> A desktop that loves you back has to know you. Personal AI has crossed the threshold where the desktop itself can be compiled per user, per device, per moment, from a small portable model of the user – what we are sketching as Kadai: an encrypted, vendor-agnostic personal kernel from which Plasma assembles each Activity.

Sounds like a gimmick that will be annoying after a week. Users who aren’t KDE developers don’t want a desktop to love them back, in fact they don’t care or want to think about their desktop beyond launching their apps and window management.

Although Xfce 1.0 appeared slightly earlier, its initial releases were proprietary, making what was originally known as the Kool Desktop Environment one of the earliest entirely FOSS desktop environments for Linux. Version 1.0 was released in July 1998.

What a weird opening.

Also, both projects were started late 1996. A lot of people were already running KDE alphas/betas before the 1.0 release, because as a desktop it was vastly better than any of the window managers at the time.

I remember installing beta 1 in 1997 was a big relief for me. It felt like the last barrier was broken for Linux taking over the desktop, now that it finally had a desktop that could compete with Windows 95. Of course, I was terribly naive, but I blame it on only being 15yo.

I don't recall Xfce being used much at all. It become more popular until 5-10 years later because people were complaining that KDE/GNOME became too heavy and they were looking for more lightweight alternatives.

A desktop is a tool, and the last thing I want from my tools is intelligence. Tools need to be predictable. My desktop should respond to my commands the exact same way every time so I can use it without conscious thought.

If you want to introduce AI to a desktop, the best way is to add an interface for automating that desktop and add a separate AI agent that can use that interface. This also benefits accessibility tools that can use the same automation interface.

So a few months ago for various reasons I started to use ubuntu/gnome again (after 15 years on MacBooks / OS X / MacOS) on my home PC. I have to say, in my case, command line llm made the desktop so much more enjoyable.

Sadly still a lot of things did not work correctly out of the box, and everything was kinda fixable, but would have needed some bigger time investment from my side.

Instead just telling llm cli that something does not work correctly, and waiting a few minutes until it fixes it is just so much better for me.

Do I want LLMs all over the place in my desktop? Probably not. But having an LLM fixing all kind of broken or miss-configured software is just so convenient.

The best way that my desktop can show me love is by being a glorified application launcher and nothing more. Also, the less meta/data that it sends back to its vendor the more love I feel that I am receiving.

It’s going to happen in windows and most likely Mac too. In both cases, all your data will be shipped off to the model providers and then out into the panopticon.

The open source desktop environments don’t have the profit motivator that drives the commercial alternatives. They’re driven to deliver their interpretation of great UI. We can each debate whether we agree with the vision or execution but that’s the motive.

If they can employ AI to help remove desktop friction - and it’s local and not shipping my data off to the surveillance machine - I’d be open to it.

In fact, if it could sort out KDE wallet’s amnesia for my wifi password, I’d give it a roll now…

> A desktop that loves you back has to know you.

Is this another downstream consequence of the COVID lockdowns? Do people need a hug?

___

Hmm. So from the actual talk description, these seem to be UX people that might have interesting ideas to share but also have completely lost the plot when it comes to the overarching narrative.

Which is unsurprising, given that their summary is clearly written by AI (albeit with human editing, but barely)

https://conf.kde.org/event/11/contributions/314/

I love making kustom KDE widgets with LLMs[0][1]. If you like customizing your desktop, I highly recommend you take advantage of this new era.

[0] https://github.com/dginovker/KDE-Mouse-Click-Counter-Widget [1] https://github.com/dginovker/KDE-AI-Usage-Tracking-Widget

Nowhere is safe, is it? All that is holy is profaned.

News piece with unnecessary annoying editorializing.

I think the idea of plasma widgets and rails fully being customized per user giving them material control on ones desktop is a good idea.

We need good ideas to counter and take on the omarchy nonsense.

Guess they never used QtQuick with LLM. LLMs are really bad at QML/QtQuick, they often use older APIs/syntaxes and generate total garbage.

Pretty sure KDE said no AI period. This won’t stand.

>In 2003, they conducted a usability study of the then newly released KDE 3.1.

>The talk has three parts: what KDE learned – or failed to learn – from that study;

This caught my eye, seems in many forums people still say 3.1 was the best KDE.

But I remember trying out KDE 3.1 on Slackware at the time. All it did was give me extremely severe eye strain after about 1 hour of use. GNOME, XFCE and various Window managers did not give me eye strain of any type.

New releases of KDE still caused that issue with me, but I found a color theme that I can use with KDE 5 that holds off eye strain after about 4 or 5 hours. Again all other DE and WMs, no issues. FWIW, I really like KDE 5, but it can still hurt my eyes :(

I have not tried 6.x and will not see it until Slackware 16 is released. I will give it a shot but with my history, I am sceptical it will be fine for me.

Sorry, but a string of ones and zeroes will never "love you back".

[dead]

I'm good, thanks. AI bros can all gather at Omarchy if they want slopOS experience.

Is it going to just be an Omarchy + KDE iteration? The article was very light on content but DHH is definitely driving more results and new contributions which is great for free software.

This article has no content whatsoever... Just sending the links of relevant materials to AI would produce a better-organized result than this.

The desktop is a solved problem. Now we are moving into the era of agents using the mouse and keyboard and interacting with the windows manager.

When I see stuff like this, I think it's clear that Linux is not going to be part of that era outside of the CLI. The Linux desktop is never going to recover from this.

Gnome and KDE just keep messing. That's why I stuck to openbox/labwc.

Because, these are not solving problems they are inventing them. Everything that makes Linux great, simply hasn't translated to the desktop GUI. And I can see that will not change. Sure there are things like Niri and thats cool, but for agent use. It's redundant.

We are done with desktops. They will be handed off to ai. Humans will use them less and less. Because they will have better things to do. And likely better interactions. After all a mouse and keyboard while I still prefer it over voice, is still from an era where that was the best we had.

Now, I actually think I prefer tmux in a terminal on my phone on a software keyboard. Over a desktop with mouse and keyboard. As clunky as it is. Because I can use that anywhere, anytime to talk to agents.