Each year the UK Electronic Information Group, a special interest group of CILIP, makes the Farradane Award. The Award honours Jason Farradane, who first made an impact on the LIS community with a paper on the ‘scientific approach to documentation’ presented at a Royal Society Scientific Information Conference in 1948. He was instrumental in establishing the Institute of Information Scientists in 1958, alongside the first academic information science courses in 1963 at the precursor to City University, London, where he became Director of the Centre for Information Science in 1966.
The winner of the 2020 Jason Farradane Award was Tom D. Wilson, Professor Emeritus, University of Sheffield, UK. I first discovered ‘information behaviour’ during a visit to the University of Sheffield in 2002 when Tom was on the staff team presenting the work of the Department (at that time it was the Department of Library and Information Studies) and as someone with a long interest in information management I will admit to having a ‘Eureka’ moment as many of the issues I had been grappling with in my work on information management started to fall into place. On learning of this award I asked Tom if he would write a feature article for Informer on his work on information behaviour.
T.D. Wilson Professor Emeritus, University of Sheffield
How people discover and relate to “information” has changed significantly over the last 50 years: when I worked in the nuclear energy industry in the late 1950s, organizing a library and doing literature searches for the scientists, there was nothing digital – everything you needed was in printed form. Discovering what my clients needed involved searching printed abstracting services such as Chemical Abstracts, Nuclear Science Abstracts, and Metallurgical Abstracts, buying journals, using the inter-library lending system, visiting the local public library and the university library.
My “information behaviour” at that time was, therefore, highly constrained by the printed form of the information resources and their local availability and the same would be true for any of the scientists or engineers who chose to do their own information searching.
Let us call this early information behaviour Search Mode I
In the late 1960s and early 1970s we saw the first impact of the computer in information management, in the development of the computer-based information services such as Dialog, ORBIT, MEDLINE, and, of course, Chemical Abstracts went online, which at least made the initial search process easier, although the telephone dial-up service, requiring some kind of teleprinter to report the results, could hardly be called fast! And, of course, the computer was a very different beast from today.
One early user of the online service records his reaction:
“The first time I used CAS Online, I felt as if I’d been given magical powers. Wildcards! Variable atoms and chain lengths! I talked everyone’s ears off about the kind of searching that could be done, and as I recall, the big issue was convincing people that the online method differed not only in degree, but in kind. You could, of course, do things that were completely impossible to do with the print edition, a fact that is mostly appreciated now by people who are at least 40 years old. That is, people who have done it the old way. Fewer and fewer do, and good riddance to it.” (Lowe, 2006)
Also, at this point, librarians and information managers were often the intermediaries because a certain skill was required in knowing which database to use and how to design a search strategy using Boolean operators. University libraries began to provide services on the basis of these systems and the cost of subscribing to the services and the telecommunication costs became significant elements in the university library’s budget. “Cost recovery” entered the library’s vocabulary at this time (see, e.g., Koch, 1982).
At the same time, professional education took on board these new developments and began to teach the use of computer-based services. At Sheffield, for example, all students in what is now the Information School acted as information researchers for staff in some other department of the University, interviewing to understand the researcher’s need, carrying out a search, taking the results back to the researcher, discussing what was relevant and possibly redoing the search, until the researcher was satisfied with the results. Sometimes the relationship took a non-professional turn and I know of at least one marriage that resulted! Online teaching aids were developed to aid instruction in the use of the databases (Wood, 1981), which, of course, became quite redundant after a few years.
The result of these developments affected the library’s aims and functions, professional skills, and researchers’ awareness of the potential of computer databases. There was, as yet, no major shift in behaviour, because the actual information resources remained in print form; we simply had a better way of finding out what was available, but we still had to photocopy journal articles, borrow books on inter-libary loan, and so on.
The situation for the academic researcher meant that he or she had to negotiate with the university library to have a search performed and the librarian might then gather the relevant material, or the researcher would take the research output and go off and find the relevant items in the library’s stacks.
The average citizen, however, would be in a different position, probably still limited to what was locally available, unless the public library also provided online searches. Business people, too, would use the local library’s reference library and probably the local chamber of commerce information services. However, we found that time was a severe constraint for the business manager, who would often give up the search for information if it could not be found within a couple of days, and would somehow manage without it (Wilson, 1987).
