As a librarian I became conscious early in my career that libraries were often not planned or operated to suit human beings. This consciousness was impressed upon me by working under a librarian who was a world authority on frustrating not only her staff but all her users to a degree one had to experience to believe. The only way many users could get what they wanted was by asking the staff, when the Librarian was not around – if she was, she would stop enquiries by saying ‚Professor X knows perfectly well how to do that’. This made me wonder what a usable library might look like, and led me into several years of research into information needs and habits, and into a lifetime’s thinking about the question: ‚What would a library look like that was designed around people?’ This may seem an obvious sort of question, but it seemed almost revolutionary in the 1960s, when many librarians apparently operated on the basis that people could be redesigned to fit libraries.
After working for many years first as an academic then a national librarian, I now operate independently, doing research at various levels, from that necessary to write scholarly articles to more selective searching in order to give talks on topics for non-professional audiences. As such, I use various types of library: special libraries, public libraries, academic libraries, and the British Library, both remotely through the Document Supply Centre and on the spot in the St Pancras building. My first-hand experience of being a user – a very demanding one at that – has reinforced my concerns.
When I enter a library that is new to me I feel a little of the confusion that must be common among untrained users. Information searching is something where one can easily make a fool of oneself; and I understand those who would rather make fools of themselves at home on the computer than in a public place.
Until about ten years ago, libraries had it more or less all their own way as a means of access to publications when online access to text reached a critical mass and users began to have an alternative. In the Good Old Days if people wanting information did not use libraries it was either because they were odd people or, much more commonly, because libraries were very unattractive to use.
‚Unattractive’ is too weak a term for some libraries that I know. I could name more than one national library that does its best to deter readers by its architectural characteristics alone. The approach is up a massive flight of steps, up which the intrepid would-be user ventures in a rather similar mood, I guess, to that of the sacrificial victims who went up the pyramidal temples of ancient Mexico. One is then confronted by a huge portico in the classical or Egyptian style, with monumental bronze doors ornamented with bas-reliefs of past literary heroes (or perhaps past sacrificial victims). Passing through these by the grace of God, one is greeted by an official with a uniform resembling that of Latin American police, who scrutinizes one’s credentials. If one gets past him, one is met by a member of library staff whose job satisfaction consists solely in either preventing or delaying further incursion. That is only the beginning. In dismay and confusion, one looks around for something familiar, such as a catalogue; and then for the appropriate reading room. Since the signposting (if there is any at all) is designed as a display of lettering rather than as guidance, all this takes time and adds to confusion. It is not surprising if many would-be users avoid the library altogether if possible. There is at least one recent proof that it is possible to construct modern national libraries that contrive to combine non-functionality with ugliness and extravagant cost.
Nor are these architectural features confined to national libraries. In some universities the library still appears to be seen by the Principal as one of the few buildings on the campus that is ideal for the exhibition of architectural talent – the first building to take distinguished visitors to, so that they can marvel at the architect’s genius and the principal’s own wisdom in appointing him (sorry, but it always seems to be ‚him’; women should in this case be pleased to be largely exempted). I have often thought that it would be costeffective to commission a building known as The Architect’s Building; this would serve no functional purpose whatever, and be used solely for display (though the architect might be compelled to live in it for one month every year) – all on condition that every other building had to be entirely suited to its function. I have seen plans for a library that had no provision for toilets, which the architect, a famous one who consistently preferred his conception of beauty over function, doubtless considered a needless concession to materialism. I have also seen an actual library in a new college that would have been lethal to staff and users if it had not been heavily modified before occupation; there were narrow balconies at two levels with nothing between them and a floor many metres below apart from one narrow rail – if anyone bent down to get a book, they were below the rail and in real danger of a nasty death (possibly more than one if readers were underneath). There was also no lift between the floors. When these deficiencies were pointed out to the architect by the new librarian, who had been appointed after the architect had completed the building, he said with astonishment ‚You’re not going to move the books once they are in place, are you?’ In general, the more famous the architect the less functional the building.
Most architects today consult with the librarian, though this is not easy when the institution is a new one, as was the case with the library just described. One British architect specialised in library buildings, and always did his best to consult, but even he was not immune to faults. One of his buildings was hexagonal In shape, with no obvious points of orientation; the inevitable result was that users regularly lost their sense of direction. I have used the library myself, though not frequently, over many years, and have never fully mastered the layout.
No longer do libraries have a monopoly on the provision of access to information. Some of the competition is physical, some virtual.
