Friday Links!
I'm 64 today and have temporarily accumulated enough physical problems in the last 30 days to make an 80-year-old blush. We have a massive links drop this week.
Leading off this week, a riveting analysis: ‘My patient was happy with her partner of 25 years – then started a torrid affair’: a psychotherapist on why people cheat.
A tragic tale: ‘There’s a dangerous epidemic in boxing’: the tragic, cautionary tale of Paul Bamba.
Public service: Everything you need to know about bird flu.
If you want to know how an incompetent fool determines tariffs (that aren't actually in any way related to tariffs), I've got you: Trump’s ‘idiotic’ and flawed tariff calculations stun economists.
From DQ Iditarod correspondent Meg McReynolds: Chasing the Iditarod Through the Wilds of Alaska. Also, a very good girl: Meet Muppy, the World’s Smallest Sled Dog.
From Wally, and this is quite stupid: People Making AI Studio Ghibli Images Are Now Producing Fake Legal Letters to Go With Their Fake Art. Fantastic: Let Britain’s magical, mythical creatures inspire a patriotism untainted by politics. Award-nominated SF stories you can read: Analytical Laboratory Finalists.
C sent this to me and it's genuinely stunning: Excess Mortality Rate in Black Children Since 1950 in the United States: A 70-Year Population-Based Study of Racial Inequalities.
From Ken P., and we live in a police state now: Surveillance shows Tufts graduate student detained. It's dangerous and we're stuck with him, for now: Politics March 24, 2025 What Was the Plan Behind This Fake CDC Website? Leading, as always: Texas is poised to make measles a nationwide epidemic, public health experts say. This is concerning (because of the model's origins): DeepSeek-V3 now runs at 20 tokens per second on Mac Studio, and that’s a nightmare for OpenAI. Not today, clowns: I won't connect my dishwasher to your stupid cloud. Amazing! Three Hundred Years Later, a Tool from Isaac Newton Gets an Update. This is mildly encouraging: Facebook to stop targeting ads at UK woman after legal fight. Very cool: New Portal pinball table may be the closest we’re gonna get to Portal 3.
From C. Lee, and refer to my previous police state comment: Weekslong lockups of European tourists at US borders spark fears of traveling to America. Useful knowledge: Here’s what you need to know about your rights when entering the US. An excellent read (or listen): How empathy came to be seen as a weakness in conservative circles. This is incredible: DOGE to Fired CISA Staff: Email Us Your Personal Data. This is helpful: Trump’s ‘climate’ purge deleted a new extreme weather risk tool. We recreated it. Confessions: Did I Really Do That? Well, that's aggressive: Impaling for Love ― Bull-Headed Shrike. A fascinating bit of history: See you in the funny papers: How superhero comics tell the story of Jewish America. A terrific read: Inside RGG Studio: Ryosuke Horii and Eiji Hamatsu Share How the Like a Dragon Series Is Developed Quickly Without Sacrificing Quality.
AI (your email)
I received this from Sean R. after my AI post earlier this week:
As a longtime reader of your blog, I'm saddened to see you posting so much A.I. content of late.
Respectfully, there are in fact many reasons not to use A.I., from the blatant theft that powers so much of what they do (Miyazaki himself is opposed), to the ways this tech is destroying the livelihood of the creative class (of which I am a member), not to mention the environmental harms their giant data centers cause, and the ways in which they routinely hallucinate answers which can have serious consequences, especially if you're using them for medical purposes.
Beyond that, there's an inherent soullessness in even the best A.I. art and "writing," as it is the product of statistical regurgitation offering (at best) second-hand imitations of true insight and actual human experience.
Compare the image you shared with any frame of any Studio Ghibli film; while the art style is similar, what is being conveyed about the characters and their world is night and day. Empty calories compared to a home-cooked feast.
Reasonable people can disagree about the extent to which A.I. tools can be used ethically and effectively, but I don't think anyone can argue that there's any way to use these generative tools in particular without causing at least some harm.
I don't agree with everything Sean says, but I think he raises some substantial points, and I'm going to write about them next week.
Oh, the Humanity
Nintendo had their Switch 2 showcase today.
Wii? I knew it was going to be great from the first moment I saw it.
Wii U? I had a bad feeling. A very bad feeling.
Switch? I was on the hype train from minute one.
Three for three, in other words.
My initial reaction to the Switch 2? A disaster.
Besides the hardware changes (which were not insignificant but also not revolutionary):
1. Paid upgrades for better graphics on certain high-profile Switch 1 games you already own.
2. Game prices of $70-80 for first-party Nintendo titles.
3. Console price of $449 in the U.S.
If the console was announced at $349, almost everyone would have been pleased. At $449, it's going to be a very, very hard sell.
Incredibly, it could have been worse.
In Europe, the equivalent conversion to USD for both the UK and the European Union is $510.
