Open science is really scary y’all
I did something stupid last week. I publicly posted my data and code.
I’ve been living in a nightmare ever since, terrified by every email and notification, fearful it’s someone pointing out how I screwed up.
What if I did it wrong? I don’t know everything. Hell, I don’t know anything, not really. Or what if there’s an error in the code? FFS, I even miscoded the outcome before Maarten spotted it. What other mistakes are lurking? It’s all too much. I can’t tell you how many times I’ve poured over it all. It’s an obsession. Where is the mistake? WHERE IS IT DAMN YOU!!!? I’m freaking out.
And speaking of code, it’s hacky AF. The annotation is weak. There are even multiple for-loops for god’s sake. FOR-LOOPS people! Then there’s the data. Just a bare-naked csv file. No code book. No metadata. Definitely not FAIR. And it’s all just shoved onto a OSF page. There’s not even a Wiki — just 3 files and a DOI. I am so bad at this.
And for what? It’s not like I want anyone to look at it, that would be terrible. Not that anyone is actually interested. The editor and reviewers for the resulting manuscript didn’t care. I know this because I didn’t realize the cited OSF page was set to private until after the paper was accepted. The higher ups at my university don’t care. Sure, we have a new code of research conduct (good) that even has a nod to reproducibility (very good) — but lacks a detailed section on it (not so good — but we’ll get there). I just checked the promotion criteria. Impact factors — check. Grant money — check. Open science methods…hmmmmmm.
But at least it’s done. And it was my first. I’ve been advocating for open science practices for a few years now. We’ve held workshops, and meet-ups, and applied for open science grants. I’ve tweeted and retweeted open science topics. I’ve been a huge hypocrite.
But you know what? Not anymore, and I am relieved.
Why didn’t I take this step sooner you might ask? The answer is simple. It’s so damn easy to do nothing. So I took a small step. But it was a step nonetheless, and now I can call myself a real-deal practitioner of open science methods and build from here.
And who cares if nobody reads it. I did it to improve my work (and maybe that of others), not to satisfy some administrator. Who cares if someone spots an error — that’s a good thing. And now I understand that my terror isn’t from posting the code — it’s from posting the code after the paper was accepted, and that’s easy enough to fix next time. Who cares if I don’t get “career credit” today. Things are changing. Funders are changing. Appreciation for the effort is certainly coming, tomorrow if not today. But most importantly, the problems we are trying to solve are real, and greater transparency is a crucial part of the solution. So take a small step today. Submit a pre-print. Open an OSF account. Share your gawd-awful code! I promise you’ll feel good about it…eventually.