Lightnews — Scholar-powered news

Reposted by Claire

Mic Wright @brokenbottleboy.bsky.social · 5d

This is a list of ways to do a really shoddy job as a journalist. As a friend just said to me, if you get AI to summarise a big report you’re almost guaranteed to miss any story that might be lurking in the detail. www.journalism.co.uk/news/how-can...

15 essential tasks GPTs can do for journalists

No, AI cannot replace original journalism. But it can remove some of the everyday tedium so you can focus on what you do best

www.journalism.co.uk

5 10 50

Reposted by Claire

John Mansfield @johnbasil.bsky.social · 9d

Social functions of language may drive faster change and faster diversification in some features.
Thanks to growing diversity in variationst sociolinguistics, I gathered studies of 63 languages from 28 families. Here I sketch some potential patterns
compass.onlinelibrary.wiley.com/doi/10.1111/...

Linguistic Diversification and Rates of Change: Insights From a Diverse Sample of Sociolinguistic Studies

Language diversification and change can be studied using phylogenetic modelling of families over thousands of years, or by close observation of changes unfolding over a few decades at the community l....

compass.onlinelibrary.wiley.com

10 16

Claire @chirila.bsky.social · 13d

Interesting! I need to look more but one thing is that the pyramids are of different parts of the plant, so it might be less meaningful than first appears

Reposted by Claire

Prisonculture @prisonculture.bsky.social · 24d

As an archivist, this keeps me up.

Conrad Hackett @conradhackett.bsky.social · 24d

LINK ROT: 38% webpages that existed in 2013 were no longer available 10 years later.

Even among pages that existed in 2021, 22% no longer accessible just two years later. This is often because individual page was deleted or removed on otherwise functional website.

Many implications for knowledge 🧪

A line chart showing that 38% of webpages from 2013 were no longer accessible one decade later.

17 160 630

Reposted by Claire

La Kinopithèque @lakinopitheque.bsky.social · 26d

Arrêtez-tout !
Mon truc c'est l'alchimie médiévale, pas les herbiers... Mais... Ne peut-on voir une ressemblance entre ce dessin dans une marge du Livre des miracles de Sainte Foy (MS022, XIe-XIIe s. Sélestat) et les plantes imaginaires du fameux manuscrit Voynich ?
#medievalsky

1 1 3

Reposted by Claire

Betsy Sneller @betsysneller.bsky.social · 28d

I am so grateful to the editors of Diachronica for letting me write this piece in memoriam of Bill Labov.

And so grateful that they let me write what I most wanted to - a piece about Bill's *goodness*, his love for humanity, and how those things *resulted in* what we think of as his genius

🐦

Diachronica journal @diachronica.bsky.social · Sep 8

www.jbe-platform.com/content/jour... is now out, @betsysneller.bsky.social 's beautiful tribute to Bill Labov

Falling in love with humanity | John Benjamins

Abstract William (Bill) Labov, pioneering scholar of language change, died on December 17, 2024. His MA and PhD theses in the mid 1960s revolutionized the field of language variation and change, and h...

www.jbe-platform.com

1 12 35

Reposted by Claire

HiPhiLangSci @hiphilangsci.bsky.social · 28d

A great opportunity to (re-)listen to episode 38 of our podcast, in which James McElvenny (@jamesmcelvenny.bsky.social) speaks with Dan Everett about the life and work of Charles Sanders Peirce.

🎙️ hiphilangsci.net/2024/04/01/p...

#Histlx

4 5

Reposted by Claire

Ben Zimmer @benzimmer.bsky.social · 29d

The @wired.com video that @mixedlinguist.bsky.social and I did a while back on the life cycle of slang got featured today on @upworthy.com! www.upworthy.com/linguists-ex...

