4.0.0 was the release of the Rust rewrite
4.0.0 was the release of the Rust rewrite


The support is bad for custom nodes and NPUs are fairly slow compared to GPUs (expect 5x to 10x longer generation time compared to 30xx+ GPUs in best case scenarios) NPUs are good at running small models efficiently, not large LLM / Image models.


Most website do not run on servo
Definitely looks worse, but I can see why they made the change, gtk is a pain to work with compared to qt


The official installation process won’t work, as Bazzite is an immutable distro.
You can install it with rpm-ostree. If you choose to do so, I invite you to look into what immutable distros are, and what problems they solve (and create).
I was curious and searched if it was more complicated than it looked for ProtonVPN. It doesn’t look complicated, but it is not very straightforward either.
You can find how to do it in the comment of this reddit post
SearXNG is really easy to setup, you should have a go at it.


https://anubis.techaro.lol/docs/user/known-broken-extensions
If you have JShelter installed, it breaks the proof of work from anubis
https://wiki.archlinux.org/title/Firefox#Hardware_video_acceleration
Firefox is a bit annoying on wayland, you often have to force enable hw accel. If you cant figure it out send me a DM, ill be glad to help you through signal/matrix/discord


xpinstall.signature.required was set back to true, seems like complaining works well


It was enabled due that zen was still a toy project and we needed people to easily open the debugger for easier bug fixing. This was due because zen was not in a daily drivable state and didn't gain any sort of popularity yet
As the dev says in the PR almost nobody was using the browser at that point. To be able to interact with the debugging server you would need to have a port open on your firewall and router. And you would need to manually start the dev server. The problem in the PR is it was not prompting the user when launching the debug server and user could turn on the debugger without touching about:config flags.
The second part is more questioning, though not exploitable without the user clicking 2 times on a security warnings. I just checked their github to see if there is an issue/pr on the subject and there is none. Might be worth making one.


Imo they are more privacy conscious than Firefox and most Chromium based browsers, and on par with Floorp/Waterfox with their provided defaults.
If someone wants a good looking browser with vertical tab, while not having to debug privacy settings breaking site or having to write custom css to have the UI they like. Zen is my recommendation.
The only telemetry they leave is the ones that provide features to their users. For example, they need to ping mozilla for addons update, firefox sync, update the tracker block list, …
Although I agree with you that the privacy part of Zen the most beautiful, productive, and privacy-focused browser out there is clickbaity.


It’s not a backdoor, it just enabled Firefox’s remote debugging tool by default, which is necessary if you want to modify the chrome of the browser on your own computer.
At the time it was in one of its first alpha, sure it was naive to ship a browser with it enabled because it was convenient for development, but it was fixed 1 week after the issue was raised, and has been for months.
They use the release candidate to test upcoming Firefox releases and see if it breaks anything, to be able to ship the update on the same day as FF (just like the majority of other forks do). None of the patches they make require extra telemetry except for their “mod” system. Most of the criticism Zen gets about “security” applies to every browser except librewolf and tor. Zen is as secure as firefox is.
All this is coming from someone who doesn’t use Zen, as my workflow is constantly broken by their UI changes and bugs (which is the main problem with the browser).


Mot really a “leak” for cursor, they are publicly available when you send they request. They just dont show it in the UI


It is open-weight, we dont have access to the training code nor the dataset.
That being said it should be safe for your computer to run Deepseeks models since the weight are .safetensors which should block any code execution from injected code in the models weight.


Ollama isn’t made by facebook, the llama models are. Ollama is juste a cli wrapper arround llama.cpp, both of which are FOSS projects.


The patch set is open source, so is the application that does the patches. I don’t feel like it’s worse than downloading most apps. Though you still rely on Google services.


https://github.com/revanced/revanced-manager for YouTube (and other App mods), you need to patch the app yourself. (If you are downloading a patched apk you are getting scammed. You have to patch it yourself) https://github.com/mihonapp/mihon for manga / manhwa I use internet archives for paywalls but their might be a better solution. F-droid has a a lot of neat open source app, I guess you know of it since you are picking graphene, but just in case. Firefox supports addons, it’s been a great browser (though its UI is a bit rough sometimes) and could be your solution for YouTube with uBlock and the background play addons. I personally don’t like graphene launcher, I personally use NovaLauncher (the paid version), their is quite a lot of FOSS ones too.
edit: sorry for the formatting, my phone f it up
It provides a lot of nice syntactic sugar that you would otherwise have to write a wrapper for on top of Fetch. Built in request interception, request transformation, (de)serialization, shared request config, timeout/retries management, …
Though this definitely comes with bloat and supply chain risks.