There are a number of different ideas floating around in my head. I have to get them down on paper before they all fall out of my head...
1. Sense Making in Organizations: I've recently been reading Chun Wei Choo's "Knowing Organization." It's quite good. I do, however, have some issues with the text. In retrospect, I could extend these same arguments to Weick. Both Choo and Weick indicate the importance of the local in sense making. Neither, however, do a great job of describing or articulating what constitutes the local. I suppose that the local is a combination of factors (i.e., Knorr Cetina's "smear" or Pickering's "mangle"). Bowker's work on Schlumberger provides one possible way of pulling apart these concepts through the use of space and time. Importantly, space and time need to be locally constructed. This space/time concept is well developed in other realms such as Ranganthan's Colon Classification or Bakhtin's articulation of the chronotope. There might be some room here to do something... I'm not too sure what that something may be. Grant Campbell's work on the chronotope and NAICS may be informative.
2. The Landscape of Information: There is always tension between the words "information" and "knowledge". I've given up any attempt to articulate the difference perhaps because I've adopted Dervin's constructivist interpretation of information. Therefore, any definition of knowledge as socially constructed or locally contingent seems moot. Regardless, I can imaging a thesis defense question from an external like: "articulate for me, your understanding of the differences between information and knowledge?" Since I'm not a big fan of any of the other definitions (although Buckland's "information as knowledge" is quite interesting), I have to create my own. For a while I was playing around with some concept of information as a human-created "other" that shapes our actions (thoughts, again, being moot). This conceptualization bundled the recent work of Knorr Cetina on trading floors with classic Belkin and then threw in a few tablespoons of Lacan. Agency, however, became a problem. Does this "other" have agency? Do we give it agency? I'm no philosopher so perhaps I need a different definition.
If not a philosopher, I am a Civil Engineer. Another possible metaphor (see Lakoff) for information is inspired by hydrology. I'm imagining a landscape that has a particular shape. The landscape is effected by various process. Some are very slow geological process like volcanism and others are relatively fast i.e., hydraulic erosion. From an information perspective, the slow change relates to various epistemes, paradigms, or developmental stages. The fast processes, however, are related to information flow. Sometimes, the information flow is just too great and just runs off. Sometimes it produces creative flowering (imagine a desert) and it always involves some sort of change. There are great ways of torturing this metaphor--the hydrological cycle, turbidity counts in rivers, tortuosity of resulting river patters, etc. The question becomes: "What is knowledge?" It's the landscape left over that responds directly to any sort of incursion and has been shaped profoundly by past information flows... Or something like that.
3. Science vs. Technology: In studying science and technology and their various documentary practices, there seems to be a discrepancy is how we approach these two fields. Granted, there are a number of needs and uses studies for both fields but studies of science seems to have a real producerly bias i.e., scientists as producers of documents. Technologists are somehow treated differently in most cases (Bowker is an exception). Instead of studying the process of production--like with scientists--we study the conditions of production or focus on the inputs to production. The actual process of creating tangible artifacts seems to be black boxed. Perhaps we're comfortable as researchers to claim that science in general, and the corresponding scientific papers, are socially constructed. This argument is a bit tougher to make in the face of an actual physical artifact such as a missile guidance system (MacKenzie), a bridge (Suchman), a producing well (Bowker), a Portuguese galleon (Law), bicycles (Bijker), or an electrical grid (Hughes). How do we account for the actual construction of real artifacts?
4. Information on the Run: I can probably run with this idea for a while but I want to capture the kernel... Bowker provides a great description of how Schlumberger was able to produce stability in a local environment by controlling the space and time of the local context. They were chasing oil. While elusive, at least it's an actual product. It is possible to apply similar concepts to a resource such as corporate information? Can we engineer a means of constructing and stabilizing information within an organization? Just as Schlumberger were able to use the USSR and Venezuela as experimental grounds for perfecting their method and growing their organization, could I--for example--work within large organizations to establish exactly what "information" is? Could the company be built? Is there an underlying commodity (perhaps money)? I'm aware that I'm trying to conflate information with information i.e., information^n, but there may be something here.
5. Child Witness: In my current work on the child witness database, I'm aware of the struggles that we're having. Stabilizing the actual product is proving to be very difficult. It doesn't want to be black boxed. The problem is that people want to disseminate the product as a standard. I now realize, however, how locally contingent this particular tool is. There are a number of problems is getting the thing to stay put. The first is the variety of actors involved e.g., local agency, court system, police, etc. Each actor has different requirements and wants different traces from the actual device. The other difficulty is the chronotopic nature of the problem. A particular "incident" involving particular parties happens at a specific time and place. As the court system takes over, however, the time of the incident is subordinated to other classification structures such as the timing of particular court dates or instances of court preparation. Space is similarly difficult to construct given the variety of localities (e.g., court rooms, agencies, addresses, and locations) and jurisdictions (e.g., police, justice, community agency, etc.) involved. Tough problem. There may, however, be a paper in here somewhere.