Very good points in this thread.

Weren't we supposed to receive money from Anthropic in compensation for them training on our books? I haven't seen a check yet.

Great thread. I've been collecting these for years (shout out to Jesse Shapiro's How to Give an Applied Micro Talk slides), and this is a good addition.

People always say "tell a story" but then don't explain how to do so! How is that helping? If we knew how to tell a story we wouldn't need the advice in the first place. 1/

More (2/2) from @clauswilke.com: blog.genesmindsmachines.com/p/read-out-a... (yes, read every word aloud... just read the post, it's good advice*)

Another thing I notice is topics where there are several camps, ideologically or methodologically. AI summaries don't always catch that dynamic and wind up in some sort of incoherent mish-mash, especially when it is not a big enough controversy to have spawned papers explaining the controversy.

On this topic: I recently reviewed a paper in my field that is completely wrong, because the authors made a subtle but critical error. When googling relevant keywords, the Google AI confidently cites this paper and repeats its main (unsupported) claim. 1/2

Great color palette, where -14C and +39C are almost the same color. Meterology training should include a mandatory course in data visualization.

The words are: one two three four (We like to hire people who can count.)

Too much chroma. There, saved you 3000 words.

Unpopular opinion: Going straight to AI limits your creatively, because it short-circuits the iterative process you need to develop new ideas.

This picture looks unreal. Moscow, today.

This picture looks unreal. Moscow, today.

Aggregating many experiments into one zero-shot protein language model score obscures that current models cannot meaningfully rank a set of fit mutations or prioritize new-to-nature functions @clauswilke.com www.biorxiv.org/content/10.6...

Our lab website is finally online! pacesalab.com You can find information about our research, publications, and on-going developments in the lab.

I'm partially face blind (prosopagnosia), so take what I say with a grain of salt, but I don't recognize Joe Rogan at all in this picture. (I can normally recognize the 2020's Joe Rogan in pictures, though he doesn't look anything like the Fear Factor Joe Rogan from the 2000's.)

I can't wait for them to sell as that'll do wonders for the SpaceX valuation.

I'm always very impressed with the BlueSky trending algorithm.

I'm in favor of limits, but I think they should be imposed on dollar amounts, not on number of grants. Why should somebody with three small grants get penalized over somebody else with one large one?

Texas wildflowers.

New paper about zero-shot predictions of mutational effects. There's a lot of misunderstanding about what such predictions can and cannot do. In brief, they're good at identifying deleterious mutations but bad at predicting fitness improvements. 1/ www.biorxiv.org/content/10.6...

Can somebody please do this for a seahorse emoji so LLMs can finally rest?

The actual challenge in slide preparation is how to structure the story and what evidence to use to support it. AI cannot help with either. We're failing students by not teaching what matters in slide preparation.

Bad advice. Ants experience massive inflation. A mature queen can lay hundreds of eggs a day. That's Zimbabwe-style devaluation. Soon you'll need billions of ants to buy a loaf of bread.

I'd like to remind people that hundreds of millions of people live within the natural range of screwworm, and it's fine. Buenos Aires, Rio, São Paulo, Cartagena, all great cities where people go on with their lives never once thinking about screwworm. It's a cow problem, human cases are marginal.

The open job postings board at ASMS is really brutal

British Airways now has free, high-speed WIFI all the way across the Atlantic. It's fantastic. I can get a full day of work in, instead of having to watch crappy movies. Currently posting from here.

They do sell in Latin, but they also sell in more useful languages, such as සිංහල (Sinhala). There are at least 16 million potential customers speaking සිංහල!

The manipulation is visible in the original image with the naked eye. Once you know it's there, it's totally obvious. Mind-boggling that somebody would decide to do such a bad image edit and then publish the image.

1/ The #Ebola situation in DRC is quite serious, but still controllable. An international consortium, led by INSP & INRB in the DRC, is curating an open-access data set and real-time dashboard to aid in the response. Why serious, but controllable? A 🧵 inrb-umie.github.io/EBOV2026_Epi...

I think this is many of us. I've requested the desert menu and then sent it back because none of the options were appealing. I'm not into ice cream, but I sympathize with the sentiment. I definitely wouldn't get the vegan panna cotta or vegan tiramisu.*

Makes sense, because accuracy and reliability matter the least when it comes to measuring things. It's fine if the AI replaces objective reality with its priors of how a company should be doing. What could possibly go wrong.

Looking forward to annual performance evals with AI employees. "You do not meet expectations. You keep making mistakes and when I point them out you just apologize but then they keep happening." "Yes, you're right. I'm so sorry. It won't happen again." "But it will." "Yes, you're right. I'm sorry."

I wish we could separate teaching and high-stakes assessment. There's an inherent conflict if one instructor needs to do both. Incentives would be much more aligned if classes focused exclusively on learning and separate standardized exams focused on final assessment.

This is hilarious. www.youtube.com/watch?v=bYPd...

"It is very important that we don't name the city in the headline or blurb, so you click on the link." The city is Santa Cruz in case you're wondering. And it's seen a population decline of 1.3% from 2024 to 2025.

I had a book that contained the complete, annotated listing of the OS of my first computer (C64), in assembly language. I would read that book to figure out how the computer worked.

These days, when you're calling an Uber in Austin, it's frequently a Waymo. They're everywhere and are taking increasing marketshare. I don't think we'll have people driving Ubers for much longer.

That the AI vibe-coding crowd can't even figure out how to run stuff in the background is too on the nose.

These things are driving around all over Austin, TX. Such a hare-brained design.

There are numerous AI tells here but the most obvious one is the bus is on the wrong side of the road.

The dog looks a bit confused.

The illustration is obviously bad, but did you notice the mustache?

If you and eight of your friends are in a room with Jeff Bezos, the average net wealth in the room is going to be ~$20 billion dollars.

As I've said before: It's mind-boggling that the biggest software company in the world is so bad at software. Microsoft software was low quality in the 1990s, and now 30 years later the software is still low quality.

I am happy to share a review I recently wrote on the design of peptide binders. It gives an overview of experimentally validated tools and discusses the challenges of why peptide design is more difficult than the design of classical protein binders. www.chimia.ch/chimia/artic...

Congrats to @jbloomlab.bsky.social for getting elected to the NAS!

Congrats to @jrossibarra.bsky.social for getting elected to the NAS!

The pogo sticks are bad for the lawn?

The last 60% of the movie is the people navigating the bureaucracy trying to get the university to accept the artwork without clearly documented provenance. The final scene is when the dean declares the the university can now finally accept the donation and it will be accessioned soon.