splittydev

Ask Claude whether a man can be a woman. No matter what political side you're on, models are censored. I'd argue American models are a lot more censored/biased on a lot more topics, and especially on things that come up a lot more than some highly specific Chinese politics. Claude even refuses to translate song lyrics because of copyright.. I wish we could simply have uncensored models from all sides, but it's pretty clear that's not happening right now.

Honestly, I think that's a good thing. A lot of questions were either duplicates of existing questions, or close derivatives of them. If I had to guess, probably 90% of SO questions already had a solution somewhere on SO. AI surfaces these solutions much quicker, so you don't have to ask. Novel questions or bugs that can't be answered or fixed by AI still get asked, and mods have less spam to deal with. I fail to see the issue here.

Wasn't the entire goal of Ladybird to have an open and independent browser engine? Making it effectively closed to contributions makes it.. Not independent anymore. It's now dependent, on few people who work on it, just like any other closed-source or corporate-controlled browser.

Alternatively, you could try compiling an Xcode project. That should do the trick as well.

This is common practice in many countries, although I don't quite understand why

Yes. And I'm sure they're having a hard time. ^

The same thing happened when we transitioned from horse carriages to cars. I'm sure a lot of people were quite outraged. But aren't we glad it happened?

Sure, you're allowed to hate whatever you want. I never said they're not allowed to hate AI. I said they're gonna have a hard time in the future if they can't accept that the times are a-changing'.

These people are going to have a really hard time coming to grips with reality in the next few years. AI is here to stay, and it's expanding very rapidly. If you can't fight them, join them.

I've had pretty much the same experience with my AI chat app. Nothing works well. Markdown rendering is slow and laggy, streaming is slow and laggy, everything locks up the UI. I've tried at least 5 of the most popular text editor components for UIKit and SwiftUI on GitHub, and all were broken in one way or another, buggy, and slow as well. It's ridiculous.

Amazing read. Blog posts rarely keep my attention like this one.

Honestly, this kind of thing seems to work quite well with vibe coding. If I remember correctly, the Ladybird JS engine was "vibe-ported" to Rust as well, and it passed 100% of the original test suite, in addition to new Rust tests.

I kind of see your point, but I also kind of don't.

Sure, it would be great if you'd immediately get hammered with hundreds of requests and start make money quickly. It would also be great if it was a bit more transparent, and you could see more stats (what counts as "idle"? Is my machine currently eligible to serve models?). But it's still very new, I'd say give it some time and let's see how it goes.

If you have it running and you get zero requests, it uses close to zero power above what your computer uses anyway. It doesn't cost you anything to have it running, and if you get requests, you make money. Seems like an easy decision to me.

They released this like a day ago, I'm not surprised that there's not enough demand right now. Give it some time to take off

Please see my other comment here :) We do not collect any private data, and we never will. We only collect very generic and fully anonymized usage data, but that does not include typed characters, words, clipboard history, snippets, or anything else that could be considered private.

Hey, we're super transparent about the data we collect.

We collect zero data about your typed words, personal dictionary data, stored contacts, clipboard history and basically anything else that's privacy-sensitive.

What we do collect is very generic and fully anonymized metrics such as: Do you use a theme, did you modify the keyboard, did you add an emoji key to your keyboard, etc.

We are not interested in typed words or any private data. We just want to know how people use the keyboard in general (which features in particular), and that's all we collect. You can opt out at any time, and all collected data is automatically deleted every 30 days because we only keep a 30-day rolling window.

If you want to be extremely safe, you can also skip enabling full access for the keyboard, which makes it impossible for us to send data from the keyboard itself to the app. But as said, we don't actually collect any privacy-sensitive data (and never will), and disabling full access comes with a few other caveats because Apple put many basic features such as vibration etc behind the full access setting as well, for whatever reason.

I'm one of the developers of Mister Keyboard. If you want, you can give it a try! Everything essential is completely free, maybe it works out for you.

Yeah, this is the great irony of it all. Germany really wants to discourage taking a car for "environmental reasons" and so on and does everything to encourage public transport.

But one thing is clear: I won't be bothered, robbed or even stabbed in my own car, and I also won't arrive in a different village lest I drive there myself. I won't arrive three hours late either, or have to stay overnight in some shitty Hotel because they couldn't find a replacement train.

The German public transport, like many other things in Germany, is an absolute fever dream for a "developed country".

All of these are incredibly obvious. If you have even the slightest idea of what you're doing and review the code before deploying it to prod, this will never succeed.

If you have absolutely no idea what you're doing, well, then it doesn't really matter in the end, does it? You're never gonna recognize any security vulnerabilities (as has happened many times with LLM-assisted "no-code" platforms and without any actual malicious intent), and you're going to deploy unsafe code either way.

Is there any context on why? Is there some controversy regarding RubyGems.org I'm not aware of?

I have contributed to a lot of open source projects for many years. They sure don't need my contribution, but if it were more accessible, I might actually contribute one day.

Given that ffmpeg was apparently using mailing lists before (which I didn't know at the time of writing), this is obviously an improvement. But it's still a lot of friction, and it seems that it's friction for no particular reason.

Hosting your code on something as obscure as Forgejo actively makes it hard for people to contribute. Most developers use GitHub, and only have experience with GitHub.

If I have an issue with something, I'll make a PR and move on with my day. But if making that PR involves creating an account on some obscure website and learning a new user interface just to contribute code to a single project, I'll simply pass. It's not worth the hassle.

GitHub is the most popular service by far for a good reason. People build their whole professional portfolio on GitHub, and it's well-known and globally recognized.

If you're afraid GitHub is gonna go down one day or kick you out, just mirror your repo to your own git instance. It's been done for decades, and it works very well, without introducing friction.

OpenRouter is great. Less privacy I guess, but you pay for usage and you have access to hundreds of models. They have free models too, albeit rate-limited.

Great, another VS Code fork that could realistically just be an extension.

Yeah, I checked the mail source too. Passed DKIM, SPF, DMARC etc, so the mail server is definitely compromised.

They seem to be using SendGrid. I pinged the CEO and CTO of Autodesk, the official Autodesk account and the SendGrid account on X about this, but now, more than 24h later, the attack is still ongoing and nobody seems to be giving a flying fuck about it.

It's ridiculous that they're not reacting to this at all.

I contacted Autodesk on X, as well as the CEO and CTO, but nobody seems to care so far.

Doesn't work for me. Pressing the button does nothing.

Holy shit for a second I misread that title as containing the word "Brainrot"

Haha same https://imgur.com/a/DHDSzPW

Fun game, but I soft-locked myself https://imgur.com/a/DHDSzPW Took a few tries to get there, but I don't think there's a way out

Of course, a well-crafted prompt and RAG or other means of retrieving relevant information can work very well.

The reason why I was asking is that as a potential customer, I kinda feel cheated reading about a langcss.fast model that doesn't actually exist. I don't think there's any need for putting this misleading (well, actually not even misleading; just plain wrong) information on the website, to make potential customers think you've trained a custom model on tailwind etc.

Honesty is a very attractive selling point, at least to me, and I'd bet it is to others too.