Discussion on "My Target Audience"

Matthijs Cox · 2023-04-18T11:46:15.511Z

What kind of people do I have in mind while writing this blog? People who share my professional mission of course! What is that mission you ask? Let me elaborate. From Research to Engineering I am a scientist who danced with startups and moved into i...

Hey Matthijs Cox! Just came across this post linked from the most recent JuliaHub blog post. Not sure if you were aware when writing this, but I actually gave a talk at the third JuliaCon on this very subject! (The comment engine won't let me link directly, but if you search YouTube for "JuliaCon 2016 The Two Cultures of Programming" it should pop right up.)

Really neat to see I'm not alone in this view :-)

Hi Joshua, I almost forgot about your talk, but I remember watching it and you definitely inspired me years ago. I should have included a link to your talk, thanks for reminding me! Let me know if you ever want to chat about this topic.

Great read! I think you should checkout the Research Software Engineering communities, for example US-RSE (I can't post a link unfortunately). It feels like target audience you are writing about.

Thanks! I recently found BSSw, will also check out US-RSE. I am also curious if there are EU counterparts.

There are! The most prominent one is UK-RSE, I think, but DE-RSE is also developing well. Matthijs Cox

great post. I feel like the set of people is larger in pure research environments. like you have "only" scientists who previously have stayed away from difficult languages like C and Fortran, leaving those use cases to a small group of people. the arrival of julia and the prospect of writing performant code in a more user friendly way has opened many doors.

I fully agree. Recently I saw Chris Rackauckas use the same kind of argument to explain why Julia development will overtake Python development (at least in niche scientific computing cases), simply because no one maintains and upgrades the fortran/c/c++ libraries anymore, let alone write new algorithms for solvers.

Great post. But after reading this I am confused about the definition of 'scientific coder'. Are there truly developers that want to, and then are allowed to by their employer, to dive into the science behind what they are 'developing'? It seems like two completely different worlds to me. I do understand it the other way around though, it is really painful to read code written by some scientists even though it 'works'. I am in academia but I spend a lot of time trying to improve the quality of my code. Mostly out of the frustration I feel when I read somebody else's code and I don't understand it. Not because it's difficult but because it's poorly written. I don't wish that on anyone

Thanks for the question actually! Perhaps my description is still too generic.

I agree that in many cases we can just teach scientists to code better. There seems to be a whole industry of converting scientists into software engineers. You can find lots of vacancies for scientific software engineer or scientific programmer if you google for it.

Those people still tend to use lots of science and numerical computing principles in their code.

Finding pure software engineers and teaching them more science is a much rarer path, though I have found some during my career.

I do intend to interview scientific coders and ask what they do, how they work and what products they make. Hopefully things will become clearer with such examples.

Matthijs Cox

Yes, that's exactly what I meant. I look forward to the interviews.

Feel free to leave a comment! I learning how this blogging platform works together with you :)

Discussion

My Target Audience

Responses(5)

Recent discussions

Search Hashnode

My Target Audience

Responses(5)

Recent discussions