Our first #ScienceCommunication editorial is published! Here, we highlight one of our recently published studies on a classic "dinosaur trackway" from the 1940s 
https://www.tandfonline.com/doi/full/10.1080/08912963.2025.2479395
The study being highlighted can be read #OpenAccess 
here: https://www.tandfonline.com/doi/full/10.1080/08912963.2025.2461068
Did #ColliderFest today on the Viper HPC stand! It was fun talking to people about supercomputing/HPC and AI, but also exhausting.
Used my UMAP word cloud demo to talk about how foundational language models (e.g. like ChatGPT, Ollama, etc) work, and how hallucinations come about.
https://starbeamrainbowlabs.com/blog/article.php?article=posts/533-research-smflooding-vis.html
“In a radical break with tradition, the National Science Foundation (NSF) this week began to search through billions of dollars of grants the agency has already awarded for anything touching on topics that President Donald Trump has criticized. And NSF has blocked grantees and trainees from accessing funds while the review is underway, wreaking havoc across the academic research community.”
#science #scicomm #OpenScience #sciencecommunication #citizenscience
If you want to see some of my #teaching and #ScienceCommunication effort, check out this video I made for high school students for their science and technology fair in 2021. I host it on my #PeerTube instance.
@Emiliagnathus@circumstances.runThis will be a challenge to the fediverse connecting abilities: I need to make a costume of a #trilobite and of #Anomalocaris. I am looking for any help: someone has made one already and is happy to share the details, a costume designed willing to take up the challenge, a handwork hobbyist with a knack for sewing/cardboard magic who always want to launch their skills onto a bigger stage. We are making a #science show for kids as part of my #ERC project MindTheGap and we will be explaining #evolution using #fossils. #Cambrian fossils 
If you have any hints, contacts, old costumes, please get in touch. We are located in NL so we probably cannot commission costumes from overseas (too much risk of delay due to customs). #Outreach #ScienceCommunication #paleontology
Such hard work in this mediascape.
Stay on message. If you get diverted, say "good question" and then bring it back to climate change.
If asked about probabilities and if this could be non climate related, the answer is a simple "no". Do not let them open the "uncertainty" door. State "no" first, then if you must, you can hedge with "the odds of that are so low...."
Every single question, go back to climate change.
If you can, throw in "start with banning private jets" or whatever snazzy statement you feel like but stirring the pot for live interviews is the best. You won't get many chances to get your message out.
Stay United. Do not undermine other researchers. They will ask leading questions and try to trap you into saying something they want printed. Do not be afraid to take control. "I think you've misunderstood, let me rephrase that to..."
Serious offer for any scientist about to do media, ping me, I'll be happy to do some basic media training before go time. Make sure you know what to expect and most importantly, any traps they may try to lay.
When poor people are getting devastated by climate change, nobody cares. This is hitting extremely rich people so this is as much media as climate change will get. We need to make use of it.
@halcy@icosahedron.websiteIn the name of #ScienceCommunication from maybe not a PR department and/or journalist trying to write a fun story, explaining the paper as simply as I can, with examples:
What we are doing here is Packet Loss Concealment for audio (mostly speech) data. When you talk on an application like Skype, Teams, Discord, whatever, any Voice-over-IP system, your voice gets split into small packets which are sent over the internet one by one. But the internet is not 100% reliable - especially when you use something like WiFi, packets can just get lost sometimes. The receiving end now has a problem: It still has to play back *something*. That something can, of course, be silence - but that sounds extremely bad an distracting. Listening to this type of audio is high effort, if it is even intelligible at all:
#introduction post on this new instance
Hello! I'm faelif, also known as Ava. I'm 18 and am studying for my A-levels here in the #UK (double #maths, #chemistry and #physics). I program in #Python and am learning #Haskell. I'm #transfemme but I don't particularly feel connected with gender.
I'm interested in #mathematics, #STEM more broadly, #trains, #politics, #history, #linguistics, #TokiPona, #ScienceCommunication.
I listen quite broadly to #KPop, #JPop, #JRock, #musicals and #pop. I also listen to #ASMR.
Though I'm not very knowledgeable about these topics yet, I'm interested in #sewing, #HistoricalFashion, #JFashion, in particular #LolitaFashion and #EGLFashion, and #manga.
I used to be @faelif@mastodon.social [mastodon.social], and I'm also at @faelif@mathstodon.xyz and @faelif@bookwyrm.social.
@luckytran From his #Seattle days -a treasured gift from friends. He's a great science communicator.
I think it was from a fundraiser for Public Television.
#introduction post on this new instance
Hello! I'm Ava, also known as faelif. I'm 18 and I'm studying for my A-levels here in the #UK (double #maths, #chemistry and #physics). I program in #Python and am learning #Haskell. I'm #transfemme but I don't really feel connected with gender at all.
I'm interested in #mathematics, #STEM more broadly, #trains, #politics, #history, #linguistics, #TokiPona and #ScienceCommunication.
I used to be on the fediverse @faelif@mastodon.social [mastodon.social]. You can also find me on Bookwyrm @faelif@bookwyrm.social.
Okay, one more time for the people in the back.
The "AI" (
) craze of the past few years is all about Large Language Models. This immediately tells us that the only thing these systems "know" is trends/patterns in the ways that people write, to the extent that those patterns are expressed in the text that was used to train the model. Even the common term, "hallucination," gives these things far too much credit: a hallucination is a departure from reality, but an LLM has no concept of reality to depart from!
