Saras Windecker

“First, reduce “statistical machismo”, or the tendency to adopt and advocate for more complex approaches for their own sake rather than because they are the best tool for the job”… I mean…

I am glad to see this piece out in Frontiers. Ecologists must consider why we do what we do in the 1st place!

(PDF) “But I can’t preregister my research”: Improving the reproducibility and transparency of ecology and conservation with adaptive preregistration for model-based research PDF | 1. Preregistration is an open-science practice which aims to improve research transparency and mitigate questionable research practices, like... | Find, read and cite all the research you need o...

What was missing:

The concepts here tie well into principles of open science, but despite talking about this approach improving reliability and credibility, the link was never explicitly made.

I'd recommend this new preprint from Elliot Gould et al: www.researchgate.net/publication/...

What we liked:

Good Habit 1: validate with simulated truth. Less common that it should be, and very useful to avoid mission creep by making you more intential about your model.

5/

Which led us to wonder, who was this paper for?

It didn't map on clearly to how we model, and unfortunately assumed a lot of familiarity with the experimentalist’s workflow that many of the team did not have. It would have benefited from a clearer worked example.

4/

Within our group we identified a third workflow that is a mix of the two described:

(3) A workflow that is devised first via theory, but it is then fit to data.

we decided this is both the most common for us and a more natural fit for thinking like an experimentalist.

3/

The author sets up two modelling workflows:

(1) "An analytical workflow that begins with data, proceeds through code and ends with a particular conclusion", and

(2) "A theoretical workflow that begins with ideas, proceeds through code and ends with a general conclusion about the concepts".

2/

Yesterday in journal club we read this paper onlinelibrary.wiley.com/doi/10.1111/.... And we had thoughts!

1/

We have the publication associated to the BlueCarbon library!

Very happy with the review process and transparency of @joss-openjournals.bsky.social
Please cite this article if your are one of the more than 1700 downloads of the package :D

bsky.app/profile/joss...

Yes good points. It wouldn’t break the old code but (perhaps worse) it would redefine what the output value means. As I write this I think it occurs to me to add another item in the output list with the new value which would mean a user to keep using it as before as well

What's the best way to go about this? I don't want to break old code, but this is based on updated best practice.

Is releasing it as a new version with an update in the NEWS enough?

@njtierney.bsky.social @bharrap.bsky.social ...

R dev friends -

I am updating a package that implements an analytical method. The update will mean a fundamental change in the outputs compared to the current version of the pkg (but is based on consult with experts).

Do you run or participate in a code club at your institution, or are you interested in starting one but don’t know where to begin? @cecibaldoni.bsky.social and I are leading an ‘unconference’ discussion at @sortee.bsky.social’s virtual conference on October 16 - come share some tips and ideas!

Very excited to be travelling to Melbourne to speak at WOMBAT2025 about community building & code review. Really looking forward to reconnecting with old friends and new. 🙌

From policy to practice: progress towards data- and code-sharing in ecology and evolution | Proceedings of the Royal Society B: Biological Sciences Data and code are essential for ensuring the credibility of scientific results and facilitating reproducibility, areas in which journal sharing policies play a crucial role. However, in ecology and evolution, we still do not know how widespread data- and ...

Our latest #SORTEE-led paper is out now in #ProcB!

We reviewed data/code-sharing policy clarity, strictness & timing across 275 EcoEvo journals

We also worked with #ProcB & #EcologyLetters to assess initial compliance with sharing mandates. Huge team effort! 👉 doi.org/10.1098/rspb...

🙌 wonderful work leading this @asanchez-tojar.bsky.social, @eivimeycook.bsky.social, and @nickmoran-sci.bsky.social.

Transparency with open data and code sharing is a critical first step in the quality control pipeline for research code.

very excited to be joining the Research Software Asia Australia conference this week #RSAA25

Speaking tomorrow "Piloting peer code review in a research consortium community of practice". Happy to be meeting with like minds 🥳

Computational image analyst (m/f/d) The Max Planck Institute for Molecular Biomedicine in Münster, Germany, has an opening for a

My lab @mpi-muenster.bsky.social is looking for a computational biologist with a passion for imaging and spatial biology. A staff scientist position with long-term perspective! Apply and spread the word 🙏 jobs.mpi-muenster.mpg.de/jobposting/0...

A couple throwback pics from past scientific lives courtesy of downloading my twitter archive ~

PhD swamp life 👩‍🎓🌱, and a classic from spotlighting for ringtail possums with Nick Golding, who I still have the pleasure of working with today 🤩

ReSA Director Michelle Barker introduces the International Research Software Conference during RSECon25.

We’re excited to announce the International Research Software Conference (IRSC)!

7–8 September 2026 | Sheffield, UK & online | Co-located with #RSECon26 @society-rse.org

Learn more: www.researchsoft.org/irsc/

For updates, join: preview.mailerlite.io/forms/778129...

Image by @danielskatz.org

rOpenSci Package Checks Check whether a package is ready for submission to rOpenSci's peer review system.

4/ lastly, ropensci's very own pkgcheck action.

This one is really interesting as it gives you a summary of points you've either hit or missed and summarises them in a new issue in your repo. Subsequent pushes show up as comments in the same issue thread.

docs.ropensci.org/pkgcheck/ind...

actions/examples at v2 · r-lib/actions GitHub Actions for the R community. Contribute to r-lib/actions development by creating an account on GitHub.

3/ pkgdown workflow

Not related to the ropensci submission, but very handy if you have a pkgdown website.

I adapted from the examples here github.com/r-lib/action.... It is great knowing the website is up to date with changes I've pushed and I didn't have to separately re-publish the website.

GitHub - codecov/codecov-action: GitHub Action that uploads coverage to Codecov GitHub Action that uploads coverage to Codecov :open_umbrella: - GitHub - codecov/codecov-action: GitHub Action that uploads coverage to Codecov

2/ I also set up an action to access my code coverage and report it on the readme. It's not as high as I'd like yet, but it's getting there!

To get this to work I first had to link the repo on codecov.io.

github.com/codecov/code...

actions/examples at v2 · r-lib/actions GitHub Actions for the R community. Contribute to r-lib/actions development by creating an account on GitHub.

1/ continuous integration

Compared to the last time I set up continuous integration, things are SO EASY now. I used the following:

github.com/r-lib/action...

a man wearing headphones says action in a snl advertisement ALT: a man wearing headphones says action in a snl advertisement

Going to start a series of threads on what I've been learning as I get an #rstats 📦 ready to submit to @ropensci.org.

Every time I think the package is "ready" I find another checklist I haven't been through, but it's making it all the better.

This first thread is on github actions!

Stopped using "X" years ago and have finally made a switch. Hope to reconnect with folks and touch back in to the social land to share and talk about open science, research software, ecology, and public health. 🙌