fg137

If I were setting up HA it is exactly because I don't want the shenanigans of Google Home and Alexa.

Seeing this, I don't think I'm going to touch it. Alexa might be more respectful of user data.

Yeah.

Seeing this, I don't think I'm going to set up Home Assistant any time soon.

My experience is that people don't typically change their behavior. Even if they do, I don't see any issue with bringing up this GitHub issue on HN which is a very suitable place to discuss technical topics (and others)

What is unforgivable?

Is there anything wrong with (not) going with a solution based on your personal belief?

You are free to make your own decision.

Was thinking of setting my own assistant using Home Assistant to replace the Google Home device I got rid of. Looks like I should slow down and take a closer look at this.

As someone whose smart home need is minimal -- I just use it to turn on/off lights with voice -- I'd rather have nothing than support such behavior.

That discussion happened way before the MOOC enthusiasm died.

It's one thing they can brag about, you know. Dumbing everything down to a single number, regardless of whether it's meaningful, feels impressive.

The vast majority of LLM users -- not the average person here on HN -- doesn't know which "model" they are using, can't tell the difference and don't care, including many software engineers. You could ask the same thing about those LLM benchmarks.

The prototype thing is quite unique among mainstream languages. It is kind of simple in principle and works well for what it intends to do, but god it causes so much confusion and is frequently a source of CVEs. valueOf combined with implicit conversion is a nightmare for production code.

> optional API and only implemented by V8

aka "Chrome implemented this and wanted everyone else to do so as well"

A few PT are getting much more lenient towards ratio, seeding time and other requirements (some used to ask you to meet certain upload "goals" in a quarter), seeing the trend and what's coming for them -- there is just less overall activity, including fewer torrents and less seeding/downloading.

Yet a few stick with all the existing, stringent rules they have, as if they are irreplaceable. I haven't touched them for a while. Let them die.

Copilot+ PC is such a terrible and wasteful marketing effort that Microsoft has given it up.

NPUs are not inherently useless, but definitely so the way they are forced onto users machines.

They intentionally skipped QuickJS because it's "too complicated" (paraphrasing) which I think is a decision they'll regret.

Been paying $12/yr for Firefox Relay Premium and have been happy with it so far. You get a dedicated (non customizable) subdomain.

> We've built a minimal JavaScript interpreter

Huh?

Well, Azure is really a piece of art.

I'll save my nice words elsewhere.

But can you use Apple Password app on an Android device?

That question has been answered multiple times here.

... looks like this should be a reply to GP not my comment.

Many of the large enterprises don't use a monorepo. Amazon doesn't have one.

For the rest that do, the vast majority of them are going to buy some Perforce licenses instead of vibe coding their own solution.

Unless you are Google/Meta where your company is large enough and its revenue supports paying a team of engineers (a lot of money) to do that, or you are Anthropic who is just vibe coding everything which can turn out to be good PR, you simply don't have the resources or can afford the risk to develop your own VCS solution.

This may be inspiring but out of question for most companies.

Yeah, that paragraph reads really weird. The entire article is basically "We could have considered Perforce but didn't even bother, instead we built our own VCS." yet they don't want to send the "wrong message".

With the way they are shipping code, source control history has no practical value. If barely any code older than 5 days still exists in the codebase, and nobody reads the commit logs, you can throw out most things in a source control system. It is basically a database of latest version of code, that's about it.

I am about 80% certain that using well configured Perforce could work just as well while saving Anthropic some money, considering the salary of Anthropic engineers.

But then Anthropic doesn't get the PR, so...

If you spent some time reading Anthropic's docs on their website, you'll notice that sometimes the same feature has different intended behavior depending on which page you are reading and may or may not match the actual software behavior. You are better off using the docs as inspiration and figure out what the software does at runtime. And I have definitely done the thing where I just ask Claude to reverse engineer its own binary. I doubt anyone at Anthropic can explain what it's supposed to do.

So basically -- it's all crap.

Sure. So if that's what you come here for,

I hereby confirm that you are not the only person who is aware of Rust's borrow checker but not aware, until now, that it has a history of serious soundness bugs.

Hope that settles this meaningless exchange.

I think you are talking about a different problem and that has nothing to do with my comment.

> I'm not the only person who ...

This is not even the first soundness bug that appeared on HN frontpage. It just means you don't pay attention to Rust development, that's it. I don't even use Rust at all yet I know a number of quirks about its compiler.

Because it's rarely used like this, just like weird bugs caused by edge cases found in all kinds of software in the world.

> is definitely something you'd easily come across in normal code.

The fact that nobody reported this before means it's not, by definition.

This is not the first soundness bug. Some bugs aren't fixed for years. The Rust community is well aware that such bugs exist, will be found, and address them just like other high impact bugs. In other words, if you paid the slightest attention to the development of Rust language, you would know this is surprising but not unexpected, and has happened many times.

> CGI's decline was justified in 2004 but doesn't hold up in 2026.

I mean, the article should not be taken seriously on any technical level. It's ok just for fun.