Let us call this Search Mode II
The Internet began to hit the academic world in the 1980s, but the real expansion came with the invention by Berners-Lee of the World Wide Web in 1989. This, together with opening up the Internet to commercial organizations led to the situation we have today. In 1997 one of the most significant developments took place: Garfield’s Science Citation Index became the Web of Knowledge (now the Web of Science) (Web of Science, 2021). This one development meant that academic libraries could now provide access to that database for its users, without the need for the librarian to act as a search intermediary.
With the invention and take-up of the World Wide Web, search engines began to emerge, although some had been developed for searching for files on the Internet before this, none of which, I think, has survived. Berners-Lee’s work at CERN was motivated by the need to manage information in one organization but he was apparently motivated by the 19th century compendium ‘Enquire within upon everything’, to the extent of calling his first search program, ENQUIRE (Berners-Lee, 1980). In effect, Berners-Lee was attempting to change information behaviour through a search process for hyperlinked documents.
‘Enquire within…’ is a kind of “Web in a book” – the topics listed in the Editor’s note can be searched for on the Web today:
modelling a flower in wax;
planning a dinner party;
making a will;
A video on YouTube on making wax roses has more than half-a-million views! (Kanigolla, 2013), and when I searched for “making a will” I got 13,800,000,000 results – plus ads!
In 1998, Google was invented by the PhD students Page and Brin and by 2000 was the most used search engine. Today it has 92.18% of the world market share (Law, 2020).
The open availability of search engines meant that computer-based searching was now available to anyone with an Internet connection – which in 2000 was only about 400 million, worldwide, but expansion has been rapid and now there are almost five billion people with access. In 2004, Google Scholar became available, meaning that the ordinary citizen now had access to the academic literature, or at least to open access documents and the abstracts of other scientific papers.
We now see the emergence over this period of Search Mode III – which is where we are today. Information discovery is not entirely dependent upon the computer and the Internet, since people still communicate directly with one another, ask questions and have them answered. The Internet connection, however, has become closer and closer to the norm, and the use of social media to ask questions and find answers is now common. Health information, in particular, is increasingly sought by these means, and by searching the Web (e.g., Fox, 2014).
It is worth noting that these developments do not instantaneously replace one another: they overlap, with people continuing with Search Mode I while others take up Search Mode II, and so on. Ultimately, however, and probably guided by the principle of least effort, and by the fact that particular search options fall out of use, everyone changes. The ease with which the Web can now be searched has probably made the shift from Search Mode II to Search Mode III much more rapid than the previous transition.
Does the fact that we are all now in Search Mode III mean that we all seek information in relation to our professional or day-to-day problems in the same way? This is unlikely: we all have different life experiences, different interests, different skills, and different roles. As individuals we fill multiple roles: the biochemistry professor is also a wife and mother, a daughter, caring for an elderly parent, a yoga enthusiast, a member of a political party, with a group of friends from her university days, and with another group of colleagues with similar research interests, scattered around the world. If an issue arises in relation to any one of these roles, which needs access to information, it is likely that she will know of different resources to draw upon to help her discover relevant information.
For example, if an issue arises in a current research project, she is likely to search Web of Science and Scopus, and perhaps Google Scholar or Microsoft Academic, to determine whether the problem has already been experienced by other researchers. But, if, in her “daughter role” she needs to find help for her elderly mother who is beginning to experience the first signs of dementia, she may search the Web for advice, or locate a discussion group on the topic, or, and perhaps more likely, contact colleagues in the medical school, or members of her friendship group to find out if any of them has experienced the same problem. Our information behaviour, in other words, is very much determined by the roles we perform, by the available search options and technologies, and by the context in which we find ourselves.
Compared with academics, other citizens are in a very different situation: academic staff and students in higher education have ready access to the Internet, at least in the developed world: the situation in developing countries is different. The ordinary citizen’s access, in the developed world, is limited primarily by his or her economic situation, giving rise to what has come to be known as the digital divide. And the most significant level of that divide is knowledge of how to use the technologies effectively, how to judge their authority, and knowing what to trust (van Dijk, 2002). However, the divide is reducing, gradually, and more and more ordinary citizens now use the Internet regularly to shop, to seek health advice, and to communicate with friends.
Given where we are now, what circumstances are likely to give rise to Search Mode IV?
