Annual Review of Information Science and Technology (ARIST) Volume 37

Citation

Meadows, J. (2003), "Annual Review of Information Science and Technology (ARIST) Volume 37", Journal of Documentation, Vol. 59 No. 4, pp. 494-496. https://doi.org/10.1108/00220410310485848

For many information scientists, the annual appearance of volumes of ARIST has been a continuing feature of their working lives. Each year, they look to see which chapters will provide overviews of relevance to their own activities. To help answer that question for this year, here, in order, are the titles of the 11 chapters in the 2003 volume: “Information retrieval and the philosophy of language”; “Natural language processing”; “Indexing and retrieval for the Web”; “Electronic journals”, “The Internet, and scholarly communication”; “Visualizing knowledge domains”; “Museum informatics”; “Music information retrieval”; “The concept of information”; “Task‐based information searching”; “The role of trust in information science and technology”; and “Information and equity”.

As usual, information retrieval (IR) forms a common theme for many of the chapters. What comes over clearly is the problem of moving from an environment where the typical user is a professional concerned with specialised databases to one where the user is a member of the public who wishes to access sources on the Web. An obvious difficulty with the Web is the rapid growth in the amount of material available. As is noted in one chapter, an additional difficulty is that search engines are simultaneously changing their characteristics, but are still falling behind in terms of the percentage of available information covered. Whether the bias towards US sources will decrease in the future remains to be seen. (The references in the 2003 volume of ARIST do quite well on this score.) Over half of current Internet users are not native speakers of English, whereas over three‐quarters of the material available is in English. This situation, which could usually be coped with by professionals, is less easily handled by the general public. One result has been a considerable growth in work on machine translation and related topics. Not only is machine translation improving rapidly, but it is also beginning to undercut human translators in cost terms.

The importance of Internet searching is indicated by the fact that searching for information comes second only to email as an Internet activity. It has become evident that traditional forms of IR query require modification for user needs on the Internet. In addition new types of query – for example, about hyperlinks – need to be formulated. Some of the problems that have long been known take on a new urgency in the online environment. For example, browsing has always been an important way of gathering information. How should this work online? Judging by the reviews here, this is not a rapidly moving field. Indeed, some of the authors still seem to assume that information gathering essentially starts with an explicitly recognised information need.

IR outreach is increasingly going beyond text: both music and museum objects are discussed in this volume. In both cases, an immediate problem is the lack of generally agreed standards. For music, this is emphasized by some lack of knowledge of traditional text‐based techniques, and by the nature of the corpus of material available for study (mainly Western tonal music). In museum informatics (and this is the first ARIST contribution on the subject), the main problem is the much greater diversity of the data to be handled as compared with libraries. However, museums are increasingly seeing themselves as depositories of knowledge, rather than of objects. This parallels the similar change in emphasis going on in libraries. Convergence between these different areas of retrieval seems a live option.

A theme common to several of the contributions is the view of users as members of socio‐technical networks that are undergoing change with time. Many of the questions relating to trust can be formulated in these terms. Do users differentiate between the systems providing information and the information provided when they allocate trust? In particular, does the level of trust change if the information comes via an automated system rather than a human agent? The review of equity notes that most research has been aimed at the division between the information‐rich and the information‐poor. This access question, it is noted, only forms one part of the problem. The other part is whether access actually confers benefit on the supposed user.

This is a significant question for the use of electronic journals. It is partly tackled in the chapter devoted to such journals, but not fully pursued. Nor is a related point: that electronic journals may be more acceptable to institutions than to the authors they employ. Differences in attitudes to electronic journals between disciplines also deserve greater mention; but the chapter justifies its initial claim that the way forward for electronic journals is now becoming a little clearer. Differences within a subject are nicely illustrated in the chapter on visualizing knowledge domains. The authors have constructed a database on the subject of the chapter, and use it to demonstrate various kinds of display. The results show that information science continues to consist of a few high peaks connected by relatively low‐lying foothills.

This volume of ARIST has a philosophical flavour: Wittgenstein, in particular, appears several times in the index. His assertion that meaning in language comes only from its everyday interactive use is seen as a useful starting point for the discussion of information. This is surely a helpful approach (although I am never quite clear how it applies to my understanding of such Lewis Carroll poems as: “Twas brillig, and the slithy toves”). The attempt to apply the philosophy of language to the problems of IR in the initial chapter of the volume is well worthwhile. The conclusions are sensible, though, almost inevitably, that tends simply to mean that they coincide with conclusions already drawn from empirical studies by workers in HCI.

The chapter on the concept of information starts from a rather similar basis, and does a good job of wending its way through the minefield of definition. I especially liked the discussion of “persuasive definition” – the tendency to define terms in order to impress other people. My main problem in reading this chapter lay in determining whether or not the authors agreed with the people they quote. For example, Norbert Wiener is twice quoted as saying that information is not matter or energy. Whether or not he was right, depends on the definition of “information” that is used, but that point does not come over clearly here. More emphasis might also have been given to time changes in the definition of information, as the role of “information” becomes more complex. For example, when biologists introduced their definition of information for DNA, it hardly affected information science; but now it is becoming a part of the information science scene.

This last chapter, like the others, contains an excellent bibliography. The editor notes in his “Introduction” that he has tried to introduce new topics and authors in this year's volume. He has not only succeeded, but has also produced a volume that is likely to contain something of interest to most people concerned with information science research.