Slashdot Mirror
Firefox 51 Arrives With HTTP Warning, WebGL 2 and FLAC Support (venturebeat.com)
Reader Krystalo writes: Mozilla today launched Firefox 51 for Windows, Mac, Linux, and Android. The new version includes a new warning for websites which collect passwords but don't use HTTPS, WebGL 2 support for better 3D graphics, and FLAC (Free Lossless Audio Codec) playback. Mozilla doesn't break out the exact numbers for Firefox, though the company does say "half a billion people around the world" use the browser. In other words, it's a major platform that web developers target -- even in a world increasingly dominated by mobile apps.
Is it really?
One of the recent updates (48/49/50( absolutely KILLED the performance. Particularly annoying is the URL bar. Autocomplete results take longer to populate and my usual pattern of opening tabs was broken. I used to type in a few characters, select the entry, and hit enter. For example: sl, down (or tab), enter, ctrl+t, ca, down (or tab), enter, ctrl+t, etc. would open up slashdot, then a new tab for my calendar, then a new tab for... Ever since the performance tanked, I couldn't do that anymore without deliberately slowing down at each step.
I even tried blowing out all of my old history (years and years of browsing data on one machine). This was particularly annoying as clearing out everything older than 6 months will do so based on the FIRST access date, not the last access date. So clearing out everything older than 6 months blows out slashdot even though I access it daily. I had to go into each subfolder in the history control and sort by last access date, then blow out everything older than a threshold of a few months back. This took almost an hour of constant work because deleting history this way causes FF to update the UI constantly. CPU usage spiked to 100% of a single core while FF deleted an entry, updated the scroll bar, scrolled the list, then deleted the next entry. To prevent FF from locking up completely and crashing I had to work in batches of a few thousand and let it stew for a couple of minutes before hitting the next batch.
And after all that work, with a history file that was in the hundreds of thousands instead of tens of millions, performance was still ass.
Webkit can probably count all the mobile users who use the default browser as part of its user count. Similarly, maybe Firefox is the browser being used on ATM screens on mall info terminals, that would add a lot of 'users' who are actually clueless as to which browser is underlying their UX.
Yet? I'll give it a chance when it's at least as sandboxed as chrome.
Although the world has largely switched to Chrome, the remaining use for Firefox is as the one browser that is still willing to support Java applets. Lots of people who work in IT have a VM or a jumpbox whose only purpose is to run Java applets inside of Firefox (for example, to do maintenance on some piece of equipment with a Java-applet-based configuration tool -- I'm talking to you, EMC) -- and *never* *run* *updates* because changing the browser or java version even slightly will break the whole thing.
Tired of FB/Google censorship? Visit UNCENSORED!
In other words, itâ(TM)s a major platform that web developers target -- even in a world increasingly dominated by mobile apps.
I'm not sure I understand this comment. I use Firefox as my main browser on both of my mobile platforms (phone and tablet). It probably just beats Twitter as my #1 app used on those devices. Why would someone imply they are some incompatible with each other?
Not relevant to the HTTP/S issue, but not too long ago, the UK was spoofing slashdot to attack their targets, so it could happen.
the preceding comment is my own and in no way reflects the opinion of the Joint Chiefs of Staff
Try with a fresh Fx profile to compare performance to your usual one. Nine times out of ten, I find that it's some obscure setting people flipped in about:config or an addon that's causing their massive performance problems. The rest are generally caused by people using the heaviest web apps out there, and wondering why their browsers aren't running as well as they did years ago before all of these resource-hungry apps.
Is there any way to tell which extensions are blocking multiprocess? My about:support page says multiprocess is disabled because of extensions, but it doesn't say which ones. It seems like they should publish this information, perhaps in a field on AMO. A Google only turns up results for developer testing or small lists, it says nothing about a complete list of incompatible extensions.
I don't know if anyone at Mozilla reads Slashdot any longer, but I think this would be a worthwhile documentation project that would help users demand extension authors make their software compatible, thus aiding the roll-out.
