Hackathon on reproducibility and longevity of R code for the Aberdeen Study Group - 2021
Reproducibility is important across all research, whether this is open access articles, repeatable results or giving access to code and data used for a paper. However, if you have ever tried to rerun old code and it has failed, you are not alone!
So what can we do maximise the longevity, and therefore reproducibility, of our code? That is what the ASG Reprothon 2021 aims to answer! Collating data on code reproducibility for the first time, the Reprothon will provide robust evidence on what factors are determining the longevity of code.
At the end of this we hope to share the results through a blog post with the aim to provide researchers with guidelines to ensure their code remains usable beyond publishing.
- [benefits of this]
- [challenges of this] - not all code is reproducible/not all data structures are.
- [Who you’re doing it for, your audience (may be same as the previous question)]
- I'm working open because....
[Some kind of graphic of the amount of data/how we're splitting up the data] [Project Status Goals:]
The Reprothon is ASG’s very own hackathon! A hackathon is an event where lots of people come together to work on a coding or programming problem. In our Reprothon, we are aiming to determine what factors affect the reproducibility of R code. To do this, participants will work through queries taken from StackOverflow, checking whether or not the accepted answer still works. Each participant will input this into the master spreadsheet along with details such as the package used, date accepted and the purpose of the code. We will track our progress live, and after the closing date, the organisers will analyse this data and produce a blog with our findings.
Where: MSTeams
When: in person 9th June 2021, 1600-1800 BST, and will remain live until TBC
If you want to get involved, we have a quick registration form to complete, which when completed will provide you with your unique Reprothon ID.
The session will be open for all contributors to pop in and out in their own time and will provide a space to chat with other contributors, but also watch real-time as the project moves forward.
*This is all open access so you will be able to take part at your own pace, whenever you want while the data is live. All data is hosted on Github and the website. Any issues can then be brought up on our github by oepning an issue. A hard deadline will be announced before the data is taken down to be analysed. [tutorial on how to open an issue - https://www.youtube.com/watch?v=N4Bl54_Xxis&t=4s&ab_channel=MozillaScienceLab] *
- good WiFi or some way of connecting to the Internet
- R version 4.0+ (if you need to update R you can do that following the guide here
- ability to install packages in R
- familiarity with StackOverlow posts
Optional: A Github Account (this is free). You will still be able to access all the infromation on Github, but if you want to propose any changes or bring up any issues
Contributors are volunteers who offer their time and expertise to the hackathon project. All contributions are central to the project as they move work forward by leaps and bounds!
As a contributor you might:
- add your own unique perspective to the project plans and ideas
- help with the research
- create graphics an mock ups
- contribute code
- contribute mock data for testing out code
You should be aware that all contributions to the project will fall under the project's license and can be reused, remixed and shared according to that license.
We will be using MS Teams and Google sheets record your findings!
After registration you will be designated a subset of the master dataset that you can work through (you don't have to go through them all!) If you want to Collaborate with other contributors you can let the organisers know in advance and we'll group you together, or if you want to collaborate with people in the moment you will have complete freedom to do so! How you split up the work and what roles people take will be entirely up to you!
[insert a gif of working the drop-down menus in the google sheet Thomas has prepared]
Finally, all results will be uploaded to the ASG Reprothon 2021 Github and Website, but you can learn more about the data in advance by checking out the data description already there [links;links]
Part of the “open” nature of the hackathon is connecting and sharing our work, processes an experiences. During the event, we encourage you to post updates– both in the MS Teams chat, and on social media.
You can follow the #reprothon2021 and #hackers2021 hashtags on Twitter to see what others are up to.
If you want a look at what a previous hackathon was like, our founder Francesca Mancini shared her experience in 2017 as a blog
Please make sure to read the general Hackathon Code of Conduct
Please make sure to also read the Contributors Code of Conduct
This work by the Aberdeen Study Group is licensed under Mozilla Public License 2.0