This article was published in The Serials Librarian, Volume 31 Number 4 1997
"In the hyperspeed reality of the Web world, a venerable institution is anything that's been around for longer than a year."
This quote is from a recent British newspaper article (Guardian Online 23 January 1997). The venerable BUBL has now been operating for six years, as outlined by Joanne Gold recently in The Serials Librarian (1). In fact BUBL is now so old that it has just had a facelift, undergone major surgery and moved to a new home in a new country. It is now entirely based at Strathclyde University in Scotland. This sadly means saying goodbye to the much-loved bubl.bath.ac.uk, but the new address is much shorter:
http://bubl.ac.uk/
This column introduces the main components of the new BUBL service, but first some basic background for those who have not seen the previous article.
A UK-based national information service intended for the higher education community, but freely available to all without registration. BUBL aims to offer fast, easy-to-use and reliable access to selected high-quality resources of academic relevance, both on its own servers and worldwide. The word BUBL is just a name, not an acronym.
Anyone who maintains links or bookmarks to BUBL should update them to the new address: http://bubl.ac.uk/. Links to specific locations, e.g. to the BUBL Subject Tree, also need changing. The new locations will be found under http://bubl.ac.uk/link/. Although the need to update links is inconvenient, the new URLs should be permanent, or at least should persist for as long as the BUBL service exists. Any existing BUBL users who have problems finding information on the new service should contact the BUBL help line by sending email to bubl@bubl.ac.uk
The aim of the new home page is to make each of the main BUBL services visible and accessible with just a single click. The main options are outlined below, though these will be extended as more services are introduced. It is planned to provide a similar look-and-feel and a consistent navigation scheme to each of these services.
This is BUBL's catalog of network resources. It is based on the long-established BUBL subject tree, but has a simple structure with a number of enhancements that are intended to offer some advantages over other subject-based catalogs available on the Web:
All resources stored in LINK are cataloged, using DDC. As well as the DDC number, entries contain abstracts, subject keywords (based on LCSH) and resource type classifications. For those familiar with the DDC system this permits a much faster and more accurate route to finding a specific subject category, as well as providing more helpful descriptive information about the resource.
All resources in LINK are searchable by a simple keyword search. However, the use of database software with catalog records allows much more refined searching if need be. It is possible to limit searches to the title, author, abstract or subject, in a similar style to a typical library OPAC. This will usually give a faster response than a more general search, as well as a more relevant result. Results are displayed via HTML pages generated dynamically from the LINK database.
Access to BUBL LINK is possible via Z39.50 as well as via HTTP, a feature offered by few other subject-based catalogs of Internet resources. This is a standard feature of the NetPublisher server software used by LINK, and the main reason for its use. Although Z39.50 usage is currently small in comparison to WWW, it is regarded by BUBL as strategically important in the longer term, as it offers the possibility of searching multiple catalogs simultaneously. For those with access to a Z39.50 client program, LINK may be accessed by connecting to database name ZPub at port 210 on link.bubl.ac.uk
In order to minimise the common problem of broken links, BUBL has a schedule for running link validation software to check the URLs in the LINK database. This does not eliminate the problem entirely, as machine faults and network delays still occur, but it should reduce it to more acceptable levels. There is however a limit to the automation of link validation at present. Whereas a validated link is known to work and can be retained, the fact that a link is broken on one or two occasions does not necessarily mean the resource itself has disappeared.
Although LINK now covers all subject areas, to a greater or lesser extent, it still has specialist coverage of library and information science. Whereas the old BUBL subject tree offered a single large file of useful LIS resources, in LINK the subject area is organised carefully according to DDC (see 020-029). For example, there is a section on acquisitions (025.2) that includes details of relevant journals and email lists, publishers' catalogs, resources on collection development, EDI, censorship and copyright issues, currency convertors and links to other relevant organisations.
One of the aims of LINK is to combine the flexibility of the popular Web search programs with a rigorous approach to resource inclusion and classification. This is intended to deliver a relatively small but highly relevant list of resources in response to search requests. To put it another way, 'less is more'. As yet the aims have only been partially met and a number of developments are planned to build on the initial service. As well as increasing the size and subject coverage of the resource base, there are plans in hand to extend classification of author types and resource types, so that it is easier to limit searches to, say, societies or electronic journals about specific subjects. Further developments should in due course allow users the option of searching specific groups of actual resources, held on remote servers, rather than simply searching the details about the resources that are stored within the LINK catalog.
