En concierto!
Remote
gabriel
@gabriel@col.social
mastodon 4.7.2Físico. Profesor de Ciencia de Datos en la Escuela Superior de Administración Pública. Lector ecléctico. Intento volver a mis raíces costeñas y tomarme la vida con suavena (con relativo poco éxito). (pronombres: él)
513 Followers
863 Following
50 Posts
Joined November 03, 2022
Pronómbres:
el/he
Página web:
gavox.srht.site
Mi esposa escuchó por TikTok a un cantante Mexicano. Le cogió cariño, se aprendió un par de canciones. Ayer fuimos a escuchar el concierto en Bogotá. Tuvimos un poco de catarsis gritando "Todo Pasa" y "Estamos Rotos"
Replying to
@pattykimura
If it is 95% of all men, then this would also imply that _at least_ 95% of all men who had sex in the last two years had multiple sexual partners. I find this figure to be quite strange, given what other studies of sexual behavior of younger generations show.
Now, I am answering in good faith, because I think we are having a conversation. But if you feel differently, feel free to block, I do not need the validation of your attention.👋🏽
Replying to
@pattykimura @amydiehl
I was shocked by the 95% figure, so I downloaded the paper. In it, I found out that the universe of the study is not "all men" but "men who identify themselves as aggressors". Now, when I claim "not-all-men" here, is not a way to deviate the conversation. Is to focus on what the intent of the paper is, and which conclusions can be taken from it. One of them, we need to know what the actual % is. Or focus on what aggressive men do, goal of this particular paper.
Replying to
@MissConstrue@mefi.social You probably know about Frances Perkins, and how the original new deal was, in fact, engineered by a her. But it's worth repeating.
Throughline: Frances Perkins Goes To Washington
Episode webpage: https://www.npr.org/2026/05/19/nx-s1-5826026/frances-perkins-goes-to-washington
Media file: https://tracking.swap.fm/track/XvDEoI11TR00olTUO8US/prfx.byspotify.com/e/play.podtrac.com/npr-510333/npr.simplecastaudio.com/923a8c19-6266-45f0-88de-3997479f3ca3/episodes/153139dc-284f-41ab-a090-106b0846cff6/audio/128/default.mp3?awCollectionId=923a8c19-6266-45f0-88de-3997479f3ca3&awEpisodeId=153139dc-284f-41ab-a090-106b0846cff6&feed=gvAZ3nqF&t=podcast&e=nx-s1-5826026&p=510333&d=968&size=15494670
Replying to
@jenzi @emilymbender @fuzzy
Say you want to discuss accuracy instead of power. In any other industry you could do it independently. For instance "accuracy of speedometer in a car". In the field of LLMs they don't let independent agencies full access to the code. And furthermore, change the model constantly. That's opacity to the consumer.
Me parece genial lo que hacen en el podcast de Economía para la Pipol; entre otras cosas acercar "role models" de mujeres exitosas. Especialmente recomiendo este episodio, en el que entrevistan a Vanessa Rosales:
https://youtu.be/cRWYFZdvtuE
Me quedaron ganas de seguir su podcast.
Replying to
@emilymbender @fuzzy
And furthermore, since there is opacity in the way these models are implemented, metrics that should be useful are actually worthless.
(As @emilymbender has discussed elsewhere)
Replying to
@abhayakara
I also concluded the same, that this could be used to argue that "men are this way". And while sexual aggression is a serious problem that is under-reported, a 95% is just too high. That's why I think it is important to clarify the number, and the context.
Replying to
@CorioPsicologia@mastodon.social @eco_amandine@mastodon.cr yo fracasé estrepitosamente con las matas de mi balcón:'(
Pero voy a doblar la apuesta, a sembrar un bosque
Replying to
@irenerd
Falta tantísimo por conocer de las Ballenas! En lugar de estar pensando en comunicarnos con alienígenas, deberíamos primero entender a los demás compañeros de esta pequeña roca.
Replying to
@Tangersoto@col.social @arturo@col.social el lío también es ese sobrecoste tan berraco. En los 2000s los colegas chinos compraban las ediciones piratas de los libros técnicos, en su tierra, por 5 dólares
Replying to
@colinpurrington do you know if this will work on culicoides? https://en.wikipedia.org/wiki/Culicoides?wprov=sfla1
Replying to
@maxleibman@beige.party so the answer is technically "yes", but it's better to skip it altogether.
Replying to
@Irenetherogue As I've said before, I'm talking in good faith. Furthermore, I think that the particular interpretation of the 95% number, besides being wrong, puts more women in danger. Since it deviates the conversation from: "there is a portion of men who are dangerous aggressors" towards "men are irredeemable". The latter closes any possibility of conversation between men and women, which terrifies me to be honest.
Replying to
@Irenetherogue @ratel @TomasHradcky
Let's ask the authors, then, what they actually meant. Then we can move on and talk about what needs to be done to end sexual aggression.
Replying to
@unchartedworlds @amydiehl they say that they picked the men from Reddit, and that they used tactics as asking them to "tell their side" of the story. What I'm seeing is a bias in the sample. It is as if I go to the LGBT friendly park in a city and ask about sexual diversity and self-image. Chances are my sample is not representative of the city population as a whole.
Replying to
@chris_spackman
I think is really valuable work, and it is important to document in the literature the strategies that are used by aggressors to force woman to have sex.
But I also think that in this particular instance finding information towards their RQ1 (what % of men..) is in conflict with their other RQ (how effective are the strategies, etc).
Replying to
@unchartedworlds
I agree, they don't claim that the subjects were chosen via online forums. I was confused with the term "online panel". I don't actually know how they chose.
The sample clearly is biased. For instance, the number of men who have more than 1 sexual partner in the US is not 95%, see https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2767066
Replying to
@lzg@mastodon.social morally better than almost everyone
As a child, my uncle's neighbor thought that I was studying to be a priest, because I always sat down to read a thick book he had on the shelf.
It was an unpublished novel by one of his friends, so I wasn't allowed to take it with me. My only chance was to read a few pages at a time, during each visit
Replying to
@emilymbender@dair-community.social
It is very interesting, thanks for sharing.
I haven't thought about "parrot" in "stochastic parrots" being related to the verb "to parrot" in English. Maybe is because it is my second language.
Thinking of that, how would one translate the term to, say, Spanish? "Loros estocàsticos" doesn't have the verb intention. "autómatas cantinflescos" would be more akin to it, using the Spanish verb "cantinflear".