Go get an ORCID iD!
Published:
What is an ORCID iD? Should you get one? What can you do with it? If you are a researcher or a scholar it might be worth you signing up for one. Here’s why …
Blog posts
All posts should be bite sized chunks of information that will take less that about a minute to read. Posts are mostly focussed on R, Git, GitHub, and provide tips for better reproducibility of work, collaboration with other, and transparency of ideas to foster open science principles.
2025
Enhancing an R package
Published:
So you want to create an R package? Well there are a ton of resources available to help with this, the best being the official book from the Posit folks, R Packages (2e). However, after the initial set up, organization, and implementation of the package there are several other related components you should think about including to make it function, look, and feel polished!
Versioning R packages
Published:
You’ve reached the point in which you want to officially start to version your R package. How do you go about doing this? When should you version? How often? What versioning scheme should you use? How do you document the changes?
2024
Semantic versioning
Published:
The official site for Semantic versioning is available in many languages indicating, in part, how important this subject is across the world for software development. It provides a clear, standardized way to communicate the nature of changes between releases, which helps manage dependencies and predict the impact of updates. By using a version format like MAJOR.MINOR.PATCH, developers indicate whether an update is a backwards-incompatible breaking change (MAJOR), a new backwards-compatible feature (MINOR), or a backwards-compatible bug fix (PATCH).
Automate versioning for R data packages
Published:
Do you find yourself downloading data from a server over and over again to make sure you have the latest data? Then struggle to work out what’s changed from the last time you downloaded it?
Error handling with tryCatch()
Published:
Error handling is not something scientists typically incorporate into their code. However, there a situations in which incorporating such a thing can save you hours of headache!
Add release tags in git
Published:
Scenario: You have a repository in which you’ve been using for a while. You use it for your scientific work and have published papers based on it. However, you never created releases of the repo corresponding to each publication and now you have no way of reproducing the work used in your publications. What can you do?
Using main, dev, feature branches
Published:
When using any type of versioning software, either collaboratively or not, there are a few best practices that should be followed regarding branching strategies. This will be a short and sweet entry. No fluff!
Using Cron jobs
Published:
Simply put, cron jobs are jobs/tasks (pieces of executable code, on Linux or Mac) that are run on a predefined schedule. For example,
Live testing regular expressions …
Published:
As some point in your career you’ll find yourself needing to parse large amounts of alpha numeric data and selecting sections which match certain string patterns. For example, maybe you are parsing the contents of a webpage looking for hyperlinks, or summarizing the contents of a log file, or searching for values of parameters in a file heavy with text.
Use Github Actions to …
Published:
If you are an R package developer, and you host your code on GitHub, there a few tasks that you will need to do repeatedly to ensure user experience is pleasant and that your code is stable.
Use DockerHub to store and retrieve your containers
Published:
So you have just built your first docker image (or podman image, but more about that in a minute). And it runs! And you are excited! So you start to tell people how cool you are, that you can run “stuff” inside a container. But then are quickly brought back down to earth when you realize that you don’t know how move this image to other platforms, like the cloud. It resides “somewhere” on your local machine, in a location that isn’t obvious.
Run your model in a container
Published:
Containers aren’t new, but to the scientific community (as of writing), they kinda are! So what are containers? Well, simply put, they are a running instance of an image! Oh Right … totally get it!
Stop using the library() function
Published:
Yes, i know, you love using the library() function in R! But it can be a real problem for several reasons:
Use markdown on GitHub
Published:
Markdown for GitHub is the language of choice when creating issues, pull requests, using discussions, or editing your README file on GitHub.
Include DESCRIPTION file in all R projects
Published:
The DESCRIPTION file is a necessary staple for all R packages. Among other things (which we won’t get into, since this isn’t a post about creating R packages) it is the location for you to list all dependencies of your package. So when your package is installed the user also gets all the dependent packages installed too! Nice and tidy!
Use reproducible examples (in R) when asking for help
Published:
Regardless of the language you are coding in, at some point, there will come a time when you just need to ask for help. When you do, you should always include a reproducible example to help explain your problem. Failing to do so will just irritate readers and limit the number of helpful responses.
Delete your last commit from GitHub
Published:
We’ve all done it, you’ll be happily commiting and pushing to GitHub from either the command line or, more than likely, your favorite IDE when you panic and realize that you just mistakenly pushed something to GitHub. You start sweating, and frenetically start googling how to reverse this. Well it turns out it is pretty simple.