I've generated some interactive quarto-based tables that look really pretty (sortable with sparklines, tooltips, other nice features), but in the end it's hard to beat the ubiquity and user-friendliness of just sending an excel file
I've generated some interactive quarto-based tables that look really pretty (sortable with sparklines, tooltips, other nice features), but in the end it's hard to beat the ubiquity and user-friendliness of just sending an excel file
R scratches that itch for me in the same wayβI've tried many other languages but something about R (probably a mix of it being functional and not overly formal) makes it feel like a "bicycle of the mind" for programming
don't ask how I ended up here #RStats
There's plenty of discourse online about the value of GenAI for OSS, but I had a package fail to install on CRAN due to CRAN-specific config issues (not found on any rhub runners) and used GenAI to help develop a fix in under 24 hours, before a single BDR email came. Now that's progress π #RStats
Fun recent weekends/evenings project: I made a native macOS app to replace a bunch of random useful Python/Perl scripts in my photography workflow.
Introducing: Pentaprism: Karl's swiss-army-knife photo workflow app. I'm really happy with this; it's got tons of UI polish. (1/6)
And to get extra mileage out of the friction analogy, the number of times I've loosened the leash on Codex and Claude just a smidge and have watched it slip and slide and fall on its face while telling me everything is going great tells me a human in the loop will be needed for the near future.
I'm glad to see someone else fall into the same sweet spot. AI's ability to remove friction has made getting projects started and moving 100x easierβthe cost of getting started with a new framework/language or debugging a cryptic error is now nil. My only problem now is juggling all my projects!
I hereby dub this meme format Zooey Dataschanel
seq_along() is indeed the inarguably correct choice
I've got to say no no no #RStats
A visualisation of the population of nine French cities
Introducing population tiles of France.
#rayshader adventures, an #rstats tale
A visualisation of the rivers of the Danube basin
Did this a while back but forgot to share: the Danube basin.
#rayshader adventures, an #rstats tale
2016: Check out my new R package!
-"You should wrap it in a Shiny app so normal people can use it"
2026: Check out my cool Shiny app!
-"You should release it as a package so GenAI can use it"
How the tibbles have turned...
#RStats
Just setting the stage to reintroduce clippy as an agent that wanders your hard drive and collects loose files like aluminum cans
::rayrender below a picture of Ferris Bueller::
oops wrong movie #RStats
The short-timeline-BDR-email-about-another-gcc-version-update-with-some-new-deprecation-to-non-CRAN-upstream-software-fix-pipeline continues #RStats
Average monthly precipitation of the world, 1970β2000
#rayshader adventures, an #rstats tale
A visualisation of the rivers of the Yangtze basin
Today's basin is the Yangtze.
#rayshader adventures, an #rstats tale
And again with more sunshine.
Okay I'm getting this into #rayrender immediately
A visualisation of the rivers of the Mississippi basin
The Mississippi basin.
#rayshader adventures, an #rstats tale
A visualisation of the rivers of the Mekong basin
Another day, another basin: the Mekong basin.
#rayshader adventures, an #rstats tale
Nothing that subtle! I'll give up the answer: the atmosphere is flipped, as the sun is moving into the northern half of the sky (at DC's latitude of 38 degrees, the midday sun should always fall into the southern hemisphere).
That's half of it!
Hint: This is Washington DC, and we're looking approximately northwest.
Writing technical blog posts about my packages is such a slog: not due to the actual writing process (which is fun!), but because I always uncover something weird and unintended in one of my #RStats packages that ends up instigating a massive rewrite somewhere. Can you spot the issue below?
Watch as a massive crack forms in the ice on Lake Erie.
An impressive view captured by GOES-19 earlier on Sunday.
I also found it interesting how much of content from the surrounding clusters has made it into my feed--treating this 2D projection as a distance metric to "regularly served content on algorithmic feeds" seems pretty accurate in my case!
Obviously I'm in the #RStats cluster, but my favorite find is the small fungi enthusiast cluster (unlabeled but annotated here), which seems to have naturally formed a fairy ring (en.wikipedia.org/wiki/Fairy_r...) around its members. Beautiful.
A visualisation of the rivers of the Colorado basin
The Colorado basin.
#rayshader adventures, an #rstats tale