The LLM simply agrees with you and you're happy. It is VERY worrying that you don't realize this, even after reading this article.

I just tell a new instance and a different provider the core idea and see if they like it too

Trouble is an LLM can test for something being logical in isolation, or coherent unto itself. It’s much weaker at anticipating what will be meaningful to other people which is usually what people are actually looking for.

Nice joke, hadn't seen it coming

Sounds like AI-written, eh? :-D

(esp last sentence?)

It can be traced, but the people who could and should be tracing this have been told not to.

Even if they found the culprits, what's a judge's verdict against a presidential pardon?

> Even if they found the culprits, what's a judge's verdict against a presidential pardon?

This is bad logic. A presidential pardon at least exposes the corruption, and exposing the corruption is more important than a prison sentence.

Is there anyone who isnt Trump4Life who still needs to be convinced theres corruption happening very high up in the government?

We've very little doubt its happening, and it does us no good if there arent actions taken against it.

It still needs to be methodically exposed. The process is important.

What if people don't care because it's their guy?

... that's the same question I responded to there.

If the culprits have been pardoned, they'd still be able to be sued by people who lost out during those trades.

So in essence they

- can give away your data for free - get hacked by nation-state such as Iran - get hacked by mercenary spyware and not notice

and their statement would still be correct. Now that's an awful lot of qualifiers. Plus that's just what they say.

iOS Safari. I see a yellow banner, the navigation bar and the rest of the screen is just a warning sign image.

Is there more..?

Checked on Chrome too, I see nothing.

iOS Chrome

What do you mean? You see nothing on the website?

I captured the full page, you can view it here: https://wormhole.app/MbljK6#qfysvKJOQh1whLcMz9JXxw

This is the screen on my phone.

https://wormhole.app/9Xv0p0#Hsq0fhLpWsr8ndJDktt2YQ

You see the little "Red hat Enterprise" at the bottom? That's the whole scrollable area. The rest is fixed and stays at the top.

Must be an iOS thing. On Blink and Gecko the page just scrolls like normal, the warning scrolls with the rest of the page when scrolling down.

They still messed up the CSS because the downloads table goes straight beyond the mobile viewport on the bottom and to the right.

1. You can add a bookmark that executes enough JavaScript to download the VSIX as usual. 2. I think you can patch the product.json from VSCodium to use VSCode. Gets overwritten on every update probably.

Honestly though, it's easier to disable ~three settings in VSCode and call it a day.

This. Their current approach is open sourced. There's no going back any more.

Also note that our use case is much simpler. The programming language tells you whether your merge created a valid document.

I've never seen that information actually being used in any merge tool, with the notable exception of Visual Studio/C# (where you get symbol resolution for the merged doc, but even there the autogenerated result is a bit hit and miss)

I think the reason is that the algorithms want to be content-agnostic.

But it's of course weird — as a user — to see a conflict resolution tool confidently return something that's not even syntactically valid.

I develop prototypes using Claude Code. The dead boring stuff.

"Implement JWT token verification and role checking in Spring Boot. Secure some endpoints with Oauth2, some with API key, some public."

C# and Java are so old, whatever solutions you find are 5 years out of date. Having an agent implement and verify the foundation is the perfect fit. There's no design, just ever-chaning framework magic. I'd do the same "Google and debug" cycle, but 10 times slower.

It's kind of funny to see you saying "whatever solution you find are 5 years out of date", while at the same time saying that the tool that was taught using those same 5 years out of date solutions as a part of its training data is actually good.

Terrible idea if you ask me. I'd suggest checking the official docs next time around, or at the very least copying them into the context window.

First, good agents do that themselves. Second, specifying an exact and current version also works. Third, I'm mostly concerned about having a working example. I'm talking about breaking changes and APIs not existing in newer framework version. As long as it compiles, it's clear the approach still works.

Well then your experience is not really relevant in this thread when the prompt is specifically asking for professional coding work now, is it?

You're not an LLM (at least I don't think you are), you're not obliged to respond with an answer even when that answer is only tangentially related to the prompt.

I'm a full-time software engineer and develop those prototypes as part of my work. Also, I won't respond any further to your comments unless they improve. If you don't like what I'm doing – fine. Just shut up. You don't have to respond to every one of my comments.