Research software is like coffee: craft and respecting origins
Posted on 1 May 2017
Research software is like coffee: craft and respecting origins
By Laurence Billingham, RSE at NERC and Software Sustainability Institute fellow
Anyone who has met me in real life knows I love coffee. In fact, I'm writing this post whilst drinking an El Salvadorian pour over in a place with exposed brickwork and a floor that looks like it used to be part of a school gym.
But this post isn't about my love for coffee per se or even about using a caffeine rush to produce awesome code really speedily. It is about a thing I'm struggling to name (naming stuff is one of the canonically hard things after all); the best I can currently come up with is:
Barista mindset
Work like a craftsperson, respect the care that others have put into your materials.
In my previous group, I was lucky enough to work alongside a bunch of talented and dedicated engineers. These folks sweat for the data they give me. Sometimes they spend weeks away from their families in lonely windswept places; making sure our instruments keep working properly and calibrating their data. I'd estimate that we spend several hundred thousands of dollars per year on instrument calibration and data quality assurance.
In a previous gig too I was fortunate to be able to witness the care and effort a bunch of really knowledgeable folk put into getting their instruments measuring correctly... before strapping them to tonnes of hydrazine and blasting them off to other worlds.
"Hang on," I can hear you muttering, "what has any of this got to do with coffee?".
The coffee we drink is the end product of a long chain of work and care put into it: from farmer; to exporter; to roaster; to barista; to, finally, the consumer. I've been struck talking to various baristas about the respect they have for where their coffee came from; for example, listen to Claire Wallace interviewed on this edition of the BBC Radio 4 Food Programme. She talks about wanting to represent all the work that has come before.
As Research Software Engineers we are also often the final link in a long chain. We filter, transform, and analyze data, before presenting the results to researchers. The chain of value between the first measurement to the final data product is only as strong as its weakest link.
Engineers worked hard and took great pains to make sure the data that reach me are correct. Like a careful barista, I want to express that work and care, to do justice by the engineers.
I try to treat coding as a craft, creating high-quality products by working with respect for those who cared for the data before me.