grusonh - Network

As the 2nd day of the Rencontres R #RR2025 is starting, I am releasing the companion post to my lightning talk on "Tips and tools for...

As the 2nd day of the Rencontres R #RR2025 is starting, I am releasing the companion post to my lightning talk on "Tips and tools for co-maintenance of R packages".

https://hugogruson.fr/posts/R-package-comaintenance

In particular, I share two GitHub Actions workflows I have been using to ensure all authors of a package get alerted when a #rstats package is at risk of archival.

This resolves the lottery factor caused by CRAN only emailing a single individual instead of the entire team to let them know about this.

Ever wondered:Will the vaccine I am getting right now be useful for another one of my future trips? 💉 🌐 orHow to create a nice...

Ever wondered:

Will the vaccine I am getting right now be useful for another one of my future trips? 💉 🌐

or

How to create a nice data.frame from data scattered across different web pages? 📊

These two questions look unrelated but I touch on both of them in my latest blog post!

"Politely mapping recommended travel vaccines"

https://hugogruson.fr/posts/travel-vaccinations/

You will hopefully learn more about travel vaccinations and web scraping in #rstats with the rvest and polite packages 📦

My last blog post in preparation for #useR2024, just in time to read it during your trip to Salzburg 😉 What can we learn from concept in...

My last blog post in preparation for #useR2024, just in time to read it during your trip to Salzburg 😉

What can we learn from concept in other domains than #rstats package development? 📦
In particular, web development has the strong challenge to adapt to browsers with various features, can we use their expertise to iteratively improve a software ecosystem?

"Improving Ecosystem Interoperability Iteratively via Progressive Enhancement"

https://epiverse-trace.github.io/posts/progressive-enhancement/

You are now convinced to use #rstats S3 classes but you aren't sure what makes a good class?In my new blog post, I discuss how to choose...

You are now convinced to use #rstats S3 classes but you aren't sure what makes a good class?

In my new blog post, I discuss how to choose a good parent class to inherit from and how you can leverage it to its maximum to simplify your code and improve sustainability and interoperability

https://epiverse-trace.github.io/posts/parent-class/

"Choosing the Right Parent for R Object Classes"

If you're interested in this kind of topic, this is a first taste of my talk at #user2024 in two weeks!

It is so much nicer as a user when an #rstats function works out of the box with both standard data.frames and #tidyverse tibbles, despite...

It is so much nicer as a user when an #rstats function works out of the box with both standard data.frames and #tidyverse tibbles, despite their small differences.

In my new blog post, I detail some minor changes to your functions to make them "compa-tibble" in most cases.

https://hugogruson.fr/posts/compa-tibble/

I end by presenting testing strategies that could help in more involved situations.

I love the renv #rstats package for #reproducibility. But I often receive feedback from frustrated colleagues who get package installation...

I love the renv #rstats package for #reproducibility.
But I often receive feedback from frustrated colleagues who get package installation errors.

I have written a blog post to dive more into the issues that can arise when using renv, where they come from (spoiler: it's not renv's fault), and possible solutions.

"Things that can go wrong when using renv"

https://epiverse-trace.github.io/posts/renv-complications/

New #epiverse #rstats blog post!"Lesser-known reasons to prefer apply() over for...

New #epiverse #rstats blog post!

"Lesser-known reasons to prefer apply() over for loops"

https://epiverse-trace.github.io/posts/for-vs-apply/
The debate on for loops vs apply() often crystallizes around (misconceptions on) performance.
But there are good reasons to prefer apply() to for loops. They revolve around readability and robustness:
1. Clarity of intent
2. Code conciseness
3. No variable leak
4. Pipeability.

Many thanks to @gaborcsardi & @maelle for their very valuable reviews and insights! 🙏

Many thanks to @gaborcsardi & @maelle for their very valuable reviews and insights! 🙏

How come you generally don't have to think about installing system dependencies when testing #rstats packages in the cloud (e.g., on...

How come you generally don't have to think about installing system dependencies when testing #rstats packages in the cloud (e.g., on GitHub Actions)? 🤔

Short answer: it's because the pak package 📦 is doing some amazing magic behind the scenes.

