I am a huge fan of strong file naming, specifically using file naming conventions. Such conventions provide consistent and descriptive names for a group of files. A good file naming convention can add order to your files while you are actively working on a project and make files easier to find later. Given the choice, I will always teach file naming conventions and their partner, the date format ISO 8601, when teaching data management basics.
There is one caveat to file naming conventions that I have not yet discussed: how to name data files that you share. I’m not talking about collaborative data sharing, where you and your colleagues are actively sharing lots of files back and forth during a research project — this scenario hugely benefits from using a naming convention. Rather, how do you name data when you are only sharing a small handful of files, such as for public data sharing? When sharing only a few files, sticking with your detailed file naming convention doesn’t make sense because it is probably difficult to interpret.
When sharing a small group of files, you should name your files for clarity. So instead of naming your file “AWA_MK01_20180222_Notes.pdf”, a better name is “InterviewTranscript.pdf”. (Based on those two names, which one gives you a better idea of what the file contains?) If you are sharing one data file with an article, you could also name your file after your article, as in, “abbreviatedArticleTitle_data”. There aren’t strict guidelines for this type of naming beyond making sure you pick a simple name that accurate reflects what is in the file.
Focusing on clarity when naming a small group of shared files is superior to using a detailed naming convention because using clear names makes it easier for someone else to quickly understand what is in those files. A lot of the problems that naming conventions solve go away when you have less than ten files. For example, differentiating similar files is much less of a problem if you only have a few files. It’s also likely that you are sharing your best data, meaning you don’t necessarily need to record a bunch of experimental parameters in the file name.
File name clarity is actually an accessibility issue as clear file names are easier for disabled people, such as those with cognitive disabilities and people who use screen readers, to understand. Actually, clear file naming helps everyone easily understand what a file contains, disabled or not! This is another case where universal design benefits everyone.
Note that the same rules about characters in file names apply no matter how you choose to name your files:
- Stick to alphabetic and numeric characters;
- Don’t use special characters like: &, $, %, #, @, etc.;
- Don’t use spaces in file names (not every operating system can handle spaces in file names);
- You can separate words in a file name with an underscore “_” or dash “-“;
- Instead of using a separator, you can alternatively use “PascalCase” or “camelCase” by capitalizing the first letter in each word (it’s your choice how to handle the first word in the name);
- Keep your file names short, ideally 25 characters or fewer.
Next time you have to share a few files, such as in support of a published article, I hope you ditch the file naming convention and opt for clarity in your data file names. You will still want to describe those files in a README.txt file, but anyone who looks at your data will appreciate the clearer names.
