flossly

It's more like the Amish not using certain technology we all take for granted.

I see this much broader: opting out of certain technologies.

As a vegan I opt out of all animal domestication/husbandry and wild catch/hunt.

And I try to stay away from refined sugar and refined starches.

Abortion is also a technology people feel strongly about.

And now we have the use of AI.

Did we also have people that did not want to use an electronic calculator when it just came out? Probably.

Linux (Linus T himself iirc) is very GPLv2.

They are not onboard with the innovation that was GPLv3. That innovation was mostly about software patents iirc. But the result was also that big corps liked GPLv3 even less than GPLv2.

On top of that it was going to be very hard (near impossible) to get all devs to agree to a v2->v3 license bump.

Are people allowed to not want islam big in their home town?

Why can every country be "proud of their culture" and "want to conserve their culture" except white/christian countries?

Why is black-pride, or gay-pride, or african-pride, or XYZ-pride all very acceptable, except white-pride and christian-pride? The pushed narrative is obvious to me.

So how is it "extremist, anti-democratic organisation"? you bring no proof. I have to believe you on it?

If you are done with mass-immigration in Germany, you get no sizable party to vote for because they all have been demonstrably in favour of the immigrant flood. Except AfD. Why no other party caters to the done-with-immigration demographic? If you deny your constituency for too long you are bound to create a new powerful player.

Now please show you proof...

I think you should be able to voice any policital opinion. But following the MSM narrative of calling every party that wants to stop mass-immigration as "far right" or "extreme right" has no meaning to me anymore.

These emails may be real. But that's yet to be confirmed.

Yesterday The Economist interviewed AfD's leader, in proper English:

https://www.youtube.com/watch?v=_cGGDu8sgXY

The BitDefender article immediately goes of with labeling AfD as far-right. This is going on all over Europe when a party is against mass-immigration and "sponsoring Ukraine till the last Ukrainian". And at the same time the streets in EU's big cities look worse than ever: unsafer, uncleaner and more un-European than ever before. Immigration is not good for us, and receiving a label "extreme rightwing" (and Fascist, and Nazi) simply because you want it to stop is loosing it's power... I dont mind to be called as such anymore (and I used to even call others this label).

In European vocabulary these "anti immigration of economically not-useful people" parties are not very rightwing: they are in favor of keeping the social benefit programs alive (that's why they want immigration of such people to stop).

We have "utilities on the market" now. So we get to choose between the color of logo on the invoice. Every year we can switch around new year. Then lots of marketing is going on: people going door to door as well...

For sure I now pay for all these marketing campaigns: they all offer a cheap first year. So I have to switch yearly to get a good "welcome present" (that the non-switchers are paying for me).

Privatizing a commodity-service, that utilities are (electrons/water/gas -- it's all very standardized) is the most ridiculous to me. The innovation a supplier can bring you is mostly it's marketing campaign.

You said:

> Rust’s type system and borrow checker are not significant reasons for its compile times

But isn't "hard" monomorphization a property of more complex type systems (generics, type variables, etc.)?

You can have naked (top-level) functions in Python right? Are they also in instances of classes?

Kotlin has optional OOP on the JVM. I still use it, but sparingly. Barely ever classes inherit from other classes in my code; I use interfaces, mostly to be able to fake services for testing.

Most of what I write are functions and data objects (records, basically).

One pattern I use is classes with one function, where the constructor takes the functions dependencies, and the function only takes non-dependency parameters.

Not sure what you mean by modern C++ code.

But Rust's type system an borrow checker are not something you want to go without in Rust, while in C++ many codebases prevent the use of some features to reduce compile time right?

Predominantly trained on atheist text maybe?

Also, would you want to know an LLM's religious/spiritual orientation? I rather have it help me code.

No, "CloudFlare going all in on Rust" means that the code they write and maintain is rewritten.

> P.S. I've tried Zig [...]

Why not Zig then, and C instead?

As a C/Zig outsider, I'd much rather read Zig.

Yes, at least Rust and Zig have answers for the remaining niches.

