1shooner
1shooner on Hacker News
But what you're calling 'talking things out' I consider to be forming thoughts. I want to do that myself upstream of the LLM. I don't want to pay per token for my half-baked stream of consciousness.
This is unfounded pessimism. The genesis of the article is civic engagement that rejected Flock, prompting elected officials, whose decision is then being supported by the police department.
>The nexus of canonical, shared reality now exists online.
I'm not entirely sure what this even means, but I belong to several communities that get together in person and talk about very non-online things. They use email to coordinate, that's about as online as they get. And it's not hard, or a big deal. They aren't neo-luddite revolutionaries or anything, the internet just doesn't come up.
>Am I so out of touch
I don't think the HN audience generally appreciates how many people have fundamental problems with AI in it's current form. Today's ingredient replacement hint is tomorrow's comprehensively enshitified content product.
Also, there's a cultural dimension to cooking; it's not just getting a tasty dish out of it, there's provenance and process, the story and the cache. To some people automating that process defeats the purpose, and I bet a lot of those people would be the audience of a subscription recipe service.
Juul was essentially this scenario.
Sooo, are they reading the md?
This is basically the needle exchange argument. Except instead of just needles, you just want to hand out the drug.
I agree that if a parent is dead-set on giving their kid endless screen time, tobacco, or heroin, they will find a way to do it. But the tobacco story demonstrates there are many interventions that can significantly decrease the net amount of abuse. If you want to take a public health argument, do it honestly.
You're asking if I have a strong argument against targeting children with tobacco products?
This is also a great argument for low-tar cigarettes for children.
The first time I heard "When you hear hoofbeats, think horses, not zebras" I was appalled. How about we just empirically determine where the hoofbeats are coming from before eliminating the killer zebras as a possibility?
>But that rewriting the base system, every couple of years
To be fair, they did it once, from v7 to v8. But yeah, a major release does break APIs. I'm concerned with how Drupal Canvas is bolted on, but it looks like it's basically a content editor UI that is separate from the site admin UI, which is a good idea.
>You can just Claude Code all kinds of shapes of software.
Drupal is used at large organizations with hundreds of sites, thousands of users, and multiple more-or-less coordinated development teams. It's actually a great target for agentic coding, because it has such a well-established internal API. It can definitely feel like an anachronism, but I think there are few projects that could replace some of it's use cases. It's the B-52 of CMSs.
It isn't quite as bad as when the codebase was switched to Symfony, but the Drupal project has just kind of abandoned it's core presentation system in favor of the new Canvas tool, and compatibility with the rest of the core seems like an afterthought. So now you have at least 3 competing display configuration solutions, which don't necessarily work together, or entirely work alone.
It's a situation entirely inline with Drupal's reputation: they couldn't just get it right, and now serious Drupal teams don't know which way to go in the ecosystem.
>why do I need a middleman for something llm do out of the box?
Why use a token-burning middleman when someone has already done it for you?
>This is just because agents don’t see ads right?
Put another way, because it undermines the revenue that is providing you the service. Working around it is not a sustainable solution.
> is not an option, unless we're ready to first stop all human driving and have the same process for those.
Why?
There are really easy to user docker app managers out there. The author says 'I already use systemd'. I mean, I guess I'm using systemd, but really I'm using e.g. runtipi with an app store, web-based config and logging, and a super easy app template. I can point claude at any project with a docker release and in a couple minutes have it show up in my app store to install, run, simple data backups, etc.
I think calling out AI is pretty 'mainstreamed' at this point, if that's our criteria.
My research on chargebacks to Amazon was that they just ban you.
This is not my experience. I had a package simply not arrive, and Amazon AI said it would replace the missing package (which was a high dollar value). Instead, it just repeated the order, double charging me. I had 3 conversations with human CS promising me they saw the mistake and would fix it, and never did, so I was double charged. After months trying to resolve it I eventually gave up and ate the cost.
I've never had such a poor experience, including being repeatedly lied to by CS, and this was the last straw for me. I honestly don't miss Amazon, I buy less stuff and feel better about it.
I agree, but I think the concept of data provenance would have a role. It should not matter if data is laundered through 'aggregation' or 'anonymization': if it was originally generated about an individual, and used to target an individual, that data should be an expensive regulatory liability.
There was a recent Show HN project that claimed to be 'self-hosted', and the installation included something like 6 service subscription dependencies. That's just a SaaS with more steps.
>Liberty also means living in an area free of crime
I don't think this is a consensus definition of liberty.
Merriam Webster via Wikipedia:
"Liberty is the state of being free within society from restrictions imposed by authority on one's way of life, behavior, or political views."
>it just (eventually) suggested the same thing a human doctor had already suggested
Of course, but what else would you think an LLM does? That is the insight: sifting through a sea of information and parsing it without the same emotional or other cognitive biases the original human recipient applied to it.
I agree, 'copilot' is overloaded even within the Microsoft portfolio.
An aside, this is a poorly-structured article. It's mostly screenshots of citations from other articles with no alt or any other way to actually search through what they're citing.
Almost 1 in 3 tracked spacecraft is Starlink, is that accurate?
I don't trust the administration to use the law in good faith, but this article is misquoting the DoJ statement. The articles omits the very relevant qualifier:
The article:
> Last week, the Justice Department instructed “citizens and noncitizens” that anyone furthering the “goals” of a foreign power in “any public activity” — including even “public demonstrations” — must formally notify the government to avoid arrest and prosecution.
The DoJ:
>“If a person acts as an agent of any foreign power to carry out any public activity, including (but not limited to) public demonstrations designed to further the propaganda or other goals of the foreign power, he or she must register with or notify the appropriate federal authorities”
I think Twitter is the best counterargument to this defeatism. I don't personally know anyone that uses the platform any more, and I don't recall needing it for anything for years. From that remove, it's now just the platform for lazy politicians, lazy journalists, and scammers of one form or another. If there's more exclusive information there I'm not aware of, I really don't need it.