I'm really looking forward to the overhaul on the road map that moves more things off the main UI thread.
I keep switching from Chrome to Firefox for reasons of principle ... and keep switching back to Chrome for reasons of performance. I hate myself, but can't help it. Firefox will briefly freeze scrolling a long page. Firefox will beach ball for 15 seconds after resuming laptop from sleep. And so on. None of it is crash/burn awful, but it adds up to be just irritating enough that it makes me mental. I tolerate it until I can't (for a couple weeks), then switch back to Chrome. Just did so yesterday, which is why I'm ranting in response to this headline.
This on OS X. Couple years ago when I was using Windows 7 all the time, Firefox was great. IME the fit/finish on OS X is not the same.
tl;dr: Firefox, I want to love you and be exclusive. Please change.
Considering extensions are the #2 reason (behind Flash) that Firefox crashes/beachballs, perhaps one would be wise to audit the (presumably) tons of extensions they have installed locally, to see if they conflict with other extensions?
If what you're saying is true, then your handful of extensions are conflicting with each other. disable one of the handful, and see if the crashes disappear. if not, disable another one.
currently using a netbook with an Atom N570. I appreciate under-powered computing. Firefox runs fine here with 5 extensions (two of which are for web development). 17 tabs are using 214.52 MB of memory. haven't had a crash or hang in a long time.
I could've written this myself. I love Firefox and I used it for years but at some point it became rather unusable and after it forgot which tabs I had pinned as app tabs for the fifth or so time (because I had to kill a completely unresponsive Firefox, this happened often) I was angry enough to actually give Chrome a try. It is SO much more responsive, even just the UI. I've kept Firefox up-to-date but I don't want to keep switching back and forth so I haven't made a switch back. I hope I'll get to, though.
This is all on a Windows 7 machine though, rather than OSX.
Firefox has come so far from the memory hogging 4.0 days. Like others here, I give it a shot ever few month and UI responsiveness always sends me back to Chrome. The resurrected electrolysis project can’t come soon enough.
The irony is what use to be its greatest flaw, memory usage/leaks, is now it's greatest strength. Around Chrome 15+ memory usage started ballooning and was too much for my 2GB netbook. A laggy UI is much better than an unresponsive system from thrashing RAM.
Yeah, I don't mind big memory usage per se. I have tons of RAM, 8GB in fact. Take 1 if you need it, if it results in a responsive application. Chrome uses something like that right now. But if I hit ctrl-t on Chrome, I INSTANTLY get a new tab. Firefox lacks that. You see Firefox moan and sweat for a few seconds, and then poop out that new tab you ordered. But that's not the worst thing: the worst thing is that it would occasionally (say once every 2 weeks) crash and upon restart completely have forgotten what tabs I had open and that is a pet peeve of mine. I want it to be more reliable.
> crash and upon restart completely have forgotten what tabs I had open
I've run into some strange behavior like this before too. When you reopen Firefox and its lost your session, try opening another window or two and see if 'Restore Previous Session' is finally available.
If that didn't work, there are also a few other tools that can help solve restoring your sessionstore.js:
On Nightly it's even better, it's aggressively compacting memory, it was running around 30% (2GB total) instead of 40+ a few weeks ago. Kudos for their effort. I wish they could replicate a tiny bit of it toward the GUI.
If you don't know already, a huge UI overhaul is planned to land sometimes between Firefox 25 or 26 [0]. As far as I know, better performance is a big reason for the change as well.
I didn't know they were trying to fix performance too. To me this was only a UX redesign. Which I tried not long ago, and I was disappointed, it doesn't add much actually. A few theme changes here and there, a nice configurable menu that personally I'll never use. I'm not sure what to think about it ... we'll see.
Google+ is a great example - it slows to a crawl after a few minutes of browsing - it particularly becomes unusable when I scroll down and it loads more content via AJAX.
Well, I am on Windows 7 and it is fine enough but at times, it does get slow. I also keep changing between Chrome and Firefox. Chrome just feels way faster.
Apart from that, Chrome has very nice features that make life easier.
The PDF preview of Chrome is great. It can rival the PDF viewers on desktops easily. Heck, it helped once when I had to get a printout at a shop and they didn't have PDF Reader installed. I just dropped PDF onto Chrome and we had a PDF reader that could print. I know about PDF.js but it is not so good!
