Sign in

Mart Roben

@martroben.bsky.social
58 followers 296 following 16 posts

data / stats / open science 🇪🇪

PostsRepliesMedia
Mart Roben @martroben.bsky.social · 12/03/2026
MacGyver --> chemistry.
000
Mart Roben @martroben.bsky.social · 24/02/2026
These classic texts are a good place to start: - pmc.ncbi.nlm.nih.gov/articles/PMC... - doi.org/10.1371/jour... For a deeper dive, see "Better Code, Better Science" by @russpoldrack.org: bettercodebetterscience.github.io/book/index.h...
bettercodebetterscience.github.io
Better Code, Better Science - Better Code, Better Science
030
Mart Roben @martroben.bsky.social · 07/11/2025
Too bad Web of Science & co wasn't tested. If they failed, this would put the "paid-access sci metrics are better quality than open"-argument to rest. If they didn't fail (i.e. knew about it) and didn't tell... well that's just evil.
030
Mart Roben @martroben.bsky.social · 29/04/2025
I might end up with something like this. 🙂
ASCII representation of repository structure
120
Mart Roben @martroben.bsky.social · 08/04/2025
Use it & love it. 1k+ pages of notes in the first year alone. The excalidraw plugin lets you create sketches & schematics with LaTeX formulas. The paid version seamlessly syncs your notes across all your devices and is well worth it.
010
Mart Roben @martroben.bsky.social · 10/01/2025
Community Improvement Advocates. It would need to be shortened of course. As in the CIA is here, they're asking about your data!
010
Mart Roben @martroben.bsky.social · 07/01/2025
5/5 ARTICLES OF INTEREST s03 - AMERICAN IVY The third season of this podcast tells the story of the origins of preppy clothing style. Information I didn't know I needed, but enjoyed thoroughly. open.spotify.com/episode/5qRZ...
open.spotify.com
American Ivy: Chapter 1
Articles of Interest · Episode
010
Mart Roben @martroben.bsky.social · 07/01/2025
4/ QUANTITUDE Imagine a universe where Beavis and Butthead got tenured as quantitative psychologists and made a stats podcast. That's Quantitude. Quantitative methods are also at the roots of science, so it's the yang to Nullius in Verba's yin. quantitudepod.org
quantitudepod.org
QuantitudePod – A podcast dedicated to all things quantitative, ranging from the relevant to the highly irrelevant.
110
Mart Roben @martroben.bsky.social · 07/01/2025
3/ NULLIUS IN VERBA Nullius in Verba investigates the roots of scientific ideals and practices. Why are some things called "scientific" and who decided that? (And a lot of Bacon quotes.) nulliusinverba.podbean.com
nulliusinverba.podbean.com
Nullius in Verba | Smriti Mehta and Daniël Lakens
Nullius in Verba is a podcast about science—what it is and what it could be. It is hosted by Smriti Mehta from UC Berkeley and Daniël Lakens from Eindhoven University of Technology. We draw inspira...
241
Mart Roben @martroben.bsky.social · 07/01/2025
2/ METASCIENCE SINCE 2012: A PERSONAL HISTORY A blog post that could be a small book. A memoir by Stuart Buck about the coming of age of open science. The most inspiring text I read last year. goodscience.substack.com/p/metascienc...
goodscience.substack.com
Metascience Since 2012: A Personal History
This essay is a personal history of the $60+ million I allocated to metascience starting in 2012 while working for the Arnold Foundation (now Arnold Ventures).
131
Mart Roben @martroben.bsky.social · 07/01/2025
Top books/blogs/podcasts from 2024. 1/ DESIGNING DATA-INTENSIVE APPLICATIONS - MARTIN KLEPPMANN A book about database technology that is the rare combination of highly readable and deeply technical. Think Feynman explaining physics, but for databases. www.goodreads.com/book/show/23...
goodreads.com
Designing Data-Intensive Applications
Data is at the center of many challenges in system desi…
120
Mart Roben @martroben.bsky.social · 04/01/2025
According to the cited source, it's 22%, not 45%. (Still too high of course!) 54.3 h work week -> 27 h spent on research -> 12 h (45% of research time) spent on research-related admin. thefdp.org/wp-content/u... p18
thefdp.org
020
Mart Roben @martroben.bsky.social · 13/12/2024
source: squareallworthy.tumblr.com/post/1637900... I thought it was gwern (gwern.net/epigram#fn1), but he cites the one above.
squareallworthy.tumblr.com
Everyone will not just
If your solution to some problem relies on “If everyone would just...” then you do not have a solution. Everyone is not going to just. At not time in the history of the universe has everyone just, and...
120
Mart Roben @martroben.bsky.social · 25/11/2024
Was R deliberate or happenstance? Is there a universe where tidyverse happened in python (or rust or elixir) instead? Also, thank you for revealing the rhythm, logic and beauty in data wrangling. In pyspark, polars, base R etc. I feel like a mechanic. In tidyverse, I feel like a poet.
090
Mart Roben @martroben.bsky.social · 24/11/2024
Suppose you record a transaction every time you share your data. You then get scooped (someone leaked your data without recording it). The scoopers say that they're the owner and you're the thief, using blockchain to legitimize your steal. Below ~70% global adoption, they'd likely get away with it?
000
Mart Roben @martroben.bsky.social · 24/11/2024
Interested, but ignorant. Is there a tl;dr about what blockchain could do for science?
210