JaggerJo

Because a LLM architecture seems way too fuzzy and unpredictable for something that should be reproducible.

Yes, this is scary.

Same, tried Claud code over the past month. Let it do different tasks.

It’s useful as an built-in quick docs / search that can spit out small code fragments.

Every time I gave it more space results were disappointing.

REST almost never is worth it. It’s a nice idea, but in practice things often are more complicated.

API quality is often not relevant to the business after it passes the “mostly works” bar.

I’ll just use plain http or RPC when it’s not important and spend more time on things that make a difference.

I really like fly.io, but you’d need to provide more information about the kind of backend and the expected load. Also what does it currently cost?

I host a few sites on fly for pennies, but they don’t see a lot of traffic.

Hah, I'm not so sure about what happens during automatic notarisation.

Sideloading would at least allow JIT compilation.

The fee only applies to alternative app stores, not simple apps.

In Europe Apple was forced to allow side loading + alternative app stores.

https://support.apple.com/en-gb/117767

Might be possible to do more (like call private apis) when apps don't need to pass app review.

F# is a really good practical FP language.

- Great real IDE support (JetBrains Rider, Visual Studio)

- Large ecosystem of packages (F# is a .NET language, so you can just use all Nuget packages out there.)

- The language is stable and only gets small updates. There is no Scala 3 scenario on the horizon.

- It is fast, multithreaded (.NET TPL) and cross platform.

- Great for building cross platform UI’s with Avalonia.

https://github.com/fsprojects/Avalonia.FuncUI

Yeah, kind of agree.

I don’t think this thought would have crossed my mind a few years ago, but I think more European countries need Nukes. Sad, but seems like the US can’t be trusted anymore.

Crazy how you can ruin a nations reputation in weeks..

I wrote a post about building document based applications.

It does not specifically go into CAD/CAM stuff, but it still applies.

https://blog.voyonic-systems.de/leveraging-immutability-and-...

Current .NET is crazy fast.

https://benchmarksgame-team.pages.debian.net/benchmarksgame/...

This is one of the first holidays where I don't feel an urge to program. Feels strange. Guess I'm getting older?

Reading: The Snows of Kilimanjaro

Project: Logic Gate Editor + Simulator (but I can't really bring myself to work on it much)

Frohe Weihnachten!

Really nice design!

Would love to have one but.. can’t find a use-case for it

Hey HN,

we've built FieldKit mostly for our own project at the start. We think it might be useful for others building embedded software.

AMA about the tool

Congrats on launching!

What language / toolchain are you using?

I note that explains what happens with the location data would also be nice.

Like is my complete health uploaded to your server? (I assume not)

I’ve used this in production. It’s great.

Wanted to check it out and was surprised that it requires an account.

I use two Ultimate Hacking Keyboards.

V1 had some issues, V2 is better.

https://ultimatehackingkeyboard.com/

I’ve used Fable in production. It just works most of the time, but it’s not seamless.

The first two I checked are also missing.

We use AvaloniaUI in production for a handfull of apps.

Couldn’t be happier.

What app are you starting?

For me it feels instant. Like opening sublime text. Faster than opening VS Code..

Depends on what you mean by "all the features". In general Avalonia UI is a .NET library, so you can use it from any .NET language.

If you have questions about specific features I can give a more detailed answer.

https://github.com/fsprojects/Avalonia.FuncUI

We are using Avalonia for embedded devices.

70%? really?

I never even had that during the alpha days of Avalonia. Give it a shot again.

> Start-up times and memory consumption are objectively bad.

Hah, interesting. On desktop (Apple M2) it pretty much feels instant to me. When targeting embedded, sure there is a delay.

Regarding memory consumption, can't really tell. It never was at the point where we had to intervene AND we target embedded devices with some apps.

What's your benchmark?

We use it in production for a bunch of apps.

Has been great so far. The core team is approachable and really cares.

When did you try it?

We use Avalonia for our Product and honestly couldn’t be happier.