Thinking about the future is a rather hazardous task: we can look at how things are now and imagine what developments might extend into the future, but we are unlikely to imagine some completely new development that changes the way we work, study, learn, or simply live our lives. No one was able to predict, with any accuracy, the arrival of the personal computer, which changed entirely how information is managed in organizations. No one predicted the fall of the Berlin wall and the associated independence movements in central and Eastern Europe, which completely changed societies in these areas. No one predicted the World Wide Web, apart, perhaps, from one or two science fiction writers who often seem to get things right, because their imaginations are not restricted by their daily circumstances.
Given the problem, however, we can make some attempt at thinking about the possible future, on the basis of what is happening now.
First, broadband coverage is gradually expanding everywhere, and we can imagine that, perhaps 10 or 20 years from now, the vast majority of people in the world will be able to have broadband access. Some countries have already achieved very high penetration of broadband, and we can expect that to continue (International Telecommunications Union, 2019). There will be pockets of poor coverage, mainly in rural areas – but some technological development, which is not yet on anyone’s horizon, may make delivery of broadband easier for such areas. Or, some countries, some cities, may decide that universal wi-fi coverage is needed as a public utility – Tallin, in Estonia, already has virtual universal wi-fi throughout the city, and, as governments rely more and more upon Internet connections to convey information to and collect data from citizents, the necessity to provide connection as a public good will become all the more obvious. The economic barrier to 100% access to the Internet will then disappear in those countries. President Biden’s plan to spend $100 billion on improving broadband coverage in the USA is an indication of how important politicians now regard this issue (Paul, 2021).
Secondly, we are now in a situation where almost 100% of current information output is in digital form: some newspapers have gone entirely digital; most journals are e-journals; books are e-books and digital audio books. Furthermore, digitisation of earlier, printed output, continues, and, while it may take some considerable time before everything held by all the national libraries of the world is digitised, it is likely to happen.
Thirdly, our personal information technology will become faster, lighter, more multi-functional. For example, we already have “smart watches” – not so different from those proposed in the American Dick Tracy comic strip of 1946 – which not only tell the time, but count our steps, monitor our heart beat, take our blood pressure (not very accurately at present), record our exercise routines, connect to our mobile phone to play music, alert us when an e-mail message is received, etc. etc. Battery life is the main limitation of such devices, and developments in battery technology, perhaps related to developments in electric cars, are likely to improve things. We also have developments in materials that may allow the tablet computer to be like a single sheet of paper, making the reading of e-books more like reading on paper than on screen, and weighing considerably less than a Kindle.
Fourthly, computer technology is likely also to change – with developments in quantum computing, already shown to be able to perform calculations much faster than present super-computers – the need for big server farms, which are a major ecological threat, may be reduced (Savage, 2020). Even personal computers may change in character, possibly using voice input to a much greater extent than at present, with the possible disappearance of the keyboard. Apple’s new “Silicon” range of machines with their own chips appear to be much faster and less power hungry than their previous models – and those developments are not going to slow down. Constant innovation is a basic requirement of the computer industry: how else could they persuade people to change the personal computer they have been using for three, five, or even ten years? We can imagine the “paper-thin” technology of the future iPad being transferred to the desktop machine. We may, at some point, have only a desk, with built in devices and a “pop-up” screen when needed – the landscape of the desk might change considerably.
Fifthly, there are developments in machine learning and artificial intelligence now taking place that will impact computer use and computer applications in coming years. Google already uses machine learning for its translation program (Caswell and Liang, 2020) and also uses a machine learning algorithm called RankBrain for searching (RankBrain, 2021). Couple these developments with future personal computing, and quantum computing, and the whole scene of data and information analysis may change significantly. Big data and quantum computing are natural partners, and coupled with machine learning could significantly affect the information that could be extracted from large data sets and, ultimately, large textual data resources.
Finally, there are associated social developments: the present pandemic has led to a changes in behaviour that are likely, to some extent, to persist into the future. One of the main developments is the increased time people spend in working from home instead of in the company office and the indications are that this has been found to be beneficial in terms of work/life balance and in productivity. Companies are now reducing their office space and introducing plans to allow home working for three days a week. At the same time, some companies are experimenting with a four-day working week, freeing up time for increased leisure activities. We have also seen an increase in online shopping, with the closure of shops in town centres (some 17,000 shops have closed permanently in the UK) – and perhaps, as a result, an increase in using price comparison sites. We can imagine then, an intensification of demand for broadband services, and an increasing need for security, as home-workers log on to company databases. An increase in online shopping also has implications for security and fraud prevention.