The physical competition comes from bookshops, which have moved more and more to an American model in recent years. Instead of the genteel and cramped shops of the past, many of the new bookshops are very inviting to step into, attractive in their layout and design, and generously laid out. Questions are positively invited of the staff, who obviously benefit themselves from working in such conditions. There are lounge areas, where customers are encouraged to browse; coffee is often available, and sometimes light refreshments. These shops are pleasant just to spend a quarter an hour or so in when one is visiting a town for shopping or other reasons; they are beginning to supersede public libraries for this purpose. I know which of the two I would rather spend a short time in, except in a few cases.
Public libraries in particular, but also academic libraries, are taking notice. Many now provide relaxed lounge space (as opposed to reading seats and tables), and some offer coffee. Obviously no drink or food can be allowed in areas where rare or valuable books are present, but the vast majority of books do not fall with this category. The problem for public libraries in particular is that they are mostly in buildings that were designed in a different era, and the adaptations that are possible usually fall well short of what is needed to compete with a modern bookshop.
The same is true of academic libraries, which also need to make themselves as attractive as possible. One might think that there is little flexibility of demand in the academic world: researchers and students need information, and they have no choice but to come to the library. This was never true, even before the Internet came along. There are numerous cases where an old building has been replaced by a new one, often much bigger as well as more attractive, and use has greatly increased, sometimes twofold, within a few months.
However, the competition they face is different from that confronting public libraries. The Internet has totally changed the whole scene of information access in the last decade. Most current scientific and technical journals are now accessible online by anyone with a computer and modem. More and more social science and humanities journals are joining them. In addition, there is a huge mass of up-to-date information of varying quality accessible on numerous websites. A student told me a few years ago that he never used the library; he ‚always managed to find something on the Internet’. When I objected that this might be unreliable as well as incomplete, he said ‚It’s enough for my purposes’. You may think this is deplorable – especially as he was studying information science – but it is a fact of life. I myself use the Internet a great deal for the purpose of preparing talks to local organisations, though I am careful to check all the facts as far as possible.
We have not yet reached the stage where e-books are commonplace, and it is possible that we never will; but it seems almost certain that computerised files of whole books or collections of passages from several books will soon be widely available in one form or another; an early use of these will be for material in heavy demand by students, thus solving a problem that libraries have never been able to solve – the problem of the same book being wanted by 20 students at the same time.
Traditional books will, I believe, never be superseded for all purposes; leisure reading in particular, but also books in such subjects as history or literary criticism that demand intensive reading. I do not recall retrieving anything off the Internet that could remotely be described as leisure or intensive reading. However, online publishing, which seems at last to be a realistic prospect, will provide conventional-looking books, ready bound, to readers on the spot; this will be especially useful in the case of books that are in so little demand that normal publication is not economic. There seems to be no fundamental a reason why libraries should not supply books in this way, as well as booksellers; the main issue is likely to be the terms on which they might do so. Another option in the case of academic institutions might be for the university bookshop to do the printing for the library. The library’s role in online book production seems to me a matter that calls for rather urgent consideration and action.
There is of course an alternative to borrowing from or consultation in libraries for books: purchase. Many books of all types are, and have been for many years, available as paperbacks, not all of them exactly cheap but hardly expensive either. So far as I know, no attempts have been made to study the effect of paperback availability on library use, but it cannot be small. (There might even be some positive effect, in that purchase might stimulate a reading habit that was then satisfied by public libraries.)
So it looks as if readers will within a few years be able to get the vast majority of what they want without going near a library. Some are doing this already, including me. Why should they want to use a library?
Well, there are things they cannot do without a library, things they do not want to do without a library or prefer to do with one, and things that libraries could do for them that are not strictly necessary – ‚added value’ services.
First, an ‚ideal’ world where everything is or could be digitised is unlikely ever to come into existence. It is not a case of either electronic information or printed matter: it is a question of what medium is best suited to what matter. There will always be documents, especially in the humanities, that must and will remain as they were produced. The cost of digitisation is simply too high, whether the source pays or whether the user pays; I personally cannot imagine acquiring scores, perhaps hundreds, of documents electronically, or expecting the library to acquire them, especially as I would not need to spend more than 30 minutes with some of them (five minutes would be enough in some cases to decide that they were of no interest). Remote consultation via a video link would be scarcely less consuming of resources.
Secondly, there may be documents that researchers need to look at in their original form; an electronic version would not do. (On the other hand, some originals can be enhanced by technology – witness The British Library’s Electronic Beowulf – but the cost of such exercises is much too high to make it generally useful.)
