Library and bibliographic research in the context of LLAS requires an ability to locate, retrieve and evaluate information relevant to the researcher’s topic, wherever it exists. Information retrieval and information organisation (the science of indexing) bear a close relationship with linguistics, and a few examples of the way they interface are given in the attached bibliography (Herrera-Viedma 2003; Iwe 2001).
With libraries no longer defined by the physical materials they stock, ideas and data are now more likely to be accessed remotely than to be found in paper form in the student’s own institution. Learners are free to explore an abundance of electronic information sources. Student exploration undertaken without essential navigational skills, can waste inordinate amounts of time, and may terminate at completely inappropriate destinations (Green 2003). Thus the skill to assess the authority and provenance of an information source is as important as being able to locate it.
Until roughly twenty years ago, journal literature and the indices to it were both in paper form. Apart from personal contact with other workers in the field, researchers could rely on a handful of paper journal indexes, such as MLA (Modern Language Association International Bibliography) and LLBA (Linguistics and Language Behaviour Abstracts) to be confident they were keeping abreast of developments in their field. The advent of computer-automated KWIC (Key Word In Context) indexing systems made little difference to this state of affairs, since the resulting indices were still output in printed form.
From the late 1970’s business information needs combined with new technology led to the widespread availability of online bibliographic databases across a dialup connection via commercial hosts such as Dialog, Datastar, and STN. These organisations provided online, fully searchable versions of major library and subject-specific paper indices, including LLBA and MLA International Bibliography. Searching was costly – charging was on a minute-by-minute basis with additional costs for every bibliographic reference retrieved – and the interface to the search engine required skill to handle. For this reason, in academic institutions as in business libraries, searching was limited to a few trained intermediaries.
Around 1989 it became possible for UK academics to consult most online University library catalogues in the world via the Internet. Then came the invention and exponential growth of hypertext transmission across the Internet (i.e. the World Wide Web) from December 1990. Online bibliographic databases, which had previously only been available by the expensive route described above, became directly available to researchers and students on an “institution-wide” basis, by annual subscription. Finally, full text electronic journals began to be widely available across the Web. The search interface to datasets became much easier, and general knowledge of the way relational databases work spread. These changes, combined with a shift in emphasis to student-centred learning in higher education generally, meant that bibliographic search expertise that had previously been considered the preserve of an elite body of intermediaries was now considered essential knowledge for all students and researchers.
But side by side with the increase in online access, first, to secondary materials (bibliographies and library catalogues) and, subsequently, to the primary material itself (research documented in electronic periodical literature), came access to an enormous amount of material which was extremely variable in value – namely, the rest of the body of data on the Internet. The dubious premise that the act of publication provided a token assurance that the material published had undergone some kind of de facto “peer review” now obviously held no meaning at all. It is this last factor that has led to the emphasis on evaluation in library and bibliographic research – something that is quite new in historical terms.
3. General methodology
Given the confusing range of electronic sources on offer, there is a natural human tendency to stick to a “tried and tested” limited number of online sources. But institutions whose web pages enumerate a comparatively small number of major, well-recognised bibliographic sources run the risk of preventing their students from discovering resources that might benefit their study.
The aim of any teaching should be to enable exploration of the full range of information sources available, while fostering a critical awareness of their academic validity, thus developing the student’s ability to make meaningful judgements about the validity of sources. To this end, all available forms of online data should be exploited to the full.
The following four sections discuss four different kinds of resource, each of which may be exploited as an aid to bibliographic search in research and teaching, and give a few pointers as to their best use.
Early on in the history of the Internet, many academic sites attempted to compile lists of online resources valuable in individual subject areas – subject “resource guides”, “gateways”, or “virtual libraries”. Because the compilation has been largely on a voluntary basis, and the number of subject-specific resources has increased exponentially, there has been a tendency for such gateways to degenerate into long, undifferentiated “grocery lists”. Portals were an attempt to introduce an element of selectivity into the process of inclusion, and to build in an element of guidance. This guidance is usually achieved by listing resources under systematic, hierarchically organized subject headings.
