Publishing code on GitHub

This semester, I’ve made an effort to get all the code I write under version control. In the past I simply maintained my codebase in Dropbox. This worked well as a backup solution and allowed me to develop the same project on my laptop and desktop without any problems (despite dealing with differences in Windows/Linux file paths). However, I’ve been involved in more collaborative coding projects this semester and Dropbox simply doesn’t cut it anymore.

Bioinformaticians as a group seem to be particularly passionate about version control and open access software – Titus Brown even says, ” If you can’t be bothered to learn how to use version control, you shouldn’t be trusted to write important software.” This goes along with the open source and open access movement academics generally tend to support. Plus, we’ve all had the experience of working with poorly maintained, documented or commented code… It can really slow down the research process and be a huge hassle.

So, I’ve made a new commitment. Every piece of code I write for an academic project will be under version control on GitHub. Code for lab work that we’ve decided to publish will also make its way there (for the time being it’s held in a private bitbucket repository, still under version control though!) This is a bit of a challenge for me – publishing code is a lot like publishing something you’ve written. You’re putting your work out there for the world to see and critique, and in a lot of cases, it’s not a finished product or something you’re quite happy with yet.

I see a lot of advantages to making code public. It should help me develop better structured, more thoughtful and well-commented code. It will allow me to share projects and ideas with anyone just by giving them my GitHub username (hint: it’s bsiranosian). I can now include my GitHub url on things like my website and business cards, and anyone can see the kind of projects I work on. I feel like this could give me a leg up when searching for jobs and the like.

I can see a few downsides too. Academic integrity is one – I don’t want someone at Brown or another university copying my code for their homework or project. After thinking about this point though, I realized the answers to most bioinformatics problems are already available at places like stackoverflow. It’s not my responsibility to make sure someone doesn’t plagiarize code. Titus Brown teaches an undergraduate class where students are required to hand in assignments on GitHub and hasn’t had any problems.

You can find my GitHub at https://github.com/bsiranosian

Posted in Coding.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.