Nate Angell
banner
xolotl.org
Nate Angell
@xolotl.org
#communications & #community @nudgital.com. Wandering IQ. Raised by wolves. Friend to cheese. he/him. All posts #CCBY. Mastodon: https://social.coop/@xolotl
Pinned
There's a deep disorder in the discourse of generative artificial intelligence: Read Are We Tripping? The Mirage of #AI Hallucinations from me and @annamillsoer.bsky.social to learn why you should replace the term "hallucination" with "mirage" in your discussions of AI: xolotl.org/are-we-tripp...
Are We Tripping? The Mirage of AI Hallucinations
There is a deep disorder in the discourse of generative artificial intelligence, aka AI — or what I like to call sparkling intelligence, because everyone is using ✨✨✨ emojis and icons to signify AI. A...
xolotl.org
Reposted by Nate Angell
Very kind review of my translation of Césaire's lost play by Laëtitia SAINT-LOUBERT. I like that she noticed a few things others had missed. Hope someday I get to thank her in person for the kindness.

www.jwilonline.org/wp-content/u...
www.jwilonline.org
November 10, 2025 at 8:22 PM
I'm old enough to remember when we used to call accounts like these "bots" and think of them as unwelcome information disorder...
Meta confirms they plan to add tons of AI-generated users to Facebook and Instagram

They will have bios, profile pics and can share content

(via Financial Times)
November 10, 2025 at 3:30 AM
Little known fact but Clinton’s entire presidential campaign was modeled on tetris, but unfortunately an opponent created a hole she couldn’t fill with a standard tetris shape…
across all wikipedias in all languages, vietnamese wikipedia's "game boy" article is the only place that uses the public domain shots of hillary clinton as a tetris addict taken on air force one
November 9, 2025 at 3:07 AM
November 9, 2025 at 1:58 AM
Reposted by Nate Angell
Someone tell Webster I want to propose a new word

errogance (n) : the state of being very confident and very wrong
August 6, 2023 at 8:16 AM
we wouldn't even need to misconfigure DNS if people would get off their lazy assess and just use numerical IP addresses
November 4, 2025 at 2:15 AM
Reposted by Nate Angell
I totally did not read this correctly the first time I
November 3, 2025 at 10:40 PM
Reposted by Nate Angell
controversial take nobody asked for: trying to hold a "middle-ground" position on the usefulness of genAI feels like being satisfied with 'regression to the mean' as acceptable standard 🤷🏾‍♀️
November 3, 2025 at 9:21 PM
Reposted by Nate Angell
Hello from Time-change-with-a-toddler Land!

THE SCIENCE SUGGESTS THAT YEAR ROUND STANDARD TIME IS BEST.

I know it’s the least of our worries right now, but let’s get some data driven policy going and end this hellscape of twice a year time change.
med.stanford.edu/news/all-new...
Study suggests most Americans would be healthier without daylight saving time
According to a new analysis by Stanford Medicine scientists, changing clocks twice a year disrupts circadian rhythms, leading to higher rates of stroke and obesity.
med.stanford.edu
November 3, 2025 at 11:43 PM
Reposted by Nate Angell
thank you for all the support; the statement from our union is now live here with some more reporting.

now that this is public I can confirm that the majority of today’s layoffs were women of color. there are no longer any Black women working at Teen Vogue.
November 3, 2025 at 11:25 PM
Reposted by Nate Angell
Gen AI is navigated through abstractions, which is ultimately reductive - and by definition, imprecise. You will never get an image of a specific thing, just some abstraction of the thing, and a problem of the 21st century was already that abstractions were flattening specificity.
October 30, 2025 at 7:16 PM
nobody expects the Monty Python Software Foundation!
October 30, 2025 at 7:52 PM
Reposted by Nate Angell
I wonder how many "the Andrew formerly known as Prince" jokes are being made globally rn
October 30, 2025 at 7:31 PM
"you blink an eye, and literally a third of your country is eating up the dreams of fascist grifters"
October 30, 2025 at 7:06 PM
Reposted by Nate Angell
No neede for expensyve consultinge firmes. Heere ys a planne for excellent higher education:

1) Hire manye facultye
2) Small class syzes
3) Emphasyze engagemente & delighte
4) Sustayne & cultivate a varietye of fieldes and disciplines, even currentlye unfashionable ones -- thys ys ed, not retail
October 29, 2025 at 4:25 PM
This info should be included in all US voting guides. The US political reality seems like it has has almost no grounding in economic reality.
I’m doing a “Seven Days of Data Stories” project, where I’m trying to get better at building interactives while sharing things that I think are important.

This is the first! Visualizing the flow of tax money, and why 11 states pay into the federal government while the rest are dependent.
The Federal-State Money Story
Interactive visualization of federal spending patterns across U.S. states. Explore which states receive more federal dollars than they pay in taxes, and how federal money flows through state economies...
dr.eamer.dev
October 29, 2025 at 2:22 AM
Reposted by Nate Angell
What could go wrong…?
October 28, 2025 at 10:51 PM
forget patent law, I’m wondering if AI can answer a question about Betteridge’s law
Can AI hold office hours? This study evaluates #AI tools’ ability to answer specific legal questions based on a #patentlaw textbook used by many law schools.

Read: spkl.io/63324Akb1M
Subscribe: spkl.io/63328Akb1u

#AICommunity #LawSky
October 28, 2025 at 4:44 PM
Reposted by Nate Angell
the mexican restaurant we like is feeding anybody who asks for free food as long as SNAP is shut off so now i'm going to eat my weight in cheese and onion enchiladas "to support them"
October 28, 2025 at 1:39 PM
The New York Rangers are the only NHL team without a mascot.
October 25, 2025 at 5:08 AM
Reposted by Nate Angell
Pride is not the only emotion that you're allowed to have about your history. There are many others that can and do apply. It should not be a failing or a threat to your sense of self to acknowledge those other emotions. To speak them aloud as you watch history unfold.
October 23, 2025 at 7:59 PM
This paper on AI’s moral status is both flawed — because it requires consciousness for moral status — and unnecessary — we don’t need it to know that humans will not treat AI well if AI has moral status. Humans already treat everything badly, regardless of moral status.
Could we conceivably know what it is like to be an AI model, and thus what its ethical status should be? Intriguing argument from Helen Yetter-Chappell that we're in a worse epistemic position than wrt animal consciousness. #aiethics
Helen Yetter-Chappell, Guessing at Ghosts in the Machine - PhilPapers
As AI grows ever more complex and ubiquitous, its moral status becomes ever more pressing. But knowing whether an AI has moral status is only part of the ethical puzzle. To ...
philpapers.org
October 23, 2025 at 2:51 AM
@wyden.senate.gov can you please talk to Jeff Merkley and get him off KOSA and Twitter?
October 22, 2025 at 5:36 AM
Oooooh. Look at what the mail brought today! Luscious Jackson album hot off the presses /cc @kates66.bsky.social
October 21, 2025 at 12:35 AM
Reposted by Nate Angell