search

nordic-rse / meetups

Proposal incubator for our online meetups.
1 stars 0 forks source link

Talk proposal - "my first R package" #7

Closed Teebusch closed 2 years ago

Teebusch commented 4 years ago

I would like to propose a talk for the online RSE get-together. See draft of abstract below. I'm thinking of submitting as a lightning talk, but pondering whether there's enough interest to make it a 20min talk, instead. I don't know what the rest of the program will be like and how high the standards are for 20min talks. My thoughts on the format:

I'd be fine with either format, but was wondering if anyone has any thoughts on this.

What I learned from publishing my first R package, and why you should write one, too.

The code I write as a researcher is single-user and single-use. It is written iteratively and usually serves a data analysis (wrangling, exploring, modeling, and reporting). It is meant to get things done quickly and will likely never be looked at again. As a result, this code is neither very reusable nor very elegant, and after doing this for a few years I'm not learning many new tricks anymore. Instead, I'm repeating myself and reinforcing my habits - good and bad ones.

In this talk, I want to speak about how writing an R package has helped me break out of this trot, and get a deeper understanding of not just R, but also software engineering best practices. Writing a package has helped me get a better grasp of software architecture, version control, continuous integration, project management, documentation, code profiling, functional and object oriented programming and more. Picking a project outside of my comfort zone has forced me to dive into unknown areas of computer science. Finally, it is a rite of passage for an aspiring software engineer.

R has a very streamlined ecosystem for package development that is perfect for beginners and makes adhering to best practices easy. That's why I would recommend using R for a first software package. However, writing code that is meant for others to use, will elevate your expertise in any programming language.

(The R package I've written is noah, it is very simple, and is not going to blow any experienced RSE's socks off. The focus of the talk is on the benefits of package authoring from a researcher / data analyst perspective)

bast commented 4 years ago

Thanks for the proposal!

bast commented 4 years ago

Just confirming also here that we are looking forward to have this in the program and we will together fine-tune the timings (we are still unsure about how much time we have for talks but we are working on it and will agree on this together with you).

Teebusch commented 4 years ago

Awesome! Thank you!

bast commented 4 years ago

dear Tobias, this is now in the schedule: https://nordic-rse.org/events/2020-online-get-together/#tuesday-december-1st

15 minutes would be ideal from the scheduling point of view. Is this OK for you? Can I use the above text as abstract to put into the session notes?

Teebusch commented 4 years ago

Hi @bast, 15 minutes is ok. I submitted an improved abstract through the abstract submission system. Could you use that instead, please? Or just copy it from here:

This talk is about how writing an R package has helped me leave my comfort zone and level up my R programming skills. The code I write as a researcher is mostly single-user and single-use. Writing and publishing code meant for others has helped me break old habits and get a better grasp of software engineering. R has a very streamlined ecosystem for package development that makes understanding and following best practices easy. I will talk about the things I have learned, why I think writing a package should be a rite of passage for any aspiring research software engineer, and why R is a great tool for this.

bast commented 4 years ago

Great! I will take the abstract from the abstract submission system. Thanks!

rantahar commented 2 years ago

Closing, this was proposed for the 2020 workshop.