The biggest problem is that Firefox updates the next time you open the app, without warning.

That's not a sane decision. Sane would be to update on closing.

Same way Windows offers to "Update and Shutdown" (you don't need the PC anymore why don't I under after you walk away) and not "Update on next Start up" (I'll wait until you need me to get in your way).

> Same way Windows offers to "Update and Shutdown" (you don't need the PC anymore why don't I under after you walk away) and not "Update on next Start up" (I'll wait until you need me to get in your way).

This presumes that shutting down is never a time-sensitive action. If I had a nickel for every time Windows decided it wants to spend a half hour installing updates when I'm trying to shut down a laptop so that I can put it in my bag and go somewhere with it, I'd be able to actually buy a Windows license instead of resorting to vlmcsd.

I run a desktop workstation with a UPS, its lead acid batteries are not infinite and don't like staying discharged, so at times shutting down quickly and cleanly is very important. The more I hear, the happier I am that I moved to Linux a decade and a half ago, and this general topic is a major reason I'm planning to go back to Debian after using Ubuntu LTS for my last two versions.

You would probably script out that shutdown sequence, and IIRC shutdown.exe can bypass the install updates step.

I should have left a "sequence shortened" warning. Windows 10/11 offer you 4 options when an update is pending. "Update and shutdown", "Update and reboot", "Shutdown" and "Reboot".

Perfect!

As a side note, I've never understood why Windows and macOS updates take so long compared to Linux updates.

I wanted to check if it is due to stacking (security) patches instead of just extracting whole tarballs and the first result I got from Google was: https://www.quora.com/Why-are-updates-in-Linux-much-faster-t...

Quora always had reputation for being shit but I am impressed by how terrible that page is. I mean there are only two 2 types of answers there: "Window$ is BAD" and stuff that looks like generated by GPT-3( or straight up from those infamous "recipe sites" a.k.a. SEO farms).

If you use win-x to access the shutdown menu, it will give you options of shutdown or update and shutdown. Although, I think that's really shutdown and update on next boot though. At least this is available in Windows 10.

Agree with yellowapple. Update on shutdown isn't great either. The sane decision to me is to let the user decide when to update their software.

Windows does, you can update whenever you life. All 4 possible options are offered when you go to shutdown, including skipping the update to shutdown/reboot now.

Windows does sometimes get it correct.

I would expect a software to close itself when I ask it to instead of hiding behind and update itself.

Just yesterday any new tabs I opened refused to load until I restarted Firefox due to a forced update. Existing tabs became unresponsive. Granted, it did restore my session upon restart, but it did not restore the DOM and data state of those tabs.

I see the "Firefox has updated, you must restart Firefox to continue using it" window every two weeks or so on Ubuntu.

Though I never seen it on Fedora or Windows, might be some quirks related to Ubuntu packaging.

This is a consequence of most Linux package managers not caring if apps are running and updating their files anyways. Firefox does this because it would crash otherwise (as most other software does when you pull its files from underneath it). If it happens frequently, you must have automatic live updates enabled - maybe you don't on Fedora and that's why you don't see that happen?

On Windows, where no standard update mechanism exists, Firefox updates itself, so it does it only when it knows it's fully shut down (before it opens or after you close all windows).

No, Fedora package manager (dnf) does the sane thing and not replaces app files from under the running apps by default - google "dnf offline upgrades".

And for Windows there is standard update mechanism - Microsoft Store - which also does the same sane thing.

So it's Ubuntu implementation problem, and not a Firefox one.

>dnf offline upgrades

Do I understand correctly that it is based on systemd's offline upgrade mode, i.e. installation is done at the following boot? If so, you can do that on Ubuntu too. I'm not sure how (perhaps by invoking apt-offline?), but at least KDE Discover does it on Kubuntu.

> Firefox does this because it would crash otherwise (as most other software does when you pull its files from underneath it).

Most software does fine, as long as it has already mapped in all of its libraries, those files can be unlinked in the filesystem, but will remain on disk in their current form. You do run into trouble if the software is part way through starting up and hasn't mapped all the libraries, or if you update the files rather than unlinking and copying new ones (at least, FreeBSD lets you update mmaped files and all processes will see the updates, which is usually not what you actually want when the files are code).

Windows makes it very hard to update program files while programs are running, so it's not surprising Firefox endeavors to do updates while it's not running.

Due to the way multi-process browsing works, Firefox at any time still needs to be able to launch new processes in order to accommodate new tabs, or thanks to site-isolation possibly even browsing to a different origin within the same tab. Once the old binaries have been unlinked, I assume there is no longer a sane way of launching a new process using the old binaries?

At start, you could start a new seed process, map all the libraries and what not, but don't do any browsing work... When it's time to launch an isolated process, fork the seed process and use the child.

This doesn't seem insane to me, but might not be an easy thing to fit in to the large body of code that's already there.

You still have issues if you're starting up during an upgrade, but that's a problem with anything dynamically linked.

Doesn't using fork make it more difficult to cleanly separate the various child processes from each other? It also makes address space layout randomisation less effective, because every forked process starts out with the same base memory layout, doesn't it?

> Doesn't using fork make it more difficult to cleanly separate the various child processes from each other?

I don't think so? As long as the main process is separate from the seed process, and the seed process is forked before specialization.

> It also makes address space layout randomisation less effective, because every forked process starts out with the same base memory layout, doesn't it?

Yeah, I think you're right. Not sure if you could maybe do some randomization post fork, but you're probably right.

Unfortunately, this breaks horribly if you're on a shitty network (for example I work a lot on the train) and/or slow hardware. The update check delays browser startup for more than a minute when connected to wifi with no internet (this happens a lot when going through rural areas) and the update after closing function has also burned me a few times when I closed the last window, then remembered I needed to check one last thing, but had to wait several minutes for the update to do its thing (a combination of old laptop, slow internet and many other things using resources).

> The update check delays browser startup

Are you sure that originates in the update check? The update itself is downloaded asynchronously while the browser runs. The browser itself checks if an pending update has been downloaded when it is launched.

I wasn't able to verify how FF actually behaved under snap, but snap's confusing notifications was enough to make it annoying. After failed tweaks, I'm glad that I no longer have to deal with this since I already purged FF that comes with Ubuntu and just used the binary from Mozilla's site instead.

This is not my experience on Ubuntu. It updates while I'm using Firefox. I then open a tab and get the stupid message saying that welp, you need to restart that browser boy because isn't this a great UX?

I don't think this is a fair decision. Virtually nothing requires such a forced update. And the browser should be sandboxed by the OS anyway.

I’m not so sure that they only download. On macOS, Firefox (and Chrome) sometimes stop working in odd ways when they do the background ”download”. For example, file pickers stop functioning or tabs stop loading. All of this starts working again after restarting and letting it update. It’s happens more than a handful of times every year. Enough so that I’ve picked up on the pattern that it only happens when I see it say “update” by the settings menu.

On my Ubuntu system it applies the update while still open. Then I open a new tab to do something and I'm told I have to restart.

> This does not interrupt your browsing in any way.

That's not been my experience. I don't know what system you're using, but with Firefox on Ubuntu, I discover that Mozilla has put out another (goddamn) update (again, already, goddamnit!) when all my tabs suddenly crash, forcing me to restart the browser.