Under these circumstances, someone, say 10 years from now, engaging in Search Mode IV to discover what they need to know, will have increased needs for fast, secure services, and will have a much more sophisticated and perhaps “intelligent” or “smarter” system upon which to draw. Already, Apple’s Siri is a voice recognition interface for computers and smartphones, but imagine how much more effective it might be if the supporting technology was driven by quantum computing and machine learning. At present, if you ask Siri a question, it delivers a list of documents it has found – in future you may get a voice reply in addition, which summarises what it has found and gives you an oral answer, along with a text transcription. If we want a regular update on a research issue or problem, we may be able to set up Siri – or its future equivalent – to give us a daily, oral report as we have breakfast.
Search Mode IV, in other words, is going to be determined by the trend towards the universal availability of broadband and Internet connection; the universal availability of information; and smarter systems for access, search and delivery. The move from Mode III to Mode IV will take time, of course, and the developments will be faster is some countries than in others. But it seems to me that the shift is inevitable – but gradual: developments will take place over time and we may not notice the changes that are taking place. No one is building a new building to provide services – everything is going on the “back office” of the telecommunications and computer industries. We will reach the point, however, at which most information resources are digital, and remotely accessible, and where the information intermediary is a software “agent”.
Paper will persist in spite of these developments: books will survive because they are still a convenient package for reading for pleasure, and readable without the need for a battery charger and Internet connection; historical paper archives will survive, even if they are also digitised, as many are today. Brewster Kahl, the founder of the Internet Archive is collecting paper copies of books in the archive, because of the uncertain life of computer storage devices. But, overall, present trends are likely to continue, influenced by the technological developments that lie ahead.
Berners-Lee, T.J. (1980). The ENQUIRE system: short description. European Organization for Nuclear Research. https://www.w3.org/History/1980/Enquire/manual/
Caswell, I. & Liang, B. (2020, June 8). Recent advances in Google Translate. Google AI Blog. https://ai.googleblog.com/2020/06/recent-advances-in-google-translate.html
Fox, S. (2014, January 15). The social life of health information. Pew Research Center. FactTank. https://www.pewresearch.org/fact-tank/2014/01/15/the-social-life-of-health-information/
International Telecommunications Union. (2019). Measuring digital development. Facts and figures 2019. ITU. https://www.itu.int/en/ITU-D/Statistics/Documents/facts/FactsFigures2019.pdf
Kanigolla, K. (2013). How to make wax roses. [Video]. YouTube. https://www.youtube.com/watch?v=LbWVnhFWQbU
Koch, J.E. (1982). A review of the costs and cost-effectiveness of online bibliographic searching. Reference Services Review, 10(1), 59-66.
Law, T.J. (2020, December 11). Meet the top 10 search engines in the world in 2021. Oberlo. https://www.oberlo.co.uk/blog/top-search-engines-world
Lowe, D. (2006, August 28). Paper on a shelf. [Blog entry]. Science Translational Medicine
Paul, K. (2021, April 1). Biden plans to spend $100bn to bring affordable internet to all Americans. The Guardian. https://www.theguardian.com/us-news/2021/mar/31/joe-biden-infrastructure-plan-affordable-internet
RankBrain. (2021). In Wikipedia. https://en.wikipedia.org/wiki/RankBrain
Savage, N. (2020, September 4). Google’s quantum computer achieves chemistry milestone. Scientific American. https://www.scientificamerican.com/article/googles-quantum-computer-achieves-chemistry-milestone/
van Dijk, J. (2002). A framework for digital divide research. The Electronic Journal of Communication, 12(1-2). http://www.cios.org/EJCPUBLIC/012/1/01211.html
Web of Science. (2021, March 10). In Wikipedia. https://en.wikipedia.org/wiki/Web_of_Science
Wilson, T.D. (1987). Information for business: the business of information. Aslib Proceedings, 39(10), 275-279. https://doi.org/10.1108/eb051066
Wood, F.E. (1981). Online teaching aids from the Department of Information Studies, University of Sheffield, England. Online Review, 5(6), 487-494. https://doi.org/10.1108/eb024085