Start Your Own Lab

We have created two geometry labs ourselves (in VA and TX), and have been involved with the senior management of a third lab (in MD). Additionally, we have consulted with others helping them create and maintain their own experimental mathematics labs.

We encourage faculty interested in creating an environment and community ripe for collaborative and experimental research that involves both undergraduate and graduate students to consider starting their own "Experimental Mathematics Lab".

We are happy to consult and collaborate: contact MEGL's director Sean Lawton for more information.

Step 1: Obtain Physical Space. Perhaps the hardest part of this processes is finding physical space. Creating a lab environment does mean having a physical space. It does not have to be big (my first lab was a converted storage closet). Be creative! This part will require talking with your Chair who will probably need to talk with your Dean. They will probably need a prospectus. We can help write such a document and write letters of support if needed.

Step 2: Design the Space. Funding helps with furniture and writable surfaces, but if you don't have funding look for donations (old white board in storage, or tables and chairs that were in surplus). My first lab was started without a promise of physical space or any funding (my second lab had the opposite), so it is possible.

Step 3: Get Equipment. Again, with funding this is easier (still there is a need to figure out what to get). Without funding you have to look to surplus and donations (computers that are about to be out of date and can be re-purposed with linux). Choice equipment includes: 3D printing/scanning, Virtual Reality, and GPU computing; but a couple workstations is all that is needed to get started. Once you start succeeding small dollar funding can be cobbled together over time to get better equipment.

Step 4: Advertising and Recruiting. This can very difficult depending on the current environment. With my first lab, I had to start a pure math major, recruit for that, then recruit from those new majors for my lab; it was very hard and time consuming. For my second lab, there was unknown demand before I got there and recruitment was easy. Some simple ways to recruit: visit upper level math classes, make fliers, send emails, and set-up a "first look recruitment day" like a seminar with pizza (if you can afford it). Also, making a website and logo (maybe even a brochure) helps make students (and your colleagues) take you more seriously.

Step 5: Forming Teams. If you have funding then you can offer hourly wage for research assistants. Otherwise, I recommend offering unpaid lab internships. These are valuable to students since it is a resume mark that employers take seriously and which honestly reflects real research experiences, working in teams, where they develop programming, collaboration, and communication skills working in an abstract and often challenging (yet fun) setting. Once you know who you want to be in your lab, you have to form teams. I recommend 2-4 undergraduates (+1 graduate student if available) per team. Interviewing the students before committing to work with them is a good idea. Also, having them sign some kind of contract outlining their responsibilities is a good idea too.

Step 6: Project Development. This could also be Step 0. I assume you have projects in mind before you went to the trouble of doing Step 1. But a good lab project is something that an enthusiastic student with A's in basic calculus, linear algebra, and intro to proofs can get started with. You want projects that lead to problems that you and the research community are interested in, but whose questions/conjectures can be motivated by analyzing patterns in computer generated data sets and visualizations. As the coding and conjectures are happening you can pair the actual work the students are doing with discussions of theory (with some black boxes as needed) and learning about more advance general content (cherry picked for relevance). As they work, they slowly figure out what they have been doing the whole time. Eventually once sufficient conjectures are formulated the team starts to try to prove them. Sometimes this requires the more experienced students or faculty mentors to take charge with, but the data and pattern recognition can be used to help direct what techniques, directions, and ideas ought to be used to prove the theorems.

Outreach and Visualization
After your lab is up and running we recommend starting Outreach and Visualization projects (in addition to experimental research). Having students participate in activities that promote the fun and excitement of mathematics to various populations (e.g. middle school) at once teaches undergraduate and graduate lab interns how to communicate in creative ways while providing an educational service to the extra-university community. See our outreach page for examples. Visualization projects (or more generally projects that create alternative ways of experiencing mathematics) serve as a bridge between experimental research and outreach.

The most important step, after creating content and activities to deliver that content, in conducting outreach is creating a network. This can be started with emails to activity coordinators at local libraries, or principals at local schools (word of mouth is your best friend). Be sure to have an activity developed and ready to implement before contacting people however, as they will likely want to meet with you and hear about what you want to do before they consider your activity.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License