Gon García-Castro
LightNews LightNews
banner
gongcastro.bsky.social
Gon García-Castro
@gongcastro.bsky.social
260 followers 330 following 79 posts
Clueless postdoc investigating language acquisition and comparative cognition • fNIRS and Bayesian stats • él/he

🌐 gongcastro.github.io

📍Institut de Recerca Sant Joan de Déu, previously at Universitat Pompeu Fabra.
Posts Replies Media Videos
Pinned
gongcastro.bsky.social
Gon García-Castro @gongcastro.bsky.social · Sep 14
I'm back with a new blog post! 📒

"Installing the MBROLA speech synthesiser"

Some months ago, we decided to synthesise some auditory linguistic stimuli in MBROLA for an experiment. Setting up MBROLA was a bit of a nightmare. In this post I share how we did it.
gongcastro.github.io/blog/pymbrol...
Gonzalo Garcia-Castro
Language, stats, R
gongcastro.github.io
gongcastro.bsky.social
A 30-40% profit margin for a company that mostly relies on unpaid labour from their own "clients" to barely do their job making science (which they did not produce) available is a good case for abolishing the whole industry.
hansonmark.bsky.social Mark A. Hanson @hansonmark.bsky.social · 3d
We wrote the Strain on scientific publishing to highlight the problems of time & trust. With a fantastic group of co-authors, we present The Drain of Scientific Publishing:

a 🧵 1/n

Drain: arxiv.org/abs/2511.04820
Strain: direct.mit.edu/qss/article/...
Oligopoly: direct.mit.edu/qss/article/...
A table showing profit margins of major publishers. A snippet of text related to this table is below. 1. The four-fold drain 1.1 Money Currently, academic publishing is dominated by profit-oriented, multinational companies for whom scientific knowledge is a commodity to be sold back to the academic community who created it. The dominant four are Elsevier, Springer Nature, Wiley and Taylor & Francis, which collectively generated over US$7.1 billion in revenue from journal publishing in 2024 alone, and over US$12 billion in profits between 2019 and 2024 (Table 1A). Their profit margins have always been over 30% in the last five years, and for the largest publisher (Elsevier) always over 37%. Against many comparators, across many sectors, scientific publishing is one of the most consistently profitable industries (Table S1). These financial arrangements make a substantial difference to science budgets. In 2024, 46% of Elsevier revenues and 53% of Taylor & Francis revenues were generated in North America, meaning that North American researchers were charged over US$2.27 billion by just two for-profit publishers. The Canadian research councils and the US National Science Foundation were allocated US$9.3 billion in that year. A figure detailing the drain on researcher time. 1. The four-fold drain 1.2 Time The number of papers published each year is growing faster than the scientific workforce, with the number of papers per researcher almost doubling between 1996 and 2022 (Figure 1A). This reflects the fact that publishers’ commercial desire to publish (sell) more material has aligned well with the competitive prestige culture in which publications help secure jobs, grants, promotions, and awards. To the extent that this growth is driven by a pressure for profit, rather than scholarly imperatives, it distorts the way researchers spend their time. The publishing system depends on unpaid reviewer labour, estimated to be over 130 million unpaid hours annually in 2020 alone (9). Researchers have complained about the demands of peer-review for decades, but the scale of the problem is now worse, with editors reporting widespread difficulties recruiting reviewers. The growth in publications involves not only the authors’ time, but that of academic editors and reviewers who are dealing with so many review demands. Even more seriously, the imperative to produce ever more articles reshapes the nature of scientific inquiry. Evidence across multiple fields shows that more papers result in ‘ossification’, not new ideas (10). It may seem paradoxical that more papers can slow progress until one considers how it affects researchers’ time. While rewards remain tied to volume, prestige, and impact of publications, researchers will be nudged away from riskier, local, interdisciplinary, and long-term work. The result is a treadmill of constant activity with limited progress whereas core scholarly practices – such as reading, reflecting and engaging with others’ contributions – is de-prioritized. What looks like productivity often masks intellectual exhaustion built on a demoralizing, narrowing scientific vision. A table of profit margins across industries. The section of text related to this table is below: 1. The four-fold drain 1.1 Money Currently, academic publishing is dominated by profit-oriented, multinational companies for whom scientific knowledge is a commodity to be sold back to the academic community who created it. The dominant four are Elsevier, Springer Nature, Wiley and Taylor & Francis, which collectively generated over US$7.1 billion in revenue from journal publishing in 2024 alone, and over US$12 billion in profits between 2019 and 2024 (Table 1A). Their profit margins have always been over 30% in the last five years, and for the largest publisher (Elsevier) always over 37%. Against many comparators, across many sectors, scientific publishing is one of the most consistently profitable industries (Table S1). These financial arrangements make a substantial difference to science budgets. In 2024, 46% of Elsevier revenues and 53% of Taylor & Francis revenues were generated in North America, meaning that North American researchers were charged over US$2.27 billion by just two for-profit publishers. The Canadian research councils and the US National Science Foundation were allocated US$9.3 billion in that year. The costs of inaction are plain: wasted public funds, lost researcher time, compromised scientific integrity and eroded public trust. Today, the system rewards commercial publishers first, and science second. Without bold action from the funders we risk continuing to pour resources into a system that prioritizes profit over the advancement of scientific knowledge.
November 13, 2025 at 8:49 AM
gongcastro.bsky.social
Por si alguien más está pensando dejar Spotify:
gongcastro.bsky.social Gon García-Castro @gongcastro.bsky.social · 5d
4 months since I changed to Tidal, very good alternative. Almost identical experience with two exceptions: no podcasts (I'm using another app for those), and offline mode is not good. Otherwise, I'm very happy with the change!

