The Quality of Information in the Web

All communities, all societies, all collaborations arise through and depend on interaction and communication among their members. There must have been concerns about the honesty, accuracy, completeness, currency, and verifiability of communications in prehistoric times. The difference now is in both the greatly increased quantity of messages and records and also our greater dependence on them.

The increase in documents

We can consider four distinct vectors of technical advances: writing, printing, telecommunications, and document copying. Writing, by recording, provides an alternative to speech by making it permanent or making it unnecessary. Writing, then, diminishes the effect of time and so provide an alternative to human memory, an ‘external memory’. Within its limits, writing, when serving like speech or gesture, has the enormous advantage of being able to counteract the effects of time and, by being portable, of distance.

Printing provides an extreme multiplication of writing, extending the effects of writing and making it available in many places simultaneously. This adds security, because there is safety in numbers when each individual record is vulnerable to destruction or falsification.

Until the nineteenth century, telecommunication was somebody traveling on foot, horse, or ship bringing news. The rise of semaphores, railways, telegraph, telephone, radio, facsimile, and now the internet has had the effect of reducing the effects of distance and so diminishing delays in communication. Much has been written on the impact of writing, printing, and telecommunications, but far less on copying. The three really important techniques were photostat (photography directly on to sensitized paper), microfilm and its variants, and electrostatic copying (also known as xerography).

The primary effect of these four vectors of technology, each amplifying the others, is to reduce the effects of both time and space. Records are increasingly accessible in any place and at any time. It is very hard now to imagine life without writing, printing, telecommunications, and copying. But this explosion of records has had two important consequences. First, to cope with this explosion a fifth vector of technical development became needed for discovering and selecting relevant documents as and when needed, variously referred to as bibliography, documentation, and information science. Second, old problems of quality and trust are made more difficulty because mediating technologies remove us from direct contact with the (greatly increased) sources.

The Division of Labor and the Need to Know

Cultures have developed from hunting and gathering to agriculture, industry, and sophisticated services. This change has been achieved through the division of labor and increased interdependence. Now we very rarely grow our own crops, raise the animals we eat, milk cows ourselves, or grow our own coffee beans. The food we eat is grown and processed by others before it reaches us. Similarly, we do not, ourselves, make the technology we use, construct the buildings we live in, or generate the energy resources we depend on. Very few of us now know how to.

This division of labor allows us a far higher standard of living overall through the development of specialized skills and greater efficiency through economies of scale, but we have become much more dependent on others in many ways. We rely increasingly on other people, on technology, and on the infrastructure of transportation, financial services, regulations, and other developments that make this interdependence possible. Others, in turn, depend on us.

The exchange of goods and services requires markets and markets depend on knowing what choices are available and on what terms. Markets are information systems. The better-informed the buyers and the sellers are, the more ‘perfect’ the market is said to be. Less often noted is that this interdependence also requires a great increase in communication and documentation. If we are to use public transportation, we need to know who provides the service, the routes, the fares, and the timetable. If we are to obey the law, we should know what the law is and the probable consequences of ignoring it. If we are to buy goods and services instead of providing our own, then we need to know who provides what we want, how much we shall need to pay, and whether what is offered is what we think it is. We can ask, of course, but mostly we depend (and anybody we ask will depend) on documents: price lists, contents, warranties, advertisements, and so on. Since we cannot ascertain entirely by ourselves everything we would like to know, we have no choice but to depend on what others tell us and so we must also decide whom and what to trust. Patrick Wilson’s Second-hand Knowledge: An Inquiry into cognitive authority (1983) provides an excellent discussion.

Scholarship evolves through interaction with the ideas of others, but generally we cannot collaborate within the same office for many practical reasons. The person whose ideas interest us may distant in space or time – and may have died. So normally the best we can do is to depend on their documents. What has that person said, done, or written? Their documents, both by them and about them, incorporate their work and their ideas, much as technology incorporates the labor of past inventors.

Documents have become the glue that enables societies to cohere. Documents have increasingly become the means for monitoring, influencing, and negotiating relationships with others. We live in a “document society” and we have to depend on our assessments of the quality and trustworthiness of the documents we find on the Web or elsewhere.

Agendas of others

It is not only own agenda that should concern us, but also the agendas of others. We have the use by governments of passports to indicate identity and to confirm citizenship in order to control movement across borders. Other examples are easy to find. Schools use textbooks to guide our learning; religions use sacred texts to inspire beliefs; artists produce images to please us and to challenge us; merchants invest heavily in advertisements to influence what we buy; politicians make statements to seek votes and campaign donations; entertainers use varied media to amuse us and to generate income from us; individuals use messages (letters, e-mail, etc.) to communicate and social media to attract attention; museums use the selective presentation and interpretation of artifacts to explain the past; mass media constantly transmit programs to entertain, to influence us, and to satisfy advertisers; and libraries provide access to collections of documents to facilitate what we learn. This list could be extended indefinitely. Anyone could easily make such a list. As the list builds up we see more and more of our lives included and we need to remember in every case that all these agencies are strongly motivated to influence us according their own objectives. Sometimes what they want is deceptive or harmful. This list characterizes the ‘information society’. The examples enumerated are less important than the cumulative evidence that our lives are permeated by messages, records, and documents.

An agenda for library science and documentation

Research in library science and documentation has focused on access to documents. We have evolved techniques and procedures for collecting, describing, arranging, indexing, and facilitating the search for relevant documents. This emphasis is practical, useful, and necessary. Much less attention has been paid to the quality of the information. In the print environment, when we add books to our collections, add entries to our bibliographies, or simply read for our own benefit, we try to select items that are at an appropriate intellectual level and we tend to depend on the reputations of publishers, authors, and reviewers. But the clues that we depend on are much less available on the Web, where anyone can make a good-looking webpage.

The phrase “information system” can be interpreted in two ways: As systems that supply information (meaning data and documents either electronic or on paper); or as systems that inform. The difference is that systems that inform need to meet additional requirements which are concerned with cognitive aspects of becoming informed. A reader may successfully find a document, but will not be informed is he does not understand it. The document must be in a suitable language, at a suitable level, and be coherently expressed. Here “suitable” depends on what the reader already knows and his cognitive skills. Also, a reader needs to decide how far to trust the document. Even if I understand what a writer has written, I need to consider how far to trust it. This, too, is situational. If I am reading for pleasure, fantasy is acceptable. Sometimes reliability is not important to me. (I do care how many moons there are around Mars, for example). But if I am depending on the information when making a decision that is important for my health, wealth, or happiness, then trust becomes very important.

"Information quality" is not easily defined but includes the honesty, accuracy, completeness, currency, and verifiability of what we read. This is generally true and especially so on the Web where the bases for assessing quality are weaker.