If you know when the packages were up to date you can install packages from that date, same with R. rig and rv should help.
06.03.2026 18:15
π 0
π 0
π¬ 0
π 0
If you know when the packages were up to date you can install packages from that date, same with R. rig and rv should help.
We used to be a proper country.
#rstats RIP, John Fox
@yihui.org just published this lovely tribute to John Fox and his work
yihui.org/en/2026/02/j...
Although it looks like they ask you to email support if you file an issue? github.com/CenterForOpe...
Yeah same, probably will use this going forward github.com/CenterForOpe...
Perhaps some kind of a big or issue tracker could be useful so we can have some sense of magnitude of these issues. I've been having similar issues intermittently and am now avoiding clicking on OSF links because I know the experience will just be bad.
Maybe try on the Stan forum, you'll have more space to explain the problem, give a reproducible example, etc.
Good one. Obviously latter 100% other opinions are wrong.
This is why I'm on social media. Let's make it happen!
- make vs targets
- one qmd file vs many r scripts
- renv vs anything else really
- feeling grumpy over all the new LLM features & packages vs accepting this is where we are now
Pick your arbitrary controversy and let's go
This is very cool, and a nice reminder to all of us that you don't always need N=9999 for psych experiments.
True, I have the same experience where sometimes pages load, sometimes they don't. I put all my materials on zenodo (works like a charm) these days but still use OSF for preprints.
This sounds really cool. brms 3.0!
It's very frustrating indeed. I think it'd be good if we reported all of these to support@osf.io so maybe things can be better in the future.
Is this a "slow website" or "slow / difficult user interface" issue? I am finding it's "both" for me on the OSF but wonder about others' experiences.
Should probably change the title eh?
I can't believe this was 10yrs ago! The 'winds' blog post is a modern social science classic.
It's also a very sensitive issue right now because in the USA superficially "open science" arguments are being made by parties advocating eating worms, injecting raw milk/bleach, etc. So this emotional/political baggage is making the "basic science" argumentation harder at the moment.
To be honest I also very much dislike being told what to do π₯΄
Maybe there is / could be some interesting work in this space on predicting attitudes to data sharing et al.
I had this vibe too but my US experience is prehistorical.
There usually is very little cost (ignoring possible environmental chaos?) in trying it out, and an LLM goes beyond writing, e.g. "find the most important flaw in my argument and help me build a better one" etc.
Indeed. It's also possible if not likely that arguing your position with someone, human or machine, might help your paper. www.frontiersin.org/journals/art...
Where do you find all these folks / listservs? I must live in a bubble because I struggle to find people who even really think twice about data sharing. @lakens.bsky.social is this a NL thing?
www.theregister.com/2026/02/16/s...
I like it when an authors voice comes thru in a piece of writing, beyond whatever (non)factual points are being made.
Sounds about right. This is why I'm happy to see funders already recommending (and more) such activities. I'm also eagerly awaiting for them to require open access publishing but not fund fool's gold OA charges.
My guess is that its uploading open data & preregistration because they are things you need to do, and many don't like to do things they're not used to doing, or at least being told what to do.
Great post worth reading.
In working on a related comment earlier today, it occurred to me that we donβt discuss often enough how the fields most concerned about science (and most involved in reforming it) are the ones most wedded to NHST as the arbiter of truth.
I'm so excited to announce the first release of my newest #Rstats package, {adrftools}! This package facilitates estimation, visualization, and testing for the causal effect of a continuous (i.e., non-discrete) treatment.
𧡠1/10
#statssky #episky #causalinference
osf.io/preprints/ps...
Interesting case where authors use simulated data (because they couldn't get the original data) to reanalyse a study and conclude that its findings were spurious.
The OSF does not currently give views / downloads via the API but we're figuring out whether this might be possible in the future.
Hi Philip, we have vuorre.com/psyarxiv-das... & psyarxivdb.vuorre.com running for PsyArXiv. Both are open source, and since we have the same host it'd be easy to adapt to SocArXiv (basically by changing psyarxiv -> socarxiv in a couple places in the code).