Thirdly, while it may be possible to use electronic versions of wanted documents, it may be slow and inconvenient. Visual scanning of a printed book to gain an impression of it or to look for items of likely interest is much easier, faster and more effective than trying to do the same with an e-book. The same applies to current journals; there is no substitute for seeing the original print on paper for keeping up in a subject or gaining new ideas. And comfortable as e-books may be or become, I doubt if they will be a good replacement for reading and absorbing, as opposed to consulting, an entire book at leisure. Nor is the Internet a perfect substitute for printed reference materials; it undoubtedly has vastly more material than any home could have, and can supplement even the richest library (especially for recent information), but 90% of the small queries people want answered can be dealt with by pulling a reference book off their own shelves (assuming they are reasonably well equipped with such books), much more quickly than they could find the same in-formation online.
Fourthly, while intelligent users learn to deal with most computer problems after a while, there are bound to be times when they would welcome the help of someone more expert. ‚Welcome’ is sometimes too weak a word: for some problems help may be needed urgently. As proficiency continues to improve, the system gets more complex – not in the sense that it is harder to use, but there are many more useful things that users can learn to do – so they never really catch up. The same applies to finding the most effective search methods and processes. It also applies to identifying websites; a knowledgeable colleague can save hours by pointing to a good website or finding short cuts to the information wanted. It is incidentally interesting that people who complain that libraries are hard to use are willing to spend several hours online (and hence expensively) struggling with web searches.
I said a few minutes ago that most information can be found without going near a library. But that does not mean that it cannot be supplied by a library remotely. As the number of distant learners registered with universities continues to increase, and some universities are contracted to offer courses remotely to other universities, the provision of remote access is no longer an optional choice.
As stated earlier, people need exposure to information. Much of what they read they do not deliberately seek; they find it, on library shelves (or in the case of new books, bookshop shelves). Even in science, much is picked up by browsing (what might be called vague searching) rather than specific searching. Browsing needs to be distinguished from serendipity, which is lighting upon things by accident; it can occur while searching (vaguely or specifically) or while not looking for anything. I am frankly worried by the thought that one day this exposure might cease to exist; after all, I can only search specifically now because in the past I was exposed to a range of material. How can one search when one has no overview or range of knowledge from which to draw search terms?
It is, incidentally, hard to browse in many research libraries, which put most of their stock in closed stacks. I would plead for much more stock on open access; I know this goes against the continental tradition (as opposed to the Anglo-American tradition), but users should take preference over tradition, however venerable.
Several attempts have been made to devise electronic surrogates for browsing and serendipity. The latter is in fact quite easy in principle: all that is needed is to flash random pieces of information on the screen from time to time. However, the interruptions would almost certainly infuriate users (just as television advertisements irritate me), who would switch the facility off. Browsing is much harder to reproduce, and I know of no successful system.
One big problem is that of identifying in advance material likely to be of interest. Every researcher must have called for items that prove useless when ob-tained: a small matter when called from the stacks of his own library, a slow and expensive one when it is obtained from elsewhere. Identifying potentially useful books could be made far easier by various additions to the basic bibliographic description. Adding contents pages is a cheap and obvious improvement, which has already been tried but not widely adopted. It does not work in all cases, since some contents pages have only chapter numbers, and some chapter headings give little or no idea of what the contents of the chapters are. Codes for audience level (e.g. children, popular, academic, etc.) should be essential elements in any description. Better still would be very brief outlines of what the book is about, followed by short reviews, as objective as possible. And is there any reason why reviews or critiques should not be included, as they are on Amazon?
If identifying useful books is hard enough, searching within individual books or journals is even harder to do in the absence of the item itself – in fact, virtually impossible. We complain about the indexing of websites when there is a lot to do by way of making ordinary printed materials more accessible – although there are solutions that could be applied. I have commented in the past on the irrationality of providing three or four subject terms in databases for a straightforward chemical single-topic journal article of nine pages while giving a book of 500 or so pages only two or three index entries in national bibliographies. The solution is to put book indexes, with all their inadequacies, into electronic files. An integrated file of large numbers of indexes would be totally unusable, so there would have to be a two-stage process: a broad search to identify potentially useful books, followed by calling up their indexes one by one for detailed searching. We have all the necessary technology for this, and publishers would probably not resist making their indexes available because it would increase usage of the books. Libraries or no libraries, remote access to the contents of books is surely becoming more important, as Amazon recognizes.
The likely relevance of journal articles could also be made easier to identify in advance. Abstracts do much of the job, but the articles themselves could be structured in a hierarchical and sectional form: list of contents – summary of each section – full text. It is the contents and summary, which require more rigorous structuring of the text, that are missing.
