iiinova — where your digital self keeps evolving

Reposted by Felipe Fontana Vieira

Isabella Velásquez @ivelasq3.bsky.social · 16h

{renv}, {rix}, rv, uvr, ir... so many new #RStats tools! I wrote up a blog post if you've want to unpack (haha) how they work and how they are different. Inspired by @rpodcast.bsky.social's wonderful Data Science Lab episode (thank you, Eric!). Check out the post here: rworks.dev/posts/r-mana...

rworks.dev

{renv}, {rix}, rv, uvr, ir: An explosion of R package and environment managers – R Works

There has been an explosion of package/environment/platform managers in the R ecosystem recently. Here, we explore {renv}, {rix}, rv, uvr, and ir.

25012

Reposted by Felipe Fontana Vieira

Nicola Rennie @nrennie.bsky.social · 17/09/2026

I love overly honest READMEs

Screenshot of GitHub README saying known issues: coding style only a mother could love

19920

Felipe Fontana Vieira @felipefv.bsky.social · 17/09/2026

I am not sure if I’m misunderstanding. But you can always work with a specific quarto extension? I always use an APA extension and done. At the end I have latex, pdf, word, html all accordingly. There may be something annoying about a notation/table here and there, but in general it is smooth.

000

Felipe Fontana Vieira @felipefv.bsky.social · 17/09/2026

There were some complaints and the development version is no longer doing that. You get only a warning now

100

Felipe Fontana Vieira @felipefv.bsky.social · 15/09/2026

Why is it that every time I ask someone in academia how they are doing, they reply that their paper is going well

000

Reposted by Felipe Fontana Vieira

Danielle Navarro @djnavarro.net · 15/09/2026

In her ongoing quest to convince #rstats users to stop invoking the cursed "rm(list=ls())" incantation at the top of their analysis scripts, she has released sessioncheck 0.2 on CRAN. Mostly unchanged, some bug fixes, some new reporting features that are most useful in the regulatory context

blog.djnavarro.net

Release: sessioncheck 0.2 – Notes from a data witch

In which the author realises she doesn’t know how to write package release posts

4439

Reposted by Felipe Fontana Vieira

Center for Open Science @cos.io · 10/09/2026

🎉 COS has established an independent subsidiary organization in Brussels! While the Belgian organization is legally independent from COS in the US, it will not operate separately—it's part of one integrated, global COS organization with a shared mission, strategy, & brand

cos.io

A New Chapter for COS in Europe

The Center for Open Science (COS) has officially established an independent subsidiary organization in Brussels, Belgium. This presence gives COS flexibility to engage with European institutions, organizations, funders, and communities.

02815

Felipe Fontana Vieira @felipefv.bsky.social · 11/09/2026

I think so, not sure, but I will ask and see where it would be available. It won't be anything new for you thou 😅

111

Felipe Fontana Vieira @felipefv.bsky.social · 11/09/2026

Credits go to people like @brodriguesco.bsky.social for {rix} and @jgeller1phd.bsky.social for all feedback/help. This is something I have been doing on my free time just because I found it useful (and spent a lot of time learning), realized many people didn’t know about it, and decided to share it.

150

Felipe Fontana Vieira @felipefv.bsky.social · 11/09/2026

Next week I will give a free webinar about how I have been using Nix (by itself) but mainly through {rix} for my current articles. If anyone is intested, my plan for the structure of the webinar (in the comments):

11913

Felipe Fontana Vieira @felipefv.bsky.social · 11/09/2026

Never interacted with you, but very much this :/ Especially annoying because I mainly follow people in research, where I (at least wanted and expected) to see something different, but not really the case.

110

Felipe Fontana Vieira @felipefv.bsky.social · 08/09/2026

Multi-threading and single-threading is also a relevant thing to be discussed when talking about computational reproducibility. I have played with this during the past months to check for speed differences and many surprising data generation/estimates differences

011

Reposted by Felipe Fontana Vieira

UGent Open Science @ugentopenscience.bsky.social · 08/09/2026

Want your R analyses to run the same way across computers, OS, and time? Join our webinar on reproducible computing in R with {rix}, a package that makes the power of Nix accessible to R users. tinyurl.com/3555uwza With @felipefv.bsky.social, co-author of a {rix} tutorial #rstats #Reproducibility

tinyurl.com

Microsoft Virtual Events Powered by Teams

0106

Felipe Fontana Vieira @felipefv.bsky.social · 05/09/2026

The bro is saying that about chess on repeat

000

Felipe Fontana Vieira @felipefv.bsky.social · 02/09/2026

Of course there is website in Germany with that precise information

000

Felipe Fontana Vieira @felipefv.bsky.social · 02/09/2026

When did Oktoberfest started to happen mainly in September?

100

Felipe Fontana Vieira @felipefv.bsky.social · 02/09/2026

Ohh, thanks, looks nice indeed. Will check it out

030

Felipe Fontana Vieira @felipefv.bsky.social · 02/09/2026

Thanks @markfabian.bsky.social, but OSF is fine now and I just downloaded it. Unless there is a new version? Let me know

020

Felipe Fontana Vieira @felipefv.bsky.social · 02/09/2026

Nice, thanks! I saw that somewhere in one of your posts, but still haven’t had the time to actually read it

130

Felipe Fontana Vieira @felipefv.bsky.social · 02/09/2026

Of course, the scale is largely targetting the second claim and one can always have a 5 while still believing you could be happier next year, but yea there is just no concrete mapping

120

Felipe Fontana Vieira @felipefv.bsky.social · 02/09/2026

My reasoning in every questionnaire I answer (e.g., 1 to 5): 5 is a claim about the world ("max. happiness, no further happiness is possible") rather than a claim about my position relative to other plausible answers. That is why I never say 5.