I used a Python script to transfer playlists:
github.com/spotify2tida...
GitHub - spotify2tidal/spotify_to_tidal: A command line tool for importing your Spotify playlists into Tidal
A command line tool for importing your Spotify playlists into Tidal - spotify2tidal/spotify_to_tidal
www.google.com
November 10, 2025 at 10:19 AM
gongcastro.bsky.social
📒 Here's a new blog post, in which I present pymbrola, a Python module that provides an interface with the MBROLA speech synthesiser:

"pymbrola: A Python interface to the MBROLA speech synthesiser"

gongcastro.github.io/blog/pymbrol...
Gonzalo Garcia-Castro
Language, stats, R
gongcastro.github.io
November 8, 2025 at 1:45 PM
gongcastro.bsky.social
La ciencia es, de forma efectiva e inevitable, una herramienta discursiva. Y el hecho de que mucha gente aún crea que podamos demarcar la ciencia es parte del problema.
sholtodavid.bsky.social Sholto David @sholtodavid.bsky.social · 23d
Happy to share a new peer-reviewed study on PubMed. Blessing eggs creates bigger and tastier chickens (p < 0.001). pubmed.ncbi.nlm.nih.gov/41035978/
October 24, 2025 at 7:40 AM
Reposted by Gon García-Castro
carlbergstrom.com
1. We ( @jbakcoleman.bsky.social, @cailinmeister.bsky.social, @jevinwest.bsky.social, and I) have a new preprint up on the arXiv.

There we explore how social media companies and other online information technology firms are able to manipulate scientific research about the effects of their products.
Three schematic diagrams. The first illustrates selective publishing of internal resection, the second selective causal focus, and the third selective access and funding for researchers.
October 24, 2025 at 12:47 AM
gongcastro.bsky.social
Claro, la subida de los precios se debe al aumento de los salarios de los trabajadores sin un incremento de la productividad.

El timón danés que se están haciendo las principales compañías tecnológicas entre ellas—invirtiendo unas en otras miles de millones sin producir nada—no tiene nada que ver.
graceblakeley.substack.com Grace Blakeley @graceblakeley.substack.com · 24d
Nvidia invests in OpenAI, OpenAI rents compute from Oracle, Oracle buys Nvidia’s hardware, and Nvidia records huge sales.

Everyone in the circle reports rising revenue, rising valuations, and surging share prices, but the cash is largely moving in circles.
graceblakeley.substack.com/p/the-ai-cir...
The AI Circular Economy
Big tech companies and wealthy investors are driving an AI bubble - but they won't be the ones feeling the pain when it bursts.
graceblakeley.substack.com
October 22, 2025 at 7:18 AM
gongcastro.bsky.social
Made some friends in the Marmara sea.
October 21, 2025 at 6:40 AM
gongcastro.bsky.social
Outsourcing your thinking process to an autocomplete machine on steroids is definitely a good thing to encourage your students to do, and is definitely how AI is supposed to push societal progress.
October 9, 2025 at 7:42 AM
gongcastro.bsky.social
Yo creo que uno no debería hacer divulgación hasta que no tenga su primer Nobel y cuarenta y cinco años cotizados. Antes de eso me parece prematuro hablar de ciencia de manera accesible y rigurosa.
October 9, 2025 at 7:31 AM
gongcastro.bsky.social
It only takes me another attempt at coding a multilevel model in Stan to remember what a blessing brms is.
October 8, 2025 at 12:54 PM
gongcastro.bsky.social
October 7, 2025 at 11:31 AM
gongcastro.bsky.social
Barcelona 🇵🇸🇵🇸🇵🇸🇵🇸🇵🇸🇵🇸
October 4, 2025 at 11:48 AM
Reposted by Gon García-Castro
ladirecta.bsky.social
🔴 ÚLTIMA HORA | L'organització de la manifestació en suport a Palestina calcula que la convocatòria ha reunit 300.000 persones. Hi ha tanta gent que algunes persones estan baixant per la Rambla i el carrer de Pau Claris. El Passeig de Gràcia està completament col·lapsat.
October 4, 2025 at 11:45 AM
Reposted by Gon García-Castro
listen2.bsky.social
UN says the genocide in Gaza won't stop as it's profitable. Palestine is a training field to test weapons & surveillance tech that are now sold worldwide.
Those making billions?
Palantir, Lockheed Martin, Alphabet, Amazon, IBM, Caterpillar, Microsoft, MIT, Blackrock, insurers, real estate firms.
September 23, 2025 at 8:51 AM
gongcastro.bsky.social
🇵🇸🇵🇸🇵🇸🇵🇸
September 14, 2025 at 7:16 PM
gongcastro.bsky.social
I'm back with a new blog post! 📒

