iambenm

"Please don't post comments saying that HN is turning into Reddit." [0]

[0] https://news.ycombinator.com/newsguidelines.html

The number of ads I have gotten lately about buying ads on Apple Maps has me glad I'm not using Apple Maps. It turns out they're just late to the ad game, not opposed to it.

Hmm. The project uses BirdNET-Go, which does inference locally on a Raspberry Pi. Then it looks up the species against a set of already existing illustrations. So in this case it seems like the "ultra expensive, data-center-driven AI" is not really involved in the core loop of this project.

Rate limiting

You can /i-have-adhd when you get a wall of text and are exasperated. You don't always want the responses that it generates when you /i-have-adhd, and things in AGENTS.md tend to lose weight as context grows.

It's Genetic Programming via LLM, cool!

How do you differentiate between mastery of a topic and mastery of using an AI in an exam setting?

Open book tests and allowed cheat sheets already solve the memorization-is-not-required gap. It seems like allowing use of AI pushes that closer to thinking-is-not-required. And if no thinking is required, then how can you prove personal mastery?

First, cloning a repo doesn't bring the hooks with it - so it's not just as simple as saying "use git hooks".

Second, if the repo had hooks and instructions for the LLM or user to blindly install/enable the hooks, we'd instead be complaining about security risks and what might happen if the repo is compromised at some point in the future.

Third, sometimes you don't want to mechanically enforce things via git hooks because it impacts your use when what you're really trying to codify and enforce are the LLM's actions. In that case you can enforce mechanically via hooks at the harness level.

And finally, git hooks are a great solution for upstream repositories to enforce quality and protect branches. But it means that the upstream is the one running the checks. It makes the upstream a potential bottleneck - better to have the leaf nodes run the checks locally and fix any issues before pushing it upstream rather than push upstream, wait for results, make changes, push upstream, wait for results, make changes.

Code appears to already be up: https://github.com/alexmorleyfinch/marlin

My read on this, beyond data and MBA's, is that all of the slack has been pulled out of the economy.

Ablation is a medical term, he's just using it generically to describe removing existing config.

Hmm, do video games fall into the same category?

India is pretty serious about being able to monitor all communications after the terror attacks in Mumbai in 2008. The terrorists used satellite phones to coordinate the attack. As a result, they banned most satellite communication devices.

I learned about this earlier this year because I had a connecting flight through India on the way to go trekking in the Himalayas. I had planned to take a Garmin inReach with me, but luckily a day or two before my flight out I came across horror stories of people getting pulled out of security and taken to JAIL in India for having these things in their checked bags when TRANSITING India. Security there when transiting from Switzerland to Nepal was no joke - every single electronic item including anything with wires had to come out of bags including electric toothbrush.

In light of that, banning point to point communication seems on brand for India. Note that Apple gets a pass for its satellite capable phones by geofencing India and disabling the feature.

I think it's more about keeping this forum one of the few places where you can have open minded conversations that don't delve into hivemind guided flamewars. Granted, I don't know what got warned or the reason.

Anything approaching nuance on other forums gets downvoted to oblivion and attacked if it doesn't fit the particular echo chamber it's in. That just reinforces the echo chamber. HN has to do quite a bit to preserve the culture and keep it from devolving into something like those other places.

There's another underlying theme when you consider Forgejo having Codeberg as a home - anti-commercialization / sensitivity to commercialization and breaking of trust in the ecosystem. I could see anti-LLM sentiment stemming from these things a bit, in terms of what the LLM's were trained on. It's like an evolution from the olden days of slashdot commenters ragging on M$ every chance they got - only now it's AI and slop.

The approach I took was to start with the $20 plan, then when it became clear it wasn't sufficient I upgraded to the $100 plan. The $20 plan didn't cover one night of coding. With the $100 plan I started bumping into the 5 hour usage limits after ramping up over a week or two.

My average API equivalent use is around $30-40/hr. I would just bite the bullet on a plan for one month, then use that to calibrate your expectations around usage and cost optimization. The plans are heavily subsidized.

Look at the subscriptions. That's how anyone using the stuff as an individual should be approaching it today if they are dealing directly with Anthropic or OpenAI.

With the $200 Claude subscription I was able to get around $13-15k of API equivalent usage in one month (note: this was during the "+50% usage" promotion that they have kept extending since May). When you hit your usage limit for a given time period you get cut off until the time period resets; don't bother paying for additional usage credits, you will be disappointed.

Spoiler alert, it seems to be like Wordle in that the same puzzle is given to everyone for that day.

What came to mind to me was a bailout shaped to benefit the public even if the company failed post-bailout, specifically the Chrysler Tech Center and how it was designed as a building that could be repurposed as a shopping mall if it fell through. Odd, interesting place.

Modeled on Chrysler, perhaps?

In the automotive industry it's not uncommon for contracts to require an on-site engineer, basically FDE.

I got it around 36, and coincidentally a friend 5-6 years younger than me in a different part of the country got it at the same time.

My research at the time led me to the same conclusion: Since we've basically eradicated chickenpox, we're not exposed to the varicella zoster virus as much as previous generations (via children that have chickenpox). Without exposure our antibodies / resistance fade out.

Since we had chickenpox, the virus is already in us and dormant. When it wakes up and decides to do its thing, our bodies have forgotten how to fight it effectively compared to previous generations - and as a result, the incidence of shingles is increasing in younger populations compared to previous generations.

My doctor said vaccination for under 50 is only indicated after multiple occurrences of shingles... so here's to hoping I'm good til the guidelines change.

Hmm - the article isn't dated and it doesn't mention which models were used for the initial slop version. The initial commit in the git repo is from Feb 15, 2026.

I wonder how the initial pass would fare now with Fable 5 or 5.6 Sol?