tobychev @tobychev@qoto.org
Not sure what this account will be about, mostly boosting things I find interesting.
Joined Nov 2022
@erikdelareguera
Hela grejen med landsomfattande protester mot ändringar av pensionerna och så viftas lagen igenom utan omröstning kanske gör att folk är lite mindre övertygande om fördelarna med mer fredliga metoder?
tobychev
boosted
TIL that „beyond the pale“ actually derives from „pale“ as in a stake driven into the ground as part of a fence (think of „impaling“ someone or „palisades“).
So its meaning of going beyond acceptable behaviour suddenly makes much more sense to me
tobychev
boosted
@Mabande
Moahahahah! Ännu ett offer till groddammen!
@philipncohen
Must have been some Chinese hackers that changed the data once it was on his computer! /s
More seriously, I guess the NYT editors didn't pass the story to a journalist with the patience to read more than the headline of the one datacolada post ...
@Mabande
Har du kollat på DiVA eller Google Scholar?
Spontant känns det dock inte som om det är så mycket där, Ben är ju en självgående karaktär medan hela golem grejen handlar om att det inte är helt lyckat att leka gud...
@franco_vazza
Some species have root networks that can be on the order of a kilometre think, the sort of thing that happens when a species can just grow new bits indefinitely
@kevinrothrock
pace yourself man! think of your poor fingers!
@kevinrothrock
seems like the crossposting mechanism can't handle longer tweets? this definitely looks like a bit got cut off
tobychev
boosted
If Ayn Rand's own life wasn't enough to show that libertarianism is a failed idea, we now have further evidence.
tobychev
boosted
@kjhealy “That’s right:
Two different people independently faked data for two different studies in a paper about dishonesty.”
😳
tobychev
boosted
tobychev
boosted
Nice, very hands-on paper on how to run surveys in market-research panels "Using Market-Research Panels for Behavioral Science: An Overview and Tutorial" https://journals.sagepub.com/doi/full/10.1177/25152459221140388 @surveystatisticsandmethodology #surveymethods #survey #surveymethodology
@bart
the url returns a 451 for me?
tobychev
boosted
Stefan Löfven. PR-konsult.
Jag vet att jag inte borde bli förvånad, men jag blir förvånad. https://www.di.se/nyheter/stefan-lofvens-nya-uppdrag-pr-konsult/
tobychev
boosted
Sixty-six million years ago, an asteroid slammed into what is today the Yucatan peninsula of Mexico. The impact was so forceful that it kicked a tremendous amount of debris out of the atmosphere, which then rained back down, blanketing the Earth's surface with a layer of dust.
All that debris re-entering the atmosphere created a pulse of heat so strong that it set the world on fire.
tobychev
boosted
Interesting paper on effect of distance to refugee accommodations on attitudes towards refugees in Germany. "we report a solid null effect and conclude that the placement of reception facilities does not influence locals’ attitudes toward refugees" https://academic.oup.com/esr/advance-article/doi/10.1093/esr/jcad028/7187857 #refugees #racism #SOEP
@erikdelareguera
Nu säger han som presenterade att han "sa fel", och att "we were training it in a simulation" bara ska förstås som något som rent hypotetiskt kan hända i framtiden....
@erikdelareguera
Det märkliga är att, med förbehållet att "människan" nog bara var en annan bott, så låter det hela som ett ganska typiskt problem, men Pentagon blånekar att något experiment ens hänt.
Den här länken ger exempel på när samma problem inträffat i andra sammanhang: https://openai.com/research/faulty-reward-functions tricket är att minnas det faktum att botten inte tar beslut om sin strategi, utan testar massa beteenden typ slumpvis och sen kör på det som ger mest poäng.
Not sure what this account will be about, mostly boosting things I find interesting.
Joined Nov 2022