Linguists explain why some slang words come and go quickly while others have staying power

"On fleek" was a flash in the pan, but "cool" has stuck around for nearly a century.

www.upworthy.com

6 27

Claire @chirila.bsky.social · 29d

James Stanford's 2014 handbook paper says this clearly. But I've assumed there are two main patterns: the teenager incrementation, and the changes that erode evidence for a system so much that kids don't acquire it (also a type of incrementation but with younger kids)

8

Claire @chirila.bsky.social · 29d

Artificial intelligence will never be a good substitute for genuine stupidity

Josie Duffy Rice @jduffyrice.bsky.social · 29d

“We believe that in the near future half the people on the planet will be AI, and we are the company that’s bringing those people to life,” said CEO Jeanine Wright, who was previously chief operating officer of podcasting company Wondery
www.hollywoodreporter.com/business/dig...

5,000 Podcasts. 3,000 Episodes a Week. $1 Cost Per Episode — Behind an AI Start Up’s Plan

Former Wondery exec Jeanine Wright is leading a new firm, Inception Point AI, that's betting on flooding the zone with audio content: “I think that people who are still referring to all AI-generated c...

www.hollywoodreporter.com

2

Reposted by Claire

Limor Raviv 🐘🤗🦒🍄🦄 @limorraviv.bsky.social · 29d

Super proud of this work and very grateful for all the amazing people who provided valuable feedback along the way @remivantrijp.bsky.social @sheinalew.bsky.social @cedricboeckx.bsky.social @inbalarnon.bsky.social ❤️❤️❤️🙏

Max Planck Institute for Psycholinguistics @mpi-nl.bsky.social · 29d

Children are not the main agents of language change. New paper by @limorraviv.bsky.social @damianblasi.bsky.social & @vkempe.bsky.social.
psycnet.apa.org/doi/10.1037/...
www.mpi.nl/publications...

7 26

Reposted by Claire

Surprised Eel Historian, PhD @greenleejw.bsky.social · Sep 7

Have you ever wondered what goes into making a map, or though about how maps wind up in the books you love?

I draw maps for living, and every week I write about how they come into being. The posts are free and available for anyone to read. So come give them a read!

patreon.com/SurprisedEel...

Custom Maps, Artwork, and History

patreon.com

3 39 170

Reposted by Claire

Diachronica journal @diachronica.bsky.social · Sep 8

www.jbe-platform.com/content/jour... is now out, @betsysneller.bsky.social 's beautiful tribute to Bill Labov

Falling in love with humanity | John Benjamins

Abstract William (Bill) Labov, pioneering scholar of language change, died on December 17, 2024. His MA and PhD theses in the mid 1960s revolutionized the field of language variation and change, and h...

www.jbe-platform.com

4 6

Reposted by Claire

Diachronica journal @diachronica.bsky.social · Sep 8

Falling in love with humanity benjamins.com/catalog/dia....

Falling in love with humanity: Remembering Bill Labov

William (Bill) Labov, pioneering scholar of language change, died on December 17, 2024. His MA and PhD theses in the mid 1960s revolutionized the field of language variation and change, and he spent t...

benjamins.com

1 2

Reposted by Claire

Jess Calarco @jessicacalarco.com · Sep 8

Ironically, it appears that AI chatbots hallucinate for the same reason that students feel compelled to use them:

They were socialized in a high-stakes testing culture that rewards guessing and maybe getting it right over admitting when there's something you just don't know.

Why Language Models Hallucinate, by Kalai et al.

Like students facing hard exam questions, large language models sometimes guess when
uncertain, producing plausible yet incorrect statements instead of admitting uncertainty. Such
“hallucinations” persist even in state-of-the-art systems and undermine trust. We argue that
language models hallucinate because the training and evaluation procedures reward guessing over
acknowledging uncertainty, and we analyze the statistical causes of hallucinations in the modern
training pipeline. Hallucinations need not be mysterious—they originate simply as errors in binary
classification. If incorrect statements cannot be distinguished from facts, then hallucinations
in pretrained language models will arise through natural statistical pressures. We then argue
that hallucinations persist due to the way most evaluations are graded—language models are
optimized to be good test-takers, and guessing when uncertain improves test performance. This
“epidemic” of penalizing uncertain responses can only be addressed through a socio-technical
mitigation: modifying the scoring of existing benchmarks that are misaligned but dominate
leaderboards, rather than introducing additional hallucination evaluations. This change may
steer the field toward more trustworthy AI systems.

