Sunday, July 23, 2006

A long term solution for how the library collection is organised?

Just read Connecting Librarian's post on "Dewey and its future in Public Libraries" and was compelled to leave a comment:
Perhaps the day will come when it doesn't matter how we label the books. Like, our customers would pick up a pair of hi-tech lens and switch on to whatever Display Mode they choose -- by DDC, by Subjects, by Themes.
Let me elaborate on the above:
  • Imagine a wearable device, where the user would input some search terms. The user walks along the shelves and would pick out items by sight.
  • The device acts as a filter, graying-out the less relevant items while the most relevant ones would appear brighter.
  • Extend this feature one step up -- in addition to a wearable device, the user could see on a screen, in plan-view (i.e. top down), the various items marked out on the entire floor of the collection area.
Discussions on how the library should organise its collections isn't new. Typically, public libraries use the DDC with the Cutter system combined (for fictional works) while academic libraries adopt the Library of Congress Classification System , or LCC (see also LCC outline).

To non-librarians, you might wonder why this discussion about classification systems. Well, if you've used the libraries and ever muttered, "Why can't libraries classify items that makes sense to ME?" then perhaps that's one reason (because ultimately, libraries want to make their collection accessible to you).

In simple terms, the "problem" is that the way libraries organise its collection isn't the most intuitive for ALL customers. I feel the root cause to this "problem" is that we all perceive things differently and/or our preception shift according to whatever context at that particular time.

Example: I might see a title "Computers in Business Applications" as "Computers" while another time I might see it as "Business". I might see Dan Brown's Da Vinci Code as "Fiction" while I might also see it as "Controversial Religious Fact-ion". You can't say it's "right" or "wrong" because that's what makes sense to me.

In my view, classification systems were never meant to be well understood by library customers. The systems were meant for librarians to organise materials so that librarians can efficiently retrieve them on behalf of the customer. It was only after computers got into the act that customers are able to routinely help themselves search for items within the collection. However, how customers search is very much dictated by how the items are organised (i.e. the classification systems used).

With improvements in technology, there's a greater awareness, and perhaps urgency, for libraries to improve the way materials are organised within the library so that more customers can help themselves. It's not that librarian are lazy. It's about improving convenience to customers.

I think a long term solution should tackle the root cause. Rather than tackle how items are organised, why not tackle the fluid nature of how preceptions about how they are organised. Hence, the idea of the "Contextual Lens/ Device".

OK, with a device like that, does this spell the end for librarians? I don't think so, because I'd like to make two assumptions:
  1. On one hand, while we reduce the reliance on librarians to retrieve information, the rate at which information is made available far exceeds the time that customers have to search for information themselves.
  2. Information needs of individual customers will become more complex and customers will become more discerning. While there is google -- which most times is "good enough" -- there will be times they will want far more than what they can find off google on their own.
It's like this -- I can already buy investment products online, and I know where to seek information. At times I'd buy direct. However, there are times I feel it's faster more comprehensive to speak to a sales representative or advisor. For instance, I want to know what's the best returns on investment based on my personal income and consumption patterns (which would differ from others, and even for myself it differs from year to year as my needs change).

Anyway, if you're really really interested in the exciting world of library classification systems, then you might want to read up more on things like Enumerative, Hierarchical, and Faceted classification systems (see this Wikipedia entry for an overview)... ... OK, I'm just cheating here and plucked the terms off Wikipedia. Frankly, my knowledge of Library Classification Systems isn't deep.

Technorati Tags: ,


  1. Your proposal of a search device reminded me of Bruce Sterling's concept of spimes.
    Not sure if the library uses RFID, but I doubt it'd be difficult to implement.

  2. I like the idea. I immediately began to think of different versions, for grocery shopping etc. Then the next library client was happy to have an EBook online for his text..used the catalogue link for

  3. I was thinking about how the NLB organises its borrowing collections: a combination of DDC with some topics like Computers and IT, Cookbooks, taken aside outside the main sequence. I suppose it's useful because those are the categories where most books are borrowed, but it's a bit disorienting, nonetheless.

  4. Hi Brandon, NLB introduced this "10 main subject categorisation" maybe 10 years ago. It was in addition to the DDC (that's still being used). See, the point was to make browsing more intuitive. Rather than tell you "300s" & "658s" are the Business/ Mgt Collection, we just place all the related Business and Mgt books under that subject, which is subsequently sorted by DDC. As what I tried to point out in my blogpost, our perspective on what is a "correct" classification depends on context, so it's definitely not perfect with the current system.


Join the conversation. Leave a comment :)

Note: only a member of this blog may post a comment.