The Howison Lab is seeking undergraduate and graduate students to assist with data collection in two ongoing research studies. RAs in both projects will gain basic skills relevant for programming and collaboration: version control, Github, Resource Description Format (aka Linked Data), and the command line as well. Research Assistants (RAs) will be paid $15/hr, with the expectation that they will work at least 10 hours per week.
In one project, Scisoft, we will be working to help increase the rewards for scientists who are building software used by other scientists. We will be reading scientific papers in a set of disciplines (esp. biology, ) and finding mentions of software. These mentions are scattered throughout the paper, sometimes in the full text, sometimes citations, sometimes URLs in footnotes etc. We'll be highlighting these and recording attributes, such as whether they give credit, include version numbers etc. Our work will build a "gold standard" dataset to be used to train supervised machine learning algorithms. If you are interested in machine learning you will find the process of building a labeled dataset interesting and one you can re-use in future. If you are interested in scientific publishing this will give insight into the need for improved guidelines for authors. If you are interested in the factors that shape how scientists work together and how incentives, rightly or wrongly, drive scientific activity, you will gain a lot from the study. For more information on this project, please read this blog post and the attached article: http://blog.impactstory.org/collaborating-635k-grant-improve-credit-research-software/.
In our other project, the Transition project, the lab is researching how software development projects funded through research grants (NSF, NIH, etc.) might transition to being community maintained and open source at the end of the grant. The goal is to identify characteristics that enable software projects to grow and exist beyond the timeline associated with their funding rather than fizzle out with the conclusion of the grant. For example, there may be differences in recruiting methods for new developers or in collaboration infrastructure like bug trackers or forums. This work will help scientists develop more sustainable software and can inform funding agencies’ policies; ultimately we hope to have more and better software for science. This work will teach qualitative research methods, including the use of a coding scheme for content analysis. If you are interested in open source projects, scientific software, or online collaborative work, this project will be a good match for you. For more information on the motivations and plans for this project, please read the article found through this link: https://www.ideals.illinois.edu/handle/2142/73439.
To be eligible:
If you meet the above eligibility criteria, please email Fan Du at firstname.lastname@example.org with 1) a copy of your resume, 2) an indication of which project(s) you would like to work on, 3) acknowledgement that you meet each of the eligibility criteria, 4) your UT EID, 5) a brief explanation of any other employment you have through UT this semester (e.g. “I’m working 5 hours per week as a research assistant in another lab.”).
The Howison Lab is in UT’s School of Information, headed by Dr. James Howison.
Friday, September 21
UT Administration Building (UTA)
1616 Guadalupe Street, Austin, TX