Notices by Mallory's Musings & Mischief (malcircuit@thingy.social)

If you're wondering how El Niño is going, not only did Polo just go from a tropical storm to a intense hurricane in under 24 hours... it's a ***Category 5 hurricane***

Likely the most rapidly intensifying hurricane EVER

Just to be clear, Hurricane Polo is currently cranking like a top in the Pacific west of Mexico. Probably will make landfall in Baja California, but that's still several days away.

Also, there still hasn't been a single hurricane in the Atlantic Ocean yet. Suppressed hurricane activity in the Atlantic is a known and expected effect of El Niño, but it's still bizarre.

Everyone in the United States needs to strap in and get ready for some crazy and exceptional weather this winter.

Hugely simplified El Niño forecast:

California, east thru Georgia: wet and stormy AF
Arkansas, east to Virginia: Snow and ice
Midwest and Great Lakes: relatively mild and dry
Northeast: ???????
Pacific Northwest: Dry?

@dalias From what I know, that's only a suspicion at this point. They could have, but no one has any real evidence of it.

That being said, I wouldn't put it past them. They're all slimy fucks.

Once again, every human endeavor is ultimately about enriching the lives of humans. If your AI bullshit doesn't do that — regardless of how "smart" it is — it doesn't matter

https://youtu.be/Nbwv5wHQoj0

The point of The Millennium Prize problems in mathematics is not to simply solve the problems, it's to *understand them*. The problems are problems because the mathematical tools that people have invented/discovered over the years cannot handle them for reasons no one can fully articulate. We need new tools.

The fact that OpenAI might have "solved" the Navier-Stokes Equation is interesting but not particularly useful, because the AI cannot articulate how that proof is interesting or novel.

It's not rocket science, folks.

https://youtube.com/shorts/eb4e0qtr0E4

Either, this "AI" stuff is so capable and powerful that there is a plausible chance it could kill everyone (in which case it should be either heavily regulated or banned entirely), or it's mostly harmless cloud software with limited applications (in which case it makes no sense to invest so much money and resources into the technology).

Which is it?

For either situation, it would make sense for these companies to be intensely scrutinized and investigated, because shit isn't adding up.

"Here's some really easy regulation: arrest people. Put Sam Altman and Dario Amodei in cuffs, take em downtown, have a chat with them about the cybercrimes that their LLMs are doing. If a regular person did these things they would be in prison." — Ed Zitron

https://youtu.be/PENnaW9zGYI

Reminder: being kind is punk as fuck

@thomasfuchs I have been using Syncthing for years now to replace Dropbox in my life. I'm not sure it's exactly the right solution for you (I don't know how it handles "small teams"), but I was able to get it installed on my QNAP NAS (via a Docker container) for backing up and it works really well.

https://syncthing.net/

You are constantly changing from moment to moment. The atoms that you were made of a second ago are different from the atoms now. The configuration of your cells isn't the same as it was a second ago. The connections between the neurons in your brain have changed from reading this.

Every instant is like a one-of-a-kind painting, unique and different from all the others. Every infinitesimal moment is its own masterpiece, complete and whole. Every moment is enough. Every you is enough.

You are enough as you are right now.

A cherry tree isn't less of a cherry tree if it does not blossom in the spring. The flowers add to its beauty, not complete it.

That doesn't mean you can't change. There is a difference between being enough and being finished. Nothing is ever finished. Everything is in a state of becoming something else. Change is not about filling a void, it's about becoming something new.

If you can't frame your argument for a better world in terms of joy, you'll never succeed in the long run

@thomasfuchs Yeah, it was definitely nerve-wracking. The whole time I was watching, there was a part of me that was like, "this rocket could turn into a giant fireball at any moment."

FYI: For anyone in North or South America, there is a partial (96%) lunar #eclipse happening on the night of Thursday, August 27! Europe, Africa, Asia, and the South Pacific will be able to see it to some degree, but not necessarily the whole thing (check local information for timing).

I know that much of Minnesota, Wisconsin, and Illinois are expected to have clear weather. So if you live there, make sure to step outside around 11pm CDT to see the maximum!

More info: https://www.timeanddate.com/eclipse/lunar/2026-august-28

@thomasfuchs @svavar

Yeah, that's exactly what I'm thinking about. I view using LLMs for coding as a more sophisticated autocomplete function. I can see how it could be useful in basically generating a skeleton of some code you're trying to implement. As in, populating some of the more generic code structures that are tedious and repetitive to type out by hand. But that stuff is often due to the quirks of the language/platform, rather than anything intrinsic to coding itself.

I'd really like to do some pixelmaxxing myself, but I keep getting cloud mogged 😶🌫️

#astrophotography

@cafuego https://misanthrope.social/@cafuego/117062280570178563

What I mean by an "anti-problem solver" is not simply someone who creates more problems. No, that's rookie shit.

The anti-problem solver is someone who makes it impossible for a problem solver to do their job. They make themselves and others believe the problem isn't real, invent imaginary unsolvable problems that consume resources for no purpose, or directly interfere with established problem-solving mechanisms.