Lisa DeBruine

Metadata!!!

I’m not sure how Scientific Reports can ever mend its reputation now.

Introduce yourself with five concerts you’ve seen

Tori Amos (first gig 1993)
Melt Banana / The Sun Ra Arkestra
Pussy Riot
Boygenius
Young Fathers (5 times, but the 2015 Art School gig was stunning)

(I saw the pixies in Paris for my 40th birthday, but I don’t want to repeat the quoted poster ;)

Oh, embarrassing for UKRIO!!

(And sorry for your wasted time)

Oh, where is the nonsense training from?!

Not wanting to disappoint you so much that it lies is the last quality I want in a computer.

🤣🤣🤣🤣🤣🙃

Now I’m wondering if this is how so much AI slop goes unnoticed

bsky.app/profile/debr...

“&runctitional features” 😩

🥲

We have a new revision on our preprint "Practice what you preach: Designing student assignments that advance open and reproducible science"! We had some of the nicest (but still critical) peer-review comments I've ever received, so that was refreshing.

osf.io/preprints/os...

Retraction Watch has covered the problem of the “national IQ” database. Should be noted I’m far from alone in working to remove these publications. The spreadsheet of pubs which use NIQ - linked to in the article - was started by @kohngregory.bsky.social; a project also worked on by Cathryn Townsend

A few years old, but this is a fascinating article about how a well-debunked method to teach reading may be at the root of American’s poor reading comprehension

www.apmreports.org/episode/2019...

Oh, I am so weary of this nonsense.

mays.tamu.edu/wp-content/u...

My draft function starts with this:

months <- "Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec|January|February|March|April|June|July|August|September|October|November|December"

pattern <- paste0(
"(", months, "|\\d{1,4})",
"[- /\\.](", months, "|\\d{1,2}),?",
"[- /\\.]\\d{2,4}"
)

I am 100% with you! I'm having a nightmare trying to parse the random-ass formats that journals use for date of submission and publication. Grobid identifies them OK, but doesn't process them at all, so I'm working on a function to make a best guess, but no idea what to do with dates like 01-02-03.

Check Research Outputs for Best Practices A modular, extendable system for automatically checking research outputs for best practices using text search, R code, and/or (optional) LLM queries.

The package formerly known as papercheck has changed its name to metacheck! We're checking more than just papers, with functions to assess OSF projects, github repos, and AsPredicted pre-registrations, with more being developed all the time.

scienceverse.github.io/metacheck/

\title[PhD Plan for CANDIDATE]{Department of \mbox{Full Luxury Galactic Metascience}\\PhD Plan for CANDIDATE} % list committee members \author[]{Committee:\\ Prof.~MEMBER1\\ Dr.~MEMBER2\\ Dr.~MEMBER3} \date{\today} \begin{document} \maketitle% this prints the handout title, author, and date \begin{marginfigure}[-5.2cm] \includegraphics[width=3.5cm]{Minerva.pdf} \end{marginfigure} \section{Phenomenon} The subject matter to be addressed. What is the question? Citation for background are useful, but don't do a full lit review (yet).\cite{HenrichMcElreath:2003} Additional references can be included at the end bibliography by using \verb|\nocite|. \nocite{NBGA2005} \section{Theoretical Perspective} The framework for explaining the phenomenon. What are the implications of theory? If more than one theory is relevant, how do the implications differ? What are the current holes in the theory or theories? \section{Approach} Modeling and empirical research approach to address the theory/ies nominated above. Important thing is to \emph{justify} the approach in terms of theory. \section{Specifics of Approach} Outline any special skills, equipment, or techniques that will be required. This includes, in the case of field research, a paragraph on the background of the field location. Data analysis approach needs a sketch. Can use citation\cite{Merowetal2014} for further detail. \section{Expected Products} A list of the (usually 3) articles expected for completion of the PhD. Think of each as a fantasy abstract. \begin{enumerate} \item \emph{The Evolution of Human Uniqueness.} There is wide agreement that humans are unique. But it is rarely asked: Have they always been unique? In this paper, we derive a general uniqueness theorem that states that any unique species must have been less unique at some point in the past. This theorem has implications for the achievement of world peace.

\item \emph{Niche Construction, Disease Transmission, and Global Warming: Chaos Deconstructs the Block Chain.} How many buzz words can be combined in a title? In this paper, we find that the natural limit on buzz words in any specific scholarly field is always given by the triangular number $T_n$ where $n$ is one more than the number of tenured faculty in the field. \item \emph{A Bayesian Hierarchical Survival Analysis of Counting Sheep.} Just like all those other models of counting sheep, but this one will be Bayesian. \end{enumerate} \section{Schedule} Outline of the calendar for the work. This is necessarily rough and subject to change. \vspace{1em} \begin{fullwidth} {\centering \begin{longtable}{lll} \toprule \makecell{Date\\(Duration)} & Phase & Comment\\ \midrule \makecell{Jan 2019\\(1 year)} & \makecell{Phase 1} & First wave of being awesome \\ \midrule \makecell{Jan 2020\\(1 year)} & \makecell{Phase 2} & \makecell{Write papers \#1 and \#2} \\ \midrule \makecell{Jan 2021\\(6 months)} & \makecell{Phase 3} & Second wave of awesomeness \\ \midrule \makecell{Jun 2021\\(6 months)} & \makecell{Phase 4} & Write paper \#3 \\ \bottomrule \end{longtable} } \end{fullwidth} \bibliography{bibliography} \bibliographystyle{abbrv}

This November, like every November, I am teaching basic research proposal writing to the new phd cohort at my institute. Here is the 2 page template we start with and adapt. Link to LaTeX github.com/rmcelreath/P...

I would like to join the Department of Full Luxury Galactic MetaScience, please

We are changing our name to metacheck in order to more accurately reflect the goals of our software and to not be associated with your software.

ladies and gentlemen...we got him

The Anxious Generation Podcast Episode · If Books Could Kill · 08/08/2024 · 2h

I was re-listening to the episode of @ifbookspod.bsky.social on The Anxious Generation, and the bit around 1:15 where they’re incredulous about “solutions” to age verification online has aged hilariously, given the recent UK online safety act changes.

podcasts.apple.com/gb/podcast/i...

The Festival of Data Science and AI is next week #DataAIFest

I'll be speaking in a panel on Using AI Ethically in Research on Monday
www.gla.ac.uk/research/az/...

And leading a workshop on Papercheck on Tuesday
www.gla.ac.uk/research/az/...

Haha, I have the same feeling, but also don't want to make my decision entirely based on an aesthetic preference :)

Not a bad name in general, but we want to move away from it just being about manuscript and also just being an R package. It may eventually evolve to something mainly web-based and running with more python code than R.

(with an R package version mainly aimed at metascientists)

Yeah, at 84K downloads, I know I can't just change defaults and function/argument names to something more sensible without breaking a bunch of people's code.

It's one of my top tips when I teach people how to write R packages: workshop your names and defaults a lot before you first submit to CRAN!

I've been thinking how to address this in my 70-person MSc coding class. Maybe take-home projects where part of the assessment is 15 minutes talking me through their code? But this adds a potential source of bias and will take ~20 hours (on top of already being well over my teaching allocation)

I'm finally writing a proper paper about faux, and have so many naming & default choices I regret (it was my first solo R package). Which is a better strategy:

1. Add new functions that work the way I wish the old ones did, but keep the old ones
2. Create a faux2 package and retire faux eventually

Haha, feels a bit egotistic :)

Very cute :) but hard to search for the right thing