintained by one person it was impossible to do more at the time and this seemed a reasonable compromise between the impossible dream of a fully classified comprehensive collection of Internet resources and having nothing at all. Now that there are many subject specific gateways, with links to evaluated and classified resources (such as EEVL [6], SOSIG [7] and HUMBUL [8]) the decision seems to have been the correct one.

Within the subject pages we developed a set of standard headings which are applied to all pages (the chemistry page includes all these standard headings [9]). Headings are left off the page only if there is no resource relevant to that heading. In addition to the standard headings, each page may have more specific subject headings which are generally related to teaching or research topics of interest at York.

There is an alphabetical index to LibWeb as a whole which was mainly designed to guide users through the whole site. It does point to the main subject pages (there is an entry for archaeology, for example), but goes no more specific than that. This was another pragmatic decision based on the lack of resources to do more and the lack of a search engine for the YorkWeb site.

Internet resources on the OPAC

We briefly considered adding some of the main Internet resources to the OPAC. There are several reasons why we have chosen not to take this approach:

• The main reason is that the Dynix WebOPAC is not sufficiently developed for us to purchase it. This means we cannot link directly to the web resource from the OPAC. We would probably want to duplicate the links on a web page while this is the case and this is seen as a waste of effort.
• Our cataloguers have a hard time keeping up with cataloguing of books and do not have the capacity to take on web sites as well.
• Web sites move about which would lead to inaccuracy in the OPAC. We feel that this is less acceptable than inaccuracy on a web page (this may be a completely spurious feeling and has no logical reasoning behind it!).
• We would rather avoid duplication of effort and rely on the national subject gateways for full catalogue entries at present.

We will certainly be reviewing this policy when we get the WebOPAC (due sometime in 1998). An obvious category of resources to add to the OPAC would be the major subject gateways since they can be looked at as the equivalent of a printed bibliography or index.

Would this mean doing away with the subject web pages completely? Perhaps. However, there is some merit in maintaining the web pages as the equivalent of browsing the library shelves. In addition, we use York Information Connections as a starting point for courses on using the Internet and the web. Our undergraduate information literacy course, ILIAD (Information Literacy In All Departments), has two hands-on practical sessions on researching and evaluating information sources with York Information Connections providing the launch point for exploration of the web (the course is run jointly with the Computing Service and also includes sessions on wordprocessing and presentation tools). We believe that the subject pages offer more than a WebOPAC.

Cataloguing e-journals

The list of full text electronic journals available on LibWeb aims to include all the e-journals which can be accessed by the University of York user [10]. This includes those which are free, those where the online version is free with a print subscription, and those which are included in a site licence agreement. Each title listed is linked to either the home page of the journal, or in the case of the HEFC titles, to the host site (Academic Press IDEAL [11], Institute of Physics Publishing [12], BIDS JournalsOnline [13]). Some other e-journal pages which provide tables of contents and/or abstracts, but not free full-text, are listed on the subject pages of LibWeb. A separate section provides links to publishers’ table of contents services [14]. We aim to continue to develop and improve the LibWeb electronic journal pages to make them the first stop for users wanting information on electronic journals.

When electronic journals became available through the HEFC agreement, we did not create catalogue records for the full text journals which were available online but which we did not already hold in print. However, we did add a note to the catalogue record of those journals which we already subscribed to in print form (and therefore already existed on the OPAC):
“Electronic version available through LibWeb URL: http://www.york.ac.uk/services/library/subjects/ejlist.htm

We decided to refer users to the electronic journals page created on LibWeb, rather than give the URL of each journal homepage. This minimizes the work involved in adding and updating URLs, which is of course already necessary on the LibWeb page. Also, the URLs of many journals can be quite lengthy and/or unstable. We thought it best to encourage our users to consult the LibWeb page, which would act as a gateway and authoritative guide to the electronic journals available to the University of York user. Links from this page also inform the University of York user of any additional information, such as passwords, that they need in order to access the journals, as well as providing brief information on host sites such as IDEAL.

When we have the WebOPAC interface to our catalogue we will be able to add the journal URL to the catalogue record; the user will then click on this link and seamlessly transfer to the journal homepage or, in the case of online journal hosts such as BIDS JournalsOnline which require users to log in with a username and password, we will provide a link to the host homepage.

So, should the OPAC have an additional role to play, other than that we have assigned to it? Is it feasible to provide catalogue records for the journals only available to our readers electronically? Can we justify extra workload of duplicating information already available from the LibWeb pages? Or, looking at this from a theoretical rather than practical point of view, what should the role of the catalogue be? The library of today and the future cannot simply be defined by its holdings (the stock housed in the library) but also includes material accessed remotely. The catalogue should therefore include material available to the user, even if it is not held on library property. Bibliographic databases held remotely on the SilverPlatter ARCPlus server are catalogued by us, so to be consistent, e-journals should be too.

