Léo Varnet @leovarnet@qoto.org

To continue on the issue of #discrimination in selection committees, it seems to me that there are two levels on which to act. First, to reduce the influence of biases in the decision, for example by choosing the selection criteria before looking at the applications. However, this kind of tools is not enough: even if we manage to eliminate judgment biases, blindly basing the decision on merit is unfair for candidates facing discrimination or coming from countries where research is less developed, and who did not have the same opportunities as others. Complementary tools are therefore proposed, for example, considering the potential of candidates and not their achievements. I think this aspect is important, but more difficult to implement: Given that not all members of the jury are sensitive to the issue of discrimination, it can be tricky to argue that we want to put forward a candidate with a lesser CV because we think he or she has greater potential... assessing achievements is a bit like "counting points", whereas assessing potential seems more subjective. Do any of you have a concrete experience of how to deal with this in a jury? @academicchatter #AcademicMastodon #genderbias

@leovarnet Our committee put together a list of tips on this here (section "You are on a selection committee and want to avoid bias in your admission/hiring procedure"): https://equalityanddiversity.dec.ens.fr/en/am-i-biased-15020

@arth This is a great explanation of git commands and data flow. I've uploaded the image with an alt-text description.
#Alt4You

@erinnacland @academicchatter The very definition of "The Matthew Effect" in science, coined by Robert Merton in 1968. https://www.science.org/doi/10.1126/science.159.3810.56
My colleague, Dr. Patricia Mabry, currently has an NSF/NIH project studying what she believes is the "Matthew Mechanism." https://www.nsf.gov/awardsearch/showAward?AWD_ID=2122232&HistoricalAwards=false
I was a co-I on this project until I quit my job at HealthPartners earlier this year.

Neuromancer 2022

The collapse of a startup leaves implanted medical devices in > 700 people

The unavailability of the proprietary SW needed to recalibrate the devices and maintain its effectiveness, and the draining of batteries, will leave them without treatment and with HW implanted in their bodies

They have to hack their own implants

And it is not an isolated case but a trend

https://www.nature.com/immersive/d41586-022-03810-5/index.html

@lanmar @StephaneDebove @RichardMonvoisin
Bonjour ! Au sujet de Steven Pinker, j'avais écrit deux billets de blog à la suite de la publication de son livre "Le Triomphe des Lumières" et de la série de conférences qu'il a donnée à cette occasion. Notez bien qu'il ne s'agit que d'une petite partie de sa production (essentiellement ses thèses concernant le progrès, la "raison", et le libéralisme, pour le dire vite). Notamment, il ne s'agit pas de sa production en tant que scientifique, mais de ses positions dans le domaine économique-politique, où il n'a pas d'expertise. C'est pourtant ces facettes de Steven Pinker qui sont les plus médiatisées.
(1/4)

En février, le Spectrum de l'IEEE alertait sur l'obsolescence technique des implants neuronaux : https://spectrum.ieee.org/bionic-eye-obsolete#toggle-gdpr - 350 patients qui avaient reçu un implant oculaire pour leur rendre en partie la vue l'ont vu s'arrêter brutalement... quand l'entreprise a fermé !

Mediapart revient sur le faux scoop du Point sur le couple Garrido-Corbière. L’hebdo avait promis de publier une enquête pour expliquer le fiasco mais l'article rédigé a été enterré. Le site d'information pointe "un journal à folle allure, avec un processus de vérification minimal". En interne, élus syndicaux et journalistes avaient alerté des mois avant sur la situation critique au sein de la rédaction.
#NotrePresseEnRevue
https://www.mediapart.fr/journal/france/081222/faux-scoop-du-point-sur-garrido-corbiere-un-accident-industriel-ecrit-d-avance

New blog post! Tagging @linguistics and #linguistics this toot

@ahnaphor had a cool Twitter thread a week or so about the "sexy baby voice" described by Lake Bell. From the examples that Bell gives, it seems like the primary features of "sexy baby voice" are bright vocal resonance and creaky voice (aka vocal fry).

There's a lot more to say about this, but I wanted to get the basic #phonetics down!

https://grieve-smith.com/blog/2022/11/listen-to-the-voices-of-the-sexy-babies/

L'horreur. Derrière Neuralink, 1500 animaux morts, notamment suite à des actes chirurgicaux réalisés dans la hâte. La science muskienne a déjà de sacrées casseroles.

https://www.numerama.com/sciences/1201284-neuralink-1-500-animaux-seraient-morts-lors-des-experiences-de-la-startup-delon-musk.html

New paper shows (as many papers before) that code that is shared will often not run. This is to be expected - few of us had training in this. But if you share code, go through checklists to prevent the most common mistakes. From the paper: https://www.nature.com/articles/s41597-022-01143-6 For some extra suggestions, see my textbook chapter on computational reproducibility: https://lakens.github.io/statistical_inferences/computationalreproducibility.html#some-points-for-improvement-in-computational-reproducibility