I have an enormous amount of affection for Nintendo because of the hundreds of hours I spent playing games with Eli 23.8. It's a wonderful company.
And they've screwed up.
Tomorrow
Sean R. sent me an excellent email in response to Monday's post about ChatGPT 4.0, and I'm going to use it tomorrow. It raises some interesting and thoughtful points. The only reason I'm not doing it today is because of Nintendo's Switch 2 Showcase.
Framing
The framer called me on Friday. Here's what the game sheet looks like:
You can't see the color inside the wood unless you click on the image, but it's quite beautiful, and the weathered nature of the gamesheet turns it into a real piece of history. Family history, anyway, and Oxford history, too.
A Day With ChatGPT 4.0
Yes, I do believe it's possible that AI will one day destroy the world.
Absolutely.
I think it's far more likely, though, that humans will destroy the world and blame it on AI.
In either case, there's no reason not to use AI in ways that make us happy. While we wait.
First off, I uploaded this picture (one of my favorite adult pictures with Eli 23.7:
Then I asked ChatGPT to create a new image in the Studio Ghibli style. Here's what it gave me:
Eli's scruffy beard is gone, but otherwise, it's amazingly good. Seeing it made me very, very happy.
I tried a few more times, but I couldn't tell the program to take "the father from the first image you created" and "the son from the third image you created" because it just doesn't work like that. What I essentially wanted it do was a basic Photoshop operation after generating a few images, which was impossible.
I sent the image to Eli and it made him happy, too.
Then I used the "deep research" option to research acid reflux, which I've been having problems with for the last few weeks (along with accompanying nausea at times). I input my symptoms, duration, and what I'd already tried. In seven minutes, it returned a 14-page document. Logically presented, thorough, and with source citations.
It would have taken me hours to do that by myself, and I'm sure I wouldn't have found much of the information.
The image generator is wonderful fun to play with, a toy, but the deep research option is a legitimate research option and an absolute game changer.
Dishes
I'm going to use dishes as a metaphor for life. Don't blame me.
When I see a dish in the sink, I'll wash it out. I don't know why, really. I just don't like seeing dishes in the sink. It's a visual thing, plus it uses space. I put most of the dishes in the dishwasher after I wash them, which is another subject entirely.
It's not something I make a conscious effort to do. It's automatic.
I realized yesterday that doing it that way is easy. Nothing ever piles up. In an organizational sense, it's clean.
Gee, that seems like an easy way to do life. Only I don't.
I struggle all the time with putting off tasks I don't want to do. I spend far more time putting them off than I spend doing them.
Part of it could be that taking care of tasks often spawns a set of side tasks. This happened all the time after Gloria's death. Tasks seemed to exponentially multiple as soon as I started on them.
It's not like that now, really. Financial and medical things are the ones most likely to spawn out of control. That's about it, though. Eli 23.7 might message and ask for a Panda by next Tuesday, but his asks always have a tight timeline attached, so they don't last for long.
I'd be interested in hearing if any of you are highly successful in this area. Maybe I (and all of us) could learn something.
Framing
I think I mentioned a few weeks ago that I was going to frame a copy of the game sheet from the Rivalry Match, and also mentioned how unusual it was of me to frame anything.
More than unusual, really. I've never done it before.
I had an excellent picture Eli 23.7 sent me, so I was planning on having it framed in the next week or two.
Last Friday, I went to get the mail and there was a padded envelope from Eli. It's very rare that he sends something from England, and I had no idea what it was.
I opened it up, and it was the game sheet. He'd sent me the original. I called him and he said it made him happy when he thought about sending it to me, so he did.
I took it on Saturday to a framing place, a real one, and spent about half an hour with a very thoughtful person discussing how framing works and what it accomplishes if you do it the right way.
I'll share a picture when I have it back. Oh, and here's what the framing place looked like:
It looks like my brain.
It's Crowded in Here
I became aware of something last week for the first time.
The only way I can write a novel is to have it all in my head at once. I work on specific sections, sure, but all 90,000 words are in my head.
At first, I feel like I'm just infinitely rearranging a warehouse of Scrabble tiles. That's why 90% of the first draft is gone now.
At some point, though, those Scrabble tiles start breathing on their own. They add color. They become vibrant. It's the best part, which is where I am now, before I see everything so clearly and realize it can never be as good as I want it to be.
It's still hard, and annoying. I spent three days looking for a crowbar to wedge into a particular chapter (that needed to be completely rewritten, of course). If I just keep showing up every day at roughly the same time, though, failure gives up.
That's not a bad way of describing the process of writing a novel: making failure give up.
Things I noticed today
Remember the 23andMe craze? On January 2, 2021, the stock closed at $258.80 a share.
Today, it closed at $0.73. That's right, seventy-three cents. And if you want a cherry on that, they filed for bankruptcy.
If you have results from them, you might want to go onto the website and delete them, if you haven't already, due to uncertainly over who might purchase the company.
