Montag, 14. Februar 2011

Some thoughts on databases & library cataloging

For some time I've been involved with the Internet Book List, which is basically an online book database which keeps track of various types of cataloging information (synopsis, contents, translations, etc). What has always particularly interested me is the potential offered by such a database for finding information which is not -- or has not been until recently -- easy to find anywhere else, at least not in any kind of systematic form.

In 2007 we implemented what we've been calling WEM (Work - Expression - Manifestation), a system based on a fairly new idea in library cataloging called Functional Requirements for Bibliographic Records (FRBR). This allows us to distinguish between different versions -- called Expressions -- of a book and to track relationships between works (i.e., adaptations of a well-known story, inclusion of a shorter work in an anthology). At the time there were scarcely any other sites trying to make use of this concept, and none which attempted (as we do) to inform the user how various expressions of a book actually differed from one another.

This is huge. Unfortunately, for my needs, the database is still too limited. At the moment we only include fiction which is available in English, and that excludes the greater part of my library. What I want -- as a scholar and someone who reads extensively in multiple languages -- is the ability to go to the page for a work and be able to see not only the original title and publication date, but whether it has been translated, and into what languages and by whom. I want to be able to find out whether a particular ISBN refers to an English translation of Sophocles' Oedipus Rex or a version in the original Greek, and whether or not it includes a commentary and in what language. I want to be able to find out -- from the work page -- whether there's critical literature on a particular book.

At present, IBList can't do any of this, and I don't see it happening in the near future, because the website has essentially been stalled for the last 2 years due to lack of time and resources. I've put far to much work into the project to simply give up on it, as several of the other administrators have already done.

But I'm frustrated. The site is not going anywhere, and I don't see any way to change that. There are a number of reasons for it. Inadequate communication among those in charge. Fairly rigid control of the data entry process, which restricts the number of people willing or able to contribute. Competition from other sites based on a similar concept.

In September I finally gave in and started cataloging my books at LibraryThing. I had mixed feelings about doing so, as I've been watching the site since it started with a mixture of admiration and envy. Because in many ways it's what I wanted IBList to be. What it could have become had things been otherwise. Had I known more, had I been better at making the project work.

LibraryThing has a "bottom-up" approach, while IBList has always been top-down. LibraryThing focuses on the manifestation level -- individual copies of books owned by its users, and I think that's one reason why it's been so successful. Copies of the same work are then grouped together ("combined") into a single entry. Some users may enter bad data, but the general principle is that enough people are entering good information that it all pools together into a very large quantity of (mostly accurate) collective data.

IBList, on the other hand, starts at the top (the work level). Our focus has always been on quality rather than quantity, and perhaps that's part of the problem. There's a certain amount of elitism inherent in our approach. Not everybody knows or cares enough to comprehend all the intricacies of WEM. But there are enough other sites out there which consist simply of lists of titles, with little additional information about any except the most recent and popular books. I don't want to turn the site into that, even if it means that our listings are less complete.

At LibraryThing I can at least keep track of my entire library. And it has enough users to support a recommendations system which is quite good, something that is a long way from happening at IBList. But I find the site chaotic in some ways, the methods for organizing data are limited by the site's user- and library-centered approach. There isn't always a direct way to combine data and often the final result is messy even after you've done so. It doesn't satisfy my desire for order and simplicity.

Just recently LibraryThing has also started creating a system for work tracking much like IBList. This is logical, and I think those involved with the site have seen the need for quite some time. I should be excited -- finally, a site that has all the features I've been wanting!'s not "my" site. Not the project which I put so much time and energy into.

It hurts. I read the discussions and I want to say "yes, but we did this first, we already sorted through these problems, this is how we solved them." I want to say -- "look, see what we did?" so that someone might appreciate our efforts. But nobody's heard of us. Nobody's pointing to us as an example. It doesn't bother me that LibraryThing is doing this. What hurts is that IBList did this before and went unnoticed. It didn't inspire anyone. The site had -- has -- so much potential that we never managed to make a reality.


Sierra hat gesagt…

This is a strange blog. I have never seen such a good quality on the internet, but there are no coments... I will remember to come back and read some of it, especially the entries on tragedy.

Brenda hat gesagt…

Hi Sierra. I'm glad you find my blog interesting! Comments are of course welcome and encouraged. Neither the eclectic content nor my occasional navel-gazing are intended to prevent visitors from sharing their thoughts -- quite the contrary, in fact.

Kommentar veröffentlichen