Some problems implicit in these attempts at systematic guidance may be briefly illustrated by a glance at two portals which, from their authorship, might be considered to be good examples of the genre. While the British Academy PORTAL lists Linguistics sources under “Linguistics and Philology”, Language is treated, at the top subject level, under two headings: “Early Modern Languages and Literature” and “Modern Languages, Literatures, and Other Media”. Area Studies is not mentioned at all. The split between early modern languages and modern languages at this top level might confuse anyone using the classification scheme (as they are invited to do) to navigate to a group of resources of interest to them. The opening menu of Humbul Humanities Hub takes a completely different approach in its classification of the general area of LLAS, and one which at first might seem more user-friendly. Humbul lists Linguistics and six language/area studies groupings (German, Spanish, Russian and Slavonic, English, French, Italian) all at the top level of its subject menus. This is obviously not an exhaustive list – a student of Celtic studies, for example, might be surprised to find that heading suddenly popping up only after they had “gone down” a level in the hierarchy. In short, there will always be a potential mismatch between the way researchers and students view their subject and the way these portals attempt to organize it structurally.
It is not advisable to make too much of these problems. They are inherent in the use of all library classification schemes, which attempt to apply a permanent, logically justifiable structure to a constantly evolving body of knowledge or taught experience. The central question that needs asking when a list of useful portals is being compiled is not: Does this portal present the structure of my subject to me in a way that I recognize as familiar and useful? Rather it should be: Through it, can I find a useful resource I need, quickly? After all, these online libraries are similar to paper libraries – where it is sometimes possible to find useful resources by browsing, but where the vital tool which needs careful evaluation is the search mechanism. The basis of the “added value” brought by portals over conventional search engines (see 3.4 below) is not the particular classification scheme they use, but the compilers’ selection criteria. Are the resources listed consistently useful? How well have they been indexed? Could a user find the information they sought – a translation of a word, a root, a theory – on this website more easily than via a good Internet search engine?
It seems to be a rule of thumb (perhaps for logistical reasons) that the less subject-specific the portal, the more easily objections may be levelled at its structure and selection criteria. Perhaps the future of subject guidance lies in narrowing the subject range of portals. In the field of Area Studies a commonly cited example of an excellent portal is LANIC, the Latin American Network Information Centre at the University of Texas. Although LANIC covers a vast area of study, it is more firmly focused than is possible for general Humanities portals such as the British Academy Portal or Humbul Humanities Hub. Using this portal, I have answered quite specific information queries with as few as four mouse clicks, without recourse to its Search facility, and this may be taken as a measure of how well it is designed and how well the content is selected.
Perhaps the best use of portals of this kind is at a local level, where good co-operation between teaching departments and library staff can ensure that the students’ needs are catered for more precisely, without degenerating into “spoonfeeding”. At Warwick University, the German Department has for many years worked closely with the Library to develop resources and training for e-learning. The Warwick German Studies Web now contains some 1200 links to scholarly resources on all areas of German Studies. But the site is careful to begin with the local (locally created bibliographies and resources for teachers), progressing outward from there to “the general”.
3.2 Virtual training suites
An example of a JISC-funded initiative, the RDN (Resource Discovery Network) Virtual Training Suite aims to help students, lecturers and researchers in UK higher and further education to develop their Internet information literacy and ICT skills. The core of the site is a series of subject-specific Internet information skills tutorials. The Modern Languages tutorial successfully combines the twin tasks that such tutorials need to address. They must impart the generic, transferable skills common to Internet research in all subjects, and at the same time provide a starting point in the form of a tour of selected subject-relevant sites.
One of the difficulties faced by all such non-local web-based tutorials is that of allowing for variations in the local subscription-based services available to any individual student. Selecting “quality” Web resources for inclusion in subject listings is a good idea in principle, but can lead to massive user frustration. Many good resources worth including cost money to subscribe to. So, for example, under “Infrastructure” resources, the RDN Modern Languages tutorial lists OCLC Firstsearch – an excellent resource, but a database unavailable to many users of the tutorial, since it costs money, and the learner’s home institution may not subscribe to it. It is important for teachers to take into account local access conditions.
