Archive for the ‘tagging’ Tag

And another thing, or, Attack of the riled cataloguer

Still on Thing 8, tagging.

I found the Thing 8 blog post hard to write as there was a lot I wanted to say and could have said. After a few days reading other blogs and people’s thoughts, I still can’t quite leave it alone.

This post started life as a comment on the lovely Cardies and Tweed’s post about tagging but ended up about much more than that, so I’ve turned it into a blog post. I actually agree with Cardies’ overall conclusions, that tagging has a place alongside traditional access points and that Thing 8 raises more questions than it answers.

I just wanted to raise a couple of points in an earlier paragraph as I think it demonstrates a problem with the Clay Shirky article and is what I tried to talk about in my post on tagging. My primary issue with his whole argument is that he conflates classification with subject cataloguing/the use of subject access points. Tagging is a user-led equivalent of the latter, while the former is a very different thing. Promoting the benefits of tagging by analysing the failings of classification schemes is attacking a straw man, in my opinion.

As a demonstration, I want to just look at some of the points raised by Cardies & Tweed:

“…we concluded that while the Library of Congress system is very good, it can often fail when you have a new subject area that just doesn’t fit into any existing category. Plus, some problems do occur when specialist subjects come up and one cataloguer would feel it necessary to place a book in one category, when another would choose differently. One example that springs to mind, after another librarian chat, is the complex world of mathematics. The librarian in question found that a lot of the existing mathematics books had been put into sub-categories that were not accurate. Of course the issue that arises here is that this librarian specialises in maths, while many other don’t, so how do we get around such a dilemma?”

I think that it’s true, new subject areas/concepts come along that are not already reflected in either LC classification or subject headings. This will always be the case and tagging does have a role to play in covering this kind of new development because tagging offers an immediacy (no decision by committee) that formal taxonomies or thesauri lack. I just wanted to state that we also have the ability to propose new LC subject headings, LCSH is not a static thing and has more flexibility than the classification scheme. Proposing a new subject heading has no implication for shelf space and has no physical limitation. New headings are added and corrections/clarifications are made on a weekly basis (see here for the most recent weekly list).

The problem of one cataloguer placing a book in a different place to another cataloguer – that is definitely possible. In fact, catch me on a different day and this one cataloguer could easily place the same book in a different place. However, to my mind, this is only a problem for classification – as Shirky says, a book can only physically be in one place in the library, so sits only in one subject area even if it might just as happily sit in a different subject area (in some cases, there is a history of assigning “alternative classmarks”, but it’s not something widely used in the UK). Whereas the whole thing about subject headings is that there can be multiple subject headings for the same book or item, you don’t have to decide it’s only about one topic. There’s no 1:1 ratio here. So multiple aspects can be reflected in the cataloguer’s choice of multiple subject headings. In fact, most subject cataloguers would specifically want to do that.

As for the maths books that are in incorrect categories, if that means their classification number then this could be reported and might get fixed but might be due to the limitations of the UL’s classification scheme (and the overlap between the collection development policy of the UL and the Moore Library). If the maths books in question were given incorrect/inappropriate subject headings or are missing the most relevant subject heading, then it absolutely should be reported so that the cataloguers can fix the records (adding extra subject headings or removing inappropriate ones as required).

There’s no doubt  that those of us classifying are not subject specialists in all areas collected by the library. How do we deal with this? Well, I have been known to call on someone with specialist knowledge when totally stumped. We are also open to the notion of crowdsourcing (ie. see what other libraries – including specialist libraries – have done). Crowdsourcing also extends to finding out what library users think. This is something that doesn’t happen often enough, but if there’s anyone (especially another librarian, but actually anyone) with specialist knowledge who feels something has been wrongly categorised in some way, then they are most welcome to report it. There’s an online comments form if the cataloguers are too scary to approach in person 🙂 We won’t mind. Cataloguers always follow Library Tip #75, as explained by the marvellous Unshelved comic strip.

Okay, this post has been a bit of a rant ramble, but what I am trying to say here is that classification is not the same thing as subject headings. It’s the flaw underlying Shirky’s whole argument. And yes, normally classification means that a book will only be assigned one classmark. However, classmarks don’t always provide subject-related access anyway. More than 50% of the stock in the UL, for example, is on closed access. A reader browsing the shelves at a particular classmark is only seeing a small section of the UL’s holdings in a particular subject area, however accurately the 3-fig classfication scheme is applied. They are not seeing closed access material, non-book material, all articles within periodicals, digital resources. Even for the bookstock, a large proportion of the closed access material has a classmark that is purely a location marker and has no subject meaning. However, all of those items (books and other types of material) have subject headings which are designed to allow users searching the catalogue to find all the items potentially of interest to them. User-generated tags would just be an extra way of searching.

Information overload, See Personal information management

Personal information management LCSH
I am starting to struggle a bit with information overload (“physician heal thyself” I hear you cry, or rather, “librarian manage thine own surfeit of information”). I have this drive to be completist, read every post in every Cam23 blog, every tweet and every link from a tweet or post. I’m just about managing this but it means I spend a lot of time thinking “ooh I read something about that… somewhere…” without being able to recall exactly where. I need to tag my thoughts.

Thing 8 is tagging. I need to declare an interest. I’m a cataloguer. My very first job as a graduate trainee was 50% cataloguing and that percentage has only gone up in my library career since then. Actually, I’ve found this Thing hardest to write about because it is so closely related to what I spend the majority of my time thinking about/doing.

The Clay Shirky article was interesting. I’ve read some of his stuff before (though obviously can’t actually recall very much about it due to lack of thought tags). I do think he’s setting up a false dichotomy though. He speaks about classification being related to the physical need for books to be located somewhere and there being restrictions on space. Yes. But classification isn’t the same as LCSH or the same as controlled vocabulary more generally. They do different things. The way in which he talks about what a “professional cataloguer” would think or feel is monumentally irritating (and inaccurate).

The simple answer is that tagging is great. It works brilliantly in Flickr or blog posts and, in a more complex form, what Shirky is describing is happening in Amazon’s “people who bought this, also bought…” recommendations and in Google searching. And we all know how successful these discovery tools have been. There is real power in crowdsourcing (if you have a big enough crowd). The Ann Arbor example shows how tags could work in a catalogue. But it’s not an either/or and I don’t know many cataloguers who think it is.

%d bloggers like this: