As you can see from the diagram below, "open science" can have many different meanings depending on the context and the person with whom you are speaking. Generally speaking, though, open science implies that your research is reproducible due to following best practices in research data management as well as accessible through open access publication. It is not always possible for researchers to publish their work in an open access journal, but publishing your code or data in an interpretable way online that is accompanied by good metadata is an excellent open science practice, for example.
Above: A diagram from the book Opening Science by S. Bartling and S. Frieske (eds.)