iiinova — where your digital self keeps evolving

Nick Tierney @njtierney.bsky.social · 4h

It sounds like the problem conda-forge is solving is making pkg installation easier and faster? I don't mean to be playing devils advocate too hard, but I am curious - what problem conda-forge is solving? The average R user doesn't really reach for the terminal to manage R pkg installation.

000

Nick Tierney @njtierney.bsky.social · 4h

So that sounds awesome! I can imagine that this is really great in situations where you want to install packages quickly and have all the dependencies resolve. Like if I was an on a HPC system? However, most R users just want to run install.packages("new-package") and be done with it.

100

Nick Tierney @njtierney.bsky.social · 4h

And if you're in Melbourne, or want to get to Melbourne a day early on Sunday 29th November, why not come along to OceaniaR? statsoc.org.au/event-6818002 It's a full day of hacking together open source packages, cleaning data, and generally working on R-focussed projects for social good. #rstats

statsoc.org.au

Statistical Society of Australia - OceaniaR 2026 presented by the Statistical Computing and Visualisation section

010

Nick Tierney @njtierney.bsky.social · 4h

Hey folks! Stoked to be giving a 3 hour workshop "A Gentle Introduction to Targets" at WOMBAT in Melbourne, November 30. Sign up at wombat2026.numbat.space/program/tuto... Early bird registration prices ends Sunday, 18th October! See the full program wombat2026.numbat.space #rstats #targets

Photo of Nick Tierney on a poster for the WOMBAT conference. An ad for his workshop, "A Gentle Introduction to Targets"

072

Nick Tierney @njtierney.bsky.social · 4h

R Universe builds R pkg binaries. You can specify via JSON, e.g., github.com/greta-dev/gr... (setup docs.r-universe.dev/publish/set-...). You can read more about the project here: ropensci.org/r-universe/ Tagging @jeroenooms.bsky.social (maintainer of R Universe) as he has more answers!

github.com

GitHub - greta-dev/greta-dev.r-universe.dev

Contribute to greta-dev/greta-dev.r-universe.dev development by creating an account on GitHub.

010

Nick Tierney @njtierney.bsky.social · 5h

Apologies! I used the wrong words - I meant conda-forge. I have been a bit confused about the use of conda-* for R packages, since I strongly associate conda with Python, and I'm still not really sure what problem conda-forge is solving? Is it building binaries for specific operating systems?

100

Nick Tierney @njtierney.bsky.social · 23h

I'm really curious about this process of more effectively using AGENTS.md and co - where did you learn about your process, and what are these tools from Wes McKinney you use?

110

Reposted by Nick Tierney

Lucy D’Agostino McGowan @lucystats.bsky.social · 14/10/2026

🎙️ @rdpeng.org has "no advice" except this low-key very profound wisdom about innovation

182

Nick Tierney @njtierney.bsky.social · 23h

v. important! I almost always upgrade to the latest version of R. R 4.7 will be out soon enough. I don't really use conda-env, does this create binaries of R packages? Is this different to R Universe? See r-spatial.r-universe.dev/sf for example

r-spatial.r-universe.dev

sf: Simple Features for R

Support for simple feature access, a standardized way to encode and analyze spatial vector data. Binds to 'GDAL' <doi:10.5281/zenodo.5884351> for reading and writing data, to 'GEOS' <doi:10.5281/zenod...

100

Reposted by Nick Tierney

shira mitchell @shiraamitchell.bsky.social · 14/10/2026

blog post: favorite textbooks ? help me find awesome survey statistics textbooks

064

Reposted by Nick Tierney

Fonti Kar @fontikar.bsky.social · 07/10/2026

💡 Are you a current Australian student or recent graduate? 💡 Are you working on an open source software? Consider submitting your software for The Statistical Computing And Visualisation Open Source Statistical Software Award! 🏆 🗓️ Applications closes soon 31 Oct 📌 statsoc.org.au/Statistical-...

064

Reposted by Nick Tierney

Adrian Barnett @aidybarnett.bsky.social · 08/10/2026

I used to regression to the mean everywhere, but not as much these days.

041

Reposted by Nick Tierney

Carlos Scheidegger @cscheid.net · 07/10/2026

So check out this PR from @gordonwoodhull.bsky.social over at Quarto 2. Is that import/export from Word preserving comments? Why yes, yes it is indeed 😮 github.com/quarto-dev/q...

