Author | Lightnews

medievalmvb

@medievalmvb.bsky.social

2K followers 1.6K following 2K posts

medievalist (researching climate in medieval literature) | working remotely at Durham University, UK | lost decade at The SunBreak & Seattlest | outdoor photographer see https://mikevanbaker.myportfolio.com/work

mikevanbaker.myportfolio.com

Posts Media Videos Starter Packs

Reposted by medievalmvb

Irina Dumitrescu @irinadumitrescu.bsky.social · 6h

I was so very pleased to blurb Peter Jones' new book on the seven deadly sins and medieval psychology. It's an unexpectedly addictive read!

www.penguin.co.uk/books/469085...

Self-Help From the Middle Ages

What can a twelfth-century monk teach us about burnout, envy, or despair? Far more than we might imagine. In Self-Help from the Middle Ages, historian Peter Jones travels through Europe’s archives and...

www.penguin.co.uk

3 7

medievalmvb @medievalmvb.bsky.social · 8h

that’s the Universe telling you to stay and have another espresso at Verdi’s

Reposted by medievalmvb

Tom the Dancing Bug, by Ruben Bolling @rubenbolling.bsky.social · 10h

I'm pretty sure it was this Dick Cheney comic that got me fired from The Washington Post. It was the first time they refused to run one of my weekly comics (after over ten years), and my comic was discontinued soon after.
From 2007:

6 82 330

medievalmvb @medievalmvb.bsky.social · 9h

Cubano Libre!

lauren @lauren.rotatingsandwiches.com · 10h

your honor, my client could have thrown a far tastier and more valuable cubano had he ventured just one block further north

medievalmvb @medievalmvb.bsky.social · 9h

real centrist Dem v MAGA vibes

Ketan Joshi @ketanjoshi.co · 1d

just....enjoy this

Reposted by medievalmvb

When the Autumn Weather Turns the Leaves to Josh @shmeppy.bsky.social · 1d

Hey girl are you a spondee because you look double stressed

3 6

medievalmvb @medievalmvb.bsky.social · 1d

Sundar Pichai on Alphabet's Q3 earnings call:

"AI Overviews drive meaningful query growth ... as users continue to learn that Google can answer more of their questions."

Huh. Let's ask Google AI Overview!

82% are skeptical of the information provided? Gosh.

Screenshot from Google AI Overview:

Key User Sentiments
• Mixed Trust and Utility: While about half of users find the summaries at least "somewhat helpful," trust is mixed. Many users (around 82%) are skeptical and only "sometimes trust" the information provided.
• Accuracy Concerns: A significant number of users (over 70% in one survey) report personally experiencing at least one significant mistake, such as misleading health advice or factual errors, leading to a general lack of confidence in the feature.
• Generational Divide: Younger users (Gen Z and Millennials) are more enthusiastic and find the feature more useful and trustworthy, whereas older generations are more skeptical and tend to skip the summaries.
• Desire for Control: A majority of users (over 55%) would like the option to turn off Al Overviews, at least for some queries, indicating a preference for traditional search results in many cases.

1 3

medievalmvb @medievalmvb.bsky.social · 1d

one thing I love about this site is people who get glimpses of the 37 or so people who have adopted whatever extreme position & think to themselves, “My god, what if everyone thought like this?!” & produce a 23-skeet thread based on this prospect

*my quarterly meta Bluesky comment, I’m allowed

medievalmvb @medievalmvb.bsky.social · 1d

SOUP SEASON!

Reposted by medievalmvb

Solomon @solomonmissouri.bsky.social · 1d

if she can wink that means she been practicing looking a mens in the face and garnering their attention!!!!!!!!

1 9 130

medievalmvb @medievalmvb.bsky.social · 1d

I was just thinking it would be fun to refuse to move on from a bygone brain metaphor: “backprop”? no no the neuron simply “telegraphs” the new information directly

1 4

Reposted by medievalmvb

The SunBreak @thesunbreak.bsky.social · 1d

Some great values at SIFF today, including the Palme d'Or winner IT WAS JUST AN ACCIDENT, which we reviewed briefly at TIFF:

2 1

Reposted by medievalmvb

ZP Rosen @zaqdelinguist.bsky.social · 1d

Answer: no. ChatGPT has no understanding of what words mean. Collocation frequencies of words can’t bring you understanding of what those words mean, even though it might help you predict which words go together.

See:

2 25 89

medievalmvb @medievalmvb.bsky.social · 1d

There’s a point this could have come up in the article but he swerves to avoid it entirely. Not great, honestly, to see in the magazine that is so proud of publishing Oliver Sacks.

medievalmvb @medievalmvb.bsky.social · 1d

They’re afflicted with pundit brain: I don’t want their opinion, I want either a lit review or a study of a model’s explanatory power. And the author never fact-checking any of his own experiences bodes ill, as they say. “I wrote an app!” Did u tho?

Reposted by medievalmvb

hammancheez @hammancheez.bsky.social · 2d

“What is dating like in your 50s”

Man sniffing womans neck “you smell wonderful what is that”

“Anti inflammatory joint cream”

3 9 74

medievalmvb @medievalmvb.bsky.social · 1d

journeyman’s work but the article turns out to be more personal essay garlanded with quotations rather than any sustained review of AI’s intersection with current neuroscience

The New Yorker @newyorker.com · 1d

The rushed and uneven rollout of A.I. has made it tempting to conclude that it’s all hype. But its advancement may have profound implications for the field of neuroscience.

The Case That A.I. Is Thinking

ChatGPT does not have an inner life. Yet it seems to know what it’s talking about.

www.newyorker.com

1 2

medievalmvb @medievalmvb.bsky.social · 2d

and that’s just one line of thought. There’s more but the coffee is wearing off. Until next time. (end)

medievalmvb @medievalmvb.bsky.social · 2d

it’s a little Mozart for Baby AI, is what I’m saying there 9)

medievalmvb @medievalmvb.bsky.social · 2d

this post hoc sandboxing of “relevant” or “trustworthy” data is the kind of intuitive (uncritical) gesture, reflexively human I suppose, that ultimately makes little sense bc the lack of trust is in the agent’s powers of discernment not vocabulary 8)

medievalmvb @medievalmvb.bsky.social · 2d

philosophy, handmaiden of Theology, was pressed into service to explain it all but it was also understood that “it just worked”—these stories people told themselves (narrative, not sermons) had real emotional impact (on ppl w active imaginations) 7)

medievalmvb @medievalmvb.bsky.social · 2d

there’s much more to this guided meditation practice, but for this 🧵let me underscore that the idea was Christ:Truth :: thinking about Christ:thinking true things (true here more salvific than fact-based) 6)

medievalmvb @medievalmvb.bsky.social · 2d

gospel meditation quite specifically was meant to provide guardrails to untrustworthy human imagination, which left to its own devices quickly fell prey to 😈 distractions

the ontological truth of the Gospels came pre-verified ✅ 5)

medievalmvb @medievalmvb.bsky.social · 2d

in the parlance of their day, gospel meditation, where practitioners were encouraged to actively imagine themselves participating in scenes either from the Gospels or “based on” them, maybe you were in the next manger over when Jesus was born 4)

medievalmvb @medievalmvb.bsky.social · 2d

in Europe & adjacent isles in the 13thC a number of religious Xian types began instructing fellow religious & lay ppl on how not to hallucinate errors by using a domain-specific training set, to use the parlance of our day 3)