sharken

Interesting, might want to try a different model the next time.

Actually the opposite, Opus 5.5 medium did the job.

Anecdotally, I have had Opus 5.5 decline a reverse engineering effort into a Zigbee gateway that I own.

DeepSeek did not have any issues with this using Claude Code as harness.

I think it's nearly impossible to explain the benefit of not seeing such a popup to most people.

It comes down to the joy of doing things, and if the joy of using said monitor depends on a popup not being shown, then so be it.

I hope that the popup can be removed :)

More often than not, what matters is transforming a concept into working code that is good enough.

The easier part is maintaining and tweaking that code, but do not underestimate the value of the former.

When using the browser, what Anthropic calls Claude.ai, the memory is stored on your account on their servers.

Claude code stores memory locally on the device, similar to how a developer stores notes.

Data retention is about storing your raw conversation data.

Capabilities and Privacy settings are used to manage memory and data retention.

In general not really, but I do understand why it's a bit underwhelming to buy the clock upfront.

With that said, the project is really cool and showcases the different watch faces well. For me that's the most interesting part, not how I construct a round LCD screen.

Well, of course politicians will try and hide what the law is about, so I find it necessary to clearly communicate what this legislation is about.

CSAM hides behind an acronym and seems benign at first, but taking away the citizens right to private chats is anything but that.

Indeed, reusing "Chat Control" and giving it a new meaning is a surefire way to cause confusion.

And the underhanded way of trying to get something as controversial as this to pass, says a lot about the politicians pushing this agenda.

Chat Control is about allowing mass surveillance and removing privacy, no should be the obvious answer here.

Claude Opus 4.8 suggests "ReGRAM", which is less bad than GRAM.

In that sense AI is a tool much like a dictionary, it enhances and I'd say improve the end result.

Given the current state of the Chrome family of browsers and the anti adblocker stance from Google, i'd think that alone would guarantee Firefox a steady user base.

Not sure how users cope with Chrome-based browsers and intrusive ads.

Not to mention a Facebook simulator :)

Also missing (I think) are anecdotes about previous experiences involving famous people or just memorable moments in history.

Also comments where the poster shares details from their own life instead of just commenting on the topic.

But I can't really describe this "human Factor" any better than through examples.

Really spooky, all comments are LLM based and something certainly feels off, but can't quite put my finger on what it is.

I think it has to do with comments that doesnt really comment on the previous comment.

Certainly one of the more interesting uses of LLMs in a while.

Well said, ideally action comes first and then these actions can be communicated.

But in the real world, you have words ie. commitment before actions and a conclusion.

Best of luck to them.