think1984

joined 1 year ago
[–] [email protected] 1 points 8 months ago (1 children)

You can force Firefox to display dark mode in web content (even with privacy tweaks enabled to resist fingerprinting or tracking), by setting the two following hidden prefs in your user.js:

// PREF: enable a Dark theme for browser and webpage content
// [TEST] https://9to5mac.com/
user_pref("ui.systemUsesDarkTheme", 1); // HIDDEN
user_pref("browser.in-content.dark-mode", true); // HIDDEN
[–] [email protected] 1 points 9 months ago (3 children)

Yes of course. I hadn't slept when I replied, how embarrassing to miss that. You can enable CNAME uncloaking in Brave, which I suspect draws them to a parallel. It would be interesting to see the test repeated with the setting enabled. Since one has to (or had to) enable it in uBO also, it would only be fair to compare apples to apples. As I said, the blocker in Brave is based on uBO anyway. To be clear, and as I've said before, I've daily driven Firefox since the beginning and run uBO in medium mode. I'm not shilling for Brave here, simply pointing out that the differences are small (much of the code is shared with uBO) and it does certainly render faster.

[–] [email protected] 1 points 9 months ago (5 children)

Brave isn't represented anywhere on the graph? Unless I've misunderstood you. That's a comparison of Firefox with various ad blockers, and uBO with and without CNAME unclocking enabled. Brave also uncloaks CNAMEs, so that's one place they are equal. Chromium based browsers do lack some abilities compared to Firefox, however. I have daily driven Firefox since the first day, but Brave and Blink/Chromium based browsers are undeniably faster at rendering (unfortunately).

[–] [email protected] 1 points 9 months ago (7 children)

And thirdly, I have seen no mention of anything like ublock’s blocking modes (block 3rd party scripts/frames). Can you quickly select an element to block in brave?

You can enter as many custom filter rules as you like, with adblock syntax support. You can select an element to block, yes.

[–] [email protected] 1 points 9 months ago (9 children)

A baked-in content blocker like that of braves loses because it isn’t extensible.

In what way? I use(d) Firefox since the very first Firebird days, and Netscape Navigator before it, and I'm practically married to uBO (don't tell my wife!). That said, Brave's 'shields' blocker is just skinned uBO with some tweaks. It can add custom cosmetic filtering rules, additional adblock format filter lists, disable or enable JS (globally or per-site) and has built in fingerprint resistance. Aside from the differing UI, I genuinely can't think of anything overtly missing as such.

[–] [email protected] 17 points 9 months ago

To speed up DNF:

/etc/dnf/dnf.conf  
...
max_parallel_downloads=5
max_parallel_connections=20

I have found that RHEL/Rocky/Alma are very much faster at updating using DNF than Fedora, though I don't know why.

[–] [email protected] 8 points 1 year ago* (last edited 1 year ago)

I'm older than some in here, I think (judging by the answers). For me, a couple of the things on steakandcheese.com and rotten.com (back in its early days, not the later milder days), cartel videos, beheadings, hangings etc. The standout though was a video called 'Terrorists, Killers & Middle East Wackos'... Grim stuff. From Wikipedia:

Terrorists, Killers and Middle-East Wackos (also Terrorists, Killers and Other Wackos in the UK) is a shockumentary video from the makers of Bumfights. It includes footage of riots, suicides, executions, and the televised suicide of R. Budd Dwyer. All the scenes included are real scenes of death and suffering. The Bumfights website store touts the video as "One hour of the sickest images ever put to film."

I was young and dumb once, but I actively avoid things like that nowadays. There's knowing how things happen in the real world, and there's traumatising and desensitising yourself to the worst depravities possible for... reasons? Just no. You can't unsee some things, and some things shouldn't ever enter your consciousness at all (unless you're really unlucky IRL...).

[–] [email protected] 1 points 1 year ago
[–] [email protected] 2 points 1 year ago* (last edited 1 year ago) (1 children)

Even Hagezi's most basic list blocks a lot more than OISD, and still no false positives. See a comparison (run over the top 10,000 websites) here.

[–] [email protected] 3 points 1 year ago (5 children)

DNS blocking is heavily dependent on the blocklist(s) you use. It's entirely possible to block >95% of crapware, and break companies' ability to track you without compromising usability.

Having used both for a lot of years, I'd say look instead at AdGuard Home. It is also FOSS but supports more out of the box; including certificate management, the ability to use encrypted DNS both upstream and downstream without need for third party software (cloudflared), the ability to use adblock filter syntax (lists are 200k lines instead of 2 million lines, but actually block more), and so on. PiHole has some improvements pending in the next version, but it's not there yet in comparison, imho.

I'd also strongly suggest you check out Hagezi's DNS blocklists, as they're pretty much set and forget. They're intended to be used as your only block list, and do an excellent job (see testing in the Discussions on their GitHub). Use the Normal list if you don't want to deal with false positives occasionally, and the Pro++ list if you don't mind getting your hands dirty (whitelisting occasionally) and want to block every last scrap of annoyance and anti-privacy crapware on the web. Both will significantly improve your online experience.

[–] [email protected] 3 points 1 year ago

Not according to PrivacyTests.org. It lacks a lot of state partitioning, amongst other things. Just FYI, the front end UI is closed source, but the backend/engine is open source, because they're just another chromium spin off.

view more: next ›