Slashdot Mirror
Firefox 55 Arrives With WebVR on Windows, Performance Panel, and Click-to-Play Flash (venturebeat.com)
Mozilla today made available a new update to Firefox for Windows to introduce support for WebVR, that the company says, will enable desktop VR users to dive into web-based experiences with ease. Firefox 55 also includes performance panel, faster startup when restoring multiple tabs, a quicker way to search across various search engines, and click-to-play Flash by default. From a report: WebVR is an experimental JavaScript API that provides support for virtual reality devices, such as the HTC Vive, Oculus Rift, and Google Cardboard. As its name implies, the technology is meant for browsers. If you find a web game or app that supports VR, just click the VR goggles icon visible on the web page to experience it using your VR headset. WebVR supports navigating and controlling VR experiences with handset controllers or your movements in physical space. [...] Firefox 55 also allows users to adjust the number of processes and how much resources they want to allocate to any of them. This setting is at the bottom of the General section in Options. In fact, if your computer has more than 8GB of RAM, Mozilla recommends "bumping up the number of content processes that Firefox uses" because it will make Firefox faster, though at the expense of using more memory. In its own tests on Windows 10, the company found that Firefox uses less memory than Chrome, even with eight content processes running.
"Firefox does not support downgrades, even though this may have worked in past versions. Users who install Firefox 55+ and later downgrade to an earlier version may experience issues with Firefox. "
WTF? Can't they just let it die completely. Geez.
Enabling the dysfunctional. Bad policy.
Moz has their head in their ass wasting time on "experimental JavaScript" features there is almost no demand for
I'm sure it will allow you to view awesome Web VR! For about 5-10 minutes until it slows to a crawl from poor memory management and then you have to restart. It's a feature!
I'm guessing this is specifically in reference to using Web VR?
Firefox 55 also allows users to adjust the number of processes and how much resources they want to allocate to any of them. This setting is at the bottom of the General section in Options. In fact, if your computer has more than 8GB of RAM, Mozilla recommends "bumping up the number of content processes that Firefox uses" because it will make Firefox faster, though at the expense of using more memory. In its own tests on Windows 10, the company found that Firefox uses less memory than Chrome, even with eight content processes running.
I have a Chrome browser open that's been running for days with multiple tabs open still as snappy as when I first opened it and it's currently using a cool 163MB of RAM. My machine has 16GB of memory but it I obviously barely need to drip into it? Oh and by the way Mozilla... Chrome just figures it out without users having to tweak their browser's memory management strategy manually. I guess you can't figure out to do that? Furthermore, I guess Firefox isn't for grandma then?
For reals, I used to use Firefox and Firebug exclusively for web development for many years before Chrome de-throned you guys. We need less spin more real value to consider switching back. C'mon Mozilla, either get in the game or admit defeat.
We'll make great pets
As a long time Firefox user, I'm scared about the upcoming Firefox 57 release. According to that Mozilla blog post, as of Firefox 57 "Firefox will only run WebExtensions." So that could mean a lot of existing extensions will no longer work.
Firefox's market share has already fallen precipitously. The most-used release, Firefox 54, only has 3.75% of the market. The next most popular release of Firefox, Firefox 52, has only 0.52%. Firefox for Android has only 0.04% of the market. Overall, Firefox has only about 5% of the market.
5% is dangerously low. Chrome, for example, is over 50%. Safari has over 10%. UC Browser for Android has over 9%. Firefox is now in the range of Opera Mini, with its 3% of the market.
I think that Firefox 57, and this switch to WebExtensions, will be what finally eliminates Firefox as a viable web browser. Broken extensions will no doubt anger many of Firefox's few remaining users. I would not be surprised if many of them will move to some other browser. And these are users that Firefox can't afford to lose.
Firefox 57 takes away one of the few remaining strong points of Firefox: its flexible and powerful extension system.
Firefox doesn't even have privacy working in its favor any longer. Firefox's privacy policy shows that it sends a lot of information to Mozilla and others. For example, it indicates that even Firefox's geolocation capabilities can use Google's service, and this involves sending information to Google.
So Firefox will soon become an almost identical clone of Chrome, including its own imitation of Chrome's UI, Chrome's extension model, and Chrome's privacy concerns. Yet Firefox still can't match Chrome's performance, even if some Firefox fanatics claim otherwise. Firefox users have clearly indicated that they find Firefox to be too slow, too bloated, and to use too much memory.
For all intents and purposes, we should probably consider Firefox to be a "dead" browser at this point. Its market share is dropping, and could very well be under 1% by this time next year. It has dropped to such a low range that web developers no longer test with it. This will likely result in more and more web sites that don't work well with Firefox, making the Firefox user experience even worse.
