date: February 12, 2013
I think it's interesting (to me) that I think a lot about what I'm doing when I tag something in a social bookmarking site (part of the reason I think about it is because it's a research area in my field, so I encounter studies on tagging quite often). For example, I use CiteULike to manage my bibliographic references, and each time I add a reference to my library, I try to imagine that my future self is looking for this article in a few years, and then I try to discern what terms I should associate with the article now so that my future self can find it then. The future self is just a device. More generally, the terms I choose usually represent, to me, what the article is about but may also indicate other aspects, such as a research design (e.g., case study) or a methodology (e.g., statistics, historical, qualitative), that are used in the article. The approach seems to work for me. That being said, I've long felt like I've been tagging wrongly or that my tagging method was missing some essential step.
I think I've found the missing bit, and part of it has to do with when I tag a reference. For the most part, my tagging process has been after the fact: First I read the article (or just find something I want to save and read later), add it to my CiteULike library, fill in any incomplete bibliographic information, tag, and then save. However, the other day I was reading an article, and while reading, I kept coming across terms that I thought would function as great tags. So I stopped reading the article in order to save it to my library and add those terms right away. But since I wasn't finished reading, I kept my CiteULike page for the article open in one browser window, the article in another, and continued to tag as I read.
What happened then was that tagging started to function as a kind of associative note taking process instead of only as an information organization and retrieval tool. I started adding a lot of tags (13 for the last article I added with usually three to five for most articles I've added in the past) and used [unique] terms that I could not imagine ever using as future retrieval terms. Instead, I'm imagining these tags will function as reminders of key parts of the text when I'm browsing my collection of bibliographic references. Another way I've been thinking about it: the old tagging method I used had a collection centered focus. That is, I used tags to organize a reference within my collection (cf. the “flexible filing system” in Emamy and Cameron, 2007). But this new method has an item centered approach or a within text approach, because I'm using tags as notes.
It's too early to tell if this method will work, but I think the two approaches can complement each other. Or perhaps I'll just generate a very noisy tag cloud. In any case, as of now, it feels good to tag in this way. I felt like my engagement with the text was augmented. Boon.
Emamy, K., & Cameron, R. (2007). Citeulike: A researcher's social bookmarking service. Ariadne, 51.