Wednesday, 15 August 2007
Misconceptions about library classification

I recently attended two presentations on 'taxonomies in the digital environment', presented by respected information management specialists.
Both presenters viewed classification as practised by librarians as a quaint anachronism unsuited to the digital world. The tone was patronising.
There appears to be a misconception that librarians practise a simple, strict taxonomy which doesn't allow for information to be described from multiple perspectives.
In a strict taxonomy an object can only occupy 1 position in that taxonomy. You only get one instance of a penguin in a scientific taxonomy of animals and that penguin is well and truly fixed within the hierarchy of that taxonomy.
An idea gaining currency at the moment is that librarians take this taxonomic perspective. The misconception is that, to a librarian, a book on penguins will always be, and only be, a book on penguins. Librarians are well aware that information, and particularly the use of information, is far more complex than that.
The myth seems to be based on the observation that libraries deal with physical objects and that as a physical object can only sit in one location on a library shelf, it must therefore only occupy one position in the hierarchy of descriptive terms.
This view is wrong, and an understanding of how library classification works will lead to an understanding of the relevance of library classification in the digital world.
Subject headings
Librarians have a structured approach to assigning subject headings. It is important to understand that a structured approach doesn't infer inflexibility.
Using structured thought processes librarians dismantle pieces of information into constituent facets, and then match these facets to subject headings. The subject heading below illustrates three facets.
Real estate business - Accounting - Problems, exercises, etc.
We have the subject matter (Real estate business); the activity associated with the main subject matter (Accounting); and the nature of the information (Problems, exercises, etc).
A disciplined process of thinking ensures that all of the important aspects of an information resource are exposed. As you can see, far from describing the information from a single facet or point of view, the subject heading is multifaceted.
It is this process of analytical thinking that brings quality to information managed by librarians.
Librarians also assign combinations of subject headings to describe information comprehensively, as shown in the sets of headings below:
Administrative agencies - United States - Accounting
Chief financial officers - United States
and
Education and state - Australia
Finance, Public - Australia - Accounting
Educational accountability - Australia
I have used the above examples to generate the following descriptors, or tags:
Real estate business
Accounting Problems, exercises, etc
Education and state
Finance, Public
Australia
Educational accountability
Administrative agencies
United States
Chief financial officers
This list translates quite well to the digital environment as tags for website searching as illustrated below:

These tags would also be useful in an index of topics, as the basis for RSS feeds, and in many other applications depending on the requirements of the particular digital environment.
Classification numbers
It is true that librarians assign a classification number to a physical object so that it can be found. This has proved to be an extremely useful practice. The number usually looks something like this: 363.7360994.
It is incorrect to state that a particular information resource is always assigned the same classification number in every library. The number can, and does vary from library to library depending on the perspective of the library.
For example, in one library a collection of photographs of formal gardens might go to the number for photography because the parent organisation has an interest in photography, but not formal gardens. Another library might assign a number for formal gardens because it is more useful from their perspective.
As with subject headings, the allocation of meaningful numbers depends on the analytical skill of the librarian and their understanding of the relevance of information in different contexts.
Classification numbers present different views of the same information. Translated to the digital environment they could be equated to the mindset of publishing from one source up through multiple presentation templates within a website information architecture.
Library model in a digital world
The library model of classification operates effectively in the digital world as demonstrated by the following:
- The digital resource equates to the physical resource in that it is a discrete entity which is (ideally) stored in 1 location
- Multiple publishing from a single source corresponds to multiple classification numbers
- Tags equate to subject headings
Here is an example of how this might translate to an intranet.
The content of the digital resource:
- LegalConsiderationsWhenExportingDangerousGoodsFromAustralia.html
- Australian intranet view
- Export channel
- Legal channel
- Goods channel
- Australian exports
- Dangerous goods
- Flammable goods
- Explosives
- Radioactive materials
- Infectious materials
- Export of dangerous goods Act 2007
- Hazchem Code
Conclusion
It is wrong to assert that library classification is unsuited to the digital environment on the basis that it is simple and one dimensional. Quite the opposite is true. Library classification provides a rich source of tagging, and multiple perspectives, which are ideally suited to the organisation, manipulation and retrieval of information in digital environments.
Librarians have developed skills which enable them to effectively and systematically analyze information and these skills should be more, not less, frequently utilized in the digital environment.
Labels: cataloguing, information organisation
Comments:
<< Home
I've always found it difficult to understand why people involved in fields closely related to librarianship are unable to recognise the value of the skills found in the library environment. Because these skills are not recognised, they are subsequently underutilised. These are very relevant skills for the digital environment. The work of many "information managers" are poorer without them. It is obvious that Maxine has been able to use her skills as a Librarian very successfully in guiding others to manage information in the digital environment.
A good example of the problem with taxonomy as information classification.
All too often we see experts present their views of how information should be classified without understanding the nuances of their audiences mind-set.
The analogy of tags represented as the title can also be seen in disciplines like records management, but the same limitation still applies -- it reflects one world-view of information, rather than allowing different aspects of the individual to show through -- something that folksonomies do well that taxonomies do not.
M
All too often we see experts present their views of how information should be classified without understanding the nuances of their audiences mind-set.
The analogy of tags represented as the title can also be seen in disciplines like records management, but the same limitation still applies -- it reflects one world-view of information, rather than allowing different aspects of the individual to show through -- something that folksonomies do well that taxonomies do not.
M
Fair comment magia3e - but I think there is a place for both taxonomies and folksonomies. Yes, folksonomies allow different aspects to be displayed, but I think they work best for "informal" information types - and that's why they work so well in the Web 2.0 sphere. Taxonomies are ideal for "formal" information types that by their very nature require more control - such as corporate documents, etc. Both approaches allow multiple points of access which in the digital realm is so important. It's not an either/or choice - we need both.
Post a Comment
<< Home
Subscribe to Posts [Atom]