3.3 Database help files
There is no substitute for a well-designed help system in teaching the mechanics of bibliographic searching. Fortunately, most database providers realise the advertising benefit of making their help systems as comprehensive and easy to use as possible. Academics should avoid duplicating an effort already well made; help files should be exploited fully where they are available. Most are capable by themselves of imparting the basics of Boolean search logic, and the essential skills needed to successfully search structured databases.
Cambridge Scientific Abstracts set a good example: their implementation of LLBA (Linguistics and Language Behaviour Abstracts) combines a simple interface with different levels of searching and search help to match the differing levels of previous experience and expectations which users bring to their search.
WebSPIRS, who provide an implementation of the Modern Language Association International Bibliography, have achieved wonders. The designers of the search interface have invested a lot of thought in making it easy for users to match natural language search terms to the database’s controlled subject thesaurus, and to develop complex search strategies fairly rapidly. However, the excellence of WebSPIRS’s help system is somewhat masked by a visual interface which can be confusing to the first-time user. Most commands open a separate browser window.
Search engines are not merely included in this review of good practice for the sake of completeness. They received a bad press from information scientists in the early days of the World Wide Web. Search engines claimed to index a far greater body of data than in fact they did. They claimed to compile their data from “web crawlers” or agents which trawled most of the World Wide Web, when in practice many of them concentrated on indexing those commercial websites whose owners paid the advertising revenue which enabled the search engine companies to continue in existence. In addition, their search mechanism was primitive when used at a basic level (retrieving astronomical quantities of irrelevant data).
But many search engines – Google is an outstanding example of one – can be used to good effect, if their advanced help files are consulted properly, and good guidance sites exist on the web for the evaluation of different search engines. Northumbria University Learning Resources offers a comparative table of Web search service features, and Search Engine Watch provides useful background data on the way individual search engines work and the extent of the data they index.
4. The way forward
From the discussion of the above examples, it is evident that the skills referred to in the Introduction (ability to locate, retrieve and evaluate information relevant to the researcher’s topic, wherever it exists) need to be a requirement of learning and actually built into the curriculum.
Virtual Learning Environments (VLE’s) such as Blackboard or WebCT provide an excellent platform for delivery of the research skills element in courses for two rather obvious reasons. Firstly, with VLE’s the learning experience is already taking place in a Web context, in which the concept of exploration is implicit. The learner is already halfway there, so to speak. Secondly, VLE’s are increasingly used in the area of LLAS teaching (a keyword search on Blackboard within this Guide yields several examples of projects reported on). They may soon represent the normal standard for delivery of undergraduate and taught postgraduate course content nation-wide.
4.1 Examples of existing training courses
The two examples of good practice at undergraduate level discussed below are not specific to LLAS. This does not alter their relevance to this Guide. Once a research skills module in a particular subject area has been devised using a VLE platform, transferability to another subject should not be a major issue – examples of specific databases, datasets, and searches can be changed while keeping the structure of the original module.
At the University of Southampton, New College has introduced the Blackboard course Information Handling Skills (Information handling skills (2003)) as a fully assessed, compulsory level 1 core unit within its Vocational Programmes, which comprise degrees in Information Technology in organizations, Political Communication and Media Management, and Sport management and leadership. These degrees are taught at two remote sites in addition to Southampton. The course was not a library initiative – it was requested specifically by the Programme authors – and has received a measure of HEFCE funding in recognition of the fact that it is breaking new ground. It carries 10 credits and aims “to develop information handling in the context of skills needed for study at higher education level”. Specific intended objectives of this course include: being able to carry out an effective literature search, being able to assess the academic validity and use of specific information sources, and being able to demonstrate competence in using a range of information skills to pursue independent research.
