search

ltguillaume / thorium-winupdater

A fork of https://github.com/ltguillaume/librewolf-winupdater to make (automatic) updating of https://thorium.rocks for Windows much easier. Can be used for installed and portable instances.
https://codeberg.org/ltguillaume/thorium-winupdater
GNU General Public License v3.0
38 stars 3 forks source link

Add Mercury Support #5

Open jet082 opened 9 months ago

jet082 commented 9 months ago

It'd be nice to have support for the full suite of browsers, Mercury included. I personally have both installed since Mercury tends to do better with pages with a ton of text.

https://github.com/Alex313031/Mercury

ltguillaume commented 8 months ago

Sorry, but this won't happen unless someone forks WinUpdater.

I'm not using Mercury and recently I've become more and more convinced that I can't endorse using Mercury or Thorium at all.

While I fully appreciate the fact that compiling and testing a browser takes a lot of time and effort, especially when you're using your own hardware for it, the amount of releases are simply too few and too irregular. These days, your browser is beyond a doubt your first line of protection against attacks and considering the vast amount of CVEs for any of the two major browsers (often describing actively abused leaks), I think it's simply irresponsible to use a browser that is months behind on those very fixes.

For Windows 7 users, I could see the appeal to using a (somewhat more privacy-friendly) browser that still receives updates beyond v109. But since a lack of steady releases is Thorium's major flaw as well, it still feels like a half-assed solution.

For those on more modern OSs, I would say the theoretical performance gains and the few handpicked tweaks don't justify the lack of updates at all.

I'm not sure about Thorium and compiling via an online CI (ideally to create reproducible builds), but it should be possible for sure with Mercury (LibreWolf currently uses GitLab for this and is looking into using Codeberg's CI).

Personally, I think using a CI to allow for more frequent releases would be the only way forward for any of such browser forks, because the development and the game of cat and mouse between browser developers and malicious actors is a very real one.

When looking at Chromium-based (desktop) browsers that do have significant benefits in terms of privacy, the only ones to consider are Brave (looking only at the end product, ignoring all the drama around the people and the product's ways to find funding) and, maybe to a lesser extent (but it's what I use as secondary browser after LibreWolf), Ungoogled Chromium.