Screenshot of https://github.com/quarto-dev/q2/pull/796

Title: "Run pandoc in the browser: Download as Word/PowerPoint/EPUB/ODT/Typst/PDF, Import from Word and friends, and margin comments"

First section of comment:

quarto-hub could render documents but could not convert them to or from other office formats. This PR adds pandoc.wasm to the hub client. With it, the hub can export the open document (or a whole book) to six formats and import five formats into a Quarto document. All of it runs in the browser, with no server-side conversion.

Comments get their own treatment. Word comments arrive on import as plain spans holding a comment, and they render as margin bubbles anchored to the commented text. That rendering is @vezwork's work on feature/span-comments. This PR carries his three commits with authorship intact (91a2b2743, 95c1364b5, b45470b17), and the import work builds directly on them.

64610

Nick Tierney @njtierney.bsky.social · 10/09/2026

Yeah, so that works! But it doesn't work inside a function - like if I write a function and then use `lm()` inside of it, when I step into `lm()`, even with debugonce(), it doesn't bring me to the lm file - just puts it into the console. I lodged an issue here: github.com/rstudio/rstu...

github.com

Debugging internal R functions in RStudio does not go to source · Issue #18754 · rstudio/rstudio

Debugging internal functions in RStudio, inside a function, does not open up R source files to debug, instead printing them to the console System details RStudio Edition : Desktop RStudio Version :...

010

Nick Tierney @njtierney.bsky.social · 09/09/2026

Hey #rstats folks! How do you debug internal R functions like `lm()` in RStudio? I want it to open up the R script so I can step through it, but currently it prints to the console. I swore there used to be a way to show this. I know positron does this automagically (!) but ppl I teach use RStudio.

121

Nick Tierney @njtierney.bsky.social · 07/09/2026

Dang, Aang. I am a bit ATLA fan, curious to give it a watch, but a shame to hear it didn't manage to hit the same notes.

000

Nick Tierney @njtierney.bsky.social · 04/09/2026

Wow I did not know that about DHH - reading his blog posts is pretty...enlightening? I don't want to post them here. Thanks for that. Won't be using "Hey".

020

Nick Tierney @njtierney.bsky.social · 04/09/2026

Yeah, I got cross w gmail b/c of the ads the tabs (promotions, social, ... etc). I "opted out" of ads by not having those tabs anymore. But now I have an unfiltered firehose of emails. I'm just tired of it all. Proton does seem to really care about security and privacy!

030

Nick Tierney @njtierney.bsky.social · 03/09/2026

What email service do you use/recommend? I saw "hey.com" and was curious about changing away from gmail. Tagging a few people I think have opinions about such things! @johnaugust.bsky.social @hadley.nz @milesmcbain.bsky.social @b0rk.jvns.ca @hspter.bsky.social @rdpeng.org @karthik.io

920

Reposted by Nick Tierney

Ben Harrap @bharrap.bsky.social · 25/07/2026

Has anyone made a gamified version of either drawing DAGs or selecting the correct covariate set from a DAG for the estimand of interest? I'm thinking of something like ggplot battles www.ggplotbattles.dev #rstats

ggplotbattles.dev

ggplot battles

3161

Reposted by Nick Tierney

Ben Harrap @bharrap.bsky.social · 20/07/2026

Any hacky #rstats workaround for the {gt} package? I want the interactive search feature to include columns that are hidden. I tried making a px(0) width column instead of hiding it, no luck. Gimme your hacky ideas!

151

Nick Tierney @njtierney.bsky.social · 21/07/2026

Thanks! I really liked @drob.bsky.social 's blog post on this -http://varianceexplained.org/r/start-blog - I couldn't find his talk, but there was a great infographic he had that was along the lines of "an idea doesn't exist if it isn't online" and had a flowchart of like idea -> draft -> blog post

050

Reposted by Nick Tierney

DuckDB @duckdb.org · 15/07/2026

Lightweight Text Analytics Workflows with DuckDB In this blog post, Petrica Leuca demonstrates how to use #DuckDB for keyword, full-text, and semantic similarity search with embeddings. Learn how to use DuckDB to efficiently perform advanced text analytics in Python: duckdb.org/2025/06/13/t...

0355

Nick Tierney @njtierney.bsky.social · 15/07/2026

I love this video. Tom Scott is a legend.

030

Nick Tierney @njtierney.bsky.social · 15/07/2026

It is great when we can make education materials free! I think there are certain caveats that bolster this argument, which I didn't get to in the slides. One is: if you are teaching free and open source software, I think it makes more sense to have your teaching materials *also* be free and open.

010

Nick Tierney @njtierney.bsky.social · 14/07/2026

I am interested in making an agent to help people transition their websites - it could be bi-directional - would also be interesting to explore {litedown}, too!

