Whoa. An update to the HTML Hell Page!

The easiest and possibly most secure server I know of is agate, written in Rust. But there are many implementations to start from…

Images and documents with any MIME type are part of the protocol. Clients decide how they do display that.

That said it is a medium primarily geared at displaying text. Not animated gifs with ads.

It can transmit any media or mime type. It just does this one document at a time.

Clients can display linked images inline of course.

The purpose of this and other limitations is to make it useless for advertising and tracking.

Trank you, I fat-fingered this without noting.

And yeah, that would be a creative yet fitting use for the “Not Safe For Work” tag 😃

Literally burning the planet with power demand from data centers but not even knowing what it could possibly be good for?

That’s eco-terrorism for lack of a better word.

Fuck you.

This is about automated coding agents.

Hard to tell whether this is a work of fiction or actual collective insanity.

Yesterday, I was at $WORK, which is producing some specialized industrial PCs. In the developer stand-up, somebody told us that it is not only hard to get RAM because prices have risen eight-fold or so, but that our supplier has cancelled the supply contract. To me, that sounds more insane than the Dutch tulip mania of 1637.

I would agree with that.

Especially, “being 70%” finished does not mean you will get a working product at all. If the fundamentale understanding is not there, you will not getting a working product without fundamental rewrites.

I have seen code from such bullshit developers myself. Vibe-coded device drivers where people do not understand the fundamentals of multi-threading. Why and when you need locks in C++. No clear API descriptions. Messaging architectures that look like a rats nest. Wild mix of synchronous and async code. Insistence that their code is self-documenting and needs neither comments nor doc. And: Agressivity when confronted with all that. Because the bullshit taints any working relationship.

Sounds like hell is freezing over…

But you might have noted that weather patterns are becoming a bit strange recently.

So, they are making their own code open source? Did that happen since this announcement in August 2021?

Or are they stating that they want to use publicly available open source code without returning much?

Guix vs Nix will be an interesting example. Nix has a way bigger user base right now but it has the whole Anduril & governance issue.

Guix has a way better configuration language and one can learn in an afternoon enough to use it productively.

What is your experience with guix like?

I am mainly using Guix as a package manager on top of Debian stable (and on top of my Arch install running in a vm). I use it mostly to have a reproducible development environment for my free time projects (which use Rust and Guile), and it works very nicely to that. It is also certainly a nice way to distribute software as source, with very little effort (just putting the own package definutions into a channel repo).

Does getting away from systemd affect things?

I have also started to run it directly on my PC as a base system. After replacing the NVidia GeForce card with an AMD Radeon one, I had no issues.

The configuration and init system work well - the only thing I would have to do is to write my own stumpwm(*) init script, which I didn’t have time for, so I use, as a fallback, i3wm and Gnome or XFce2, what I use at work, too.

(*) Stumpwm is a highly configurable tiling window manager written in Common Lisp. Similar to i3, but using key chords, and window manager actions are just lisp functions one can program and extend - they are called via key chords like Emacs commands.

In respect to the init systems, I have to confess that I am mostly agnostic. As long as it works, I am fine. I think Guix is the more modern and better approach.

Here a summary on Guix I posted a while ago:

https://feddit.org/post/23120439

But also one can’t rewrite many many complex apps by themselves.

Well, ressources are limited. Especially the amount of stuff other people will do for free in their free time.

My point is it wouldn’t be an issue as much if there wasn’t as much fragmentation. If it was easy to write for both qt and gtk at once then people wouldn’t have to complain about one or the other all the time.

Specifically with this, I don’t see the issue. Qt apps run fine under GNOME and vice versa.

But sometimes unification across the desktop is something some people want.

Yeah but these same people are not going to do anything about it.

In theory, it would be nice to get some solid public funding for making desktop apps more accessible. With our rapidly aging population in Japan or large parts of Europe, that would absolutely make sense. But I don’t see the job offers for SW engineers to do that.

I don’t think you can just invalidate what I’m saying with the just go write it then.

Well, I get that you want that.

But who should do that? On whoms time? With which money? Or for free?

