helle253

as the son of a (former) mailman, i would have loved an @usps.com email

My wife and I always argue about that, she loves Socrates and I find him utterly insufferable. Plato didn't do him any favors.

A daily note, with a bunch of checkbox items

theres a plugin that automatically rolls every unchecked item into the next day's daily note.

these two features fulfill 99% of my 'todo list' needs.

Given Rail's maturity, i would have expected otherwise - there is tons of Ruby/Rails code to train on, but... yeah.

OTOH, doing some side-project stuff in TS, and the difference is a little mindblowing. I can see the hype behind vibecoding WAY more.

I recently migrated a featureset from one Rails project into another, as a mounted engine, and ensuring isolation (but not requiring it!) has been tremendously helpful.

Two communities with distinct cultures, whose membership nonetheless overlaps, are still two distinct cultures.

They may influence each other through that overlapping membership, but that does not mean they're the same.

i was at the playground, trying to convince my daughter to go down the slide on her own.

She kept saying it was too scary, so I went down first to show her it wasnt scary. Then, still not convinced, she said there were monsters in the slide! I, of course, told her I got rid of them on the way down. She pondered for a moment, then decided it wasn't so scary anymore. Shortly thereafter she went down the slide herself!

It was a funny, insightful moment, negotiating her fears without invalidating them.

I don't know, its frightening to think about. I don't think its LIKELY, yet. but i cant say its UNTHINKABLE anymore.

Do implementation with minimal (not none, tbf) AI support, ask copilot if there are any obvious issues with and otherwise check my work.

This workflow has helped me catch quite a few gotchas I would have otherwise missed.

Coding assistants are really helpful for validating output, I've had much more mixed results trying to use it to generate novel outputs.

It's clear they're taking a 'the Court has written their law. Now let them enforce it.' approach, and it's not clear anybody is going to stop them.

I've always felt that the 'there are only two internet cultures: 4chan and tumblr' has felt somewhat accurate. Unfortunately moreso now that /pol/ and /r9k/ have taken over broad swathes of the internet.

It's sad to see how far this old haunt has fallen. Lurking /v/ in my early/mid teens was a formative experience for me. It wasn't as hateful as it was, until Gamergate.

I still miss hanging out on /v/ and /fa/. When they split /vg/ out into its own board, the colour started to drain from my experience.

Often, when we're having trouble persuading her to do something, I can tell her Eloise wants to do that thing. I'll use teeth brushing as an example.

Eloise, being an independent 'person' of sorts in my daughter's mind, accepts the premise that Eloise would want to brush her teeth, (we remind her, Eloise doesnt want cavities, and my daughter agrees with Eloise: cavities are bad!). So we go and pretend to brush Eloise's teeth in the bathroom. Once Eloise is done, my daughter is usually all too ready to brush her own teeth.

As one might imagine, this tactic would not work on most adults. but toddler logic ontology imbues her stuffed animals with a sort of pseudo-agency (sometimes daughter insists Eloise does NOT want to brush her teeth or whatever, but that doesnt happen often, funnily enough).

sometimes the chain of logic and premise is grounded in a different ontology entirely (this is what i am referring to as 'toddler logic' for shorthand). this doesn't invalidate the conclusions or chain of logic used to derive them, when you (adult) is using different ontological anchors, which leads to different conclusions.

being able to inhabit the Other's (toddler's) ontological world and navigate it with them helps sharpen their reasoning skills!

they can figure out the Adult Ontology stuff later and apply those reasoning skills then. It's important to let kids be kids sometimes :) Encouraging their conclusions, and building confidence in their reasoning abilities, these are important endeavors in their own right.

i have a 2.5 year old and its extremely interesting just how much her thinking is clearly 'logical', but it operates in what i like to call 'toddler logic'.

If you ask kids questions, and really listen, their lines of thinking are very clearly following an internal logic. It just isnt one that is entirely compatible with 'adult' reasoning - unless the adult in the room intentionally chooses to communicate in 'toddler logic' operators. Most adults are, sadly and imo, not very good at this!

That doesn't preclude quality from coming out of it eventually - one of my favorite examples is Walter Miller's 'Canticle for Leibowitz' which was reworked from several of his pulp short stories into a single novel. It's a really great book! I think the incentive structure encouraged volume over quality, but quality still did matter. There wasn't the same algorithm-hacking incentive at play, and you still had to get your works past an editor and approved. Those are important distinctions!

that said, as an aside, im a little nonplussed by the inclusion of hiphop as opposed to, say, 'pop' or 'country', music.

You could always settle in the Mojave, plant a grove of bristlecone pines instead.

I don't know if more boreal climates support longevity, I think it has to be some combination of temperate + dry + safe from natural disasters like wildfires and hurricanes

Differentiating between 5d chess and delusional behavior is not easy. I'm personally inclined to think its the latter, but there's admittedly room for debate.

The reason I think this is because it is failure mode that is not uncommon in highly driven, egocentric geniuses. Especially one individual who has stated on record that they are taking ketamine, a powerful dissociative drug.

I found some other articles that go into further detail, and name one of the lead researchers involved.

I've also found a paper published on this, though I haven't reviewed it myself: https://strathprints.strath.ac.uk/74530/1/El_Rai_etal_SPIE20...

I did a quick search, and found a couple other articles that expound on the details.

https://www.cnn.com/science/artificial-intelligence-archaeol...

> The article was written with the assistance of a news analysis system.

I wonder if there was any human involvement in writing this jpost article at all beyond a cursory review.

Funnily enough, the only place where this problem is well acknowledged is law/ government. It's a lot harder to greenfield a new project in that realm though!

Without tooling to run a serverless service locally, this is always going to be a sticking point. This is fine for hobby projects where you can push to prod in order to test (which is what I've ended up doing) but if you want stronger safeguards, it's a real problem.