170

Felipe Fontana Vieira @felipefv.bsky.social · 02/09/2026

Could be nice to develop something that uses AI but I honestly feel like a boomer when it comes to that so…

110

Felipe Fontana Vieira @felipefv.bsky.social · 02/09/2026

Clinical trials have something like that as far as I know! Across other fields we usually have similar stuff that are tailored to data analyses/stats, but only at the level of the files (snakemake etc). Definitely a good idea. I will check more if we have an automatized version of that in the wild.

110

Felipe Fontana Vieira @felipefv.bsky.social · 01/09/2026

Good work and it was nice to meet you!

110

Felipe Fontana Vieira @felipefv.bsky.social · 01/09/2026

Yea, I don’t disagree! I think my main concern at the moment is just the lack of consensus/clarity/expectation on what constitutes “good enough” in this pipeline-reproducibility (and reproducibility in general)

110

Felipe Fontana Vieira @felipefv.bsky.social · 01/09/2026

Hopefully it was not too bad 😅 About the take, do you mean in terms of the organization/tidiness of your code?

110

Felipe Fontana Vieira @felipefv.bsky.social · 31/08/2026

Imo they should even get two degrees

060

Felipe Fontana Vieira @felipefv.bsky.social · 31/08/2026

And then breaking news: the mediation in those cases ends up just being two regressions instead of one

250

Reposted by Felipe Fontana Vieira

JsonGeller @jgeller1phd.bsky.social · 30/08/2026

Didn't everyone p hack in the early 90s?

041

Felipe Fontana Vieira @felipefv.bsky.social · 30/08/2026

Yup, the water in Zurich is as clean as they say

000

Felipe Fontana Vieira @felipefv.bsky.social · 27/08/2026

I knew I should have asked for the autograph when I had the chance

030

Reposted by Felipe Fontana Vieira

Felipe Fontana Vieira @felipefv.bsky.social · 25/08/2026

Anyone into #reproducibility who'd like to bring CI/GitHub Actions into their R workflow? I'm looking for some people to try out something I'm working on that rebuilds your Nix environment (e.g., built via rix). It is genuinely meant to be a simple tool.

001

Reposted by Felipe Fontana Vieira

Pieter T. L. Beck @ptrbck.bsky.social · 25/08/2026

We have written a statement in support of Ghent Uni's decision to suspend Nathan Cofnas. We resist attempts to frame this as a punishment for "whistleblowing" or censorship of "controversial research". We invite all academics who share our views to sign our letter. forms.gle/qwN8xzEM5C7Q...

forms.gle

Open letter on the suspension of Nathan Cofnas at Ghent University

We, colleagues of Nathan Cofnas - supported by many colleagues internationally - want to respond to the negative and incorrect framing of Ghent University’s decision to suspend Nathan Cofnas. With th...

421493642

Reposted by Felipe Fontana Vieira

Eiko Fried @eikofried.bsky.social · 21/08/2026

Defending Cofnas, who had written outrageously racist stuff for years (e.g., that in a meritocracy there would be next to no Black professors because they have lower IQs), is an odd hill to die on. Yet the 100-plus academics who signed this letter are clearly willing to die on it :/

1020248

Felipe Fontana Vieira @felipefv.bsky.social · 25/08/2026

001

Reposted by Felipe Fontana Vieira

Michel Nivard @michelnivard.bsky.social · 20/08/2026

🚨6 weeks till World Congress of Psychiatric Genetics ! We (me and @drmoab.bsky.social ) are organizing a 5km fun run. Conversational pace, coffee after and plenty of time to change and get to the venue. Let me know if you’re down for this and we’ll build an WhatsApp group! 🚨

2105

Felipe Fontana Vieira @felipefv.bsky.social · 19/08/2026

You have someone developing uvr: nbafrank.github.io/uvr/#benchma... But, as mentioned by @modrakm.bsky.social, I do believe the best we have right now is combining Nix (for reproducible-builds/reproducibility) and Docker (for portability)

010

Felipe Fontana Vieira @felipefv.bsky.social · 13/08/2026

Yes, this is very much true and got me confused in the beginning. A few books make this explicit or at least hint more clearly at it (e.g., latent variable models and factor analysis), but the large bulk of articles out there do not.

010

Felipe Fontana Vieira @felipefv.bsky.social · 07/08/2026

Read it today and liked it a lot! Also, Figure 1 is aesthetically very pleasing 😆

120

Reposted by Felipe Fontana Vieira

Ben Hanowell @hanowell.me · 04/08/2026

Tracing the usage of the term "latent" in science is frickin WILD, dude. It looks like it all goes back to (arguably) the start of the scientific method, Bacon's 1620 Novum Organum, in which he discussed the processus latens.

262

Felipe Fontana Vieira @felipefv.bsky.social · 07/08/2026

Briefly: I would say it is about an ontological commitment. Within statistics, I see latent variables as just a modelling choice (i.e., random effects etc). Within psych/metrics, there is a strong ontological commitment, where some ppl are more careful than others w.r.t it. What is your conclusion?

110

Felipe Fontana Vieira @felipefv.bsky.social · 03/08/2026

Sometimes I wonder if we have all been using the same set.seed() value this whole time

001

Reposted by Felipe Fontana Vieira

Eiko Fried @eikofried.bsky.social · 15/07/2026

Join our Many-Analysts Simulation Study—yes, you read that correctly—accepted as Stage 1 Registered Report in (AMPPS) @psychscience.bsky.social. We're trying to find out how researchers' simulation design choices shape methodological conclusions. Further info: osf.io/8vcxh/files/...

25132

Felipe Fontana Vieira @felipefv.bsky.social · 15/07/2026

This looks cool! Is the email correct?

210