Next, the UI could be improved. Firefox has its upsides as well. I love the hidden status bar. I have dropped all my extension icons there. Now, with one shortcut, they are accessible and for normal browsing, don't come in between. But Chrome's way of managing extension icons is great, and one thing that particulary drives mad with FF is that there are no visible zoom controls. I have been using it on and off for a year now but I still don't know how to find if I'm zoomed in or not.
Chrome has really good Flash integration and I didn't know that I did not have Flash Player installed till I started using Firefox. What is stopping them from bundling a plugin like Chrome?
Well, I'm a student and lot of searching for me ends on PDFs. Some with figures etc. are broken and display fine in Chrome. FF is good but Chrome is much better, that was comparison I was making!
I have a 1.07GHz Core-i5 laptop (with Nvidia Optimus) at home and 3GHz Core-i5 desktop at work. I happily use Firefox (Aurora) at work, but at home I have to use Chrome because Firefox is simply way too laggy (it improves a bit when run with Nvidia).
Same boat. I switched recently (when switching feed readers!) and cannot believe how lousy the performance is and inelegant the UI. Very jerky scrolling, flashes to re-render CSS, beachballs, stalls, ancient UI paradigms...
After using chrome and even safari, the UI feels so "thick". I think maybe a release or 3 focused entirely on performance and UI cleaning may be in order.
I love firefox too. But for me it always starts slowing down half way through the day (about the same time as I do). Maybe its firebug? This has happened on windows and mac. Restarting it makes it work better, like myself after lunch.
If you hover that button with your mouse pointer you can read the following explanation:
Send three "heap-minimize" notifications in a row. Each notification triggers a global garbage collection followed by a cycle collection, and causes the process to reduce memory usage in other ways, e.g. by flushing various caches.
Is there any good reason for removing the distinctive behavior of an element that so many older websites use, even if it is considered a design faux pas? History should not be destroyed.
I agree that history should not be destroyed and in sites that have old pages with the blink tag it won't be. But I think its fine for the browser makers to remove old cruft. Especially in this case where the behavior can easily be replicated in Javascript.
Those first few (dozen?) releases people constantly complained about the version number changes, the rapid release cycle, and there was quite a bit of fragmention.
There days there are a billion people getting a new HTML5 browser every 6 weeks.
> Support for new scrollbar style in Mac OS X 10.7 and newer
Finally. That one took quite a long time for them. It looks simple in the change log, but I suspect that it was actually quite a big change under the hood.
EDIT: Woops, it seems that this is an error on release page, since they've hit a regression [1] and they've pulled the feature, it's due to come back in v24. And is seems I was right that this was actually a pretty big change under the hood [2].
EDIT 2: And they now removed the error about new scrollbar from the article.
I know that the release notes still says Firefox Beta when I submit this, but my browser just updated itself to version 23. The beta release notes usually are pretty accurate when it comes which features will be in the next stable version.
I believe the new baseline compiler is included in this release as described here:
I didn't understand why they would disable this feature which I love so much, but reading the bug explained that it was to prevent malicious apps from being able to change the default search in a way that is not easily reverted by users (keyword.URL can only be changed through about:config). Makes sense to me, and fortunately we have add-ons which can restore this functionality :)
I see that Mozilla is now signing hashes of Firefox downloads with a 4096-bit RSA key. But the key used to sign firefox-23.0.tar.bz2 (id 0x057CC3EB15A0A4BC) is only self-signed and was created three weeks ago. It's not on pgp.mit.edu. Is there any actual way to "validate the authenticity of these keys in an out-of-band manner" as the KEYS file recommends?
the primary key is 2B90 598A 745E 992F 315E 22C5 8AB1 3296 3A06 537A not 5445 390E F5D0 C2EC FB8A 6201 057C C3EB 15A0 A4BC
if you search for the primary its there (at least it's on pgp.mit.edu as far as I can see)
it's the primary that you need to trust for the trust to work
For FF 23:
gpg --verify SHA1SUMS.asc
gpg: Signature made Tue 30 Jul 2013 09:32:39 PM PDT using RSA key ID 15A0A4BC
gpg: Good signature from "Mozilla Software Releases <releases@mozilla.org>"
gpg: WARNING: This key is not certified with a trusted signature!
gpg: There is no indication that the signature belongs to the owner.