In its structure, Information handling skills (2003) follows the “seven pillars of wisdom” Critical Success Factors module outlined at a SCONUL (Standing Conference of National and University Libraries) conference at the University of Warwick in July 2000 (SCONUL (2000)). The seven critical success factors are (1) Recognise information need, (2) Distinguish ways of addressing gap, (3) Construct strategies for locating (4) Locate and access, (5) Compare and evaluate, (6) Organise, apply and communicate, and (7) Synthesise and create. The sessions which comprise the core activities of the course begin with a definition of terms (What is information?), and a discussion of use of different types of information sources. Subsequent sessions focus on specific kinds of materials (journals, electronic indexes and electronic journal articles), literature searching strategy, effective use of the Internet as a whole, finishing with a virtual/practical visit to an academic library and evaluation of its information finding systems. The Blackboard model has been so successful that already other course designers at Southampton have asked to be allowed to use elements of it in Humanities programmes.
The Open University has written, for its own students, a web-based information skills package, Safari – Skills in Accessing, Finding, and Reviewing Information. Safari could equally well have been discussed under heading 3.2 above (Virtual Training Suites), because it is not assessed as part of the student’s course. Nor is access to Safari restricted to OU students (although what you get from guest access is of course restricted). But it is included here because it is closely tailored to the needs of students at one institution (the OU). Like Information handling skills (2003), Safari follows the SCONUL “seven pillars” module very closely. Its seven sections are headed: Understanding information, Unpacking information, Planning a search, Searching for information, Evaluating information, Organising information and Where do I go from here?
4.2 Other work in progress
The UCML Collaboration Programme in Modern Languages in Higher Education consists of ten projects several of which will promote the development of bibliographic research skills in their end product. The three most relevant are the Web materials in European Area Studies project, the Database of current research in Modern Languages at Oxford Brookes University, and Project Hi-Web (Materials for postgraduate research training in Romance Studies) – the last of which is one of two projects that received extra funding at the end of 2002. The LLAS Subject Centre also organises various research training events related to research skills – for example, the Resources Information Day in Linguistics organised for 4 June 2003.
4.3 A look into the future
Technology has the capacity to alter many aspects of the bibliographic search process immeasurably, simplifying searching of disparate types of materials and eliminating much of the “drudgery” element.
BUILDER (Birmingham University Integrated Library Development and Electronic Resource), a three year JISC-funded project based at the University of Birmingham which finished in December 2000, was one of a number of “hybrid library” projects which formed part of phase 3 of the Electronic Libraries (eLib) programme. BUILDER is important because it explored and elaborated many elements in the possible future design of libraries as mixtures of traditional and digital resources (hybrid libraries). Its hybrid library demonstrator “aims to demonstrate ‘seamlessness’, where all resources are accessible through the same interface, and profiling, where the resources offered are relevant to the needs of the individual user”, while its metadata index demonstrator “aims… to produce a high level metadata index to printed and electronic information resources held in a variety of formats (for example, Web-based resources, electronic online databases and CD-ROMs).” Both these demonstrators are available online from the BUILDER site.
Another international project that has been in progress for some while, and closely related to BUILDER’s metadata index in concept, is the Dublin Core Metadata Initiative (DCMI). DCMI is “an organization dedicated to fostering the widespread adoption of interoperable metadata standards and promoting the development of specialized metadata vocabularies for describing resources to enable more intelligent resource discovery systems.” Since 1995, DCMI “has been committed to the continual refinement of a “core” foundation of property types and values to provide vertically specific (semantic) information about Web resources, much in the same way a library card catalog provides indexes of book properties.” The initiative provides a set of 15 descriptive semantic definitions , representing “a core set of elements likely to be useful across a broad range of vertical industries and disciplines of study”. If universally applied, Dublin Core Metadata could be used to supplement existing methods for searching and indexing Web-based metadata, regardless of whether the corresponding resource is an electronic document or a “real” physical object. Its success (or not) depends on the degree to which it is adopted as a standard, not just by governments, but by the academic community as a whole.
