Unfortunately by sending DNT you are merely suggesting to the server that you wish to not be tracked. There's no requirement for the server to actually care about you at all.

Now, if DNT were actually legally binding though - that would indeed be very cool.

If a website has a compatible PWA manifest the there will be an item labelled "install" in the three-dot menu of Firefox in place of usual "add to homescreen" item.

Edit: There's a few other requirements as well for the website to be considered installable as PWA, such as it must have a registered service worker so it can work offline. But regardless, if the website provides all the requirements then it can just be installed straight from the menu.

None.

I don't think it's a good idea to take some huge collection of prefs and just apply them blindly.

Instead, make the changes that you actually want to do, so that you actually know what changes you are causing. If you want to put those into your user.js file then feel free, but in my opinion it's just better to change them in about:config directly - that is, unless you need to apply the exact same set of changes to multiple profiles.

Right, but then you shouldn't be shocked to find out that a feature was removed because nobody seemed to be using it.

My first guess would be that this is caused by the website implementing its own navigation/history behavior using History API. That can easily mess things up, or at least not behave like you might want.

Well the feature development is certainly progressing - here is the tracking bug for it.

You can nowadays just test it in normal nightly without special build - it's extremely incomplete, but you can test it if you wish. It's tied to revorked sidebar which you need to enable in about:config.

Also, with mv3 extensions, the extensions that are not actively doing some work are not really "running" in the first place but are just waiting for some event to happen that they have previously told Firefox to inform them about, but there isn't any persistent execution context that is constantly running.

I believe one reason why extension support was not-fully enabled earlier was because mv2 extensions required persistent background context for each extension and that could cause issues if Android just decided to kill that process. But with mv3 the extensions are required to be able to be suspended and then woken up on demand.

IIRC the old tab groups feature was eventually removed because telemetry showed that only very few people used it...

I think this really comes to what exactly you want to separate. You say "I often need to use two different profiles". Okay, why do you need to use separate profiles though? Maybe separate profiles are not a great solution in the first place for your purpose?

Firefox profiles are amazing because you can be sure that no data is shared between the two profiles (unless you sync them of course) - for whatever reason one might want that. But if you just need some session separation then containers would be a much better fit.

In that case the issue is likely that files on disk are being modified by whatever mechanism your IT uses to push updates to devices. If the program files are modified while Firefox is running then you will unavoidably get this prompt.

I suppose the best you can do is to ask your IT folks to not update programs that are currently running.

Yeah, history is extremely valuable feature. I think I would rather get rid of bookmarks and maybe even tabs rather than history.

I guess this style might fit the bill

I mean, first you need to set up using custom style sheets - some guidance here, but after you've set that up you can try the style I mentioned.

There isn't anything to specifically update the favicon of a tab. But you can do the exact same thing websites can update their favicon with; run a content script on the web page that modifies the favicon. Firefox will then show whatever the new icon image is in the tab bar.

I don't have a better example but the content script of an extension I wrote updates the favicon during media playback in a tab. As an overview it creates a canvas of the original favicon, creates a new link element as the "new favicon", updates that canvas during playback, and then sets the created link element href attribute to the data uri created from the canvas.

Right, and that's fine. The one good thing about these "collections" is when they describe what the pref does (I mean, so does official source typically). But that matters only as long as the audience actually reads those descriptions. But then if you just pick the ones you actually care about (which you should totally do) it becomes irrelevant from which "collection" you found about it from.

Okay, since it doesn't like it's your main computer or anything, you might be interested to try taskbar profile grouping. Go to about:config and while there create a new boolean pref named taskbar.grouping.useprofile and set it to true. Doing that the two profiles should have their own group in taskbar. It's a very crude feature though, since for example the right-click jump list items are not separate and you can't set different icons for them (unless you do that via Windows somehow), but it sort of works.

This should be pretty simple to do, although what exactly do you consider a "duplicate"? Does it count as duplicate if the bookmarks have the same name, or same url, or does both name and url need to be equal?

Edit: So the extension could work like this - it's awaiting review, but the eventual AMO page would be here

Not at all the same thing. For one, you can open a history entry and then navigate back from that to the page you came from to that page - which there may be several. Tabs preserve per-tab history which makes it superior in many ways to both history and bookmarks.

You just have the link texts in a text editor one at each line, then select all and drag the selection to tabs toolbar.

But yeah, it does become an issue if you try it with thousands of tabs... It should work, but probably chokes quite a bit.

Notepad++ works great for taking notes. Besides, it's open 24/7 anyway so dedicating a tab or two (or dozen lol) feels pretty natural.

I only use Joplin if I have some "very structured" notes about some topic, and while that is also open 24/7 np++ is always my go-to because "it just works" without having to care at all about formatting nor anything fancy.

Indeed, but what I don't get is why on earth do people spew this damn crap about manifest v3 as a whole, when the actual issue is just the removal of "webRequestBlocking" feature that Google is about to bring along with their implementation of mv3. Why the hell aren't folks mad about the actual issue but instead just want to be mad about the whole bloody thing, which actually also does bring some very real privacy improvements among other nice things.

This isn't true anymore. Any extensions that are explictly marked as Android compatible (by the author) should now be installable from AMO. I'm pretty sure that the extension needs to use manifest v3 to be able to be Android compatible, but as long as that is true and the author has set the Android compatibility flag, you should be able to add it normally.

Installing from AMO seems to work fine for me - although I'm using Nightly variant, which certainly could behave differently. However, Firefox 120 is supposed to work the same I believe.

I don't know if the syncing part will work for unsigned extensions, but you need to use the same storage area on both endpoints. So if you set kitten key to sync area (browser.storage.sync) - then you need to also read it from the sync area, but in your example you are reading from local area which is for local not-to-be-synced data only.

That ain't normal. You can certainly customize Firefox to behave that way with custom userChrome.css file, but you would have to do that on purpose.

I suppose it's also possible that if Firefox is installed via your package manager then they might do some customization to it. I've seen some linux distros package these sorts of custom Firefox builds that apply various changes to official Mozilla builds.

I mean, do whatever works for you, but that sounds kinda unnecessary when you can just use an already existing feature - tabs.

Fair. For something like that containers don't work, and indeed profiles are probably the way to go. I sure wouldn't mind if about:profiles had a button to create new icon for that specific profile which then would also be in its own taskbar group, but I doubt I would want it as default for new profiles.

At any rate, having multiple profiles per same install on same Windows user poses some issues. Like what profile are links in other applications supposed to open in?

Huh. I dunno what the actual requirements for Android compatibility are then. Anyhow, being recommended or in a custom collection isn't strictly a requirement anymore.

Yeah, that indeed sounds like a bug.

I guess that kinda depends on what you mean by "upload". If you simply mean if user can point the extension to load some local file eg. via `` then sure, that works just fine.

It sounds like the issue you mention regarding uBO Lite is not about if it can read a local file, but rather that it can't use that data to update the active filter lists - perhaps because it already has too many dynamically inserted filter rules, but that's just a guess.

I have no clue about hacking macOS

Take this with a grain of salt, but I believe I've read that standard backdrop-filter won't work here and this would instead require OS compositor level mechanism because menupopup and panels are technically separate windows (or window-like widgets) from OS perspective.