nilsb

That's not the point of why those environments are sandboxed. They're restricted so that during training the LLM doesn't "cheat", for example by looking up answers on Google (simplified example). You wouldn't want an LLM that responds to "fix this bug in my repo" with replacing your code with something it found on the web.

Who needs testing when apologizing to your customers is cheaper?

I'd hope that it's based on calendar months. Wouldn't invoices that span multiple years make things a lot more complicated in terms of accounting in general?

As far as UI/UX is concerned their tvOS app is in a really rough shape. The web UI has a fairly reasonable program guide (edit: for Live TV), while the app doesn't. Also navigation in the app doesn't feel polished at all unfortunately.

Doesn’t the ESP32 have internal pull-up resistors, thereby making that extra resistor unnecessary?

Maybe that‘s been their strategy all along? I.e. demand the impossible and then back down to more „reasonable“ prices to appease the masses.

A request for a list of personal data they’re processing would be interesting. How would they even comply with such a request?

You won't need people for that once AI systems start improving themselves.

I've been looking at their website for the past 10 minutes or so and I still don't know what _exactly_ this does.

It seems like this is a local firewall (remember ZoneAlarm of the '00s with all the alerts about connections it has blocked?) paired with a VPN.

Didder is fantastic. I'm using it to convert images to grayscale for use with Waveshare e-ink displays. I've found no other tool that both lets you specify a custom palette (without which the resulting image looks pretty bad) and is reasonably quick.

For future reference, this works beautifully with their 9.7" display: didder --palette '000000 111111 222222 333333 444444 555555 666666 777777 888888 999999 aaaaaa bbbbbb cccccc dddddd eeeeee ffffff' -g -i - -o - edm --serpentine FloydSteinberg

Yes, it does. The delay of the first start of an app is quite noticeable. But the transpiled binary is apparently cached somewhere.

While technically true it would be more accurate to say that no features are _planned_ in advance and it’s entirely up to individual contributors to decide what they work on.

Is it just me or is understanding those examples virtually impossible if you're not intimately familiar with that particular framework? Then again... I guess that's the author's point.

Also, there's "different" vs. "diferent" in the examples:

source:2:2: error: _Static_assert failed "types are diferent"

2 |static_assert(typeid(a) == typeid(double [10]), "types are diferent"); |^