Usage Models in Context

posted by Eleanor Wynn on December 20, 2006

There is a lot of discussion about usage models lately; and more than one view on how to detect, elicit, construct and create them. I have been working for about 30 years in this area, though it keeps changing names. I am still involved. I started my work doing “micro-ethnography”—technical term is “ethnomethodology”. That involves putting a microscope to what people say in natural conversation and pulling out the implications and assumptions that show how they “construct the world”—especially how they do it together because the work is not on monologues but dialogues. Then I worked my way up the ladder of assumptions about everyting social, including the organization. I would like to present my most recent thoughts on the organization as a context for usage models, or anything resembling a usage model.

Context of Use. Usage models for consumers are different than usage models for corporate employees. Someone could create a “persona” of me, I suppose, and pitch products to me based on that. Even as an individual chooser I have constraints: mainly in my budget, but also in changes in my lifestyle, or perhaps if I started screening my purchases through my significant other, prioritizing family needs differently, etc.

A “user” in an organization is far more constrained than that. Someone else chooses the hardware and software the person will use, how often it will refresh, whether it will be consistent across enterprise applications or not (those choices being made by the people who purchase them, usually different organizations). People in organizations work on one or many teams, have managers, policies, paperwork, etc. Basically, they have people on all sides laterally, above and possibly below, to coordinate with. They do have leeway and they do determine how their own work will be done and that is the essence of the corporate usage model: how does this person get his or her work done?

How we put together the context picture for usage. I recently had to summarize all of the enterprise-focused research and innovation that IT Research and Innovation had done in the area of Workforce for a high-level overview and present in a very limited set of (5) posters. I was allocated 20 minutes to cover the entire subject, but being last on both occasions, I actually did it in about 7 minutes. For this reason, the story is very clear in my head.

Here is my run-down of corporate IT Research’s Workforce Productivity and Innovation areas.

1. Enterprise system dynamic model. I saw this model at Boeing PhantomWorks and it depicts the agile military. News to executives: you are also users! There is no changing the body without changing the head. The PhantomWorks model comes from the network-centric organizational model outlined by David Albers and Robert Hayes in their book Power to the Edge (free at CCRP website). Whether or not you agree with the model in every detail, it has a lot to teach us about the relationship between collaboration at the top over policy, rapid communication of that policy through socialization and training, tactical independence in the field and rapid communication of results back to the top. There is IT at every stage of this model and there are people at every stage of this model.

2. Communities. Across the enterprise we have peers, or would-be peers, people who interact outside the bounds of hierarchy, across organizations, to exchange and seek information. One term used for some of these is “communities of practice” though strictly speaking that is a misuse. Don’t argue with me, I helped create the term! Communities of interest or communities of practitioners would be good descriptions. This level is a kind of glue that binds the organization together outside of normal structures. Communities are supported in wikis, distribution lists, social networking technologies, as well as in corporate communications and blogs. Any “knowledge management” app also is in this area. Since Intel is a large and totally global company, the more we know about these self-forming communities and provide access and scaffolding (lightweight infrastructure) to them, the better.

3. Teams. Where the deliverables get delivered. Structured and unstructured work processes with defined groups of people, in a hierarchy or matrixed structure with expected outcomes at expected times in accordance with corporate or departmental objectives. These also have to be supported globally. We have lately devoted more effort into novel approaches to support large global design teams who need better visualization and more communication redundancy based on their high-value, complex, distributed work that is done cross-culturally and on critical schedules.

4. The Individual User. The individual sitting or moving around at/with his/her desktop/laptop/portable device. This is where much usage model research is focused. IT has a large group of researchers in this area as do our product groups with many reports to make. One unique are we have lately concentrated on in IT Research is understanding the facts about multi-tasking, using a machine learning tool from Oregon State University Department of Electrical Engineering and Computer Science. The tool is called TaskTracer which runs in background and tracks activity mostly mathematically. TaskTracer so far has shown the high overhead the average user incurs just in shifting contexts. Not only are people on multiple teams, not only do they multi-task, but, what was not obvious: each task can use multiple applications. That is a lot of overhead just navigating your desktop! I took a random screenshot of my laptop while preparing a poster and I had ten visible windows not to mention what was minimized. OK I am less organized than my betters. But there were many nods of recognition when I showed this. So at the micro-level we have a problem. We used that to argue for our innovative 3D user interface, Miramar.

……..

Here is the slot for all the other usage model work that has taken place in IT. There is plenty of room for that and I hope my colleagues will leave lengthy comments because I can’t cover it all. I know David Sward is dying to dive into this space, and I will leave it to him

Bottom line: the user exists and acts with others in a complex system. The user moves in that system like a fish in a school of fish, only maybe a school of fish that lives in a sunken battleship, ie has an artificial structure around it to navigate through. We can’t understand the individual fish without understanding flows, swarming and patterns of fish communication. And we need to take the battleship into account, too. (We can make it a submarine if “battleship” seems too bulky.)

Comments (2)

tagged: communities, complexity, enterprise, ethnography, layers, organization, system dynamics, teams, usage