Primary key fingerprint: 2B90 598A 745E 992F 315E 22C5 8AB1 3296 3A06 537A
Subkey fingerprint: 5445 390E F5D0 C2EC FB8A 6201 057C C3EB 15A0 A4BC
gpg --sign-key 0x8AB132963A06537A
gpg --verify SHA1SUMS.asc
gpg: Signature made Tue 30 Jul 2013 09:32:39 PM PDT using RSA key ID 15A0A4BC
gpg: Good signature from "Mozilla Software Releases <releases@mozilla.org>"
It is however, indeed, only self-signed right now as far as I can see.
Thanks; I had searched on pgp.mit.edu for the primary key but forgot to add the hex prefix 0x.
Curiously, the 1024 bit DSA key used for some previous releases (0x7f4d66451ebcab3a) has been signed by "Someone at Mozilla Should Sign the Release Key (so users can verify the key's owner!) <anonymous@lulz.example.com>".
The biggy for me is <input type="range"/> support. I’ve already ripped jQuery UI out of an internal project today; it was only there because of that omission.
For those who upgraded and still see the old scrollbar: despite what Beta Release Note said, Lion-style scrollbar is not in this release due to regression. It will be available in Firefox 24 instead[1]:
>The new scrollbars has been disabled on Beta for Firefox 23 due to a serious regression. It is now scheduled to be released in Firefox 24 as long as there are no serious regressions.
Never had any stability problems for more then a year now, on both work Windows 7 x64 and various Linux variants at home.
A year ago there were a few addon problems here and there, but these days addon authors are used to the new release cycle, so I don't have even these problems anymore.
Same here. However, it's still not showing local resources that are not being loaded because the path or the name of the script or stylesheet is wrong.
The idea is that non-technical users shouldn't have options in preferences that will break the web for them, and that technical users will use about:config or an add-on (e.g. NoScript). I find it difficult to argue with that.
"Non-technical users" is a contradiction in terms. If you're using technology, you're a "technical user".
If you don't know how certain options work, you should leave them alone; but if you don't leave them alone and something breaks, it takes only common sense and not specialized knowledge to suspect that the thing you changed has some relationship to the thing that broke.
De-featuring software for the presumptive convenience of a presumptive lowest-common-denominator user is not a good practice.
It's true that users should not break things without some mind as to how to fix them, but it's also irrelevant. If a user breaks your product, it makes no difference to him or her who broke it. It's still broken. Users will have a bad experience.
This doesn't mean advanced options shouldn't exist, but it does mean that they should be out of the way for non-advanced users. In this case, the setting is moved to about:config or to add-ons, so I don't see how we've lost anything.
Are you opining that some users may not be able to find the setting in about:config? Would they not just enter "How to disable JavaScript in new Firefox" into a search engine?
>"Non-technical users" is a contradiction in terms. If you're using technology, you're a "technical user".
This is a debate about semantics. "Non-technical" is a colloquialism. I see no benefit to this train of thought.
Users will have a "bad experience" as a consequence of an action that they took, and can undo just as easily.
A good user experience can be achieved by providing users with an efficient way to understand and fix problems that they encounter.
De-featuring a product to prevent users from ever encountering a problem reduces the utility of the product, which produces a worse experience for everyone, not just the ones who would have had a problem they couldn't solve.
Those who would have had a problem they couldn't solve will still ultimately have a bad experience, because it's their own pattern of usage, and not any defect in the software, that's ultimately getting them stuck.
Moving an "advanced" setting to a different interface is fine, but the release notes don't say that the JavaScript toggle was moved to about:config; the notes say that it was removed (and that user-set values will be reset to default, which, IMO, is an unacceptable thing for an update patch to ever do.)
>A good user experience can be achieved by providing users with an efficient way to understand and fix problems that they encounter.
Agreed.
>De-featuring a product
If I were building a browser today, I wouldn't want the option to be where Mozilla had placed it. I think of this more as a correction than as removing a feature. I don't think we're going to agree on this one, which is fine.
>produces a worse experience for everyone
Not at all. The majority of users do not need this feature, and those that do are likely to be sufficiently experienced to find it anyway.
>Those who _would_ have had a problem they couldn't solve will _still_ ultimately have a bad experience, because it's their own pattern of usage, and not any defect in the software, that's ultimately getting them stuck.
This is debatable. Good UX means preventing these kinds of pitfalls.
>the notes say that it was removed
This is a failure on their part to communicate the move, which sucks. It's also possibly a direct attempt to prevent people from using this feature for whatever reason, which is kind of sleazy.
