View All Pages

Design Example: Adminima

As I've become more senior at the University of Washington, I've taken on more administrative work, such as overseeing Informatics, or all of the iSchool's academics. That's meant that I've not only had to understand our bureaucracies, but also be responsible for maintaining and evolving them. I wrote this piece to help clarify my ideas about how I want to change it.

Adminima

Amy J. Ko

No one likes bureaucracy. The word's connotations says all anyone needs to know, invoking feelings of complexity, injustice, confusion, dysfunction, and more. And yet, in any non-trivial group of people with some shared purpose, it seems inevitable: to provide some kind of order amidst the chaos of large groups, we seem to need to create rules, processes, and procedures, and these rules, processes, and procedures seem to always be a source of pain when interacting with an organization. Anthropologists, for example, have found that much of this pain comes from the documents that we create to structure these rules, processes, and procedures (Hull, 2012). They seem to be central actors in shaping and structuring the frustration in our work.

My own experience with the iSchool's bureaucracies has given me an up close view of exactly how our processes and policies can obstruct student learning, rather than help it. For example, I recently started a large scale independent study to support an open source project I'm working on. Student services had a form for students to fill out to get an "add code" to register. But this form did not align with the information I asked students to share with me. And so this led to more than 50 emails of students asking how they should fill out the form. And then, after the form was filled out, students did not know if it was received or if they would ever receive a reply, and so I received 50 more emails asking for an update. I wrote several emails to student services, asking for an estimate of when students would receive a reply, but they took weeks to reply (because they were on vacation), and so students and I waited with confusion while the requests sat in an inbox, unexamined. None of this was transparent to students or I; and none of my unique goals or their conflicts with our independent study registration were transparent to the staff.

I believe that the root cause of this and other kinds of UW bureaucratic pain stems from a lack of transparency in process, and the status of these procedures. Faculty and students often do not know what the process is, and staff have no easy way of communicating it or providing updates on its status. This leaves students without the information they need to learn and faculty without the information they need to teach. And it results in a constant flow of information requests from students to staff and faculty to get this information.

Existing tools do not help with this. To do lists and project management tools are generally private. And using public documents and web pages to make processes transparent ignores the inherent structure of procedures, and would require manual email sending to notify people of updates.

I propose to create a new information system to address this transparency problem. It will be a web application, called Adminima, which will transparently document internal procedures and faculty and staff process on completing them. The design is as follows:

  • The application will have the concept of a Role, which is a set of Procedures that are completed by someone in that role. A Role can have one or more responsible people for the role.
  • A Procedure is the heart of the application, and includes a set of people responsible for completing the procedure, hierarchy of Tasks, and a start date. If the Procedure is recurring, it also has a schedule on which it repeats (e.g., every week, every month, every quarter). A Procedure is done when all of its tasks are done.
  • Each Task is a rich text description of some discrete, completable piece of work, the people responsible for it, and links to any online resources they'll need to complete it. It also includes a rationale for the task, which explains why the step is necessary. Tasks are either complete or not.
  • Procedures can have watchers, which are people who want to be notified when a Task in a procedure is completed. Watches supply an email address and they are notified when a Task or Procedure is completed (or undone).
  • Every Procedure has a unique URL, allowing it to be easily shared.

The application would have a timeline view that shows all of the Procedures and when they start. This would provide a high level overview of a Role's procedures, and the other Procedures they are doing in a parallel, to give transparency to the Role's workload. A Procedure view would give clear visibility to the tasks remaining before the procedure is done.

This application would have a number of clear benefits to transparency. Using Adminima for the independent study registration, for example, would have made it clear to students that the form they submit goes to student services, and not me, and that they are reviewed by an advisor before the start of the quarter, and not by me. It would have made it clear to me that the person who reviews them would not review for a while, since I knew he was on vacation, and would have allowed me to ask a different advisor to do the work. Students would not have to have written for updates, because they could have watched the quarterly procedure to get updates. And the advisor who was on vacation could have used the list of watches to send a manual update, noting the delay due to vacation. This whole experience would have reduced the student and faculty workload by 100 emails, reduced student anxiety. It would have also made clear that the independent study registration procedure had a single point of failure, since there was only one person responsible for processing them, and he was away.

There would be other benefits to the organization. For example, when staff leave the organization, they would no longer take knowledge of procedures with them, since it would be documented, and it would be clear that someone new needs to be responsible for them. It would also allow for faculty and students to advocate for procedure improvements. Having visibility into the volume of procedures a role is responsible for would also allow for better workload audits and reallocations, to avoid staff burnout.

While Adminima would have many benefits to faculty, students, and staff, there are some stakeholders that might be harmed by the system. For example, when procedures are conducted exclusively by people, and not written down, there is a degree of flexibility in how they happen. This is important to responding to exceptional circumstances, which can be more common amongst students whose lives do not conform to dominant norms. Documenting procedures in this way risks "hardening" procedures, making them even harder to design for equity, as revising a document is harder to do than just changing behavior.

That said, even if there is some initial harm, making bureaucracy transparent may still have long term benefits to equity, allowing it to be examined, critiqued, and studied as subject of advocacy. Maybe, with enough cultural flexibility in bureaucracy, we can start shifting the connotation of the word from one of suffering, to at least one of function.

Hull, M. S. (2012). Documents and bureaucracy. Annual review of anthropology, 41, 251-267.