SPEC Kit 340: Open Source Software (July 2014)
Page11(11 of 184)
SPEC Kit 340: Open Source Software · 11 EXECUTIVE SUMMARY Introduction Open source software (OSS) “licenses must permit non-exclusive commercial exploitation of the licensed work, must make available the work’s source code, and must permit the creation of derivative works from the work itself.” [St. Laurent, Andrew M. (2008). Understanding Open Source and Free Software Licensing. O’Reilly Media, p 8. ISBN 9780596553951]. The emergence of OSS increases collaboration among research libraries, providing greater control of library tools, as well as improving usability and quality of library resources. This collaborative ap- proach fits neatly with the knowledge and resource sharing ideology of libraries. While OSS is ostensibly “free,” adoption of OSS within an organization is not without significant support, integration, and develop- ment costs. The purpose of this survey was to study ARL member libraries’ adoption and/or development of OSS for functions such as an integrated library system (ILS), discovery layer, electronic resource manage- ment, inter-library loan, digital asset management, institutional repository, course reserve, streaming media, study room scheduler, digital preservation, publishing, floor maps, data warehouse, and other library-related purposes. We wanted to understand organizational factors that affect decisions to adopt OSS, the cost of OSS, and the awareness of OSS sys- tems already in use. With regard to development of OSS, we wanted to understand: 1) research libraries’ policies and practices on open sourcing their code 2) the frequency of research library contributions to open source projects 3) the reluctance of research li- braries to make their code openly available and 4) the most common benefits and challenges encountered when research libraries open source their code. This survey was distributed to the 125 ARL mem- ber libraries in February 2014. Seventy-seven libraries (62%) responded to the survey by the March 17, 2014 deadline. Library IT Staff The 66 responding academic and public libraries reported between two and 50 staff with IT respon- sibilities as all or part of their duties, with an aver- age of 16 and a median of 14. Three national libraries reported between 130 and 350 IT staff. This bimodal distribution is stark, with the national libraries an order of magnitude larger than their university coun- terparts. Despite this difference in staff size, we find no statistically significant differences in the relative participation in OSS projects. Seventy respondents (91%) develop software in- house. Of those, the most common software develop- ment practices include using version control (86%) and performing usability tests (86%). The least common practices include the use of independent quality as- surance (24%), adherence to a formal, written code reuse policy (10%), and the presence of a committee or working group to encourage code reuse (7%). The most common other software practices mentioned by respondents were agile/scrum development method- ologies (5 of 15 respondents) and pair programming (2 respondents). Most respondents reported that their library IT staff are encouraged to experiment with new technologies (75 or 99%), and prototype potential projects (62 or 82%).