43 410 1.5K

Reposted by Claire

Iris van Rooij 💭 @irisvanrooij.bsky.social · Sep 7

Cognitive scientists and AI researchers make a forceful call to reject “uncritical adoption" of AI in academia
www.bloodinthemachine.com/p/cognitive-...

Cognitive scientists and AI researchers make a forceful call to reject “uncritical adoption" of AI in academia

A new paper calls on academia to repel rampant AI in university departments and classrooms.

www.bloodinthemachine.com

1 44 100

Reposted by Claire

Jonathan Kasstan @jrkasstan.bsky.social · Sep 5

Haverford College is looking for a tenure-track Assistant Professor of #Linguistics, with a primary specialization in #Sociolinguistics, secondary specialization(s) open -> www.haverford.edu/provost/news...

Tenure-track Assistant Professor of Sociolinguistics

Haverford College seeks a tenure-track Assistant Professor of Linguistics.

www.haverford.edu

28 32

Claire @chirila.bsky.social · Sep 6

Another lefty here who uses fountain pens with no problems!

1 2

Claire @chirila.bsky.social · Sep 4

It can't be worse than the people with Chinese character tattoos that they think say "spiritual harmony" and in fact say "compostable plastic". If you're willing to risk a tattoo that says "bile duct" or "powdered goat dung" go for it

2

Claire @chirila.bsky.social · Aug 25

That's what I send to the fakely friendly spam about scheduling a consult to use generative ai in my writing

1

Reposted by Claire

Schoenberg Institute for Manuscript Studies @sims-mss.bsky.social · Aug 15

Did you miss yesterday's #CoffeeWithACodex with @lisafdavis.bsky.social, featuring the realistic facsimile of the Voynich Manuscript? Fear not! It was recorded, and I just posted it to YouTube. Take an almost-look at one of the world's most inscrutable manuscripts.

youtu.be/LlwjwLYTJug

Coffee with a Codex: Voynich Facsimile (Beinecke Library (Yale) MS 408)

YouTube video by SchoenbergInstitute

youtu.be

1 12 32

Reposted by Claire

Ordo Fraterna Fibonacci @ordofibonacci.bsky.social · Aug 16

Coffee with a Codex: Voynich Facsimile (Beinecke Library (Yale) MS 408)
SchoenbergInstitute
youtu.be/LlwjwLYTJug?...

Coffee with a Codex: Voynich Facsimile (Beinecke Library (Yale) MS 408)

YouTube video by SchoenbergInstitute

youtu.be

1 2

Reposted by Claire

Dominic @dmncschmtz.bsky.social · Aug 15

Excited to see Gender Linguistics officially launched during LILG 2025 here at @hhu.de – and to share this moment with so many colleagues passionate about language and gender.

Looking forward to your submissions!

#linguistics

div-ling e.V. @div-ling.org · Aug 14

One highlight of LILG 2025: the official launch of the new open-access journal Gender Linguistics!

Published by our association, the journal will be a space for research at all linguistic intersections of language and gender.

Learn more: gender-linguistics.org

#linguistics #journal #launch

1 6 17

Reposted by Claire

Heidi Harley @heidiharley.bsky.social · Aug 15

STAR's first issue is out! (STAR is an open-access journal publishing syntactic research from the same editorial team that brought us Syntax) #Linguistics

star-linguistics.org

Syntactic Theory and Research

star-linguistics.org

1 20 46

Claire @chirila.bsky.social · Aug 14

He was born a hundred years after the manuscript was written