There’s a new issue of Dlib out with an article called “Some Philosophical Problems with Folksonomy”. There has been some discussion in the library blog arena about it. Here is a round-up and some thoughts. The article goes over traditional cataloging and folksonomy and concludes:
and then also makes the statement:
I didn’t really see much evidence of “better results” in the article and I haven’t seen it in real life either. Some of the arguments that I’ve seen include that the article misses the social aspect and the “wisdom of crowds” as well as having an either-or mentality when they could augment each other. From dystmesis:
And from Catalogablog, though a commenter disagrees that the paper has an either-or position:
I agree with some of the arguments for strict classification and that there are problems with tagging en masse, though I agree with others that there is probably a critical mass where useless tags get washed out. I don’t agree, however, that using relative meaning is inheritently bad or invalid. In terms of discoverability and findability, all possible interpretations can be helpful. It can also provide context for works such as ones which had a political reason for being published even though the work itself was not about politics. I think the most useful benefit of tagging is the relationships between items that can be made apparent through the interpretations. I do agree, however, that there will also be the need for controlled vocabularies in organizations.
It might be my background in science that plays a roll. With the advent of genetics it became apparent that some classifications were probably not correct, though it’s sometimes difficult to change. And whether controlled or not, some decisions in classification can be arbitrary. And when you get down to some systems what something is or isn’t turns into a statistic and the idea of duality becomes easier to fathom.
In the end I think both systems do an ok job though I tend to find more with tags on sites such as LibraryThing, though this might be the display and not the classification at work. Having more than one “classifier” means that the results can be weighted depending on how many have tagged it such.
I guess I can’t end without an example. On Worldcat, 1984 has only two subjects which include London fiction and Totalitarianism. LibraryThing has some overlap but adds things such as dystopia, political fiction and social commentary. Sure there are some erroneous tags in there but the popular tags are quite useful, especially with creating the “similarly tagged” items which give decent recommendations of relatedness from a quick glance. Similar to Worldcat’s results for the totalitarianism subject though with likely a bit more weighting. In the end, both are useful.
I’d also like to point out that dystmesis also has a nice writeup on folksonomies and LibraryThing that’s worth a read.