Submitted by Erica.Johns on Wed, 03/23/2016 - 14:37
Typically, proprietary software is not shared within a data package due to licensing, yet the documentation accompanying the dataset needs to contain metadata regarding the software and the version used to analyze and later access the data (these may be different). It is always good to encourage users to use non-proprietary software when possible. It is important to note that often, while proprietary software is being used for analysis of the data, there are non-proprietary software packages or viewers that allow access to the data and some limited analysis, as well. Ask researchers who use proprietary software to identify open source alternatives that can facilitate access to their data if they exist, and be sure to include this within the documentation. If the researcher wrote the software or the software is open-source, then it may be possible to share the software within the data package. If it is possible to store the open source software with the data, one must verify they have the right to redistribute said software. Many open source softwares will have stipulations against commercial reuse, thus any original clauses/restrictions must be followed. When archiving data, one may place a copy of the software (proprietary or not) within the package and include a line in the public record saying 'please contact PI/librarian/applicable person if you need software assistance' to enable long-term access while also adhering to licensing.