One of the most difficult tasks in an organisation is to identify the information that it needs, whether to solve a problem, investigate the current situation, initiate a process or educate staff. An understanding of the context of the problem and what information is available both inside and outside the organisation is required. While external information resources are often fairly well codified and known, the information resources within an organisation are often not well known.
Of importance to Knowledge Management is the idea of Expertise Databases. These are listings which contain an employee directory with staff biographies and photographs, areas of expertise, client experience and a client database (past and present clients), and the work the firm has done for them. A more sophisticated expertise database might be a more dynamic product which allows the development of communities of practice between remote members of an organisation involving communication of ideas, knowledge and practice. Some organisations that operate on a global level use a number of these kinds of tools to share knowledge and ideas and to make sure that the most appropriate information is available to all employees.
back to the top
There are a number of ways of acquiring or capturing information in an organisation. Internal information is created either as a by-product of the business processes or is more directly created through the accumulation of material about customers, orders and sales. Information is also acquired by buying in selected material e.g. research material on products, markets, competitors, innovations, new products and industries.
The processes involved for acquiring information this way include scanning the environment and using both automated and human centred approaches. Scanning can produce filtered information for users either as structured information from recognised sources, e.g. published works – books, web pages, article, conference papers, market reports, statistical surveys, or from recognised authorities either individuals or institutions. Information is also generated as a result of the informal interactions between employees and groups within the organisation.
back to the top
Once the data or information is acquired or created, it needs to be organised so that users can have access to it. In some organisations, libraries exist for this purpose. More often these days, as well as libraries and information centres, organisations have knowledge repositories which can be distributed systems rather than static places. The process of organising information involves cataloguing – describing the information using a set of descriptors that are listed and known to all the users.
This information includes:
* who wrote / created the information,
* publisher or copyright owner,
* when it was published / created,
* the title / name / listing.
Classifying the information involves grouping information into like categories so that all the information about a certain subject can be accessed at the same time. Assigning subject headings to the groups allows for easer access (see material on Taxonomies, Thesauri).
The procedures used in a library for classifying and cataloging information so that people can find it and use it are a useful metaphor for the information and data stored and organised in an organisation. Consider the access to electronic information environments that are the norm for most organisations.
A search engine or a browser could then use all this metadata to refine the display of search results. Some search engines do this. For instance AltaVista uses the data contained in the description element to display in the results list instead of the first few lines of text on the webpage. This often makes much more sense to the person trying to read the list of results, especially when the first few lines of text from a page are just a list of headings and not really meant to be read as a sentence.
Most large search engines such as Google and Alta Vista do not use the DC metadata element set as a way of searching for web pages as there are still not many web pages on the web that actually have metadata built into them.
This happens for two reasons. First, people who build web pages often have technical skills but not information management skills and therefore do not know about metadata. So the search engine companies are not going to invest a lot of money into making their search engines search for metadata until more people put it into their pages. Secondly, people think it is not worth the effort of putting metadata into their pages if the search engines do not go looking for it. This is a classic chicken and egg problem. It is unlikely to be solved while the web as a whole is the free and easy place that it currently is.
The problem becomes much less difficult when it is constrained to a single organisation. It is much easier for a particular organisation to say to all the people who are going to contribute pages to the organisational website that they have to use metadata. Then a search engine that is limited to searching that particular site can be configured to look for a certain schema of metadata elements in every page. This allows the website search engine to act much more like a library catalogue.
The same sort of decision can be made by a particular sector or industry. This has happened in Australia with regard to Australian government electronic publications. Every website put up by every government department in Australia has to use a metadata schema called AGLS.
AGLS (Australian Government Locator Service) is a metadata schema based on the Dublin Core schema but with four extra elements. The four extra elements
he problem of language – classification, taxonomies and thesauri
One of the most significant issues for information and knowledge managers to deal with is the problem of language. Words are complex things. People use words in all sorts of different ways. Sometimes people use the same words to mean different things and sometimes people use different words to mean the same thing.
If you look up the word ‘stress’ on a search engine you may find items relating to:
* The stress placed on a beam by forces acting on it (engineering);
The stress of living in a fast-paced chaotic environment (psychology or biology);
* The stress placed on a word when uttered with emphasis (linguistics).
It would be nice if you could be sure which of these you were looking for and which you were getting back in your search results. Different sections of organisations sometimes use language differently. For example, consider - what is 'a customer'? A large retail business may have slightly different definitions of 'a customer':
* On the shop floor a customer may be anyone who walks in the door, whether they actually buy anything or not;
* For the accounting department a customer may be someone who actually pays money for a product;
* For the marketing department a customer may be someone who has an account with the business and therefore whose address is known.
This means that when the different sections of the business report on customers then they may not be reporting on exactly the same thing. If the information manager in the organisation is not keeping track of these differences and allowing for their use in the various reporting activities, then the business could be led into making incorrect decisions.