Users won't waste their time with Firefox when they can just use Chrome instead, and get the same UI and privacy experience, but with much better performance and reliability.
The worst part of all of this is that it didn't have to be like this. Firefox's developers didn't need to copy Chrome. They didn't need to ruin the Firefox user experience that Firefox users had come to love. Firefox could have been its own independent browser. Yet all of this potential has been discarded, and the end result is disturbing: Firefox is, or soon will be, an unusable browser for most of its users.
another vulnerability.
What "world"? Out of 15 extensions I have enabled 14 are currently are showing [LEGACY] flag on the options page.
Looks to me like firefox is about to die.
... and less than 5% of users care.
#DeleteFacebook
I use FF with multiprocess enabled, and I find it a lot slower than Chrome. This is true even when I try a fresh profile, and even after installing ad blocking and other extensions. Everything about FF feels so slow to me. The UI feels sluggish. Pages feel like they take longer to load. Scrolling a loaded page doesn't feel smooth. JS intensive pages that work fine in Chrome lag badly for me in FF. More and more I find myself using Chrome because it feels to much faster than Chrome. I don't like using Google products, and I don't even like Chrome, but Chrome is just so much more usable than FF these days. Have these performance problems beem fixed in this release?
Really now, Mozilla. Who on Earth wants VR support in their browser?
Stop dumbing it down. Stop adding useless garbage bloat. Make the damn browser better.
rm -f /usr/local/bin/firefox
rm -rf ~/.mozilla
See ya!
And is there a list of features and corresponding prefs entries to disable them?
Do people actually use this phrase outside of TV comedy shows?
you should know the issue:
-In About:Config you disable Pocket and buffering pages when hovering over links.
-you upgrade to the latest version of FF
-Pocket is back and hovering over links buffers the page of said links.
Why does Mozilla force us to live in FF's version of Groundhog day?
Obama's legacy: (N)othing (S)ecure (A)nywhere and (T)error (S)imulation (A)dministration
this AC has 53 addons, 50 are marked as [legacy], i think Mozilla is truly underestimating the damage webextensions and the depreciation of the XUL will do to the browser, millions of dev man hours and unique functionality up the shitter.
And the only instances I can think of where things were "broken" in Firefox wasn't so much that they were broken as that the layout wasn't quite right due to slight differences in CSS.
Counterexample: Custom emoji uploading in Discordapp.com (a without-charge replacement for Slack) works in Chrome but not Firefox. A user with "Manage Emojis" privileges on a server can rename or delete emojis using Chrome or Firefox, but uploading new ones works only in Chrome. Clicking the "Upload Emoji" button in Firefox just makes the button go in and out; it doesn't display a file chooser as expected. I found it surprising because avatar uploading and photo attachment uploading work fine.
because it will make Firefox faster, though at the expense of using more memory.
Firefox already uses an obscene amount of memory. The longer it runs, the more ridiculous its memory consumption gets, as it gets slower and slower to the point where it becomes unusable. Then crash, restart, repeat.
This still has not been fixed and now they're thinking up new ways to make it even worse? Firefox has become the Hummer of browsers. No wonder Chrome is taking over. NoScript is Firefox's sole redeeming feature at this point.
The process must be very hard to convert an add-on to the new standard. Even Mozilla's own "Add-on Compatibility Reporter" is listed as legacy.
https://addons.mozilla.org/en-...
If Mozilla can't even do it then how do they expect others to do it it??
Most all my add-ins are listed as Legacy. Some are listed a compatible with Multiprocess and some are not.
Well, welcome to the wonderfull world of FireFox ESR.
Firefox ESR 52 will allow you to keep your old time extensions until Aprox April next year.
Also, your NPAPI plug-ins (Like the ones you have to use for SabaMeeting, Cisco WebEx, and al sorts of ILO plugins for server and network gear).
Just install (or downgrade to) Firefox ESR52.
While this is not optimal, It will bid you time so that your Plug-ins and AdOns are ported to the new FireFox framework (which is SIMILAR BUT NOT EQUAL to that of chrome). Or ported to chrome, or whatever other solution your provider of said Plug-ins or AdOns considers...
I've been on the ESR channel since its inception (I can not have my workflow disrupted every three months or so, when the firefox devs decide to change another thing).
While is not a bed of roses (specially at the end of the life of the ESR, when pretty much all sites believe your browser is "out of date and insecure" [which it is NOT]), is better than the alternative for people like us who use the browser as a WORK tool first and foremost, with recreational uses in the backseat...
