The role of IA in classifying ephemera

I was reading a thread on Drop about classifying/categorizing chat. In the comments, the Shifted Librarian, pointed to John Robb's discussion, The I in K-Logs, in which he talks about the concept of blogging and classifying such things as instant messages, email, etc. for knowledge management purposes. This thread raised some new questions in my mind about the role Information Architecture could play in helping afford access to ephemera within the enterprise.

Robb's discussion got me interested in the concept of classifying ephemera such as chat sessions and wondering where this fits in with Knowledge Management. I read the emails on the K-Log list and see that a lot of people are doing the weblog thing inside the enterprise to narrow-cast their stuff to their department/organization or whatever. In a sense, that is what I do with iaslash as well. The bulk of what I point to is ephemera -- weblog discussion about some topic that someone has posted somewhere, timely newspaper articles about some event or critique of some site. But what becomes of this recorded knowledge if it simply goes into a database and is not organized. Do we trust that search engines will help us in discovering knowledge held in these repositories? Does IA-type organization support learning or facilitate better discovery?

Ephemeral items are a special kind of entity for description. They're sometimes too difficult to catalog (the original creator or publication information may not be known, for example). Often people don't want to expend the effort on cataloguing or classifying them because of their ephemeral nature. But I think there is great value in categorizing these types of material, especially when dealing with documents available via the web.

I learned a good lesson from an experience I once had categorizing ephemera as a grad student. I was involved with a project at the Metropolitan Museum of Art's research library where I had to go through ephemeral printed matter -- newspaper and magazine clippings, photographs, exhibition announcements -- and then categorize them using a system of flat files that organized that ephemera based on access points that could be empirically determined such as artist name, organization name, etc. This seemed to me a rather superficial way to categorize these things. I thought a better method would be to follow a richer model like the The Getty Information Institute's Categories for the Description of Works of Art. The concept for me was that, although the intellectual work of identifying some rich description for ephemera may be difficult, archiving those ephemeral materials is of no consequence if it cannot be retrieved later. This is the library and information science/information retrieval part of me coming to the surface. What I was always interested in, but never privy to, was how frequently this information was used and with what success. It turned out that the collection could only be navigated by someone very familiar with its contents -- an expert.

This brings me back to the role of IA. To me everything is ephemeral and everything could do with some rich description -- at the least using somthing like Dublin Core fields, and hopefully some subject classification scheme. Sure, a good search engine might help you find what you're looking for ... if you know what it is you're looking for. But learning/exploring is best supported, in my opinion, by subject classfication. Maybe auto-classifying applications do some good using applications like Semio and Verity Intelligent Classifier. I don't know. But needless to say, the effectiveness of those tools is measured by the effectiveness of the person administering the searches matched to the taxonomies they've built. For all intents and purposes that person is an IA.

This concept of being able to categorize/classify everything/anything has been on my mind again lately because of the IA-org Library, and because I want to eventually turn this blog into a bibliographic tool like ProCite rather than continue to use it just to blog. There is value, I believe, in the expertise we can bring to information retrieval and discovery applications (web sites, databases, etc.) within the enterprise and on customer sites. Lately I feel like classification is the hammer that I want take to all of these problems, ephemeral or long-lasting.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Related discussion

A (somewhat) related discussion can be found on I've been trying to develop a personal schema/taxonomy

Kika --

Thanks, Kika

I saw that thread on kottke, but didn't get to the comments until I saw your post. A lot of discussion and interesting links floating around there. I find some of the suggestions to be pretty humorous.

Using Ranganathan's facets of personality, matter, etc. to categorize your stuff? Interesting. But useful for a personal organization scheme? I don't know about that.