You may have seen ‘README’ files before. And, like most people, you may have completely ignored them! It’s true that in most applications they contain information that’s not really of interest to the viewer, or instructions that aren’t really necessary. But in the context of complex scientific datasets, they really can provide a useful key for researchers who are unfamiliar with your data management protocols. Or, indeed, for you too, some years after you’ve archived your data and can’t remember what’s in a bunch of old directories.
1) README files don’t have to be long or verbose. In fact it’s important that a separate, small README file is put into every directory that requires explanation, rather than putting one comprehensive explanation at the top level of your directory structure.
2) The main function is simply to describe, in plain English, what data is contained in the folder, for the benefit of future researchers who might be unfamiliar with the work.
3) Include scientific keywords and metadata to assist search engines.
4) Include author and collaborator names, and if possible a contact email address that will remain active permanently.
5) Include descriptions of any modifications to the directory contents, and date of those modifications.
6) The README file should always be in plain text format (.TXT), so that it can be opened by any device, now and in the future.