I don't need a bunch of annoying security messages that either tell me something I already know, or that tell me something I have already decided to ignore. What, I think, _everybody_ wants from a browser is: - decent memory management (I don't care if I page takes 2ms more to load, especially if the price is to have a browser taking 2gb of RAM and making my system unusable) - minimalist interface - fast startup time - fast page load Extra points if it also has: - add-on support - sound and videos on page load disabled by default - address autocomplete That's it. Really. It's a fucking browser. It's sole purpose is to render web-pages. I don't need another operating system. If I wanted that, I'd install a VM or emacs.
I don't need a bunch of annoying security messages that either tell me something I already know, or that tell me something I have already decided to ignore. What, I think _everybody_ wants from a browser is:
- decent memory management (I don't care if I page takes 2ms more to load, especially if the price is to have a browser taking 2gb of RAM and making my system unusable)
- minimalist interface
- fast startup time
- fast page load
Extra points if it also has:
- add-on support
- sound and videos on page load disabled by default
- address autocomplete That's it. Really. It's a fucking browser. It's sole purpose is to render web-pages. I don't need another operating system. If I wanted that, I'd install a VM or emacs.
The problem is likely the internal sqlite database fragmentation. On Windows you can run tools like "speedyfox" in order to defrag the database.
Bad performance is likely caused by having lots of extensions enabled.
Christ. Heaven forbid someone points out that problems are often solved by troubleshooting, rather than pissing and moaning.
>Why do I have to reset my Firefox profile every couple of months just to get it to kind of work better, while I never have to do that for Chrome?
I don't know. Why do so many other people NOT have to do that, yet you have to? If you don't want to find out, then fuck off with your holier-than-thou attitude about how Firefox sucks because it's just not stroking you off the right way.
>I've used the same Chrome profile for years now
>Chrome works perfectly.
Yes, and there are many people who would say the same about Firefox, because people's needs from a browser aren't exactly the same as yours.
Look. I'm glad Chrome works perfectly for you. I'm also perfectly willing to accept that Firefox doesn't work for you. The question is still why? And why are you so against figuring that out, instead of felating Chrome like a cheap whore and then accusing others of doing the same for Firefox?
>Why is Firefox so much worse than Chrome? Why are Firefox's users constantly told to use fresh profiles, while Chrome's users don't have to do this?
Gee, I don't know, maybe it's because Firefox allows you to customize far more than Chrome, to the point of letting you screw up? Or that it has many more third party apps like shitty A/V software fucking with it? Perhaps because it's is so much older than Chrome that it has needed extreme overhauls, while not breaking everything for all of its users as it does so?
I seriously wonder if you guys who say these things have even tried to reset your profiles, or if you're just avoiding any troubleshooting at all. Maybe if you collectively spent less time bitching in vain online while stroking Chrome's ego, you could have helped to diagnose whatever problems you're having, they would have been fixed, and all of this finger-waggling of yours wouldn't be necessary anymore.
The problem with the database Firefox uses is it doesn't actually delete entries when you delete them. It just marks them as deleted so apparently database queries still spend time skipping over them. They aren't actually removed until you perform a vacuum operation on the database. As you found out, there are tools like Speedyfox which do that for you.
Firefox has tons of issues. I'm amazed it's gotten as far as it has with so many fundamental design mistakes. But I guess that really just shows you the sorry state browsers are in. Granted they're required to do a lot, but they implement the basics so poorly.
So somehow they're saying that there are 10 billion people in the world?
Yup. And of those 10 billion, at last four billion love, love, love Australis. And another two billion want Pocket integrated into Firefox. The only downside to the Mozilla... sorry, MI//lla://: whatever it's called now good-news tour is that one or possibly two people have complained about problems with memory leaks, but luckily that affects so few people that it's not worth addressing.
Firefox does seem to automatically clean places.sqlite. Not sure what the interval (or trigger) is, but I've got an about:config value, storage.vacuum.last.places.sqlite, which seems to store an epoch time for the last time it was done. In my case, it corresponds to 24 Dec. 2016, or a little under a month ago. So they're not leaving it wholly uncleaned it appears.
$_ = "wftedskaebjgdpjgidbsmnjgcdwatb"; tr/a-z/oh, turtleneck Phrase Jar!/; print
Did they really roll out Firefox 51 for Android?
F-Droid, Google Play, and their own direct download link all seem to still have 50.1.0. Clicking the "check for updates" on the "About Firefox" screen says "no updates available".