I'm going to finish this manuscript by the end of the week, I told myself

The narrator of my audiobook keeps pronouncing Xhosa as [hoʊsə], with initial [h]! - Usual English pronunciationː [ˈkoʊ.sə] - In Xhosaː isiXhosa [isì‖ʰɔ́sa] I don't expect audiobook narrators to know orthographies of Bantu languages, but I do expect them to look up unfamiliar words. #linguistics

Having fun with some animal sounds for practice stimuli in our AXB task. 🔊sound on!

We present a simplified, p5.js version of the vocal tract synthesizer from Neil Thapen's popular Pink Trombone, adapted for use in a creative coding assignment concerned with interaction design and absurd puppeteering. Interactive: editor.p5js.org/golan/sketch... Writeup: github.com/golanlevin/6...

when they close it all up, I do believe this will go down as the funniest video on the internet

I think the only hope is that eventually they’ll run so many ‘inconclusive’ UBI trials that everyone will de facto be on UBI even though they won’t ever decide to implement it.

This exists… At least I think it does?

Space-station specific accents are going to be so cool.

Doctors are AMAZED at these 10 ways to BOOST your immune system: The Tetanus vaccine The Measles vaccine The Polio vaccine The Mumps vaccine The HPV vaccine The Pneumonia vaccine The Hepatitis B vaccine The Rubella vaccine The Diphtheria vaccine The Covid-19 vaccine - @hislopmd.bsky.social

Descartes gave us all the tools we need to survive in an era of misinformation. Simply begin with the understanding that you are real but everything else is fairly likely to be the cruel games of a demon sent to torment you specifically.

@wmgbennett.bsky.social and I are teaching "Topics in the Phonology of African Languages" at the Seshegopuo Winter Linguistics School later this month!

Can you train a performant language model using only openly licensed text? We are thrilled to announce the Common Pile v0.1, an 8TB dataset of openly licensed and public domain text. We train 7B models for 1T and 2T tokens and match the performance similar models like LLaMA 1 & 2

Well this is a first: tornado warning during a plenary talk! (Everyone is fine; we all evacuated downstairs; warning is over). #acal56

Folks attending ACAL— Is there a bsky hashtag or feed? #acal56 perhaps?

Penny Eckert, on Bill's reaction to her not turning in problem sets during her 1st semester in grad school (bc she was worried they weren't good enough): After a few weeks, he made me fork them over, saying "If you give me shit at least I'll know you're capable of shit"

Happy Friday! It's #WorldPenguinDay so I thought I'd remind you that penguins do have knees, they just aren't where you think they should be.

New visualization tool alert! The vayr package version 1.0.0 is now on CRAN. It contains position adjustments for ggplot2 that help with overplotting in pleasing ways. My favorite is position_sunflower(). - install.packages("vayr") - alexandercoppock.com/vayr #rstats #ggplot2 #dataviz

Dolphins can be researchers now?!

Do you do Phonetics? Human-Computer Interaction? Anything sound-y plus computer-y? Submit to our special issue of Phonetica! Just a little one-page abstract due by 4/30, and you get to work with very cool special issue editors! www.degruyterbrill.com/journal/key/...

(One of) my issues with LLMs is that we don't understand the steps the model takes to respond to a given query. Researchers at Anthropic have developed tools to see how their model comes up with a response—and the results are much better than just asking the model "tell me how you did that."

Just got served this ad for an app called “Kiki” - complete with spiky mascot! #linguistics

Does anyone have (or know of) #linguistics sample job materials for PhD students going on the market for the first time? Looking to provide some good examples to my students.

Georgian words look so much like pretend African language words that I had to make a video about it. www.youtube.com/watch?v=B7E-... #georgian #chishona #lingcomm #scicomm #langsky #🗣️💬

This paper from CogSci last year has my favorite figure for explaining transparent rule ordering in phonology: escholarship.org/uc/item/1kg1...

Check out @lingsocam.bsky.social's great statement on why making English the official language of the US is a bad idea.

Tashelhiyt Berber is the best.

#rstats "Your outie enjoys programming in base R."

Paging @gretchenmcc.bsky.social #linguistics #giftlink (With apologies for the politics-adjacent content)

hate that almost every explainer is a video now, i can't ctrl+F a video

YOUNG MAN, there's no need to feel down i said YOUNG MAN, pagliacci the clown you can SEE HIM he is playing in town you should go and have a good time

🚨 The "penguins" data is coming to the base #rstats{datasets} in 4.5.0! @ellakaye.co.uk & @HeathrTurnr.fosstodon.org.ap.brid.gy r prepared an adapted and a raw version of the data set based on the {palmerpenguins} 📦 by @allisonhorst.bsky.social @apreshill.com and Kristen Gorman. 📈 Scatter plot:

I really liked this idea of using a histogram as a legend in a choropleth map (since land isn't unemployed; people are), so I made a little guide to doing it with #rstats, {ggplot2}, and {patchwork} www.andrewheiss.com/blog/2025/02...

Me, working on my little articles and book projects in February 2025:

❓The visual shapes of letters, contrary to phonemes, lack a unified description. 🔍 Using crowdsourcing, the letters of 43 scripts are completely described as sets of binary features. 👉 Compared to phonemes, letters require more features, which are still less informative in letters than in phonemes.

LSA joined dozens of scholarly societies in signing a letter that begins, "American science and innovation have advanced humanity for generations... But today, science is under threat." We commit to championing integrity and conveying the importance of science. unitedsciencealliance.org

New video just went up! #linguistics #L2 #phonetics #phonology #perturbedspeech

Oh hey @abraver.bsky.social, look, it’s that collection we’re editing!

Submit a chapter proposal for @wmgbennett.bsky.social and my edited collection on labphon/quantitative/experimental approaches to African language phonologies! [1-2pp abstract, deadline March 31] #linguistics

New YouTube video up here youtu.be/iZmGkLQFVXw #linguistics #phonetics #langsky

Tips on working with sign language data in R

Suppose you generated a sequence of 100 random numbers. Then one year later, you did it again. Do you think we could predict one sequence from the other? It turns out, we can! Now in press @ JEP:G with @samiyousif.bsky.social @actlab.bsky.social @robbrutledge.bsky.social; osf.io/preprints/ps...

Check out @wmgbennett.bsky.social's new linguistics YouTube channel, y'all!

Pronouncing “zotero” like “Totoro”

We’re excited to announce the SignGPT project, a large language model for sign languages led by Professor Richard Bowden (pictured). We’re building this LLM from the ground up, creating the world’s largest sign-language dataset in the process. www.surrey.ac.uk/news/signgpt...

One of my fave facts: The “R” sound that we have in English appears naturally in almost no other language except for the Beijing dialect of Mandarin, so it’s in almost no languages but used by a near majority of humans, since those are the most common languages spoken.

Say what you want about the merger of De Gruyter and Brill, but their new logo is clever. #linguistics

Shameless #LSA2025 plug: Come to our* session, ~Illustrating African Phonologies~! Phonetic/experimental/quantitative/labphon approaches to African phonology. Saturday 1:45–3:15 and 3:30–5:00 in Franklin 1. Schedule of talks/abstracts: aaronbraver.com/lsa2025 * @wmgbennett.bsky.social + me

Writing tip: After you finish, go back & cut all the adverbs. Then all the adjectives. Then all the other words. Eat the paper. Now you're free.