Since BUBL began way back in 1991, an important part of its service has been the provision of electronic newsletters, journals, abstracts, and tables of contents. These have now been reorganised and brought together into the distinct BUBL Journals service. The content has been weeded and streamlined so that all titles that are no longer issued or no longer received by BUBL have been stored away in the BUBL Archive (see below). This leaves a total of 201 current titles which have abstracts and tables of contents stored on BUBL. Just over half of these are related to library and information science. There is also good coverage of titles covering social, medical and business topics. In the last three months of 1996 the most popular twenty titles, as judged by number of accesses, were as follows:
Many of these titles are published by Haworth Press, and BUBL is fortunate to have built up a good working relationship with Haworth, who allow BUBL to hold this information. It would of course be an even better service for users if they could simply click on the abstract of an article to access the full text, but leaving aside technical issues, this could have an adverse impact on publishers' subscriptions and document delivery services.
There are now a number of services, such as BIDS JournalsOnline (2), SwetsNet (3), and Ideal from Academic Press (4) which do offer the full text of a limited number of journal titles via the Web. These use different charging models, and it remains to be seen how they will develop. The provision of electronic access to full-text journals, and investigation of the associated control and copyright problems, is one of the major issues being address by eLib, the UK Electronic Libraries Programme (5). At present all material in BUBL Journals is freely available to all users without registration and without the need for any software other than a WWW browser.
For many titles in BUBL Journals coverage goes back to 1992 or 1993, though some have been introduced more recently. BUBL has always offered a current awareness service, and effort has been spent on covering recent issues of a large spread of titles rather than extending archive coverage of a limited number. Some of the work on transcribing abstracts is still carried out by volunteers at other institutions, so it is not easy to guarantee full coverage of all titles. However, recent attempts to locate and add missing issues from important titles have been reasonably successful.
Not all the 201 titles are refereed academic journals, as BUBL Journals includes contents and abstracts from a selection of newsletters and magazines. There are also several further titles where the full text is stored on BUBL, though these are a mixture of electronic newsletters and magazines, not refereed journals.
Searching is a vital part of BUBL. The BUBL Search service brings all the options together in one place. You can of course search the LINK database for specific keywords, or the complete body of information held within BUBL. Alternatively you can limit searches to a specific journal or to all the journal abstracts relevant to a particular subject area. Boolean searching is a standard feature throughout, but stem searching, using wild-card characters, is at present only available via BUBL LINK. BUBL Search also offers direct links to selected external search services; not just the most popular general WWW search programs, but also to many of the specialised search services which are now available.
This is a new service introduced in March 1997, to coincide with the launch of the new BUBL service. It is not a complete index of all UK WWW sites, nor is it a search service or a subject-based index to resources held within the UK. Instead it may be described as an institution tree, or even a UK home page. It provides links to the Web sites of numerous well-known institutions which have a real existence in addition to a virtual one. It is highly selective, and therefore provides a quick and easy method of finding important UK sites. Initial categories of institution include:
The links and categories will be expanded and revised, but the total number of links will remain relatively small, so that it is always easy to find sites by browsing as well as by searching.
BUBL began life as the Bulletin Board for Libraries, and this element of the service is retained in BUBL News. This service holds details of job vacancies (mostly UK or USA), forthcoming conferences and workshops, details of current offers such as surplus journal disposals, and other current news items relevant to the academic community.
BUBL maintains a number of mailing lists for the library community, including the well-known lis-link, which is one of the most active of the several thousand lists managed by the UK Mailbase service (6). Details of this and other lists and how to subscribe to them are available via http://bubl.ac.uk/mail/
This is BUBL's attic, a home for all sorts of old files that are probably no longer needed yet are somehow difficult to throw away. There are four main sections to the archive: Journals, Internet, LIS and Other Subjects. The journals archive is the most likely to be of interest, as it contains abstracts and full texts from over 100 different titles that are no longer held by the main BUBL Journals service. For some titles only a small number of issues are held.
This service contains details about BUBL itself: contacts, FAQ, philosophy, help files, funding details, copyright information and so on. Not perhaps of interest to most casual users, but the place to look for anyone wanting to know more about BUBL itself or how it operates. And if you can't find what you are looking for, or you have any other questions or comments, please feel free to send a message to bubl@bubl.ac.uk. We'd be pleased to hear from you.