An LLM does exactly one thing: you give it a chunk of text, and it predicts which word will come next after the end of the chunk. That's it. An LLM-powered chatbot will then stick that word onto the end of the chunk and feed the resulting, slightly longer chunk back into the model to predict the next word, and then do it again for the next, etc. Such a chatbot's output is unreliable by design, because there are many linguistically valid continuations to any chunk of text, and the model usually reflects that by having an output that means, "There is a 63% chance that the next word is X, a 14% chance that it's Y, etc." The text produced by these chatbots is often not even correlated with factual correctness, because the models are trained on works of fiction and non-fiction alike.
For example, when you ask a chatbot what 2 + 2 is, it will usually say it's 4, but not because the model knows anything about math. It's because when people write about asking that question, the text that they write next is usually a statement that the answer is 4. But if the model's training data includes Orwell's Nineteen Eighty-Four (or certain texts that discuss the book or its ideas), then the chatbot will very rarely say that the answer is 5 instead, because convincing people that that is the answer is a plot point in the book.
If you're still having trouble, you can think of it this way: when you ask one of these chatbots a question, it does not give you the answer; it gives you an example of what—linguistically speaking—an answer might look like. Or, to put it even more succinctly: these things are not the Star Trek ship's computer; they are very impressive autocomplete.
So LLMs are fundamentally a poor fit for any task that is some form of, "producing factually correct information." But if you really wanted to try to force it and damn the torpedos, then I'd say you basically have two options. I'll tell you what they are in a reply. 
The majority of scientists (57% female cf 46% male) - have been attacked after reporting on research in the media.
Unsurprisingly, musk’s X has done the heavy lifting. Likely due to the scientific illiterate manfluencer cult the platform now attracts.
The popular story of evolution is oversimplified and misleading. I used to be frustrated that nobody was telling a better story. Then I discovered many brilliant scientists have been writing about this, extensively, for decades. There are fantastic books about the full complexity of evolution and how different it is from the dominant story. It's just hard to find them. You have to know where to look, who to trust.
This raises an interesting challenge. How do you change that dominant narrative? It's not like the good ideas aren't out there. I could write more about it. Run experiments to support my arguments. But how do you get people to take notice? How do you change the default explanations and references that everyone's familiar with? How do you get folks talking about the good stuff in secondary school, not just graduate level biology?
It seems like "more science" and "more writing" is not the most important thing here.
Historical document: Carl Sagan explains the greenhouse effect to the US congress and suggests to stop subsidizing fossil fuels in 1985. https://www.youtube.com/watch?v=Wp-WiNXH6hI
I’ve been gaining some new followers recently, so it’s time for a reintroduction!
I’m a scientist and macrophotographer, fascinated by all the amazing creatures that live in the soil! I use my photography and media appearances to do science communication on soil biodiversity, to spread awareness and love for the world beneath our feet. Oh, and I’ve got a major book deal to write a popular science book on this subject, which I’m extremely excited about!
Follow my page for interesting soil animal facts, macrophotography and updates!
NEW PTEROSAUR!
Inabtanin alarabia was a fairly large azhdarchoid pterosaur found in Jordan. Its generic name literally means “grape dragon”, or “grape colored”, which explains my color choice! 
If you like my little pterosaur cartoons, the best way you can support me and my work is by favoriting my posts and reblogging them!
Seriously, it means a lot! 
#newpterosaur #sciencecommunication #inabtanin #inabtaninalarabia
NEW PTEROSAUR!
Haliskia peterseni was an anhanguerid pterosaur from the early Cretaceous of Australia! It had an estimated wingspan of about 4.6 meters (≈15 feet), and might’ve ate soft ocean critters!
If you want to see more paleoart like this, please consider favoriting and reblogging my work! It really means a ton!!
NEW PTEROSAUR!
Torukjara bandeirae is a new species of tapejarid pterosaur from the famous Pterosaur Graveyard site in southern Brazil! It was named both after the Toruks from Avatar, and Kamila Bandeira, a Brazilian paleontologist who first noticed a distinction between what is now two separate pterosaur genera!
I’ll post the next new pterosaur soon! So be sure to stick around to see Haliskia! 
NEW PTEROSAUR!
Inabtanin alarabia was a fairly large azhdarchoid pterosaur found in Jordan. Its generic name literally means “grape dragon”, or “grape colored”, which explains my color choice!
If you like my little pterosaur cartoons, the best way you can support me and my work is by favoriting my posts and reblogging them!
Seriously, it means a lot!
Season 4 episode 5: Scientists Tell Stories Too (and That's a Good Thing), with Prof. Jenni Barclay
In this episode, Prof. Jenni Barclay explains the importance of storytelling by scientists to themselves, other scientists, and the general public.
Has that got you thinking, hey, wait, WHAT?! Everyone knows that scientists should never tell stories! If we expect them to show up like Back off man, I’m a scientist! and guide us through difficulties, then they’d best stick strictly and dryly to the facts, because everyone knows that scientists should never tell stories, right? But scientists even need to tell stories to themselves and to each other to more effectively process the information contained in their data, observations, and experiences. Human beings are not computers: we need stories to grasp the meanings of things, and that also goes for scientific facts. This means scientists need to be storytellers, too, if they want people to understand not just what the facts are, but what they mean for society and the world at large. Then people would better be able to see what our options are for responding to environmental and technological developments and emergencies.
