mikeocool

This case not withstanding, Apple regularly includes patches to the built in CLI tools in security updates, so it's a little disingenuous to suggest those tools are not really supported.

If this is Jev’s trough of disillusionment, I’ll happily trade them for my plateau of productivity.

Is that really different from anything else?

If you’re not upgrading your deps and rebuilding your rust project this morning, you’re pretty likely exposed to eMudhra. If your project was “done” a few years ago that’s likely not going to be easy.

Kind of, except as product manager you're held accountable for delivering the right product, while your engineering team is held accountable for things like performance, incidents, and security issues caused by their code.

It's hard to hold an LLM accountable for anything.

I am generally not onboard with the “all you need is SQLite” articles that pop up on HN regularly.

But I recently looked into self hosting and several things each wanted their own Postgres container. Immich wants a Postgres and a redis instance.

If you’re building apps targeted at self hosters, SQLite is really all you need.

I think you'll find most people on hacker news don't want to read re-packaged content, they want to read primary sources or original reporting.

This link seems to be re-packaging news stories as an ad for whatever auravision is.

The actual sources for this (that actually list the 19 car brands) are: https://news.northeastern.edu/2026/09/29/connected-car-priva...

https://www.consumerreports.org/electronics/personal-informa...

The real advantage of cloudflare workers is you don’t pay for the processing time while you’re waiting for upstream network responses.

I believe Cloudflare is the only FaaS that offers that. It makes it fantastic for putting a light weight layer in front of a few upstream API, like say inference providers.

Many business insurance policies come with requirements for physical security, if you've skimped on them, you're not getting paid out.

Sadly, even if this were the case, there is no mechanism for one NYC city agency to fine another one.

This is why the sidewalks next to city parks take forever to get shoveled after a big snow storm.

Finally came across while searching for “S3 server in front of file system.”

It seems like most other options for self-hosted S3 (Garage/SeaweedFS/RustFS) want to hide my files away in some sort of custom format.

Some of these requirements are so ridiculous that they create the need for everyone to do what Vance is accusing Microsoft of.

Many companies fulfill the intent of the law. They run a real hiring process — where they advertise online and interview people with work authorization/citizens and people who would require sponsorship.

But if someone requiring sponsorship was the best candidate for the job, then they have to run an additional “fake” process that fulfills these requirements, since no one actually hires tech workers from the newspaper.

I was a volunteer EMT for a while — and didn’t take it further for some similar/some different reasons.

One recommendation, if you want to take the class but aren’t planning on doing it professionally (and have any interest in outdoor activities), take a Wilderness First Responder class. I found I learned a lot more in that class, and much it is practical in many circumstances, not just riding on an ambulance. It’s also a lot more fun, as you spend a lot of time running around in the woods learning to macguyver splints out of all kinds of stuff.

The other thing I found was that to really be good at it/feel confident treating anything beyond the most basic patients, you’re going to have to work on an ambulance full time in an urban setting. And in most places, to do that professionally, you’re going to have to become a paramedic.

Ha, try the “turn off the viewing light” button.

What are you building? Is it successful? Does it support you in the lifestyle you’re aiming for?

Note that the author denies the claim. And it also appears that AI detectors are split on the matter, and likely don’t have a rich training corpus of Haitian-French to draw on.

His publisher also said the original draft was submitted in 2019, though they’ve also become lukewarm in their backing of the author lately (he was also just accused of classic plagiarism for an unrelated short story).

So hard to say this one is settled.

It seems like jev's major advantage over existing classifiers is that I dont have train it.

If I have to gather and tag data to fine-tune Jev, I can probably just train an "old school" classifier model and make it even cheaper, faster, and just as accurate.

Except there already was "something" that the creators of MCP just ignored.

Imagine a world where you could:

* Configure your favorite harness/chat client with any OpenAPI spec for an API that supports Oauth2.

* The harness would walk you through the Oauth flow and securely store the token.

* And then insert some tools for discovering the API methods and making requests in to the context.

* The agent could then formulate a request, call the request tool, and the harness would 1) makes sure it's allowed to make a request to that API, and 2) insert the Auth Token into the request.

It would be basically exactly the same way MCP is setup today, except all you would need is an OpenAPI spec. You wouldn't have setup a server for a janky new standard that's half implemented slightly differently by every harness/chat client.

Pretty sure Apple must dedicate a product manager to coming up with new ways to break spaces in each OS release.

Why is “are you using the latest model?” the automatic response to any even mildly critical of LLM coding?

OP wasn’t even actually critical of LLMs, they were just saying that plan mode was helpful to stop the model from making incorrect assumptions when you want you don’t specify everything you should.

I'm not sure it's strictly red state/blue state. I've live in Massachusetts and Minnesota for a while, where they made it very easy to vote so that tracks.

However, now I live in New York, where they seem to go out of their way to make it difficult. The registration cutoff is several weeks before the election and for at least one election, I found myself wrongfully expunged from the voter roles.

I think it may just be how corrupt the state government is.

Seriously, once you realize you’re paying a minimum of 30% extra, it becomes hard to stomach.

Weird, I've seen countless instagram posts telling me the five prompts I need to have ChatGPT run my account and grow it to millions of followers.

In my experience, the docs for something like setting up a dev env are typically greatly improved by the second person who sets up the dev env, not the personal who originally wrote the docs.

In that case, when the docs are not associated with a code change, you want to make getting those improvements into the docs as frictionless as possible, otherwise the changes aren't going to get made.

Personally, I've found that making docs updates incredibly fast + easy to be far more valuable than anything you get from forcing doc changes through the full SDLC process. If someone has feedback on your docs changes they would have shared in a review, they can just update the docs instead.

Why do they need real influencers? Don’t they own some of the world’s best AI models that are really good at generating content?

That's the load-bearing seam in this blog post.

I think the point of those comments is this could have been much simpler for everyone if the model providers had given us some tools around existing standards.

Instead they invented something new/weird/complex standard, and then every client implemented slightly differently (and different parts of it).

Yeah, from a design perspective MCP upsets me, because it’s a poorly designed standard and creating a good one could have been much easier.

But you’re right, since clients don’t have a nicely sandboxed “make api request” tool, it’s basically the way to go for a lot of use cases.

All of these things are perfectly possible with a plain REST API with an Open API spec and using some standard auth options, and an AI client that implements a reasonable “make api request tool” (just like the AI clients implement MCP today).

I think the real value of MCP is that it allowed companies to say “we’re doing AI!” When they built an MCP server. Just saying “use our api” was a lot less exciting.

Giving it a different name probably also helped cut through politics at companies where non-technical people didn’t want to open up user data with an API, but they did want to do AI.

I believe you can use your codex subscription with Pi, and it’s officially endorsed by OpenAI and uses your subscription usage.

Though for cc you are correct, using it with your sub draws from extra usage.