Rust's type system is making it's compile times bigger than C, C++ and Zig. This does keep it away from some code that's currently C or C++.

Cannot really compare C# to Rust.

C# to me compares to Java and Kotlin (they have a large runtime or VM and implement GC).

Rust compares to C, C++, Zig or Odin (compile to native, memory is more self-managed).

Best part of Golang ihmo; first time I saw a language mandate it's formatting. Fast compiles are also nice, but not exclusive to Go (C, C++, OCaml, also compile fast).

I find it interesting that Google uses Rust. I know they work on Rust-based kernel code for Android. I'm curious to know if they also use it for networked application code, and why they picked it over Go.

Personally I prefer Rust over Go for it's types, and I never use it in big projects, so compile time does not bother me. But Google basically created Go for it's networked application code.

...and a community the size of Elm.

> I want regular OOP

Why would you want that? Just to continue the habit? I'd say regular OOP is, after implicit `null`, the biggest mistake in software.

Crystal is nice, but for network programming, Go simply cannot be beaten by a language that has a equally limited type system.

just bolt your SSG onto the many CMS backends that are available. like Ghost, WriteFreely, Strapi or Payload.

We can do better than PHP, EAV-style db schemas, year 2000ish CMSes and, thus, Drupal.

It' reputation is earned.

A fresh web stack with a traditional (non-EAV) db schema and proper programming layers is a much better DX, I'd say. All very doable with LLM-assisted coding.

I'm more insterested in the new CSS scoping feature that still not supported. HAving CSS only work on a scope of the DOM would make dealing with CSS a lot easier.

While waiting for that I decided to use Tailwind so I can basically "not do CSS", but have it locally defined in my HTML. It buys me a similar freedom to the CSS scoping feature, as the cost of having to run a bunch of tools to get my final CSS asset file for deployment.

> bad taste

They are only as good as their training data.

I expect them to have badder taste for PHP or JS, than, say, Kotlin or Ruby.

That's be because I prefer the code the latter communities have produced over the former.

Ruby is token efficient because it is terse, and has little special-characters. That why it sometimes "reads like english".

But also it has a bolted on type system.

I dont mind burning tokens on code that is well typed.

Just like I dont mind burning tokens on TDD.

I think maintainers here listened to their gut feeling and trusted their experience.

On top of that I think this is actually great FLOSS news, knowing how many commercial products have sacrificed proper professional programming on the altar of shareholder expectations.

> We might need different kind of rules/constraints because at the end we rely on LLMs to respect or use those rules

I want everyone to respect those rules. LLM and human alike.

Hence, I like architecture tests. Most of my arch-tests double as a point in a CLAUDE.md (or sub) file.

Architecture tests are understood both by LLM and human, and test everyone's code. How's this not what you need?

Saying we need a different language, to me, is saying you do not understand LLMs well: those algorithms basically can do what they do due to prior art. A completely new language is not what I'd expect an LLM to be great at.

Look into architecture tests. By leveraging those you can get any project to have those benefits.

Is it?

When you have to understand something yourself, do you rather read C, JS or C#?

In the end of the day I have to be able to read it, in anger, when things go south...

Some libs have amazing APIs (many in C# and Kotlin are well done); some libs have horrible APIs (many libs in PHP and JS look horrible to me).

In the end of the day I have to be able to read it, in anger, when things go south...

I found that what me/ my project/ my LLM needed to arrange this was very elaborate architecture tests: tests that check if some architecture rules are still upheld on the main (and test!) codebases.

If I find a violation of a arch rule (e.g.: a db query defined in HTTP-handler code), I codify that arch rule in the architecture tests.

Libraries exist to help writing arch tests, but you can also let your LLM come up with something custom.

Also for CSS? I really like how Tailwind puts the CSS in the same place as my HTML, and also I like how well LLMs write it.

Also my JS is very "contained" as I write SSR web applications. I also really like how Alpine.js makes my use of JS very "close to the HTML". I'd say using Alpine helps my LLM to deliver more maintainable JS code (in case of SSR web apps).