Having said that, I believe that the Direction Firefox is taking under the hood (I will NOT enter a UI/UX holly war) in order to increase performance and security is the right one, and a little pain in the short term is whorthwile for the performance and security rewards that will be collcted later on....
*** Suerte a todos y Feliz dia!
Ok, so extensions.pocket.enabled is obvious. I just searched for buffer and do not get any settings that appear to handle your second concern, so what setting is it?
Broken extensions will be EXTREMELY destructive to Firefox, in my opinion. Broken extensions will be as though Mozilla Foundation spent $100 million on advertising to kill Firefox. Extensions are the main reason I use Firefox and Pale Moon (Pale Moon had a 64-bit version before Firefox).
I installed Google's Chrome browser a long time ago. I discovered Chrome had installed 3 system services. So Chrome and Google had more control over my computer than I normally allow myself. Now, no more Chrome on any of my computers.
Why do software company managers become self-destructive? Firefox managers are EXTREMELY self-destructive, in my opinion. Google is rapidly traveling from "Do no evil" to "Do evil if it make money" if that initially makes money, in my opinion.
My Firefox and Pale Moon extensions
The first is a Pale Moon ad-blocker. Some Firefox extensions don't work in Pale Moon:
"This add-on will stop working when Firefox 57 arrives in November 2017."
"This add-on will stop working when Firefox 57 arrives in November 2017 and Mozilla drops support for XUL / XPCOM / legacy add-ons. It should still work on Firefox 52 ESR until ESR moves to Firefox 59 ESR in 2018 (~Q2).
"There is no 'please port it' or 'please add support for it' this time, because the entire add-on eco system changes and the technology behind this kind of add-on gets dropped without replacement."
USE THIS: ghostery-5.4.10-sm+an+fx.xpi Link: Version 5.4.10
Ghostery sells data it collects. (Business Insider, Jun 18, 2013)
Ghostery web site
Something like the following should do the trick:
PROFILE_NAME="23k5nqzr.default-1404250880109"
PROFILE_DIR="~/.mozilla/firefox/${PROFILE_NAME}"
sudo for i in "browser.pocket.enabled"
do
if grep $i ${PROFILE_DIR}/prefs.js
then
sed -i -e 's/^user_pref($i, \(true\|false\));$/user_pref($i, false);/' ${PROFILE_DIR}/prefs.js
else
echo "user_pref($i, true);" >> ${PROFILE_DIR}/prefs.js
fi
done
sudo for i in "network.http.speculative-parallel-limit"
do
if grep $i ${PROFILE_DIR}/prefs.js
then
sed -i -e 's/^user_pref($i, \(0\|1\));$/user_pref($i, 0);/' ${PROFILE_DIR}/prefs.js
else
echo "user_pref($i, true);" >> ${PROFILE_DIR}/prefs.js
fi
done
Not ready to jump from Chrome to Firefox, although I do use Firefox from time to time. Its better and will be watching to see how FF 57 comes out of the gate. Will it stumble with loads of problems? Or be a real refresh capable of winning over Chrome users? Guess we will see.
Search for "specul". It's internally called speculative buffering.
I've been trying to use firefox again due to privacy concerns with Chrome. I've been really frustrated by the poor performance, despite the new multi-process.
I just realized, now that there's a GUI (performance panel) that I multi-process wasn't enabled because I had a single legacy addon. I disabled that addon and now FireFox is MUCH faster.
I know everyone hates that the legacy addons are going away--and I do think FireFox needs to do something about this--support them longer, fund development of the most popular addons, something... but MultiProcess FF is amazingly faster than before. I would never want to go back...
-=Lothsahn=-
Genuine question : what do you mean with "hovering over links buffers the page of said links"? Thanks.
Genuine question : what do you mean with "hovering over links buffers the page of said links"? Thanks.
https://news.slashdot.org/story/15/08/14/2321202/how-to-quash-firefoxs-silent-requests
It's sad to see a once respectable team deliberately torpedo their own project. But they have done so with their users openly opposing the change. They deserve it.
I use Pale Moon on Linux and, for now, Brave on Android. Frankly, my web experience is so bad I don't really enjoy browsing any more. Gopherspace is enjoying a small surge in popularity from real people instead of corporations. IRC and GNU social replace proprietary chat and Twitter.
Remind me why the modern Web is good for users? It's terrible. I have to either submit to privacy-destroying scripts, or run multiple extensions (that FF57 will kill) and pray they work.
Sorry, the Web isn't fit for purpose.
Different AC here, but I think it means that when you hover the mouse cursor over a link, FF will go off in the background and prefetch the page contents, so that when you click on the link there's less of a delay to show you the page.
Of course, this can end up loading a lot of things you didn't explicitly ask to load, and arguably has privacy implications too since it gives away information that used to be only local. Also if you have bandwidth limits it can eat into that for things you will never view.
