“Open source software is software that can be freely accessed, used, changed, and shared (in modified or unmodified form) by anyone” (cp. https://opensource.org/osd). So open source software (OSS) is actually something that one or more people can work on, improve it, refine it, change it, adapt it and share or use it. Why would anyone support such a feature? Examples from the industry show that this is a valid approach for many software products. Prominent open source projects are in use worldwide on an everyday basis, including the Apache Web Server, the Linux Kernel, the GNU Compiler Collection, Samba, OpenSSL, and MySQL. For industry this means not only re-using components, and libraries, but also being able to fix them, adapt them to their needs and hire people who are already familiar with the tools. Business models based on open source software focus more on services than products and ensure the longevity of the software as even if companies vanish, the open source software is here to stay.
In academia open source provides a way to employ well-known methods as a base line or a starting point without having to re-invent the wheel by programming algorithms and methods all over again. This is especially popular in multimedia research, which would not be as agile and forward looking if it weren’t for OpenCV, ffmpeg, Caffe, and SciPy and NumPy, just to name a few. In research the need for publishing source code and data along with the scientific publication to ensure reproducibility has been identified recently (cp. ACM Artifact Review and Badging, https://www.acm.org/publications/policies/artifact-review-badging). This of course includes stronger support for releasing software and data artifacts based on open licenses.
The SIGMM community has been very active in this regard, since ACM Intl. Conference on Multimedia hosts the Open Source Software Competition since 2004; this competition has attracted in the latest years an increasing number of submissions and, according to Google Scholar, two of the currently three top cited papers in the last 5 years of the conference were submitted to this competition. This year also the ACM Intl. Conference on Multimedia Retrieval has introduced an OSS track.
Our aim for SIGMM Records is to point out recent development, announce interesting releases, share insights from the community and actively support knowledge transfer from research to industry based on open source software and open data four times a year. If you are interested in writing for the open source column, or have something you would like to know more about in this area, please do not hesitate to contact the editors. Examples are articles on open source frameworks or projects like the Menpo project, the Siva Suite, or the Yael library.
The SIGMM Records editors responsible for the open source are dedicated to the cause and have quite some history with open source in academia and industry.
Marco Bertini (https://github.com/mbertini) is associate professor at the University of Florence and long term open source supporter, especially by having served as chair and co-chair of the open source software competition at ACM Intl. Conference on Multimedia.
Mathias Lux (https://github.com/dermotte) has participated in the very same challenge with several open source projects. He’s associate professor at Klagenfurt University and dedicated to open source in research and teaching and main contributor to several open source projects.