The above improvements would all aid researchers. They would be valuable both in small libraries, which lack much of the material that researchers may need, and in large libraries, where much of the stock may be on closed access and the sheer size of the collection inhibits shelf browsing. All the measures are easily feasible technically. Some are easy to put into practice, others less so, but they are all practicable if the will is there. They are not browsing, but they would be very useful surrogates for browsing. They could be accessible at home – at a price. Most are completely out of the hands of libraries, except that since libraries constitute a main market for information services they are in a strong position to plead for such improvements, and they should combine to do so. Dare we also hope that an entrepreneurial librarian might have the courage to leave the profession and go into business developing some of these services? Eugene Garfield did this with his citation indexes, and there must be more than one Garfield in the world.
Scientists and technologists are far more able to manage on their own without libraries than other researchers. Most of what they want is already in digital form and readily (if expensively) accessible. Social scientists working in current areas too are not badly off, whether they are sociologists wanting journal articles or economists using large databases of statistics. As noted, people may use libraries for reasons of necessity, convenience, or desire for assistance. Another reason is that it is both tedious and unstimulating to work at home all the time – a point on which I expend later. But one of the most important reasons is cost:
It is usually much cheaper to consult material in a library or borrow it than call it up at home from electronic versions, even if the substantial cost (more than £1,000, or 1,600 euros, a year, I reckon) of maintaining an up-to-date computer is discounted, as it probably should be – it is on the way to becoming a household necessity. Cost is also a major factor if the numerous charges – for searching databases of references or texts, for downloading texts, for acquiring books on demand – are not borne by the library. In the case of electronic journals research libraries commonly have site licences that allow them to provide access to designated journals on agreed conditions. One would expect similar commercial arrangements to be made for e-books. However, downloading texts of whole books, or having them printed out on de-mand, could become very expensive, and most libraries would not consider that there was enough in it for them unless the books were likely to be wanted by several people – even supposing retention of a downloaded copy for multiple use were to be permitted by publishers without a substantial additional charge. The question of what libraries will pay for, what they will charge to departments if they are attached to universities, and what they will pass on to individual users has still to be fully answered, but it is becoming fairly common practice for libraries to give their users access to journal articles free. This risks some abuse (not all use of the web is academic), but the hassle of monitoring use and charging selectively is presumably too great to be worth it.
That is all fine for researchers who are attached to universities in some formal way, but many (like myself) are not. In the UK, it is usually possible to become an external member of an academic library with full or partial privileges. But this costs money – £100 a year is a rough average fee for access. If this includes borrowing, it is usually restricted to a few items at once, on the very reasonable grounds that the university’s own members must have priority. However, some site licences now permit free access to electronic material for external members, and it is up to the individual university to give them this right or not; there seems no reason not to, since there is little conflict here with local needs unless (as may well be the case) there are not enough computer terminals. For those who cannot for one reason or another become external users of academic libraries, public libraries must surely provide the same sort of facilities of access to electronic resources – which means that they should combine to form consortia and seek site licences.
There is no such thing as a typical user; users are people, all different, and therefore with different preferences; libraries should cater for these. One prospect offered by technology is the individualisation of services on a scale hitherto inconceivable – the provision of tailor-made information to the individual user. This applies to all kinds of research library, but especially to the academic library, which has a more coherent and consistent body of users whose needs can be understood and served over a sustained period; most national libraries have a much larger and vaguer clientele, and many people will use them only once or twice. It will clearly never be possible to serve everyone’s specific needs in their preferred ways. In this sense, tailor-made services are impossible. What can be done is to provide facilities that are easily usable and that can be adapted by everyone according to their own preferences.
This may be compared to providing food; a large store is set out with a huge range of foods, from which everyone can make their own choice and which they can mix in different combinations, cook and otherwise prepare in their own way. They need some basic knowledge of cooking, and they can if they wish find out how other people prepare food, by consulting recipe books. They may sometimes prefer not to cook for themselves but to go to a restaurant where they can order from a more or less generous menu.
Libraries have a more difficult task than restaurants. We will accept a menu that is not too varied so long as it has something we like; we do not go into restaurants, or for that matter food shops, demanding one specific type of rare vegetable; and we certainly do not demand a particular individual vegetable. We want of restaurants something we like, not something we must have. The same is true of a good deal of public library use, but not of any academic library. Libraries have to deal not with multiples of a few items but with vast numbers of individual ones – far more than bookshops have to deal with, since no-one expects bookshops to supply specific periodical articles or a wide range of older books (not even secondhand bookshops). I cannot think of a parallel in any other sphere. It makes librarianship more demanding than almost any other service.
