Some Linux desktop environments were developed over decades, others in just a few years. Starling, which made the rounds online this week, claims to have been built by just one person in six months using AI coding tools.
What’s more, a preview build is available as a .deb to install on Ubuntu 26.04 LTS. Yup, this isn’t a ChatGPT mockup but a real working demo.
According to the project’s GitHub and website1, a developer used AI2 to rewrite Flutter’s framework layer from Dart to Swift, and left its C++ rendering engine unchanged. A traditional desktop uses a windowing system but here the engine drives the display via DRM/KMS.
A Wayland compositor and X server, built for this project, take each app’s window and hand it to the framework directly off the GPU, with no copying involved. The framework then treats that window the same way it treats its own dock or buttons or window bars.
Now, zero-copy GPU buffer sharing is a Wayland protocol, and mixing native and X11 clients in what Xwayland does. What’s actually new here is narrower: treating that composited ‘buffer’ as a widget inside the framework’s own layout system.
Novel? As a feat, certainly. It’s a working desktop made in less than six months with technical choices and an adapted codebase. But is that – done by AI – enough to make it compelling for end-users?
Starling desktop was ‘written by AI’
I’ll be honest: I’ve been writing and rewriting this post for several days because it is hard to get a handle on what Starling is aiming to be and who it’s for. Any human impetus that existed has been buried beneath a mountain of ‘profundity slop’ on its website and GitHub.
The homepage gives more time to justifying its claim as “the first desktop written by AI” than the “why” of what was written by AI matters. A comparison box sees it ranks itself again mockups and screenshots. The rest of the text is breathlessly hollow narrative framing – it’s headache inducing.
Take this:
A curious person with a bold idea is not new. Being able to act on it alone is.
Starling.build website
It’s size 8 “polyester prose”; words an LLM arranges to look flashy, but whose logic is frayed (and yes, it leaves me feeling itchy). I wonder if the prompt that led to such glib garbage was: “write a manifesto in the style of a tai-chi-walking scam ad, but with less sincerity”.
To wit, all of the ‘this was never before possible’ claims fall apart.
Enlightenment desktop exists. Its founder built the first version alone in 1997, decades before AI tools existed. And it grew into a full desktop running on its own EFL toolkit.
Starling’s own “why” page briefly mentions it, but rather than use it as an anchor to explain how it differs, it just slips back to the the hard sell slop that pretends this effort, made by one person with AI, was never-before possible.
Parodic TedTalk-isms are out in such force I did wonder if this was a satire project. Many of the one page statements skewer or undermine the entire premise of why Starling purports to be worthy of gushing praise.
LLMs gonna LLM, and I don’t feel confident quoting rationale that isn’t, frankly, rational!
But I digress…
Starling desktop’s visual style looks familiar…
Screenshots of Starling on the website showcase a modern, if traditional, desktop layout with a UI style I’d charitably describe as “Temu macOS Mojave in dark mode”.
I’d charitably describe Starling desktop’s UI as “Temu macOS Mojave”
Dock, panel, clock, window switcher. There’s little user-facing that can’t be recreated on any Linux desktop install using a few apps and a stale macOS GTK theme.
Once you’re past the vibe-coded engineering newspeg, you’re left with a patchwork of pastiche atop a ported codebase, in effect.
What benefit is on offer comes from the fact tiling, spaces and the dock glass-effect work on apps like Google Chrome without the need to explicitly write code to support them.
To my mind, more time spent on detailing these practical differentiations, and much less on the heavy self-apotheosising – telling me it’s important without showing me it is – would do wonders. Nr’y a mention of the first-party apps available, including a file manager and calculator.
You can’t prompt commitment®
Starling looks passable at first glance. It’s doing something novel on the backend, sure. My cynicism – provenance of and ethics of AI generated code notwithstanding – is more on the fleeting nature of vibe-coded ware in general.
This could buck the trend and become a properly maintained, resourced Linux desktop that attracts users and contributors en masse. It could mature in new and interesting ways. Or, like a lot of “AI slop”, it may just drip down the walls of existence before evaporating.
AI coding tools bypass constraints, yet constraints matter. Some software is good and some software is not because someone, somewhere had to make a choice because of constraints (self-imposed or otherwise) and then carry the ‘cost’ of that decision forward.
‘What costs nothing to make, costs nothing to abandon’ is an aphorism I’m willing to mangle to make a point because it rings true for vibe-coded software. You can’t prompt worth into existence, no matter how flashy the ‘polyester prose’ you dress it in.
How to install Starling desktop on Ubuntu 26.04
If you want to try Starling yourself, you can. Download the .deb from the project GitHub and install it on Ubuntu 26.04 (not a container; Starling needs a real AMD GPU to work). Then, pick the Starling session at GDM login screen and log in as normal.
If you’re a DE-hopper, look at the list of known issues on the GitHub as it lacks support for Intel and NVIDIA GPUs, third-party apps, working audio, a screen lock.
I will add a note of caution.
Installing random deb packages with root permissions on a stable system originating from developers you don’t know or trust is a risk. When software is a double-whammy of pre-stable state and vibe-coded… One person who tried it filed a bug to say it corrupted their install.
Does this effort give you good vibes?