>and that user-set values will be reset to default, which, IMO, is an unacceptable thing for an update patch to _ever_ do.
I'm pretty sure we're in agreement here. I take no issue with moving the option to about:config. I also wouldn't take issue with it being removed entirely, because add-ons such as NoScript are superior to that feature anyway.
So the people who do act that way are fake people?
I've worked a helpdesk, and I know where the stereotypes are coming from. I also know that selection bias is selection bias, and that there's never an excuse to remove functionality from software.
Also a dealbreaker, no "disable HTML" option. It's critical to my daily workflow to be able to turn off all content and browse the web in peace and silence.
Although it's not a deal breaker for me (I am savvy enough to use about:config, NoScript, etc), I think this is bad for users in general. Many of them are not tech-savvy enough to dig into hidden configuration options. Though they certainly deserve better security when they browse the web.
The very large banner at the top of the page seems to indicate that this is a beta release, so shouldn't that be indicated in the submission title? I was excited about a new release of Firefox (namely to see if performance improved), but if this is beta then I won't be touching it.
That was supposed to be switched on by default in Firefox 22, but Mozilla noticed some issue with it. The feature is there in FF22, but you have to switch it on yourself. If you you want to know more about the issues, you can read what Brendan Eich had to say:
It feels like OSX will be on Mavericks before the Lion scroll bars are supported in Firefox.
I've been using Safari a lot lately. Only thing I'm not crazy about is the inspector. It's easily the worst of all the browsers, but at least they are fixing that in the next update.
That's interesting - I prefer Safari's web inspector to any other browser... The updates look cool, but currently I find it way more useful than Firefox or Chrome's inspectors.
I just don't like Chrome, for one. But, I really like the scrolling, pinch to zoom, swipe to go backwards, bookmark syncing across my multiple computers, etc.
I prefer Safari because it integrates into the Apple ecosystem, including e.g. iCloud tabs, Keychain integration, gesture support; because it's a native application; because it's perfectly performant; because it doesn't phone home. I don't write web applications or even use many, so my needs are well satisfied.
It blows the other browsers away in scrolling performance on a retina display laptop. Chrome has gotten a lot better recently, but it's still worse enough that it's worth putting up with Safari's weaknesses.
Some of configuration setting has been removed or burried deep in about:config. Now I can not find TLS 1.0 checkbox, because I am hoping this release support TLS 1.1 and 1.2. But, there is no sign of this support.
Beside that, I love FF. Thank you for this release and everything.
I'm on XP at work. I get a new computer in January though and all the new ones have W7. I must have been one of the last people to get an XP computer here.
This OS is what, 12 years old now? Linux was on version 2.4.x when XP came out. OSX wasn't around yet, or was brand spanking new. I hate MS as much as anyone but that's still pretty impressive.
Technically each SP refreshed the code base and many other libraries are independent of the platform, which is why it has lasted so long.
SP3 was released late 2008, so it's "only" 5 years old.
I'm not a gamer so DX9 is fine, all the .net libraries work on it, it has NTFS support just as fine as W7 and W7 32-bit cannot support any more memory than XP can.
XP is just a way for me to launch programs, W7 is not going to launch them any better.
In that inexperienced users are not well-informed of the option to disallow it, and might not understand that indiscriminate, promiscuous execution of JavaScript as a default behavior can be hazardous. ...at least as hazardous as connecting to unencrypted Wi-Fi.
I have an omegle exploit that exists entirely in javascript. it was written for Chrome (Windows/Mac specifically), but still 100% javascript. I'd be happy to email a copy of it (in .zip form) to anyone interested.
Also, the recent Tor Browser Bundle exploit (RIP Tormail) was javascript-based.
I keep switching from Chrome to Firefox for reasons of principle ... and keep switching back to Chrome for reasons of performance. I hate myself, but can't help it. Firefox will briefly freeze scrolling a long page. Firefox will beach ball for 15 seconds after resuming laptop from sleep. And so on. None of it is crash/burn awful, but it adds up to be just irritating enough that it makes me mental. I tolerate it until I can't (for a couple weeks), then switch back to Chrome. Just did so yesterday, which is why I'm ranting in response to this headline.
This on OS X. Couple years ago when I was using Windows 7 all the time, Firefox was great. IME the fit/finish on OS X is not the same.
tl;dr: Firefox, I want to love you and be exclusive. Please change.