Adding e-journals to the catalogue as access is arranged and as they are added to LibWeb does not seem unreasonable. The problem facing us is how to deal with this retrospectively. We do not know at the moment how long free access will be available; we are now 2 years through the HEFC scheme with one year to go. There are over 100 Academic Press titles available through IDEAL which we do not hold in print form. Should all these be catalogued? If they are not freely available next year, they would need to be deleted from the catalogue. We have now begun to catalogue Internet-only journals such as D-Lib [15] and Internet Archaeology [16] which are freely available. The jury is still out on what to do with the Academic Press titles!

We have taken the decision to use one catalogue record to describe both the print and electronic formats of a journal. But should there be one catalogue record for the print version of a journal and another for the electronic version? After all, they do have separate ISSNs and due to the increased functionality of the web versions they often contain different material. Take Ariadne as an example. But there is a major problem regarding e-journals - the administration of them is in addition to the work already involved with managing print journals. Libraries simply do not have the resources (specifically, time) to duplicate cataloguing. In many cases anyway, access to the electronic version is freely available for a limited period only. So, we do not envisage changing our practice in this regard and creating even more work for ourselves.

One advantage of cataloguing e-journals is to provide subject classification for them. They are not arranged by subject on LibWeb. As the full-text alphabetical list becomes longer and more unwieldy, it would be very helpful for our users to having a subject listing as well as an alphabetical one. We would actually prefer to dedicate time to this, rather than adding electronic-only journals to the catalogue. The electronic journal pages on LibWeb would then provide alphabetical and subject indices to available titles, although they would not be searchable in the same way as the catalogue. The catalogue tends to be regarded as the authoritative inventory of all the library’s resources. If something is available, the expectation is it should be included there.

Future developments

Our plan for the next academic year is to add to the catalogue a record for each e-journal listed on the full text LibWeb page (except for IDEAL and IOP titles, where we will wait and see what happens next year). We will continue to add the LibWeb URL to records rather than the e-journal URL, as we want to encourage LibWeb to be seen as the comprehensive guide and index to e-journals. We also plan to add a subject listing of our full-text journals to LibWeb, in addition to the alphabetical list. This will be linked to from the alphabetical list, and from the subject pages already set up.

We are interested in the electronic journal management systems being offered by organisations such as Blackwells [17], Swets [18] and OCLC [19] and are hopeful that a CHEST deal will be announced soon. We are also monitoring developments in the use of metadata and standards such as the Dublin Core [20] for cataloguing Internet pages. These will influence any future decisions we make on cataloguing Internet resources. In conclusion, we can say that the decisions we have made on cataloguing Internet resources have been influenced chiefly by practical considerations. We have weighed the time involved against the benefits for our users and come to a compromise solution which we believe fulfills a useful function while avoiding too much duplication of others’ efforts.

References

[1] LibWeb: University of York Library and Information Services Home Page
http://www.york.ac.uk/services/library/

[2] BUBL Information Service
http://www.bubl.ac.uk/

[3] NISS Information Gateway
http://www.niss.ac.uk/

[4] University of York Library: York Information Connections
http://www.york.ac.uk/services/library/subjects/subjects.htm

[5] University of York Library: Women’s studies information sources
http://www.york.ac.uk/services/library/subjects/womenint.htm

[6] EEVL: Edinburgh Engineering Virtual Library. The UK gateway to engineering information on the Internet
http://www.eevl.ac.uk/

[7] SOSIG: Social Science Information Gateway
http://www.sosig.ac.uk/

[8] The HUMBUL Gateway: International resources for the humanities
http://info.ox.ac.uk/oucs/humanities/international.html

[9] University of York Library: Chemistry information resources
http://www.york.ac.uk/services/library/subjects/chemint.htm

[10] University of York Library: Full text electronic journals
http://www.york.ac.uk/services/library/subjects/ejlist.htm

[11] IDEAL: International Digital Electronic Access Library from Academic Press
http://www.janet.idealibrary.com/

[12] Institute of Physics Publishing: Electronic journals
http://www.iop.org/EJ/welcome/

[13] BIDS JournalsOnline
http://www.journalsonline.bids.ac.uk/JournalsOnline

[14] University of York Library: Electronic tables of contents and abstract services
http://www.york.ac.uk/services/library/subjects/ejtoc.htm

[15] D-Lib Magazine: the magazine of digital library research
http://hosted.ukoln.ac.uk/mirrored/lis-journals/dlib/dlib/

[16] Internet Archaeology
http://intarch.york.ac.uk/

[17] Blackwell’s Electronic Journal Navigator
http://navigator.blackwell.co.uk/

[18] SwetsNet
http://www.swetsnet.nl/

[19] OCLC FirstSearch Electronic Collections Online
http://www.oclc.org/oclc/menu/eco.htm

[20] Andy Powell, 1997. Dublin Core Management. Ariadne: the Web version, Issue 10.
http://www.ariadne.ac.uk/issue10/dublin/

Author details