creator Heston, Matthew advisor Hemphill, Libby We collect data for 13,383 rst time code contributions from 45 projects on the website GitHub and analyze behavior of developers before submitting code as well as community response to code contributions. Our ndings di er from previous research on open source software communities and social theories of learning in communities of practice. We nd most users do not participate in GitHub peripheral activities before submitting code changes. We also nd that community response to these submitted code changes is a poor predictor of whether or not the code is accepted. Submitted by Liana Khananashvili (khananashvili@iit.edu) on 2014-10-17T19:24:37Z No. of bitstreams: 2 Heston_Matthew_Thesis.pdf: 201868 bytes, checksum: 673ec44be736e50723795797274bcdf9 (MD5) signed_cover.pdf: 79002 bytes, checksum: 7b03bf146ad9d33687fae0bf6b6d0946 (MD5) Made available in DSpace on 2014-10-17T19:24:37Z (GMT). No. of bitstreams: 2 Heston_Matthew_Thesis.pdf: 201868 bytes, checksum: 673ec44be736e50723795797274bcdf9 (MD5) signed_cover.pdf: 79002 bytes, checksum: 7b03bf146ad9d33687fae0bf6b6d0946 (MD5) Previous issue date: 2014-05 M.S. in Information Architecture, May 2014 2014 2014-05 http://hdl.handle.net/10560/3282 en Thesis born digital application/pdf In Copyright http://rightsstatements.org/page/InC/1.0/ Restricted Access Other Illinois Institute of Technology Affiliated department