Although no library can provide tailor-made information services for every individual, people can do much of their own tailoring by the well-established method of constructing a profile of their interests and matching it regularly against references on (or added to) databases. This is most common in the sciences, where terminology is much more precise and internationally standardised than in the humanities and (especially) in the social sciences. Although most users who have the time do most of this tailoring for themselves, help is sometimes sought, often wanted and still more often needed; much time and therefore money can be wasted by inefficient searching.
These represent the value-added services that libraries – and no other sort of organisation – can give. They would of course be of value to all users, but especially to independent researchers, who do not have substitutes for them in the form of ready-to-hand colleagues.
Similar principles of individualisation can be applied to physical features such as lighting. Some users like natural light, some artificial; some like silence, some like a fair amount of noise around them. The number of possible combinations of these and other features is large. In practice only a few can be provided, but, in larger libraries at least, some attempt should be made. To suggest that libraries should offer features ranging from those of a night club to those of a cemetery might be extreme, but the need for variety should not be ignored.
However good direct electronic access to information becomes in future, there will always be one element missing. Many people need to mix with other human beings: to stimulate the generation of new ideas and the rethinking of old ones, to illuminate shady areas, and to provide a forum where formal information can be tested, organised and assimilated into their internal information stores. While their information digestive system will, given time, process information, accepting and absorbing what is useful and rejecting the rest, the process is greatly speeded up and enhanced by interaction with colleagues. And the interaction has to be face-to-face: remote interaction is not nearly so effective. I profoundly believe that the library should serve as a social place, a place where interaction can not only occur but is actively encouraged.
Not only users have to be considered when a library is designed. The sort of accommodation provided for staff can make a great deal of difference to the well-being and cohesion of staff, the efficiency of the library’s operations, and the nature and quality of the service the library offers. This is too big a topic to go into here. I would only say from my experience that I strongly favour an open plan for areas restricted to staff, since this greatly aids communication and cohesion. However, open planned work spaces have to be designed with great care. Everyone needs some personal space. Desks should have dividers, and be generously spaced; there should be sound-dampening, and provision for confidential conversations when they are necessary. Some separate offices are unavoidable, but there should be very few that do not have windows facing the inside of the building, or at least doors with glass (which need not be clear glass). No-one should be able to hide; and since many services to users will involve the assistance of staff, many of them should be not only visible and accessible to users but should positively invite consultation. The basic principle should be ‚Down with barriers’ – barriers between staff, and barriers between staff and users.
The future academic library should aim to become a one-stop information shop: a source both of material and expertise. Decisions as to how particular pieces of information should be made available – by on-the-spot collections, or by access to remote resources, if a choice exists at all – should be determined by user and cost criteria. The relative balance between acquisitions and services (which will include access to remote resources) will almost certainly change, as may the allocation of money between staff and other expenditure heads.
The research library should be a social place, where researchers and others can meet one another across subject boundaries and gain mutual stimulus and inspiration – an intellectual recreation centre. Perhaps some of the visits lost to libraries because they are not necessary for access to information could be replaced by ‚social’ visits. For this to become possible, the use of space will need to change in order to provide facilities such as comfortable lounge areas with coffee – the sort of provision that many bookshops are now making.
Finally, the future research library must be a remote information supplier, able to serve users at a distance – any distance – as readily as those on the spot.
I have tried to identify ways information needs that demand to be met in the new information world, and possible roles of the library in meeting them. Very briefly, the latter are:
Another set of requirements is:
Most of these have implications for the design of the building, some obvious, some less so; it would far too long to spell them all out here. I recognise that some are hard to meet by adaptation of an existing building designed in another era, but I must also say that I have seen some very good attempts, alongside new buildings that seem to have been planned for a world that has disappeared.
If all these requirements are met, the library can indeed be a real hub of the university, rather than a dispensable appendage. They are however only my own opinion, which may not be that of other users. Why do we not ask users what sort of libraries they like to work in or use, and why? There are many studies of users and their behaviour and needs, but very few studies of their likes and dislikes where buildings are concerned. Actually, we could get quite a lot of useful information by noting down every irritation we ourselves experience in libraries other than our own. Most are trivial and soon forgotten (e.g. surfaces which are slippery when walked on with wet shoes); but the cumulation of trivial annoyances experienced by many observers may amount to one very large annoyance. In my consultancy work I have sometimes got a few staff from the library to pay a ‚critical visit’ to another library of the same type, and to receive a return visit; it is amazing how many faults staff find, ones they are blind to in their own libraries because when they have been there for a few weeks they no longer see them.
I conclude with a list of what users do not want in buildings:
I have no doubt that this list could be extended almost ad infinitum.