One potential workaround, if you’re okay with using extra software, is making a script with AutoHotKey to intercept Ctrl+Alt+X and fire a different key combo in its stead.

This doesn’t fix the issue with the Firefox but, you know, it’s something.

Firefox will upgrade page loads to HTTPS and gracefully fall back to HTTP if that does not succeed. This behavior is known as HTTPS-First.

Wasn’t this already the case? For a while?! I don’t know whether this means HTTPS sites will load more reliably, or HTTP sites will.

AI Chatbot access is now being gradually rolled out to all users.

Hooray

To use this optional feature, choose AI Chatbot from the sidebar

Oh it’s migrated to be even more prominent now

or from Firefox Labs

… Despite still being labeled experimental

Pro tip: Long-clicking the Back button gives you a mini history to choose from. It works on the desktop and on mobile Firefox too, in the overflow menu’s Back button.

Not quite the same as this fix, but this knowledge has saved me from hunting through browsing history a few times.

Seeing the perspective of somebody who’s not particularly well versed in Android forks is interesting, though.

I found the part around 2:45 to be interesting, where the YouTuber says the thought of the OS getting compromised was scary. This is a sort of privacy paradox where Calyx looks worse than other, less honest, alternatives.

Could a rouge employee compromise Calyx? I guess, but Calyx has the best possible setup to avoid it. And Android itself is basically compromised by default, which should be far more concerning. The biggest reason people aren’t concerned is because Google understands PR, and they know how to spin things in the most positive light possible.

Mozilla FakeSpot sells your private data (profile, location, browsing and search history) to advertising companies.

https://lemm.ee/comment/11216210

And when Mozilla released the Orbit AI assistant, they linked to FakeSpot for some reason…

Edit 2

The official Deepfake Detector privacy policy is missing from the FakeSpot website. This is the one Mozilla links from their extension policy.

Edit

Firefox may be bad in many ways, but the biggest alternatives are generally worse. There are some reasonably good forks like Zen and LibreWolf, but if you have the wherewithal to patch Firefox’s user.js on a desktop, you’ll probably be better off with a patched Firefox.

https://lemmy.world/post/4064988

https://dataswamp.org/~solene/2023-09-24-harden-firefox-with-arkenfox.html

The best service I’m aware of is archive.is, which has been around for at least a decade, and does a good job taking snapshots of pages the way they’ve rendered. Beggars can’t be choosers, though, and the alternatives are sparse. (Their site does recommend a couple others, though.)

Typically I would prefer The Web Archive, but currently their site is experiencing issues with Twitter URLs.

For clarification, is there a reason you would prefer xcancel links in particular over other frontends? I’m entirely uninformed on the matter, outside of seeing this service used relatively often and recently, compared to other Nitter instances.

I’m not the person you asked, but in my opinion, archiving services are more reliable than simple frontends, since they will continue to work even if the tweet in question is from an account that deletes or protects the tweets later, or if the account is suspended by Twitter. Considering the tumultuous relationship Twitter has with both reality and its users, this might be worth consideration too.

Your writing is better than serviceable, I just had a brain fart while reading it

I thought you said “What’s next, a few ads in the URL bar?” …because Mozilla has put a few ads in the URL bar.

This sounds like a good change! It’s bound to be frustrating to a lot of people, but it’s good to see protecting your passwords on their servers is the number one goal of this company.

I don’t see how enabling federation will fix the problem of not knowing what is running on their servers. You’ve just introduced a new problem: other servers, with their own rules, which may also be peppered with requests for data and gag orders.

This change will impact how you set up Signal on your desktop computer. Previously, after linking your desktop to your phone, you would be presented with basically an empty window.

This change will allow you to, optionally, synchronize your message history from your phone to your desktop, filling it with your previous messages, making it much easier to pick up where you left off with your conversations.

Pictures and videos that were sent will also synchronize, as long as they are from the past month and a half.

Kind of funny they list their built-in, paid VPN as a positive feature and not a negative. Maybe they were running out of good things to say about… Themselves.

Granted, Mozilla also shot themselves in the foot by saying Firefox was better for not blocking ads by default, but that’s a different story for a different day

Brave still does include ads enabled by default. You need to disable sponsored images in the New Tab page.

News feeds seem to be a symptom of enshittifiaction. At least you can still get a functional and minimalist homepage on Firefox by disabling it.

Remember Google Now, that “homepage for you” on Android that showed you the weather, reminders, calendar events, etc… But eventually Google removed all the functionality and replaced it with an infinite feed of news slop.

Slop. Feed. Rather synonymous.

fake news drama storm

Uh… No?

Proton’s CEO just hijacked the company account, wrote a bunch of stuff that said “Our team” at the beginning. Then he claimed he had accidentally used the wrong account and accidentally spoken for the entire team.

I could have been 100% on board with everything the CEO said, but then his rapid denial of obvious facts is a huge deal in itself. Proton’s entire existence exists upon being trustworthy, and if somebody’s going to clearly lie, trust gets broken fast.

YouTube memory leak issues allegedly fixed.

It looks like GPC spec creates the same sort of tracking signal that DNT did, but it requests less protection: there is specifically a carve-out in GPC that says websites can track you, including for advertising purposes.

GPC is also not intended to limit a first party’s use of personal information within the first-party context (such as a publisher targeting ads to a user on its website based on that user’s previous activity on that same site).

I think the article made a typo that claims GPC is the same as DNT.

When you enable the feature, the GPC sends a signal… This signal is sent via a special HTTP header called DNT: 1 (Do Not Track)

But the GPC spec does say it sends a new signal: Another header (like DNT) and a JavaScript variable the client would set. I don’t see why this couldn’t be used for tracking too.

A user agent MUST generate a Sec-GPC header

So if it generates a header, it can still be used for fingerprinting, but this header is actually less restrictive for what the receiver must do.

DNT was “do not track,” and GPC is "do not sell:

GPC is also not intended to limit a first party’s use of personal information within the first-party context (such as a publisher targeting ads to a user on its website based on that user’s previous activity on that same site).