Even things like the real time Linux project, which is extremely relevant for industry (including defense) is not funded in any sensible way.

Myself, I am an expert in signal processing and renewable energy topics. It is extremely relevant for energetic independence of Europe, and climate protection. That’s not funded either. What is funded instead are “audio sound design” for combustion engine cars (that is, artificial simulation of engine noise). And this is bad politics - not something FOSS developers can solve by putting in more work for free.

Now there are wishes that “open source developers” put more (free) work into software security. Who exactly should do that?

I think most desktop stuff like KDE is done by people in their free time. They already do great work, including in the domain of UI. The negativity you transpire is unwarranted. These people do A LOT.

These people have a life outside programning, other responsibilities, and other things to do. Complaining about them not doing even more work will not motivate them.

Lots of foss projects that look ugly and are not designed with ui in mind because it’s made by devs who are used to doing lower level stuff.

Maybe I am one of these developers because my domain is signal processing.

I don’t care. I write mostly CLI apps.

Plus the fragmentation of qt and gtk.

I don’t care whether something is qt or gtk. I use what fits me. I might write a GUI in Swing or,JavaFX because it fits nicely with Clojure or Common Lisp on the JVM, or I might write a Rust app with a Racket GUI because it is actually native and cross-platform. It is so liberating not having to deal with corporate bullshit.

It is not that I am an enemy of aesthetics. Actually, I like to do art! But I do that in wood and metal and other materials - not on the computer.

Winning would be easily building applications that could be native across DEs and actually look nice.

And about winning and having apps that suit your taste: Go ahead and write them. Scratch your own itch - that’s how great FOSS software is created. But don’t expect from others to spend THEIR free time on things YOU want. In return, you can do anything you like.

Depends on the license.

And yes, you could say that the relationship between companies and open source contributors is a bit parasitic at times. But a part of them don’t care because for a start, they are not motivated by money to write open source software.

Nail on head! Exactly!

Here a summary I posted a while ago:

https://feddit.org/post/23120439

I think that one reason why the proportion of open source code grows is software quality:

Companies would love to own all their code. So, when they employ people who work on proprietary code, the amount of proprietary code should grow, shouldn’t it?

Except that companies have mostly very short-term goals. And this affects quality: A lot of proprietary code has quite shit quality and is not really maintainable. Which has the effect that either the project dies, or becomes very slow to develop further, because of tons of technical debt.

So, the company eventually will resort to rewrite that project. But that is like walking on a threadmill; it always takes a long time until a rewrite of an old project matches the predecessor projects features and stability. And the current GenAI craze will only make that threadmill rotate faster…

FOSS projects do not have this obsessive constraint on short-term returns, so they often have better quality. Which makes it more likely that these projects live and prosper a bit longer. The short-term difference might not be even large - but the process goes year for year, round for round, and it becomes an evolutionary advantage.

In the end, everyone uses that Finnish students former hobby kernel project, and nobody uses Windows 95 - or wants to use its shitty successors.

(And this is why I also think that Guix will win in the long term: The capability to re-produce all components of a program or system from freely available source is, in the long run, an overwhelming evolutionary advantage.)

I think that one reason why the proportion of open source code grows is software quality:

Companies would love to own all their code. So, when they employ people who work on proprietary code, the amount of proprietary code should grow, shouldn’t it?

Except that companies have mostly very short-term goals. And this affects quality: A lot of proprietary code has quite shit quality and is not really maintainable. Which has the effect that either the project dies, or becomes very slow to develop further, because of tons of technical debt.

FOSS projects do not have this constraint on short-term returns, so they often have better quality. Which makes it more likely that these projects live and prosper a bit longer. The short-term difference might not be even large - but the process goes year for year, round for round, and it becomes an evolutionary advantage.

In the end, everyone uses that Finnish students former hobby kernel project, and nobody uses Windows 95 - or wants to use its shitty successors.

(And this is why I also think that Guix will win in the long term: The capability to re-produce all components of a program or system from freely available source is, in the long run, an overwhelming evolutionary advantage.)

Copilot told him: “you are absolutely right!”