This illustrates how the definition of one word can cause a problem. In fact it is often these apparently small differences that can cause the biggest problems. When people use quite different words, or have quite different meanings for the same words, then it is usually obvious to everyone involved in the communication that something does not quite match. In that case usually someone will ask for clarification of what is being said and the problem can be sorted out. But when the differences are very small, as they are in the above retail business example, then it is easy for people to overlook the differences and assume that they are talking about the same things when they are not. In this case the problem does not get sorted out and mistakes occur.
For nearly all words the same sorts of problems apply. This ambiguity is one of the aspects that make language interesting and fun. Many jokes would not be possible if such ambiguity did not exist. But for information management purposes it is a problem. The problem arises because when we are searching for information we are really looking for something based on a concept or idea that we have in mind. But the only way we have of expressing that concept or idea is with words. Yet words and concepts are not quite the same thing.
One of the tools that information and knowledge managers use to try and control this ambiguity of language is a taxonomy or thesaurus.
Taxonomy, or thesaurus
A taxonomy, or thesaurus, is a list of approved words and their meanings and relationships. For instance, a taxonomy may define exactly what the word ‘customer’ means in the retail business above. It may break the term customer down into specific types of customers and give rules as to when those more specific types are to be used.
A knowledge manager will often be required to develop a taxonomy or thesaurus of terms used by an organisation so that everyone in the organisation can use the agreed meanings. This is a way around the problem of slightly different meanings being used by different units of the organisation.
Taxonomies are useful in many different situations. On websites they are useful to ensure that the words used for links and other navigation cues match the expectations of the users. In databases they can be useful to ensure that data entry is consistent when different people are entering data into the same field at different sites or times.
In metadata they are almost essential. If you consider the DC subject metadata elements mentioned in the section above there is an obvious problem. Although the DCMI says that the subject of the webpage should be put in to this element it does not tell us what words should be used to express that subject. There are too many possibilities for the DCMI to cover so they leave it up to each person or organisation filling in the metadata to devise a set of rules that are appropriate for their own situation. Thesauri and taxonomies are those rules. They are often used in constrained situations in order to help people know what words they should be using in this metadata element.
For instance, if you are writing a webpage about a type of wolf known as the timber wolf should you put the subject in the metadata as ‘wolf’, ‘wolves’, ‘timber wolf’, ‘timber wolves’, ‘timber-wolf’, ‘timber-wolves’, or ‘lupus’? (Lupus is the scientific name for wolves). There is no right answer to this problem for all situations. But if you work for a zoo, and the zoo has a thesaurus that you can use to check which you should use, then it will be much easier to enter the metadata. It will also be much easier for someone else to come along later and find the webpage if they have access to the thesaurus too.
It would be even better of course if the thesaurus was built into the search engine on the website and, no matter which of the terms the person searched under, the search engine always pointed them to the right one and to the webpage. Taxonomies and thesauri are sometimes built into search engines in just this manner specifically to allow for searchers to use whatever words they happen to think of to search under. The search engine will still take them to the correct term used in the taxonomy or thesaurus.
Developing a taxonomy or thesaurus takes a lot of work. There are two major reasons why this is the case. First, the work of finding out all the words used and their exact meanings and the relationships between them is a slow and intellectually challenging business. Secondly, there is the politically charged work of deciding whose meaning will be used and whose will not be used. How will the organisation be affected if some sections of the organisation are not able to continue to use the language as they have been? Being too prescriptive can be counterproductive if sections of the organisation will refuse to use the taxonomy at all. Being too laissez-faire can be counterproductive in that even if the people do use it there is little benefit gained.
In most organisational environments the storage of information and data is now done electronically, mainly in large database systems. The manner of the storage will depend on the organisation and how well they manage and perceive their information assets. back to the top
Disseminating and using information
Connecting users with the information they need is one of the most crucial processes in any organisation. Creating a culture of information sharing is part of this connecting and an integral part of creating an intelligent organisation. The information strategies employed by an organisation are usually a mixture of human and computer generated. Computer generated information appears to offer more timely delivery of information and sophisticated delivery strategies can tailor information rapidly. The development of Knowledge Management and Content Management Systems stems from this need to disseminated information in a timely and appropriate manner. The development of large Web-based information management systems in organisations has led to the proliferation of Content Management Systems in order to control the large amount of information an organisation uses to function productively.
Underlying the use and dissemination is an evaluation of the end product and the processes. Questions include:
* What information did I use and was it appropriate?
* What else do I need?
* How do I share the results of this work?
* How was information accessed?
If there is an information sharing culture in place, then it will be easy to store and share the information and knowledge generated from any particular project.
The process approach to information management sets an information rich culture with appropriate mechanisms for information sharing and the building of knowledge in an organisation. Development of these processes in a particular environment using a vast array of information technology precuts specifically developed for many different contexts.
While these IT products offer many services and options, some idea of what the business uses and needs for information resources, access, dissemination and specific knowledge is important as a first step.
The Information Auditing procedures described in the next Topic will help in the development of these elements.