Nicola Romanò @nicolaromano@qoto.org

Because they bring me such joy, I will share with you all the software testing videos I share with my Software Design and Development students.

Video 1/3: “We don’t need user testing! We already know our users”

"It's difficult to judge the quality of someone's work," you tell me, "especially these days when so much is being published every day."

And you're right. Good work in many areas or research is difficult to pick out, obscured by a rising flood of self-promoting bullshit. /1

#statstab #64 Using tidybayes with the posterior package by @mjskay

Thoughts: Going #bayesian means a lot of wrangling the posteriors to get the info you want. A crucial skills I'm still learning, but invaluable.

#bayes #rstats #tidybayes #brms

https://mjskay.github.io/tidybayes/articles/tidy-posterior.html

Reminder: MIT Libraries have been out of contract with Elsevier for four years now.

Bravo MIT!

More like this please. Paying the ransom only encourages the robber-barons to crank-up the ransom price the next year. We need to break the cycle. Don't feed the monster.

Professors: you can use yt-dlp to download YouTube videos and add them to directly your presentations.

I've seen so many lectures get derailed by an embedded YouTube video breaking in PowerPoint. Or they minimize their presentation, open up a browser and navigate to YouTube (revealing embarrassing personal recommendations), then make a lecture hall full of people sit through an ad for some fly-by-night home security system or whatever.

https://github.com/yt-dlp/yt-dlp

#education #teaching #academia

Dans "La vie de laboratoire", classique de #sociologie des #science de Bruno Latour, une note en bas de pagelnous rappelle que le web scraping et les techniques de #MachineLearning ont finalement assez peu évolués en 45 ans...

#shitpost

@nicolaromano Exactly! The specific language is less important than the ability to write dependable and understandable code. R wasn't developed as a general purpose programming language and it's not fair to judge it that way (Python on the other hand...). Even then, the R community adapted the language and practice to better support making standalone applications, to make documentation and testing easier, etc. There's at least movement in a good direction. Excel just keeps adding less and less useful features without addressing any of its core shortcomings. Excel doesn't improve, it just gets bigger and its UI keeps churning.

It's only Monday and I'm already done with this week. I watched a work issue turn increasingly acromonious over either miscommunication, misunderstanding procedure, or being too wed to a tool to admit it's problematic. I drafted about half a sentence in chat, then stopped because I didn't think I was adding anything necessary and didn't want to make conflict any worse.

I'm increasingly convinced that Excel spreadsheet files should self-destruct or become permanently read-only after 90 days. That's plenty of time to migrate data into a proper database or to migrate calculations into a programming language amenable to auditing, verification, and change control. Excel should be treated as an attractive hazard with a time limit on how long an unverifiable and hostile-to-revision-control worksheet should be suffered to live.

Sadly, these two thoughts are related. Spreadsheets in their current form are too dangerous to be used for engineering work.

I finished my summer internship this week! I spent 4 months working with a team of biologists and statisticians at The Jackson Laboratory, building visualization software for their research needs. The lab I worked with focuses on analyzing genetics, the microbiome, and addiction-related traits. Here's an overview of some of the EDA tools I built for them in #Python ☺️ #genetics #biology #addiction #dataviz #computerscience #research #phd

If you're doing cool work with data vis + bio, lmk!

Black hat hackers demo keycard vulnerability in millions of hotel rooms worldwide:

"They merely tap those two cards on a lock, the first rewrites a certain piece of the lock's data, and the second opens it."

Interesting read. Time to change the hotel lock tech? Again?

https://www.wired.com/story/saflok-hotel-lock-unsaflok-hack-technique/
#Security #Hacking #Travel

A small rant about zombie ideas and the tendency to keep looking for modifications of study methods to avoid concluding that a null result is really null. http://deevybee.blogspot.com/2024/03/just-make-it-stop-when-will-we-say-that.html
#research #nullresults #laterality #handedness #publicationbias

I've been writing serverside SQLite applications for several years now and I still picked things up from this article, which is extremely good. https://kerkour.com/sqlite-for-servers

@futurebird Someone once told me "It's not Kafkaesque to wake up an insect. If you wake up an insect and your first thought is that this might make you late for work...*that* is Kafkaesque." That completely changed my perspective on a lot of things.

Excellent piece from Grégory Miras on why new tools that change #accents in real time are harmful and problematic - they erase diversity - and make us less able to appreciate and listen to that diversity.

#linguistics #AccentBias

https://theconversation.com/why-ai-software-softening-accents-is-problematic-197751

Metrics reloaded. "A comprehensive framework guiding researchers in the problem-aware selection of metrics."
A must-read for everyone doing #ImageAnalysis in #biology
https://arxiv.org/abs/2206.01653

One of the largest science funders, the Bill & Melinda Gates Foundation, will cease paying journal “article processing charges” and instead asks funded researchers to publish their work as preprints. This is fantastic. The costs of the current publishing system drain research funds and exclude too many scientists solely due to financial constraints. Funders are in a much better position to rock the “publishing” boat than researchers.
https://gatesfoundationoa.zendesk.com/hc/en-us/articles/24810787662100-Policy-Refresh-2025-Overview

#academia #OpenScience #research

Have to output a bunch of dense scatterplots to PDF/SVG, but don't want the filesize to bloat with thousands of symbols to draw the points?

Learn from my mistakes instead of making your own:

```r
geom_point <- function(...) {
ggrastr::rasterise(ggplot2::geom_point(...), dpi = 300)
}
```
#rstats

I'm reviewing code that I wrote more than 1yr ago for a manuscript. I had a script with ~1000 lines and I just reduced it to 200 by using `purrr::map()` and my custom functions. It's crazy how if you just keep coding you eventually get better. I could go to 100-150 but I need to move on so I'll just leave it as is.