What does this mean: sources or binaries?

Open source software knows two ways to distribute the software: in the form of binaries and in the form of sources. Mostly it is up to you to decide whether you want to distribute only the binaries or whether you are intentionally going to distribute the sources (too).

At a first glance, the concepts ’sources’ and ’binaries’ seems to be clearly distinguished. On the one hand, compiled sources should be taken as binaries. On the other hand, editable pieces of software are denoted by the concept ’sources’. But sometimes the difference is not as clear as wished: For example, you can modify even already compiled object files by using an hex-editor. Or it is very difficult to modify the minimized versions of javascript files even if they are indeed text files.

Therefore, the OSLiC ’reuses’ a famous rule of thumb: "The source code for a work means the preferred form of the work for making modifications to it." All other forms are denoted by the concept of 'binaries'.

Based on this specification, you can respect some special conditions if you want to distribute the sources and/or the binaries.