Jo Potts

I did recently see an echidna swimming in Richmond (Tas) and do know of a lonely wallaby once trapped in the northern end of bribie island but wow a 1km swim from Bribie to the mainland is hectic for a tiny echidna!

Came for the political comics and stayed for the animal facts, it’s awesome :)

My 2026 office calendar is up! Thanks @firstdogonthemoon.bsky.social

(Pride and place underneath a tapa painting gifted to me from Vanuatu during a visit)

‘Unfathomable loss’: University of Melbourne’s vice chancellor dies at 52 Professor Emma Johnston, the University of Melbourne’s first female vice chancellor, has died less than 12 months after taking the role.

A tragic loss, first and foremost for her children and family, and secondly science, marine biology, science communication, the academic sector broadly 😢

www.theage.com.au/national/vic...

ggridges turns to grudges for predictive text

At the recent Aust stats conference @bharrap.bsky.social showed me some cool ridge plots and I am very very amused (perhaps more than I should publicly admit) that predictive text turns ggridges to grudges.... #rstats #ggplot

Robodebt and robotax wasn’t terrible enough it seems…..

Thanks Cam! My talk seemed to resonate with old and young alike 🤣 it’s was a fun talk to deliver, and the feedback has been really positive. What a great conference!

Big day! Keynote at the Australian Statistics Conference in Perth, talking about my journey and future of stats in the age of AI. Feeling nervous but have my favourite dress on #ASC2025 #stats

Come and join the @significancemag.bsky.social editorial board!

reverse assignment example in R

how did I get this old and not know about reverse assignment in R? #rstats

(arguably if I have lasted this long... maybe it isn't that useful!?)

Yes I did wonder about using a downloaded button. I got a work around to not use any sf (eg coord_sf and use geom_path on a df instead of a shapefile)

I suspect the download button might be more robust but presumably you lose plotly features like the zoom etc?

I am not 100% convinced this is a general issue, rather than a me-issue, but anyways, it seems ok now!

OK nothing like proclaiming in a public space and then quickly finding solution 🤪 If anyone else is stuck on this, sf seems to not work with plotly. If you put your coords in as a df, and use geom_path to plot with geom_tile it renders (and saves output correctly with .png) #rstats soln next comment

Following on from this issue the SEAPODYM dashboard when downloading all the climate data has the same issue🫠 must be a shiny/ggplotly issue(?) or a Joanne-coding-issue(!!!)

analyticaledge.shinyapps.io/tae-ofp-fema...

Good question - yes. In R, I can render the plot, click the 'camera' icon and it downloads and .png with the tiles. Must be something inside the shiny app then....? (I was thinking it was a plotly issue)

I've tried geom_raster... I render using plotlyOutput in the main page....and create output in the server using renderPlotly...

Convinced the answer is something super straightforward (as usual!) but I can't work it out. Is this a plotly thing?

Ok #rstats #shiny hive mind.
When rendering geom_tile to a plotly in shiny, and press the download button the resulting .png is missing the tiles? What gives! Any advice gratefully received

(These data are fake!)

It's long past time for state and federal authorities alike to put an end to Stephen Miller's jerkoff fantasy of masked thugs kidnapping people of color off the street with impunity.

Yes, there are Americans who want this. But they are a woefully small minority. Why are we still letting it happen?

Case_when() really does deserve its own appreciation society #rstats

Quarto 1.8 – Quarto Quarto 1.8 improves brand support, introduces brand extensions, adds HTML accessibility checks, and gives access to execution context.

Quarto 1.8 is available! 💡

New features include integrated Axe-core checks for HTML documents, access to environment variables via code cells, light/dark mode logos and colors, and Brand Extensions for sharing your org's brand.

Learn more: quarto.org/docs/blog/po...

#Quarto #Python #RStats

Missing Values were dropped;
Mean Imputation was applied;
Data available on request;
Code available on request;

#rstats

It’s really not the best infographic hey

💯. How anyone can claim this was not anticipated is lying or living on another planet

It's bizarre. I can think of at least 3 scientific publications that predict (i.e., model) the uncontrolled explosion of deer across Tas. None of them were referred to in the report. weirdddd

'Not anticipated': New survey shows deer population boom in Tasmania The first survey for feral deer in Tasmania in five years reveals an unanticipated increase of nearly 20,000 to the population.

“No anticipated” except by all of us that have been saying for ever that the #feral #deer management plan is pathetic. 🌎 #wildoz

www.abc.net.au/news/2025-10...

Are you a #maths teacher? We are delighted to share a brand new set of materials for teaching #statistics for students aged 11-16, and we'd love your feedback on them! #ITeachMath #MathsWeekScot rss.org.uk/resources/ed...

Hey #rstats friends! 👋

Only one week to go until my "Introduction to R Packages" tutorial at #WOMBAT2025!

This is a gentle introduction to making R packages, going from initial code to `devtools::check()`

Still plenty of spots available :)

Sign up at: wombat2025.numbat.space

eight years as a statistical consultant and ecologists still manage to collect weird data that I'm not immediately sure what to do with

(it's fun though!)

hmmTMB: Hidden Markov Models with Flexible Covariate Effects in R by Théo Michelot <p>Hidden Markov models (HMMs) are widely applied in studies where a discrete-valued process of interest is observed indirectly. They have for example been used to model behavior from human and animal...

The hmmTMB paper is finally out in the Journal of Statistical Software!

An R package for hidden Markov models with random effects, flexible spline-based covariate effects, and fast inference using TMB or Stan.

Check out the GIthub repository for more examples.

doi.org/10.18637/jss...

output from a GAM in the linked essay

Simon Wood, the GOAT of generalized additive models & creator of the mgcv #rstats package, has an Annual Review of Statistics essay on GAMs, available open access #statssky #mlsky

www.annualreviews.org/content/jour...

This looks like a great further education opportunity #rstats #stats

ONE-YEAR MSC (RESEARCH) SCHOLARSHIPS IN SPATIAL STATISTICS AT THE UNIVERSITY OF OTAGO

statsoc.org.au/forum-job-va...