Second, as the Patron Saint of Lost Causes, I'm of course attracted to the upcoming baseball game between Yeshiva University and Lehman College on April 8.
Why? Because they've lost a combined 124 games in a row.
I was in NYC only a few weeks ago. If only I'd gone in April. You know I'd be there for the entire doubleheader (that's right--two games, not one).
I'm not sure when the other three (!) episodes will be available, but it's worth the wait.
GPT Shanties, day three
ChatGPT
said:
Got it!
Thank you for the clarity. I’ll keep it tighter, more rhythmic, and stripped
down—no excess words. I’ll aim to match the simplicity and flow you’re looking
for with "Tide's Lullaby."
Let’s try
again with a more concise approach, focusing on rhythm and purpose.
1. "Roll
With the Tide"
(Verse
1)
Roll with the tide, the wind’s on our side,
The sea’s in our blood, and the stars are our guide.
We’ve sailed through the storm, we’ve weathered the night,
Roll with the tide, and keep our course right.
(Chorus)
So haul, lads, haul, the ship’s steady and true,
The sea is our mistress, we’re ready for you!
With a shout and a cheer, and the wind in our sails,
Roll with the tide, and set the wind’s trail!
2. "The
Ocean's Call"
(Verse
1)
The ocean’s call is loud and clear,
She pulls at our hearts, draws us near.
Through the mist and the roar, we’ll sail on and on,
The ocean’s call is where we belong.
(Chorus)
So haul, lads, haul, don’t hesitate,
The ocean’s our mistress, it’s never too late!
With the waves at our feet, and the stars in our eyes,
We’ll follow her call, ‘neath the endless skies!
3. "Sail
to the End"
(Verse
1)
Sail to the end, through the fog and the rain,
The sea’s our home, we’ll sail once again.
With the wind in our sails and the stars overhead,
Sail to the end, where the horizon’s red.
(Chorus)
So haul, lads, haul, the course is set,
The sea’s our home, we’ve no regrets!
With the tide in our veins, and the wind at our back,
Sail to the end, we’re on the right track!
These are
more streamlined and should flow better in rhythm. Let me know if any of them
work or if you'd like further adjustments!
You
said:
Sail To
the End is quite good. Thank you.
ChatGPT
said:
You're
very welcome! I'm glad "Sail to the End" hit the mark. If you have
any more tweaks or want to try another round, just let me know. I'm always
ready to help you fine-tune it further. Happy sailing, and may the wind be at
your back!
__________________________________________________________________________________
Taking the lyrics from ChatGPT, I then went to Udio and went through three iterations of song generation. Which I would share for you now but for the radical stupidity of multiple platforms involved in doing this. I'll try to figure it out over the weekend and have it for you on Monday.
GTP Shanties Final Product
After a ridiculous series of steps comparable only to the Manhattan Project in complexity, here's the final result: Sail to the Horizon.
GPT Shanties, day two
ChatGPT
said:
Here’s my
revised set of shanties based on your input:
1. "Sail
Away, Bold and Free"
(Verse
1)
The wind it blows, the waves they rise,
We sail for treasure 'neath open skies.
With a song on our lips, and a heart full of cheer,
Sail away, bold and free, no fear!
(Chorus)
So haul, lads, haul, don’t let her stray,
The ocean’s callin’, we’re on our way!
With the wind in our sails and the stars to guide,
We sail away, bold and free, to the other side!
2. "Heave,
My Hearties"
(Verse
1)
Heave, my hearties, the shore’s far away,
The storm’s on the horizon, but we’ll sail through the day.
We’ve weathered the winds and the heavy, dark seas,
So heave, my hearties, and follow the breeze!
(Chorus)
Heave, lads, heave, we’re strong and we’re true,
The sea’s our mistress, and she’ll see us through!
With a shout and a cheer, we’ll chart our own course,
Heave, my hearties, to the wind and the force!
3. "Tide’s
Lullaby"
(Verse
1)
The tide’s lullaby is callin’ us near,
The ship’s hummin’ softly, no more fear.
The stars are our lanterns, the night’s our friend,
We’ll sail ‘til the dawn, and then sail again!
(Chorus)
So lift your voices, and sing with me,
We’ll sail the wide world, just wait and see!
With the tide at our back, and the wind in our hair,
We’ll chase down the horizon, no matter where!
I’ve kept
the lines tighter and more rhythmic to suit the feel of "Roll, You
Rogues" while making sure they still have that adventurous, bold energy.
Let me know which one you like best or if you have any tweaks to suggest!
You
said:
Sail Away,
Bold and Free is a mess. It's as if your mischievous younger brother broke into
your room and accessed your computer. Heave, My hearties, is still too verbose.
Tide's Lullaby is the best. Remember, these have to be sung in a rhythm, and
any unnecessary words act as friction and slow the words down. If you're going
to introduce friction in a song, do it with purpose, not as a matter of course.
Try again.