Desk Allocation at CCRMA
Here is the algorithm used for allocating desks to PhD students and other people who might want a desk at CCRMA. As of 10/23/18, this process is still under development and this page represents a draft of the process.
When to Assign Desks
Desks should be (re-)assigned when there is a vacancy coupled with several people interested in having a desk. Since there are not enough desks at CCRMA, this is expected to happen every time there is a vacancy. If multiple desks are expected to become available around the same time, the process should be postponed until all of the desks are free; in the interim, the free desks can be treated as an open work space for PhD students.
Values of Assigning Desks
- Don't disrupt the work life of people who currently have desks. People grow to depend on having their own work space and so a desk assignment should not be forcibly removed from someone (unless they are not in fact using their desk).
- For new desk assignments, prioritize PhD students at key later milestones of their PhD, followed by PhD students who are currently TAing for the department.
- People who take a desk should work in it for a significant portion of the day at least 3 days a week on a normal week, or arrange to share their desk with someone else.
How to Assign Desks
The priority order for assigning new desks is:
- 5th year PhD students
- 4th year PhD students
- 6th year PhD students
- 3rd year PhD students
- 2nd year PhD students
- 7th year PhD students
- 8th year PhD students
- 1st year PhD students
- Visiting scholars (note: the CCRMA page on visiting scholars explicitly states that visiting scholars will NOT be assigned an office space and should use the open space office environment)
If multiple people from the same priority class want desks, have them flip a coin to determine who picks first.
Here is the procedure for assigning new desks:
- PhD students who have a desk already can keep their desk and not participate in the process. It is understood that Jonathan Abel, an Adjunct Professor at CCRMA, will keep his desk in the upstairs desk area.
- Obtain a list of people who might want a desk, including all current PhD students and upcoming visiting scholars. Circulate the announcement that desk assignments will happen soon.
- Determine who wants a new desk, including students who do and don't already have an assignment.
- Create and circulate a map of the current desk assignments, indicating which desks are free. For those students looking to switch desks, free up their space on the desk map.
- Perform a "mock" assignment. In order, ask each student in the priority-ordered list of people who might want a desk to select a desk they would hypothetically want to take.
- Once the mock assignment is complete, each student who currently has a desk and opted to join the mock bids can decide whether they want to pull out of the process and keep their current desk.
- Perform the actual assignment. Once again, in order, ask each student in the priority-ordered list of people who definitely want a new desk to select their desk from the list of remaining options.
- What should we do if someone cannot use their desk for a full quarter or more? Some possibilities include:
- going through a round of desk selection ever term, and possibly the person leaving for a term gets bumped from their desk
- yielding the desk for the absent term but having the right to reclaim it upon returning
- allocate the desk for a temporary person, such as a visiting scholar
- Should 6th years be prioritized above 5th years?
- "Normative" funding time is to finish in 5 years, so we shouldn't allow people who are staying longer to get top priority in getting desks
- But! most people don't finish in 5 years, so maybe we should allow 6th years a slightly higher priority in picking a new desk?
- How to make future modifications to this process?
- If more than 50% of students currently affected by the process give a written "yes" to a proposed change, it will be enacted?
- How do we tell if someone is officially not using their desk enough and should lose it?