I need some advice about referencing source code in an academic journal. I rarely see it happen and I don’t know why.
I’m building a website that connects academic researchers with software developers interested in helping scientists write code. My goal is for these researchers to be able to reference any new source code in the articles they publish -- much like a “gene accession number” or a “PDB code”.
Unfortunately, I don’t see any code repositories referenced in high profile journals like Science or PNAS. I’m guessing it’s because the code in the repositories isn’t permanent and may be deleted anytime? Or perhaps a DOI needs to be assigned?
So my question to the group is:
What criteria is necessary for a code repository or database to be eligible for referencing in scientific academic journals?
Some ideas I have based on looking at the Protein Databank and Genbank are:
1) The entry is permanent -- we can’t delete articles once they’ve been published, same is true for entries in the PDB and Genbank
2) The entry gives credit to all authors and contributors
3) The entry has a DOI
4) The entry has a simple accession number - PDB is a four character code, Genebank number is six characters.
Is there anything I’m missing? Any advice would be greatly appreciated.
Heather Claxton-Douglas, PhD