However successful hybrid libraries and metadata element set applications turn out to be, it is likely that one key requirement for bibliographic research skills will not change in the foreseeable future. The ability to evaluate published material and to relate it to the subject being studied in a meaningful way is a difficult skill to impart. We can imagine tools that will make its application easier (for example, a universal academic website “seal of approval” might be developed and applied universally alongside the “Dublin core” metadata elements), but it is difficult to conceive of technological developments that will eliminate the necessity for human evaluative skill as a principal element in the bibliographic research process.
Brown University Library (2009) Bibliographic managers. Available at: https://www.youtube.com/watch?v=DiMpNDwwL9s (Accessed: 13 November 2016).
Conner-Sax, K. (1999). The Whole Internet: The Next Generation. Cambridge, Mass: O’Reilly.
Green, R. (2003). “E” is for Exploration. Interactions 7, 1. Available at: http://www.warwick.ac.uk/ETS/interactions/vol7no1/
Herrera-Viedma, E. (2003). Evaluating the Informative Quality of Documents in SGML Format from Judgements by means of Fuzzy Linguistic Techniques based on Computing with Words. Information Processing & Management 39, 2:233-49.
Information Handling Skills (2003). Information Handling Skills: University of Southampton New College Unit handbook ZA102: Information Handling Skills Semester One 2002/2003. Southampton: University of Southampton New College.
Iwe, J. I. (2001). Linguistics and Information Processing: Provision of Syntactic and Semantic Consistency in the Language of Library of Congress Subject Headings (LCSH) Pertaining to Literature and Librarianship: A Comparative Analysis. Cataloging and Classification Quarterly 32, 2:107-26.
SCONUL (2000). Seven pillars of Wisdom? Good practice in information skills development: proceedings of a conference held at the University of Warwick on 6 – 7 July 2000. Ed. S. Corrall & H. Hathaway. London: SCONUL. ISBN 0900 210 230
Shaw, D. (1995). Bibliographic Database Searching by Graduate Students in Language and Literature: Search Strategies, System Interfaces, and Relevance Judgments. Library and Information Science Research 17, 4:327-46.
Blackboard. Available at: http://www.blackboard.com/
BUILDER. Available at: http://builder.bham.ac.uk/
The British Academy PORTAL. Available at: http://www.britac.ac.uk/portal/
Cambridge Scientific Abstracts (for Linguistics and Language Behaviour Abstracts). Available at: http://www.csa1.co.uk
Dublin Core Metadata Initiative. Available at: http://dublincore.org/
Google UK. Available at: http://www.google.co.uk/
Humbul Humanities Hub. Available at: http://www.humbul.ac.uk
Lanic: Latin American Network Information Centre. Available at: http://lanic.utexas.edu/
OCLC FirstSearch. Available at: http://newfirstsearch.uk.oclc.org/
RDN (Resource Discovery Network) Virtual Training Suite. Available at:
Resources Information Day in Linguistics. Available at: http://www.llas.ac.uk/events/archive/595
Safari – Skills in Accessing, Finding, and Reviewing Information. Available at: http://sorbus.open.ac.uk/safari/signpostframe.htm
Search Engine Watch. Available at: http://searchenginewatch.com/
Tecla: Texts for learners and speakers of Spanish. Available at: http://www.sgci.mec.es/uk/Pub/tecla.html
UCML Collaboration programme in Modern Languages in Higher Education. Available at: http://www.llas.ac.uk/collaboration.html
The virtual environment learning blackboard: uses and limitations in the teaching and learning of four languages. Available at: http://www.llas.ac.uk/resources/1422
Warwick German Studies Web. Available at: http://www2.warwick.ac.uk/fac/arts/german/resources/
Web search service features: comparative table from Northumbria University Learning Resources. Available at: http://www.unn.ac.uk/central/isd/features.htm
WebCT. Available at: http://www.webct.com/
WebSPIRS (for Modern Language Association International Bibliography). Available at: http://arc.ucl.ac.uk:8590/