Live data from Hacker News

Firefox 90

mozilla.org

101–110 of 351 posts

Re: Firefox 90

#101
post #24

Earlier quoted context omitted.

> wish they'd just become a privacy-focused Chromium fork because These already exist, and Chromium is far from perfect. I'm glad there's a viable competitor.

In my experience, Firefox actually has better performance than Chrome nowadays — particularly on Apple Silicon, where the difference is stark and dramatic. And with the new "Proton" interface, IMO Firefox finally looks at least as nice as Chrome out of the box. Firefox mobile still has some catching up to do on the overall UI/UX, but it's not too far off, and the fact that it allows extensions is nice.

this is interesting - i've had the opposite experience on both x86 and ARM Macs to the extent that it's not really a viable browser for me :/ purely anecdotal of course, and i haven't found a lot of information supporting my experience...

Re: Firefox 90

#102

Background updates! Hooray! That may sound like a basic table stakes feature but honestly it's a huge improvement. And a lot of other apps do a terrible job of it, particularly on Windows. It's remarkable how many apps require 8, 10 mouse clicks and a restart to install a minor update. Firefox didn't require any clicks but the nag / need to restart was a nuisance. Actually now that I type that I wonder what has chang…

The key here is "while Firefox is not running." They have a maintenance service now that can check for updates when the browser is closed. Personally, I wish they hadn't done that. First, because my browser is always open. But also because I don't need another service.

The maintenance service is not checking for updates. It is still Firefox itself that checks for updates, but instead of launching the updater.exe (which triggers the UAC dialog), Firefox will start the maintenance service and tell the service there is a new update. The service will stop Firefox, run the updater to install the new version of Firefox (which does not require a UAC because the service already has elevated permissions), restart Firefox (updated version), and then stop itself. So this is not your typical update service such as Java or Adobe have which continuously runs in the background, but rather an interesting trick to run executables with elevated on-demand of a non-elevated executable.

Re: Firefox 90

#103
post #87

Honestly I'm still not used to the new tab style that was introduced last time. 90 seems fine, but not much to get excited about from the user perspective.

loving 'Firefox-UI-Fix' which fixes all the density changes introduced as well https://github.com/black7375/Firefox-UI-Fix/

Agreed - this makes the ui much better.

It's annoying have to manually enable compact in about:config to get back to the compact layout in new profiles.

Re: Firefox 90

#104

> FTP support has been removed sigh

Honestly, I'm more than happy to see older/niche protocols removed from web browsers. God knows the average browser contains way, way too much stuff anyway[0], not to mention there are FTP clients that are likely far better for that use case than Firefox would ever be. I wouldn't mind if FF (or other browsers) tightened their scope a little more often and adhered to the Unix philosophy a little more closely.

0: https://drewdevault.com/2020/03/18/Reckless-limitless-scope....

Re: Firefox 90

#105

> Now, SmartBlock 2.0 in Firefox 90 eliminates this login problem. Initially, Facebook scripts are all blocked, just as before, ensuring your privacy is preserved. But when you click on the “Continue with Facebook” button to sign in, SmartBlock reacts by quickly unblocking the Facebook login script just in time for the sign-in to proceed smoothly. I wonder how it does that, and if it's open to exploitation by a malic…

Basically, if a site uses the Facebook SDK to start the authenticaion flow, it will be broken because it's blocked.

SmartBlock just stands in for that blocked script, acting just enough like the real API to prevent some common site breakage.

This lets it detect when sites try to open the popup-based authenticaion flow, unload the stand-in script, load the real one, and continue the login flow.

It only unblocks some Facebook resources, as listed here: https://searchfox.org/mozilla-central/source/browser/extensi...

It only unblocks it for that specific website (the domain in the tab URL), and only until Firefox is closed. Other trackers continue to be blocked, and other tracking protections stay in place (as opposed to turning off ETP for the site entirely).

Re: Firefox 90

#106
post #78
post #63

Earlier quoted context omitted.

I mean sure, but realistically speaking, how can we afford to continue supporting Firefox? I can't make a plea to the CTO of "Please let us keep investing money in compatibility because it may have a minor impact on the open web." Firefox is in a downward trajectory for marketshare and mindshare. I would love to hear a realistic solution to this problem. Maybe switching to WebKit?

Given how many discussions I had in the past about people insisting to support legacy, unsupported browsers I'm surprised "5% of our users use it" isn't enough to convince almost everyone to support it.

If you mean Internet Explorer, it's because the deprecation of that is starting now because Edge supports IE mode for legacy apps and is otherwise Chromium.

Re: Firefox 90

#107

Earlier quoted context omitted.

Did you use it?

Obviously not as much as HTTP(s) but still there are FTP links which show up after a google search from time to time...I guess now I have to context switch to the file explorer...Thunar is OK but does Windows Explorer even support FTP?

Personally, I've always just associated ftp:// with my preferred client and the browser just opens my client for me when I hit one. Works pretty good.

Re: Firefox 90

#108
post #12
post #5

I think Firefox has ~5% of our user base. It's hard to justify running separate tests for it anymore. I wish they'd just become a privacy-focused Chromium fork because I don't see them crawling back to a meaningful marketshare at this rate. (Yeah I know multiple implementations and all that, but we don't have many people running FreeBSD just to give Linux some competition.)

I wish more people would run FreeBSD because it would make both Linux and FreeBSD stronger. Single platforms stagnate, we’ve seen it before with IE6 and we’re starting to see it again.

Ironically, as much as I love FreeBSD, I am being forced to migrate to Linux because the USB mass storage support on FreeBSD is broken on Dell rack servers. I recently migrated to a newer generation Dell server w/ 16TB disks to replace my aging fleet of R710s and cut my power usage considerably, and found I could no longer reliably keep a server online if I boot from USB. There's several threads about this on the FreeBSD forums, including one I started, and the general consensus of the community is "don't do that thing you're doing that works perfectly fine on every other OS, even Windows." not "FreeBSD is broken and this bug should be fixed."

I've fought the good fight for FreeBSD for more than a decade, and my last system running it is getting Debian installed on it in the near-term future.

Re: Firefox 90

#110
post #92
post #5

I think Firefox has ~5% of our user base. It's hard to justify running separate tests for it anymore. I wish they'd just become a privacy-focused Chromium fork because I don't see them crawling back to a meaningful marketshare at this rate. (Yeah I know multiple implementations and all that, but we don't have many people running FreeBSD just to give Linux some competition.)

Were you around during the IE and ActiveX days? Because that's what happens when a single vendor gets to dictate web standards.

You say that like Google doesn't pretty much already dictate web standards.
Post reply on HN