"Installing the MBROLA speech synthesiser"

Some months ago, we decided to synthesise some auditory linguistic stimuli in MBROLA for an experiment. Setting up MBROLA was a bit of a nightmare. In this post I share how we did it.
gongcastro.github.io/blog/pymbrol...
Gonzalo Garcia-Castro
Language, stats, R
gongcastro.github.io
September 14, 2025 at 3:12 PM
gongcastro.bsky.social
This week I taught an introductory lecture on fNIRS at the Deep Tech Brain Summer School at @ub.edu.

We covered some theory, experimental design in dev populations and a live demonstration of data acquisition and preprocessing using @nirx-nirs.bsky.social's NIRSport2. Thanks to the organisers!
July 23, 2025 at 7:17 AM
Reposted by Gon García-Castro
rmcelreath.bsky.social
Wouldn't this method show that classic social psychology garbage was highly replicable? Besides, false findings can replicate even without pub bias - hot hand "fallacy" replicated but data analysis was incoherent.

After a couple decades of science reform, I'd like to think we've moved beyond this
scientificdiscovery.dev Saloni @scientificdiscovery.dev · Jul 17
I'm not sure if this is a normal way to assess replication, but it wasn't what I expected and seems prone to being affected by publication bias...

(paper here: www.biorxiv.org/content/10.1...)
The team looked at 1,006 results from 400 papers published in the field between 1959 and 2011 — up until Jules Hoffmann won a share of the 2011 medicine Nobel prize for work on the Drosophila immune response. They then went through the published literature looking for later work that either supported or refuted each claim. Rather than looking for studies that exactly replicated the methods of an experiment behind a result, or reproduced an identical effect, they searched to see if the claims were successfully built on by other studies, were broadly reproduced using more-modern techniques or still stood up to scrutiny now that knowledge of the field has moved on. Using their expertise of the field, the researchers classified claims as either verified, unchallenged by other papers, challenged or partly verified.
July 17, 2025 at 1:12 PM
gongcastro.bsky.social
Three years into the "formal" onset of AI hype and the only domains where AI has objectively impacted people's lives is that we now get identified, located, made redundant, and killed more easily.
July 15, 2025 at 7:54 AM
Reposted by Gon García-Castro
ramiismail.com
Bye forever, WeTransfer.
WeTransfer TOS says they can use all your uploaded content for genAI
July 14, 2025 at 11:57 PM
gongcastro.bsky.social
La actual era de la austeridad y los recortes ha sido rebautizada como "la era de la abundancia". Naturalmente, no nos queda otra que volver a la austeridad y los recortes.
elpais.com EL PAÍS @elpais.com · Jun 30
Se acabó la era de la abundancia. La combinación de pensiones, deuda y defensa aboca a España a décadas de austeridad. La Autoridad Fiscal calcula que España necesitará un ajuste de 30.000 millones por llevar el gasto militar al 3,5% del PIB social.elpais.com/pofp25
Deuda, pensiones y defensa: el fin de la abundancia
La Autoridad Fiscal calcula que España necesitará un ajuste de 30.000 millones por llevar el gasto militar al 3,5% del PIB. El BCE estima que la deuda aumentará en 10 puntos por el mayor gasto en defensa
social.elpais.com
July 2, 2025 at 6:53 AM
Reposted by Gon García-Castro
olivia.science
Hello & HAPPY FRIDAY, as promised an open letter by my lovely colleagues and myself. Everybody, feel free to sign this even if you are not NL-based and ‼️ share ‼️ (anonymous signature is poss):

Open Letter: Stop the Uncritical Adoption of AI Technologies in Academia

openletter.earth/open-letter-...
Open Letter: Stop the Uncritical Adoption of AI Technologies in Academia Our demands, see link
June 27, 2025 at 11:38 AM
gongcastro.bsky.social
Solo me quedan dos episodios de #Andor. Es muy una pasada que la saga con la que crecí desde los cinco años haya albergado algo una serie tan interesante, potente y políticamente comprometida.
June 29, 2025 at 7:45 PM
gongcastro.bsky.social
There's some audacity in this statement after a decade of promoting the mantra that there is a "replicability *crisis*" that should make everyone doubt the credibility of science, as a whole.

Now it turns out it's been fueling anti-science discourses without improving the credibility of science?
"If we define a crisis as something that’s changing for the worse, we don’t have the evidence to say that’s happening." Brian Nosek, executive director, Center for Open Science.
June 29, 2025 at 10:07 AM
Reposted by Gon García-Castro
devezer.bsky.social
"Nosek also wants academia to create clear career paths for people doing replication studies—it should be something that university bosses admire and want to promote."

Also NO. This is someone's pet theory for how science works, not actually how science works.
June 29, 2025 at 6:47 AM