000

Nick Tierney @njtierney.bsky.social · 14/07/2026

Just migrated my #rstats blog from blogdown/hugodown to quarto using claude! Was quite straightforward, if a bit iterative. Have a look at the new (nearly identical) website here: www.njtierney.com

njtierney.com

Credibly Curious

1110

Nick Tierney @njtierney.bsky.social · 14/07/2026

So great to see you in person, it's been a minute!

000

Nick Tierney @njtierney.bsky.social · 14/07/2026

Thanks! That's another point I make - free advertising!

010

Nick Tierney @njtierney.bsky.social · 14/07/2026

Related, I am delivering a suite of data science training courses later this year. You can register your interest at training.njtierney.com Course materials are free and online Live courses will be: - Priced competitively - Run online in small groups (up to 6) - Have a $6 scholarship per course

training.njtierney.com

Data Science Training

072

Nick Tierney @njtierney.bsky.social · 14/07/2026

As part of this talk, I wrote a blog post detailing how to share your delivered course as a quarto book, and put it online - you can read this here: www.njtierney.com/post/2026/07... And see the further accompanying github repo: github.com/njtierney/co... #rstats

njtierney.com

How to Put your Course Book Online | Credibly Curious

182

Nick Tierney @njtierney.bsky.social · 14/07/2026

I recently gave a talk, "The value in teaching is not the content it’s the teacher" (njtierney.github.io/talk-teacher...) at "Essential skills for statistics leadership in consultancy and research" My argument was: your time is time higher value than content, so put your content up for free #rstats

njtierney.github.io

The value in teaching is not the content it’s the teacher

4204

Reposted by Nick Tierney

Matt Cooper @nzcoops.bsky.social · 11/07/2026

Was great to cross paths with @njtierney.bsky.social @emitanaka.org (virtually) @aidybarnett.bsky.social (virtually) @analyticaledge.bsky.social And please 🙏🏼 @ tag (below) anyone else that was there as I'd love to stayed connected with everyone but searching @s is ... 🫣🤫

152

Nick Tierney @njtierney.bsky.social · 02/07/2026

Is this Doom like "Doom and Gloom" or Doom like, "that amazing game"?

210

Nick Tierney @njtierney.bsky.social · 18/06/2026

Looks like a great talk! Do you think you will share a recording or slides from this?

100

Reposted by Nick Tierney

Isabella Velásquez @ivelasq3.bsky.social · 17/06/2026

If you're still wondering about the slick new functions in dplyr 1.2.0, join me for a "Working smarter with dplyr 1.2.0" redux with R-Ladies Abuja! 📅 Wednesday, 24 June 2026 🕐️ 6pm WAT 🔗 www.meetup.com/rladies-abuj... See you there! ⚒️ #RStats

meetup.com

Working Smarter with {dplyr} 1.2.0, Wed, Jun 24, 2026, 6:00 PM | Meetup

Discover how the latest features in **{dplyr} 1.2.0** can help you write cleaner, faster, and more maintainable data transformation workflows. This session explores practic

12511

Nick Tierney @njtierney.bsky.social · 10/06/2026

I'm curious on your thoughts on this @hadley.nz - how do you see S7 fitting in to the tibble landscape?

000

Nick Tierney @njtierney.bsky.social · 10/06/2026

Kitchen scissors are The Way. We should put everything in holsters. A bandolier of tasting spoons.

110

Nick Tierney @njtierney.bsky.social · 09/06/2026

I really think the "AI echo-hype chamber" is A Thing. Example: You talk to an AI about Your Idea and it is your hype-person: AI: "Yeah! That's a great idea, and the thing is NO ONE ELSE is doing it, you're really leading the field" Your idea: A pizza slicer holster. (It's still a good idea)

140

Nick Tierney @njtierney.bsky.social · 09/06/2026

Ah I mean when you do dat$var_long_name - if you've defined `$` method to access the data variables then you don't get autocomplete, but dat$var_long_name returns the vector of interest.

000

Nick Tierney @njtierney.bsky.social · 09/06/2026

I do think all the getter/setter and validation benefits are really nice, but then when you do dat$very_long_var_name you don't get autocomplete, which feels kind of like a dealbreaker for me!

100

Nick Tierney @njtierney.bsky.social · 09/06/2026

What are the current views on extending #rstats data.frames with S7 vs S3? I like the idea of attributes on my data frames, so I could do: dat@attribute_1 Instead of an accessor: attribute_1(dat) But then you need to do: dat@data to get the df, unless I define a `$` method?

322

Nick Tierney @njtierney.bsky.social · 09/06/2026

Thank youu! These are my favourite themes!

010