You can disable the behavior by setting network.http.speculative-parallel-limit to 0 in the about:config page. I do that routinely when setting up a new FF profile. There's a LOT of things you have to change to stop leaking data all over, but... at least you kinda can, in FF, which beats most other browsers.
A javascript API for VR.. in a web browser. VR already taxes hardware and requires tight synchronization of sensor inputs over USB3 to video framebuffer output to produce a usable experience. Adding javascript, another layer of abstraction, and frankly, a shitty wannabe operating system masquerading as a web browser into the mix IF FUCKING STUPID AND A WASTE OF TIME.
I believe it is network.predictor.enabled
Repeat after me: "We put our developer efforts into replacing public CAs and implemented DNSSEC DANE TLSA."
There is nothing else that matters. You've dragged your feet for 6 years on one open issue in your issue tracker and, at one point closed the ticket as "WONTFIX". Public CA infrastructure is an eyesore and a serious security issue that you clearly aren't taking seriously. You know, Netscape gave us the first public CA infrastructure with Verisign and Mozilla inherited the problem when you took the Netscape source code for the basis of Mozilla browser. You bought the problem hook, line, and sinker. It is up to you to fix it. DANE TLSA came along and your response has been worse than tepid: You got in bed with the EFF and had a Let's Encrypt baby which just further propagates the problem but does not solve it! In addition, the sponsors of Let's Encrypt spend a collective $1.1 million USD annually of which Mozilla and the EFF are the premiere sponsors but you've only committed $20K to DANE TLSA...and that to a third party. You cannot absolve yourself of the responsibility beholden to you. ANY announcement that you make other than an implementation of DNSSEC DANE TLSA is a complete and absolute failure of your organization and the developers who work for it.
Firefox has been moving in the right direction as far as multi process/performance updates. I don't care about the VR atm but the performance panel could be a positive addition. On the desktop Firefox is now close to- or at parity with Chrome in most benchmarks so I cant complain but I want them to focus on the baisics and forget about the flashy extras. (I hide pocket and the like first thing when I set up Firefox.)
FYI, Firefox 55 reduces memory usage dramatically (e.g. that link shows it going from over 2GB to under 500MB for the author's 1691 tab test case; the start up time difference is even more dramatic).
So this particular release actually addresses your memory usage complaint.
Now I have a reason to pick up my 32" monitor and glue it into my face! Oh I can't wait for the excitement of spinning my head in the room for the full VR experience, while my monitor smashing everything in sight. /s
Thanks for the info, but I don't understand the connection with pocket.
I think what you says about prefetching is independent from pocket, or I'm wrong?
Comment removed based on user account deletion
WebVR, huh... so they are bringing this stuff back. And everyone in the mainstream is painting this as some kind of wonderful news about innovation https://software.informer.com/... It's like that every time a useless gimmick is introduced, like that facebook webpage ranking https://droidinformer.org/Stor... I mean, who gives a damn? But this WebVr is going to be paraded around until people are tied of it. And then it will be obsolete and forgotten.
faster startup when restoring multiple tabs
The long wait on startup with multiple tabs open appeared several versions ago; since I often keep lots of tabs open, it was quite noticeable. So let's see: they made a regression, took a very long time to fix it, and now we're supposed to cheer? I can run an innovative software company like that...
There is a major deadlock bug so updates have stopped. Expect fixed version soon.
Java is owned by Oracle. The license of the Java Language Specification appears to prohibit distributing an incomplete implementation from scratch of the Java platform, whether called Java or called something else. This means it's a breach (and therefore a copyright infringement) for a group of hobbyists to reimplement the Java platform in a public repository. Instead, a reimplementation of the Java platform must be performed behind closed doors, distributed to the public only once it becomes a "Compliant Implementation".
ECMAScript is an Ecma standard, and HTML, CSS, and the HTML DOM are W3C Recommendations. Unlike Java, JavaScript (the combination of ECMAScript with the HTML DOM) allows making a partial implementation public.
In what way do the advantages of Java over JavaScript outweigh the disadvantage of ownership by One Rich American Called Larry Ellison?
PCs from the dial-up era were fast enough to play SWF animations from the dial-up era, and Flash Player let the user dial the FSAA up or down. Nowadays, CPUs are faster, and GPU acceleration is more common. Adobe is the limiting factor in making Flash Player not suck, and I concede that the company has recently punted on that.
No one said about distribution. And if you care much about distribution, then distribute the compliant implementation, OpenJDK/JRE, problem solved. If the distribution size is the issue, luckily Java9 will solve this issue with its modules paradigm.