A research career requires publication in peer-reviewed journals but what if some of your scholarly output is in the form of software? The Journal of Open Source Software attempts to solve that by essentially wrapping peer-reviewed software packages up as lightweight papers. My pyuca library was just accepted for publication by the journal.

pyuca is a Python implementation of the Unicode Collation Algorithm and is a vital part of most of my Greek work because it lets me properly sort Greek words. It’s not limited to Greek, though, and the library is potentially useful for anyone doing text processing using Python on natural languages other than English.

pyuca has always been citable in an ad-hoc fashion, but thanks to publication in The Journal of Open Source Software, it can now be cited as a peer-reviewed journal article.

The submission process was straightforward. I dug up an ORCID (a persistent identifier for researchers) I’d acquired a while ago but never used and set up my GitHub repo on Zenodo so a Digital Object Identifier (DOI) gets minted for each release.

I then added a specially-formatted paper.md file to the repo (including my ORCID, abstract about the software and any references) and submitted the repo for consideration.

JOSS reviews are done openly using GitHub issues. A reviewer stepped up and gave some excellent feedback on the usage example in my README and on adding contributor guidelines. Once I’d addressed that feedback, the paper was accepted by the reviewer and the editor-in-chief and a new DOI was minted for the paper itself.

I also got a notification from ORCID that Crossref had found a new work to be added to my ORCID record.

Of course, I could at some point write an article about pyuca but an article about software is not the same as the software itself (they would likely have quite different audiences) and so citing an article about particular software is not the same as citing the software itself. Thanks to JOSS, the distinction can be maintained while still keeping within a framework of peer-reviewed journal articles.

I’m particularly excited that JOSS accepted software with a digital humanities application rather than their typical scientific computing applications.

So if you publish a work that made use of pyuca, you can now cite it as:

Tauber, J. K. (2016). pyuca: a Python implementation of the Unicode Collation Algorithm. The Journal of Open Source Software. DOI: 10.21105/joss.00021