Long answer: jump down the rabbit hole with this new blog post where we explore the various layers of complexity pak deals with to achieve this result:

https://blog.r-hub.io/2023/09/26/system-dependency/

(cross-posted on #Epiverse blog: https://epiverse-trace.github.io/posts/system-dependencies/)

Senior Software Sustainability Engineer #remote #job open for international hiring 🌐. #rstats package development 📦 and ecosystem...

Senior Software Sustainability Engineer #remote #job open for international hiring 🌐. #rstats package development 📦 and ecosystem building on the #Epiverse project. #remotejobs
https://data.org/careers/senior-software-sustainability-engineer/
Come work with me in a supportive environment with passionate, nice and helpful colleagues at data.org.

And I'm even happier because this is my first post on my personal blog!I've been writing many posts lately but they were posted on...

And I'm even happier because this is my first post on my personal blog!
I've been writing many posts lately but they were posted on different organization blogs. I'm proud to be able to also move my personal projects forward 🎉

I am very excited to announce the first CRAN release of the xlcutter #rstats package 📦!https://cran.r-project.org/package=xlcutterWe all...

I am very excited to announce the first CRAN release of the xlcutter #rstats package 📦!
https://cran.r-project.org/package=xlcutter
We all love rectangular or tidy data but we don't always get it. So, how to deal with a folder of excel files that have been used as forms rather than true tables?
The xlcutter package can help you. Its main strength is the fact that templates can be edited by non-programmers. Data producers can directly give you the excel forms and the key to read them! 🤩
https://hugogruson.fr/posts/xlcutter-0-1-0/

@darwin

@darwin

This post is the companion blog post for the presentation I'll give today at #deRSE23: "From disconnected elements to a harmonious...

This post is the companion blog post for the presentation I'll give today at #deRSE23: "From disconnected elements to a harmonious ecosystem: The Epiverse-TRACE project".
If you're at the conference, please come and say hi 😄.
If you're not at the conference, I'll share the recording as soon as it's available.

#rstats S3 system is fun and interesting because it's so minimal that users may not even realize they are using object orientation.This...

#rstats S3 system is fun and interesting because it's so minimal that users may not even realize they are using object orientation.
This means you can *almost* invisibly convert a standard function to an S3 generic.
Why almost? Because you should pay attention to little details regarding documentation, debugging, etc.
Read my new blog post to learn more: https://epiverse-trace.github.io/posts/s3-generic/

There are interesting considerations about how this relates to #codereview. I believe they can compliment each other, with automated tests...

There are interesting considerations about how this relates to #codereview. I believe they can compliment each other, with automated tests reducing the workload of reviewers.

Excited to share this blog post that has been maturing for a while! 🎉 "Ensuring & Showcasing the Statistical Correctness of your...

Excited to share this blog post that has been maturing for a while! 🎉
"Ensuring & Showcasing the Statistical Correctness of your #rstats Package 📦"
https://epiverse-trace.github.io/posts/statistical-correctness/
I present a couple of automated checks you should add to your statistical software testing suite to increase certainty it's computing what it's supposed to.

RT @joinmastodon@twitter.comWhat if I told you that on Mastodon, you can verify yourself, by yourself, for free... 🤫...

RT @joinmastodon@twitter.com

What if I told you that on Mastodon, you can verify yourself, by yourself, for free... 🤫 https://twitter.com/JPFosterson/status/1588957399208976385

🐦🔗: https://twitter.com/joinmastodon/status/1588959499708432384

RT @GullumLuvl@twitter.com@grimmrad@twitter.com @joelmcglothlin@twitter.com I agree there is no good index of all the instances.For science,...

RT @GullumLuvl@twitter.com

@grimmrad@twitter.com @joelmcglothlin@twitter.com I agree there is no good index of all the instances.
For science, I've seen http://scholar.social, http://fediscience.org, http://mstdn.science or more specialized ones like http://genomic.social, http://ecoevo.social.
Also http://scicomm.xyz for outreach I guess.

🐦🔗: https://twitter.com/GullumLuvl/status/1588516086840197123

RT @alx_sim@twitter.comI already boosted this one but will share it again as it is full of really important points to understand about...

RT @alx_sim@twitter.com

I already boosted this one but will share it again as it is full of really important points to understand about Mastodon
https://wordsmith.social/elilla/a-futuristic-mastodon-introduction-for-2021

🐦🔗: https://twitter.com/alx_sim/status/1587166516583153664