Flash was everywhere. Now it's gone. Sound familiar?
Flash is just part of the story of VANISHING CULTURE, a new book exploring the fight to preserve our fragile digital history. 📖
Join us for the book launch!
📅 Apr 23
🕠 Doors 5:30 PM
📍 300 Funston Ave, SF
🎟️ www.eventbrite.com/e/vanishing-...
Humanity did that. Science did that. Publicly-funded research did that. Excellent universities did that. Diversity did that. International cooperation did that.
Artemis II is a perfect example of what we can do at our best.
Welcome home, Integrity crew!
I think we do not yet appreciate how completely overpowered people picking up agentic coding in their thirties today will be in their sixties and seventies.
Juniors have always had the boost of youth in performance. But that is now moot really. Productive aging is possible.
books: the rectangle that can’t send you push notifications. Try: books.
That may have only mattered before the models had sufficient context. And before compute was cheap.
I am imagining that it will be a lot more of a challenge for "attention to abundance" than say "rocket steering".
If you use tikz-cd as usual it should be as good as we can expect. One day they may have some additional annotations possible for each diagram package.
For now we try to do the heavy lifting on the development side of things, mapping the CDs to SVG+MathML .
We indeed have a page: info.arxiv.org/help/submit_...
It boils down to using the most semantic markup that LaTeX classes and packages provide, and aiming to mostly fit the supported list of packages in LaTeXML.
We still have a road ahead of us to succeed in more cases, so it's also a dev burden.
🗓️ The March 2026 arXiv articles are now in ar5iv.
In March, arXiv passed
👉3 million total articles
👉30K monthly submissions
Roughly 2️⃣ million of them have error-free HTML in the ar5iv Lab.
— and, even worse, it turns out there’s an uncanny valley for “worth spending time on”
Sorry for the screenshot from the other place, but this is the best thing I’ve learned from the Claude Code leak so far
Another first in this project. I've written MathML for my blog post to explain some of the math that is involved with projecting walls in 3D space.
Me reassuring myself in an act of self-care: You worry too much about what people think of you. I'm sure that no one noticed that your outfit didn't match the other day.
Me reading the author list of a paper on the arxiv this morning: Oh that's the guy who wore those weird pants that one time.
If there are many layers between the people using the software and people working on it, that iteration can't happen. If users don't have any choice in the matter, then that iteration won't be forced to happen. Relatively trivial things, like our experience, stay broken forever.
The software industry knows how to make software good. Pay close attention to your users to see if they're succeeding at their tasks and happy with it, through some combination of interviewing them directly and monitoring how they use it, then fix broken things.
Just a bug sadly... But in any case, the submitted TeX sources are publicly available so comments, keys, macro names could all lead to embarrassment eventually.
Elsevier spends $2B/year on tech. Springer Nature has spent hundreds of millions. Sobering numbers for non-profit tech initiatives trying to compete newsletter.journalology.com/p/journalolo...
Why are people adopting AI to write?
The last few weeks I have witnessed a number of interesting discussions breaking out on social media. A couple of weeks ago a US-based academic admitted using AI in some of his writing, which prompted an angry response from a prominent AI researcher. I'm not…
Every tech person I know that’s gone deep in AI comes away realizing that Claude can already do a surprising chunk of their job and this is the worst the tech will ever be.
How far behind tech is the rest of the world? Optimistically 3-5 years, but big tech's pace since November has surprised me.
In the penultimate talk of the meetup, Kyano rings the alarm bell: HTML with packages is the wild west! What standards could the ecosystem converge upon? Let's kickstart a discussion before it's too late.
www.youtube.com/watch?v=KETM...
Hit refresh: redesigning our technical infrastructure
Our core systems handle 2 billion API requests a month — and some of that code dates back to 2005. We're redesigning it piece by piece to serve a community that's grown from 318 to 24,000+ members.
More about the redesign: https://doi.org/10.64000/8mckt-w8m69
Further details on arXiv going independent: like us the aim is improved sustainability, tech, governance, and global engagement tech.cornell.edu/arxiv/
As the leading AI models move into longer-horizon and more specialized tasks, mediated by complex and expensive gate-keepers in the U.S. economy (e.g. legal or healthcare systems), I expect large gaps in performance to appear. Coding can largely be mostly “solved” with careful data processes, scraping GitHub, and clever environments. The economies of scale and foci of training are moving into domains that are not on the public web, so they are far harder to replicate than early language models.
Go read @natolambert.bsky.social's latest post. (www.interconnects.ai/p/the-next-p...) Very smart and sobering about the challenges/opportunities ahead. Open intellectual commons is not going to happen automatically.
Germany does not lack talent, and it does not lack funding. But we are trapping 21st-century minds inside 19th-century academic hierarchies. We are asking brilliant young scientists to build the future of the German economy, but refusing to give them the lab space, the job security, or the scientific independence to actually do it. If we want to reclaim our place as an industrial superpower, we have to stop the rat race of trying to keep every technology and structure alive that made us successful in the 20th century. Instead, we must fix our system that pushes our most ambitious scientists away. The money is there. The talent can be there. Now, we also need the courage to fix what’s broken.
“we are trapping 21st-century minds inside 19th-century academic hierarchies.” This essay gets a lot right about problems with German science. I would add that the hierarchies and precarious contracts lead also to systemic abuse and scientific misconduct. open.substack.com/pub/realimag...
Не мога да кажа. Там има 35 години история които само повърхностно разбирам. Надявам се някой ден да напише книга за архива.
All those AI companies destroying our scholarly infrastructures will definitely be queuing up to support arXiv now it's becoming an independent organisation and seeking financial sustainability, yeah?
"I got my PhD by writing prompts instead of doing research, I'm winning"
got some bad news, there still no jobs and now you also know nothing
